United States
Environmental Protection
Agency EPA/600/R-13/202
First External Review Draft
November 2013
www.epa.gov/ncea/isa
Integrated Science Assessment
for Oxides of Nitrogen-
Health Criteria
(First External Review Draft)
November 2013
National Center for Environmental Assessment-RTF Division
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
-------
DISCLAIMER
This document is the first external review draft, for review purposes only. This information is
distributed solely for the purpose of pre-dissemination peer review under applicable information
quality guidelines. It has not been formally disseminated by EPA. It does not represent and should
not be construed to represent any Agency determination or policy. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
November 2013 ii DRAFT: Do Not Cite or Quote
-------
CONTENTS
ISA TEAM FOR OXIDES OF NITROGEN xiii
AUTHORS, CONTRIBUTORS, AND REVIEWERS xvi
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE OXIDES OF NITROGEN NAAQS
REVIEW PANEL xx
ACRONYMS AND ABBREVIATIONS xxi
PREAMBLE xxx
1 Process of ISA Development xxx
Figure I Schematic of the key steps in the process of the review of National
Ambient Air Quality Standards. xxxi
Figure II Characterization of the general process of ISA development. xxx///
2. Literature Search xxxiv
Figure III Illustration of processes for literature search and study selection
used for development of ISAs. xxxv
3. Study Selection xxxv
4. Evaluation of Individual Study Quality xxxv/
a. Atmospheric Science and Exposure Assessment xxxvii
b. Epidemiology xxxviii
c. Controlled Human Exposure and Animal Toxicology xxxix
d. Ecological Effects xl
5. Evaluation, Synthesis, and Integration Across Disciplines and Development of Scientific
Conclusions and Causal Determinations xli
a. Evaluation, Synthesis, and Integration of Evidence Across Disciplines xlii
b. Application of Framework for Scientific Conclusions and Causal Determinations xlvi
Table I Aspects to aid in judging causality. xlvii
c. Determination of Causality xlviii
Table II Weight of evidence for causal determination. /
6. Public Health Impact
7. Approach to Classifying At-Risk Factors
Table III Classification of evidence for potential at-risk factors.
8. Quantitative Relationships: Effects on Welfare liv
9. Concepts in Evaluating Adversity liv
a. Evaluating Adversity of Health Effects liv
b. Evaluating Adversity of Ecological Effects Iv
References for Preamble Mi
PREFACE MX
Legislative Requirements for the Primary NAAQS Review lix
Introduction to the NAAQS for Nitrogen Dioxide (NO2) Ixi
History of the Review of Air Quality Criteria for the Oxides of Nitrogen and the NAAQS for NO2 Ixi
Table IV History of the National Ambient Air Quality Standards for NO2 during
the Period 1971-2012. Ixii
References for Preface Ixv
EXECUTIVE SUMMARY Ixvi
Purpose of the Integrated Science Assessment Ixvi
Scope and Methods Ixvi
Sources of Oxides of Nitrogen to Human Exposure Ixviii
Health Effects of Oxides of Nitrogen /xx/
November 2013 iii DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Table ES-1 Causal determinations for NO2 by exposure duration and health
effect category from the 2008 ISA and current draft ISA for oxides of
nitrogen. Ixxiii
Health Effects Associated with Short-term NC>2 Exposure Ixxiv
Policy-relevant Considerations for Evaluating Health Effects Associated with Short-term NO2
Exposure Ixxvi
Health Effects Associated with Long-term NO2 Exposure Ixxviii
Policy Relevant Considerations for Evaluating Health Effects Associated with Long-term NO2
Exposure Ixxix
Conclusions Ixxx
References for Executive Summary Ixxxi
CHAPTER 1 INTEGRATED SUMMARY 1-1
1.1 ISA Development and Scope 1-2
1.2 Organization of the ISA 1-5
1.3 Sources of Oxides of Nitrogen to Human Exposure 1-6
1.3.1 Sources of Oxides of Nitrogen 1-6
1.3.2 Atmospheric Chemistry and Fate of Oxides of Nitrogen 1-7
1.3.3 Ambient Concentrations - Temporal and Spatial Trends 1-7
1.3.4 Human Exposure to Oxides of Nitrogen 1-9
1.4 Health Effects of Oxides of Nitrogen 1-11
1.4.1 Dosimetry and Modes of Action Informing Respiratory Effects 1-12
1.4.2 Respiratory Effects 1-13
1.4.3 Dosimetry and Modes of Action Informing Extrapulmonary Effects 1-18
1.4.4 Cardiovascular Effects 1-19
1.4.5 Total Mortality 1-22
1.4.6 Reproductive and Developmental Effects 1-24
1.4.7 Cancer 1-27
Table 1-1 Key evidence contributing to causal determinations for NO2
exposure and health effects evaluated in the current draft ISA for
1.
1.
1.
Oxides of Nitrogen.
5 Evaluation of the Independent Effects of NO 2
1.5.1 Potential confounding by time-varying factors and individual- or population-level
characteristics
1.5.2 Potential confounding by copollutant exposures
1.5.3 Summary of Evaluation of the Independent Effects of NO2 Exposure
6 Policy-Relevant Considerations
1.6.1 NO2 Exposure Metrics
1.6.2 NO2 Laq Structure in Epidemioloqic Studies
1.6.3 Concentration-Response Relationships and Thresholds
1.6.4 Regional Heterogeneity in Effect Estimates
1.6.5 Public Health Significance
7 Conclusions
References for Chapter 1
1-29
1-35
1-36
1-37
1-42
1-43
1-43
1-45
1-46
1-48
1-49
1-54
1-56
CHAPTER 2 ATMOSPHERIC CHEMISTRY AND EXPOSURE TO OXIDES OF
NITROGEN 2-1
2.1 Introduction 2-1
2.2 Atmospheric Chemistry and Fate 2-1
Figure 2-1 Schematic diagram of the cycle of reactive, oxidized nitrogen
species in the atmosphere.
2.3 Sources
2.3.1 Overview
Figure 2-2 Major sources of A/Ox averaged over the United States, 2008.
Figure 2-3 National average A/Ox emissions attributed to on-road vehicles and
fuel combustion, from 1990 to 2012.
2.3.2 Highway Vehicles
2.3.3 Off-Highway Vehicles
2.3.4 Aviation Emissions
2.3.5 Shipping Emissions
2-2
2-8
2-8
2-9
2-10
2-11
2-12
2-13
2-14
November 2013 iv DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
2.3.6 Locomotive Emissions 2-14
2.3.7 Fuel Combustion for Electrical Utilities and Industrial Use 2-15
2.3.8 Biogenics and Wildfires 2-16
2.3.9 Lightning 2-17
2.3.10 Oil and Gas Development 2-18
2.4 Measurement Methods 2-19
2.4.1 Federal Reference and Equivalent Methods 2-19
2.4.2 Other Methods for Measuring NO2 2-21
Figure 2-4 Comparison of NC>2 measured by CAPS (Cavity Attenuated Phase
Shin) spectroscopy to NO2 measured by chemiluminescence. 2-22
Figure 2-5 Comparison of NO2 measured by QC-TILDAS (Quantum Cascade-
Tunable Infrared Differential Absorption Spectroscopy) to NO2
measured by chemiluminescence with photolytic converter. 2-23
2.4.3 Satellite Measurements of NO2 2-24
Figure 2-6 Seasonal average tropospheric column abundances for NO2
(1015 molecules/cm2) derived by OMI for winter (upper panel) and
summer (lower panel), for 2005 to 2007. 2-25
Figure 2-7 Seasonal average tropospheric column abundances for NO2
(1015 molecules/cm2) derived by OMI for winter (upper panel) and
summer (lower panel), for 2010 to 2012. 2-26
2.4.4 Measurements of Total Oxidized Nitrogen Compounds (NOY) in the Atmosphere 2-27
2.4.5 Ambient Sampling Network Design 2-28
Figure 2-8 Map of oxides of nitrogen monitoring sites in the U.S., from four
networks (SLAMS, Castnet, A/Core, and SEARCH). 2-30
2.5 Ambient Concentrations of Oxides of Nitrogen 2-30
2.5.1 National Scale Spatial Variability 2-31
Figure 2-9 Annual average ambient NO concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors. 2-32
Figure 2-10 Annual average ambient NO2 concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors. 2-33
Figure 2-11 Annual average ambient NOX concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors. 2-34
Table 2-1 Summary statistics for 1-hour daily maximum NO2 concentrations
based on 295 SLAMS monitoring sites (ppb). 2-35
Table 2-2 Summary statistics for NO2, NO and NOX annual average
concentrations based on 295 SLAMS monitoring sites (ppb). 2-36
Figure 2-12 Seasonal average surface NO2 concentrations in ppb for winter
(upper panel) and summer (lower panel) derived by OMI/GEOS-
Chem, for 2009-2011. 2-38
2.5.2 Urban Scale Spatial Variability 2-38
2.5.3 Micro-to-neighborhood Scale Spatial Variability, Including Near Roads 2-39
Figure 2-13 Spatial distributions of CO, NO, NO2, and NOX concentrations at
the "near" (15 meters), "far" (80 meters), and Del Amo (background)
sites during winter and summer in southern California. 2-42
2.5.4 Seasonal, Weekday/Weekend and Diurnal Trends 2-47
Figure 2-14 January and July hourly profiles of NO and NO2 (ppb) for Atlanta,
Georgia (site with maximum NO2 levels). 2-48
Figure 2-15 WeekendAA/eekday hourly profiles of NO and NO2 (ppb) for Atlanta,
Georgia (site with maximum NO2 levels). 2-48
2.5.5 Multi-year Trends in Oxides of Nitrogen 2-49
Figure 2-16 National annual average ambient NO2 concentration trends,
1990-2012. 2-50
2.5.6 Background Concentrations 2-50
Figure 2-17 Simulated Asian contributions to CO, Os, PAN, and A/Ox
concentrations at 800 hPa in April, 2010. 2-52
Figure 2-18 Simulated total CO, O3, PAN, and NOX concentrations at 800 hPa
in April, 2010. 2-53
2.6 Exposure Assessment 2-53
2.6.1 Conceptual Model 2-53
2.6.2 Spatially Resolved Models for Use in Exposure Assessment 2-56
2.6.3 Personal Sampling Considerations 2-64
November 2013 v DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Table 2-3 Indoor NO2 and MONO concentrations in the presence and absence
of combustion. 2-67
2.6.4 Oxides of Nitrogen in a Multipollutant Context 2-69
Table 2-4 Synthesis of NO2 ambient-ambient copollutant correlations reported
in the literature. 2-73
Figure 2-19 Summary of copollutant correlation coefficients reported in studies in
Table 2-4. 2-77
Table 2-5 Pearson correlation coefficients between ambient NO2 and personal
copollutants. 2-78
Table 2-6 Pearson correlation coefficients between personal NO2 and ambient
copollutants. 2-78
Table 2-7 Pearson correlation coefficients between personal NO2 and
personal copollutants. 2-78
Table 2-8 Correlation coefficients between indoor NO2 and indoor
copollutants. 2-79
Figure 2-20 Spatial variability in concentrations of near-road pollutants, including
NO2, A/Ox, CO, PM2.s, and UFP. NO2, NO, and NOX concentration
gradients are presented in the center panel. 2-82
2.6.5 Considerations for Use of Exposure Metrics in Epidemiology 2-83
Table 2-9 Ambient, outdoor, transport, indoor, and personal NO2
measurements (ppb) across studies. 2-85
Table 2-10 Correlations between measured NO2 concentrations from personal,
indoor, outdoor, and ambient monitors. 2-90
Table 2-11 Meta-regression results from 15 studies examining the relationship
between personal exposure measurements and ambient
concentrations. 2-93
Table 2-12 Exposure measurement error metrics for comparing central site
monitoring data and various monitor averages compared with values
computed from a CTM. 2-96
Figure 2-21 Distribution of time sample population spends in various
environments, from the U.S. National Human Activity Pattern Survey
(alleges). 2-98
2.7 Summary and Conclusions 2-102
References for Chapter 2 2-104
CHAPTER 3 DOSIMETRY AND MODES OF ACTION FOR INHALED OXIDES OF
NITROGEN 3-1
3.1 Introduction 3-1
3.2 Dosimetry of Inhaled Oxides of Nitrogen 3-2
3.2.1 Introduction 3-2
3.2.2 Dosimetry of NO2 3-3
Table 3-1 Small molecular weight antioxidant concentrations in ELF and
predicted penetration distances for NO2. 3-9
3.2.3 Dosimetry of NO 3-18
3.2.4 Metabolism, Distribution, and Elimination of Products Derived from Inhaled Oxides of
Nitrogen 3-21
3.2.5 Summary 3-22
3.3 Modes of Action for Inhaled Oxides of Nitrogen 3-23
3.3.1 Introduction 3-23
Table 3-2 Chemical properties of NO2 and NO that inform modes of action. 3-24
3.3.2 NO2 3-25
3.3.3 NO 3-46
3.3.4 Metabolites of NO and NO2 3-50
3.3.5 Summary 3-52
Table 3-3 Biological pathways, key events and endpoints. 3-56
3.4 Summary 3-57
References for Chapter 3 3-58
November 2013 vi DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
CHAPTER 4 INTEGRATED HEALTH EFFECTS OF SHORT-TERM EXPOSURE TO
OXIDES OF NITROGEN
4.1 Introduction
4.2 Respiratory Effects
4.2.1 Introduction
4.2.2 Airway Hyperresponsiveness_
Table 4-1
Table 4-2
Table 4-3
Table 4-4
Table 4-5
Resting exposures to NC>2 and airway responsiveness in subjects
with asthma.
Exercising exposures to NO2 and airway responsiveness in subjects
with asthma.
Fraction of subjects with asthma having NO2-induced increase in
airway hyperresponsiveness to a nonspecific challenge.
4.2.3 Lung Function_
Table 4-6
Figure 4-1
Table 4-7
Fraction of subjects with asthma having NO2-induced increase in
specific airway hyperresponsiveness to an allergen challenge.
Fraction of subjects with asthma having NO2-induced increase in
airway hyperresponsiveness regardless of challenge types.
Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Associations between ambient or personal NO2 concentrations and
FEVi in children with asthma.
Epidemiologic studies of lung function in children and adults with
respiratory disease.
Table 4-8 Controlled human exposure studies of NO2 and lung function.
4.2.4 Pulmonary Inflammation, Injury, and Oxidative Stress
Table 4-9
Table 4-10
Table 4-11
Table 4-12
Table 4-13
Figure 4-2
Table 4-14
4.2.5 Host Defense
Controlled human exposure studies of NO2 and pulmonary
inflammation, injury, and oxidative stress.
Animal toxicological studies of NO2 and pulmonary inflammation,
injury, and oxidative stress. _
Controlled human exposure studies of NO2 and allergic
inflammation.
Animal toxicological studies of NO2 and allergic inflammation.
Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of pulmonary inflammation and oxidative
stress.
Associations of personal or ambient NO2, NO, or NOX with exhaled
nitric oxide (eNO) in various populations.
Epidemiologic studies of pulmonary inflammation, injury, and
oxidative stress.
4.2.6
Table 4-15 Animal toxicological studies of NO2 and lung host defense.
Table 4-16 Controlled human exposure studies of lung host defense.
Respiratory Symptoms and Asthma Medication Use
Table 4-17 Mean and upper percentile concentrations of oxides of nitrogen in
Figure 4-3
Table 4-18
epidemiologic studies of respiratory symptoms.
Associations of ambient NO2 concentrations with respiratory
symptoms and asthma medication use in children with asthma.
Epidemiologic studies of respiratory symptoms and asthma
medication use in children.
4.2.7
Table 4-19 Controlled human exposure studies of respiratory symptoms.
Respiratory Hospital Admissions and Emergency Department Visits
Table 4-20 Mean and upper percentile concentrations of respiratory-related
Figure 4-4
Table 4-21
hospital admission and emergency department visit studies
published since the 2008 ISA for Oxides of Nitrogen._
Percent increase in respiratory-related hospital admissions for a 20
ppb increse in 24-h avg or 30-ppb increase in 1-h max NO2
concentrations from U. S. and Canadian studies evaluated in the
2008 ISA for Oxides of Nitrogen and recent studies.
Corresponding percent increase (95% Cl) for studies presented in
Figure 4-4.
_ 4-1
_ 4-3
_ 4-3
_ 4-4
_ 4-8
4-10
4-15
4-15
4-16
4-25
4-26
4-32
4-33
4-61
4-65
4-68
4-72
4-78
4-80
_ 4-82
_ 4-86
_ 4-87
4-108
4-116
4-118
4-119
4--720
4-124
4-125
4-142
4-146
4-148
4-163
4-164
November 201:
VII
DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Figure 4-5 Risk ratio and 95% CIs for associations between various lag 1 NO2
metrics and respiratory ED visits. 4-167
Figure 4-6 Spatial correlations for NO2 metrics in the Atlanta, GA area. 4-171
Figure 4-7 LOESS Concentration-Response estimates (solid line) and twice-
standard error estimates (dashed lines) from generalized additive
models for associations between 3-day avg (lag 0-2) NO2
concentrations and ED visits for pediatric asthma at the 5th to 95th
percentile of NO2 concentrations in the Atlanta, GA area. 4-173
Figure 4-8 Age-specific NO2 asthma ED visit effect estimates from copollutant
models with CO in Edmonton, Canada. 4-175
Figure 4-9 Percent increase in respiratory-related ED visits for a 20-ppb
increase in 24-h avg or 30-ppb increase in 1-h max NO2
concentrations from U. S. and Canadian studies evaluated in the
2008 ISA for Oxides of Nitrogen and recent studies in all-year and
seasonal analyses. 4-177
Table 4-22 Corresponding percent increase (95% Cl) for studies presented in
Figure 4-9. 4-177
4.2.8 Respiratory Mortality 4-180
4.2.9 Summary and Causal Determination 4-181
Figure 4-10 Associations of ambient NO2 or NO with respiratory outcomes
adjusted for PM'w, PM2,5, or PM 10-2.5- 4-187
Figure 4-11 Associations of ambient NO2 or A/Ox with respiratory outcomes
adjusted for elemental carbon (EC) or black carbon (BC), ultrafine
particles (UFP), or particle number concentration (PNC). 4-188
Table 4-23 Summary of evidence supporting a causal relationship between
short-term NO2 exposure and respiratory effects.
4.3 Cardiovascular Effects
4.3.1 Introduction
4.3.2 Arrhythmia and Cardiac Arrest
Table 4-24 Epidemioloqic studies of arrhythmia and cardiac arrest.
4.3.3 Heart Rate/Heart Rate Variability
Table 4-25 Epidemioloqic studies of heart rate/heart rate variability.
4.3.4 ST-Seqment Amplitude and QT-lnterval Duration
4-189
4-193
4-193
4-193
4-195
4-196
4-200
4-206
Table 4-26 Epidemiologic studies of ST-segment amplitude and QT-interval
duration. 4-208
4.3.5 Blood Pressure 4-209
Table 4-27 Epidemiologic studies of blood pressure. 4-211
4.3.6 Blood Biomarkers of Cardiovascular Effects 4-215
Table 4-28 Epidemiologic studies ofbiomarkers of cardiovascular effects. 4-218
Table 4-29 Controlled human exposure studies of short-term NO2 exposure and
cardiovascular effects. 4-227
Table 4-30 Animal toxicological studies of short-term NO2 exposure and
cardiovascular effects. 4-237
4.3.7 Hospital Admissions and Emergency Department Visits 4-232
Figure 4-12 Results of studies of short-term exposure to oxides of nitrogen and
hospital admissions for all cardiovascular disease. 4-233
Table 4-31 Corresponding effect estimates for hospital admissions for all
cardiovascular disease studies presented in Figure 4-12. 4-234
Figure 4-13 Results of studies of short-term exposure to oxides of nitrogen and
hospital admissions for cardiac disease. 4-238
Table 4-32 Corresponding risk estimates for hospital admissions for cardiac
disease for studies presented in Figure 4-13. 4-239
Figure 4-14 Results of studies of short-term exposure to oxides of nitrogen and
hospital admissions for cerebrovascular disease and stroke. 4-243
Table 4-33 Corresponding risk estimates for hospital admissions for
cerebrovascular disease and stroke for studies presented in
Figure 4-14. 4-244
4.3.8 Cardiovascular Mortality 4-246
4.3.9 Summary and Causal Determination 4-247
Figure 4-15 Results of single-pollutant and copollutants models of short-term
exposure to NO2 or A/Ox with and without PM and hospital
admissions for cardiovascular disease. 4-251
November 2013 viii DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Table 4-34 Corresponding risk estimates of ambient NO2 or NOX for hospital
admissions for cardiovascular disease in studies conducting
copollutant models with PM for presented in Figure 4-15. 4-252
Figure 4-16 Results of single-pollutant and copollutants models of short-term
exposure to NO2 (withCO [triangles] and without CO [circles]) and
hospital admissions for cardiovascular disease. 4-254
Table 4-35 Corresponding risk estimates of ambient NO2 for hospital
admissions for cardiovascular disease in studies conducting
copollutant models with CO presented in Figure 4-16. 4-255
Table 4-36 Summary of evidence supporting a likely to be a causal relationship
between short-term NO2 exposure and cardiovascular effects. 4-256
4.4 Total Mortality 4-258
4.4.1 Introduction and Summary of 2008 ISA for Oxides of Nitrogen 4-258
4.4.2 Associations between Short-term NO2 Exposure and Mortality 4-259
4.4.3 Associations between Short-term NC>2 Exposure and Mortality in All-Year Analyses 4-260
Table 4-37 Air quality characteristics of studies evaluated in the 2008 ISA for
Oxides of Nitrogen and recently published multicity and select
single-city studies. 4-267
Figure 4-17 Summary of multicity studies evaluated in the 2008 ISA for Oxides
of Nitrogen (black circles) and recently published (red circles) that
examined the association between short-term NO2 exposure and
total mortality. 4-263
Table 4-38 Corresponding percent increase in total mortality (95% Cl) for
Figure 4-17. 4-264
Figure 4-18 Percent increase in total, cardiovascular, and respiratory mortality
from multicity studies for a 20-ppb increase in 24-h avg or 30-ppb
increase in 1-h max NO2 concentrations. 4-265
Table 4-39 Corresponding percent increase (95% Cl) for Figure 4-18. 4-266
4.4.4 Potential Confounding of the NO2-Mortality Relationship 4-266
Table 4-40 Percent increase in total and cause-specific mortality for a 20-ppb
increase in 24-h avg NO2 concentrations in single- and copollutant
models with PMio in all-year analyses or 03 in summer season
analyses. 4-268
4.4.5 Modification of the NO2-Mortality Relationship 4-270
4.4.6 Potential Seasonal Differences in the NO2-Mortality Relationship 4-272
4.4.7 NO2-Mortality Concentration-Response (C-R) Relationship and Related Issues 4-274
Figure 4-19 Percent increase in total and cause-specific mortality due to short-
term NO2 exposure at single day lags, individual lag days of a
constrained polynomial distributed lag model, and multi-day lags of
an unconstrained distributed lag model. 4-275
Figure 4-20 Percent increase in total and cause-specific mortality due to short-
term NO2 exposure in single- and multi-day lag models. 4-276
Figure 4-21 Flexible ambient C-R relationship between short-term NO2 (ppb)
exposure and mortality at lag day 1. Pointwise means and 95% CIs
adjusted for size of the bootstrap sample. 4-278
Figure 4-22 CAPES C-R curve for the association between total and cause-
specific mortality and 24-h avg NO2 concentrations at lag 0-1 days.
NO2 concentrations on the x-axis are in the unit of/jg/m . 4-279
Figure 4-23 C-R curve for association between total mortality and 24-h avg NO2
concentrations at lag 0-1 days in the four cities of the PAPA study. 4-287
4.4.8 Summary and Causal Determination 4-281
Table 4-41 Summary of evidence supporting a likely to be a causal relationship
between short-term NO2 exposure and total mortality. 4-285
References for Chapter 4 4-286
CHAPTER 5 INTEGRATED HEALTH EFFECTS OF LONG-TERM EXPOSURE TO
OXIDES OF NITROGEN 5-1
5.7 Introduction 5-7
5.2 Respiratory Effects 5-3
5.2.1 Introduction 5-3
5.2.2 Asthma/Chronic Bronchitis Incidence 5-4
November 2013 ix DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Figure 5-1 Distribution of NO2 exposure in first year of life in GALA II/SAGE. 5-6
Figure 5-2 Adjusted overall and age-specific association between annual
average levels of air pollution at the birth address and asthma
during the first 8 years of life. 5-9
Table 5-1 Prospective long-term NO2 exposure new onset asthma in children
cohort studies. 5-12
Figure 5-3 Risk ratio estimates of asthma incidence from prospective studies. 5-19
Figure 5-4 Adjusted ratios of new asthma in ECRHS II (1999-2001) for every
10 pg/m3 (5.3 ppb) NO2 increase by center in subjects with no
asthma in ECRHS I (1991-1993). 5-21
5.2.3 Pulmonary Function 5-24
Table 5-2 NO2 long-term pulmonary function growth children cohort
prospective studies. 5-27
Figure 5-5 Associations of NO2 or NOX with lung function indices from
prospective studies. 5-33
5.2.4 Hospital Admissions 5-36
5.2.5 Respiratory Symptoms 5-38
Table 5-3 Long-term NO2 exposure prospective children studies: respiratory
symptoms. 5-41
5.2.6 Allergic sensitization 5-46
5.2.7 Pulmonary Inflammation and Oxidative Stress 5-49
Table 5-4 Animal Toxicological Studies of the Respiratory Effects of Long-term
NO2 Exposure. 5-52
5.2.8 Toxicological Studies of Airway Hyperresponsiveness 5-55
5.2.9 Toxicological Studies of Host Defense 5-55
5.2.10 Toxicological Studies of Respiratory Morphology 5-57
5.2.11 Gene-Environment Interactions 5-58
5.2.12 Concentration-Response 5-60
Figure 5-6 Concentration-response relationships between health outcome and
NO2 (log concentration as a continuous variable) illustrated with
constrained, natural spline functions (solid lines) with 95%
confidence limits (small dashed lines) and threshold function (bold
dashed line) from fully adjusted, hierarchical ordered logistic
regression models. 5-61
Figure 5-7 Community-specific average growth in FEVi (mL) among girls and
boys during the eight-year period from 1993 to 2001, plotted against
average NO2 levels from 1994 through 2000. 5-62
Figure 5-8 Community-specific proportion of 18-year-olds with a FEVi below
80 percent of the predicted value, plotted against the average levels
ofNO2 from 1994 through 2000. 5-63
5.2.13 Analysis of copollutants 5-63
Table 5-5 Studies that provide evidence for NO2 and also provide analysis of
copollutants (PM, O3, SO2, CO). 5-64
5.2.14 Mixtures: Traffic-related Pollutants 5-67
Table 5-6 Studies reporting NO2 results and results for traffic measures. 5-69
5.2.15 Indoor Studies 5-70
5.2.16 Surrogate for ambient NC>2 or other pollutants as a mixture 5-70
5.2.17 Summary and Causal Determination 5-73
Table 5-7 Annual ambient NO2 concentrations in prospective studies
examining relationships in children with respiratory health effects in
children. 5-77
Table 5-8 Annual ambient NO2 concentrations prospective studies examining
relationships with respiratory health effects in adults. 5-79
Table 5-9 Summary of evidence supporting a likely to be a causal relationship
between long-term NO2 exposure and respiratory effects. 5-80
5.3 Cardiovascular Effects 5-83
5.3.1 Introduction 5-83
5.3.2 Cardiovascular Diseases 5-83
Table 5-10 Epidemiologic studies of long-term exposure to NO2 or A/Ox and
effects on the cardiovascular system. 5-86
5.3.3 Markers of Cardiovascular Disease Risk 5-87
5.3.4 Inflammation and Oxidative Stress 5-88
November 2013 x DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
Table 5-11 Study details for toxicological studies examining cardiovascular
effects from long-term NO2 exposure 5-90
5.3.5 Cardiovascular Mortality 5-91
5.3.6 Summary and Causal Determination 5-91
Table 5-12 Summary of evidence supporting a suggestive relationship between
long-term NC>2 exposure and cardiovascular effects. 5-93
5.4 Reproductive and Developmental Effects 5-94
5.4.1 Introduction 5-94
5.4.2 Fertility, Reproduction, and Pregnancy 5-96
5.4.3 Birth Outcomes 5-100
5.4.4 Postnatal Development 5-107
Table 5-13 Key reproductive and developmental epidemiologic studies for NO2. 5-117
Table 5-14 Reproductive and developmental toxicological studies for NC>2. 5-123
5.4.5 Summary and Causal Determination 5-123
Table 5-15 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and reproductive and
developmental effects. 5-126
5.5 Mortality. 5-728
5.5.1 Review of Mortality Evidence from 2008 ISA for Oxides of Nitrogen 5-129
5.5.2 Recent Evidence for Mortality from Long-term Exposure to Oxides of Nitrogen 5-131
Figure 5-9 Results of studies of long-term exposure to NO2 or NOX and all-
cause mortality. 5-134
Table 5-16 Corresponding risk estimates for Figure 5-9. 5-735
Figure 5-10 Results of studies of long-term exposure to NO2, NO, or NOX and
cardiovascular mortality. 5-736
Table 5-17 Corresponding risk estimates for Figure 5-10. 5-737
Figure 5-11 Results of studies of long-term exposure to NO2 or A/Ox and
respiratory mortality. 5-739
Table 5-18 Corresponding risk estimates for Figure 5-11. 5-740
5.5.3 Summary and Causal Determination 5-141
Table 5-19 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and total mortality. 5-743
5.6 Cancer 5-144
5.6.1 Lunq Cancer Incidence
5.6.2 Lunq Cancer Mortality
5.6.3 Leukemia Incidence and Mortality
5.6.4 Bladder Cancer Mortality
5.6.5 Breast Cancer Incidence
5.6.6 Prostate Cancer Incidence
5.6.7 Animal and In Vitro Carcinoqenicity and Genotoxicity Studies
5.6.8 Animal Toxicoloqy Studies of Co-exposure with Known Carcinoqens
5.6.9 Studies in Animals with Spontaneous Hiqh Tumor Rates
5.6.10 Facilitation of Metastases
5.6. 1 1 Production of N-Nitroso Compounds and other Nitro Derivatives
5-144
5-145
5-147
5-148
5-148
5-149
5-149
5-150
5-151
5-151
5-151
Table 5-20 Animal toxicological studies of carcinogenicity and genotoxicity with
exposure to NO2. 5-753
5.6.12 Summary and Causal Determination 5-155
Table 5-21 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and cancer. 5-756
References for Chapter 5 5-758
CHAPTER 6 POPULATIONS POTENTIALLY AT INCREASED RISK FOR HEALTH
EFFECTS RELATED TO EXPOSURE TO OXIDES OF NITROGEN 6-1
6.7 Introduction 6-7
Table 6-1 Classification of Evidence for Potential At-Risk Factors. 6-3
6.2 Genetic Factors 6-3
Table 6-2 Summary of epidemiologic studies evaluating effect modification by
genetic variants. 6-6
6.3 Pre-existing Disease/Conditions 6-7
Table 6-3 Prevalence of respiratory diseases, cardiovascular diseases, and
diabetes among adults by age and region in the U.S. in 2010. 6-8
November 2013 xi DRAFT: Do Not Cite or Quote
-------
CONTENTS (Continued)
6.3.1 Asthma 6-8
6.3.2 Chronic Obstructive Pulmonary Disease (COPD) 6-11
6.3.3 Cardiovascular Disease (CVD) 6-12
6.3.4 Diabetes 6-13
6.4 Sociodemographic Factors 6-14
6.4.1 Lifestage 6-14
6.4.2 Socioeconomic Status 6-18
6.4.3 Race/Ethnicity 6-22
6.4.4 Sex 6-23
6.5 Behavioral and Other Factors 6-24
6.5.1 Diet 6-24
6.5.2 Obesity 6-25
6.5.3 Smoking 6-26
6.5.4 Residential Location 6-27
6.6 Summary 6-28
Table 6-4 Summary of evidence for potential increased risk of health effects
related to exposure to oxides of nitrogen. 6-29
References for Chapter 6 6-31
November 2013 xii DRAFT: Do Not Cite or Quote
-------
ISA TEAM FOR OXIDES OF NITROGEN
Executive Direction
Dr. John Vandenberg (Director)—National Center for Environmental Assessment-RTF
Division, Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC
Dr. Reeder Sams II (Acting Deputy Director)—National Center for Environmental
Assessment-RTP Division, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Mary Ross (Branch Chief)—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr. Steven Dutton (Acting Branch Chief)—National Center for Environmental Assessment,
Office of Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, NC
Scientific Staff
Dr. Molini M. Patel (Team Leader, ISA for Oxides of Nitrogen)—National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. James Brown—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Barbara Buckley—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Erin Hines—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Ellen Kirrane—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Thomas Luben—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Stephen McDow—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Ms. Connie Meacham—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Jennifer Nichols—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
November 2013 xiii DRAFT: Do Not Cite or Quote
-------
Dr. Michelle Oakes—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Elizabeth Oesterling Owens—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr Joseph P. Pinto—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Jennifer Richmond-Bryant—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Mr. Jason Sacks—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Tina Stevens—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. David Svendsgaard—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr. Lisa Vinikoor-Imler—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Ms. Brianna Young—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Technical Support Staff
Ms. Marieka Boyd—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Mr. Kenneth J. Breito—Senior Environmental Employment Program, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Mr. Nathan Ellenfield—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Mr. Gerald Gurevich—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Ms. Katie Jelen—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Mr. Ryan Jones—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
November 2013 xiv DRAFT: Do Not Cite or Quote
-------
Ms. Diane LeBlond—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Ms. Danielle Moore—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Ms. Candis O'Neal—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Ms. Sandy Pham—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Mr. Richard N. Wilson—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Ms. Barbara Wright—Senior Environmental Employment Program, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
November 2013 xv DRAFT: Do Not Cite or Quote
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Authors
Dr. Molini M. Patel (Team Leader, ISA for Oxides of Nitrogen)—National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. James Brown—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Barbara Buckley—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Rachelle Duvall—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Erin Hines—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Ellen Kirrane—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Thomas Luben—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Stephen McDow—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Jennifer Nichols—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Michelle Oakes—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Elizabeth Oesterling Owens—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr. Jennifer Peel—Department of Environmental and Radiological Health Sciences,
Colorado School of Public Health, Colorado State University, Fort Collins, CO
Dr Joseph P. Pinto—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Edward Postlethwait—Department of Environmental Health Sciences, Ryals School for
Public Health, University of Alabama at Birmingham, Birmingham, AL
Dr. Jennifer Richmond-Bryant—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
November 2013 xvi DRAFT: Do Not Cite or Quote
-------
Mr. Jason Sacks—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Tina Stevens—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Giuseppe Squadrito—Department of Environmental Health Sciences, Ryals School for
Public Health, University of Alabama at Birmingham, Birmingham, AL
Dr. David Svendsgaard—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr. George Thurston—Department of Environmental Medicine, New York University
School of Medicine, Tuxedo, NY
Dr. Lisa Vinikoor-Imler—National Center for Environmental Assessment, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Dr. Gregory Wellenius—Department of Community Health (Epidemiology Section), Brown
University, Providence, RI
Ms. Brianna Young—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Contributors
Mr. Evan Coffman—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Ms. Laura Datko—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Ms. Meagan Madden—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
Dr. Lok Namsal—Universities Space Research Association, NASA Goddard Space Flight
Center, Greenbelt, MD
Dr. Havala Pye—National Exposure Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC
Ms. Adrien Wilkie—Oak Ridge Institute for Science and Education, National Center for
Environmental Assessment, Office of Research and Development, U.S. Environmental
Protection Agency, Research Triangle Park, NC
November 2013 xvii DRAFT: Do Not Cite or Quote
-------
Reviewers
Dr. Lisa Baxter—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Melinda Beaver—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Mr. Roger Erode—Office of Air Quality Planning and Standards, Office of Air and
Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Philip Bromberg—School of Medicine, University of North Carolina, Chapel Hill, NC
Dr. Steven Dutton—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Mark Frampton—Pulmonary and Critical Care Division, Department of Medicine,
University of Rochester School of Medicine, Rochester, NY
Dr. Terry Gordon—Department of Environmental Medicine, New York University School
of Medicine, Tuxedo, NY
Dr. Gary Hatch—National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Ms. Beth Hassett-Sipple—Office of Air Quality Planning and Standards, Office of Air and
Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
Mr. Marc Houyoux—Office of Air Quality Planning and Standards, Office of Air and
Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Thomas Long—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Deborah Luecken—National Exposure Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Ms. Connie Meacham—National Center for Environmental Assessment, Office of Research
and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Mr. David Orlin—Air and Radiation Law Office, Office of General Counsel, U.S.
Environmental Protection Agency, Washington, DC
Dr. Jennifer Peel—Department of Environmental and Radiological Health Sciences,
Colorado School of Public Health, Colorado State University, Fort Collins, CO
Dr. Edward Postlethwait—Department of Environmental Health Sciences, Ryals School for
Public Health, University of Alabama at Birmingham, Birmingham, AL
Dr. Havala Pye—National Exposure Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC
Ms. Elise Richman—Office of Children's Health Protection, U.S. Environmental Protection
Agency, Washington, DC
Dr. Mary Ross—National Center for Environmental Assessment, Office of Research and
Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
November 2013 xviii DRAFT: Do Not Cite or Quote
-------
Dr. Reader Sams II—National Center for Environmental Assessment-RTF Division, Office
of Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, NC
Dr. Stephanie Sarnat—Department of Environmental Health, Rollins School of Public
Health, Emory University, Atlanta, GA
Dr. Matthew Strickland—Department of Environmental Health, Rollins School of Public
Health, Emory University, Atlanta, GA
Dr. John Vandenberg—National Center for Environmental Assessment-RTP Division,
Office of Research and Development, U.S. Environmental Protection Agency, Research
Triangle Park, NC
Ms. Debra Walsh—National Center for Environmental Assessment-RTP Division, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC
Mr. Nealson Watkins—Office of Air Quality Planning and Standards, Office of Air and
Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
Dr. Margaret Zawacki—Office of Transportation and Air Quality, Office of Air and
Radiation, U.S. Environmental Protection Agency, Ann Arbor, MI
November 2013 xix DRAFT: Do Not Cite or Quote
-------
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE
OXIDES OF NITROGEN NAAQS REVIEW PANEL
Chair of the Oxides of Nitrogen Review Panel
Dr. H. Christopher Frey**, North Carolina State University, Raleigh, NC
Oxides of Nitrogen Review Panel Members
Mr. George A. Allen*, Northeast States for Coordinated Air Use Management
(NESCAUM), Boston, MA
Dr. Matthew Campen, University of New Mexico, Albuquerque, NM
Dr. Ronald Cohen, University of California, Berkeley, Berkeley, CA
Dr. Douglas Dockery, Harvard University, Boston, MA
Dr. Philip Fine, South Coast Air Quality Management District, Diamond Bar, CA
Dr. Panos Georgopoulos, UMDNJ-Robert Wood Johnson Medical School, Piscataway, NJ
Dr. Jack Harkema*, Michigan State University, East Lansing, MI
Dr. Michael Jerrett, University of California, Berkeley, Berkeley, CA
Dr. Joel Kaufman, University of Washington, Seattle, WA
Dr. Patrick Kinney, Columbia University, New York, NY
Dr. Michael T. Kleinman, University of California, Irvine, Irvine, CA
Dr. Timothy V. Larson, University of Washington, Seattle, WA
Dr. Jeremy Sarnat, Emory University, Atlanta, GA
Dr. Richard Schlesinger, Pace University, New York, NY
Dr. Elizabeth A. (Lianne) Sheppard, University of Washington, Seattle, WA
Dr. Helen Suh*, Northeastern University, Boston, MA
Dr. Ronald Wyzga*, Electric Power Research Institute, Palo Alto, CA
Dr. Junfeng (Jim) Zhang, Duke University, Durham, NC
* Members of the statutory Clean Air Scientific Advisory Committee (CASAC) appointed
by the EPA Administrator
** Chair of the statutory Clean Air Scientific Advisory Committee (CASAC) appointed by
the EPA Administrator
Science Advisory Board Staff
Mr. Aaron Yeow, Designated Federal Officer, U.S. Environmental Protection Agency,
Science Advisory Board (1400R), 1200 Pennsylvania Avenue, NW, Washington, D.C.
20460-0001, Phone: 202-564-2050, Fax: 202-565-2098, (yeow.aaron@,epa.gov)
November 2013 xx DRAFT: Do Not Cite or Quote
-------
ACRONYMS AND ABBREVIATIONS
Acronym/Abbreviation Meaning
a-ATD
AADT
Abs
ABTS
ACS
ADHD
ADRB2
AERMOD
AHR
AIRES
AK
AKR/J
AL
ALRI
a.m.
AM
AM3
AMF
AMs
APEX
APHEA
APHEA-2
alpha, single term defined to
express the influence of time-
weighting and infiltration on
NO2 exposure
alpha 1 -antitrypsin deficiency
annual average daily traffic
absorbance coefficient
2,2'-azino-bis(3-
ethylbenzothiazoline-6-
sulphonic acid)
American Cancer Society
attention deficit hyperactivity
disorder
beta-2-adrenergic receptor
American Meteorological
Society/Environmental
Protection Agency Regulatory
Model
airway hyperresponsiveness
air exchange rate
Atlanta Aerosol Research
Inhalation Epidemiology
Study
Alaska
mice strain with short life-
span; often used as model for
aging
Alabama
acute lower respiratory
infection
ante meridiem (before noon)
alveolar macrophages
global scale, three-
dimensional chemical tracer
model
air mass factor
alveolar macrophages
Air Pollution Exposure
Air Pollution and Health: a
European Approach
second, more recent APHEA
study with more cities
Acronym/Abbreviation Meaning
APHENA
AR
AQCD
AQI
AQM
ATS
avg
AZ
P
BAL
BALF
BC
BHPN
BL
BMI
BP
BS
BSA
BTEX
BW
C
Ca
CA
Ca
c
^ a,csm
CAA
CALINE4
Air Pollution and Health: A
European and North American
Approach study
apolipoprotein E knockout
Arkansas
Air Quality Criteria
Document
Air Quality Index
air quality model
American Thoracic Society
average
Arizona
beta
bronchoalveolar lavage
bronchoalveolar lavage fluid
black carbon
N-bis (2-hydroxy-propyl)
nitrosamine
bronchial lavage
body mass index
blood pressure
bromide
black smoke
body surface area
benzene, toluene,
ethybenzene, xylene: traffic
related VOCs
body weight
Celsius, microenvironmental
concentration
calcium
California
ambient concentration
ambient concentration at a
central site monitor
Clean Air Act
California Department of
Transportation's most recent
line dispersion model
November 201:
xxi
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation Meaning
CALPUFF
CAMP
CAPES
CAPs
CAPS
CASAC
CASNET
CBSA
CC16
CDC
CDPFs
GEMS
CFD
CHAD
CHD
CHF
CHIMERE
CI(s)
Cf
CL/MC
CL/PC
C1NO2
CMAQ
CMSA
CO
COPD
C-R
Non-steady-state
meterological and air quality
modeling system developed
by the Atmospheric Studies
Group at TRC
Childhood Asthma
Management Program
China Air Pollution and
Health Effects Study
concentrated ambient particles
cavity attenuated phase shift
Clean Air Scientific Advisory
Committee
Clean Air Status and Trends
Network
Core Based Statistical Area
Clara Cell secretory protein
Centers for Disease Control
catalyzed diesel particle filters
Continuous Emission
Monitoring System
computational fluid dynamics
Consolidated Human Activity
Database
coronary heart disease
congestive heart failure
regional chemistry transport
model
average NO2 concentration in
the rth microenvironment
confidence interval(s)
chloride
catalytic converter
photolytic converter
nitryl chloride
Community Multiscale Air
Quality
Consolidated Metropolitan
Statistical Area
carbon monoxide; Colorado
chronic obstructive
pulmonary disease
concentration-response
(relationship)
Acronym/Abbreviation Meaning
CRDS
CRP
CT
CTM
CVD
Cys'
DBF
DC
D-dimer
DE
DEARS
DEP
df
DHA
DL
DLM
DOAS
DOCs
e.g.
Ea
EEC
EC
EGG
ECP
ED
EGR
ELF
eNO
diode laser based cavity ring
down spectroscopy
C-reactive protein
Connecticut
chemical transport models
cardiovascular disease
cysteine radical
diastolic blood pressure
District of Columbia
blood indicator of thrombosis
Deleware
Detroit Exposure and Aerosol
Research Study
diesel exhaust particles
degrees of freedom
dehydroascorbate
distributed lag
polynomial distributed lag
model
differential optical absorption
spectroscopy
diesel oxidation catalysts
exempli gratia; for example
ambient NO2 exposure
average exposure to ambient
NO 2
expired (exhaled) breath
condensate
elemental carbon
electrocardiography
eosinophil cationic protein
emergency department
exhaust gas recirculation
indoor NO2 exposures in the
rth microenvironment
epithelial lining fluid,
extracellular lining fluid
non-ambient NO 2 exposure
exhaled nitric oxide,
endogenous nitric oxide
November 201:
xxu
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation Meaning
eNOS
Eo
EPA
E-selectin
ESR
ET
ET-1
EXPOLIS
FA
factor VII
FEF25%
FEF25.75./0
FEF50%
FEM
FeNO
FEVj
FL
FRM
FVC
y
g
g/bhp-h
GA
GAM
GCLC
GCLM
GEE
GIS
endothelial nitric oxide
synthase
outdoor microenvironmental
NO2 exposures
U.S. Environmental
Protection Agency
indicator of inflammation
erythrocyte sedimentation rate
total personal exposure
vasoconstrictor endothelin-1
exposure in polis or cities
filtered air
enzyme in the coagulation
cascade
forced expiratory flow at 25%
of forced vital capacity
forced expiratory flow at
25-75% of exhaled volume
forced expiratory flow at 50%
of forced vital capacity
Federal Equivalent Method
fractional exhaled nitric oxide
forced expiratory volume in
1 second
Florida
Federal Reference Method
forced vital capacity
gamma; uptake coefficients
gram
grams per brake horsepower-
hour
Georgia
generalized additive models
gene that encodes the catalytic
subunit for the human enzyme
glutamate-cysteine ligase
gene that encodes the
regulatory subunit for the
human enzyme glutamate-
cysteine ligase
generalized estimating
equations
geographic information
systems
Acronym/Abbreviation
GLM
GLMM
Meaning
generalized linear model
generalized linear mixed
model
GPx
GS*
GSH
GSR
GSS
GST
GSTM1
GSTP1
h
hCAEC
HDL
HDM
H&E staining
HECT
HEI
HERO
HEV
HF
HI
HNO2
HNO3
HNO4
HO-1
H02
H02N02
HONO
HOONO
HR
HRV
HSC
H2SO4
glutathione peroxidase
glutathione radical
glutathione
glutathione reductase
glutathione synthetase
glutathione S-transferase
glutathione s-transferase Mu 1
glutathione s-transferase P
hour, hours
human coronary artery
endothelial cell
high-density lipoprotein
house dust mite
hematoxylin and eosin stain
for histology analysis
hand eye coordination test
Health Effects Institute
Health and Environmental
Research Online
hold-out evaluation
high frequency component of
HRV
Hawai'i
nitrous acid
nitric acid
peroxynitric acid
heme oxidase-1, heme
oxygenase-1
hydroperoxyl radical,
perhydroxyl radical
peroxynitric acid (PNA)
nitrous acid
pernitrous acid, peroxynitrous
acid
hazard ratio(s)
heart rate variability
Harvard Six Cities
sulfuric acid
November 201:
XXlll
DRAFT: Do Not Cite or Quote
-------
Acronym/ Abbreviation
IA
i.e.
ICARTT
ICAS
ICD
ICR
ICS
ID
IDW
Ig
IgA
IgE
IgG
IgM
IHD
IL
IL-1
IL-6
IL-8
He
IMSI
IN
INDAIR
INF
iNOS
INs
IQR
ISA
ISC3
IUGR
IVF
k
kg
Meaning
Iowa
id est; that is
International Consortium for
atmospheric research on
Transport and Transformation
Inner-City Asthma Study
International Classification of
Diseases
mice strain
inhaled corticosteroids
Idaho
inverse distance weighting
immunoglobulin
immunoglobulin A
immunoglobulin E
immunoglobulin G
immunoglobulin M
ischemic heart disease
interleukin; Illinois
interleukin- 1
interleukin-6
interleukin- 8
isoleucine
Integrated Mobile Source
Indicator
Indiana
probabilistic model for indoor
pollution exposures
infiltration of outdoor NO2
inducible nitric oxide
synthase
isoprene nitrates
interquartile range
Integrated Science
Assessment
Industrial Source Complex
dispersion model
intrauterine growth restriction
in-vitro fertilization
reaction rate
kilogram
Acronym/ Abbreviation
k,
km
KS
KY
LA
LEW
LDH
LF
LF/HF
LIF
LOOCV
LOPAP
LOX-1
Lp-PLA2
LRI
LRS
LRTI
LT
LTO
LUR
ug/m3
m
MA
max
MCP-1
MD
MDA
ME
MESA
MI
MI
min
Meaning
decay rate
kilometer(s)
Kansas
Kentucky
Louisiana; Los Angeles
low birth weight
lactate dehydrogenase
low-frequency component of
HRV
ratio of LF and HF
components of HRV
laser induced fluorescence
leave one out cross-validation
long path absorption
photometer
lectin-like oxLDL receptor
lipoprotein-associated
phospholipase A2
lower respiratory infection
lower respiratory symptoms
lower respiratory tract
infection
leukotrienes
landing and take-off cycles
land use regression
mu; micro
micrograms per cubic meter
meter
Massachusetts
maximum
monocyte chemoattractant
protein-1
Maryland
malondialdehyde
Maine
Multi-Ethnic Study of
Atherosclerosis
myocardial infarction, "heart
attack"; myocardial ischemia
Michigan
minimum
November 201:
xxiv
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation
MLI
MM5
MMEF
MMP
MMP-2
MMP-9
MN
mo
MO
MoOx
MPO
mRNA
ms; msec
MS
MSA
MSCA
MT
n
N
NAAQS
NAB
NaCl
NADPH
NAL
NC
NCEA
NCICAS
NCORE
ND
NDMA
NE
NES
Meaning
mean linear intercept
meteorological mesoscale
model
maximum (or maximal)
midexpiratory flow
matrix metalloproteinase
matrix metalloproteinase-2
matrix metalloproteinase-9
Minnesota
month, months
Missouri
molybdenum oxide
myeloperoxidase
messenger ribonucleic acid
millisecond
Mississippi
Metropolitan Statistical Area
McCarthy Scales of
Children's Abilities
Montana
sample size; total number of
microenvironments
nitrogen; population number
National Ambient Air Quality
Standards
North American Background
sodium chloride
reduced nicotinamide adenine
dinucleotide phosphate
nasal lavage
North Carolina
National Center for
Environmental Assessment
National Cooperative Inner-
City Asthma Study
National Core network
North Dakota
N-nitrosodimethylamine
Nebraska
Neurobehavioral Evaluation
System
Acronym/ Abbreviation
NH
NH3
(NH4)2SO4
NHAPS
NJ
nm
NM
NMMAPS
NMOR
NO
NO 2
N2O5
NOS
NOX
NOY
NOZ
NQO1
NR
n.s.
NSR
NV
NY
NYC
-------
Acronym/Abbreviation
OK
OMI
ONOCf
ONOOCO2~
OR
ORD
OTAG
P
PA
PAH(s)
PAN
PAPA
Pb
PEL
PC
PCA
PCO
PCX
PD
PDX
PEACE
PEF
PFK
P,
PK
p.m.
PM
Meaning
Oklahoma
Ozone Monitoring Instrument
peroxy nitrite
nitrosoperoxylcarbonate anion
odds ratio(s); Oregon
Office of Research and
Development
Ozone Transport Assessment
Group
probability
Pennsylvania
poly cyclic aromatic
hydrocarbon(s)
peroxyacetyl nitrate
Public Health and Air
Pollution in Asia
lead
planetary boundary layer
provocative concentration
principal component analysis
protein carbonyl
provocative concentration
required to reduce/increase an
effect by X%
provocative dose
provocative dose required to
reduce/increase an effect by
X%
Pollution Effects on
Asthmatic Children in Europe
peak expiratory flow
phosphofructokinase
air pollutant penetration
pyruvate kinase
post meridiem (after noon)
particulate matter
Acronym/Abbreviation Meaning
PM,
PM,
In general terms, particulate
matter with a nominal
aerodynamic diameter less
than or equal to 10 um; a
measurement of thoracic
particles (i.e., that subset of
inhalable particles thought
small enough to penetrate
beyond the larynx into the
thoracic region of the
respiratory tract) in regulatory
terms, particles with an upper
50% cut-point of 10 ± 0.5 um
aerodynamic diameter (the
50% cut point diameter is the
diameter at which the sampler
collects 50% of the particles
and rejects 50% of the
particles) and a penetration
curve as measured by a
reference method based on
Appendix J of 40 CFR Part 50
and designated in accordance
with 40 CFR Part 53 or by an
equivalent method designated
in accordance with 40 CFR
Part 53.
In general terms, particulate
matter with a nominal
aerodynamic diameter less
than or equal to 10 um and
greater than a nominal
2.5 um; a measurement of
thoracic coarse particulate
matter or the coarse fraction
of PMjo in regulatory terms,
particles with an upper 50%
cut-point of 10 um
aerodynamic diameter and a
lower 50% cut-point of
2.5 um aerodynamic diameter
(the 50% cut point diameter is
the diameter at which the
sampler collects 50% of the
particles and rejects 50% of
the particles) as measured by
a reference method based on
Appendix O of 40 CFR Part
50 and designated in
accordance with 40 CFR Part
53 or by an equivalent method
designated in accordance with
40 CFR Part 53.
November 201:
xxvi
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation Meaning
Acronym/Abbreviation Meaning
PMA
PMF
PMN(s)
PNC
PNN50
pNO3
ppb
ppm
PPN
ppt
PSDs
P-selectin
PTB
QC-TILDAS
QT interval
P
In general terms, participate
matter with a nominal
aerodynamic diameter less
than or equal to 2.5 um; a
measurement of fine particles
in regulatory terms, particles
with an upper 50% cut-point
of 2.5 um aerodynamic
diameter (the 50% cut point
diameter is the diameter at
which the sampler collects
50% of the particles and
rejects 50% of the particles)
and a penetration curve as
measured by a reference
method based on Appendix L
of 40 CFR Part 50 and
designated in accordance with
40 CFR Part 53, by an
equivalent method designated
in accordance with 40 CFR
Part 53, or by an approved
regional method designated in
accordance with Appendix C
of 40 CFR Part 58.
phorbol myristate acetate
positive matrix factorization
polymorphonuclear cell(s),
polymorphonuclear leukocyte
particle number concentration
proportion of successive NN
intervals with difference
>50 msec (NN50) out of the
total number of NN intervals
particulate nitrate
parts per billion
parts per million
peroxypropionyl nitrate
parts per trillion
passive sampling devices
platelet selectin, a marker of
platelet activation
preterm birth
quantum cascade - tunable
infrared laser differential
absorption spectrometer
time between start of Q wave
and end of T wave in EGG
rho, Spearman correlation
coefficient
RANTES
RBC
RC(=0)OON02
REA
RI
RMS
rMSSD
RNS
RONO2
ROO*
ROS
RR
RSNO
s
S/N
SA-LUR
SAT
SBL
SBP
SC
sCD62P
SCR
SD
SDNN
SEARCH
sec
SEER
Pearson correlation
coefficient; Spearman
correlation coefficient
regulated on activation,
normal T cell expressed and
secreted (aka chemokine
ligand 5, CCL5)
red blood cells
peroxynitrates, peroxyacetyl
nitrates
Risk and Exposure
Assessment
Rhode Island
ratios of the mean asthma
scores
root mean square of
successive differences; a
measure of HRV
reactive nitrogen species
organic nitrates
organic peroxyl radical
reactive oxygen species
risk ratio(s), relative risk
S-nitrosothiols
second(s)
signal to noise ratio
Source-Area land use
regression
Switching Attention Test
stable boundary layer
systolic blood pressure
South Carolina
platelet activation biomarker
selective catalytic reduction
standard deviation; South
Dakota
standard deviation of beat-to-
beat (NN) intervals, an index
of total HRV
Southeast Aerosol Research
Characterization
second(s)
Surveillance, Epidemiology,
and End Results
November 201:
xxvii
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation
SES
SGA
SHARP
SHEDS
SHS
sIC AM-1
SLAMS
SMWAOs
SNCR
SNP
SNR
S02
SOA
SOD
SPM
Sp02
sRaw
SRTT
s-TNFa-RII
ST-segment
sVCAM-1
TEARS
TC
TEA
Th2
t,
Meaning
socioeconomic status
small for gestational age
Study of Houston
Atmospheric Radical
Precursors
Stochastic Human Exposure
and Dose Simulation
secondhand smoke
soluble intercellular adhesion
molecule-1
State and Local Air
Monitoring Stations
small molecular weight
antioxidants
selective non-catalytic
reduction
single nucleotide
polymorphism
selective NOX recirculation
sulfur dioxide
secondary organic aerosols
superoxide dismutase
suspended PM, suspended
particulate matter
blood oxygen saturation
specific airway resistance
Simple Reaction Time Test
soluble tumor necrosis factor
a receptor II
measured from the J point to
the end of the T wave in an
EGG
soluble vascular adhesion
molecule-1
fraction of time spent in a
microenvironment, time
thiobabituric acid reactive
substances (species)
total carbon
triethanolamine
T-derived lymphocyte helper
2
fraction of total time spent in
the ith microenvironment
Acronym/Abbreviation Meaning
TIMP-1
TIMP-2
TLR2
TLR4
TN
TNF
TP
TRP
TX
UFP
UH2-
UK
U.K.
ULTRA
URI
U.S.; USA
use
U.S. EPA
UT
VA
Val
VE
VOCs
VPTB
VT
vWF
WHO
WI
wk
WRE-chem
WV
tissue inhibitor of matrix
metalloproteinase-1
tissue inhibitor of matrix
metalloproteinase-2
Toll-like receptor 2
toll-like receptor 4
Tennessee
tumor necrosis factor
total power of heart rate signal
in an EGG
traffic-related pollution
Texas
ultrafine particles
urate
universal kriging
United Kingdom
The Exposure and Risk
Assessment for Fine and
Ultrafine Particles in Ambient
Air Study conducted in
Europe
upper respiratory infection
United States of America
U.S. Code
U.S. Environmental
Protection Agency
Utah
Virginia
valine
minute volume
volatile organic compounds
very preterm birth
Vermont
von Willbrand factor
World Health Organization
Wisconsin
week, weeks
Weather Research and
Forecast model with
chemistry
West Virginia
November 201:
XXVlll
DRAFT: Do Not Cite or Quote
-------
Acronym/Abbreviation Meaning
WY Wyoming
Yi time spent indoors
y0 fraction of the day spent
outdoors
yr year(s)
November 2013 xxix DRAFT: Do Not Cite or Quote
-------
PREAMBLE
1. Process of ISA Development
1 This preamble outlines the general process for developing an Integrated Science
2 Assessment (ISA) including the framework for evaluating weight of evidence and
3 drawing scientific conclusions and causal judgments. The ISA provides a concise review,
4 synthesis, and evaluation of the most policy-relevant science to serve as a scientific
5 foundation for the review of the National Ambient Air Quality Standards (NAAQS)1 for
6 the criteria air pollutants (i.e., carbon monoxide [CO], lead [Pb], nitrogen oxides, ozone
7 [O3], particulate matter [PM] and sulfur oxides) as defined by the Clean Air Act (CAA.
8 1990a. b). Figure I depicts the general NAAQS review process, and information for
9 individual NAAQS reviews is available online2.
10 The ISA is preceded by the release of an Integrated Review Plan (IRP) that discusses the
11 planned scope and organization of the key NAAQS assessment documents (e.g., ISA),
12 including policy-relevant questions, approaches for preparing documents, and the
13 schedule for release and review of the documents. The policy-relevant questions included
14 in the IRP serve to clarify and focus the NAAQS review on the critical scientific and
15 policy issues, including uncertainties discussed during the previous review and newly
16 emerging literature. The IRP is informed by an EPA hosted public "kick-off workshop"
17 that seeks input on the current state of the science and engages stakeholders and experts
18 in discussion of the policy-relevant science that should be considered in the ISA.
1 The general process for NAAQS reviews is described at http://www.epa. gov/ttn/naaqs/review.html.
2 Information for individual NAAQS reviews is available at www.epa.gov/ttn/naaqs.
November 2013 xxx DRAFT: Do Not Cite or Quote
-------
Workshop on
science-policy issues
EPA
proposed
decisions on
Integrated Review Plan (IRP): timeline and key
policy-relevant issues and scientific questions
Integrated Science Assessment (ISA): evaluation and
synthesis of most policy-relevant studies
Risk/Exposure Assessment (REA):
quantitative assessment, as warranted, focused
on key results, observations, and uncertainties
Policy Assessment (PA): staff analysis of
policy options based on integration and
interpretation of information in the ISA and REA
Figure I
Public hearings
and comments
on proposal
Agency decision
making and draft
final notice
Interagency
review
Clean Air Scientific
Advisory Committee
(C ASAC) review
Public comment
Schematic of the key steps in the process of the review of
National Ambient Air Quality Standards.
i
2
3
4
9
10
11
12
This preamble is a general discussion of the basic steps and criteria used in developing an
ISA. Details and considerations specific to an individual ISA are included in the Preface
and introductory section for that assessment. The general process for ISA development is
illustrated in Figure II.
The fundamental process for developing an ISA includes:
• literature searches;
• study selection;
• evaluation of individual study quality;
• evaluation, synthesis, and integration of the evidence; and
• development of scientific conclusions and causal determinations.
In developing an ISA, the U.S. Environmental Protection Agency (EPA) reviews and
summarizes the evidence from studies on atmospheric sciences, human exposure, animal
November 2013
XXX!
DRAFT: Do Not Cite or Quote
-------
1 toxicology, controlled human exposure, epidemiology, and ecology and other welfare1
2 effects. In the process of developing the first draft ISA, EPA may convene a peer input
3 meeting in which the scientific content of preliminary draft materials is reviewed to
4 ensure that the ISA is up to date and is focused on the most policy-relevant findings, and
5 to assist EPA with integration of evidence within and across disciplines.
6 EPA integrates the evidence from across scientific disciplines or study types and
7 characterizes the weight of evidence for relationships between the pollutant and various
8 outcomes. The integration of evidence on health or welfare effects, involves collaboration
9 between scientists from various disciplines. As an example, an evaluation of health
10 effects evidence would include the integration of the results from epidemiologic,
11 controlled human exposure, and toxicological studies, consideration of exposure
12 assessment, and application of the causal framework (described below) to draw
13 conclusions.
14 Integration of results on health or ecological effects that are logically or mechanistically
15 connected (e.g., effects on the respiratory system) informs judgments of causality. Using
16 the causal framework described in this Preamble, EPA scientists consider aspects such as
17 strength, consistency, coherence, and biological plausibility of the evidence and develop
18 causality determinations on the nature of the relationships. Causality determinations often
19 entail an iterative process of review and evaluation of the evidence. Two drafts of the ISA
20 are typically released for review by the Clean Air Scientific Advisory Committee
21 (CASAC) and the public, and comments received on the characterization of the science
22 as well as the implementation of the causal framework are carefully considered in
23 revising and completing the final ISA.
1 Welfare effects as defined in Clean Air Act (CAA) Section 302(h) [42 U.S.C. 7602(h)] include, but are not limited
to, "effects on soils, water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate,
damage to and deterioration of property, and hazards to transportation, as well as effects on economic values and on
personal comfort and well-being."
November 2013 xxxii DRAFT: Do Not Cite or Quote
-------
Literature Search and
Study Selection
(See Figure 3)
Evaluation of Individual Stud'
After study selection, the quality of individual studies is evaluated by EPA or outside experts in the fields of
atmospheric science, exposure assessment, dosimetry, animal toxicplogy, controlled human exposure studies,
epidemiology, ecology and other welfare effects, considering the design, methods, conduct, and documentation of
each study. Strengths and limitations of individual studies that may affect the interpretation of the study are
considered.
Develop Initial Sections
Review and summarize new study results and
findings and conclusions from previous
assessments by category of outcome/effect and
by discipline, e.g., toxicological studies of lung
function.
Peer Input Consultation
Review of initial draft materials by scientists
from both outside and within EPA in public
meeting or public teleconference.
Evaluation, Synthesis and Integration of Evidence
Integrate evidence from scientific disciplines-for example, toxicological, controlled human exposure and
epidemiologic study findings for particular health outcome. Evaluate evidence for related groups of endpoints or
outcomes to draw conclusions regarding health or welfare effect categories, integrating health or welfare effects
evidence with information on mode of action and exposure assessment.
Development of Scientific Conclusions and Causal Determinations
Characterize weight of evidence and devejop judgments regarding causality for health or welfare effect categories.
Develop conclusions regarding concentration- or dose-response relationships, potentially at-risk populations or
ecosystems.
Draft Integrated Science Assessment
Evaluation and integration of newly published studies
after each draft
Clean Air Scientific Advisory Committee
Independent review of draft documents for scientific
quality and sound implementation of causal
framework; anticipated review of two drafts of ISA in
public meetings.
Public Comments
Comments on draft ISA solicited by EPA
Final Integrated Science Assessment
Figure II
Characterization of the general process of ISA development.
November 2013
XXXlll
DRAFT: Do Not Cite or Quote
-------
2. Literature Search
1 An initial step in the literature search process is publication of a call for information in
2 the Federal Register that invites the public to provide information relevant to the
3 assessment, such as new or recent publications on health or welfare effects of the
4 pollutant. The EPA maintains an ongoing literature search process for identification of
5 relevant scientific studies published since the last ISA for a given criteria pollutant.
6 Search strategies are designed a priori for pollutants and scientific disciplines and
7 iteratively modified to optimize identification of pertinent publications. In addition,
8 papers are identified for inclusion in several ways: specialized searches on specific
9 topics; identification of new publications by relational searches conducted using citations
10 from previous assessments; review of tables of contents for journals in which relevant
11 papers may be published; identification of relevant literature by expert scientists; review
12 of citations in previous assessments and recommendations by the public and CASAC
13 during the call for information and external review processes. References identified
14 through the multipronged search strategy are screened by title and abstract. Those
15 references that are potentially relevant after reading the title are "considered" for
16 inclusion in the ISA and are added to the Health and Environmental Research Online
17 (HERO) database developed by EPA (http://hero.epa.gov/). The references cited in the
18 ISA include a hyperlink to the HERO database. This literature search and study selection
19 process is depicted in Figure III.
20 Studies and reports that have undergone scientific peer review and have been published
21 (or accepted for publication) are considered for inclusion in the ISA. All relevant
22 epidemiologic, controlled human exposure, toxicological, and ecological and welfare
23 effects studies published since the last review are considered, including those related to
24 exposure-response relationships, mode(s) of action (MOA), and potentially at-risk
25 populations and lifestages. Studies and data anlyses on atmospheric chemistry, air quality
26 and emissions, environmental fate and transport, dosimetry, toxicokinetics and exposure
27 are also considered for inclusion in the document. References considered for inclusion in
28 a specific ISA can be found using the HERO website (http://hero.epa.gov).
November 2013 xxxiv DRAFT: Do Not Cite or Quote
-------
Literature
Search
Strategies
Citations from
Past Assessments
Peer Review
Recommendations
Figure III Illustration of processes for literature search and study selection
used for development of ISAs.
1 Each ISA builds upon the conclusions of previous assessments for the pollutant under
2 review. EPA focuses on peer reviewed literature published following the completion of
3 the previous review and on any new interpretations of previous literature, integrating the
4 results of recent scientific studies with previous findings. Important earlier studies may
5 be discussed in detail to reinforce key concepts and conclusions or for reinterpretation in
6 light of newer data. Earlier studies also are the primary focus in some areas of the
7 document where research efforts have subsided, or if these earlier studies remain the
8 definitive works available in the literature.
3. Study Selection
9 Considered references undergo abstract and full-text review to determine if they will be
10 included in the ISA. The selection process is based on the extent to which the study is
November 2013 xxxv DRAFT: Do Not Cite or Quote
-------
1 informative and policy-relevant. Informative and policy-relevant studies include those
2 that provide a basis for or describe the relationship between the criteria pollutant and
3 effects, including studies that offer innovation in method or design and studies that
4 reduce uncertainty on critical issues. Emphasis is placed on studies that examine effects
5 associated with pollutant concentrations relevant to current human population and
6 ecosystem exposures, and particularly those pertaining to concentrations currently found
7 in ambient air. Other studies are included if they contain unique data, such as a
8 previously unreported effect or MOA for an observed effect, or examine multiple
9 concentrations to elucidate exposure-response relationships.
4. Evaluation of Individual Study Quality
10 After selecting studies for inclusion, the individual study quality is evaluated by
11 considering the design, methods, conduct, and documentation of each study, but not
12 whether the results are positive, negative, or null. This uniform approach aims to consider
13 the strengths, limitations, and possible roles of chance, confounding, and other biases that
14 may affect the interpretation of individual studies.
15 These criteria provide standards for evaluating various studies and for focusing on the
16 policy-relevant studies in assessing the body of health, ecological and welfare effects
17 evidence. As stated initially, the intent of the ISA is to provide a concise review,
18 synthesis, and evaluation of the most policy-relevant science to serve as a scientific
19 foundation for the review of the NAAQS, not extensive summaries of all health,
20 ecological and other welfare effects studies for a pollutant. Of most relevance for
21 inclusion of studies is whether they provide useful qualitative or quantitative information
22 on exposure-effect or exposure-response relationships for effects associated with
23 pollutant exposures at doses or concentrations relevant to ambient conditions that can
24 inform decisions on whether to retain or revise the standards.
25 In general, in assessing the scientific quality of studies on health and welfare effects, the
26 following considerations have been taken into account.
27 • Were study design, study groups, methods, data, and results clearly presented
28 to allow for study evaluation?
29 • Were the ecosystems, study site(s), study populations, subjects, or organism
30 models adequately selected, and are they sufficiently well defined to allow for
31 meaningful comparisons between study or exposure groups?
32 • Are the air quality data, exposure, or dose metrics of adequate quality and
33 sufficiently representative of information regarding ambient conditions?
November 2013 xxxvi DRAFT: Do Not Cite or Quote
-------
1 • Are the health, ecological or welfare effect measurements meaningful, valid
2 and reliable?
3 • Were likely covariates or modifying factors adequately controlled or taken
4 into account in the study design and statistical analysis?
5 • Do the analytical methods provide adequate sensitivity and precision to
6 support conclusions?
7 • Were the statistical analyses appropriate, properly performed, and properly
8 interpreted?
9 Additional considerations specific to particular disciplines are discussed below.
a. Atmospheric Science and Exposure Assessment
10 Considered atmospheric science and exposure assessment studies focus on measurement
11 of, behavior of, and exposure to ambient air pollution using quality-assured field,
12 experimental, and/or modeling techniques. The most informative measurement-based
13 studies will include detailed descriptive statistics for high-quality measurements taken at
14 varying spatial and temporal scales. These studies will also include a clear and
15 comprehensive description of measurement techniques and quality control procedures
16 used. Quality control metrics (e.g., method detection limits) and quantitative relationships
17 between and within pollutant measurements (e.g., regression slopes, intercepts, and fit
18 statistics) should be provided when appropriate. Measurements including contrasting
19 conditions for various time periods (e.g., weekday/weekend, season), populations,
20 regions, and categories (e.g., urban/rural) are particularly useful. The most informative
21 mode ling-based studies will incorporate appropriate chemistry, transport, dispersion,
22 and/or exposure modeling techniques with a clear and comprehensive description of
23 model evaluation procedures and metrics.
24 Exposure measurement error, which refers to the uncertainty associated with the exposure
25 metrics used to represent exposure of an individual or population, can be an important
26 contributor to uncertainty in air pollution epidemiologic study results. Exposure
27 measurement error can influence observed epidemiologic associations between ambient
28 pollutant concentrations and health outcomes by biasing effect estimates toward or away
29 from the null and widening confidence intervals around those estimates (Zeger et al.,
30 2000). Factors that could influence exposure estimates include, but are not limited to,
31 nonambient sources of exposure, topography of the natural and built environment,
32 meteorology, instrument errors, time-activity patterns, and the infiltration into indoor
33 environments. Additional information present in high-quality exposure studies includes
34 location and activity information from diaries, questionnaires, global positioning system
November 2013 xxxvii DRAFT: Do Not Cite or Quote
-------
1 data, or other means, as well as information on commuting patterns. In general,
2 atmospheric science and exposure studies focusing on locations pertinent to the U.S. will
3 have maximum value in informing review of the NAAQS.
b. Epidemiology
4 In selecting epidemiologic studies, EPA considers, in addition to the general quality
5 considerations discussed previously, whether a given study: (1) presents information on
6 associations with short- or long-term pollutant exposures at or near conditions relevant to
7 ambient exposures; (2) addresses potential confounding by other pollutants; (3) assesses
8 potential effect modifiers; (4) evaluates health endpoints and populations not previously
9 extensively researched; and (5) evaluates important methodological issues related to
10 interpretation of the health evidence (e.g., lag or time period between exposure and
11 effects, model specifications, thresholds).
12 In the evaluation of epidemiologic evidence, one important consideration is potential
13 confounding. Confounding is "... a confusion of effects. Specifically, the apparent effect
14 of the exposure of interest is distorted because the effect of an extraneous factor is
15 mistaken for or mixed with the actual exposure effect (which may be null)" (Rothman
16 and Greenland. 1998). A confounder is associated with both the exposure and the effect;
17 for example, confounding can occur between correlated pollutants that are associated
18 with the same effect. One approach to remove spurious associations due to possible
19 confounders is to control for characteristics that may differ between exposed and
20 unexposed persons; this is frequently termed "adjustment." Scientific judgment is needed
21 to evaluate likely sources and extent of confounding, together with consideration of how
22 well the existing constellation of study designs, results, and analyses address the potential
23 for erroneous inferences.
24 Several statistical methods are available to detect and control for potential confounders;
25 however, none of these methods is completely satisfactory. Multivariable regression
26 models constitute one tool for estimating the association between exposure and outcome
27 after adjusting for characteristics of participants that might confound the results. The use
28 of copollutant regression models has been the prevailing approach for controlling
29 potential confounding by copollutants in air pollution health effects studies. Trying to
30 determine if an individual pollutant is independently associated with the health outcome
31 of interest from copollutant regression models is made difficult by the possibility that one
32 or more air pollutants may be acting as a surrogate for an unmeasured or poorly measured
33 pollutant or for a particular mixture of pollutants. In addition, pollutants may
34 independently exert effects on the same system; for example, several pollutants may be
November 2013 xxxviii DRAFT: Do Not Cite or Quote
-------
1 associated with a respiratory effect through either the same or different modes of action.
2 Despite these limitations, the use of copollutant models is still the prevailing approach
3 employed in most air pollution epidemiologic studies and provides some insight into the
4 potential for confounding or interaction among pollutants.
5 Confidence that unmeasured confounders are not producing the findings is increased
6 when multiple studies are conducted in various settings using different subjects or
7 exposures, each of which might eliminate another source of confounding from
8 consideration. For example, multicity studies can provide insight on potential
9 confounding through the use of a consistent method to analyze data from across locations
10 with different concentrations of copollutants and other covariates. Intervention studies,
11 because of their quasi-experimental nature, can be particularly useful in characterizing
12 causation.
13 Another important consideration in the evaluation of epidemiologic studies is effect
14 measure modification, which occurs when the effect differs between subgroups or strata;
15 for example, effect estimates that vary by age group or a potential risk factor. As stated
16 by Rothman and Greenland (1998):
"Effect-measure modification differs from confounding in several ways. The main
difference is that, whereas confounding is a bias that the investigator hopes to prevent or
remove from the effect estimate, effect-measure modification is a property of the effect
under study ... In epidemiologic analysis one tries to eliminate confounding but one tries
to detect and estimate effect-measure modification."
17 When a risk factor is a confounder, it is the true cause of the association observed
18 between the exposure and the outcome; when a risk factor is an effect modifier, it
19 changes the magnitude of the association between the exposure and the outcome in
20 stratified analyses. For example, the presence of a pre-existing disease or indicator of low
21 socioeconomic status may act as effect modifiers if they are associated with increased
22 risk of effects related to air pollution exposure. It is often possible to stratify the
23 relationship between health outcome and exposure by one or more of these potential
24 effect modifiers. For variables that modify the association, effect estimates in each
25 stratum will be different from one another and different from the overall estimate,
26 indicating a different exposure-response relationship may exist in populations represented
27 by these variables.
c. Controlled Human Exposure and Animal Toxicology
28 Controlled human exposure and animal toxicological studies experimentally evaluate the
29 health effects of administered exposures in human volunteers and animal models under
30 highly controlled laboratory conditions. Controlled human exposure studies are also
November 2013 xxxix DRAFT: Do Not Cite or Quote
-------
1 referred to as human clinical studies. These experiments allow investigators to expose
2 subjects to known concentrations of air pollutants under carefully regulated
3 environmental conditions and activity levels. In addition to the general quality
4 considerations discussed previously, evaluation of controlled human exposure and animal
5 toxicological studies includes assessing the design and methodology of each study with
6 focus on (1) characterization of the intake dose, dosing regimen, and exposure route; (2)
7 characterization of the pollutant(s); (3) sample size and statistical power to detect
8 differences; and (4) control of other variables that could influence the occurrence of
9 effects. The evaluation of study design generally includes consideration of factors that
10 minimize bias in results such as randomization, blinding and allocation concealment of
11 study subjects, investigators, and research staff, and unexplained loss of animals or
12 withdrawal/exclusion of subjects. Additionally, studies must include appropriate control
13 groups and exposures to allow for accurate interpretation of results relative to exposure.
14 Emphasis is placed on studies that address concentration-dependent responses or time-
15 course of responses and studies that investigate potentially at-risk populations (e.g., age
16 or pre-existing disease).
17 Controlled human exposure or animal toxicological studies that approximate expected
18 human exposures in terms of concentration, duration, and route of exposure are of
19 particular interest. Relevant pollutant exposures are considered to be those generally
20 within two orders of magnitude of ambient concentrations, which may vary in animal
21 studies depending on dosimetry, toxicokinetics, and biological sensitivity of the species
22 or strain. Studies using higher concentration exposures or doses will be considered to the
23 extent that they provide information relevant to understanding MOA or mechanisms,
24 interspecies variation, or at-risk human populations. In vitro studies may be included if
25 they provide mechanistic insight or support results demonstrated in vivo.
d. Ecological Effects
26 In evaluating studies that consider ecological effects, in addition to assessing the general
27 quality considerations discussed previously, emphasis is placed on studies that evaluate
28 effects at or near ambient concentrations of the criteria air pollutants. Studies at higher
29 concentrations are used to evaluate ecological effects only when they are part of a range
30 of concentrations that also included more typical values, or when they inform
31 understanding of modes of action and illustrate the wide range of sensitivity to air
32 pollutants across taxa or across biomes and ecoregions. Studies conducted in any country
33 that contribute significantly to the general understanding of air pollutant effects are
34 considered for inclusion. In evaluating quantitative exposure-response relationships,
35 emphasis is placed on findings from studies conducted in the U.S. and Canada as having
November 2013 xl DRAFT: Do Not Cite or Quote
-------
1 ecological and climatic conditions most relevant for review of the NAAQS. The type of
2 experimental approach used in the study (e.g., controlled laboratory exposure, growth
3 chamber, open-top chamber, mesocosm, gradient, field study, etc.) is also evaluated when
4 considering the applicability of the results to the review of criteria air pollutant effects.
5. Evaluation, Synthesis, and Integration Across Disciplines and
Development of Scientific Conclusions and Causal
Determinations
5 EPA has developed a consistent and transparent basis for integration of scientific
6 evidence and evaluation of the causal nature of air pollution-related health or welfare
7 effects for use in developing ISAs. The framework described below establishes uniform
8 language concerning causality and brings specificity to the conclusions. This
9 standardized language was drawn from sources across the federal government and wider
10 scientific community, especially the U.S. EPA Guidelines for Carcinogen Risk
11 Assessment (2005) and National Academy of Sciences (NAS) Institute of Medicine
12 (IOM) document, Improving the Presumptive Disability Decision-Making Process for
13 Veterans (2008). a comprehensive report on evaluating causality.
14 This framework:
15 • describes the kinds of scientific evidence used in establishing a general causal
16 relationship between exposure and health effects;
17 • characterizes the process for integration and evaluation of evidence necessary
18 to reach a conclusion about the existence of a causal relationship;
19 • identifies issues and approaches related to uncertainty; and
20 • provides a framework for classifying and characterizing the weight of
21 evidence in support of a general causal relationship.
22 Approaches to assessing the separate and combined lines of human health evidence
23 (e-g-, epidemiologic, controlled human exposure, and animal toxicological studies) have
24 been formulated by a number of regulatory and science agencies, including the IOM of
25 the NAS (IOM. 2008). the International Agency for Research on Cancer (IARC. 2006).
26 the U.S. EPA (2005). and the Centers for Disease Control and Prevention (CDC. 2004).
27 Causal inference criteria have also been described for ecological effects evidence (U.S.
28 EPA. 1998a; Fox. 1991). These formalized approaches offer guidance for assessing
29 causality. The frameworks are similar in nature, although adapted to different purposes,
30 and have proven effective in providing a uniform structure and language for causal
31 determinations.
November 2013 xli DRAFT: Do Not Cite or Quote
-------
1 The 1964 Surgeon General's report defined "cause" as a "significant, effectual
2 relationship between an agent and an associated disorder or disease in the host" (HEW.
3 1964). More generally, a cause is defined as an agent that brings about an effect or a
4 result. An association is the statistical relationship among variables; alone, however, it is
5 insufficient proof of a causal relationship between an exposure and a health outcome.
6 Unlike an association, a causal claim supports the creation of counterfactual claims; that
7 is, a claim about what the world would have been like under different or changed
8 circumstances (TOM. 2008).
9 Many of the health and environmental outcomes reported in these studies have complex
10 etiologies. Diseases such as asthma, coronary heart disease or cancer are typically
11 initiated by multiple agents. Outcomes depend on a variety of factors, such as age,
12 genetic background, nutritional status, immune competence, and social factors (TOM.
13 2008; Gee and Payne-Sturges. 2004). Effects on ecosystems are also often multifactorial
14 with a complex web of causation. Further, exposure to a combination of agents could
15 cause synergistic or antagonistic effects. Thus, the observed risk may represent the net
16 effect of many actions and counteractions.
a. Evaluation, Synthesis, and Integration of Evidence
Across Disciplines
17 Moving from association to causation involves the elimination of alternative explanations
18 for the association. The ISA focuses on evaluation of the findings from the body of
19 evidence across disciplines, drawing upon the results of all studies determined to meet the
20 criteria described previously. Evidence from across scientific disciplines for related and
21 similar health or welfare effects is evaluated, synthesized, and integrated to develop
22 conclusions and causality determinations. This includes the evaluation of strengths and
23 weaknesses in the overall collection of studies across disciplines. Confidence in the body
24 of evidence is based on evaluation of study design and quality. The relative importance of
25 different types of evidence to the conclusions varies by pollutant or assessment, as does
26 the availability of different types of evidence for causality determination. Consideration
27 of human health effects are informed by controlled human exposure, epidemiologic, and
28 toxicological studies. Evidence on ecological effects may be drawn from a variety of
29 experimental approaches (e.g., greenhouse, laboratory, field) and numerous disciplines
30 (e-g-, community ecology, biogeochemistry and paleontological/historical
31 reconstructions). Other evidence including mechanistic, toxicokinetics, and exposure
32 assessment may be highlighted if it is relevant to the evaluation of health and ecological
33 effects and if it is of sufficient importance to affect the overall evaluation.
November 2013 xlii DRAFT: Do Not Cite or Quote
-------
1 Evaluation and integration of evidence must also include consideration of uncertainty,
2 which is inherent in scientific findings. "Uncertainty" can be defined as a deficit of
3 knowledge to describe the existing state or future outcome with accuracy and precision,
4 e.g., the lack of knowledge about the correct value for a specific measure or estimate.
5 Uncertainty analysis may be qualitative or quantitative in nature. In many cases, the
6 analysis is qualitative and can include professional judgment or inferences based on
7 analogy with similar situations. Quantitative uncertainty analysis may include use of
8 simple measures (e.g., ranges) and analytical techniques. Quantitative uncertainty
9 analysis might progress to more complex measures and techniques, if needed for decision
10 support. Various approaches to evaluating uncertainty include classical statistical
11 methods, sensitivity analysis, or probabilistic uncertainty analysis, in order of increasing
12 complexity and data requirements. However, data may not be available for all aspects of
13 an assessment, and those data that are available may be of questionable or unknown
14 quality. Ultimately, the assessment is based on a number of assumptions with varying
15 degrees of uncertainty. While the ISA may include quantitative analysis approaches, such
16 as meta-regression, in some situations, generally qualitative evaluation of uncertainties is
17 used in assessing the evidence from across studies.
18 Publication bias is another source of uncertainty that can impact the magnitude of health
19 risk estimates. It is well understood that studies reporting non-null findings are more
20 likely to be published than reports of null findings. Publication bias can result in
21 overestimation of effect estimate sizes (loannidis. 2008). For example, effect estimates
22 from single-city epidemiologic studies have been found to be generally larger than those
23 from multicity studies which is an indication of publication bias in that null or negative
24 single-city results may be reported in multicity analyses but might not be published
25 independently (Bell et al.. 2005).
26 Potential strengths and limitations of the body of studies can vary across disciplines and
27 are evaluated during data synthesis and integration. Direct evidence of a relationship
28 between pollutant exposures and human health effects may come from controlled human
29 exposure studies. These studies can also provide important information on the biological
30 plausibility of associations observed in epidemiologic studies and inform determinations
31 of response modifying factors that may increase or decrease the risk of health effects in
32 certain populations. In some instances, controlled human exposure studies can be used to
33 characterize concentration-response relationships at pollutant concentrations relevant to
34 ambient conditions. Controlled human exposures are typically conducted using a
35 randomized crossover design, with subjects exposed both to the pollutant and a clean air
36 control. In this way, subjects serve as their own experimental controls, effectively
37 limiting the variance associated with potential inter-individual confounders. Limitations
38 that must be considered in evaluating controlled human study findings include the
November 2013 xliii DRAFT: Do Not Cite or Quote
-------
1 generally small sample size and short exposure time used in experimental studies, and
2 that severe health outcomes are not assessed. By experimental design, controlled human
3 exposure studies are structured to evaluate physiological or biomolecular outcomes in
4 response to exposure to a specific air pollutant and/or combination of pollutants. In
5 addition, the study design generally precludes inclusion of subjects with serious health
6 conditions, and therefore the results often cannot be generalized to an entire population.
7 Although some controlled human exposure studies have included health-compromised
8 individuals such as those with respiratory or cardiovascular disease, these individuals
9 may also be relatively healthy and may not represent the most sensitive individuals in the
10 population. Thus, observed effects in these studies may underestimate the response in
11 certain populations. In addition, the study design is limited to exposures and endpoints
12 that are not expected to result in severe health outcomes.
13 Epidemiologic studies provide important information on the associations between health
14 effects and exposure of human populations to ambient air pollution. In epidemiologic or
15 observational studies of humans, the investigator does not control exposures or intervene
16 with the study population. Broadly, observational studies can describe associations
17 between exposures and effects. These studies fall into several categories:
18 e.g., cross-sectional, prospective cohort, panel, and time-series studies, and have various
19 strengths and limitations. Cross-sectional ecologic studies use health outcome, exposure
20 and covariate data available at the community level (e.g., annual mortality rates and
21 pollutant concentrations), but do not have individual-level data. Prospective cohort
22 studies include some data collected at the individual level, which is typically health
23 outcome data, and in some cases individual-level data on exposure and covariates are
24 collected. Time-series and case-crossover studies are often used to evaluate the
25 relationship between day-to-day changes in air pollution exposures and a specific health
26 outcome at the population-level (i.e., mortality, hospital admissions or emergency
27 department visits). Panel studies include repeated measurements of health outcomes, such
28 as respiratory symptoms or heart rate variability, at the individual level. "Natural
29 experiments" offer the opportunity to investigate changes in health related to a change in
30 exposure, such as closure of a pollution source.
31 When evaluating the collective body of epidemiologic studies, consideration of many
32 study design factors and limitations must be taken into account to properly inform their
33 interpretation. One key consideration is the evaluation of the potential independent
34 contribution of the pollutant to a health outcome when it is a component of a complex air
35 pollutant mixture. Reported effect estimates in epidemiologic studies may reflect
36 (1) independent effects on health outcomes; (2) effects of the pollutant acting as an
37 indicator of a copollutant or a complex ambient air pollution mixture; and (3) effects
38 resulting from interactions between that pollutant and copollutants.
November 2013 xliv DRAFT: Do Not Cite or Quote
-------
1 The third main type of health effects evidence, animal toxicological studies, provides
2 information on the pollutant's biological action under controlled and monitored exposure
3 circumstances. Taking into account physiological differences of the experimental species
4 from humans, these studies inform characterization of health effects of concern,
5 exposure-response relationships and MOAs. Further, animal models can inform
6 determinations of response modifying factors that may increase or decrease the risk of
7 health effects in certain populations. These studies evaluate the effects of exposures to a
8 variety of pollutants in a highly controlled laboratory setting and allow exploration of
9 toxicological pathways or mechanisms by which a pollutant may cause effects.
10 Understanding the biological mechanisms underlying various health outcomes can prove
11 crucial in establishing or negating causality. In the absence of human studies data,
12 extensive, well-conducted animal toxicological studies can support determinations of
13 causality, if the evidence base indicates that similar responses are expected in humans
14 under ambient exposure conditions.
15 Interpretations of animal toxicological studies are affected by limitations associated with
16 extrapolation between animal and human responses. The differences between humans
17 and other species have to be taken into consideration, including metabolism, hormonal
18 regulation, breathing pattern, and differences in lung structure and anatomy. Also, in spite
19 of a high degree of homology and the existence of a high percentage of orthologous
20 genes across humans and rodents (particularly mice), extrapolation of molecular
21 alterations at the gene or protein level is complicated by species-specific differences in
22 transcriptional regulation and/or signaling. Given these differences, there are
23 uncertainties associated with quantitative extrapolations of observed pollutant-induced
24 pathophysiological alterations between laboratory animals and humans, as those
25 alterations are under the control of widely varying biochemical, endocrine, and neuronal
26 factors.
27 For ecological effects assessment, both laboratory and field studies (including field
28 experiments and observational studies) can provide useful data for causality
29 determination. Because conditions can be controlled in laboratory studies, responses may
30 be less variable and smaller effects may be easier to detect. However, the control
31 conditions may limit the range of responses (e.g., animals may not be able to seek
32 alternative food sources) or incompletely reflect pollutant bioavailability, so they may not
33 reflect responses that would occur in the natural environment. In addition, larger-scale
34 processes are difficult to reproduce in the laboratory.
35 Field observational studies measure biological changes in uncontrolled situations with
36 high natural variability (in organismal genetics, or in abiotic seasonal, climatic, or soil-
37 related factors) and describe an association between a disturbance and an ecological
November 2013 xlv DRAFT: Do Not Cite or Quote
-------
1 effect. Field data can provide important information for assessments of multiple stressors
2 or where site-specific factors significantly influence exposure. They are also often useful
3 for analyses of pollutant effects at larger geographic scales and higher levels of biological
4 organization. However, because conditions are not controlled, variability of the response
5 is expected to be higher and may mask effects. Field surveys are most useful for linking
6 stressors with effects when stressor and effect levels are measured concurrently. The
7 presence of confounding factors can make it difficult to attribute observed effects to
8 specific stressors.
9 Ecological impacts of pollutants are also evaluated in studies "intermediate" between the
10 lower variability typically associated with laboratory exposures and high natural
11 variability usually found in field studies. Some use environmental media collected from
12 the field to examine the biological responses under controlled laboratory conditions.
13 Others are experiments that are performed in the natural environment while controlling
14 for some, but not all, of the environmental or genetic variability (i.e., mesocosm studies).
15 This type of study in manipulated natural environments can be considered a hybrid
16 between a field experiment and laboratory study since some sources of response variation
17 are removed through use of control conditions while others are included to mimic natural
18 variation. They make it possible to observe community and/or ecosystem dynamics, and
19 provide strong evidence for causality when combined with findings of studies that have
20 been made under more controlled conditions.
b. Application of Framework for Scientific Conclusions and
Causal Determinations
21 In its evaluation and integration of the scientific evidence on health or welfare effects of
22 criteria pollutants, EPA determines the weight of evidence in support of causation and
23 characterizes the strength of any resulting causal classification. EPA also evaluates the
24 quantitative evidence and draws scientific conclusions, to the extent possible, regarding
25 the concentration-response relationships and the loads to ecosystems, exposures, doses or
26 concentrations, exposure duration, and pattern of exposures at which effects are observed.
November 2013 xlvi DRAFT: Do Not Cite or Quote
-------
Table I
Aspects to aid in judging causality.
Aspect
Description
Consistency of the
observed association
An inference of causality is strengthened when a pattern of elevated risks is observed
across several independent studies. The reproducibility of findings constitutes one of
the strongest arguments for causality. If there are discordant results among
investigations, possible reasons such as differences in exposure, confounding
factors, and the power of the study are considered.
Coherence An inference of causality from one line of evidence (e.g., epidemiologic, clinical, or
animal studies) may be strengthened by other lines of evidence that support a
cause-and-effect interpretation of the association. For example, evidence on welfare
effects may be drawn from a variety of experimental approaches (e.g., greenhouse,
laboratory, and field) and subdisciplines of ecology (e.g., community ecology,
biogeochemistry, and paleontological/historical reconstructions). The coherence of
evidence from various fields greatly adds to the strength of an inference of causality.
In addition, there may be coherence in demonstrating effects across multiple study
designs or related health endpoints within one scientific line of evidence.
Biological plausibility.
Biological gradient
(exposure-response
relationship)
Strength of the observed
association
Experimental evidence
Temporal relationship of
the observed association
An inference of causality tends to be strengthened by consistency with data from
experimental studies or other sources demonstrating plausible biological
mechanisms. A proposed mechanism linking between an effect and exposure to the
agent is an important source of support for causality, especially when data
establishing the existence and functioning of those mechanistic links are available.
A well-characterized exposure-response relationship (e.g., increasing effects
associated with greater exposure) strongly suggests cause and effect, especially
when such relationships are also observed for duration of exposure (e.g., increasing
effects observed following longer exposure times).
The finding of large, precise risks increases confidence that the association is not
likely due to chance, bias, or other factors. However, it is noted that a small
magnitude in an effect estimate may represent a substantial effect in a population.
Strong evidence for causality can be provided through "natural experiments" when a
change in exposure is found to result in a change in occurrence or frequency of
health or welfare effects.
Evidence of a temporal sequence between the introduction of an agent, and
appearance of the effect, constitutes another argument in favor of causality.
Specificity of the
observed association
Evidence linking a specific outcome to an exposure can provide a strong argument
for causation. However, it must be recognized that rarely, if ever, does exposure to a
pollutant invariably predict the occurrence of an outcome, and that a given outcome
may have multiple causes.
Analogy Structure activity relationships and information on the agent's structural analogs can
provide insight into whether an association is causal. Similarly, information on mode
of action for a chemical, as one of many structural analogs, can inform decisions
regarding likely causality.
November 2013
xlvii
DRAFT: Do Not Cite or Quote
-------
1 To aid judgment, various "aspects"1 of causality have been discussed by many
2 philosophers and scientists. The 1964 Surgeon General's report on tobacco smoking
3 discussed criteria for the evaluation of epidemiologic studies, focusing on consistency,
4 strength, specificity, temporal relationship, and coherence (HEW. 1964). Sir Austin
5 Bradford Hill (Hill, 1965) articulated aspects of causality in epidemiology and public
6 health that have been widely used (TOM. 2008: IARC. 2006: U.S. EPA. 2005: CDC.
7 2004). These aspects (Hill, 1965) have been modified (Table I) for use in causal
8 determinations specific to health and welfare effects for pollutant exposures (U.S. EPA.
9 2009a).2 Although these aspects provide a framework for assessing the evidence, they do
10 not lend themselves to being considered in terms of simple formulas or fixed rules of
11 evidence leading to conclusions about causality (Hill. 1965). For example, one cannot
12 simply count the number of studies reporting statistically significant results or
13 statistically nonsignificant results and reach credible conclusions about the relative
14 weight of the evidence and the likelihood of causality. Rather, these aspects provide a
15 framework for systematic appraisal of the body of evidence, informed by peer and public
16 comment and advice, which includes weighing alternative views on controversial issues.
17 In addition, it is important to note that the aspects in Table I cannot be used as a strict
18 checklist, but rather to determine the weight of the evidence for inferring causality. In
19 particular, not meeting one or more of the principles does not automatically preclude a
20 determination of causality [see discussion in (CDC. 2004)1.
c. Determination of Causality
21 In the ISA, EPA assesses the body of relevant literature, building upon evidence available
22 during previous NAAQS reviews, to draw conclusions on the causal relationships
23 between relevant pollutant exposures and health or environmental effects. IS As use a
24 five-level hierarchy that classifies the weight of evidence for causation3. In developing
25 this hierarchy, EPA has drawn on the work of previous evaluations, most prominently the
26 lOM's Improving the Presumptive Disability Decision-Making Process for Veterans
27 (TOM. 2008). EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005). and
28 the U.S. Surgeon General's smoking report (CDC. 2004). This weight of evidence
1 The "aspects" described by Sir Austin Bradford Hill (Hill. 1965) have become, in the subsequent literature, more
commonly described as "criteria." The original term "aspects" is used here to avoid confusion with "criteria" as it is
used, with different meaning, in the Clean Air Act.
2 The Hill aspects were developed for interpretation of epidemiologic results. They have been modified here for use
with a broader array of data, i.e., epidemiologic, controlled human exposure, ecological, and animal toxicological
studies, as well as in vitro data, and to be more consistent with the EPA Guidelines for Carcinogen Risk Assessment.
3 The Center for Disease Control (CDC) and IOM frameworks use a four-category hierarchy for the strength of the
evidence. A five-level hierarchy is used here to be consistent with the EPA Guidelines for Carcinogen Risk
Assessment and to provide a more nuanced set of categories.
November 2013 xlviii DRAFT: Do Not Cite or Quote
-------
1 evaluation is based on integration of findings from various lines of evidence from across
2 the health and environmental effects disciplines. These separate judgments are integrated
3 into a qualitative statement about the overall weight of the evidence and causality. The
4 five descriptors for causal determination are described in Table II.
5 Determination of causality involves the evaluation and integration of evidence for
6 different types of health, ecological or welfare effects associated with short- and long-
7 term exposure periods. In making determinations of causality, evidence is evaluated for
8 major outcome categories or groups of related endpoints (e.g., respiratory effects,
9 vegetation growth), integrating evidence from across disciplines, and evaluating the
10 coherence of evidence across a spectrum of related endpoints to draw conclusions
11 regarding causality. In discussing the causal determination, EPA characterizes the
12 evidence on which the judgment is based, including strength of evidence for individual
13 endpoints within the outcome category or group of related endpoints.
14 In drawing judgments regarding causality for the criteria air pollutants, the ISA focuses
15 on evidence of effects in the range of relevant pollutant exposures or doses, and not on
16 determination of causality at any dose. Emphasis is placed on evidence of effects at doses
17 (e.g., blood Pb concentration) or exposures (e.g., air concentrations) that are relevant to,
18 or somewhat above, those currently experienced by the population. The extent to which
19 studies of higher concentrations are considered varies by pollutant and major outcome
20 category, but generally includes those with doses or exposures in the range of one to two
21 orders of magnitude above current or ambient conditions. Studies that use higher doses or
22 exposures may also be considered to the extent that they provide useful information to
23 inform understanding of mode of action, interspecies differences, or factors that may
24 increase risk of effects for a population. Thus, a causality determination is based on
25 weight of evidence evaluation for health or welfare effects, focusing on the evidence
26 from exposures or doses generally ranging from current levels to one or two orders of
27 magnitude above current levels.
28 In addition, EPA evaluates evidence relevant to understand the quantitative relationships
29 between pollutant exposures and health or welfare effects. This includes evaluating the
30 form of concentration-response or dose-response relationships and, to the extent possible,
31 drawing conclusions on the levels at which effects are observed. The ISA also draws
32 scientific conclusions regarding important exposure conditions for effects and
33 populations that may be at greater risk for effects, as described in the following section.
November 2013 xlix DRAFT: Do Not Cite or Quote
-------
Table II Weight of evidence for causal determination.
Health Effects
Ecological and Welfare Effects
Causal Evidence is sufficient to conclude that there is a
relationship causal relationship with relevant pollutant exposures
(e.g., doses or exposures generally within one to two
orders of magnitude of current levels). That is, the
pollutant has been shown to result in health effects in
studies in which chance, confounding, and other
biases could be ruled out with reasonable confidence.
For example: (1) controlled human exposure studies
that demonstrate consistent effects; or (2)
observational studies that cannot be explained by
plausible alternatives or that are supported by other
lines of evidence (e.g., animal studies or mode of
action information). Generally, the determination is
based on multiple high-quality studies conducted by
multiple research groups.
Evidence is sufficient to conclude that there is a
causal relationship with relevant pollutant exposures
(e.g., doses or exposures generally within one to two
orders of magnitude of current levels). That is, the
pollutant has been shown to result in effects in
studies in which chance, confounding, and other
biases could be ruled out with reasonable
confidence. Controlled exposure studies (laboratory
or small- to medium-scale field studies) provide the
strongest evidence for causality, but the scope of
inference may be limited. Generally, the
determination is based on multiple studies conducted
by multiple research groups, and evidence that is
considered sufficient to infer a causal relationship is
usually obtained from the joint consideration of many
lines of evidence that reinforce each other.
Likely to be
a causal
relationship
Evidence is sufficient to conclude that a causal
relationship is likely to exist with relevant pollutant
exposures. That is, the pollutant has been shown to
result in health effects in studies where results are not
explained by chance, confounding, and other biases,
but uncertainties remain in the evidence overall. For
example: (1) observational studies show an
association, but copollutant exposures are difficult to
address and/or other lines of evidence (controlled
human exposure, animal, or mode of action
information) are limited or inconsistent; or (2) animal
toxicological evidence from multiple studies from
different laboratories demonstrate effects, but limited
or no human data are available. Generally, the
determination is based on multiple high-quality
studies.
Evidence is sufficient to conclude that there is a likely
causal association with relevant pollutant exposures.
That is, an association has been observed between
the pollutant and the outcome in studies in which
chance, confounding, and other biases are
minimized, but uncertainties remain. For example,
field studies show a relationship, but suspected
interacting factors cannot be controlled, and other
lines of evidence are limited or inconsistent.
Generally, the determination is based on multiple
studies by multiple research groups.
Suggestive
of a causal
relationship
Evidence is suggestive of a causal relationship with
relevant pollutant exposures, but is limited. For
example, (1) at least one high-quality epidemiologic
study shows an association with a given health
outcome although inconsistencies remain across other
studies that are or are not of comparable quality; or (2)
a well-conducted toxicological study, such as those
conducted in the National Toxicology Program (NTP),
shows effects relevant to humans in animal species.
Evidence is suggestive of a causal relationship with
relevant pollutant exposures, but chance,
confounding, and other biases cannot be ruled out.
For example, at least one high-quality study shows
an effect, but the results of other studies are
inconsistent.
Inadequate
to infer a
causal
relationship
Evidence is inadequate to determine that a causal
relationship exists with relevant pollutant exposures.
The available studies are of insufficient quantity,
quality, consistency, or statistical power to permit a
conclusion regarding the presence or absence of an
effect.
The available studies are of insufficient quality,
consistency, or statistical power to permit a
conclusion regarding the presence or absence of an
effect.
Not likely to
be a causal
relationship
Evidence indicates there is no causal relationship with
relevant pollutant exposures. Several adequate
studies, covering the full range of levels of exposure
that human beings are known to encounter and
considering at-risk populations and lifestages, are
mutually consistent in not showing an effect at any
level of exposure.
Several adequate studies, examining relationships
with relevant exposures, are consistent in failing to
show an effect at any level of exposure.
November 2013
DRAFT: Do Not Cite or Quote
-------
6. Public Health Impact
1 Once a determination is made regarding the causal relationship between the pollutant and
2 outcome category, important questions regarding the public health impact include:
3 • What is the concentration-response, exposure-response, or dose-response
4 relationship in the human population?
5 • What is the interrelationship between incidence and severity of effect?
6 • What exposure conditions (dose or exposure, duration and pattern) are
7 important?
8 • What populations and lifestages appear to be differentially affected (i.e., at
9 greater or less risk of experiencing effects)?
10 In order to address these questions, the entirety of quantitative evidence is evaluated to
11 characterize pollutant concentrations and exposure durations at which effects were
12 observed for exposed populations, including populations and lifestages potentially at
13 increased risk. To accomplish this, evidence is considered from multiple and diverse
14 types of studies, and a study or set of studies that best approximates the concentration-
15 response relationships between health outcomes and the pollutant may be identified.
16 Controlled human exposure studies provide the most direct and quantifiable exposure-
17 response data on the human health effects of pollutant exposures. To the extent available,
18 the ISA evaluates results from epidemiologic studies that characterize the form of
19 relationships between the pollutant and health outcomes and draws conclusions on the
20 shape of these relationships. Animal data may also inform evaluation of
21 concentration-response relationships, particularly relative to MOAs and characteristics of
22 at-risk populations.
23 An important consideration in characterizing the public health impacts associated with
24 exposure to a pollutant is whether the concentration-response relationship is linear across
25 the range of concentrations or if nonlinear relationships exist along any part of this range.
26 The shape of the concentration-response curve at and below the level of the current
27 standards is of particular interest. Various sources of variability and uncertainty, such as
28 low data density in the lower concentration range, possible influence of exposure
29 measurement error, and variability between individuals in susceptibility to air pollution
30 health effects, tend to smooth and "linearize" the concentration-response function and
31 thus can obscure the existence of a threshold or nonlinear relationship. Since individual
32 thresholds vary from person to person due to individual differences such as genetic level
33 susceptibility or pre-existing disease conditions (and even can vary from one time to
34 another for a given person), it can be difficult to demonstrate that a threshold exists in a
35 population study. These sources of variability and uncertainty may explain why the
November 2013 li DRAFT: Do Not Cite or Quote
-------
1 available human data at ambient concentrations for some environmental pollutants
2 (e-g-, particulate matter [PM], O3, lead [Pb], environmental tobacco smoke [ETS],
3 radiation) do not exhibit population-level thresholds for cancer or noncancer health
4 effects, even though likely mechanisms include nonlinear processes for some key events.
5 Finally, identification of the population groups or lifestages that may be at greater risk, or
6 in some cases decreased risk, of health effects from air pollutant exposures contributes to
7 an understanding of the public health impact of pollutant exposures. In the ISA, the term
8 "at-risk population" is used to encompass characteristics of populations or lifestages that
9 have a greater, or decreased, likelihood of experiencing health effects related to exposure
10 to an air pollutant due to a variety of risk modifying factors.lt should be noted that other
11 terms have often been used in the literature to identify these populations and lifestages,
12 including susceptible, vulnerable, and sensitive.
13 It is recognized that these factors may be intrinsic due to an increase in risk for an effect
14 through a biological mechanism, such as genetic or developmental factors, race, sex,
15 lifestage, or the presence of pre-existing diseases. In general, people in this category
16 would have a steeper concentration-risk relationship, compared to those not in the
17 category. Additionally, the factors may be extrinsic due to an increase in risk for an effect
18 through an external, non-biological factor, such as socioeconomic status (SES) (e.g.,
19 educational attainment, income, access to healthcare, etc.), activity pattern and exercise
20 level, reduced access to health care, low educational attainment, or increased pollutant
21 exposures (e.g., near roadways). Some groups are at risk of increased internal dose at a
22 given exposure concentration, which includes individuals that have a greater dose of
23 delivered pollutant because of breathing pattern. This category would include children
24 who are typically more active outdoors. In addition, some groups could have greater
25 exposure (concentration x time) regardless of the delivered dose, such as outdoor
26 workers. Finally, there are those who might be placed at increased risk for experiencing a
27 greater exposure by being exposed at a higher concentration. Some factors described
28 above are multifaceted and may influence the risk of an air pollutant related health effect
29 through a combination of avenues. The emphasis is to identify and understand the factors
30 that potentially increase, or in some cases decrease, the risk of air pollutant-related health
31 effects, regardless of whether the increased risk is due to intrinsic factors, extrinsic
32 factors, increased dose/exposure, or a combination due to the often interconnectedness of
33 factors.
7. Approach to Classifying At-Risk Factors
34 To identify at-risk factors that potentially lead to some populations or lifestages being at
35 increased or decreased risk of air pollution-related health effects, the evidence is
November 2013 lii DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
systematically evaluated across relevant scientific disciplines (i.e., exposure sciences,
dosimetry, toxicology, and epidemiology). An evaluation of studies first consists of
focusing on studies that conducted stratified analyses (i.e., epidemiologic or controlled
human exposure) to compare populations or lifestages exposed to similar air pollutant
concentrations within the same study design. Experimental studies also provide important
lines of evidence in the evaluation of at-risk factors that may lead to increased or
decreased risk of an air pollutant related-health effect. Toxicological studies conducted
using animal models of disease and controlled human exposure studies that examine
individuals with underlying disease or genetic polymorphisms may provide evidence in
the absence of stratified epidemiologic analyses. Additionally these studies can provide
support for coherence with the health effects observed in epidemiologic studies as well as
an understanding of biological plausibility. The potential increased or decreased risk of
an air pollutant-related health effect may also be determined from studies that examined
at-risk factors that result in differential air pollutant exposures. Building on the causal
framework discussed in detail above, conclusions are reached regarding the strength of
evidence across scientific disciplines for each at-risk factor that may contribute to
increased or decreased risk of an air pollutant-related health effect. The conclusions
drawn consider the "Aspects to Aid in Judging Causality" discussed in Table I. The
categories considered for evaluating the potential increased risk of an air pollutant-related
health effect are "adequate evidence," "suggestive evidence," "inadequate evidence," and
"evidence of no effect." They are described in more detail in Table III.
Table
Classification of evidence for potential at-risk factors.
Classification
Adequate
evidence
Suggestive
evidence
Inadequate
evidence
Evidence of
no effect
Health Effects
There is substantial, consistent evidence within a discipline to conclude that a factor results in
a population or lifestage being at increased or decreased risk of air pollutant-related health
effect(s) relative to some reference population or lifestage. Where applicable this includes
coherence across disciplines. Evidence includes multiple high-quality studies.
The collective evidence suggests that a factor results in a population or lifestage being at
increased or decreased risk of an air pollutant-related health effect relative to some reference
population or lifestage, but the evidence is limited due to some inconsistency within a
discipline or, where applicable, a lack of coherence across disciplines.
The collective evidence is inadequate to determine if a factor results in a population or
lifestage being at increased or decreased risk of an air pollutant-related health effect relative
to some reference population or lifestage. The available studies are of insufficient quantity,
quality, consistency, and/or statistical power to permit a conclusion to be drawn.
There is substantial, consistent evidence within a discipline to conclude that a factor does not
result in a population or lifestage being at increased or decreased risk of air pollutant-related
health effect(s) relative to some reference population or lifestage. Where applicable this
includes coherence across disciplines. Evidence includes multiple high-quality studies.
November 2013
liii
DRAFT: Do Not Cite or Quote
-------
8. Quantitative Relationships: Effects on Welfare
1 Key questions for understanding the quantitative relationships between exposure (or
2 concentration or deposition) to a pollutant and risk to ecosystems or other welfare effects
3 include:
4 • What elements of the ecosystem (e.g., types, regions, taxonomic groups,
5 populations, functions, etc.) appear to be affected, or are more sensitive to
6 effects? Are there differences between locations or materials in welfare effects
7 responses, such as impaired visibility or materials damage?
8 • Under what exposure conditions (amount deposited or concentration, duration
9 and pattern) are effects seen?
10 • What is the shape of the concentration-response or exposure-response
11 relationship?
12 Evaluations of causality generally consider the probability of quantitative changes in
13 welfare effects in response to exposure. A challenge to the quantification of exposure-
14 response relationships for ecological effects is the great regional and local spatial
15 variability, as well as temporal variability, in ecosystems. Thus, exposure-response
16 relationships are often determined for a specific ecological system and scale, rather than
17 at the national or even regional scale. Quantitative relationships therefore are estimated
18 site by site and may differ greatly between ecosystems.
9. Concepts in Evaluating Adversity
a. Evaluating Adversity of Health Effects
19 In evaluating health evidence, a number of factors can be considered in delineating
20 between adverse and nonadverse health effects resulting from exposure to air pollution.
21 Some health outcomes, such as hospitalization for respiratory or cardiovascular diseases,
22 are clearly considered adverse. It is more difficult to determine the extent of change that
23 constitutes adversity in more subtle health measures. These include a wide variety of
24 responses, such as alterations in markers of inflammation or oxidative stress, changes in
25 pulmonary function or heart rate variability, or alterations in neurocognitive function
26 measures. The challenge is determining the magnitude of change in these measures when
27 there is no clear point at which a change becomes adverse. The extent to which a change
28 in health measure constitutes an adverse health effect may vary between populations.
29 Some changes that may not be considered adverse in healthy individuals would be
30 potentially adverse in more at-risk individuals.
November 2013 liv DRAFT: Do Not Cite or Quote
-------
1 Professional scientific societies may evaluate the magnitude of change in an outcome or
2 event that is considered adverse. For example, the extent to which changes in lung
3 function are adverse has been discussed by the American Thoracic Society (ATS) in an
4 official statement titled What Constitutes an Adverse Health Effect of Air Pollution?
5 (ATS. 2000b). An air pollution-induced shift in the population distribution of a given risk
6 factor for a health outcome was viewed as adverse, even though it may not increase the
7 risk of any one individual to an unacceptable level. For example, a population of
8 asthmatics could have a distribution of lung function such that no identifiable individual
9 has a level associated with significant impairment. Exposure to air pollution could shift
10 the distribution such that no identifiable individual experiences clinically relevant effects.
11 This shift toward decreased lung function, however, would be considered adverse
12 because individuals within the population would have diminished reserve function and
13 therefore would be at increased risk to further environmental insult. The committee also
14 observed that elevations of biomarkers, such as cell number and types, cytokines and
15 reactive oxygen species, may signal risk for ongoing injury and clinical effects or may
16 simply indicate transient responses that can provide insights into mechanisms of injury,
17 thus illustrating the lack of clear boundaries that separate adverse from nonadverse
18 effects.
19 The more subtle health outcomes may be connected mechanistically to health events that
20 are clearly adverse. For example, air pollution may affect markers of transient myocardial
21 ischemia such as ST-segment abnormalities or onset of exertional angina. These effects
22 may not be apparent to the individual, yet may still increase the risk of a number of
23 cardiac events, including myocardial infarction and sudden death. Thus, small changes in
24 physiological measures may not appear to be clearly adverse when considered alone, but
25 may be a part of a coherent and biologically plausible chain of related health outcomes
26 that range up to responses that are very clearly adverse, such as hospitalization or
27 mortality.
b. Evaluating Adversity of Ecological Effects
28 Adversity of ecological effects can be understood in terms ranging in biological level of
29 organization; from the cellular level to the individual organism and to the population,
30 community, and ecosystem levels. In the context of ecology, a population is a group of
31 individuals of the same species, and a community is an assemblage of populations of
32 different species that inhabit an area and interact with one another. An ecosystem is the
33 interactive system formed from all living organisms and their abiotic (physical and
34 chemical) environment within a given area (IPCC. 2007). The boundaries of what could
35 be called an ecosystem are somewhat arbitrary, depending on the focus of interest or
November 2013 Iv DRAFT: Do Not Cite or Quote
-------
1 study. Thus, the extent of an ecosystem may range from very small spatial scales to,
2 ultimately, the entire Earth (IPCC. 2007V
3 Effects on an individual organism are generally not considered to be adverse to public
4 welfare. However if effects occur to enough individuals within a population, then
5 communities and ecosystems may be disrupted. Changes to populations, communities,
6 and ecosystems can in turn result in an alteration of ecosystem processes. Ecosystem
7 processes are defined as the metabolic functions of ecosystems including energy flow,
8 elemental cycling, and the production, consumption and decomposition of organic matter
9 (U.S. EPA. 2002). Growth, reproduction, and mortality are species-level endpoints that
10 may be clearly linked to community and ecosystem effects and are considered to be
11 adverse when negatively affected. Other endpoints such as changes in behavior and
12 physiological stress can decrease ecological fitness of an organism, but are harder to link
13 unequivocally to effects at the population, community, and ecosystem level. Support for
14 consideration of adversity beyond the species level by making explicit the linkages
15 between stress-related effects at the species and effects at the ecosystem level is found in
16 A Framework for Assessing and Reporting on Ecological Condition: an SAB report (U.S.
17 EPA. 2002). Additionally, the National Acid Precipitation Assessment Program
18 (TSfAPAP. 1991) uses the following working definition of "adverse ecological effects" in
19 the preparation of reports to Congress mandated by the Clean Air Act: "any injury
20 (i.e., loss of chemical or physical quality or viability) to any ecological or ecosystem
21 component, up to and including at the regional level, over both long and short terms."
22 Beyond the level of species-level impacts, consideration of ecosystem services allows for
23 evaluation of how pollutant exposure may adversely impact species or processes of
24 particular economic or cultural importance to humans. On a broader scale, ecosystem
25 services may provide indicators for ecological impacts. Ecosystem services are the
26 benefits that people obtain from ecosystems (UNEP. 2003). According to the Millennium
27 Ecosystem Assessment, ecosystem services include: "provisioning services such as food
28 and water; regulating services such as regulation of floods, drought, land degradation,
29 and disease; supporting services such as soil formation and nutrient cycling; and cultural
30 services such as recreational, spiritual, religious and other nonmaterial benefits." For
31 example, a more subtle ecological effect of pollution exposure may result in a clearly
32 adverse impact on ecosystem services if it results in a population decline in a species that
33 is recreationally or culturally important.
November 2013 Ivi DRAFT: Do Not Cite or Quote
-------
References for Preamble
ATS (American Thoracic Society). (2000b). What constitutes an adverse health effect of air pollution? Am
J Respir Crit Care Med 161: 665-673. http://dx.doi.Org/10.1164/ajrccm.161.2.ats4-00
Bell. ML; Dominici. F; Samet. JM. (2005). A meta-analysis of time-series studies of ozone and mortality
with comparison to the national morbidity, mortality, and air pollution study. Epidemiology 16: 436-
445. http://dx.doi.org/10.1097/01.ede.0000165817.40152.85
CAA. Clean Air Act, as amended by Pub. L. No. 101-549. section 108: Air quality criteria and control
techniques. 42 USC § 7408 (1990a). http://www.law.cornell.edu/uscode/text/42/7408
CAA. Clean Air Act, as amended by Pub. L. No. 101-549. section 109: National primary and secondary
ambient air quality standards. 42 USC § 7409 (1990b). http://www.epa.gov/air/caa/title 1 .html#ia
CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A report
of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
http ://www. surgeongeneral. gov/library/smokingconsequences/
Fox. GA. (1991). Practical causal inference for ecoepidemiologists. JToxicol Environ Health A 33: 359-
373. http://dx.doi.org/10.1080/15287399109531535
Gee. GC: Pavne-Sturges. DC. (2004). Environmental health disparities: A framework integrating
psychosocial and environmental concepts [Review]. Environ Health Perspect 112: 1645-1653.
http://dx.doi.org/10.1289/ehp.7074
HEW (U.S. Department of Health, Education and Welfare). (1964). Smoking and health: Report of the
advisory committee to the surgeon general of the public health service. Washington, DC: U.S.
Department of Health, Education, and Welfare.
http://proFiles.nlm.nih.gov/ps/retrieve/ResourceMetadata/NNBBMQ
Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.
IARC (International Agency for Research on Cancer). (2006). Preamble to the IARC monographs. Lyon,
France. http://monographs.iarc.fr/ENG/Preamble/
loannidis. JPA. (2008). Why most discovered true associations are inflated [Review]. Epidemiology 19:
640-648. http://dx.doi.org/10.1097/EDE.Ob013e31818131e7
IOM (Institute of Medicine). (2008). Improving the presumptive disability decision-making process for
veterans. In JM Samet; CC Bodurow (Eds.). Washington, DC: National Academies Press.
http://www.nap.edu/openbook.php7record id=l 1908
IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Impacts, adaptation and
vulnerability. Cambridge, UK: Cambridge University Press, http://www.ipcc.ch/ipccreports/ar4-
wg2.htm
NAPAP (National Acid Precipitation Assessment Program). (1991). The experience and legacy of NAPAP:
Report of the Oversight Review Board of the National Acid Precipitation Assessment Program.
Washington, DC.
Rothman. KJ; Greenland. S. (1998). Modern epidemiology (2nd ed.). Philadelphia, PA: Lippincott,
Williams, & Wilkins.
U.S. EPA (U.S. Environmental Protection Agency). (1998a). Guidelines for ecological risk assessment
[EPA Report]. (EPA/630/R-95/002F). Washington, DC. http://www.epa.gov/raf/publications/guidelines-
ecological-risk-assessment.htm
November 2013 Ivii DRAFT: Do Not Cite or Quote
-------
U.S. EPA (U.S. Environmental Protection Agency). (2002). A framework for assessing and reporting on
ecological condition: An SAB report [EPA Report]. (EPA-SAB-EPEC-02-009). Washington, DC.
http://www.ntis.gov/search/product. aspx?ABBR=PB2004100741
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment [EPA
Report]. (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum, http://www.epa.gov/cancerguidelines/
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
UNEP (United Nations Environment Programme). (2003). Ecosystems and human well-being: A
framework for assessment. Washington, DC: Island Press.
Zeger. SL; Thomas. D; Dominici. F; Samet. JM; Schwartz. J; Dockery. D; Cohen. A. (2000). Exposure
measurement error in time-series studies of air pollution: Concepts and consequences. Environ Health
Perspect 108: 419-426.
November 2013 Iviii DRAFT: Do Not Cite or Quote
-------
PREFACE
Legislative Requirements for the Primary NAAQS Review
1 Two sections of the Clean Air Act (CAA) govern the establishment and revisions of the
2 National Ambient Air Quality Standards (NAAQS). Section 108 (42:U.S.C.:7408) directs
3 the Administrator to identify and list certain air pollutants and then to issue air quality
4 criteria for those pollutants. The Administrator is to list those air pollutants that in her
5 "judgment, cause or contribute to air pollution which may reasonably be anticipated to
6 endanger public health or welfare;" "... the presence of which in the ambient air results
7 from numerous or diverse mobile or stationary sources;" and "... for which ... [the
8 Administrator] plans to issue air quality criteria ..." (CAA. 1990a). Air quality criteria
9 are intended to "accurately reflect the latest scientific knowledge useful in indicating the
10 kind and extent of all identifiable effects on public health or welfare, which may be
11 expected from the presence of [a] pollutant in the ambient air ..." (42:11.S.C.:7408([b]).
12 Section 109 (42:U.S.C.:7409) (CAA. 1990b) directs the Administrator to propose and
13 promulgate "primary" and "secondary" NAAQS for pollutants for which air quality
14 criteria are issued. Section 109(b)(l) defines a primary standard as one "the attainment
15 and maintenance of which in the judgment of the Administrator, based on such criteria
16 and allowing an adequate margin of safety, are requisite to protect the public health."1
17 The legislative history of Section 109 indicates that a primary standard is to be set at
18 "... the maximum permissible ambient air level ... which will protect the health of any
19 [sensitive] group of the population," and that for this purpose "... reference should be
20 made to a representative sample of persons comprising the sensitive group rather than to
21 a single person in such agroup ..." (s. Rep. No. 91:1196, 91st Cong., 2d Sess. 10
22 [1970]). A secondary standard, as defined in Section 109(b)(2), must "specify a level of
23 air quality the attainment and maintenance of which, in the judgment of the
24 Administrator, based on such criteria, is requisite to protect the public welfare from any
25 known or anticipated adverse effects associated with the presence of [the] air pollutant in
26 the ambient air."2
1 The legislative history of Section 109 indicates that a primary standard is to be set at".. .the maximum permissible
ambient air level.. .which will protect the health of any [sensitive] group of the population," and that for this purpose
"... reference should be made to a representative sample of persons comprising the sensitive group rather than to a
single person in such a group..." (s. Rep. No. 91:1196, 91st Cong., 2d Sess. 10 [1970]).
2 Welfare effects (as defined in Section 302(h); 42:U.S.C.:7602[h]) include, but are not limited to, "...effects on
soils, water, crops, vegetation, man-made materials, animals, wildlife, weather ,visibility and climate, damage to and
deterioration of property, and hazards to transportation, as well as effects on economic values and on personal
comfort and well-being"(CAA. 2005).
November 2013 lix DRAFT: Do Not Cite or Quote
-------
1 The requirement that primary standards provide an adequate margin of safety was
2 intended to address uncertainties associated with inconclusive scientific and technical
3 information available at the time of standard setting. It was also intended to provide a
4 reasonable degree of protection against hazards that research has not yet identified. See
5 Lead Industries Association v. EPA, 647 F.2d 1130, 1154 (D.C. Cir 1980); American
6 Petroleum Institute v. Costle, 665 F.2d 1176, 1186 (D.C. Cir. \9%\); American Farm
7 Bureau Federation v. EPA, 559 F. 3d 512, 533 (D.C. Cir. 2009); Association of Battery
8 Recyclers v. EPA, 604 F. 3d 613, 617-18 (D.C. Cir. 2010). Both kinds of uncertainty are
9 components of the risk associated with pollution at levels below those at which human
10 health effects can be said to occur with reasonable scientific certainty. Thus, in selecting
11 primary standards that provide an adequate margin of safety, the Administrator is seeking
12 not only to prevent pollution levels that have been demonstrated to be harmful but also to
13 prevent lower pollutant levels that may pose an unacceptable risk of harm, even if the risk
14 is not precisely identified as to nature or degree. The CAA does not require the
15 Administrator to establish a primary NAAQS at a zero-risk level or at background
16 concentration levels, see Lead Industries Association v. EPA, 647 F.2d at 1156 n.51, but
17 rather at a level that reduces risk sufficiently so as to protect public health with an
18 adequate margin of safety.
19 In addressing the requirement for an adequate margin of safety, the EPA considers such
20 factors as the nature and severity of the health effects involved, the size of at-risk
21 population(s), and the kind and degree of the uncertainties that must be addressed. The
22 selection of any particular approach to providing an adequate margin of safety is a policy
23 choice left specifically to the Administrator's judgment. See Lead Industries Association
24 v. EPA, 647 F.2d at 1161-1162; Whitman v. American Trucking Associations, 531 U.S.
25 457,495(2001).
26 In setting standards that are "requisite" to protect public health and welfare as provided in
27 Section 109(b), EPA's task is to establish standards that are neither more nor less
28 stringent than necessary for these purposes. In so doing, EPA may not consider the costs
29 of implementing the standards. See generally, Whitman v. American Trucking
30 Associations, 531 U.S. 457, 465-472, 475-476 (2001. Likewise, "[a]ttainability and
31 technological feasibility are not relevant considerations in the promulgation of national
32 ambient air quality standards." American Petroleum Institute v. Costle, 665 F. 2d at 1185.
33 Section 109(d)(l) requires that "not later than December 31, 1980, and at 5-year intervals
34 thereafter, the Administrator shall complete a thorough review of the criteria published
35 under section 108 and the national ambient air quality standards... and shall make such
36 revisions in such criteria and standards and promulgate such new standards as may be
37 appropriate..." Section 109(d)(2) requires that an independent scientific review
November 2013 Ix DRAFT: Do Not Cite or Quote
-------
1 committee "shall complete a review of the criteria... and the national primary and
2 secondary ambient air quality standards... and shall recommend to the Administrator any
3 new... standards and revisions of existing criteria and standards as may be
4 appropriate...." Since the early 1980s, this independent review function has been
5 performed by the Clean Air Scientific Advisory Committee (CASAC).1
Introduction to the NAAQS for Nitrogen Dioxide (NO2)
6 NAAQS comprise four basic elements: indicator, averaging time, level, and form. The
7 indicator defines the pollutant to be measured in the ambient air for the purpose of
8 determining compliance with the standard. The averaging time defines the time period
9 over which air quality measurements are to be obtained and averaged or cumulated,
10 considering evidence of effects associated with various time periods of exposure. The
11 level of a standard defines the air quality concentration used (i.e., an ambient
12 concentration of the indicator pollutant) in determining whether the standard is achieved.
13 The form of the standard specifies the air quality measurements that are to be used for
14 compliance purposes and whether the statistic is to be averaged across multiple years
15 (e-g-, the annual fourth-highest daily maximum 8-hour concentration, averaged over 3
16 years). These four elements together determine the degree of public health and welfare
17 protection afforded by the NAAQS.
18 Nitrogen dioxide (NO2) is the indicator for a broad category of oxides of nitrogen. As
19 specified in Section 108(c) of the CAA (42:U.S.C.21:7408(c)), EPA considers the term
20 oxides of nitrogen to refer to all forms of oxidized nitrogen including multiple gaseous
21 species (e.g., NO2, nitric oxide [NO]) and particulate species (e.g., nitrates). EPA has
22 evaluated the atmospheric chemistry, exposure, and health effects associated with
23 nitrogen compounds present in particulate matter (PM) in the Agency's review of the
24 NAAQS for particulate matter (PM). Thus, the review of the NAAQS for NO2 focuses on
25 the gaseous oxides of nitrogen.
History of the Review of Air Quality Criteria for the Oxides of
Nitrogen and the NAAQS for NO2
26 On April 30, 1971 the EPA initially promulgated primary and secondary NAAQS for
27 NO2, under section 109 of the Act (36 FR 8186). Both primary and secondary standards
28 were set at 0.053 parts per million (ppm), annual average. The standards were based on
29 scientific information contained the 1971 Air Quality Criteria Document for Nitrogen
30 Oxides (U.S. EPA, 1971). Since then, the Agency has completed multiple reviews of the
1 Lists of CASAC members and of members of the CASAC Oxides of Nitrogen Primary NAAQS Review Panel are
available at: http://vosemite.epa.gov/sab/sabproduct.nsfAVebCASAC/CommitteesandMembership7OpenDocument.
November 2013 Ixi DPvAFT: Do Not Cite or Quote
-------
air quality criteria upon which the NAAQS are set and the standards themselves.
Table IV provides a brief summary of these reviews.
Table IV History of the National Ambient Air Quality Standards for NO2 during
the Period 1971-2012.
Final Rule
1971
36 FR 81 86
Apr, 30, 1971
1985
50 FR 25532
Jun 19, 1985
1996
61 FR 52852
Oct8, 1996
2010
FR 74 34404
Feb9, 2010
2012
77 FR 2021 8
Apr3, 2012
Primary/Secondary
Primary and
Secondary
Indicator
NO2
Averaging
Time
Annual
Level
53 ppb1
Form
Annual arithmetic average
Primary and secondary NO2 standards retained, without revision.
Primary and secondary NC>2 standards retained, without revision.
Primary
Secondary
NO2
1-hour
100 ppb
98th percentile,
averaged over 3 years2
Primary annual NO2 standard retained, without revision.
Secondary annual standard retained, without revision.
3
4
5
6
7
8
9
10
11
12
The EPA retained the primary and secondary NO2 standards, without revision, in reviews
completed in 1985 and 1996 (50 FR 25532, June 19, 1985; 61 FR 52852, October 8,
1996). These decisions were informed by scientific information contained in the 1982 Air
Quality Criteria Document for Oxides of Nitrogen (U.S. EPA. 1982) which updated the
scientific criteria upon which the initial NO2 standards were based and the 1993 Air
Quality Criteria Document for the Oxides of Nitrogen (U.S. EPA. 1993).
The most recent review of the air quality criteria for oxides of nitrogen (health criteria)
and the primary NO2 standard was initiated in December 2005 (70 FR 73236,
December 9, 2005).3 The Agency's plans for conducting the review were contained the
Integrated Review Plan for the Primary National Ambient Air Quality Standard for NO2
1 The initial standard level of the annual NO2 standard was 0.053 ppm, which is equal to 53 ppb.
2 The form of the 1 -hour standard is the 3 -year average of the 98th percentile of the yearly distribution of 1 -hour
daily maximum NO2 concentrations.
3 Documents related to reviews completed in 2010 and 1996 are available at:
http://www.epa.gOv/ttn/naaqs/standards/nox/s nox index.html.
November 2013
Ixii
DRAFT: Do Not Cite or Quote
-------
1 (U.S. EPA. 2007a)1 which included consideration of comments received during a
2 CASAC consultation as well as public comment on a draft IRP. The science assessment
3 for the review was described in the 2008 Integrated Science Assessment for Oxides of
4 Nitrogen - Health Criteria (U.S. EPA. 2008c). multiple drafts of which received review
5 by CASAC and the public. The EPA also conducted quantitative human risk and
6 exposure assessments, after consultation with CASAC and receiving public comment on
7 a draft analysis plan (U.S. EPA. 2007b). These technical analyses were presented in the
8 Risk and Exposure Assessment (REA) to Support the Review of the NO2 Primary
9 National Ambient Air Quality Standard (U.S. EPA. 2008d). multiple drafts of which
10 received CASAC and public review.
11 In the course of reviewing the second draft REA, CASAC expressed the view that the
12 document would be incomplete without the addition of a policy assessment chapter
13 presenting an integration of evidence-based considerations and risk and exposure
14 assessment results. CASAC stated that such a chapter would be "critical for considering
15 options for the NAAQS for NO2" (Samet 2008). In addition, within the period of
16 CASAC review's of the second draft REA, the EPA's Deputy Administrator indicated in
17 a letter to the chair of CASAC chair, addressing earlier CASAC comments on the
18 NAAQS review process, that the risk and exposure assessment will include "a broader
19 discussion of the science and how uncertainties may effect decisions on the standard" and
20 "all analyses and approaches for considering the level of the standard under review,
21 including risk assessment and weight of evidence methodologies" (Peacock. 2008).
22 Accordingly, the final REA included a new policy assessment chapter. This policy
23 assessment chapter considered the scientific evidence in the ISA and the exposure and
24 risk characterization results presented in other chapters of the REA as they related to the
25 adequacy of the then current primary NO2 standard and potential alternative primary
26 standards for consideration.2 CASAC discussed the final version of the REA, with an
27 emphasis on the policy assessment chapter during a public teleconference on
28 December 5, 2008. Following that teleconference, CASAC offered comments and advice
29 on the primary NO2 standard in a letter to the Administrator (Samet 2008).
30 At the time of the last review, the epidemiological evidence had grown substantially since
31 the completion of the 1995 Staff Paper, with the addition of field and panel studies,
32 intervention studies, and time-series studies of effect (75 FR 6488). After considering an
1 The EPA conducted a separate review of the secondary NO2 NAAQS jointly with a review of the secondary sulfur
dioxide (SO2) NAAQS. The Agency retained those secondary standards, without revision, to address the direct
effects on vegetation of exposure to gaseous oxides of nitrogen and sulfur (77 FR 20218, April 3, 2012).
2 Subsequent to the completion of the 2008 NO2 REA, EPA Administrator Jackson called for key changes to the
NAAQS review process including reinstating a policy assessment document that contains staff analysis of the
scientific bases for alternative policy options for consideration by senior Agency management prior to rulemaking
(Jackson. 2009).
November 2013 Ixiii DRAFT: Do Not Cite or Quote
-------
1 integrative synthesis of the entire body of this evidence on human health effects
2 associated with presence of NO2 in the air, the Administrator determined that the existing
3 primary NO2 NAAQS, based on an annual arithmetic average, was not sufficient to
4 protect the public health from the array of effects that could occur following short-term
5 exposures to ambient NO2. The Administrator noted a particular concern with the
6 potential for adverse health effects following exposures to elevated NO2 concentrations
7 that can occur around major roads. Id at 6482. On July 15, 2009, the EPA proposed to
8 supplement the existing annual, primary standard for NO2 by establishing a new short-
9 term standard (75 FR 34404). On February 9, 2010, the EPA finalized a new short-term
10 standard with a level of 100 ppb, based on the 3-year average of the 98th percentile of the
11 yearly distribution of 1-hour daily maximum concentrations. The EPA also retained the
12 existing primary annual NO2 standard with a level of 53 ppb (75 FR 6474).
13 Revisions to the NAAQS were accompanied by revisions to the data handling
14 procedures, the ambient air monitoring and reporting requirements, and the Air Quality
15 Index (AQI).1 One aspect of the new monitoring network requirements included
16 requirements for States to locate monitors within 50 meters of major roadways in large
17 urban areas, and in other locations maximum NO2 concentrations can occur. Subsequent
18 to the 2010 rulemaking, the EPA revised the deadlines by which the near-road monitors
19 are to be operational in order to implement a phased deployment approach (78 FR 16184,
20 March 14, 2013). The near-road NO2 monitors will become operational between January
21 1, 2014 and January 1, 2017.
1 The current federal regulatory measurement methods for NO2 are specified in 40 CFR part 50, Appendix F and 40
CFR part 53. Consideration of ambient air measurements with regard to judging attainment of the standards is
specified in 40 CFR part 50, Appendix S. The NO2 monitoring network requirements are specified in 40 CFR part
58, Appendix D, section 4.3. The EPA revised the AQI for NO2 to be consistent with the revised primary NO2
NAAQS as specified in 40 CFR part 58 Appendix G. Guidance on the approach for implementation of the new
standards was described in the Federal Register notices for the proposed and final rules (74 FR 34404; 75 FR 6474).
November 2013 Ixiv DRAFT: Do Not Cite or Quote
-------
References for Preface
CAA. Clean Air Act as amended by Pub. L. No. 101-549. section 108: Air quality criteria and control
techniques. 42 USC § 7408 (1990a). http://www.law.cornell.edu/uscode/text/42/7408
CAA. Clean Air Act, as amended by Pub. L. No. 101-549. section 109: National primary and secondary
ambient air quality standards. 42 USC § 7409 (1990b). http://www.epa.gov/air/caa/title 1 .html#ia
CAA. Clean Air Act, section 302: Definitions. 42 USC § 7602 (2005).
http://www.gpo.gov/fdsvs/pkg/USCODE-2005-title42/pdf/USCODE-2005-title42-chap85-subchapIII-
sec7602.pdf
Jackson. LP. (2009). Letter from EPA Administrator Lisa P. Jackson to Dr. Jonathan M. Samet. Subject:
Changes to the NAAQS review process. Available online at
http://vosemite.epa.gov/sab/sabproduct.nsf/WebCASAC/Jackson%2005-21-
09/$File/NAAOS%20Letter%20to%20CASAC%20Chair-Mav%202009.pdf
Peacock. MC. (2008). Letter from Marcus C. Peacock to Rogene Henderson (regarding CASAC comments
on application of new NAAQS process). Available online
Samet. JM. (2008). Letter from Dr. Jonathan M. Samet to The Honorable Stephen L. Johnson. Subject:
Clean Air Scientific Advisory Committees (CASAC) peer review of draft chapter 8 of EPAs risk and
exposure assessment to support the review of the NO2 primary National Ambient Air Quality Standard.
Available online
U.S. EPA (U.S. Environmental Protection Agency). (1971). Air quality criteria for nitrogen oxides [EPA
Report]. (AP-84). Washington DC. http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=20013K3B.txt
U.S. EPA (U.S. Environmental Protection Agency). (1982). Air quality criteria for oxides of nitrogen [EPA
Report]. (EPA/600/8-82-026). Research Triangle Park, NC.
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2007a). Integrated review plan for the Primary
National Ambient Air Quality Standard for Nitrogen Dioxide [EPA Report]. Research Triangle Park,
NC.
U.S. EPA (U.S. Environmental Protection Agency). (2007b). Nitrogen dioxide health assessment plan:
Scope and methods for exposure and risk assessment. Draft [EPA Report]. Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
http://www.epa.gov/ttn/naaqs/standards/nox/data/20070927 risk exposure scope.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2008d). Risk and exposure assessment to support the
review of the NO2 primary national ambient air quality standard. (EPA-452/R-08-008a). Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, http://www.epa.gov/ttn/naaqs/standards/nox/data/20081121 NO2 REA final.pdf
November 2013 Ixv DRAFT: Do Not Cite or Quote
-------
EXECUTIVE SUMMARY
Purpose of the Integrated Science Assessment
1 This Integrated Science Assessment (ISA) is a comprehensive evaluation and synthesis of
2 the most policy-relevant science that serves as the scientific foundation for the review of
3 the primary (health-based) National Ambient Air Quality Standards (NAAQS) for
4 nitrogen dioxide (NO2). The Clean Air Act requires the Environmental Protection
5 Agency (EPA), every five years, to review the NAAQS including the science upon which
6 the NAAQS are based. Scientific information and conclusions presented in the ISA guide
7 the development of quantitative assessments in EPA's Risk and Exposure Assessment (if
8 one is warranted). Information in the ISA, and in the Risk and Exposure Assessment, is
9 integrated and interpreted in the EPA Policy Assessment to frame the broadest range of
10 policy options that can be supported by the available scientific and technical information
11 to guide decisions made by the EPA Administrator on whether to retain or revise the
12 NAAQS (see Figure I, Preamble to the ISA).
13 The most recent review of the primary NAAQS for NO2 was completed in 2010. EPA
14 retained the annual standard with a level of 53 parts per billion (ppb), annual average
15 (avg) concentration, to protect against health effects potentially associated with long-term
16 NO2 exposures. EPA established a new 1-hour (h) standard at a level of 100 ppb, based
17 on the 3-year average of the 98th percentile of yearly 1-h daily maximum (max)
18 concentrations. The 1-hour standard was established to protect against a broad range of
19 respiratory effects associated with short-term NO2 exposures in potential at-risk
20 populations such as people with asthma and people who spend time on or near major
21 roadways. EPA also set requirements for a monitoring network that includes monitors
22 near major roadways, where maximum NO2 concentrations are expected to occur.
Scope and Methods
23 Oxides of nitrogen are one of six criteria pollutants for which EPA has established
24 NAAQS. Oxides of nitrogen include all oxidized nitrogen compounds, including gases
25 such as NO2 and nitric oxide (NO)1 and several particle species (Figure 2-1. Section
26 2.2)2. The NAAQS are specified in terms of the indicator NO2. As this ISA serves as the
27 scientific foundation for the review of the primary NO2 NAAQS, it evaluates scientific
28 information on the atmospheric chemistry and human exposure to the gaseous forms of
29 oxides of nitrogen and associated health effects only. The particle species (e.g., nitrates)
Gaseous oxides of nitrogen also include NOX, which refers to the sum of NO2 and NO.
2Section 108(c) of the Clear Air Act refers to oxides of nitrogen as all forms of oxidized nitrogen including multiple
gaseous and paniculate species. 42. U.S.C. 21 7408(c).
November 2013 Ixvi DRAFT: Do Not Cite or Quote
-------
1 were most recently examined in the 2009 ISA for Particulate Matter (PM) (U.S. EPA.
2 2009a). The ecological and other welfare effects of oxides of nitrogen are being
3 considered in a separate assessment conducted as part of the review of the secondary
4 (welfare-based) NAAQS for NO2 and sulfur dioxide (SO2) (U.S. EPA. 2013).
5 EPA uses a structured and robust process for evaluating available scientific information
6 and drawing conclusions about the health effects associated with air pollution exposure.
7 This process includes criteria for identifying and selecting relevant recent peer-reviewed
8 literature published since the previous ISA (i.e., studies published starting in 2008 for this
9 ISA), assessing quality of scientific information, and integrating information. EPA also
10 uses a consistent and transparent framework for drawing conclusions about the causal
11 nature of air pollution-related health effects in the ISA (Section ,5, Preamble to the ISA).
12 Evidence is integrated across epidemiologic, controlled human exposure, and
13 toxicological studies and across related outcomes to make a determination about
14 causation, not just association. Conclusions are formed by synthesizing findings from
15 studies reviewed in previous assessments and recent studies. Based on judgments of
16 aspects such as the consistency, coherence, and biological plausibility of observed effects
17 (i.e., evidence for the direct effect on a health outcome or key events that inform the
18 mode of action), the evidence is classified according to a five-level hierarchy:
19 • Causal relationship
20 • Likely to be a causal relationship
21 • Suggestive of a causal relationship
22 • Inadequate to infer a causal relationship
23 • Not likely to be a causal relationship
24 In addition to describing causal determinations, the ISA addresses policy-relevant issues
25 such as: (1) exposure concentrations, durations, and patterns associated with health
26 effects; (2) the concentration-response relationship(s), including information related to
27 identifying thresholds for effects; and (3) lifestages or populations at increased risk for
28 health effects related to exposure to oxides of nitrogen. In the evaluation of the scientific
29 information and policy-relevant issues, the ISA also describes uncertainties and
30 limitations in the scientific evidence base including the potential for NO2 to serve
31 primarily as an indicator of another ambient air pollutant or mixture.
November 2013 Ixvii DRAFT: Do Not Cite or Quote
-------
Sources of Oxides of Nitrogen to Human Exposure
1 Ambient concentrations of oxides of nitrogen are determined by the types and strength of
2 emissions sources present, the chemical transformations that occur in the atmosphere,
3 weather conditions, deposition to surfaces, and transport to other locations. Based on the
4 2008 National Emissions Inventory1, the major emissions source categories for the U.S.
5 as a whole are highway vehicles (39%), off-highway vehicles (19%), fuel combustion by
6 electric utilities (17%), and industrial fuel combustion (8%) (Section 2.3. Figure 2-2).
7 Smaller source categories include other industrial operations and microbial processes in
8 the soil. Specific sources that can affect local air quality include on-road vehicles,
9 airports, railyards, shipping ports, home wood burning, intense industrial and chemical
10 processes, activities for oil and gas development, and wildfires. NO2 transported from
11 continents outside of North America (i.e., North American Background [NAB])
12 contributes less than 1% to ambient concentrations in the U.S. (Section 2.5.6).
13 Direct emissions consist of mostly NO but also include some NO2. A major chemical
14 transformation in the air is the reaction of NO and ozone (O3) to form NO2 (Section 2.2.
15 Figure 2-1). Rather than direct emissions, this reaction is the main source of the ambient
16 air NO2 concentrations measured in most urban locations. Chemical reactions also
17 remove NO2 and NO from the atmosphere. For example, reactions between NO2 and
18 volatile organic compounds (VOCs) form O3. NO and NO2 also are transformed into
19 other oxides of nitrogen by reactions with reactive radical species and O3. Other major
20 mechanisms by which oxides of nitrogen are removed from the atmosphere are
21 deposition including impaction with surfaces, reactions with plants, diffusion into cloud
22 droplets, and washout in falling rain. Chemical transformations and deposition vary
23 according to several environmental factors such as season, time of day, air circulation
24 patterns, and ambient concentrations of other pollutants.
25 There are differences across locations and time in factors such as the sources present,
26 chemical reactions that occur, atmospheric conditions, and topography. Thus, ambient
27 concentrations of oxides of nitrogen are highly variable. This variability occurs across
28 spatial scales, including national, regional, urban, neighborhood, and microscale
29 environments (Section 2.5). Variability also occurs across time scales such as years,
30 seasons, days of the week, and time of day.
31 Much of the information on ambient concentrations of oxides of nitrogen in the U.S. is
32 for NO2 and comes from the State and Local Monitoring Air Stations Network of about
33 500 sites. This monitoring network serves many purposes: assessing compliance with the
34 NAAQS, providing the public with air pollution data in a timely manner, and supporting
Information on emissions sources will be updated in the Second External Review Draft of the ISA for Oxides of
Nitrogen with data from the 2011 National Emissions Inventory, which became available in November 2013.
November 2013 Ixviii DRAFT: Do Not Cite or Quote
-------
1 air pollution research studies. NO and NOX also are measured at many of these stations,
2 but other oxides of nitrogen are not routinely measured. Across the U.S. during
3 2009-2011, the mean and 99th percentile for 1-h daily maximum ambient NO2
4 concentrations were 20 ppb and 57 ppb, respectively (Table 2-1). During the same time
5 period, the mean and 99th percentile for annual average NO2 concentrations were 9.4 ppb
6 and 25 ppb, respectively (Table 2-2). NO2 concentrations are higher in urban areas than
7 in nonurban areas (Figure 2-10 and Figure 2-12, for annual average and seasonal average,
8 respectively). In microscale environments, ambient NO2, NO, and NOX concentrations
9 have been shown to be 30% to 200% higher at locations within 15 m of a roadway
10 (averaged over hours to weeks) compared with locations farther away from the road. The
11 nature of the gradient varies according to factors such as traffic volume, time of day,
12 meteorology, and local topography (Section 2.5.3). Concentrations can be higher in urban
13 street canyons or in locations where built or natural topographical features influence air
14 circulation.
15 With respect to time trends, annual average ambient NO2 concentrations in the U.S. as a
16 whole decreased by 48% from 1990 to 2012 (Figure 2-16). mainly due to decreases in
17 NOX emissions from on-road vehicles and electric utilities (Figure 2-3). On shorter time
18 scales, ambient NO2 concentrations typically are higher in winter than summer, on
19 weekdays than weekends, and during early mornings (corresponding with rush hour) than
20 other times of the day (including evening rush hour) (Section 2.5.4).
21 The variability in ambient NO2 concentrations observed across spatial and time scales as
22 specified above are important influences on human exposure to ambient NO2. Human
23 exposure to ambient NO2 is determined by the concentrations in various ambient
24 microenvironments, including in vehicles, and the time spent in those microenvironments
25 (Section 2.6.1). Another component of total exposure is indoor NO2 exposure, which is
26 influenced by indoor sources (e.g., gas stoves, gas heaters, oil furnaces, wood burning
27 stoves, kerosene heaters, smoking). Ambient concentrations penetrate indoors (Section
28 2.6.3.3). Ventilation characteristics (e.g., air conditioning, open windows) can affect the
29 amount of NO2 that penetrates indoors and thus contribute to variability in human
30 exposure to ambient NO2. Understanding human exposure to ambient NO2 and the
31 relationships between exposure and ambient concentrations is essential for interpreting
32 scientific information on relationships between ambient NO2 concentrations and health
33 effects and in turn, for informing the review of the primary NO2 NAAQS.
34 Many epidemiologic studies assess human exposure to NO2 using ambient concentrations
35 obtained from central site monitors. The siting of monitors does not cover all locations
36 where people live or spend their time. Ambient concentrations may not be available at the
37 microenvironment of interest (Section 2.6.5). for example near roads. Data from the near-
November 2013 Ixix DRAFT: Do Not Cite or Quote
-------
1 road monitoring network are not available yet. Thus, the use of ambient concentrations
2 obtained from central site monitors to represent human exposure is associated with
3 measurement error. The extent of exposure measurement error is influenced by the
4 relationships between central site ambient concentrations and personal exposure. Across
5 the population, there is wide variation in personal-ambient NO2 relationships, indicating
6 that there is heterogeneity among individuals in how well spatial and temporal variability
7 in ambient concentrations correlate to personal exposure, accounting for varying time-
8 activity patterns. Personal-ambient relationships may vary by age, season, and local
9 sources (Sections 2.6.4 and 2.6.5.1).
10 Exposure measurement error can have important implications for the relationships
11 observed between ambient NO2 concentrations and health effects. For example, some
12 studies estimated larger respiratory effects in association with more spatially-resolved
13 NO2 exposure estimates than with NO2 concentrations obtained from a single central site
14 monitor or averaged over area monitors (Sections 2.6.5.2 and 4.2.4.4). More spatially-
15 resolved estimates included personal exposure measures, outdoor school measurements,
16 and ambient concentrations at the nearest central site monitor. Health effects also are
17 associated with more spatially-resolved measures of long-term NO2 or NOX exposure
18 estimated from models. Measurement error is a component of the various exposure
19 assessment methods, and is influenced by the spatial and temporal variability in NO2
20 concentrations, time-activity patterns, air exchange characteristics of locations, and
21 accuracy and precision of instrumentation. Measurement methods of regulatory networks
22 are found to overestimate NO2 concentrations because of interference from other oxides
23 of nitrogen, but the impact typically is less than 10% in urban areas and near sources
24 (Section 2.4.1). New measurement methods are being developed. Exposure measurement
25 error may reduce the magnitude and/or precision (i.e., widen confidence intervals) of
26 health effect associations (Section 2.6.5.3). However, the effect of measurement error
27 resulting from the use of central site NO2 to represent near-road exposures is not known.
28 Associations between NO2 and health effects observed in epidemiologic studies may
29 represent an independent effect of NO2 or the effect of another air pollutant or mixture
30 that is related to both the health effect being examined and NO2 concentrations. A wide
31 range of correlations is reported between NO2 and copollutants such as O3, SO2,
32 particulate matter (PM2 5, PM^)1, elemental or black carbon (EC or BC), ultrafine
33 particles (UFP), and carbon monoxide (CO) (Table 2-4. Figure 2-19). Correlations
34 generally are higher for EC, UFP, and CO, which like NO2, are emitted from motor
35 vehicles. The relationship between NO2 and a given copollutant varies across locations
1 PM2 5: particulate matter with mean aerodynamic diameter less than or equal to a nominal 2.5 iim.
PM10: particulate matter with mean aerodynamic diameter less than or equal to a nominal 10 iim.
November 2013 Ixx DRAFT: Do Not Cite or Quote
-------
1 and exposure assessment methods, which indicates that there is variability in the potential
2 for a copollutant to bias (i.e., confound) associations between NO2 and health effects.
Health Effects of Oxides of Nitrogen
3 This ISA evaluates relationships between oxides of nitrogen and a broad range of
4 outcomes related to respiratory effects, cardiovascular effects, reproductive and
5 developmental effects, total mortality, and cancer as examined in epidemiologic,
6 controlled human exposure, and toxicological studies. For experimental studies, emphasis
7 is placed on studies with exposures that are relevant to human ambient exposures, defined
8 as concentrations no greater than 5,000 ppb, which is about one to two orders of
9 magnitude higher than peak concentrations of NO2, NO, or NOX that humans experience
10 on roads (Section 2.5.3). Causal determinations are based on evidence that is integrated
11 across scientific disciplines and related outcomes, including information about potential
12 modes of action. Relevant information on exposure and dosimetry also is considered.
13 Separate causal determinations are made for health effects related to short-term (minutes
14 up to 1 month, Chapter 4) and long-term (more than 1 month to years, Chapter 5)
15 exposures. Although the scope of this ISA includes all gaseous oxides of nitrogen,
16 information on the dosimetry, modes of action, and health effects is available for NO2,
17 NO, or NOX exposures. This may be explained, in part, by the little information available
18 on ambient exposures to other oxides of nitrogen. In this and the 2008 ISA for Oxides of
19 Nitrogen, the majority of scientific information is available for health effects related to
20 NO2 exposure; thus, causal determinations are formed only for NO2. Because there is at
21 least some biological plausibility for negative health effects resulting from ambient-
22 relevant NO2 exposures but not from ambient-relevant NO exposures, associations
23 between health effects and ambient NOX are considered to be reflecting associations with
24 NO2 and are considered in causal determinations for NO2.
25 The causal determinations from the 2008 ISA for Oxides of Nitrogen and this ISA are
26 presented in Table ES-1. Integrated with previous evidence, results from recent studies
27 are the basis for strengthening the causal determinations for all of the evaluated health
28 effect categories. In some cases, recent studies show health effects associated with NO2
29 exposure where previous results were inconsistent or showed no association. In other
30 cases, recent studies expand on previous supporting evidence. Most of the recent
31 literature base consists of epidemiologic studies. The previous body of controlled human
32 exposure and toxicological studies and the relatively small body of recent studies inform
33 the biological plausibility for health effects of NO2 exposure with information on the
34 direct effects on health outcomes or biological processes by which effects may occur. In
35 the 2008 ISA for Oxides of Nitrogen, a major uncertainty noted for the relationships
36 between NO2 exposure and several health effect categories was the difficulty in
November 2013 Ixxi DRAFT: Do Not Cite or Quote
-------
1 distinguishing whether the epidemiologic associations observed with ambient NO2
2 concentrations were independent of the effects of another traffic-related air pollutant or
3 mixture. This uncertainty regarding the potential for copollutant confounding or the
4 potential for NO2 to serve primarily as an indicator for another traffic-related pollutant or
5 mixture remains for some health effects. For other health effects, evidence from recent
6 studies reduces this uncertainty by indicating the independent effects of NO2 exposure.
November 2013 Ixxii DRAFT: Do Not Cite or Quote
-------
Table ES-1 Causal determinations for NO2 by exposure duration and health
effect category from the 2008 ISA and current draft ISA for oxides of
nitrogen.
Exposure Duration and Health
Effect Category3
Causal Determination13
Current Draft ISA for
2008 ISA for Oxides of Nitrogen Oxides of Nitrogen
Short-term NO2 Exposure
Respiratory Effects
Section 4.2, Table 4-23
Cardiovascular Effects
Section 4.3, Table 4-36
Total Mortality
Section 4.4, Table 4-41
Sufficient to Infer a Likely Causal
Relationship
Inadequate to Infer the Presence or
Absence of a Causal Relationship
Suggestive but not Sufficient to
Infer a Causal Relationship
Causal Relationship
Likely to be a Causal Relationship
Likely to be a Causal Relationship
Long-term NO2 Exposure
Respiratory Effects
Section 5.2, Table 5-9
Cardiovascular Effects
Section 5.3, Table 5-12
Reproductive and Developmental
Effects0
Sections 5.4.2, 5.4.3, 5.4.4, and
Table 5-1 5
Total Mortality
Suggestive but not Sufficient to
Infer a Causal Relationship
Inadequate to Infer the Presence or
Absence of a Causal Relationship
Inadequate to Infer the Presence or
Absence of a Causal Relationship
Inadequate to Infer the Presence or
Likely to be a Causal Relationship
Suggestive of a Causal Relationship
Fertility, Reproduction, and Pregnancy
Suggestive of a Causal Relationship
Birth Outcomes
Suggestive of a Causal Relationship
Postnatal Development
Suggestive of a Causal Relationship
Suggestive of a Causal Relationship
Section 5.5, Table 5-19
Absence of a Causal Relationship
Cancer
Section 5.6, Table 5-21
Inadequate to Infer the Presence or
Absence of a Causal Relationship
Suggestive of a Causal Relationship
aA spectrum of outcomes is evaluated as part of a broad health effect category including physiological measures (e.g., airway
responsiveness, lung function), clinical outcomes (e.g., respiratory symptoms, hospital admissions), and cause-specific mortality.
Total mortality includes all nonaccidental causes of mortality and is informed by the nature of the evidence for the spectrum of
morbidity effects (e.g., respiratory, cardiovascular) that can lead to mortality. The sections and tables referenced include a detailed
discussion of the available evidence that supports the causal determinations presented in the current draft ISA and the NO2
concentrations with which health effects have been associated. Summary information also is presented in Table 1-1.
bSince the completion of the 2008 ISA for Oxides of Nitrogen, the phrasing of causal determinations has changed slightly, and the
weight of evidence that describes each level in the hierarchy of the causal framework has been more explicitly characterized.
°ln the current draft ISA, reproductive and developmental effects are separated into smaller subcategories of outcomes based on
varying underlying biological processes and exposure patterns over different lifestages.
November 2013
Ixxiii
DRAFT: Do Not Cite or Quote
-------
1 Providing insight into the processes by which exposure to NO2 or NO can lead to health
2 effects are the dosimetry and modes of action of NO2 and NO. Once inhaled, NO2 leaves
3 the gas phase and enters the extracellular lining fluid of the lung. In the extracellular
4 lining fluid, NO2 readily reacts with substances such as antioxidants. The formation of
5 secondary oxidation products and the cascade of events that follow are likely responsible
6 for the health effects associated with NO2 exposure. These products can initiate oxidative
7 stress, alter permeability of the alveolar capillary barrier, and change the activity of
8 enzymes in the respiratory tract (Section 3.2.2). In the respiratory tract, the biological
9 responses demonstrated to occur with NO2 exposure include, but are not limited to,
10 initiation of inflammation, enhancement of bronchial smooth muscle reactivity,
11 modification of immune responses, and remodeling of airways (Section 3.3.2). These
12 observations provide support for the respiratory effects observed in relation to NO2
13 exposure across disciplines by providing insight into potential underlying biological
14 processes. The processes by which inhaled NO2 may lead to health effects outside the
15 respiratory system are not well characterized. Some products formed by reactions of
16 inhaled NO2 have been found in the blood of animal models but with higher than
17 ambient-relevant NO2 exposure concentrations. Limited evidence suggests that mediators
18 may spillover from the respiratory tract into the blood in response to NO2 exposure
19 (Section 3.3.2.8). which could explain cardiovascular effects.
20 Unlike NO2, NO is not transformed by reactions in the extracellular lining fluid of the
21 respiratory tract and can diffuse into the blood. However, NO is formed naturally in the
22 body from biological processes and from nitrates and nitrites that are present in foods
23 consumed. It is not clear whether ambient-relevant concentrations of inhaled NO alter
24 biological processes that are affected by endogenous or diet-derived NO. Thus, it is not
25 clear by what processes inhaled NO may induce health effects (Section 3.3.3).
Health Effects Associated with Short-term NO2 Exposure
26 Across the array of health effects evaluated in this ISA, evidence is more robust for
27 health effects associated with short-term NO2 exposure than long-term NO2 exposure.
28 For several outcomes related to respiratory and cardiovascular effects and for total
29 mortality, epidemiologic associations are more consistently observed for short-term than
30 long-term NO2 exposure. Differences in causal determinations for these health effects
31 relate to the extent to which there is biological plausibility for the effects of NO2
32 exposure.
33 The strongest evidence is for respiratory effects, and it indicates that there is a
34 causal relationship with short-term NO2 exposure (Section 4.2.9). This conclusion is
November 2013 Ixxiv DRAFT: Do Not Cite or Quote
-------
1 based on the consistency, coherence, and biological plausibility of evidence integrated
2 across epidemiologic, controlled human exposure, and animal toxicological studies
3 indicating increases in asthma exacerbations. Epidemiologic studies consistently show
4 associations between short-term increases in ambient NO2 concentration and increases in
5 hospital admissions and emergency department (ED) visits for asthma. Associations also
6 are found with respiratory symptoms, pulmonary inflammation, and decreases in lung
7 function in children with asthma. Epidemiologic associations are demonstrated in studies
8 conducted in diverse geographical locations and using varied designs, including multicity
9 analyses. Evidence from controlled human exposure and animal toxicological studies for
10 NO2-induced increases in airway responsiveness in adults with asthma and increases in
11 allergic inflammation and oxidative stress demonstrate that the effects of NO2 exposure
12 on asthma exacerbations are biologically plausible. NO2 exposure showed effects on
13 many of the same endpoints in humans and experimental animals, indicating similar
14 mechanisms across species. There also is evidence for NO2-related increases in
15 respiratory infection and chronic obstructive pulmonary disease (COPD) exacerbations,
16 but inconsistencies are found across scientific disciplines or related outcomes. Recent
17 epidemiologic studies also support the effects of ambient NO2 exposure on a continuum
18 of respiratory outcomes by demonstrating associations with respiratory mortality.
19 Much of the evidence described above was available in the 2008 ISA for Oxides of
20 Nitrogen. In the current ISA, the causal determination is strengthened from likely to be a
21 causal relationship to causal relationship because the recent epidemiologic evidence
22 reduces the previously identified uncertainty regarding confounding by other traffic-
23 related pollutants. Recent epidemiologic studies add to the evidence that associations of
24 ambient NO2 with asthma-related effects and other respiratory effects remain positive in
25 copollutant models that statistically adjust for the effects of another pollutant such as
26 PMio, PM2 5, SO2, O3, or examined in fewer studies, CO, EC, BC, or UFP (e.g., Figure
27 4-10 and Figure 4-11). Further, previous and recent studies demonstrate associations of
28 indoor NO2 with respiratory symptoms in children with asthma. The epidemiologic
29 evidence and the experimental evidence together provide sufficient evidence that short-
30 term NO2 exposure has an independent, causal relationship with respiratory effects.
31 For cardiovascular effects, there is likely to be a causal relationship with short-term
32 NO2 exposure (Section 4.3.9) based strongly on associations consistently found between
33 short-term increases in ambient NO2 concentration and hospital admissions for ischemic
34 heart disease (IHD) and cardiovascular mortality. The conclusion is supported by the
35 concurrence between findings for cardiovascular mortality and hospital admissions for
36 IHD, which is a leading cause of death. Recent epidemiologic studies reduce previous
37 uncertainty regarding copollutant confounding by demonstrating that NO2-associated
38 cardiovascular hospital admissions and mortality remain positive in copollutant models
November 2013 Ixxv DRAFT: Do Not Cite or Quote
-------
1 with PM10, SO2, O3, or in some but not all locations, PM2 5 or CO (e.g., Figure 4-15 and
2 Figure 4-16). Recent findings from epidemiologic and controlled human exposure studies
3 for NO2-related decreases in heart rate variability and changes in ventricular
4 repolarization provide some biological plausibility. Findings from some experimental
5 studies for NO2-induced increases in inflammation and oxidative stress describe effects
6 on other key events that inform the modes of action for IHD and cardiovascular
7 mortality. However, since experimental evidence is inconsistent and analysis of
8 confounding does not include the array of potentially correlated copollutants, the
9 collective evidence is not sufficient to conclusively demonstrate the independent
10 cardiovascular effects of NO2 exposure.
11 For total mortality, there is likely to be a causal relationship with short-term NO2
12 exposure (Section 4.4.8) based on associations consistently observed between short-term
13 increases in ambient NO2 concentration and increases in mortality from all nonaccidental
14 causes. Recent epidemiologic studies that pool data across various cities demonstrate the
15 robustness of association and reduce previous uncertainty regarding confounding by SO2,
16 PMio, or O3. The findings for NO2-related increases in cardiovascular hospital
17 admissions provide some understanding of the biological processes by which NO2
18 exposure may lead to mortality. Cardiovascular diseases account for a large portion of
19 mortality (e.g., 35% in the U.S.). However, the limited findings for effects on measures
20 of cardiovascular physiology produce some uncertainty regarding the spectrum of
21 cardiovascular effects that NO2 exposure may induce to lead to mortality. The robust
22 evidence for the effects of NO2 on asthma exacerbations but limited evidence for effects
23 on COPD exacerbations and respiratory infection produces some uncertainty regarding
24 the spectrum of respiratory effects that NO2 exposure may induce to lead to mortality.
25 The limited biological plausibility and limited analysis of potential copollutant
26 confounding are not sufficient to conclusively demonstrate the independent effects of
27 NO2 exposure on total mortality.
Policy-relevant Considerations for Evaluating Health
Effects Associated with Short-term NO2 Exposure
28 Recent epidemiologic studies continue to find respiratory and cardiovascular effects as
29 well as increases in total mortality in association with 24-h avg NO2 and shorter
30 averaging times, including 1-h max, 3-h max, and 8-h max NO2 (Section 1.6.1). No
31 consistent difference is demonstrated in the magnitude of health effects associated with
32 24-h avg ambient NO2 versus shorter averaging times. However, potential differences in
33 exposure measurement error may obscure differences in health effects associated with
34 various NO2 exposure metrics. Controlled human exposure studies demonstrate increases
November 2013 Ixxvi DRAFT: Do Not Cite or Quote
-------
1 in airway responsiveness of adults with asthma following NO2 exposures of 30 minutes
2 and 1 hour to concentrations of 200 to 300 ppb, and 100 ppb, respectively. Also,
3 respiratory effects are found in association with ambient NO2 exposures of 2 to 5 hours
4 in adults with asthma and healthy adults subject to outdoor exposures at various traffic
5 and nontraffic locations. Across health effects, epidemiologic associations are found with
6 ambient NO2 concentrations lagged 0 to 7 days or averaged over 2 to 7 days (Section
7 1.6.2). For several respiratory outcomes, larger effects are estimated for multiday
8 averages of NO2 than single-day averages. NO2 exposures during time spent outdoors are
9 associated with increases in respiratory effects immediately after exposure that persist to
10 the following day. The shape of the NO2 concentration-response relationship was
11 formally evaluated mostly for respiratory ED visits and total mortality. These studies
12 found evidence for linear relationships, and results do not identify a threshold for effects
13 (Section 1.6.3).
14 The public health significance of NO2-related health effects is supported by the large
15 percentage of the population living near major roads and potentially having elevated
16 exposures to NO2 (Section 1.6.5). Further, short-term ambient NO2 exposure is
17 consistently associated with effects that are clearly adverse such as hospital admissions,
18 ED visits, and mortality. NO2-related increases in effects such as airway responsiveness
19 in adults with asthma and changes in cardiovascular physiology in adults with
20 cardiovascular disease are considered adverse on a population level because a shift in the
21 distribution of the outcome can increase the proportion of individuals with clinically-
22 important effects such as asthma exacerbations. The presence of at-risk lifestages and
23 populations also informs the public health significance, as the NAAQS are intended to
24 protect public health for at-risk populations. There is suggestive evidence that
25 NO 2 -related health effects differ by pre-existing asthma, pre-existing COPD, genetic
26 variants for oxidative metabolism enzymes, dietary antioxidant intake, sex, and SES
27 (Chapter 6). There is adequate evidence that children (ages 0-14 years) and older adults
28 (ages 65 years and older) are at increased risk of NO2-related health effects. The reasons
29 for their increased risk (e.g., higher NO2 exposure, biological susceptibility) are not clear.
30 However, co-occurring risk factors in children and older adults may magnify the public
31 health impact of NO2-related health effects. For example, asthma is the leading chronic
32 illness among U.S. children, and cardiovascular disease is prevalent in older adults (Table
33 6-3). The large potential for elevated exposures to NO2, the increased risk for children
34 and older adults, and prevalence of asthma in children and other chronic diseases in older
35 adults can translate into a large number of people affected by NO2 exposure
November 2013 Ixxvii DRAFT: Do Not Cite or Quote
-------
Health Effects Associated with Long-term NO2 Exposure
1 A broad range of health effects has been evaluated for relationships with long-term NO2
2 exposure. The strongest evidence is for respiratory effects, and it indicates that there
3 is likely to be a causal relationship with long-term NO2 exposure (Section 5.2.17).
4 The key supporting evidence includes consistent recent epidemiologic findings for
5 associations between long-term ambient NO2 concentrations and asthma incidence in
6 children. Also, recent epidemiologic studies continue to demonstrate associations with
7 decreases in lung function and partially irreversible decreases in lung function growth in
8 children. The recent evidence reduces previous uncertainty regarding a relationship with
9 asthma incidence. A relationship between NO2 exposure and asthma incidence also is
10 supported by recent epidemiologic associations observed with related outcomes such as
11 respiratory symptoms in children with asthma and development of allergy in children.
12 Biological plausibility for effects on asthma is provided by previous findings for
13 increases in airway responsiveness and T-derived lymphocyte helper (Th)2 immune
14 responses in guinea pigs induced by long-term NO2 exposure and Th2 immune responses
15 in humans and guinea pigs induced by short-term NO2 exposure. Airway responsiveness
16 and Th2 immune responses are key processes involved in the development of asthma and
17 allergy. Previous and recent epidemiologic studies found NO2-related respiratory effects
18 in copollutant models with O3, SO2, PM, or EC. However, the limited examination of
19 potential confounding by copollutants that often are highly correlated with NO2 and
20 limited experimental evidence are not sufficient to conclusively demonstrate an
21 independent relationship between long-term NO2 exposure and respiratory effects.
22 For the other health effects examined, cardiovascular effects (Section 5.3.6).
23 reproductive and developmental effects (Section 5.4.5). total mortality (Section
24 5.5.3), and cancer (Section 5.6.12). evidence is suggestive of a causal relationship
25 with long-term NO2 exposure. The evidence base and uncertainties informing each of
26 these causal determinations share common characteristics. The causal determination for
27 each health effect category is strengthened from that made in the 2008 ISA for Oxides of
28 Nitrogen based on some recent epidemiologic studies showing associations with long-
29 term NO2 or NOX exposure. However, for each of the health effect categories, some
30 epidemiologic studies show no association. For cardiovascular effects, some recent
31 studies showed associations between long-term NO2 or NOX concentrations and heart
32 failure, myocardial infarction, stroke, or cardiovascular mortality. Reproductive and
33 developmental effects are evaluated as three subcategories that likely occur by different
34 biological processes and exposure patterns over different lifestages. There is limited
35 recent evidence for NO2-related effects on fertility, reproduction, and pregnancy
36 indicated as increases in pre-eclampsia. There is limited recent evidence for NO2-related
37 effects on birth outcomes indicated as fetal growth restriction. For postnatal development,
November 2013 Ixxviii DRAFT: Do Not Cite or Quote
-------
1 the strongest evidence is for partially irreversible decreases in lung function growth in
2 children with less consistent evidence for decreases in cognitive function in children.
3 Some recent epidemiologic studies also show associations of long-term NO2 or NOX
4 exposure with total mortality and lung cancer incidence and mortality. For each of these
5 health effect categories, there also is uncertainty in the relationship with long-term NO2
6 exposure because of limited or no supporting evidence in experimental animals to
7 provide biological plausibility.
Policy Relevant Considerations for Evaluating Health
Effects Associated with Long-term NO2 Exposure
8 Asthma incidence, decreases in lung function, and partially irreversible decreases in lung
9 function growth in children are found in association with various long-term averages of
10 NO2, including 6-month average and NO2 averaged over 1 to 10 (representing lifetime
11 exposure) years. Relationships between long-term NO2 exposure and outcomes such as
12 asthma incidence and decreases in lung function growth in children or COPD hospital
13 admissions in adults appear to be linear (Section 1.6.3). Most studies did not conduct
14 analyses to formally evaluate the shape of the concentration-response relationship or
15 evaluate whether there is evidence for a threshold for effects related to long-term NO2
16 exposure. A U.S. study did not find strong regional heterogeneity in the association
17 between NO2 exposure and asthma evaluated in several U.S. cities, New York, NY;
18 Chicago, IL; Houston, TX; and San Francisco, CA, as well as Puerto Rico (Section
19 1.6.4). The public health significance of the effects of long-term NO2 exposure is
20 supported by the evidence for increases in asthma incidence in children (Section 1.6.5).
21 Asthma is a leading cause of missed school days and hospital admissions in children.
22 Also, the NO2-related decreases in lung function growth found in children could have
23 implications for higher risk of mortality and cardiovascular morbidity in adulthood.
24 Regarding at-risk lifestages and populations, evidence indicates that the risk of
25 developing asthma or allergy and the magnitude of decreases in lung function growth
26 may be larger for long-term NO2 exposure around birth or infancy compared with
27 exposure later in childhood. Results for associations of long-term NO2 exposure with
28 asthma and lung function growth also informed the conclusion that there is suggestive
29 evidence for genetic variants for oxidative metabolizing enzymes increasing risk of
30 NO2-related health effects.
November 2013 Ixxix DRAFT: Do Not Cite or Quote
-------
Conclusions
1 Based on the 2008 National Emissions Inventory, the major NOX emissions source
2 categories in the U.S. are highway and off-highway vehicles and fuel combustion by
3 electric utilities. Ambient concentrations of NO2, NO, and NOX show spatial and
4 temporal heterogeneity at multiple scales and have been shown to be 30% to 200% higher
5 at locations within 15m of a roadway (averaged over hours to weeks) compared with
6 locations farther away from the road. Emissions of NOX and ambient concentrations of
7 NO2 have decreased over the past 20 years in the U.S. Relationships between NO2
8 concentrations obtained from ambient monitors and personal exposures vary in the
9 population, and exposure measurement error resulting from the use of ambient
10 concentrations has been shown to reduce epidemiologic associations observed with health
11 effects.
12 Recent studies, most of which are epidemiologic, expand on findings reported in the 2008
13 ISA for Oxides of Nitrogen and previous assessments. The consistency, coherence, and
14 biological plausibility of evidence integrated across scientific disciplines and outcomes
15 related to asthma exacerbations indicate that there is a causal relationship between
16 short-term exposure to NO2 and respiratory effects. Evidence indicates there is likely
17 to be a causal relationship between short-term exposure to NO2 and cardiovascular
18 effects as well as total mortality. There is likely to be a causal relationship between
19 long-term NO2 exposure and respiratory effects based strongly on findings in children
20 for asthma incidence and decreases in lung function. Evidence is suggestive of a causal
21 relationship between long-term NO2 exposure and cardiovascular effects,
22 reproductive and developmental effects, total mortality, and cancer.
23 A major uncertainty in the 2008 ISA for Oxides of Nitrogen was the extent to which
24 evidence indicated that NO2 has effects on health that are independent of effects of
25 another traffic-related pollutant or mixture. For respiratory effects, cardiovascular effects,
26 and total mortality related to short-term exposure and respiratory effects related to long-
27 term exposure, recent epidemiologic studies reduce this uncertainty with additional
28 results for associations with NO2 that remain positive in copollutant models. However,
29 analysis of confounding by the array of potentially correlated copollutants, in particular
30 CO, UFP, EC, and BC, which also are emitted from vehicles, is limited. Therefore, other
31 lines of evidence that inform biological plausibility are key in addressing limitations of
32 the epidemiologic evidence. For cardiovascular effects and total mortality related to
33 short-term exposure and respiratory effects related to long-term exposure, biological
34 plausibility is limited, and evidence is not sufficient to conclusively demonstrate effects
35 of NO2 exposure that are independent of those of other traffic-related pollutants.
November 2013 Ixxx DRAFT: Do Not Cite or Quote
-------
1 There is adequate evidence that children (ages 0-14 years) and older adults (ages 65 years
2 and older) have increased risk for NO 2 -related health effects. A large proportion of the
3 population lives near major roads; thus, the potential for elevated exposures to oxides of
4 nitrogen, relative to people living 500 meters or more from roads, is large. There is
5 suggestive evidence that the risk of NO2-related health effects differs by pre-existing
6 asthma, pre-existing COPD, genetic variants for oxidative metabolism enzymes, dietary
7 antioxidant intake, sex, and SES. Daily average and 1-h max NO2 concentrations as well
8 as concentrations averaged over 30 minutes to a few hours are associated with health
9 effects. For many respiratory outcomes, larger effects are estimated for multiday averages
10 of ambient NO2 concentrations than single-day concentrations. For long-term exposure,
11 respiratory effects are associated with 6-month average NO2 and NO2 averaged over
12 1 year to 10 (representing lifetime exposure) years. The concentration-response
13 relationship for associations of short-term ambient NO2 exposure with respiratory-related
14 ED visits and total mortality is found to be linear, and results do not identify a threshold
15 for effects.
References for Executive Summary
U.S. EPA (U.S. Environmental Protection Agency). (2008b). Integrated science assessment for oxides of
nitrogen and sulfur: Ecological criteria [EPA Report]. (EPA/600/R-08/082F). Research Triangle Park,
NC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=201485
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for particulate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA. (2013). Notice of workshop and call for information on integrated science assessment for oxides of
nitrogen and oxides of sulfur. Fed Reg 78: 53452-53454.
November 2013 Ixxxi DRAFT: Do Not Cite or Quote
-------
CHAPTER 1 INTEGRATED SUMMARY
1 The Integrated Science Assessment (ISA) is a comprehensive evaluation and synthesis of
2 the most policy-relevant science "... useful in indicating the kind and extent of
3 identifiable effects on public health or welfare which may be expected from the presence
4 of [a] pollutant in ambient air" (CAA. 1990a). This ISA serves as the scientific
5 foundation for the review of the health criteria for a broad category of oxides of nitrogen,
6 which includes nitrogen dioxide (NO2). As such, it communicates critical science
7 judgments to inform the review of the current primary (health-based) National Ambient
8 Air Quality Standards (NAAQS) for NO2. This ISA incorporates key information and
9 judgments contained in the 1993 Air Quality Criteria Document (AQCD) and the 2008
10 ISA for Oxides of Nitrogen (U.S. EPA. 2008c. 1993) to provide the foundation for the
11 review of recent studies. Additional details of the relevant scientific literature published
12 since the previous ISA, as well as key studies from previous assessments, are included.
13 Thus, this ISA serves to update the evaluation of the scientific evidence that was
14 available at the time of completion of the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
15 2008c).
16 The most recent review of the primary NO2 NAAQS was completed in 2010. EPA
17 retained the annual standard with a level of 53 parts per billion (ppb) NO2, annual
18 average (avg) concentration, to protect against health effects potentially associated with
19 long-term exposure. EPA established a new 1-hour (h) standard at a level of 100 ppb
20 NO2 based on the 3-year average of the 98th percentile of the yearly distribution of
21 1-h daily maximum (max) concentrations. The 1-hour standard was established to protect
22 against a broad range of respiratory effects associated with short-term exposures in
23 potential at-risk populations such as people with asthma and people who spend time on or
24 near major roads. In 2010, EPA also established requirements for a monitoring network
25 that includes monitors near major roads in urban areas, locations where maximum NO2
26 concentrations are expected to occur (U.S. EPA. 2010c). Additional information on the
27 legislative requirements and historical background for the NO2 NAAQS is contained in
28 the Preface to this ISA.
29 This chapter provides a summary and synthesis of the scientific evidence reviewed in the
30 ISA and the conclusions and findings that best inform many policy-relevant questions
31 that frame the review of the NO2 NAAQS as identified in "The Integrated Review Plan
32 for the Primary National Ambient Air Quality Standard for Nitrogen Dioxide." To that
33 end, this chapter includes:
34 • An evaluation of the evidence for health effects associated with short-term
35 (minutes up to 1 month) and long-term (more than 1 month to years)
November 2013 1-1 DRAFT: Do Not Cite or Quote
-------
1 exposure to oxides of nitrogen based on the integration of findings across
2 various scientific disciplines and across related health outcomes as well as a
3 discussion of important uncertainties identified in the interpretation of the
4 scientific evidence, including the role of NO2 within the broader ambient
5 mixture of pollutants.
6 • Discussion of policy-relevant considerations such as: exposure averaging
7 times and lags associated with health effects; concentration-response
8 relationships and thresholds below which effects do not occur; and lifestages
9 and populations potentially with increased exposure to oxides of nitrogen
10 and/or risk of associated health effects.
1.1 ISA Development and Scope
11 EPA uses a structured and transparent process for evaluating the scientific evidence and
12 forming conclusions and causal determinations for the relationships between air pollution
13 exposures and health effects. The ISA development process describes approaches for
14 literature searches, criteria for selecting and evaluating relevant studies, guidelines for
15 evaluating the weight of the evidence, and a framework for forming causal
16 determinations. As part of this process, the ISA is reviewed by the Clean Air Scientific
17 Advisory Committee, a formal independent panel of scientific experts, and by the public.
18 The ISA development process and causal framework are described in detail in the
19 Preamble to the ISA and are summarized below.
20 This ISA evaluates scientific information for gaseous species of oxides of nitrogen1.
21 Oxides of nitrogen consist of all forms of oxidized nitrogen compounds, including gases
22 such as NO2 and nitric oxide (NO) as well as their gaseous and particulate reaction
23 products (e.g., organic and inorganic nitrates and nitrites, nitro-polycyclic aromatic
24 hydrocarbons) (Section 2.2. Figure 2-1)2. The particle species (e.g., nitrates, nitro-
25 poly cyclic aromatic hydrocarbons) are not the focus of this ISA and were reviewed most
26 recently in the 2009 ISA for PM (U.S. EPA. 2009a). When referring to the group of
27 gaseous oxidized nitrogen compounds as a whole, the ISA uses the term oxides of
28 nitrogen. Based on the definition commonly used in the scientific literature, this ISA uses
29 the abbreviation NOX to refer specifically to the sum of NO2 and NO concentrations.
1 The other criteria pollutants are ozone (O3), particulate matter (PM), oxides of sulfur (SOX and SO2),
carbon monoxide (CO), and lead (Pb).
2 Section 108(c) of the Clear Air Act refers to oxides of nitrogen as all forms of oxidized nitrogen including multiple
gaseous and particulate species. 42. U.S.C. 21 7408(c).
November 2013 1-2 DRAFT: Do Not Cite or Quote
-------
1 As this ISA informs the review of the primary NO2 NAAQS, it evaluates information on
2 potential relationships of oxides of nitrogen with health effects as reported in
3 epidemiologic, controlled human exposure, and toxicological studies. Relevant studies
4 also include those informing concentration-response relationships, modes of action, and
5 potential at-risk lifestages and populations for exposure to oxides of nitrogen and/or
6 related health effects. Also relevant to this ISA are studies on atmospheric chemistry and
7 fate of emissions as well as EPA analyses of air quality and emissions data. The
8 ecological and other welfare effects of oxides of nitrogen are being evaluated in a
9 separate assessment conducted as part of the review of the secondary (welfare-based)
10 NAAQS for NO2 and sulfur dioxide (SO2) (U.S. EPA. 2013).
11 EPA initiated the current review of the primary NAAQS for NO2 in February 2012 with
12 a call for information from the public (U.S. EPA. 2012). From that time, literature
13 searches were conducted routinely to identify peer-reviewed studies published since the
14 previous ISA (i.e., studies published starting in 2008). Multiple search methods were
15 used (Section 2, Preamble), and relevant studies were also identified by referrals from the
16 public and scientific experts. Some studies were judged to be irrelevant (see preceding
17 paragraph) based on title and were excluded. Studies that were judged as potentially
18 relevant based on review beyond the title and "considered" for inclusion in the ISA are
19 documented and can be found at the Health and Environmental Research Online (HERO)
20 website. The HERO project page (http://hero.epa.gov/oxides-of-nitrogen) for the ISA for
21 Oxides of Nitrogen lists the references that are cited in the ISA as well as the references
22 that were considered for inclusion but not cited, and also contains electronic links to
23 bibliographic information and abstracts.
24 Health effects were considered for evaluation in this ISA if examined in previous
25 assessments for oxides of nitrogen or if examined in multiple recent studies in more than
26 one location (e.g., neurodevelopmental effects). Literature searches identified one or two
27 recently published epidemiologic studies each on outcomes such as epilepsy, headache,
28 depression, ocular effects, gastrointestinal effects, and bone density [Supplemental Table
29 S1 -1 (U.S. EPA. 2013e)1. These health effects are not evaluated in the current draft ISA
30 because of the large potential for publication bias. These studies were conducted in areas
31 and populations for which associations between oxides of nitrogen and other health
32 effects have been demonstrated. Thus, the exclusion of these studies does not exclude the
33 assessment of particular geographic locations, potential at-risk lifestages or populations,
34 or range of ambient concentrations of oxides of nitrogen.
35 The Preamble describes the general framework for evaluating scientific information,
36 including criteria for assessing study quality and findings and developing scientific
37 conclusions. Greater weight is placed on those studies most relevant to the review of the
November 2013 1-3 DRAFT: Do Not Cite or Quote
-------
1 NAAQS. For epidemiologic studies, this includes high-quality studies that can be
2 characterized as: (1) studies that provide understanding of the quantitative relationships
3 between varying concentrations of oxides of nitrogen and health effects; (2) studies that
4 examine oxides of nitrogen as a component of a complex mixture of air pollutants and
5 consider other potential confounding factors; (3) studies with a priori aims to examine
6 potential at-risk lifestages and populations; or (4) multicity studies that employ
7 standardized analytical methods for evaluating health effects of oxides of nitrogen across
8 locations and provide overall estimates for effects by pooling information across multiple
9 cities. With respect to the evaluation of controlled human exposure and toxicological
10 studies, emphasis is placed on studies that examine effects relevant to humans and
11 concentrations of oxides of nitrogen that are relevant to human ambient exposures.
12 Ambient-relevant exposures are defined as those no greater than 5,000 ppb, which is one
13 to two orders of magnitude higher than peak concentrations of NO2, NO, or NOX that
14 humans experience on roads (Section 2.5.3). Studies with higher exposure concentrations
15 are included in cases where results inform potential modes of action. For the evaluation
16 of human exposure to ambient oxides of nitrogen, emphasis is placed on studies that
17 examine the quality of data sources used to assess exposures such as central site
18 monitors, land use regression (LUR) models, and personal exposure monitors. The ISA
19 also emphasizes studies that examine factors that influence exposure such as time-activity
20 patterns and building ventilation characteristics.
21 The ISA uses a formal causal framework to classify the weight of evidence according to a
22 five-level hierarchy (Table II of the Preamble). Conclusions are drawn based on
23 information integrated across scientific disciplines and related endpoints and the
24 synthesis of evidence from previous and recent studies. Determinations are made for
25 causation not just association and are based on judgments of aspects such as the
26 consistency, coherence, and biological plausibility of observed effects (i.e., evidence for
27 the direct effect of a pollutant on a health outcome or key events that inform the mode of
28 action) as well as related uncertainties.
29 • Causal relationship: the consistency and coherence of evidence integrated
30 across scientific disciplines and related outcomes are sufficient to rule out
31 chance, confounding, and other biases with reasonable confidence.
32 • Likely to be a causal relationship: several studies show effects that do not
33 appear to be explained by chance, confounding, and other biases, but
34 uncertainty remains in the evidence base. For example, there may be some
35 uncertainty regarding potential confounding by other pollutants in the
36 ambient mixture or biological plausibility because of some inconsistency of
37 evidence among different disciplines.
November 2013 1-4 DRAFT: Do Not Cite or Quote
-------
1 • Suggestive of a causal relationship: evidence overall is limited, for
2 example, to a high-quality epidemiologic or animal toxicological study with
3 effects relevant to humans. Or, effects are found in some high-quality studies
4 but not in other studies that are or are not of comparable quality.
5 • Inadequate to infer a causal relationship: there is insufficient quantity,
6 quality, consistency, or statistical power of results from studies.
7 • Not likely to be a causal relationship: several adequate studies, examining
8 the full range of human exposure concentrations and potential at-risk
9 lifestages and populations, consistently show no effect.
10 Beyond forming causal determinations for relationships between pollutant exposures and
11 health effects, the ISA aims to address questions relevant to quantifying health risks using
12 information on the quantitative relationships between pollutant exposures and health
13 effects. These questions include:
14 • What is the nature of the concentration-response, exposure-response, or
15 dose-response relationship?
16 • Under what exposure conditions (dose or concentration, duration, and
17 pattern) are effects observed?
18 • What lifestages or populations appear to have increased exposure to oxides
19 of nitrogen and/or risk of associated health effects?
1.2 Organization of the ISA
20 The ISA comprises the Preamble. Preface (with Legislative Requirements and History of
21 the NAAQS for NO2), Executive Summary, and six chapters. Subsequent sections of
22 Chapter 1 synthesize the scientific evidence that best informs policy-relevant questions
23 that frame this review of the primary NO2 NAAQS. Section 1.3 summarizes information
24 on the sources, atmospheric chemistry, ambient concentrations, and human exposure to
25 oxides of nitrogen. Section 1.4 summarizes the causal determinations for health effects
26 associated with short-term and long-term NO2 exposure and the key contributing
27 evidence, including information on the dosimetry of NO2 and NO and the key events that
28 inform the modes of action underlying health effects. Section 1.5 is an integrated
29 discussion of the scientific information addressing the independent effects of exposure to
30 oxides of nitrogen on various health outcomes, which was an important uncertainty
31 identified in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008cV The discussion
32 addresses potential confounding by copollutants and other factors and whether oxides of
November 2013 1-5 DRAFT: Do Not Cite or Quote
-------
1 nitrogen have independent effects or serve primarily as an indicator of other traffic-
2 related air pollutants. Section 1.6 presents a discussion of policy-relevant considerations,
3 such as exposure metrics associated with health effects, concentration-response
4 relationships for NO2; and potential at-risk lifestages and populations.
5 Chapter 2 characterizes the sources, atmospheric chemistry, and fate of oxides of nitrogen
6 in the environment; trends in ambient concentrations; and factors influencing human
7 exposure to oxides of nitrogen. Chapter 3 describes the dosimetry and modes of action
8 for health effects related to NO2 and NO. Chapter 4 and Chapter 5 evaluate and integrate
9 epidemiologic, controlled human exposure, and toxicological evidence for the health
10 effects related to short-term and long-term exposure to oxides of nitrogen, respectively.
11 Chapter 6 evaluates the evidence for potential at-risk lifestages and populations.
1.3 Sources of Oxides of Nitrogen to Human Exposure
1.3.1 Sources of Oxides of Nitrogen
12 Direct emissions of oxides of nitrogen from sources comprise a mix of NO and NO2,
13 with a ratio in favor of NO. Based on the 2008 National Emissions Inventory1, the major
14 NOX emissions categories in the U.S. are related to combustion processes, dominated by
15 highway vehicles (39%) and followed by off-highway vehicles (19%), fuel combustion
16 by electric utilities (17%), and industrial fuel combustion (8%) (Section 2.3. Figure 2-2).
17 Smaller source categories (each less than 5% of the inventory) include other industrial
18 operations and microbial processes in soil. Specific NOX emissions sources that can
19 affect local air quality include on-road vehicles, airports, railyards, shipping ports, home
20 wood burning, intense industrial and chemical processes, activities for oil and gas
21 development, and wildfires (Section 2.3). Some of these specific sources can emit large,
22 transient peaks of NOX. Specific locations vary in both the presence and the mix of
23 specific emissions sources that contribute to total emissions.
24 Emissions from natural and anthropogenic sources from continents other than North
25 America contribute to North American Background NO2 concentrations. Seasonal mean
26 background concentrations are estimated to be less than 0.3 ppb over most of the
27 continental U.S. and account for a small fraction of ambient concentrations in the U.S.
28 For example, in the Eastern U.S. where ambient NO2 concentrations are the highest,
:The data presented on emissions sources will be updated in the Second External Review Draft of the ISA for
Oxides of Nitrogen using the 2011 National Emissions Inventory, which became available in November 2013.
November 2013 1-6 DRAFT: Do Not Cite or Quote
-------
1 North American Background accounts for less than 1% of the total concentration (Section
2 2.5.6V
1.3.2 Atmospheric Chemistry and Fate of Oxides of Nitrogen
3 In addition to emissions sources, ambient concentrations of oxides of nitrogen are
4 determined by chemical transformations, transport to other locations, meteorology, and
5 deposition to surfaces. A major transformation is the rapid (i.e., minutes) reaction that
6 occurs between direct NO emissions and ozone (O3) to form NO2 (Section 2.2, Figure
7 2-1). Rather than direct emissions, the reaction of NO with O3 is the main source of NO2
8 concentrations obtained from most ambient air monitors in U.S. urban locations.
9 NO and NO2 also are transformed into other oxides of nitrogen. Reactions with gas phase
10 hydroxyl radicals, hydroperoxy radicals, organic peroxyl radicals, and O3 form
11 compounds such as nitric acid (HNO3), peroxyacetyl nitrate (PAN), nitrous acid
12 (HONO), and particulate nitrates. NO and NO2 also are involved in reaction cycles with
13 radicals produced from volatile organic compounds (VOCs) to form O3 (Section 2.2).
14 The major gas-phase products of the reactions of NO and NO2 are PAN and HNO3.
15 Morning rush hour NO and NO2 emissions can be completely converted to HNO3 and
16 other products by late afternoon under warm, sunny conditions via reaction with highly
17 abundant hydroxyl radicals. The abundance of nitrate radicals at night favors formation
18 of gas phase organic nitrates, secondary organic aerosols, and dinitrogen pentoxide. PAN
19 and isoprene nitrates can be transported to distant locations, where they decompose to
20 release NO2, which is then available to participate in O3 formation. HNO3 acts similarly
21 but is highly soluble and has a high deposition rate. The transformations of NO and NO2
22 into other oxides of nitrogen followed by dry deposition (i.e., impaction with surfaces or
23 gas exchange with plants), and wet deposition (i.e., diffusion into cloud droplets, washout
24 by impaction with falling rain drops) are the major processes by which oxides of nitrogen
25 are removed from the atmosphere.
1.3.3 Ambient Concentrations - Temporal and Spatial Trends
26 Information on ambient concentrations of NO2, NO, and NOX in the U.S. is provided
27 primarily by the State and Local Air Monitoring Stations (SLAMS) Network of about
28 500 sites (Section 2.4.5). and most information is for NO2. This network serves many
29 objectives: determining compliance with the NAAQS, providing the public with air
30 pollution data in a timely manner, and providing estimates of human ambient exposure
31 for many epidemiologic studies of health effects. The regulatory network monitors use
November 2013 1-7 DRAFT: Do Not Cite or Quote
-------
1 chemiluminescence techniques that directly measure NO and NOX and then report a
2 calculated NO2 as the difference between NOX and NO. Monitoring sites are located
3 within U.S. Metropolitan Statistical Areas or urban areas. The near-road network
4 promulgated as part of the 2010 decision on the primary NO2 NAAQS is being phased in,
5 with the first of three phases of monitoring scheduled to begin in January 2014. Data
6 from this network are not available yet.
7 Across U.S. SLAMS, the mean and 99th percentile for 1-h daily maximum ambient NO2
8 concentrations for 2009-2011 were 20 ppb and 57 ppb, respectively (Table 2-1). During
9 the same time period, the mean and 99th percentile for annual average NO2
10 concentrations were 9.4 ppb and 25 ppb, respectively (Table 2-2). Although ambient
11 concentrations of species such as HNO3 and HONO can be higher than those of NO2 far
12 downwind of sources, the limited available data indicate that with typical ambient
13 concentrations of NO2, ambient concentrations of HNO3 and HONO range from less than
14 0.1 ppb to a few ppb. With respect to long-term temporal trends, U.S.-wide annual
15 average NO2 concentrations decreased by 48% from 1990 to 2012 (Figure 2-16). This
16 decrease is attributed to a decrease in NOX emissions, which declined by more than 50%
17 in the U.S. from 1990 to 2012 (Figure 2-3). Emissions were reduced for on-road vehicles
18 and electric utilities due, in part, to the use of pollution control technologies (Sections
19 2.3.2 and 2.3.7). In addition to long-term trends, ambient NO2 concentrations show
20 seasonal trends, with higher concentrations measured in the winter than summer. In urban
21 areas, ambient NO and NO2 concentrations rise during the night when atmospheric
22 mixing is reduced because of low wind speeds and low mixing layer heights. Ambient
23 NO and NO2 concentrations peak in early mornings corresponding with morning rush
24 hours, decrease until late afternoon, then increase again in early evening. Ambient
25 concentrations at most urban sites are higher on weekdays versus weekends.
26 Ambient NO2 concentrations vary on multiple spatial scales, including regional, urban,
27 neighborhood, and microscale environments (Section 2.5). Corroborating previous data,
28 data from monitoring networks for 2009-2011 indicate higher NO2 concentrations in
29 urban locations than less populated nonurban locations (Figure 2-10 and Figure 2-12 for
30 annual average and seasonal average, respectively). There are more monitors in urban
31 areas, which may make it difficult to assess regional trends in ambient NO2
32 concentrations. Several lines of evidence demonstrate the large variability in ambient
33 NO2 concentrations with distance to roads. Across geographic areas, ambient NO, NO2,
34 and NOX concentrations decrease exponentially with increasing distance from the
35 roadway, reaching background concentrations within 100 to 500 meters downwind of the
36 roadway (Section 2.5.3). LUR models show that at the urban scale, ambient NO2
37 concentrations can be predicted well by roadway proximity and in some cases, proximity
38 to industrial sources, which are other major sources of NO2 (Section 2.6.2.3).
November 2013 1-8 DRAFT: Do Not Cite or Quote
-------
1 Neighborhood-scale and smaller near-road scale ambient concentrations of oxides of
2 nitrogen can be affected by temporal variation in emissions sources, atmospheric
3 conditions, and characteristics of the built environment. Characteristics of the built
4 environment such as vehicle direction and speed, roadway structure (barriers versus open
5 terrain), slopes of roadways, and height of buildings adjacent to roadways can affect
6 ambient concentrations by affecting movement of air.
1.3.4 Human Exposure to Oxides of Nitrogen
7 The spatial and temporal variability in ambient concentrations measured at various scales
8 are important influences on human exposure to ambient oxides of nitrogen. Human
9 exposure is determined by concentrations in specific ambient, indoor, and in-vehicle,
10 microenvironments and time spent in those microenvironments (Section 2.6.1).
11 Characterizing human exposure to ambient oxides of nitrogen is of primary importance to
12 the review of the NO2 NAAQS. Among oxides of nitrogen, human exposure to ambient
13 NO 2 is the most well characterized, and key findings described below are supported by
14 recent studies and those reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
15 2008c). A recent analysis found that a time-weighted average of NO2 concentrations in
16 various microenvironments corresponded well with total personal NO2 exposures
17 (Section 2.6.5.2). Such findings indicate that variability in ambient concentrations and
18 time-activity patterns affect how exposure varies within individuals as they move across
19 locations in time and how exposure varies among individuals in the population who are in
20 different locations. Human exposure also is determined by indoor NO2 concentrations.
21 Indoor concentrations not only are affected by indoor sources such as gas stoves and
22 heaters, oil furnaces, wood burning stoves, kerosene heaters, and smoking (Section
23 2.6.3.3) but also the penetration of ambient concentrations. The penetration of ambient
24 concentrations indoors and in turn, human exposure, can vary according to ventilation
25 characteristics (e.g., open windows, air conditioning, building air exchange rate). The
26 relative contribution of NO2 exposure in specific microenvironments, including in
27 vehicles, to peak or total exposure is not widely characterized. However, a recent study in
28 the U.K. found that indoor exposures made up the majority of total NO2 exposure in
29 adults, with smaller contributions coming from in-vehicle and outdoor exposures.
30 In many epidemiologic studies, human exposure to ambient NO2 is estimated using
31 ambient concentrations obtained from central site monitors. The siting of monitors
32 indicates that ambient NO2 concentrations likely represent area-wide exposures at the
33 regional, urban, or neighborhood scale. Thus, central site monitors do not cover all
34 locations where people live or spend at least part of their time. Ambient concentrations
35 may not be available for the microenvironment of interest, for example near roads. Data
November 2013 1-9 DRAFT: Do Not Cite or Quote
-------
1 from the near-road monitoring network are not available yet. Hence, the use of ambient
2 concentrations obtained from central site monitors to represent human exposure to
3 ambient NO2 is associated with measurement error. The extent of exposure measurement
4 error is influenced by the relationship between personal and central site ambient
5 concentrations. These relationships are not consistently characterized for NO2,
6 particularly for long-term NO2 exposures. Further, examination generally is limited to
7 relationships between ambient NO2 concentrations and total personal NO2 exposure,
8 rather than the ambient component of personal exposure (Section 2.6.5.1). In some cases,
9 personal exposure measures are comparable to corresponding central site monitor
10 concentrations. In other cases, the two exposure metrics vary considerably. An analysis
11 combining several studies found personal and ambient NO2 concentrations to have low to
12 moderate correlation from 0.16 to 0.45 (Table 2-11). Thus, there is heterogeneity among
13 individuals in how well the spatial or temporal variability in ambient concentrations
14 obtained from central monitors correlate with personal exposure, taking into account
15 varying time-activity patterns. A recent meta-analysis found that personal-ambient NO2
16 relationships may be influenced by season, age, and local sources (Section 2.6.5.2).
17 Higher personal-ambient correlations are reported for adults than for children.
18 Several studies have aimed to characterize the exposure error that is due to the spatial or
19 temporal variability in ambient NO2 concentrations. A study of cardiovascular-related
20 emergency department (ED) visits added an error term to a time-series health effects
21 model that accounted for correlations of concentrations across monitor sites and found
22 that associations with NO2 were attenuated toward the null. Some studies estimated
23 larger respiratory effects in association with more spatially-resolved estimates of short-
24 term NO2 exposure than with NO2 concentrations obtained from a single city central site
25 monitor or averaged over area monitors (Sections 2.6.5.3 and 4.2.4.5). Exposure
26 estimates with higher spatial resolution included personal NO2 measures, outdoor school
27 measurements, and ambient concentrations at the nearest central site monitor. Health
28 effects also are associated with measures of long-term NO2 or NOX exposure that are
29 more spatially resolved than central site concentrations. For example, LUR or dispersion
30 models and spatial interpolation methods have been used to estimate ambient NO2
31 concentrations at the neighborhood scale or at the level of an individual's residence.
32 Measurement error is a component of each NO2 exposure assessment method. Error can
33 depend on the accuracy and precision of the instrumentation. For example, the
34 chemiluminescence method used by regulatory networks tends to overestimate ambient
35 NO2 concentrations because of interference from other oxides of nitrogen. However,
36 interference generally is less than 10% in urban locations and near sources, and new
37 measurement methods are being developed (Section 2.4.1). Spatially-resolved estimates
38 of NO2 exposure may include measurement error because they do not incorporate
November 2013 1-10 DRAFT: Do Not Cite or Quote
-------
1 information on exposure in the range of potentially important microenvironments.
2 Modeled residential ambient NO2 may account better for differences among individuals
3 in distance to sources but may estimate total ambient exposure with error because they
4 may not account for time-activity patterns (i.e., movement of individuals across areas).
5 On the other hand, concentrations averaged across multiple monitors may represent the
6 mean ambient exposures of the population moving within the area during a given time
7 period especially if there are few or well-dispersed local sources of NO2.
8 Exposure measurement error can have important implications for the relationships
9 observed between NO2 concentrations and health effects. Study design can influence the
10 nature of the effect of exposure measurement error (Section 2.6.5.3). Repeated measures
11 time-series or panel studies examine relationships between temporal patterns in exposures
12 and outcomes. If the temporal variation in ambient NO2 concentrations is the same as the
13 temporal variation in true ambient exposure, there may be little effect on associations
14 with health. But, differences in the correlation across people and time, for example due to
15 varying nonambient exposures, may decrease the magnitude or precision of associations.
16 Cohort or cross-sectional studies tend to compare exposures and outcomes among people
17 who live in different locations. In these studies, estimating exposure with ambient
18 concentrations from central site monitors can decrease the magnitude and/or precision of
19 effect estimates if the monitor is not located close to the study population and/or if local
20 emission sources are not well dispersed. The impact on epidemiologic associations of
21 exposure measurement error resulting from the use of central site ambient NO2
22 concentrations to represent near-road exposures is not well characterized. However, the
23 contribution of near-road exposure to total ambient NO2 exposure or to the health effects
24 found in association with ambient NO2 concentrations is not well characterized either.
1.4 Health Effects of Oxides of Nitrogen
25 This ISA evaluates relationships between a broad range of health effects and short-term
26 (Chapter 4) and long-term (Chapter 5) exposures to oxides of nitrogen as examined in
27 epidemiologic, controlled human exposure, and animal toxicological studies. Short-term
28 exposures are defined as those with durations of minutes up to 1 month. Most of these
29 studies examine effects related to exposures in the range of 1 hour to 1 week; a few
30 controlled human exposure and animal toxicological studies examine exposures of less
31 than 1 hour. Long-term exposures are defined as those with durations of more than 1
32 month to years, with 4 weeks used as the cut-off in most toxicological studies. The
33 evidence informing the causal determinations is described in detail in Chapter 4 and
34 Chapter 5, and is summarized in this section and in Table 1-1. Across disciplines, studies
November 2013 1-11 DRAFT: Do Not Cite or Quote
-------
1 examined health effects primarily in relation to NO2; information is limited for NO or
2 NOX. Thus, causal determinations are formed only forNO2.
3 An uncertainty noted in the 2008 ISA for Oxides of Nitrogen for the relationships
4 between NO2 exposure and health effects was the potential for the effects observed with
5 NO2 to be biased (i.e., confounded) by the effects of another traffic-related pollutant or
6 mixture that is highly correlated with NO2 concentrations. Section 1.5 presents an
7 integrated evaluation of confounding, and the discussions of the health effects evidence
8 in the sections that follow also describe the extent to which recent studies inform this
9 uncertainty. The copollutants most frequently examined include PM with aerodynamic
10 diameter less than or equal to a nominal 2.5 (im (PM2 5), PM with aerodynamic diameter
11 less than or equal to a nominal 10 (im (PMi0), SO2, and O3. Examination of potential
12 confounding by VOCs, carbon monoxide (CO), black or elemental carbon (BC or EC),
13 and ultrafine particles (UFPs) is more limited and varies across health effects.
1.4.1 Dosimetry and Modes of Action Informing Respiratory Effects
14 Linking NO2, NO, or NOX exposure to the occurrence of health effects are the dosimetry
15 (Section 3.2) and modes of action (Section 3.3) of these pollutants. Relevant to all health
16 effects, inhaled NO2 at ambient-relevant concentrations is unlikely to penetrate the
17 extracellular lining fluid (ELF) to reach underlying sites in the respiratory tract
18 epithelium. During absorption into the ELF, NO2 becomes a solute and rapidly reacts
19 with antioxidants and other ELF constituents (Section 3.2.2). The formation of secondary
20 oxidation products likely is the initiating event in the sequence of key events comprising
21 the mode of action for NO2. These products can mediate oxidation of cell membrane
22 lipids which may lead to alterations in permeability of the alveolar capillary barrier. They
23 also can mediate thiol oxidation which can alter enzyme activity and the antioxidant-
24 oxidant balance. Subsequent responses at the cellular, tissue, or organ level likely lead to
25 the health effects associated with NO2 exposure. Key events that pertain to specific
26 health effects are described in the sections that follow. NO2 is produced endogenously by
27 enzymatic and nonenzymatic pathways that are enhanced during inflammation and other
28 immune responses. It is not clear how ambient-relevant NO2 exposures compare with
29 endogenous concentrations or rate of production (Section 3.2.2.4).
30 Although not as extensively examined as NO2, ambient NO or NOX exposures also are
31 found to be associated with health effects. Inhaled NO is present in the respiratory tract in
32 the gas phase and is not transformed by reactions in the ELF. Ambient NO concentrations
33 generally are in the range of endogenous NO concentrations found in the respiratory tract.
34 Thus, it is not clear whether inhaled NO at ambient-relevant concentrations significantly
November 2013 1-12 DRAFT: Do Not Cite or Quote
-------
1 affects the absorption, metabolism, or downstream biological processes of endogenous
2 NO (Section 3.2.3). Because there is biological plausibility for negative respiratory
3 effects occurring with ambient-relevant NO2 exposures but not with ambient-relevant NO
4 exposures, respiratory effects associated with ambient NOX are considered to be
5 reflecting associations with NO2 and are considered in causal determinations for NO2.
1.4.2 Respiratory Effects
6 The strongest evidence for a relationship between NO2 exposure and respiratory effects is
7 for asthma exacerbations in children and adults related to short-term exposure and for
8 asthma development in children related to long-term exposure. Biological plausibility is
9 provided by findings for NO2-induced increases in airway responsiveness and effects on
10 other key events informing the mode of action such as inflammation and oxidative stress.
11 NO2 showed effects on many of the same endpoints in humans and experimental animals,
12 indicating similar biological processes across species. The respiratory effects of NO2
13 exposure also are supported by the characterization of the uptake of inhaled NO2 in the
14 respiratory tract and reactions to form secondary oxidation products (Section 1.4.1).
15 This insight into potential biological processes by which the inhalation of NO2 may lead
16 to asthma exacerbations and asthma development supports the causal determinations
17 made for relationships of respiratory effects with both short-term and long-term NO2
18 exposure (Table 1-1). Each of the causal determinations is strengthened from the 2008
19 ISA for Oxides of Nitrogen based on evidence from recent epidemiologic studies that
20 reduces previously identified uncertainties. For short-term exposure, the causal
21 determination is strengthened from likely to be a causal relationship to a causal
22 relationship because recent epidemiologic studies demonstrate that associations of
23 ambient NO2 concentrations with asthma and other respiratory effects remain positive
24 with adjustment for various copollutants. For long-term exposure, the causal
25 determination is strengthened from suggestive of a causal relationship to likely to be a
26 causal relationship based on recent epidemiologic evidence consistently demonstrating
27 NO2-related increases in asthma incidence in children. Previous studies did not
28 consistently find NO2-related increases in asthma incidence in children. A difference in
29 the evidence bases for respiratory effects related to short-term and long-term NO2
30 exposure is the extent to which evidence supports the independent effects of NO2,
31 including findings from experimental studies and findings from epidemiologic studies
32 that examine associations with NO2 in models with a copollutant.
November 2013 1-13 DRAFT: Do Not Cite or Quote
-------
Respiratory Effects Associated with Short-term NO2 Exposure
1 The strongest evidence in support of a causal relationship between short-term NO2
2 exposure and respiratory effects comprises results across scientific disciplines indicating
3 effects of NO2 exposure on asthma exacerbations (Section 4.2.9. Table 4-23). Recent
4 epidemiologic studies in diverse geographic locations and of varied study designs expand
5 evidence for NO 2 -related increases in hospital admissions and ED visits for asthma as
6 well as increases in respiratory symptoms in children with asthma. As uncontrolled
7 symptoms are a major reason for seeking medical treatment for asthma, the coherence of
8 findings for these outcomes further supports a relationship between NO2 exposure and
9 respiratory effects. These studies are considered to be high quality based on analyses that
10 apply a consistent statistical model to data pooled across multiple cities and across
11 several years, thus minimizing the potential for publication bias. Several panel studies
12 finding NO 2 -associated respiratory effects in children with asthma are noteworthy for
13 spatially-resolved estimates of exposure, including personal exposure, modeled outdoor
14 concentrations, and outdoor school concentrations. Both time-series and panel studies
15 adjust for potential confounding by temporal factors such as meteorology and long-term
16 time trends. Recent studies of hospital admissions, ED visits, and respiratory symptoms
17 reduce previous uncertainty regarding copollutant confounding with additional findings
18 that associations of NO2 remain positive with adjustment for a copollutant among PM2 5,
19 PM10, SO2, O3, or as examined in fewer studies, CO, EC, BC, or UFP. These
20 copollutant-adjusted associations were found across geographic locations that vary from
21 each other in correlations between NO2 and copollutants, pointing to varied air pollution
22 mixtures. Associations are found in studies with mean 24-h avg NO2 concentrations 18 to
23 29 ppb and maximum concentrations 48 to 106 ppb. Associations are found in studies
24 with mean 1-h max NO2 concentrations 22to 66 ppb and maximum concentrations 59 to
25 298 ppb.
26 Biological plausibility for the independent effects of NO2 exposure on asthma
27 exacerbations is demonstrated by findings from previous controlled human exposure
28 studies that NO2 exposures of 200 to 300 ppb for 30 minutes, and 100 ppb for 1 hour
29 induced increases in airway responsiveness in adults with asthma. Airway hyper-
30 responsiveness is a key pathophysiological characteristic of asthma and can lead to
31 respiratory symptoms and asthma exacerbations. Studies across disciplines also
32 characterized key events informing the modes of action for airway responsiveness and
33 respiratory symptoms, including allergic inflammation and oxidative stress. Some but not
34 all controlled human exposure and animal studies found NO2-induced oxidative stress,
35 and results are inconsistent for the effects of ambient-relevant NO2 exposures on lung
36 permeability. However, in experimental studies of both humans and rats, NO2 induced
37 increases in indicators of allergic inflammation such as T-derived lymphocyte helper
November 2013 1-14 DRAFT: Do Not Cite or Quote
-------
1 (Th)2 cytokines, immunoglobulin E, activated eosinophils, and neutrophils.
2 Epidemiologic studies found ambient NO2-related increases in pulmonary inflammation
3 in children with asthma, and several study populations had high prevalence of allergy.
4 These epidemiologic associations were found in several studies characterized as having
5 strong exposure assessment with school or personal monitoring or time-weighted
6 estimates of outdoor exposure. The observations of NO2-related increases in allergic
7 inflammation support the findings of NO2-induced increases in airway responsiveness in
8 adults with asthma and increases in respiratory symptoms found in association with
9 ambient NO2 in epidemiologic studies of children with asthma and allergy.
10 NO2-related decreases in lung function were not found consistently in adults with asthma
11 in controlled human exposure studies but were found in recent epidemiologic studies of
12 children with asthma. Several of the study populations of children with asthma had high
13 prevalence of allergy. The epidemiologic findings are supported by observations that
14 NO2-induced lung function decrements may be mediated by increases in mast cell
15 degranulation and airway obstruction. The cascade of events leading from NO2-related
16 increases in allergic inflammation and airway obstruction to decreases in lung function
17 may provide an additional explanation for NO2-related increases in respiratory symptoms
18 and hospital admissions and ED visits for asthma.
19 A causal relationship between short-term NO2 exposure and respiratory effects is
20 supported by evidence for other specific respiratory effects such as impaired host defense
21 and exacerbations of chronic obstructive pulmonary disease (COPD) as well as hospital
22 admissions, ED visits, and mortality for all respiratory causes combined. There is clear
23 evidence for impaired host defense demonstrated by NO2-induced (1,500 to 5,000 ppb
24 for 1 to 8 hours) mortality in animal models following bacterial or viral infection, and
25 support from associations observed between ambient NO2 concentrations and respiratory
26 infections in children. Some experimental evidence describes effects of NO2 exposure on
27 key events informing the mode of action, including NO2-induced decreases in alveolar
28 macrophage function and increases in pulmonary inflammation. Effects on pulmonary
29 clearance were more variable. Evidence also supports associations of ambient NO2
30 concentrations with COPD hospital admissions and ED visits. However, NO2 was not
31 consistently associated with increases in respiratory symptoms or decreases in lung
32 function in epidemiologic or controlled human exposure studies of adults with COPD.
33 Recent epidemiologic studies consistently indicate ambient NO2-associated increases in
34 respiratory mortality, which demonstrates the effects of NO2 exposure on a continuum of
35 respiratory effects. However, it is not entirely clear what changes in respiratory morbidity
36 NO2 exposure may induce to lead to increases in respiratory mortality. Strong evidence
37 demonstrates NO2-related effects on asthma exacerbations, but asthma is not a leading
38 cause of mortality. There is limited coherence among lines of evidence indicating
November 2013 1-15 DRAFT: Do Not Cite or Quote
-------
1 NO 2-related effects on COPD exacerbations and respiratory infections, which are larger
2 causes of mortality.
3 In conclusion, evidence indicates that there is a causal relationship between short-term
4 NO 2 exposure and respiratory effects, based strongly on the consistency, coherence, and
5 biological plausibility of findings for the effects on exacerbations of asthma. There is
6 some evidence for relationships of NO2 exposure with impaired host defense, COPD
7 exacerbations, and respiratory mortality but limited coherence among outcomes or
8 disciplines. Previous uncertainty regarding copollutant confounding is reduced with
9 additional recent epidemiologic results showing that associations between ambient NO2
10 concentrations and respiratory effects remain positive in copollutant models with PM25,
11 PMio, SO2, O3, or as examined in fewer studies, BC, EC, UFP, or CO. Copollutant
12 models have limitations (Section 1.5) and were not analyzed for every potentially
13 correlated copollutant or study. Therefore, evidence for NO2-induced increases in airway
14 responsiveness in adults with asthma from controlled human exposure studies is key in
15 providing biological plausibility for effects on asthma exacerbations. Further, evidence
16 for NO2-related oxidative stress and inflammation (including allergic inflammation)
17 describes other biological processes informing the modes of action for exacerbations of
18 asthma. Thus, the epidemiologic and experimental evidence together demonstrate the
19 independent respiratory effects of short-term NO2 exposure and together are the basis of
20 concluding a causal relationship.
Respiratory Effects Associated with Long-term NO2 Exposure
21 The strongest evidence indicating that there is likely to be a causal relationship between
22 long-term NO2 exposure and respiratory effects comprises the associations consistently
23 found between ambient NO2 concentrations and asthma incidence in children in diverse
24 geographical locations (Section 5.2.17. Table 5-9). This recent evidence is provided by
25 several high-quality single- and multi-city studies characterized by prospective follow-up
26 of children, in several cases from birth to ages 8-12 years. Associations were found with
27 adjustment for potential confounding by socioeconomic status (SES), smoking exposure,
28 housing characteristics, and gas stove use. Associations were found with NO2 obtained
29 from central site monitors and more spatially-resolved residential outdoor NO2 estimated
30 with LUR or dispersion models. These models predicted ambient concentrations that
31 correlated well with measured concentrations (R2 = 0.42 to 0.69). Asthma incidence was
32 associated with the average NO2 from the first year of life and NO2 averaged over
33 multiple years (study mean concentrations: 14 to 21 ppb). A relationship between long-
34 term NO2 exposure and asthma is supported by evidence for respiratory effects
35 associated with short-term exposure. Several epidemiologic found increases in respiratory
36 symptoms and pulmonary inflammation in children in the general population associated
November 2013 1-16 DRAFT: Do Not Cite or Quote
-------
1 with short-term increases in ambient NO2 concentrations. Recurrent episodes of
2 inflammation and respiratory symptoms are diagnostic indicators of asthma. An effect on
3 asthma incidence also is supported by evidence from prospective studies showing
4 increases in respiratory symptoms in children with asthma in association with long-term
5 NO2 exposure.
6 There is limited biological plausibility for NO 2 -related increases in asthma incidence,
7 provided by the small body of results showing increased airway responsiveness in guinea
8 pigs in response to short- and long-term NO2 exposure (1,000 to 4,000 ppb). NO2
9 exposure (2,000 to 4,000 ppb) also induced Th2 immune responses with short-term
10 exposure in a controlled human exposure study of healthy adults and with short- and
11 long-term NO2 exposure in guinea pigs. Enhanced Th2 immune responses can contribute
12 to the development of asthma. This evidence also provides biological plausibility for the
13 NO2-associated increases in allergic sensitization found in a few prospective studies of
14 children. Results from a few toxicological studies also describe other key events that
15 inform modes of action for NO2-related asthma development. NO2 induced airway
16 responsiveness with increased airway resistance suggesting the involvement of airway
17 obstruction and airway remodeling. A few animal studies found increased oxidative stress
18 and inflammation with long-term NO2 exposure but not consistently across studies.
19 Continued evidence in children for decreases in lung function and partially irreversible
20 decreases in lung function growth with long-term NO2 exposure also supports a likely to
21 be a causal relationship. Single- and multi-city studies found associations with 6-month
22 and annual average NO2 before lung function measurement, the average NO2 in the first
23 year of life, and 10-year lifetime average NO2 (study mean concentrations: 14 to 34 ppb).
24 Changes in lung function may represent the effect of short-term exposure occurring at the
25 time of measurement. Some studies reduced the potential for bias from the effects of
26 short-term exposure with observations that long-term NO2-associated decreases in lung
27 function persisted with adjustment for short-term NO2 exposure. Limited biological
28 plausibility for decreases in lung function resulting from long-term NO2 exposure is
29 provided by findings for increases in airway responsiveness in guinea pigs induced by
30 short- and long-term NO2 exposure (1,000 to 4,000 ppb). NO2 exposure did not alter lung
31 function in experimental animals, and the morphological effects induced by NO2
32 exposure in adult animals such as hyperproliferation of lung epithelial cells and fibrosis
33 are not directly related to the effects on lung function described in studies of children.
34 Recent epidemiologic studies also provide new evidence for associations between long-
35 term NO2 exposure and respiratory effects in adults, with the most consistent findings for
36 asthma. Associations are less consistent for hospital admissions for COPD or asthma, and
37 studies did not adjust for effects of short-term exposure. Results also are inconsistent for
November 2013 1-17 DRAFT: Do Not Cite or Quote
-------
1 respiratory mortality associated with long-term NO2 exposure. A recent case-control
2 study found higher NO2 exposures among adults with hospital admissions or ED visits
3 for pneumonia, but the most robust evidence for impaired host defense consists of the
4 findings in experimental animals for NO2-induced increases in mortality following
5 bacterial or viral infection and changes in alveolar macrophage function.
6 For any given respiratory effect, copollutant-adjusted results are available for one or two
7 epidemiologic studies. These studies of lung function, respiratory symptoms, and asthma
8 generally found associations with long-term NO2 exposure to change little when adjusted
9 for copollutants such as O3, SO2, PMi0, PM2 5, or EC. However, because of the limited
10 examination of copollutant confounding and the limited biological plausibility, there
11 remains some uncertainty regarding the independent effects of long-term NO2 exposure.
12 In conclusion, evidence indicates that there is likely to be a causal relationship between
13 long-term NO2 exposure and respiratory effects based primarily on recent epidemiologic
14 findings in children for increases in asthma incidence and collective results for decreases
15 in lung function and partially irreversible decreases in lung function growth. Supporting
16 evidence includes NO2-related increases in respiratory symptoms in children with
17 asthma, allergic sensitization in children, asthma in adults, and impaired host defense in
18 animal models. NO2 associations with respiratory effects remain positive in copollutant
19 models, and findings for NO2-induced airway responsiveness and development of Th2
20 immune responses in experimental studies provide biological plausibility. However, the
21 limited nature of such evidence does not conclusively demonstrate that respiratory effects
22 of long-term NO2 exposure are independent of other traffic-related pollutants.
1.4.3 Dosimetry and Modes of Action Informing Extrapulmonary Effects
23 Health effects in various organ systems have been found in relation with NO2, NO, and
24 NOX exposure in epidemiologic studies and NO2 exposure in experimental studies.
25 However, in contrast with the respiratory tract, there is weak characterization of how
26 ambient-relevant exposures to NO2 or NO affect processes that may underlie the health
27 effects observed beyond the respiratory system (Section 3.2.2). As described in Section
28 1.4.1. NO2 is transformed by reactions in the ELF. A major reaction product, nitrite, can
29 gain access to the blood, where it can react with red blood cell hemoglobin to form
30 nitrosylhemoglobin, methemoglobin, and nitrate. Some of these reaction products of NO2
31 have been found in the blood of experimental animals but with higher than ambient-
32 relevant NO2 exposure concentrations (Section 3.2.2.4). Further, nitrite has not been
33 shown to have negative health effects. Another process that could mediate
34 extrapulmonary effects of inhaled NO2 is the spillover of mediators from the respiratory
November 2013 1-18 DRAFT: Do Not Cite or Quote
-------
1 tract into the blood. For example, a recent controlled human exposure study found an
2 increase in a vasoactive mediator in the plasma of NO2-exposed adults (Section 3.3.5).
3 Inhaled NO can diffuse across the alveolar capillary barrier into the blood and bind with
4 hemoglobin in red blood cells to form nitrosylhemoglobin, methemoglobin, and nitrate
5 (Section 1.4.1). Methemoglobin has been linked with health effects, but increases in
6 blood levels of methemoglobin and nitrosylhemoglobin are not consistently found with
7 inhalation of ambient-relevant concentrations of NO (Section 3.3.3). NO is produced
8 endogenously from nitrates and nitrites derived from diet and enzymatic pathways that
9 are enhanced during inflammation. Endogenous NO can affect diverse physiological
10 processes through interactions with heme proteins, other transition metal-containing
11 proteins, and radical species. However, it is not clear if ambient-relevant concentrations
12 of inhaled NO alter physiological processes that are affected by endogenous NO, in part,
13 because endogenous concentrations of NO in the respiratory tract are similar to those
14 found in ambient air. Because there is at least some biological plausibility for negative
15 effects resulting from ambient-relevant NO2 exposure but not from ambient-relevant NO
16 exposure, extrapulmonary health effects associated with ambient NOX are considered to
17 be reflecting associations with NO2 and are considered in causal determinations for NO2.
1.4.4 Cardiovascular Effects
18 Although the biological processes mediating the extrapulmonary effects of ambient-
19 relevant NO2 exposures are not well understood (Section 1.4.3). other lines of evidence
20 support relationships between both short-term and long-term NO2 exposure and
21 cardiovascular effects. Results from recent epidemiologic studies are the primary basis
22 for strengthening the causal determinations from the 2008 ISA for Oxides of Nitrogen
23 (Table 1-1). For short-term exposure, the causal determination is strengthened from
24 inadequate to infer a causal relationship to likely to be a causal relationship because
25 recent epidemiologic studies consistently find cardiovascular hospital admissions and
26 mortality and changes in measures of cardiovascular physiology in association with
27 ambient NO2 concentrations and reduce previous uncertainty about copollutant
28 confounding. For long-term exposure, the conclusion is strengthened from inadequate to
29 infer a causal relationship to suggestive of a causal relationship because of associations
30 with heart failure, myocardial infarction, stroke, and cardiovascular mortality found in
31 some recent epidemiologic studies. The single previous long-term exposure study did not
32 find NO2-associated increases in cardiovascular events. A common uncertainty in the
33 relationships with cardiovascular effects is the limited biological plausibility, including
34 the effects of inhaled NO2 on changes in mediators in the blood and subsequent events
35 that inform the modes of action for cardiovascular effects. Epidemiologic evidence is
November 2013 1-19 DRAFT: Do Not Cite or Quote
-------
1 more consistent for effects associated with short-term NO2 exposure than with long-term
2 exposure, and more copollutant-adjusted results indicate independent NO2 associations.
Cardiovascular Effects Associated with Short-term NO2 Exposure
3 Evidence indicates that there is likely to be a causal relationship between short-term NO2
4 exposure and cardiovascular effects based strongly on recent epidemiologic studies
5 consistently indicating associations of NO2 with increases in cardiovascular hospital
6 admissions and mortality in diverse geographic locations (Section 4.3.9. Table 4-36). The
7 evidence for cardiovascular hospital admissions and mortality is substantiated by the
8 associations found in high-quality studies that were conducted over several years and
9 adjusted for potential confounding by weather and long-term time trends. Particularly for
10 cardiovascular-related mortality, but also for hospital admissions, there are multicity
11 studies demonstrating the robustness of association with short-term increases in ambient
12 NO2 in data pooled across cities. Recent studies of cardiovascular hospital admissions
13 and mortality reduce an important uncertainty identified in the 2008 ISA regarding the
14 potential for copollutant confounding. Several studies showed that associations with NO2
15 remain positive with adjustment for copollutants such as PMi0, SO2, O3, or in some but
16 not all locations, PM25 or CO. In studies of cardiovascular hospital admissions and
17 mortality, city-specific mean 24-h avg NO2 were 12 to 41 ppb, 90th percentiles were 22
18 to 100 ppb, and maximum concentrations were 19 to 132 ppb. For 1-h maxNO2, overall
19 study mean concentrations were 43 and 46 ppb, and 90th percentiles were 66 and 68 ppb.
20 The coherence between evidence for increases in cardiovascular mortality and hospital
21 admissions further supports a relationship between cardiovascular effects and short-term
22 NO2 exposure. The strongest evidence for cardiovascular hospital admissions is for
23 ischemic heart disease (IHD), the leading cause of death in the world (Finegold et al..
24 2013). Limited biological plausibility is provided by results showing that NO2 exposure
25 is related to changes in cardiovascular physiology that may lead to cardiovascular
26 hospital admissions or mortality. Decreases in heart rate variability (HRV) have been
27 linked with premature mortality and are found in association with increases in ambient
28 NO2 in recent epidemiologic studies of adults with cardiovascular disease and in a recent
29 controlled human exposure study of healthy adults. Consistent with hospital admissions
30 for IHD, a few recent epidemiologic studies in adults with coronary artery disease found
31 NO2-associated changes in ventricular repolarization, which are markers of myocardial
32 ischemia. However, neither epidemiologic nor controlled human exposure studies
33 consistently demonstrate effects on cerebrovascular diseases, arrhythmia, or blood
34 pressure. Some experimental and epidemiologic studies indicate NO2-related increases in
35 inflammation and oxidative stress. Different mediators were examined in humans and
36 rodents, but these results indicate effects on other key events that inform the modes of
November 2013 1 -20 DRAFT: Do Not Cite or Quote
-------
1 action for IHD and cardiovascular mortality. However, because cardiovascular effects are
2 not consistently demonstrated in experimental studies, the available evidence does not
3 conclusively demonstrate the independent cardiovascular effects of NO2 exposure.
4 In conclusion, evidence indicates that there is likely to be a causal relationship between
5 short-term NO2 exposure and cardiovascular effects, based strongly on epidemiologic
6 evidence from single- and multi-city studies for NO2-related increases in cardiovascular
7 hospital admissions, particularly for IHD, and mortality. Previous uncertainty regarding
8 copollutant confounding is reduced by recent results showing that NO 2 -associated
9 increases in cardiovascular hospital admissions and mortality remain positive with
10 adjustment for PMi0, SO2, O3, or in some but not all locations, PM25 or CO. Biological
11 plausibility is provided by some findings for NO2-associated changes in ventricular
12 repolarization and HRV. However, because experimental evidence is inconsistent and
13 epidemiologic studies did not clearly exclude confounding by the array of potentially
14 correlated traffic-related copollutants, some uncertainty remains as to whether the
15 cardiovascular effects of NO2 exposure are independent of other traffic-related pollutants.
Cardiovascular Effects Associated with Long-term NO2 Exposure
16 Evidence is suggestive of a causal relationship between long-term NO2 exposure and
17 cardiovascular effects based on some recent epidemiologic studies indicating associations
18 of NO2 or NOX with myocardial infarction, heart failure, and stroke but other studies
19 showing no associations (Section 5.3.6. Table 5-12). The recent epidemiologic studies
20 contributing to the evidence base are considered to be high quality based on their large
21 sample sizes, prospective follow up of subjects (up to 9 years), and adjustment for
22 potential confounding by age, sex, SES, cardiovascular disease, and other comorbid
23 factors. Associations were found with ambient NO2 or NOX averaged over 1 to 9 years
24 (study mean concentrations: 12 and 34 ppb NO2 and 96 ppb NOX). From the small
25 number of studies, it is difficult to assess whether the studies finding no association
26 clearly differed in NO2 or NOX concentrations, duration of follow up, or other factors.
27 The cardiovascular morbidity findings have limited support from mortality results. IHD
28 includes myocardial infarction and can lead to heart failure, and some but not all studies
29 found associations between long-term NO2 exposure and mortality from cardiovascular
30 causes, including IHD. Support is provided by the associations observed between short-
31 term increases in ambient NO2 concentration and increases in IHD hospital admissions.
32 In addition to the uncertainty due to inconsistent epidemiologic evidence for
33 cardiovascular effects associated with long-term NO2 exposure, there is uncertainty in the
34 relationship because of limited biological plausibility. The few available recent
35 epidemiologic studies indicate NO2-associated increases in arterial stiffness and
November 2013 1-21 DRAFT: Do Not Cite or Quote
-------
1 decreases in HRV (study mean concentrations: 18 or 23 ppb for annual average), which
2 are related to myocardial infarction and mortality, respectively. Long-term NO2 exposure
3 of rats induced dyslipidemia, a risk factor for IHD. There is weak evidence describing
4 other key events that inform modes of action for NO2-related cardiovascular effects.
5 Epidemiologic evidence is inconsistent for associations between long-term NO2 exposure
6 and increases in inflammation in adults, and short-term NO2 exposure induced increases
7 in indicators of inflammation and oxidative stress in some but not all experimental
8 studies.
9 In summary, evidence is suggestive of a causal relationship between long-term NO2
10 exposure and cardiovascular effects based on associations between NO2 or NOX and
11 myocardial infarction, heart failure, and stroke found in some recent epidemiologic
12 studies but no association found in other studies. There also is uncertainty because there
13 is inconsistent evidence across disciplines for effects on an array of key events that
14 inform modes of action for cardiovascular effects to provide biological plausibility.
1.4.5 Total Mortality
15 The causal determinations for relationships of total mortality with both short-term and
16 long-term NO2 exposure are strengthened from the 2008 ISA based on evidence from
17 recent epidemiologic studies. Conclusions are drawn by evaluating mortality from all
18 nonaccidental causes. For short-term exposure, the conclusion for total mortality is
19 strengthened from suggestive of a causal relationship to likely to be a causal relationship
20 because recent high-quality epidemiologic studies add evidence for associations with
21 NO2 and reduce previous uncertainty regarding copollutant confounding (Table 1-1). For
22 long-term exposure, the conclusion is strengthened from inadequate to infer a causal
23 relationship to suggestive of a causal relationship because some recent high-quality
24 studies reported associations with NO2 or NOX, whereas the previous evidence was more
25 limited and inconsistent. A common uncertainty in the relationships for short-term and
26 long-term NO2 exposure is the limited biological plausibility, which is informed by the
27 extent to which evidence indicates effects on a spectrum of cardiovascular and respiratory
28 morbidity and mortality outcomes. The limited coherence of findings across scientific
29 disciplines and across the array of respiratory (Section 1.4.2) and cardiovascular (Section
30 1.4.4) morbidity and mortality effects produces some uncertainty as to what spectrum of
31 effects NO2 exposure may induce to lead to increases in mortality. Evidence is more
32 consistent for total mortality associated with short-term NO2 exposure than with long-
33 term exposure, and more results indicate NO2 associations with copollutant adjustment.
November 2013 1 -22 DRAFT: Do Not Cite or Quote
-------
Total Mortality Associated with Short-term NO2 Exposure
1 Evidence indicates that there is likely to be a causal relationship between short-term NO2
2 exposure and total mortality based on consistent evidence from several recent high-
3 quality epidemiologic studies conducted in diverse geographic locations (Section 4.4.8.
4 Table 4-41). These studies are considered to be high quality based on multicity analyses
5 that apply a consistent statistical model to data pooled across cities, thus minimizing the
6 potential for publication bias. These studies also adequately adjust for potential
7 confounding by weather and long-term time trends. Individual cities had mean 24-h avg
8 NO2 concentrations 9.2 to 55 ppb and maximum concentrations 55 to 161 ppb. City-
9 specific 1-h max NO2 mean concentrations were 16 to 81 ppb, 90th percentile
10 concentrations were 33 to 133 ppb, and maximum concentrations were 55 to 161 ppb.
11 Recent studies reduce the previously identified uncertainty regarding copollutant
12 confounding by showing that NO2-related mortality effect estimates are similar with
13 adjustment for PMi0, SO2, or O3. A relationship between short-term NO2 exposure and
14 total mortality also is supported by the consistent evidence for NO2-associated increases
15 in hospital admissions for cardiovascular diseases, which are the leading cause of deaths
16 in the U.S. (35% as cited in Hoyert and Xu. 2012). However, evidence is inconsistent for
17 NO2-related changes in measures of cardiovascular physiology such as HRV, arrhythmia,
18 and blood pressure (Section 1.4.4). Respiratory causes comprise a smaller fraction of
19 mortality (9%); however, COPD and respiratory infections are among the leading causes
20 of mortality in the world. There is limited coherence among lines of evidence indicating
21 NO2-related effects on COPD and respiratory infection (Section 1.4.2). Strong evidence
22 demonstrates NO 2-related exacerbations of asthma, but asthma is not a leading cause of
23 mortality. Thus, is not entirely clear what spectrum of cardiovascular and respiratory
24 effects NO2 exposure may induce to lead to mortality.
25 In summary, evidence indicates that there is likely to be a causal relationship between
26 short-term NO2 exposure and total mortality based on consistent evidence of association
27 in previous and recent multicity studies. Results show robust NO2 associations with
28 adjustment for PM10, SO2, or O3. Because the available evidence does not conclusively
29 identify an independent effect of NO2 from those of other measured or unmeasured
30 traffic-related pollutants or clearly characterize the biological processes by which NO2
31 exposure may lead to total mortality, there remains some uncertainty in the relationship
32 between short-term NO2 exposure and total mortality.
Total Mortality Associated with Long-term NO2 Exposure
33 Evidence is suggestive of a causal relationship between long-term NO2 exposure and
34 total mortality based on some high-quality recent studies in the U.S. and Europe showing
November 2013 1 -23 DRAFT: Do Not Cite or Quote
-------
1 associations with NO2 or NOX but other high-quality studies showing no association
2 (Section 5.5). Studies are considered to be high quality based on large sample sizes; long
3 periods of follow-up up to 26 years; and adjustment for potential confounding by age,
4 sex, smoking, education, comorbid factors, and in some cases, community-level
5 characteristics. Results are inconsistent between the Harvard Six Cities and the American
6 Cancer Society cohorts, both of which are considered to be seminal air pollution studies
7 of multiple U.S. cities. Increases in total mortality were found in association with NO2
8 concentrations averaged over 1 to 16 years and assessed for the year of death and for
9 periods up to 20 years before death. Study mean ambient concentrations were 14 to
10 34 ppb. There is no clear indication that the mean ambient NO2 concentrations or
11 exposure period examined differed in the studies finding no association. As in the 2008
12 ISA, associations were found with copollutants such as PM2 5, and copollutant models
13 generally were not analyzed. However, some recent studies found associations with NO2
14 with adjustment for traffic, indicating that NO2 was not only serving as an indicator of
15 traffic. Similar to mortality related to short-term NO2 exposure, there is limited evidence
16 to explain the biological processes by which long-term NO2 exposure may lead to
17 mortality. As described in Section 1.4.4. there is inconsistent epidemiologic evidence for
18 associations between long-term NO2 or NOX exposure and cardiovascular morbidity and
19 limited biological plausibility. There is limited available evidence for associations
20 between long-term NO2 exposure and increases in asthma and hospital admissions for
21 COPD in adults to support associations found between long-term exposure and total
22 mortality.
23 In summary, evidence is suggestive of a causal relationship between long-term NO2
24 exposure and total mortality based on associations found with ambient NO2 or NOX
25 concentrations in some previous and recent high-quality studies but not in other studies of
26 comparable quality. The limited coherence of findings across a spectrum of
27 cardiovascular and respiratory morbidity outcomes also produces uncertainty regarding
28 the biological processes by which long-term NO2 exposure may lead to mortality.
1.4.6 Reproductive and Developmental Effects
29 The 2008 ISA for Oxides of Nitrogen formed a single causal determination for the
30 heterogeneous group of reproductive and developmental effects. The current draft ISA
31 presents separate conclusions for more defined subcategories of outcomes that are likely
32 to occur by different biological processes and exposure patterns over different lifestages:
33 (1) fertility, reproduction, and pregnancy (Section 5.4.2); (2) birth outcomes (Section
34 5.4.3); and (3) postnatal development (Section 5.4.4). The causal determination is
35 strengthened from inadequate to infer a causal relationship for the broad category to
November 2013 1 -24 DRAFT: Do Not Cite or Quote
-------
1 suggestive of a causal relationship with long-term NO2 exposure for each of the three
2 subcategories, based on some high-quality studies finding associations with monitored or
3 modeled ambient NO2 or NOX concentrations (Table 1-1). Previous epidemiologic
4 evidence was limited and inconsistent for effects on birth outcomes and there was weak
5 evidence in experimental animals to provide biological plausibility. This weak biological
6 plausibility remains an uncertainty for all three subcategories of reproductive and
7 developmental effects, as there are no recent animal toxicological studies to consider.
8 Another uncertainty that pertains to all three subcategories is the lack of association
9 found in some recent high-quality epidemiologic studies.
Fertility, Reproduction, and Pregnancy
10 Evidence is suggestive of a causal relationship between long-term NO2 exposure and
11 effects on fertility, reproduction, and pregnancy based primarily on limited but consistent
12 epidemiologic evidence for associations of pregnancy NO2 or NOX exposure with
13 pre-eclampsia (Section 5.4.5. Table 5-15). a pregnancy complication related to
14 hypertension and protein in the urine of pregnant women. Single- and multi-city studies
15 found pre-eclampsia in association with NO2 (mean 23 ppb for entire pregnancy) and
16 NOX (means: 7.2 for entire pregnancy, 7.5 ppb for 3rd trimester) modeled for residential
17 locations and with adjustment for potential confounding by maternal age, smoking, SES,
18 diabetes, and parity. There are no toxicological studies on effects related to pre-eclampsia
19 to inform biological plausibility, and epidemiologic evidence for pregnancy-induced
20 hypertension is inconsistent. Reduced fertility was found in association with short- and
21 long-term NO2 exposure in a recent epidemiologic study of women undergoing in vitro
22 fertilization, but a study in rats found no effect on fertility. Epidemiologic evidence does
23 not consistently indicate associations with gestational diabetes or reduced placental
24 growth and function, and epidemiologic and toxicological evidence does not indicate
25 effects on sperm quality.
Birth Outcomes
26 Evidence is suggestive of a causal relationship between NO2 exposure and effects on
27 birth outcomes because while epidemiologic studies found associations of higher prenatal
28 NO2 exposure with fetal growth restriction, associations with other outcomes are not
29 consistent (Section 5.4.5. Table 5-15). The studies of fetal growth restriction are
30 considered to be high quality because they adjust for potential confounding by maternal
31 age, SES, smoking, alcohol use, and season of conception and assess fetal growth
32 restriction with fetal or neonatal physical measurements. Recent epidemiologic studies
33 found fetal growth restriction in association with ambient residential NO2 (means: 7.8 to
November 2013 1 -25 DRAFT: Do Not Cite or Quote
-------
1 21 ppb) estimated from LUR or dispersion models. Associations were stronger among
2 children with mothers who spent more time at home and less outdoor time in locations
3 other than home. Consideration of time activity patterns may have improved the
4 relationship of modeled exposure estimates to personal exposures. Associations were
5 found with early pregnancy, 3rd trimester, and entire pregnancy NO2 exposure, without a
6 clear indication of risk differing among exposure periods. Birth weight was inconsistently
7 associated with NO2 exposure estimated using LUR and central site concentrations, but
8 there is limited biological plausibility provided by findings of decreased birth weight in a
9 study of rats. Effects on other key events informing modes of action are unclear. Prenatal
10 ambient NO2 exposure was associated with an indicator of inflammation in human cord
11 blood, but its role in influencing birth outcomes is not clear. Epidemiologic evidence for
12 effects on other birth outcomes such as preterm birth and birth defects is inconsistent.
Postnatal Development
13 The evidence is suggestive of a causal relationship between NO2 exposure and effects on
14 postnatal development based mostly on previous and recent epidemiologic associations
15 observed between NO2 exposure (means: 34 ppb for 6-mo avg, 14 to 21 ppb for annual
16 avg) and partially irreversible decreases in lung function growth in children (Section
17 5.4.5. Table 5-15). Studies adjusted for potential confounding by age, body mass index,
18 and smoking exposure, but SES was not examined. Copollutants also were associated
19 with decreases in lung function growth, and few studies adjusted for copollutants.
20 Because the changes in lung morphology induced by NO2 exposure in experimental
21 animals do not inform the changes observed in children, uncertainty remains regarding
22 the independent effect of NO2 exposure. Another line of evidence contributing to the
23 causal determination consists of decreases in cognitive function in children found in
24 association with prenatal (mean: 16 ppb) or concurrent outdoor school annual average
25 (mean: 17 ppb) NO2 in some recent epidemiologic studies. Other studies with
26 comparable mean NO2 concentrations did not find NO2-associated decreases in cognitive
27 function, and evidence was absent or inconsistent for other neurodevelopmental effects
28 such as attention, motor function, and psychological distress. Associations with decreases
29 in cognitive function were found with adjustment for SES, and in one study, noise.
30 However, potential confounding by copollutants including lead or PM was not examined.
31 Results from animal toxicological studies do not clearly indicate the effects of NO2
32 exposure on neurodevelopmental outcomes, and it is not clear how comparable the
33 endpoints examined in children and rodents are. Effects of prenatal or infancy NO2
34 exposure on other postnatal development effects also are variable. Both epidemiologic
35 and toxicological evidence for infant mortality is inconsistent. Physical development was
November 2013 1 -26 DRAFT: Do Not Cite or Quote
-------
1 not affected by NO2 exposure in all animal toxicological studies and not examined in
2 epidemiologic studies.
1.4.7 Cancer
3 For cancer, the causal determination is strengthened from inadequate to infer a causal
4 relationship in the 2008 ISA for Oxides of Nitrogen to suggestive of a causal relationship
5 with NO 2 exposure in the current draft ISA because among the several recent
6 epidemiologic studies that examined lung cancer incidence or mortality, some high-
7 quality studies found associations with monitored or modeled ambient NO2 or NOX
8 concentrations (Table 1-1, Table 5-21. Section 5.6.12). Previous epidemiologic evidence
9 was more limited, and experimental evidence that NO2 is a complete carcinogen was
10 lacking. There is uncertainty regarding a relationship with NO2 because some recent
11 high-quality epidemiologic studies did not find associations with lung cancer incidence or
12 mortality. Studies are considered to be high quality based on the large numbers of cancer
13 cases examined, the follow-up of adults over 7-30 years, and adjustment for several
14 potential confounding factors such as SES, smoking, diet, and occupational exposures.
15 Associations were found with a wide range of exposure durations: NO2 or NOX averaged
16 over 1 year at the beginning of follow-up to a 30-year average before the outcome
17 (means: 14- to 23 ppb for NO2 and 11 to 42 (ig/m3 for NOX). Models used to estimate
18 exposure were validated to ensure that the data were of sufficient quality. Studies not
19 finding associations did not differ in mean NO2 or NOX concentrations, exposure
20 duration examined, or exposure assessment method (central site monitors or modeled
21 estimates of exposure at residences). Several studies finding associations of NO2 or NOX
22 with lung cancer incidence or mortality also reported associations with copollutants such
23 as PM2 5, PMio, or CO. Another uncertainty in the relationship between NO2 exposure
24 and cancer is the limited biological plausibility. NO2 did not independently induce lung
25 tumor formation in various animal models, but a potential role for high-concentration
26 exposures in tumor promotion is indicated by findings of NO2 exposures of 4,000 to
27 10,000 ppb increasing lung tumors in mice with spontaneously high tumor rates or with
28 co-exposure to diesel exhaust particles or known carcinogens. A few findings of
29 formation of secondary oxidation products in the respiratory tract (Section 1.4.1) and
30 NO2-induced increases in hyperplasia of the lung epithelium of rodents describe plausible
31 biological processes mediating NO2-related lung cancer effects. While NO2 exposure
32 impaired host defense in animal models (Section 5.2.9). parameters more directly linked
33 to antitumor immunity such as cytotoxic or regulatory T cells and interferon-gamma were
34 not studied.
November 2013 1 -27 DRAFT: Do Not Cite or Quote
-------
1 A few recent epidemiologic studies indicate associations between NO2 exposure and
2 leukemia, bladder cancer, prostate cancer, and breast cancer. However, there is a lack of
3 biological plausibility for these findings. There are inconsistent findings for NO2-induced
4 (higher than ambient-relevant concentrations) mutagenicity and carcinogenicity in bone
5 marrow, spermatocytes, and lymphocytes. Further, the effects of inhaled NO2 on
6 transforming other chemicals in the body into mutagens or carcinogens are found only
7 with higher than ambient-relevant NO2 exposure concentrations, i.e., 16,500 to 20,000
8 ppb.
9 In conclusion, evidence is suggestive of a causal relationship between long-term NO2
10 exposure and cancer based primarily on associations between ambient NO2 or NOX
11 concentrations and lung cancer incidence and mortality found in some previous and
12 recent high-quality epidemiologic studies but not in other studies of comparable quality.
13 There also is uncertainty in the relationship between NO2 exposure and lung cancer
14 because biological plausibility is limited. There are findings for lung tumor promotion
15 and hyperplasia of lung epithelial cells with NO2 exposure, some at higher than ambient-
16 relevant concentrations, but no evidence for direct effects on carcinogenesis
17
November 2013 1 -28 DRAFT: Do Not Cite or Quote
-------
Table 1-1 Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Causal Determination
NO2 Concentrations
Associated with Effects0
Respiratory Effects
Short-term Exposure (Section 4.2)
2008 ISA - Sufficient to Infer a Likely Causal Relationship
Current draft ISA - Causal Relationship
Key Evidence:
(Table 4-23)
Strongest evidence is for NO2-related increases in asthma exacerbations indicated as increases in
asthma hospital admissions and ED visits in single- and multi-city studies in diverse populations as well
as increases in respiratory symptoms and pulmonary inflammation and decreases in lung function in
children with asthma. Associations found with adjustment for weather, time trends and in copollutant
models for PM-io, PIVb.s, SC>2, or O^, or examined in fewer locations, EC, BC, CO, or UFP.
Biological plausibility demonstrated by NC>2-induced increases in airway responsiveness of adults with
asthma in controlled human exposure studies and effects on other key events informing modes of action
including initiation of inflammation, allergic inflammation and oxidative stress.
Clear evidence for impaired host defense in experimental animals, with some epidemiologic associations
with respiratory infections in children. Some evidence for effects on COPD indicated as NO2-related
increases in hospital admissions and ED visits. Inconsistent findings for respiratory symptoms or
decreases in lung function in adults with COPD. Consistent epidemiologic evidence for increases in
respiratory symptoms and pulmonary inflammation in children in the general population. Consistent
NO2-related increases in respiratory mortality.
Overall study ambient
maximums:
24-h avg: 52 to 80 ppb
1-h max: 59 to 298 ppb
24-h avg personal
maximums: 48, 106 ppb
Airway responsiveness: 200
to 300 ppb for 30 min, 100
ppb for 1 hour
Mortality from infection in
animals: 1,500 to 5,000 ppb
Recent studies add:
Additional epidemiologic evidence that associations for NO2 are independent of many traffic-related
copollutants; new epidemiologic evidence for associations with decreases in lung function in children
with asthma.
November 2013
1-29
DRAFT: Do Not Cite or Quorte
-------
Table 1-1 (Continued): Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Causal Determination
NC<2 Concentrations
Associated with Effects0
Respiratory Effects
Long-term Exposure (Section 5.2)
2008 ISA - Suggestive but not Sufficient to Infer a Causal Relationship
Current draft ISA - Likely to be a Causal Relationship
Key Evidence:
(Table 5-9)
Strongest evidence is for associations of ambient NO2 averaged over 1-3 yr with asthma incidence in
several diverse cohorts of children, some followed from birth. Supporting evidence for increases in
respiratory symptoms in children with asthma and allergic sensitization in children. Associations found
with adjustment for SES, smoking exposure and housing characteristics. Supporting evidence for
asthma, respiratory symptoms and COPD hospital admissions in adults.
Limited biological plausibility demonstrated by increases in airway responsiveness and Th2 responses in
guinea pigs with NC>2 long-term exposure; development of Th2 allergic phenotype with short-term
exposure in humans and guinea pigs.
Consistent evidence in children for decreases in lung function and partially irreversible decreases in lung
function growth. Lung edema, hypertrophy, fibrotic changes found in adult experimental animals not
related to changes in children. Clear animal toxicological evidence for impaired host defense.
NO2 associations for lung function, bronchitic symptoms and asthma remain robust in the few
copollutant models analyzed with PM-io, PlVb.s, O^, SC>2, or EC.
Recent studies add: New evidence for asthma incidence and respiratory symptoms in children; respiratory effects in adults.
Uncertainty/Limitation:
Independent effect of NC>2 from copollutants not widely characterized; limited biological plausibility, i.e.,
characterization of spectrum of key events informing mode of action.
Overall study ambient
means:
Children: 14 to 21 ppb
annual avg, 34 ppb 6-mo
avg Adults: 8 to 20 ppb
annual avg
Th2 responses: 2,000 ppb
in humans, 4 days and
3,000 ppb in guinea pigs,
2 weeks
Airway responsiveness:
1,000 to 4,000 ppb in
guinea pigs, 6, 12 weeks
Impaired host defense in
animals: 500, 5,000 ppb,
200 ppb base + 800 ppb
spike
Cardiovascular
Effects
Short-term Exposure (Section 4.3)
2008 ISA- Inadequate to Infer the Presence or Absence of a Causal Relationship
Current draft ISA - Likely to be a Causal Relationship
Key Evidence:
(Table 4-36)
Recent studies add:
Consistent evidence for increases in cardiovascular hospital admissions and ED visits, particularly for
IHD, as well as cardiovascular mortality in single- and multi-city studies in diverse populations.
Associations found with adjustment for weather, time trends and in copollutant models for PM-io, SO2,
Os, or in some but not all locations, PlVb.s or CO.
Limited biological plausibility demonstrated by decreases in HRV and changes in ventricular
repolarization in epidemiologic studies of adults with cardiovascular disease but inconsistent changes in
HRV in controlled human exposure studies. Changes in other cardiovascular effects generally not found
in epidemiologic or experimental studies. Weak evidence to describe other key events informing mode of
action with observations of oxidative stress and endothelial inflammation in some experimental studies
and associations with indicators of inflammation in some but not all epidemiologic studies.
Epidemiologic associations of NO2 with cardiovascular hospital admissions, ED visits, and mortality that
are independent of many copollutants, consistent epidemiologic evidence for decreases in HRV.
Individual city ambient
24-havg: 90th: 22 to 100
ppb, Maximum: 19 to 132
ppb
Overall study ambient
1-h max: 90th: 66 and 68
ppb
Oxidative stress in rats:
5,320 ppb, Inflammation in
rats: 2,660 and 5,320 ppb
Inflammation in human cells
exposed to human plasma,
November 2013
1-30
DRAFT: Do Not Cite or Quote
-------
Table 1-1 (Continued): Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Uncertainty/Limitation:
Cardiovascular
Effects
Key Evidence:
(Table 5-1 2)
Recent studies add:
Uncertainty/Limitation:
Causal Determination13
Limited biological plausibility, i.e., characterization of spectrum of key events informing mode of action.
Long-term Exposure (Section 5.3)
2008 ISA- Inadequate to Infer the Presence or Absence of a Causal Relationship
Current draft ISA - Suggestive of a Causal Relationship
Evidence from some recent high-quality, large cohort studies for associations of NC>2 or NOx averaged
over approximately 1 to 9 yr with myocardial infarction, heart failure and stroke. Associations found with
adjustment for age, sex, SES, cardiovascular disease and other comorbid factors. Limited support from
increases in cardiovascular mortality. No association found with myocardial infarction in some recent
studies of comparable quality or with cardiovascular events in a previous study.
Limited biological plausibility demonstrated by findings for arterial stiffness or decrease in HRV in a few
epidemiologic studies and dyslipidemia in a study of rats. Weak evidence to describe other key events
informing mode of action. Increased oxidative stress and endothelial inflammation found in some but not
all experimental studies with short-term exposure. Inconsistent epidemiologic evidence for long-term
NC>2-associated increases in indicators of systemic inflammation in adults.
New evidence for myocardial infarction and heart failure in some recent high-quality cohort studies.
Inconsistent epidemiologic evidence, limited biological plausibility, i.e., characterization of spectrum of
key events informing mode of action.
NC<2 Concentrations
Associated with Effects0
oxidative stress in human
plasma: 500 ppb
Overall study ambient
means:
Annual avg: 12 to 23 ppb
9.5-yr avg: 34 ppb NC>2 and
96 ppb NOX in cases
Dyslipidemia in rats: 160
ppb
Oxidative stress in rats:
5,320 ppb, Inflammation in
rats. Z,DDU ana o,ozu ppo
Inflammation in human cells
exposed to human plasma,
oxidative stress in human
plasma: 500 ppb
Total Mortality
Short-term Exposure (Section 4.4)
2008 ISA - Suggestive but not Sufficient to Infer a Causal Relationship
Current draft ISA - Likely to be a Causal Relationship
Key Evidence:
(Table 4-41)
Recent studies add:
Uncertainty/Limitation:
Consistent evidence for NO2-related increases in total mortality from multicity studies in diverse
locations. Associations found with adjustment for weather, time trends and are robust to adjustment
using various methods and varying degrees of freedom to specify temporal trends. Several studies found
robust NO2 associations in copollutant models with PM-io, SO2, or O^.
Biological processes leading to NO2-related mortality not entirely clear. Large percentage of mortality is
due to cardiovascular causes, but there is limited coherence of evidence across the spectrum of
cardiovascular morbidity outcomes. The strongest evidence for respiratory morbidity is for asthma and is
more limited or inconsistent for COPD and respiratory infection, which are larger causes of mortality in
adults.
Additional evidence from multicity studies and additional evidence for associations of NO2 that are
independent of the few copollutants that are examined.
Biological processes (i.e., effects on morbidity) by which NO2 exposure leads to mortality not well
characterized.
Individual city ambient
24-h avg maximums: 55 to
161 ppb
Individual city ambient
1-h max:
90th: 33 to 133 ppb,
Maximums: 96 to 112 ppb
November 2013
1-31
DRAFT: Do Not Cite or Quote
-------
Table 1-1 (Continued): Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Causal Determination
NC<2 Concentrations
Associated with Effects0
Total Mortality
Long-term Exposure (Section 5.5)
2008 ISA- Inadequate to Infer the Presence or Absence of a Causal Relationship
Current draft ISA - Suggestive of a Causal Relationship
Key Evidence:
(Table 5-19)
NO2-related mortality found in the Harvard Six Cities cohort but not full American Cancer Society cohort,
seminal studies of multiple U.S. cities. Consistent evidence in single-city studies in diverse locations but
inconsistent evidence among other large cohorts of multiple U.S. locations. Associations found with NO2
averaged over 1 to 16 years for periods 0 to 20 yr before death. Similar results found with extended
follow-up of cohorts up to 26 years. Associations found with adjustment for age, sex, smoking,
education, comorbid factors and in some cases, neighborhood-level SES. A few studies adjust for a
copollutant or traffic and generally find positive NO2 associations.
Biological processes by which NO2 exposure leads to mortality not entirely clear because of limited
coherence among findings for respiratory and cardiovascular morbidity outcomes.
Recent studies add: Evidence from additional cohorts and similar results with extended follow-up of previous cohorts.
Uncertainty/Limitation: Lack of association in some high-quality U.S. cohort studies; biological processes (i.e., effects on
morbidity) by which NO2 exposure leads to mortality not well characterized.
Reproductive and Developmental Effects Long-term Exposure01
2008 ISA - Inadequate to Infer the Presence or Absence of a Causal Relationship for broad category
Fertility, Reproduction, and Pregnancy (Section 5.4.2)
Current draft ISA - Suggestive of a Causal Relationship
Overall study ambient
means:
14 to 34 ppb
Key Evidence:
(Table 5-15)
Consistent but limited evidence for increases in pre-eclampsia in association with pregnancy NO2
exposure in a few epidemiologic studies with adjustment for maternal age, smoking, SES, diabetes,
parity. Lack of toxicological studies to inform biological plausibility.
Decreased odds of live birth found in an in vitro fertilization study in association with short-term NO2
exposure in periods leading up to in vitro fertilization and long-term exposure up to birth. Lack of
biological plausibility with no effect on fertility found in a rat study.
Inconsistent epidemiologic evidence for changes in blood pressure in pregnancy, gestational diabetes,
effects on placenta. No effects found on sperm count or quality in epidemiologic or animal toxicological
studies.
Recent Studies add: New epidemiologic investigation of these outcomes with some supporting evidence.
Uncertainty/Limitation:
Inconsistent and limited evidence for several outcomes, weak biological plausibility, i.e., characterization
of spectrum of key events informing mode of action.
Overall study ambient
means:
NO2:
12-to 14-day avg: 19 ppb
Entire pregnancy: 23 ppb
NOX:
Entire pregnancy mean: 7.2
ppb
3rd trimester median: 7.5
ppb
November 2013
1-32
DRAFT: Do Not Cite or Quote
-------
Table 1-1 (Continued): Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Causal Determination
NC<2 Concentrations
Associated with Effects0
Birth outcomes (Section 5.4.3)
Current draft ISA - Suggestive of a Causal Relationship
Key Evidence:
(Table 5-15)
Recent studies add:
Uncertainty/Limitation:
Consistent evidence from high-quality studies for associations of prenatal NO2 exposure with fetal
growth restriction particularly, as assessed with fetal or neonatal physical measurements. Associations
found with adjustment for maternal age, SES, smoking, alcohol use, season of conception. Associations
found with copollutants, with examination in only a few studies. Key events to inform mode of action not
clearly characterized.
Decreased birth weight found in a few high-quality epidemiologic studies but not in other studies. Limited
biological plausibility provided by findings of decreased birth weight in a rat study.
Inconsistent epidemiologic evidence for associations with preterm birth, birth defects. Findings for
decreased litter size in rodents are inconsistent and do not directly inform epidemiologic observations.
Large body of epidemiologic investigation of birth outcomes, with new evidence for fetal growth
restriction.
Inconsistent evidence for some outcomes, lack of clear biological plausibility, i.e., characterization of
spectrum of key events informing mode of action.
Overall study ambient
means:
Entire pregnancy: 16 to 20
ppb
Specific trimesters: 7.8 to
21 ppb
Decreased birth weight in
rats: 1,300 to 5,300 ppb
Postnatal development (Section 5.4.4)
Current draft ISA - Suggestive of a Causal Relationship
Key Evidence:
(Table 5-15)
Evidence for partially irreversible decreases in lung function growth in a few cohorts of children in
association with NC>2 averaged over 6 mo, 1 or 8 yr. Associations found with adjustment for age, body
mass index, smoking exposure. SES not examined. Associations also found with copollutants.
Impairments in lung morphology found in experimental animals not related to changes in children.
Some evidence for decreases in cognitive function in association with concurrent annual avg or prenatal
NO2 exposure. Inconsistencies found across the various neurodevelopmental effects examined.
Associations found with adjustment for SES and in one study, noise, but potential confounding by lead or
other pollutants not examined. Weak biological plausibility due to inconsistent findings in rodents for
effects on emotional responses and motor function. Relationship between outcomes examined in
children and rodents not clear.
Inconsistent epidemiologic and toxicological evidence for postnatal mortality. Limited and inconsistent
evidence for impaired physical development in rats and no analogous epidemiologic investigation.
Across outcomes, effects on key events to inform mode of action not characterized.
Recent studies add: New epidemiologic investigation of neurodevelopmental effects with some supporting evidence.
Overall study ambient
means:
Lung function growth
Annual avg: 14 to 21 ppb
6-mo avg: 34 ppb
Cognitive function
Concurrent annual avg: 17
ppb
Prenatal: 15 ppb
Uncertainty/Limitation:
Inconsistent evidence for some outcomes; weak biological plausibility, i.e., characterization of the
spectrum of key events informing mode of action.
November 2013
1-33
DRAFT: Do Not Cite or Quote
-------
Table 1-1 (Continued): Key evidence contributing to causal determinations for NO2 exposure and health effects evaluated in
the current draft ISA for Oxides of Nitrogen.
Health Effect
Category3
Causal Determination
NC<2 Concentrations
Associated with Effects0
Cancer
Long-term Exposure (Section 5.6)
2008 ISA- Inadequate to Infer the Presence or Absence of a Causal Relationship
Current draft ISA - Suggestive of a Causal Relationship
Key Evidence:
(Table 5-21)
Evidence from some high-quality studies for increases in lung cancer incidence and mortality, but no
association in other studies of comparable quality. Cohorts followed for 7-30 yr, with NC>2 or NOx
exposures assessed for 1- to 30-yr periods. Associations found with adjustment for smoking, diet, SES
and occupational exposures. Associations also found with copollutants. Lack of evidence in experimental
animals for direct effect of NO2 in lung tumor induction, but limited biological plausibility provided by
findings that high NC>2 exposures promote lung tumors with co-exposure to diesel exhaust particles or
known carcinogens. Limited evidence for other key events informing mode of action with findings of
hyperplasia of lung epithelium and formation of secondary oxidation products in the respiratory tract.
Limited epidemiologic evidence for associations with cancers of other sites. Weak evidence to describe
key events informing mode of action with mixed findings for mutagenic and genotoxic effects in
experimental animals and in vitro.
Recent studies add: Evidence in some studies for lung cancer incidence and mortality.
Uncertainty/Limitation: Lack of evidence that NO2 acts as a direct carcinogen, weak evidence for key events informing mode of
action.
aA spectrum of outcomes is evaluated as part of a broad health effect category including physiological measures (e.g., airway responsiveness, lung function), clinical outcomes (e.g.,
respiratory symptoms, hospital admissions), and cause-specific mortality. Total mortality includes all nonaccidental causes of mortality and is informed by the nature of the evidence for
the spectrum of morbidity effects (e.g., respiratory, cardiovascular) that can lead to mortality. The sections and tables referenced include a detailed discussion of the available evidence
that supports the causal determinations.
bSince the completion of the 2008 ISA for Oxides of Nitrogen, the phrasing of causal determinations has changed slightly, and the weight of evidence that describes each level in the
hierarchy of the causal framework has been more explicitly characterized.
The concentrations refer to NO2 unless otherwise specified.
dln the current draft ISA, separate causal determinations are formed for smaller subcategories of reproductive and developmental effects based on varying underlying biological
processes and exposure patterns over different lifestages.
Overall study ambient
means:
Lung cancer
NOX: 11 ug/m3 for 5-yr avg
in all subjects, 32 and 42
ug/m3 averaged over follow-
up of 9.6 or 6.7 yr in cases
NO2: 14 to 23 ppb for 1-yr
or 5-y avg
Lung tumor promotion in
rodents: 4,000 to10,000
ppb
November 2013
1-34
DRAFT: Do Not Cite or Quote
-------
1.5 Evaluation of the Independent Effects of NO2
1 As described in the preceding section, a key consideration in the causal determinations is
2 the extent to which the evidence demonstrates that NO2 exposure has an independent
3 effect on health outcomes. The evaluation of the independent effect of NO2 exposure has
4 two major components. One is the extent to which epidemiologic studies account for
5 potential confounding factors. The other is the extent to which controlled human
6 exposure and animal toxicological studies demonstrate a direct effect of NO2 exposure on
7 the key events that inform the mode of action for a health outcome. In the 2008 and this
8 ISA for Oxides of Nitrogen, potential confounding by other traffic-related pollutants is
9 identified as an uncertainty in characterizing relationships between NO2 exposure and
10 many health effects. Other factors can potentially confound associations between ambient
11 NO2 concentrations and health effects. Across the health effects examined, epidemiologic
12 studies are noted for consistently adjusting for potential confounding by factors such as
13 meteorology and time trends in analyses of short-term NO2 exposure, and for factors
14 such as SES and smoking exposure in analyses of long-term NO2 exposure. This section
15 assesses the epidemiologic evidence for the independent effect of NO2 exposure on
16 health outcomes by integrating information across various lines of investigation.
17 In epidemiologic studies evaluated in this ISA, confounding was assessed primarily using
18 multivariable models that include NO2 concentrations and potential confounders in the
19 same model. The NO2 effect estimate represents the effect of NO2 keeping the level of
20 the covariate constant. In the ISA, confounding is assessed by examining the change in
21 the magnitude and precision of the effect estimate for NO2 in multivariable models, not a
22 change in statistical significance. There are limitations to multivariable models. If NO2
23 and the potential confounder are highly correlated, the collinearity (i.e., covariates predict
24 each other) introduced by including them in the same model can misleadingly decrease or
25 increase the magnitude or precision of the effect estimates for NO2 or the potential
26 confounder. Collinearity can occur, for example, if pollutants are from the same sources
27 or are derived from NO2 (e.g., O3), or if meteorology affects formation of both
28 pollutants. Adding correlated but noncausal variables can produce models that fit the data
29 poorly, and residual confounding is possible if confounders are excluded or poorly
30 measured.
November 2013 1-35 DRAFT: Do Not Cite or Quote
-------
1.5.1 Potential confounding by time-varying factors and individual- or
population-level characteristics
1 Most epidemiologic studies of short-term NO2 exposure evaluated in the ISA are
2 repeated measures population-level time series or panel studies that examine correlation
3 between time-varying patterns in NO2 and outcome. Using linear terms, interaction
4 terms, and splines, these studies found associations with health effects with adjustment
5 for factors such as temperature, humidity, day of the week, season, and long-term time
6 trends, which can have similar temporal patterns as both ambient NO2 concentrations and
7 health outcomes. Time-series and panel studies accounted for serially correlated errors
8 that can arise with repeated measures of ambient NO2 and outcomes. Studies reported
9 adjusting for temperature and humidity using splines with 3 or 4 degrees of freedom, and
10 associations between NO2 and respiratory hospital admissions in 8 Korean cities were
11 robust to using 3 to 6 degrees of freedom for adjustment (Son et al., 2013). Studies
12 reported adjusting for time trends using natural splines with 4 to 12 degrees of freedom
13 per year. Effect estimates were robust to using 6 to 10 degrees of freedom to control for
14 time trends for NO2-associated respiratory hospital admissions (Son et al.. 2013) and 4 to
15 14 degrees of freedom for NO2-associated total mortality (Wong etal.. 2010; Stieb et al..
16 2008). Evidence indicates adequate control for confounding by weather or time trends.
17 Most epidemiologic studies of long-term exposure compare individuals living in varying
18 locations. These studies found associations with health effects with adjustment for factors
19 that vary among individuals including SES, smoking, and other health conditions.
20 Confounding also can occur by factors that vary among locations and affect the spatial
21 pattern of health effects. Many studies of asthma and lung function in children adjusted
22 for community of residence or accounted for spatial variation in outcome. A few of
23 studies of respiratory effects and mortality found associations with NO2 exposure with
24 adjustment for community-level SES factors such as the percentage of individuals with
25 low or high education. Neither noise nor stress, which like NO2 concentrations are found
26 to be higher near traffic, was widely examined as a potential confounder. A recent study
27 found that the association between residential ambient NO2 and asthma in children was
28 limited to those with higher exposure to violence (Clougherty et al.. 2007). If exposure to
29 violence is only a confounder, the association for NO2 would tend to be similar by level
30 of exposure to violence. Individual-level stress was not examined as a potential
31 confounder, but results from the analysis of SES and violence provide some support for
32 respiratory effects of NO2 exposure being independent of psychosocial stress. Although
33 associations for NO2 were inconsistent overall, one study each found associations of
34 ambient NO2 with poorer memory or decreased fetal growth with adjustment for aircraft
35 or road traffic noise (van den Hooven et al., 2012b; van Kempen et al., 2012).
November 2013 1-36 DRAFT: Do Not Cite or Quote
-------
1.5.2 Potential confounding by copollutant exposures
1 The near-road environment is characterized as having higher concentrations of NO2 and
2 other traffic-related pollutants such as EC, CO, and UFP (Section 2.6.4.1). Studies
3 examining the near-road environment found spatial gradients in NO2 concentrations that
4 are correlated with gradients in pollutants such as UFP, VOCs such as BTEX (sum of
5 benzene, toluene, ethylbenzene, xylene), and polycyclic aromatic hydrocarbons (Section
6 2.6.4.1). These observations indicate the potential for NO2-related health effects to be
7 confounded by other traffic-related pollutants or NO2 to serve as a surrogate for a
8 mixture of traffic-related pollution. NO2 concentrations decrease less sharply with
9 increasing distance from the roadway than concentrations of UFP and CO, suggesting
10 that pollutants other than NO2 may serve as better indicators of near-road pollution
11 gradients. However, NO2 may capture spatial and temporal trends in traffic pollution
12 better than PM2 5 concentrations (Section 2.6.4.3). Gradients of traffic-related pollutants
13 are influenced by factors other than distance to roadway including traffic volumes, local
14 topography, meteorology, and conditions affecting chemical transformations.
15 These differences in spatial patterns among pollutants may explain the wide range of
16 correlations found between ambient concentrations of NO2 and copollutants (Table 2-4).
17 In studies of health effects (Chapter 4 and Chapter 5). moderate correlations often were
18 observed for SO2 (Spearman or Pearson r = 0.12 to 0.74 for 25th-75th percentiles), PM2 5
19 (r = 0.38 to 0.65), and PMi0 (r = 0.26 to 0.62), whereas higher correlations were found
20 for CO (r = 0.66 to 0.82) and EC (r = 0.25 to 0.92) (Figure 2-19). Ozone generally is
21 poorly or inversely correlated with NO2 (r = -0.16 to 0.41 for 25th-75th percentiles) even
22 during the summer, when O3 concentrations are higher (Section 2.6.4.2) (Table 2-4). The
23 data described above also indicate that the magnitude of correlation varies by location.
24 The few available studies show weaker correlations between personal exposures of NO2
25 and PM2 5 or EC (Pearson r = 0.06 to 0.49) (Table 2-7). Thus, the potential for copollutant
26 confounding may vary by pollutant, location, and exposure assessment method.
27 Epidemiologic studies evaluated in this ISA examined copollutant confounding primarily
28 with copollutant (i.e., two pollutants) models. The ISA does not consider multipollutant
29 (i.e., three or more pollutants) models because multicollinearity among three or more
30 pollutants can produce less reliable and precise effect estimates. Respiratory effects were
31 consistently associated with short-term increases in ambient NO2 concentration across
32 locations with adjustment for copollutants such as PMi0, PM2 5, SO2, or O3, or as
33 examined in fewer studies, CO, EC, BC, UFP, or VOCs [Figure 4-10. Figure 4-11. and
34 S4-1 (U.S. EPA. 2013d)1. An independent effect of NO2 exposure also is supported by
35 findings from some studies that NO2 but not copollutants such as PM2 5, PM10, EC, CO,
36 or SO2 were associated with decreases in lung function or increases in pulmonary
November 2013 1-37 DRAFT: Do Not Cite or Quote
-------
1 inflammation in single-pollutant models (Sarnat et al.. 2012; Oftedal et al.. 2008; Holguin
2 et al.. 2007; Lagorio et al.. 2006). In these studies, NO2 concentrations were weakly to
3 moderately correlated with copollutants (Spearman r = 0.05 to 0.51).
4 As described above, NO2 is more highly correlated with CO and EC than other
5 pollutants. Studies of respiratory effects provide limited evidence that associations
6 observed with NO2 are independent of CO, EC, or BC. In studies conducted in Atlanta,
7 GA, Canada, and Australia, associations between NO2 and increases in respiratory ED
8 visits remained positive with adjustment for CO (Jalaludin et al., 2008; Tolbert et al..
9 2007; Villeneuve et al.. 2007). In the Australian study, NO2 showed a stronger
10 association in the warm season, when no association was found for CO (Jalaludin et al..
11 2008). Associations between NO2 and cardiovascular hospital admissions were robust to
12 adjustment for CO in studies conducted in Taiwan but not other locations (Figure 4-16).
13 In some cases, effect estimates for CO were attenuated with adjustment for NO2,
14 indicating confounding by NO2. NO2 and CO were moderately to highly correlated
15 (Spearman or Pearson r = 0.55 to 0.74), which may limit the implications of some of the
16 copollutant model results.
17 Several panel studies characterized as having strong exposure assessment with personal
18 exposure monitoring or measurement of ambient NO2 and copollutants at the location
19 and time of subjects' exposures in outdoor locations found associations of NO2 with
20 respiratory effects with adjustment for EC or BC (Strak et al.. 2012; McCreanor et al..
21 2007; Delfino et al.. 2006). Among children with asthma in Southern California, personal
22 NO2 but not EC was associated with decreases in lung function (Delfino et al.. 2008a).
23 The correlations of NO2 with EC or BC varied widely among these studies from the
24 weak correlations typically found with personal exposures (r = 0.21, 0.38) to the higher
25 correlations typically found with near-road exposure (r = 0.58 to 0.67). In Delfino et al.
26 (2006). the effect estimate for central site NO2 was reduced with adjustment for central
27 site EC but remained positive. In the Children's Health Study, long-term NO2
28 concentrations were associated with bronchitic symptoms in children with asthma with
29 adjustment for EC (McConnell et al.. 2003). In most studies, effect estimates for EC or
30 BC were robust to adjustment for NO2. However, in a few cases, effect estimates for EC
31 were reduced with adjustment for NO2 (Strak et al.. 2012; McConnell et al.. 2003).
32 Among children in Beijing, China, the association between NO2 and pulmonary
33 inflammation was reduced with adjustment for BC but remained positive (Lin et al..
34 2011). indicating that BC may explain some but not all of the effects of NO2.
35 Similar to NO2, UFP and VOCs show gradients with distance from roads. Limited
36 available evidence indicates that health effects are associated with NO2 independently of
37 UFP or VOCs. Also, a recent review indicated a lack of consistent association between
November 2013 1-38 DRAFT: Do Not Cite or Quote
-------
1 UFP and health effects, although deficiencies in data and heterogeneity in study designs
2 were noted as limitations to drawing conclusions (HEI Review Panel on Ultrafine
3 Particles, 2013). Panel studies found NO2-associated decreases in lung function and
4 increases in asthma medication use with adjustment for UFP (Figure 4-11). This evidence
5 is substantiated by the strong exposure assessment of studies with measurement of NO2
6 and UFP (Spearman r = 0.56, 0.58) at the time and location of subjects' outdoor
7 exposures in traffic and other locations (Strak et al.. 2012; McCreanor et al.. 2007). Some
8 studies did indicate that UFP confounded associations of NO2 with particular outcomes,
9 for example, lung function but not exhaled nitric oxide (eNO), the pro-inflammatory
10 cytokine IL-6, or protein in nasal lavage fluid (Steenhof et al.. 2013; Strak et al.. 2012)
11 and wheeze but not asthma medication use (von Klot et al.. 2002). Some effect estimates
12 for UFP were robust to adjustment for NO2; others were attenuated. In Copenhagen,
13 Denmark, the association between NOX and cardiovascular hospital admissions largely
14 decreased and became imprecise with adjustment for UFP (Figure 4-15). Potential
15 confounding of NO2-associated health effects by VOCs has been little examined. NO2,
16 BTEX, and individual VOCs were associated with pulmonary inflammation and lung
17 function in children with asthma or wheeze (Greenwald et al.. 2013; Martins et al.. 2012).
18 NO2 concentrations showed a wide range of correlations with VOCs (r = -0.43 to 0.77).
19 Martins et al. (2012) examined copollutant models and found that associations of NO2
20 with pulmonary inflammation were robust to VOC adjustment. The association between
21 NO2 and FEVi was attenuated with adjustment for benzene but not ethylbenzene.
22 Correlations for NO2 with PMi0, PM2 5, and SO2 are variable across locations and are
23 low or inverse for O3. A few results indicate that these copollutants confound
24 associations with NO2. For example, NO2 was not associated with asthma hospital
25 admissions in Greece (Samoli et al.. 2011) or with lung function in children with asthma
26 in Canada (Liu et al.. 2009b) with adjustment for PMi0, PM25, or SO2. Most evidence
27 shows that associations of short-term NO2 exposure with respiratory effects,
28 cardiovascular hospital admissions, and total mortality remain positive with adjustment
29 for PM10, SO2, or O3 (Figure 4-10 and Figure 4-15. and Section 4.4.4). NO2 associations
30 with respiratory effects, and to a more limited extent, cardiovascular effects remain
31 positive with adjustment for PM2 5. The few studies of long-term exposure that adjusted
32 for these copollutants found robust associations between NO2 and bronchitic symptoms
33 and lung function growth (Hwang and Lee. 2010; Rojas-Martinez et al.. 2007a;
34 McConnell et al.. 2003). Limited available results do not indicate that associations of
35 short-term NO2 exposure with wheeze or pulmonary inflammation (Strak et al.. 2012;
36 Mann et al.. 2010) or long-term NO2 exposure with bronchitic symptoms are confounded
37 by PM10_2.5 (McConnell et al.. 2003). Adjustment for NO2 had varying effects on the
38 associations for PMi0, PM25, SO2, and O3. Across the aforementioned studies,
39 copollutant associations did not change with adjustment for NO2 in some cases but were
November 2013 1-39 DRAFT: Do Not Cite or Quote
-------
1 attenuated in other cases. For respiratory and cardiovascular hospital admissions and
2 mortality, several multicity studies found robust associations for NO2 with adjustment for
3 PM10, SO2, or O3. These results pooled across cities add to the evidence for the
4 independent effects of NO2 exposure by indicating an effect of NO2 across locations that
5 vary from each other in the correlations between NO2 and copollutants (Faustini et al..
6 2013: Chenetal.. 2012b: Wong etal. 2010).
7 In some locations, ambient concentrations of gases are associated more strongly with
8 personal PM than personal exposures to gases, suggesting that ambient gases may serve
9 as a surrogate for personal PM exposure (Sarnat et al.. 2001). However, limited recent
10 data show weak to moderate correlations between personal NO2 and ambient copollutants
11 and between ambient NO2 and personal copollutant exposures (r = -0.30 to 0.44) (Table
12 2-5 and Table 2-6). This is true of Suh and Zanobetti (201 Ob) in Atlanta, GA where
13 several studies link ambient NO2 with asthma and respiratory ED visits (Section 4.2.7.4).
14 A recent meta-analysis found that in some cases, ambient NO2 concentrations were more
15 strongly related with personal PM2 5 than personal NO2 concentrations. In other cases,
16 ambient and personal NO2 were well correlated. Among children with asthma in
17 Southern California, the association of personal NO2 with decreases in lung function was
18 reduced with adjustment for central site PM25; however, there still was evidence for
19 association with ambient NO2 (Delfino et al., 2008a). The collective data do not indicate
20 that ambient NO2 concentrations serve only as surrogates for personal exposures to PM.
21 Analyses other than copollutant modeling provide support for the independent effects of
22 NO2 from PM. Studies of respiratory-related ED visits and mortality conducted in the
23 U.S., Canada, Australia, Europe, and Asia showed stronger associations with NO2 in the
24 warm season than cold season (Figure 4-9, Section 4.4.6). Although these results may
25 point to lower exposure measurement error because of more time outdoors, they also
26 could support the independent effects of NO2 from PM2 5 or PM10 since lower
27 correlations are reported for the warm season (Section 2.6.4.1). In the multicontinent
28 APHENA study, PM10-total mortality risk estimates were higher with higher (75th versus
29 25th percentile) mean ambient NO2 concentration (Katsouvanni et al.. 2009). If PMi0 and
30 NO2 were only confounders of each other, risk estimates would tend to be similar by
31 NO2 concentration.
Indoor NO2
32 Indoor NO2 exposures (averaged over 3 to 7 days or 4 weeks) are consistently associated
33 with respiratory symptoms in children (Section 4.2.6.1). In the 2008 ISA for Oxides of
34 Nitrogen, results indicating reductions in respiratory symptoms after an intervention in
35 classrooms to reduce NO2 concentrations with use of flued gas heaters (Pilotto et al..
November 2013 1 -40 DRAFT: Do Not Cite or Quote
-------
1 2004) were used as support for the independent effects of NO2 exposure. Few studies
2 examined other indoor pollutants. In children with asthma, pulmonary inflammation was
3 associated with both indoor school NO2 and various PM metrics (Sarnat et al., 2012). In
4 another study, indoor home NO2 was associated with respiratory symptoms with
5 adjustment for indoor PM25 (Hansel et al., 2008). Sarnat etal. (2012) found correlations
6 between NO2 and copollutants such as BC and PM to differ in magnitude or direction
7 between the indoor and outdoor school environments, suggesting that NO2 may exist as
8 part of a different pollutant mixture in indoor and outdoor environments.
NO2 as an Indicator for Traffic-related Pollution or Proximity to Traffic
9 Vehicles make up the largest single emission source of NOX, and vehicle NOX emissions
10 or factors such as distance to major roadway or roadway density are important inputs into
11 LUR models that predict ambient NO2 or NOX concentrations (Sections 2.6.2.2 and
12 2.6.2.3). A recent review of near-road studies concluded that residence near busy roads is
13 consistently associated with risk of development of asthma in children and risk of asthma
14 exacerbations (HEI. 2010). Some evidence also was reported for lung function, mortality,
15 and cardiovascular effects. However, several lines of evidence indicate that NO2 may not
16 serve only as an indicator of traffic pollution. NO2 concentrations display different
17 gradients with distance from the roadway from other pollutants, so using NO2
18 concentrations as indicator for traffic may misrepresent concentration gradients for UFP,
19 CO, and PM2 5 (Sections 2.5.3 and 2.6.4.3). Further, since other sources contribute to
20 ambient NO2 concentrations, including electric utilities, airports, wildfires, and shipping
21 ports (Section 2.3), NO2 is not unique to vehicle emissions. The relative contributions of
22 various sources to ambient concentrations of NO2 are difficult to distinguish.
23 Several studies evaluated in this ISA found associations of effects such as lung function,
24 pulmonary inflammation, asthma, and mortality with both NO2 and traffic indicators
25 such as roadway proximity or density and traffic density. However, other studies reported
26 associations with outcomes such as decreases in lung function in children with asthma or
27 hospital admissions in adults with COPD with NO2 but not roadway proximity or density
28 (Andersen et al.. 2011; Holguin et al., 2007). Andersen et al. (2011) reported low to
29 moderate correlations between traffic variables and 1-year or 25-year averages of NO2
30 (Spearman r = 0.30 to 0.49). Studies also found associations of ambient (Gauderman et
31 al.. 2007) or indoor (Hansel et al.. 2008) NO2 concentrations with respiratory effects to
32 persist with adjustment for traffic variables such as distance to freeway, distance to curb,
33 or type of street in front of the home. McConnell et al. (2010) found that the association
34 between ambient NO2 and asthma in children was attenuated with adjustment for
35 modeled NOX, which was highly correlated with NO2 and other pollutants. The
36 association between modeled NOX and asthma was robust to adjustment for NO2
November 2013 1-41 DRAFT: Do Not Cite or Quote
-------
1 concentrations. These results suggest that associations for modeled NOX may reflect
2 effects of NO2 and other traffic-related pollutants. Although long-term NO2 exposure
3 was not consistently associated with total mortality, a few studies indicated associations
4 with NO2 with adjustment for traffic proximity (Jerrett et al.. 2009) or modification of the
5 NO2 association by traffic density (Lipfert et al., 2009). Source apportionment models
6 often combine NO2 with traffic-related PM species in a single factor (Section 2.6.4.3):
7 however, there is support for an effect of NO2 distinct from that of a mixture of traffic-
8 related pollution. Among children in southern New England, NO2 and a source
9 apportionment factor of EC, zinc, lead, copper, and selenium (labeled a motor vehicle
10 source) were moderately correlated (Pearson r = 0.49), and each was associated with
11 asthma symptoms in a copollutant model (Gent et al.. 2003). The observations from
12 several studies indicating that NO2 associations are independent of those for measures of
13 roadway or traffic proximity and density or a mixture of traffic-related pollutants provide
14 evidence that NO2 does not serve only as a surrogate for traffic-related pollution.
1.5.3 Summary of Evaluation of the Independent Effects of NO2 Exposure
15 Several lines of epidemiologic evidence indicate that associations of NO2 exposure with
16 health outcomes are independent of other correlated factors such as meteorology, season,
17 and factors associated with traffic and other emission sources including SES and
18 exposure to other pollutants. Multivariable models indicate health effects in association
19 with short-term NO2 exposure that are independent of temperature, humidity, day of the
20 week, season, and long-term time trends. Associations between long-term NO2 exposure
21 and health effects are found to be independent of individual- and community-level SES
22 measures, smoking exposure, and other health conditions. Potential confounding by stress
23 or noise was not widely examined, particularly for health effects consistently associated
24 with NO2. The array of potential confounders were measured with methods widely used
25 in the literature, but residual confounding is possible if factors are measured with error.
26 NO2 and copollutants such as PM10, PM25, SO2, O3, EC or BC, UFP, CO, and VOCs do
27 not always show similar trends in ambient concentrations or associations with health
28 effects. Examination of potential confounding by EC, BC, UFP, CO, and VOCs is limited
29 overall, particularly for cardiovascular effects, and is absent for mortality. Epidemiologic
30 studies consistently report associations between short-term increases in ambient NO2
31 concentration and an array of respiratory effects, cardiovascular hospital admissions, and
32 total mortality with copollutant adjustment across locations with varying correlations
33 between NO2 and copollutants. Although examined in fewer studies, associations
34 between long-term NO2 exposure and respiratory effects were found with copollutant
November 2013 1 -42 DRAFT: Do Not Cite or Quote
-------
1 adjustment. There also is evidence for health effects associated with long-term NO2
2 exposure but not measures of traffic or with adjustment for traffic proximity.
3 Epidemiology is limited in its ability to demonstrate the independent effects of NO2
4 exposure because not all potential confounding factors are examined, including the full
5 array of traffic-related pollutants potentially correlated with NO2. Further, multivariable
6 models can produce biased or unreliable effect estimates, including those used to adjust
7 for multiple copollutants together. Thus, the evidence from experimental studies is key
8 for informing the independent effect of a pollutant. Differences in determinations of a
9 causal versus likely to be a causal relationship relate to the extent to which there is
10 coherence among various lines of evidence to provide biological plausibility for the
11 effects of NO 2 exposure. For cardiovascular effects and total mortality related to short-
12 term NO2 exposure and respiratory effects related to long-term NO2 exposure, there
13 remains some uncertainty regarding an independent effect of NO2 exposure because there
14 is limited or inconsistent evidence from experimental studies or across a spectrum of
15 related outcomes. For respiratory effects of short-term NO2 exposure, epidemiologic
16 evidence for copollutant-adjusted results and indoor NO2 together with experimental
17 evidence provide sufficient evidence for the independent effects of NO2 exposure.
1.6 Policy-Relevant Considerations
18 A key policy-relevant issue that frames the review of the NAAQS as described in detail
19 in the Integrated Review Plan is how the available scientific evidence informs decisions
20 on the basic elements of the NAAQS: indicator, averaging time, level, and form (Preface
21 to the ISA, Section 1.1). The NAAQS are required to provide an adequate margin of
22 safety, and thus understanding of the adverse nature of health effects and of at-risk
23 lifestages and populations also is a key policy-relevant consideration. This ISA addresses
24 the key policy-relevant considerations with the health effects for which the evidence
25 indicates there is a causal or likely to be a causal relationship with NO2 exposure. The
26 discussion focuses on respiratory effects, cardiovascular effects, and total mortality of
27 short-term NO2 exposure and respiratory effects of long-term NO2 exposure.
1.6.1 NO2 Exposure Metrics
28 The primary short-term and long-term NO2 NAAQS are based on 1-h daily max
29 concentrations and annual average concentrations, respectively (Preface to the ISA).
30 These standards were set to protect against a broad range of respiratory effects associated
31 with short-term NO2 exposures and health effects potentially associated with long-term
November 2013 1 -43 DRAFT: Do Not Cite or Quote
-------
1 exposure. Thus, an important consideration in the review of the NAAQS is evaluation of
2 the health effects evidence for various averaging times of NO2 exposure.
3 For short-term exposure, the majority of previous and recent evidence is for health effects
4 associated with 24-h avg ambient NO2, but the smaller body of evidence is equally
5 consistent for subdaily averages such as 1-h or 8-h max NO2 and NO2 averaged over
6 periods of 2 to 10 hours. These subdaily averages of NO2 concentrations were associated
7 with a spectrum of effects related to asthma exacerbations, measures of cardiovascular
8 physiology and hospital admissions, as well as total mortality. NO2 exposures occurring
9 over 2 to 5 hours in outdoor traffic and nontraffic locations were associated with
10 decreases in lung function and pulmonary inflammation in adults (Strak et al.. 2012;
11 McCreanor et al., 2007). This evidence is substantiated by strong exposure assessment
12 with measurement of NO2 at the locations of adults' outdoor exposures. Biological
13 plausibility is provided by demonstrations of airway responsiveness (Section 4.2.2.2) and
14 allergic inflammation (Section 4.2.4.3) in adults with asthma or animal models of allergic
15 disease induced by NO2 exposures in the range of 30 minutes to 6 hours.
16 Across epidemiologic studies, the robustness of evidence for associations with respiratory
17 effects, cardiovascular effects, and total mortality is similar for 24-h avg and 1-h max
18 NO2 concentrations. Based on the few within-study comparisons, the magnitude of
19 association with health effects did not clearly differ between 24-h avg and 1-h max NO2.
20 In some cases, associations with respiratory or cardiovascular effects were larger for
21 1-h max NO2 than 24-h avg NO2 (Carlsen et al.. 2012; Ballester et al.. 2006). In other
22 cases, associations were stronger for 24-h avg NO2 than 1-h max NO2 (Rodriguez et al..
23 2007; Morgan et al.. 1998). For asthma-related ED visits in Atlanta, GA, associations
24 were similar for 1-h max and 24-h avg NO2 with a 1-day lag, and slightly larger for
25 6-h nighttime avg NO2 (12 a.m.-6 a.m.) (Darrow et al.. 201 la). The NO2 averaging times
26 varied in the distribution of concentrations and spatial heterogeneity, which may account
27 for differences in associations with asthma ED visits. Nighttime avg NO2 had a wider
28 distribution of concentrations than 24-h avg NO2. Nighttime avg NO2 was similarly
29 spatially heterogeneous as 1-h max NO2 but was lower in concentrations. The spatial
30 heterogeneity in ambient NO2 concentrations within urban areas (Section 2.5.2) and with
31 distance to roadways (Section 2.5.3) and diurnal trends with higher concentrations
32 measured during morning rush hours (Section 2.5.4) are not unique to Atlanta, GA. This
33 heterogeneity in ambient NO2 concentrations along with diurnal variation in time-activity
34 patterns suggest that exposure measurement error can vary among different NO2
35 averaging times, which could obscure true differences in association with health effects.
36 Various long-term NO2 exposure metrics were associated with respiratory effects in
37 children, including 6-month, 1-year, 3-year, 4-year, and 10-year lifetime average ambient
November 2013 1 -44 DRAFT: Do Not Cite or Quote
-------
1 NO2 concentrations (Section 5.2). Associations did not consistently differ among long-
2 term exposure metrics. However, stronger associations were found with 1-year NO2
3 averages that corresponded with exposure in the first year of life (Gruzieva et al., 2013;
4 Gruzieva etal.. 2012; Schultz et al.. 2012).
1.6.2 NO2 Lag Structure in Epidemiologic Studies
5 Characterization of the NO2 exposure durations and lags associated with health effects
6 can increase the understanding of the nature of relationships between NO2 exposure and
7 health effects. The lag structure for NO2 exposure may vary among health effects
8 depending on differences in the time course by which various biological processes occur.
9 Identifying important lag structures can depend on whether the lag structure varies within
10 the population according to differences in time activity patterns, pre-existing disease, or
11 other factors that influence exposure and responses to exposure. Associations among
12 exposure lags, particularly single-day and multiday averages of NO2, may vary because
13 the spatial and temporal heterogeneity in ambient NO2 concentrations can result in
14 differences in exposure measurement error. The lag structure was examined in studies of
15 short-term NO2 exposure for an array of respiratory effects, cardiovascular effects, and
16 total mortality. While no particular lag of NO2 exposure was more strongly associated
17 with cardiovascular effects, evidence indicates that NO2-associated respiratory effects
18 and total mortality generally are larger for multiday exposures than single-day exposures.
19 Epidemiologic panel studies of children with asthma found increases in pulmonary
20 inflammation and respiratory symptoms and decreases in lung function in association
21 with increases in NO2 lagged 0 or 1 day and multiday averages of 2 to 7 days. Increases
22 in respiratory symptoms also were associated with NO2 lagged 2 to 7 days. Consistent
23 with these findings, increases in respiratory hospital admissions and ED visits were found
24 in association with NO2 lagged 0 or 1 day or averaged over 2 to 7 days. Whereas no
25 particular lag of exposure was more strongly associated with decreases in lung function,
26 several studies indicated larger increases in pulmonary inflammation, respiratory
27 symptoms, and respiratory hospital admissions and ED visits for increases in multiday
28 averages of NO2 than single-day lags. Multiple studies indicate the largest increases in
29 total mortality occurring with a lag of 1 day; however, several studies found associations
30 with NO2 averaged over 2 to 7 days.
31 Studies in which adults with asthma and healthy adults were exposed for 2 to 5 hours in
32 outdoor traffic and nontraffic locations indicated decreases in lung function and increases
33 in pulmonary inflammation immediately or 2 hours after exposures (Strak et al., 2012;
34 McCreanor et al.. 2007). In both populations, decreases in lung function also were found
November 2013 1 -45 DRAFT: Do Not Cite or Quote
-------
1 the day after exposures. In healthy adults, increases in pulmonary inflammation did not
2 persist the day after outdoor exposure (Strak et al.. 2012). These data based on strong
3 exposure assessment support other epidemiologic findings showing increases in
4 respiratory effects at lag 0 or 1 day of NO2 exposure and also indicate a similar lag
5 structure for people with and without asthma. NO2 exposure appears to affect the
6 biological processes underlying the effects observed in epidemiologic studies on a similar
7 time frame. Controlled human exposure studies found airway responsiveness in adults
8 with asthma to increase immediately after or 20 minutes to 4 hours after a single NO2
9 exposure and over 4 days of repeated exposure (Sections 4.2.2.2 and 4.2.2.3). In
10 experimental studies, NO2 induced allergic inflammation 30 minutes up to 19 hours after
11 a single or 2-day exposure in humans and 7 days after exposure in rats.
1.6.3 Concentration-Response Relationships and Thresholds
12 Characterizing the shape of the concentration-response relationship aids in understanding
13 the public health impacts of NO2 exposure. Of particular interest for the review of the
14 NO2 NAAQS is whether the relationship is linear across the full range of ambient
15 concentrations or whether there are deviations from linearity at and below the levels of
16 the current 1-hour standard of 100 ppb and annual average standard of 53 ppb. The true
17 concentration-response relationship may be obscured by fewer observations and greater
18 exposure measurement error in the lower than upper range of the ambient concentration
19 distribution, the influence of other determinants or risk factors for the health effect, and
20 heterogeneity among individuals in the population in their response to air pollution
21 exposures.
22 The shape of the concentration-response relationship for NO 2-associated health effects
23 was examined in a limited number of epidemiologic studies and is better characterized
24 for respiratory hospital admissions and ED visits and total mortality than other outcomes.
25 Consistent with evidence reported in the 2008 ISA for Oxides of Nitrogen, results from
26 recent studies indicate a linear concentration-response relationship for respiratory
27 hospital admissions and ED visits and total mortality using various methods, including
28 analysis of splines, higher order terms for NO2 (e.g., quadratic, cubic), and quantiles of
29 NO2. The collective evidence for short-term exposure does not identify a threshold for
30 the effects of NO2 exposure.
31 Recent studies continue to support a linear relationship between short-term NO2 exposure
32 and asthma ED visits in U.S. cities. For 1-h max NO2 (lag 0-2 day avg) combined across
33 urban monitors by population-weighting, a linear association was indicated with asthma
34 ED visits in Atlanta, GA during 1993-2004 (Strickland etal. 2010). Relative risks
November 2013 1 -46 DRAFT: Do Not Cite or Quote
-------
1 increased across quintiles of NO2 between 28 and 181 ppb (with concentrations less than
2 28 ppb as the reference), and models with nonparametric smoothing showed increasing
3 risk with increasing 1-h max NO2 between the 5th and 95th percentile of concentrations
4 (11 to 37 ppb) (Strickland. 2013). The concentration-response relationship was not
5 examined for 1-h max NO2 concentrations less than 11 ppb. For the relationship between
6 24-h avg NO2 and pediatric asthma ED visits in Detroit, MI, there was not evidence for
7 the association differing below and above a threshold set by the investigators at 23 ppb
8 NO2 (between the 82nd and 85th percentiles) in conditional logistic regression models.
9 The risk was not assumed to be zero below 23 ppb, and the threshold model did not fit
10 the data better than the linear model did (Li et al., 20 lib).
11 Linear concentration-response relationships also are indicated for mortality associated
12 with short-term NO2 exposure (lag 1 day or 0-1 day avg) in the U.S., Canada, and Asia
13 based on comparisons of linear and various nonlinear models with natural (Moolgavkar et
14 al.. 2013; Wong et al.. 2008b) and cubic (Chen et al.. 2012b) splines or quadratic and
15 cubic terms for NO2 (Stieb et al.. 2008). Most results are for 24-h avg NO2, with limited
16 evidence for 3-h max NO2 (Stieb et al.. 2008). The results do not identify a threshold for
17 NO2-related mortality. The analysis of 85 U.S. cities indicated less certainty in the shape
18 of the concentration-response at 24-h avg NO2 concentrations less than 20 ppb, where the
19 density of data was low and 95% CIs were wide (Moolgavkar et al., 2013).
20 A few previous results point to nonlinear concentration-response relationships but for
21 outcomes for which the concentration-response relationship has not been widely
22 examined, including cough in children or cardiovascular hospital admissions in adults.
23 The studies tended to find NO2-related increases in effects that were larger in magnitude
24 per increment in NO2 in the lower range of NO2 concentrations than in the upper range
25 of concentrations.
26 The shape of the concentration-response relationship was not formally evaluated in
27 previous or recent studies of long-term NO2 exposure and is not well characterized.
28 Limited available evidence from Europe indicates increasing respiratory effects with
29 increasing ambient NO2 concentration in analyses of tertiles (concentrations not reported)
30 of modeled ambient residential NO2 for asthma incidence in children (Modig et al.. 2009)
31 and a cubic spline of ambient NO2 concentrations for asthma hospital admissions in
32 adults (Andersen et al.. 2012) (Section 5.2.12). These studies reported annual average
33 NO2 concentrations with ranges 1.8 to 24 ppb and 5.3 to 21 ppb.
34 In summary, the shape of the concentration-response relationship is better characterized
35 for short-term NO2 exposure than for long-term exposure. Previous and recent evidence
36 indicates a linear relationship between short-term NO2 exposure and respiratory hospital
37 admissions or ED visits and mortality. Evidence is available primarily for 24-h avg NO2
November 2013 1 -47 DRAFT: Do Not Cite or Quote
-------
1 but also 1-h and 3-h max NO2 and for NO2 averaged over 2 to 5 days or lagged 1 day.
2 There is uncertainty in the shape of the concentration-response in the low range of the
3 distribution of NO2 concentrations where data density is low. However, results do not
4 identify a threshold for respiratory hospital admissions or ED visits or mortality. Results
5 from U.S. studies indicate uncertainty in the shape of the concentration-response
6 relationship for asthma ED visits at 1-h max NO2 concentrations less than 11 ppb and for
7 total mortality at 24-h avg ambient NO2 concentrations less than 20 ppb.
1.6.4 Regional Heterogeneity in Effect Estimates
8 In addition to examining the shape of the concentration-response relationship for
9 NO 2 -related health effects across the distribution of concentrations, studies have
10 examined whether the concentration-response varies across geographical regions. Such
11 information is limited largely to European and Asian cities. In the only U.S. study, a test
12 for heterogeneity was not statistically significant for the association between ambient
13 NO2 concentrations (for the first year or first 3 years of life) and asthma among Latino
14 and African American individuals ages 8-21 years in Chicago, Houston, San Francisco,
15 New York, and Puerto Rico (Nishimura et al., 2013a). Comparisons of odds ratios do
16 indicate differences among cities, namely a larger association in the San Francisco cohort
17 comprising only African American children, an imprecise association in New York for
18 NO2 in the first 3 years of life and no association in Houston, which had a much smaller
19 sample size. San Francisco had lower ambient NO2 and SO2 concentrations than New
20 York. PM2 5 and SO2 were associated with asthma in Houston but not New York or San
21 Francisco. There was not strong indication of regional heterogeneity in associations of
22 short-term or long-term NO2 exposure with respiratory effects among European or
23 Korean cities (Jacquemin et al., 2009b; Moon et al., 2009; Timonen et al., 2004).
24 Regional heterogeneity is indicated among European and Asian cities in the relationship
25 between short-term NO2 exposure and total mortality. Larger risk estimates were found
26 for European cities with lower prevalence of smoking and greater household use of gas
27 (Samoli et al.. 2006). Larger risk estimates were found for Asian cities with a larger
28 population of older adults and higher concentrations of PMi0 and cities with higher
29 temperature, populations with more time outdoors, less air conditioning use, higher
30 mortality from infection, more deaths among younger people, and lower mean ambient
31 concentrations of NO2 and other pollutants (Wong et al.. 2008b).
32 Studies also did not clearly show heterogeneity in NO2-related respiratory effects
33 between neighboring communities. Higher NO2 concentrations are found in urban areas
34 than nonurban areas; however, NO2-related respiratory effects do not consistently differ
November 2013 1 -48 DRAFT: Do Not Cite or Quote
-------
1 between urban and suburban communities in Europe (Section 6.5.4). A recent study
2 found larger NO2-related increases in pulmonary inflammation among children with
3 asthma in Ciudad Juarez, Mexico schools than nearby El Paso, TX schools (Sarnat et al.,
4 2012). The reasons for the heterogeneity were not explicitly analyzed.
5 In summary, there is not clear evidence for regional heterogeneity in the relationship
6 between short-term or long-term NO2 exposure and respiratory effects, including the only
7 U.S. study, which examined asthma. Studies of short-term NO2 exposure and mortality
8 found heterogeneity across cities in Europe and cities in Asia and indicated that
9 heterogeneity may be due to differences among cities in factors that influence exposure to
10 air pollution such as time outdoors or air conditioning use, differences in exposure to
11 other pollutants, or differences in the distribution of other indicators of health.
1.6.5 Public Health Significance
12 The public health significance of air pollution-related health effects is informed by the
13 adverse nature of the health effects that are observed, the size of the population exposed
14 to air pollution or affected by the health outcome, and the presence of populations or
15 lifestages with higher exposure or greater risk of air pollution-related health effects.
Evaluating Adversity of Health Effects
16 Both the World Health Organization (WHO) and the American Thoracic Society (ATS)
17 have addressed what health effects may be considered adverse. In defining health as "the
18 state of complete physical, mental, and social well-being and not merely the absence of
19 disease or infirmity" (WHO. 1948). WHO acknowledges that changes in health outcomes
20 that are not severe enough to result in a diagnosis of a clinical outcome can be adverse if
21 they affect the well-being of an individual. ATS also considered a wide range of health
22 outcomes in defining adverse effects. Distinguishing between individual and population
23 risk, ATS indicated that small air pollution-related changes in an outcome observed in
24 individuals can be considered adverse on a population level since a shift in the
25 distribution of population responses due to higher air pollution exposure can increase the
26 proportion of the population with clinically important effects or at increased risk of a
27 clinically important effect that can be caused by another risk factor (ATS, 2000b).
28 Increases in ambient NO2 concentrations are associated with a broad spectrum of health
29 effects, including those characterized as adverse by ATS such as mortality, the
30 development of asthma, and asthma exacerbations (ATS. 2000b). Ambient NO2 exposure
31 also is associated with more subtle changes in function such as increases in airway
November 2013 1 -49 DRAFT: Do Not Cite or Quote
-------
1 responsiveness and pulmonary inflammation and decreases in lung function (Section
2 1.4.2). Increases in airway responsiveness and pulmonary inflammation are key events
3 informing the mode of action for acute asthma exacerbations and asthma development.
4 While evidence is not consistent across cardiovascular endpoints, ambient NO2
5 concentrations are associated with increases in cardiovascular mortality and hospital
6 admissions as well as decreases in HRV, and decreases in HRV have been linked with
7 increased risk of life-threatening cardiovascular events (Section 1.4.4). These
8 physiological measures show a distribution within populations, and NO2-associated
9 changes in airway responsiveness, pulmonary inflammation, or HRV may be considered
10 adverse on a population level because they can increase the proportion of the population
11 with clinically important changes that can lead to exacerbation or development of asthma
12 and cardiovascular events, respectively.
At-risk Lifestages or Populations for Exposure of Oxides of Nitrogen or
Related Health Effects
13 The NAAQS are intended to protect public health with an adequate margin of safety, and
14 protection is provided for the population as a whole and groups at increased risk for
15 health effects from exposure to the pollutant for which each NAAQS is set (Preface to the
16 ISA). Hence, the public health significance of health effects related to NO2 exposure also
17 is informed by whether specific lifestages or groups in the population are identified as
18 having higher NO2 exposure or NO2-related health effects. The 2009 American Housing
19 Survey reports that 17.5% of occupied housing units in the U.S. are within 90 meters of
20 NOX emissions sources such as a four-lane highway, railroad, or airport (U.S. Census
21 Bureau. 2009). In Los Angeles, CA, 44% of the population was found to live within
22 100 meters of a major road (HEI. 2010). Such proximity to roadways can be
23 characterized by higher concentrations of NO2 than background (Section 2.5.3). Thus, a
24 large proportion of the U.S. population has the potential for elevated ambient NO2
25 exposures and for increased risk of health effects that are related to higher NO2 exposure.
26 At-risk lifestages or populations also can be characterized by specific biological,
27 sociodemographic, or behavioral factors among others. Since the 2008 ISA for Oxides of
28 Nitrogen and as used in the recent IS As for O3 (U.S. EPA. 2013b) and Lead (U.S. EPA.
29 2013a). EPA has developed a framework for drawing conclusions about the role of such
30 factors in modifying risk of air pollution-related health effects or the magnitude of
31 physiological responses to air pollution exposure (Table III of the Preamble). Similar to
32 the causal framework, conclusions about at-risk factors are based on judgments of the
33 consistency and coherence of evidence within and across disciplines (Chapter 6).
34 Conclusions on at-risk factors are based on NO2 exposure since other oxides of nitrogen
35 were examined in few studies. Briefly, the evaluation includes analysis of studies that
November 2013 1 -50 DRAFT: Do Not Cite or Quote
-------
1 compared exposure or health effect relationships among different groups (e.g., people in
2 different age categories, people with and without asthma) and studies conducted in a
3 population or animal model with a particular factor or pathophysiological condition.
4 Where available, information on exposure, dosimetry, and modes of action is included to
5 assess coherence with evidence for health effects and inform how a particular factor may
6 modify NO2-related risk of health effects or physiological responses (e.g., exposure
7 differences, differences in biological effect for a given dose). Because the framework for
8 at-risk factors was not available for the 2008 ISA for Oxides of Nitrogen, previous and
9 recent studies are considered in the current conclusions.
10 There is adequate evidence to indicate that children (ages 0-14 years) and older adults
11 (ages > 65 years) are at increased risk for NO2-related health effects. There is suggestive
12 evidence that genetic variants (primarily in antioxidant genes), asthma, COPD, SES, sex,
13 and diet modify health effects associated with NO2 exposure based on a limited evidence
14 base or inconsistent results within a discipline. In most cases, there is coherence with
15 evidence from another discipline. Because of insufficient consistency and quantity of
16 evidence within a discipline and lack of information from another discipline, there is
17 inadequate evidence to determine whether cardiovascular disease, diabetes, obesity,
18 race/ethnicity, smoking, or urban/nonurban residence modifies NO2-related health
19 effects.
20 Children are identified as an at-risk lifestage based on consistent evidence for larger risks
21 of asthma hospital admissions and ED visits associated with short-term NO2 exposure in
22 children (Section 6.4.1.1). Most studies compared children ages 0-14 years with people of
23 all ages or adults. Compared with people ages 15-64 years, children ages 0-14 years had
24 2- to 3-fold higher risk of asthma hospital admissions or ED visits for the same increase
25 in 24-h avg NO2 concentrations. Substantiating the evidence for NO2 specifically,
26 evidence indicates that NO2 exposure has independent effects on asthma exacerbations
27 (Section 1.4.2). The reasons for the increased risk for children are not clear. Compared
28 with adults, children have developing respiratory systems, oronasal breathing, higher
29 ventilation rates (Section 3.2.2.3). and different time activity patterns characterized by
30 more time outdoors and more vigorous activity (Section 6.4.1). However, it is not clear
31 whether these physiological and behavioral characteristics result in differences in NO2
32 exposure, uptake in the respiratory tract, or exposure measurement error for children
33 (Sections 2.6.5.2 and 3.2.5). Limited data do not clearly indicate higher personal NO2
34 exposures in children (Table 2-9). Many studies reported a higher number of asthma ED
35 visits or hospital admissions among children than other age groups. Thus, higher
36 incidence of asthma exacerbations in children may be a reason for their increased risk.
November 2013 1-51 DRAFT: Do Not Cite or Quote
-------
1 Among children, several studies found that associations with asthma, allergic
2 sensitization, and decreases in lung function in individuals ages 4-21 years (mostly
3 children) were larger for NO2 exposure around birth or infancy compared with NO2
4 exposure in the first four years of life, year before diagnosis, or lifetime average NO2
5 exposure (Sections 5.2 and 6.4.1) These results suggest that the prenatal period or infancy
6 may represent a critical time window of exposure for NO2-related respiratory effects in
7 children. Lung development begins during the prenatal period and continues throughout
8 childhood; however, it is not clear whether differences in lung development contribute to
9 higher risk associated with NO2 exposure during the prenatal period or infancy.
10 Children ages 18 years and younger not only comprise a large proportion of the U.S.
11 population (24% in the 2010 U.S. census), but also have a higher rate of asthma health
12 care encounters than adults (e.g., 10.7 versus 7.0 per 100 persons with asthma)1. Further,
13 asthma is the leading chronic illness (9.5% prevalence) and reason for school
14 absenteeism in children in the U.S. Although there is only suggestive evidence for people
15 with asthma having increased risk for NO 2-related health effects because of inconsistent
16 epidemiologic evidence, there are some studies showing larger NO 2-related increases in
17 respiratory symptoms or decreases in lung function in children with asthma than without
18 asthma (Section 6.3.1). Based on the large number of children in the U.S. population and
19 the high prevalence of asthma morbidity among children, even slightly higher risks of
20 asthma exacerbations for children compared with adults can translate into large numbers
21 affected, magnifying the potential public health impact of NO2 exposure.
22 There is adequate evidence that older adults, i.e., those ages 65 years and older, have
23 increased risk for NO2-related health effects compared with younger adults (Section
24 6.4.1.2). This conclusion is based mainly on evidence for hospital admissions for asthma
25 or COPD and for mortality. Substantiating this conclusion, evidence indicates that NO2
26 exposure has independent effects on asthma exacerbations. Studies showed a wide range
27 in difference in magnitude of risk for older adults, from <1- to 3-fold higher risk of
28 NO2-related respiratory hospital admissions to 2- to 6-fold higher for NO2-related
29 mortality. As with children, the reasons for the increased risk for older adults are not well
30 understood. Older adults did not consistently have higher absolute numbers of respiratory
31 hospital admissions compared with younger adults, so higher incidence of the health
32 effect does not seem to explain their higher NO2-related risk estimates. Studies show
33 different time activity patterns (Section 6.4.1) and rates of ventilation in older than
34 younger adults; however, it is not known whether these factors contribute to differential
35 exposure and uptake of NO2 in older adults. Regardless of the reasons for increased risk
36 of NO2-related mortality, the higher incidence of mortality in older adults than other age
National Center for Health Care Statistics Data Brief. Available: http://www.cdc.gov/nchs/data/databriefs/db94.htm
November 2013 1 -52 DRAFT: Do Not Cite or Quote
-------
1 groups, and the growing proportion of older adults in the U.S. magnify the public health
2 impact of increased risk of mortality from NO2 exposure.
3 At-risk lifestages and populations likely are not characterized by a single factor.
4 Cardiovascular diseases and diabetes are more prevalent in adults ages 65 years and older
5 (Table 6-3). and comorbid factors potentially could contribute to their higher risk of
6 NO2-related health effects. Compared with younger adults, older adults had larger risks
7 of NO2-associated COPD hospital admissions in the studies reviewed in this ISA. A
8 controlled human exposure study of older adults found a larger NO2-induced decrease in
9 lung function among adults with COPD (mean age 60 yr) than among healthy older
10 adults (mean age 61 yr) (Section 6.3.2). While pre-existing cardiovascular disease did not
11 consistently modify all NO2-related health effects, there is evidence for cardiovascular
12 disease increasing the risk of NO 2 -associated mortality. These various lines of evidence
13 suggest that comorbidities in older adults could contribute to their higher risk of
14 NO2-related health effects. In children, there is some evidence for larger NO2-related
15 respiratory effects in lower SES groups than higher SES groups (Section 6.4.2). Lower
16 SES also is associated with higher asthma prevalence and exacerbations as well as higher
17 NO2 or NOX exposure. Co-occurring risk factors in a lifestage or population may
18 influence their risk of NO2-related health effects. Such inter-relationships among
19 potential risk factors have not been well examined for NO2-related health effects.
20 In summary, the public health significance of NO2-related health effects is supported by
21 many lines of evidence. A large proportion of the U.S. population lives near major roads,
22 resulting in a large number of people potentially with elevated ambient NO2 exposure.
23 There is evidence for relationships with effects that are clearly adverse such as premature
24 mortality, hospital admissions, ED visits, and asthma incidence. More subtle NO2-related
25 effects such as increases in airway responsiveness and pulmonary inflammation or
26 decreases in lung function and lung function growth can be considered adverse on a
27 population level because higher NO2 exposure can lead to an increase in the proportion
28 of the population with clinically important effects. The public health significance of
29 NO 2 -related health effects also is supported by the increased risk for children (ages 0-14
30 years) compared with adults and increased risk for older adults (ages 65 years and older)
31 compared with younger adults. Children and older adults differ from other lifestages in
32 behavior, physiology, and comorbidities; however, it is not clear whether these
33 characteristics contribute to their increased risk of NO2-related health effects. The large
34 proportions of children and older adults in the U.S. population and the higher prevalence
35 of co-occurring factors such as COPD and cardiovascular disease in older adults and
36 asthma in children can translate into a large number of people affected by ambient NO2
37 exposure and thus magnify the public health impact of ambient NO2 exposure.
November 2013 1-53 DRAFT: Do Not Cite or Quote
-------
1.7 Conclusions
1 The major emissions sources of NOX in the U.S. based on the 2008 National Emissions
2 Inventory are motor vehicles and electric utilities. The distribution of emissions sources
3 and chemical transformation, transport, and deposition of these emissions contribute to
4 spatial and temporal heterogeneity in ambient concentrations and human exposure to
5 NO2, NO, and NOX. In the U.S., NOX emissions and ambient NO2 concentrations have
6 decreased over the past 20 years. Ambient concentrations have been shown to be 30% to
7 200% higher at locations within 15m of a roadway (averaged over hours to weeks)
8 compared with locations farther away from the road. Relationships between ambient NO2
9 concentrations and personal exposures vary in the population, and exposure measurement
10 error resulting from the use of ambient concentrations can reduce epidemiologic
11 associations observed with health effects. Once inhaled, NO2 is transformed in the
12 respiratory tract to secondary oxidation products, which can initiate a cascade of events,
13 including inflammation, bronchial smooth muscle reactivity, and modification of immune
14 responses. The processes by which ambient-relevant NO2 exposures lead to effects
15 outside of the respiratory system are not well characterized.
16 Recent studies, most of which are epidemiologic, expand on findings reported in the 2008
17 ISA for Oxides of Nitrogen and earlier assessments. The consistency, coherence, and
18 biological plausibility of evidence integrated across disciplines and outcomes related to
19 asthma exacerbations indicate that there is a causal relationship between short-term NO2
20 exposure and respiratory effects. Evidence indicates there is likely to be a causal
21 relationship between short-term NO2 exposure and cardiovascular effects as well as total
22 mortality. There is likely to be a causal relationship between long-term NO2 exposure
23 and respiratory effects based strongly on findings in children for asthma incidence and
24 decreases in lung function. Epidemiologic studies provide compelling evidence for the
25 independent effects of NO2 exposure with associations of health effects with NO2 in
26 copollutant models across locations that vary in copollutant relationships and across
27 exposure assessment methods, associations with indoor NO2, or differences in effects
28 with traffic proximity or intensity. However, the extent of examination of potential
29 copollutant confounding varies across health effects, particularly confounding of the
30 relationships of short-term NO2 exposure with cardiovascular effects and total mortality
31 by CO, UFP, EC, and BC. For these relationships, the limited biological plausibility does
32 not conclusively demonstrate that NO2 exposure has an effect independent of the effects
33 of another traffic-related pollutant or mixture. Evidence is suggestive of a causal
34 relationship between long-term NO2 exposure and cardiovascular effects, reproductive
35 and developmental effects, total mortality, and cancer.
November 2013 1 -54 DRAFT: Do Not Cite or Quote
-------
1 There is adequate evidence that children (ages 0-14 years) and older adults (ages 65 years
2 and older) have increased risk for NO 2 -related health effects. Large numbers of people in
3 the U.S. live near major roads and potentially have elevated exposures to ambient NO2
4 compared with people living 500 meters or more from roads. There is suggestive
5 evidence that risk of NO2-related health effects differs by pre-existing asthma,
6 pre-existing COPD, genetic variants for oxidative metabolism enzymes, dietary
7 antioxidant intake, SES, and sex. Daily average and 1-hour maximum NO2 as well as
8 concentrations averaged over 30 minutes to a few hours are associated with health
9 effects. For many respiratory outcomes, larger effects are estimated for multiday averages
10 of ambient NO2 concentrations than single-day concentrations. Respiratory effects are
11 associated with long-term NO2 concentrations averaged over 6 months and 1 year to 10
12 (representing lifetime exposure) years. Evidence indicates that the concentration-response
13 relationship for relationships of short-term ambient NO2 exposure with respiratory
14 hospital admissions and ED visits and mortality is linear, and results do not identify a
15 threshold for effects.
November 2013 1-55 DRAFT: Do Not Cite or Quote
-------
References for Chapter 1
Adamkiewicz. G: Ebelt. S: Syring. M; Slater. J; Speizer. FE; Schwartz. J; Suh. H; Gold. DR. (2004).
Association between air pollution exposure and exhaled nitric oxide in an elderly population. Thorax
59: 204-209. http://dx.doi.org/10.1136/thorax.2003.006445
Alver. K; Meyer. HE; Falch. JA; S0gaard. AJ. (2010). Outdoor air pollution, bone density and self-reported
forearm fracture: The Oslo Health Study. Osteoporos Int 21: 1751-1760.
htrp://dx.doi.org/10.1007/s00198-009-1130-8
Andersen. ZJ: B0nnelykke. K; Hvidberg. M; Jensen. SS: Ketzel. M; Loft. S: S0rensen. M; Tj0nneland. A;
Overvad. K; Raaschou-Nielsen. O. (2012). Long-term exposure to air pollution and asthma
hospitalisations in older adults: A cohort study. Thorax 67: 6-11. http://dx.doi.org/10.1136/thoraxjnl-
2011-200711
Andersen. ZJ: Hvidberg. M; Jensen. SS: Ketzel. M; Loft. S: S0rensen. M; Tj0nneland. A; Overvad. K;
Raaschou-Nielsen. O. (2011). Chronic obstructive pulmonary disease and long-term exposure to traffic-
related air pollution: A cohort study. Am J Respir Crit Care Med 183: 455-461.
http://dx.doi.org/10.1164/rccm.201006-0937OC
ATS (American Thoracic Society). (2000b). What constitutes an adverse health effect of air pollution? Am
J Respir Crit Care Med 161: 665-673. http://dx.doi.Org/10.1164/ajrccm.161.2.ats4-00
Bagheri Lankarani. N; Kreis. I; Griffiths. DA. (2010). Air pollution effects on peak expiratory flow rate in
children. Iran J Allergy Asthma Immunol 9: 117-126.
Ballester. F: Rodriguez. P; Iniguez. C: Saez. M; Daponte. A; Galan. I; Taracido. M; Arribas. F: Bellido. J:
Cirarda. FB; Canada. A; Guillen. JJ; Guillen-Grima. F: Lopez. E; Perez-Hoyos. S: Lertxundi. A; Toro.
S.. (2006). Air pollution and cardiovascular admisisons association in Spain: Results within the
EMECAS project. J Epidemiol Community Health 60: 328-336.
htrp://dx.doi.org/10.1136/iech.20Q5.037978
CAA. Clean Air Act, as amended by Pub. L. No. 101-549. section 108: Air quality criteria and control
techniques. 42 USC § 7408 (1990a). http://www.law.cornell.edu/uscode/text/42/7408
Cakmak. S: Dales. RE; Vidal. CB. (2010). Air pollution and hospitalization for epilepsy in Chile. Environ
Int 36: 501-505. http://dx.doi.Org/10.1016/i.envint.2010.03.008
Carlsen. HK; Zoega. H; Valdimarsdottir. U; Gislason. T; Hrafnkelsson. B. (2012). Hydrogen sulfide and
particle matter levels associated with increased dispensing of anti-asthma drugs in Iceland's capital.
Environ Res 113: 33-39. http://dx.doi.Org/10.1016/j.envres.2011.10.010
Carlsten. C: Dybuncio. A; Becker. A; Chan-Yeung. M; Brauer. M. (2011c). Traffic-related air pollution and
incident asthma in a high-risk birth cohort. Occup Environ Med 68: 291-295.
http://dx.doi.org/10.1136/oem.2010.055152
CDC (Centers for Disease Control and Prevention). (2012). Exposure to nitrogen dioxide in an indoor ice
arena - New Hampshire, 2011. MMWR Morb Mortal Wkly Rep 61: 139-142.
Chen. R; Samoli. E; Wong. CM; Huang. W: Wang. Z; Chen. B; Kan. H. (2012b). Associations between
short-term exposure to nitrogen dioxide and mortality in 17 Chinese cities: The China Air Pollution and
Health Effects Study (CAPES). Environ Int 45: 32-38. http://dx.doi.Org/10.1016/i.envint.2012.04.008
Cloughertv. JE; Lew. JI; Kubzansky. LD; Ryan. PB; Suglia. SF; Canner. MJ; Wright. RJ. (2007).
Synergistic effects of traffic-related air pollution and exposure to violence on urban asthma etiology.
Environ Health Perspect 115: 1140-1146. http://dx.doi.org/10.1289/ehp.9863
Dales. RE; Cakmak. S: Vidal. CB. (2009b). Air pollution and hospitalization for headache in Chile. Am J
Epidemiol 170: 1057-1066. http://dx.doi.org/10.1093/aje/kwp217
November 2013 1 -56 DRAFT: Do Not Cite or Quote
-------
Darrow. LA; Klein. M; Sarnat. JA; Mulholland. JA; Strickland. MJ; Sarnat. SE; Russell. AG: Tolbert. PE.
(2011 a). The use of alternative pollutant metrics in time-series studies of ambient air pollution and
respiratory emergency department visits. J Expo Sci Environ Epidemiol 21: 10-19.
http://dx.doi.org/10.1038/ies.2009.49
Delfino. RJ: Staimer. N; Gillen. D; Tjoa. T; Sioutas. C: Fung. K; George. SC: Kleinman. MT. (2006).
Personal and ambient air pollution is associated with increased exhaled nitric oxide in children with
asthma. Environ Health Perspect 114: 1736-1743.
Delfino. RJ: Staimer. N; Tjoa. T; Gillen. D; Kleinman. MT; Sioutas. C: Cooper. D. (2008a). Personal and
ambient air pollution exposures and lung function decrements in children with asthma. Environ Health
Perspect 116: 550-558. http://dx.doi.org/10.1289/ehp.10911
Faustini. A; Stafoggia. M; Colais. P; Berti. G: Bisanti. L; Cadum. E; Cernigliaro. A; Mallone. S: Scarnato.
C: Forastiere. F. (2013). Air pollution and multiple acute respiratory outcomes. Eur Respir J 42: 304-
313. http://dx.doi.org/10.1183/09031936.00128712
Finegold. JA: Asaria. P; Francis. DP. (2013). Mortality from ischaemic heart disease by country, region, and
age: Statistics from World Health Organisation and United Nations. Int J Cardiol 168: 934-945.
http://dx.doi.0rg/10.1016/i.iicard.2012.10.046
Gauderman. WJ: Vora. H; McConnell. R; Berhane. K; Gilliland. F; Thomas. D: Lurmann. F; Avol. E;
Kunzli. N; Jerrett. M; Peters. J. (2007). Effect of exposure to traffic on lung development from 10 to 18
years of age: A cohort study. Lancet 369: 571-577. http://dx.doi.org/10.1016/80140-6736(07)60037-3
Gent. JF; Triche. EW: Holford. TR; Belanger. K; Bracken. MB; Beckett. WS: Leaderer. BP (2003).
Association of low-level ozone and fine particles with respiratory symptoms in children with asthma.
JAMA 290: 1859-1867. http://dx.doi.org/10.1001/iama.290.14.1859
Greenwald. R; Sarnat. SB; Ravsoni. AU; Li. WW: Johnson. BA; Stock. TH; Holguin. F; Sosa. T; Sarnat.
JA. (2013). Associations between source-indicative pollution metrics and increases in pulmonary
inflammation and reduced lung function in a panel of asthmatic children. Air Qua! Atmos Health 6: 487-
499. http://dx.doi.org/10.1007/sll869-012-0186-3
Gruzieva. O: Bellander. T; Eneroth. K; Kull. I; Melen. E; Nordling. E; van Hage. M; Wickman. M;
Moskalenko. V; Hulchiy. O; Pershagen. G. (2012). Traffic-related air pollution and development of
allergic sensitization in children during the first 8 years of life. J Allergy Clin Immunol 129: 240-246.
http://dx.doi.0rg/10.1016/i.jaci.2011.ll.001
Gruzieva. O: Bergstrom. A; Hulchiy. O: Kull. I; Lind. T; Melen. E; Moskalenko. V; Pershagen. G:
Bellander. T. (2013). Exposure to air pollution from traffic and childhood asthma until 12 years of age.
Epidemiology 24: 54-61. http://dx.doi.org/10.1097/EDE.Ob013e318276clea
Hansel NN: Brevsse. PN: McCormack. MC: Matsui EC: Curtin-Brosnan. J: Williams. DAL: Moore. JL:
Cuhran. JL: Diette. GB. (2008). A longitudinal study of indoor nitrogen dioxide levels and respiratory
symptoms in inner-city children with asthma. Environ Health Perspect 116: 1428-1432.
http://dx.doi.org/10.1289/ehp. 11349
HEI (Health Effects Institute). (2010). Traffic-related air pollution: A critical review of the literature on
emissions, exposure, and health effects [HEI]. (Special Report 17). Boston, MA.
http://pubs.healtheffects.org/view.php?id=334
HEI Review Panel on Ultrafine Particles. (2013). Understanding the health effects of ambient ultrafme
particles. (HEI Perspectives 3). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=394
Holguin. F; Flores. S: Ross. Z; Cortez. M; Molina. M; Molina. L; Rincon. C: Jerrett. M; Berhane. K;
Granados. A; Romieu. I. (2007). Traffic-related exposures, airway function, inflammation, and
respiratory symptoms in children. Am J Respir Crit Care Med 176: 1236-1242.
http://dx.doi.org/10.1164/rccm.200611-1616OC
Hovert. PL: Xu. J. (2012). Deaths: Preliminary data for 2011. US Department of Health and Human Services
National Center for Health Statistics National Vital Statistics Reports 61: 1-51.
November 2013 1 -57 DRAFT: Do Not Cite or Quote
-------
Hwang. BF; Lee. YL. (2010). Air pollution and prevalence of bronchitic symptoms among children in
Taiwan. Chest 138: 956-964. http://dx.doi.org/10.1378/chest.09-2600
Jacquemin. B; Sunyer. J; Forsberg. B; Aguilera. I; Briggs. D; Garcia-Esteban. R; Gotschi. T; Heinrich. J;
Jarvholm. B; Jarvis. D; Vienneau. D; Kiinzli. N. (2009b). Home outdoor NO2 and new onset of self-
reported asthma in adults. Epidemiology 20: 119-126.
http://dx.doi.org/10.1097/EDE.Ob013e3181886e76
Jalaludin. B; Khalaj. B; Sheppeard. V: Morgan. G. (2008). Air pollution and ED visits for asthma in
Australian children: A case-crossover analysis. Int Arch Occup Environ Health 81: 967-974.
http://dx.doi.org/10.1007/s00420-007-0290-0
Jerrett. M; Finkelstein. MM; Brook. JR; Arain. MA; Kanaroglou. P; Stieb. DM; Gilbert. NL; Verma. D:
Finkelstein. N; Chapman. KR; Sears. MR. (2009). A cohort study of traffic-related air pollution and
mortality in Toronto, Ontario, Canada. Environ Health Perspect 117: 772-777.
http://dx.doi.org/10.1289/ehp.11533
Kaplan. GG: Dixon. E; Panaccione. R; Fong. A; Chen. L; Szyszkowicz. M; Wheeler. A; Maclean. A; Buie.
WD; Leung. T; Heitman. SJ: Villeneuve. PJ. (2009). Effect of ambient air pollution on the incidence of
appendicitis. CMAJ 181: 591-597. http://dx.doi.org/10.1503/cmai.082068
Kaplan. GG: Hubbard. J: Korzenik. J: Sands. BE; Panaccione. R: Ghosh. S: Wheeler. AJ; Villaneuve. PJ.
(2010). The inflammatory bowel diseases and ambient air pollution: A novel association. Am J
Gastroenterol 105: 2412-2419. http://dx.doi.org/10.1038/ajg.2010.252
Katsouvanni. K; Samet. JM; Anderson. HR; Atkinson. R; Le Tertre. A; Medina. S: Samoli. E; Touloumi. G:
Burnett. RT; Krewski. D; Ramsay. T; Dominici. F; Peng. RD; Schwartz. J: Zanobetti. A. (2009). Air
pollution and health: A European and North American approach (APHENA) (pp. 5-90). (ISSN 1041-
5505 Research Report 142). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=327
Lagorio. S: Forastiere. F; Pistelli. R: lavarone. I; Michelozzi. P; Fano. V; Marconi. A; Ziemacki. G: Ostro.
BD. (2006). Air pollution and lung function among susceptible adult subjects: A panel study. Environ
Health 5: 11. http://dx.doi.org/10.1186/1476-069X-5-ll
Li. S: Batterman. S: Wasilevich. E; Wahl. R; Wirtfa. J: Su. FC: Mukherjee. B. (20lib). Association of daily
asthma emergency department visits and hospital admissions with ambient air pollutants among the
pediatric Medicaid population in Detroit: Time-series and time-stratified case-crossover analyses with
threshold effects. Environ Res 111: 1137-1147. http://dx.doi.Org/10.1016/j.envres.2011.06.002
Lim. YH; Kim. H; Kim. JH; Bae. S: Park. HY; Hong. YC. (2012). Air pollution and symptoms of
depression in elderly adults. Environ Health Perspect 120: 1023-1028.
http://dx.doi.org/10.1289/ehp.1104100
Lin. W: Huang. W: Zhu. T: Hu. M: Brunekreef. B: Zhang. Y: Liu. X: Cheng. H: Gehring. U: Li. C: Tang.
2L (2011). Acute respiratory inflammation in children and black carbon in ambient air before and during
the 2008 Beijing Olympics. Environ Health Perspect 119: 1507-1512.
http://dx.doi.org/10.1289/ehp.1103461
Lipfert F; Wyzga. R: Batv. J: Miller. J. (2009). Air pollution and survival within the Washington
University-EPRI veterans cohort: Risks based on modeled estimates of ambient levels of hazardous and
criteria air pollutants. J Air Waste Manag Assoc 59: 473-489.
Liu. L; Poon. R; Chen. L; Frescura. AM; Montuschi. P; Ciabattoni. G: Wheeler. A; Dales. R. (2009b).
Acute effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in
asthmatic children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehpll813
Malerbi. FK; Martins. LC: Saldiva. PHN; Braga. ALF. (2012). Ambient levels of air pollution induce
clinical worsening of blepharitis. Environ Res 112: 199-203.
http://dx.doi.0rg/10.1016/i.envres.2011.ll.010
November 2013 1-58 DRAFT: Do Not Cite or Quote
-------
Mann. JK; Balmes. JR: Bruckner. TA; Mortimer. KM; Margolis. HG: Pratt. B; Hammond. SK; Lurmarm. F;
Tager. IB. (2010). Short-term effects of air pollution on wheeze in asthmatic children in Fresno,
California. Environ Health Perspect 118: 1497-1502. http://dx.doi.org/10.1289/ehp.0901292
Martins. PC: Valente. J: Papoila. AL; Caires. I; Araujo-Martins. J: Mata. P; Lopes. M; Torres. S: Rosado-
Pinto. J: Borrego. C: Annesi-Maesano. I; Neuparth. N. (2012). Airways changes related to air pollution
exposure in wheezing children. Eur Respir J 39: 246-253. http://dx.doi.org/10.1183/09031936.00025111
McConnell. R; Berhane. K; Gilliland. F; Molitor. J: Thomas. D; Lurmann. F; Avol. E; Gauderman. WJ:
Peters. JM. (2003). Prospective study of air pollution and bronchitic symptoms in children with asthma.
Am J Respir Crit Care Med 168: 790-797. http://dx.doi.org/10.1164/rccm.200304-466OC
McConnell. R: Islam. T: Shankardass. K: Jerrett. M: Lurmann. F: Gilliland. F: Gauderman. J: Avol. E: Kuenzlj
N: Yao. L: Peters. J: Berhane. K. (2010). Childhood incident asthma and traffic-related air pollution at home
and school. Environ Health Perspect 118: 1021-1026. http://dx.doi.org/10.1289/ehp.0901232
McCreanor. J: Cullinan. P; Nieuwenhuijsen. MJ; Stewart-Evans. J: Malliarou. E; Jarup. L; Harrington. R;
Svartengren. M; Han. IK; Ohman-Strickland. P; Chung. KF; Zhang. J. (2007). Respiratory effects of
exposure to diesel traffic in persons with asthma. N Engl J Med 357: 2348-2358.
http://dx.doi.org/10.1056/NEJMoa071535
Modig. L; Toren. K; Janson. C: Jarvholm. B; Forsberg. B. (2009). Vehicle exhaust outside the home and
onset of asthma among adults. Eur Respir J 33: 1261-1267.
http://dx.doi.org/10.1183/09031936.00101108
Moolgavkar. SH; Mcclellan. RO; Dewanji. A; Turim. J: Luebeck. EG: Edwards. M. (2013). Time-series
analyses of air pollution and mortality in the United States: A subsampling approach. Environ Health
Perspect 121: 73-78. http://dx.doi.org/10.1289/ehp.1104507
Moon. JS: Kim. YS: Kim. JH; Son. BS: Kim. PS: Yang. W. (2009). Respiratory health effects among
schoolchildren and their relationship to air pollutants in Korea. Int J Environ Health Res 19: 31-48.
http://dx.doi.org/10.1080/09603120802272201
Morgan. G: Corbett. S: Wlodarczyk. J. (1998). Air pollution and hospital admissions in Sydney, Australia,
1990 to 1994. Am J Public Health 88: 1761-1766. http://dx.doi.org/10.2105/AJPH.88.12.1761
Nishimura. KK; Galanter. JM; Roth. LA; Oh. SS: Thakur. N; Nguyen. EA; Thyne. S: Farber. HJ;
Serebrisky. D; Kumar. R; Brigino-Buenaventura. E; Davis. A; LeNoir. MA; Meade. K; Rodriguez-
Cintron. W: Avila. PC: Borrell. LN; Bibbins-Domingo. K; Rodriguez-Santana. JR: Sen. S: Lurmann. F;
Balmes. JR: Burchard. EG. (2013a). Early-life air pollution and asthma risk in minority children: The
GALA II and SAGE II studies. Am J Respir Crit Care Med 188: 309-318.
http://dx.doi.org/10.1164/rccm.201302-0264OC
Novaes. P: Do Nascimento Saldiva. PH; Karajose. N; Macchione. M; Matsuda. M; Racca. L; Berra. A.
(2007). Ambient levels of air pollution induce goblet-cell hyperplasia in human conjunctival epithelium.
Environ Health Perspect 115: 1753-1756. http://dx.doi.org/10.1289/ehp.10363
Novaes. P: Saldiva. PHN; Matsuda. M; Macchione. M; Rangel. MP; Kara-Jose. N; Berra. A. (2010). The
effects of chronic exposure to traffic derived air pollution on the ocular surface. Environ Res 110: 372-
374. http://dx.doi.0rg/10.1016/j.envres.2010.03.003
Oftedal. B; Brunekreef. B; Nystad. W: Madsen. C: Walker. SE; Nafstad. P. (2008). Residential outdoor air
pollution and lung function in schoolchildren. Epidemiology 19: 129-137.
http://dx.doi.org/10.1097/EDE.Ob013e31815c0827
Pilotto. LS: Nitschke. M; Smith. BJ; Pisaniello. D: Ruffin. RE; McElroy. HJ: Martin. J: Hiller. JE. (2004).
Randomized controlled trial of unflued gas heater replacement on respiratory health of asthmatic
schoolchildren. Int J Epidemiol 33: 208-214.
Rodriguez. C: Tonkin. R; Hevwortfa. J: Kusel. M; De Klerk. N; Sly. PD; Franklin. P; Runnion. T; Blocklev.
A; Landau. L; Hinwood. AL. (2007). The relationship between outdoor air quality and respiratory
symptoms in young children. Int J Environ Health Res 17: 351-360.
http://dx.doi.org/10.1080/09603120701628669
November 2013 1-59 DRAFT: Do Not Cite or Quote
-------
Roias-Martinez. R; Perez-Padilla. R; Olaiz-Fernandez. G: Mendoza-Alvarado. L; Moreno-Macias. H;
Fortoul. T; McDonnell. W: Loomis. D; Romieu. I. (2007a). Lung function growth in children with long-
term exposure to air pollutants in Mexico City. Am J Respir Crit Care Med 176: 377-384.
http://dx.doi.org/10.1164/rccm.200510-1678OC
Samoli. E; Aga. E; Touloumi. G: Nisiotis. K; Forsberg. B; Lefranc. A; Pekkanen. J; Wojtyniak. B;
Schindler. C: Niciu. E; Brunstein. R; Dodic Fikfak. M; Schwartz. J; Katsouvanni. K. (2006). Short-term
effects of nitrogen dioxide on mortality: An analysis within the APHEA project. Eur Respir J 27: 1129-
1137. http://dx.doi.org/10.1183/09031936.06.00143905
Samoli. E; Nastos. PT; Paliatsos. AG: Katsouvanni. K; Priftis. KN. (2011). Acute effects of air pollution on
pediatric asthma exacerbation: Evidence of association and effect modification. Environ Res 111: 418-
424. http://dx.doi.0rg/10.1016/j.envres.2011.01.014
Sarnat. JA; Schwartz. J: Catalano. PJ; Suh. HH. (2001). Gaseous pollutants in paniculate matter
epidemiology: Confounders or surrogates? Environ Health Perspect 109: 1053-1061.
Sarnat. SE; Ravsoni. AU; Li. WW: Holguin. F; Johnson. BA; Flores Luevano. S: Garcia. JH; Sarnat. JA.
(2012). Air pollution and acute respiratory response in a panel of asthmatic children along the U.S.-
Mexico border. Environ Health Perspect 120: 437444. http://dx.doi.org/10.1289/ehp.1003169
Schultz. ES: Gruzieva. O: Bellander. T; Bottai. M; Hallberg. J: Kull. I; Svartengren. M; Melen. E;
Pershagen. G. (2012). Traffic-related air pollution and lung function in children at 8 years of age: A
birth cohort study. Am J Respir Crit Care Med 186: 1286-1291. http://dx.doi.org/10.1164/rccm.201206-
1045OC
Schwartz. J: Dockerv. DW: Neas. LM; Wvpij. D; Ware. JH; Spengler. JD; Koutrakis. P; Speizer. FE; Ferris.
BG. Jr. (1994). Acute effects of summer air pollution on respiratory symptom reporting in children. Am
J Respir Crit Care Med 150: 1234-1242. http://dx.doi.Org/10.1164/airccm.150.5.7952546
Son. JY; Lee. JT; Park. YH; Bell. ML. (2013). Short-term effects of air pollution on hospital admissions in
Korea. Epidemiology 24: 545-554. http://dx.doi.org/10.1097/EDE.Ob013e3182953244
Steenhof M; Mudwav. IS: Gosens. I; Hoek. G: Godri. KJ; Kelly. FJ; Harrison. RM; Pieters. RH; Cassee.
FR; Lebret. E; Brunekreef. BA; Strak. M; Janssen. NA. (2013). Acute nasal pro-inflammatory response
to air pollution depends on characteristics other than particle mass concentration or oxidative potential:
the RAPTES project. Occup Environ Med 70: 341-348. http://dx.doi.org/10.1136/oemed-2012-100993
Steerenberg. PA; Nierkens. S: Fischer. PH; Van Loveren. H; Opperhuizen. A; Vos. JG: van Amsterdam. JG.
(2001). Traffic-related air pollution affects peak expiratory flow, exhaled nitric oxide, and inflammatory
nasal markers. Arch Environ Health 56: 167-174. http://dx.doi.org/10.1080/00039890109604069
Stieb. DM; Burnett. RT; Smith-Doiron. M; Brion. O; Shin. HH; Economou. V. (2008). A new
multipollutant, no-threshold air quality health index based on short-term associations observed in daily
time-series analyses. J Air Waste Manag Assoc 58: 435-450.
Strak. M; Janssen. NA; Godri. KJ: Gosens. I; Mudwav. IS: Cassee. FR; Lebret. E; Kelly. FJ: Harrison. RM;
Brunekreef. B; Steenhof. M; Hoek. G. (2012). Respiratory health effects of airborne particulate matter:
The role of particle size, composition and oxidative potential - The RAPTES project. Environ Health
Perspect 120: 1183-1189. http://dx.doi.org/10.1289/ehp.1104389
Strickland. MJ; Darrow. LA; Klein. M; Flanders. WD; Sarnat. JA; Waller. LA; Sarnat. SE; Mulholland. JA;
Tolbert. PE. (2010). Short-term associations between ambient air pollutants and pediatric asthma
emergency department visits. Am J Respir Crit Care Med 182: 307-316.
http://dx.doi.org/10.1164/rccm.200908-1201OC
Strickland. MJ. (2013). Email from Dr Strickland to Dr Patel; Response to question on 2010 AJRCCM paper.
Available online
Suh. HH; Zanobetti. A. (201 Ob). Exposure error masks the relationship between traffic-related air pollution
and heart rate variability. J Occup Environ Med 52: 685-692.
http://dx.doi.org/10.1097/JOM.Ob013e3181e8071f
November 2013 1 -60 DRAFT: Do Not Cite or Quote
-------
Szvszkowicz. M. (2007a). Air pollution and emergency department visits for depression in Edmonton,
Canada. Int J Occup Med Environ Health 20: 241-245. http://dx.doi.org/10.2478/vl0001-007-0024-2
Szyszkowicz. M. (2008a). Ambient air pollution and daily emergency department visits for headache in
Ottawa, Canada. Headache 48: 1076-1082. http://dx.doi.Org/10.llll/i.1526-4610.2007.01039.x
Timonen. KL; Hoek. G: Heinrich. J; Bernard. A; Brunekreef. B; De Hartog. J: Hameri. K; Ibald-Mulli. A;
Mirme. A; Peters. A; Tiittanen. P; Kreyling. WG: Pekkanen. J. (2004). Daily variation in fine and
ultrafine paniculate air pollution and urinary concentrations of lung Clara cell protein CC16. Occup
Environ Med 61: 908-914. http://dx.doi.org/10.1136/oem.2004.012849
Tolbert. PE; Klein. M; Peel. JL; Sarnat SE; Sarnat. JA. (2007). Multipollutant modeling issues in a study of
ambient air quality and emergency department visits in Atlanta. J Expo Sci Environ Epidemiol 17: S29-
S35. http://dx.doi.org/10.1038/si.ies.750Q625
U.S. Census Bureau. (2009). American housing survey for the United States: 2009. Washington, DC: U.S.
Government Printing Office, http://www.census.gov/prod/2011pubs/hl50-09.pdf
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park,NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2008b). Integrated science assessment for oxides of
nitrogen and sulfur: Ecological criteria [EPA Report]. (EPA/600/R-08/082F). Research Triangle Park,
NC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=201485
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2010c). Primary National Ambient Air Quality
Standards for nitrogen dioxide. Fed Reg 75: 6474-6537.
U.S. EPA (U.S. Environmental Protection Agency). (2012). Notice of workshop and call for information on
integrated science assessment for oxides of nitrogen. Fed Reg 77: 7149-7151.
U.S. EPA (U.S. Environmental Protection Agency). (2013a). Integrated science assessment for lead [EPA
Report]. (EPA/600/R-10/075F). Research Triangle Park, NC.
http://ofmpub.epa.gov/eims/eimscomm.getfile7p download id=514513
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for ozone and
related photochemical oxidants (final report) [EPA Report]. (EPA/600/R-10/076F). Washington, DC.
http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=247492
U.S. EPA (U.S. Environmental Protection Agency). (2013d). Supplemental figure S4-1. Associations of
ambient NO2 with respiratory outcomes adjusted for various gaseous copollutants [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013e). Supplemental table Sl-1. Epidemiologic
studies of health effects not evaluated in the ISA for oxides for nitrogen [EPA Report].
van den Hooven. EH; Pierik. FH; de Kluizenaa. Y; Willemsen. SP; Hofman. A; van Ratingen. SW:
Zandvel. PYJ; Mackenbach. JP; Steegers. EAP; Miedema. HME; Jaddoe. VWV. (2012b). Air pollution
exposure during pregnancy, ultrasound measures of fetal growth, and adverse birth outcomes: A
prospective cohort study. Environ Health Perspect 120: 150-156. http://dx.doi.org/10.1289/ehp.1003316
van Kempen. E; Fischer. P; Janssen. N; Houthuijs. D; van Kamp. I; Stansfeld. S: Cassee. F. (2012).
Neurobehavioral effects of exposure to traffic-related air pollution and transportation noise in primary
schoolchildren. Environ Res 115: 18-25. http://dx.doi.Org/10.1016/i.envres.2012.03.002
November 2013 1-61 DRAFT: Do Not Cite or Quote
-------
Villeneuve. PJ: Chen. L; Rowe. BH; Coates. F. (2007). Outdoor air pollution and emergency department
visits for asthma among children and adults: A case-crossover study in northern Alberta, Canada.
Environ Health Global Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40
von Klot. S: Wolke. G: Tuch. T; Heinrich. J; Dockery. DW: Schwartz. J; Kreyling. WG: Wichmann. HE;
Peters. A. (2002). Increased asthma medication use in association with ambient fine and ultrafine
particles. Eur Respir J 20: 691-702. http://dx.doi.org/10.1183/09031936.02.01402001
WHO (World Health Organization). (1948). Preamble to the Constitution of the World Health Organization
as adopted by the International Health Conference, New York, 19-22 June, 1946. In Constitution of the
World Health Organization (pp. 2). Geneva, Switzerland.
http://whqlibdoc.who.int/hist/official records/constitution.pdf
Wong. CM; Vichit-Vadakan. N; Kan. H; Qian. Z. (2008b). Public Health and Air Pollution in Asia (PAPA):
Amulticity study of short-term effects of air pollution on mortality. Environ Health Perspect 116: 1195-
1202. http://dx.doi.org/10.1289/ehp. 11257
Wong. CM; Vichit-Vadakan. N; Vajanapoom. N; Ostro. B; Thach. TO; Chau. PY; Chan. EK; Chung. RY;
Ou. CO: Yang. L: Peiris. JS: Thomas. GN: Lam. TH: Wong. TW: Hedlev. AJ: Kan. H: Chen. B: Zhao.
N; London. SJ: Song. G: Chen. G: Zhang. Y; Jiang. L; Qian. Z; He. 0: Lin. HM; Kong. L; Zhou. D;
Liang. S: Zhu. Z: Liao. D: Liu. W: Bentlev. CM: Dan. J: Wang. B: Yang. N: Xu. S: Gong. J: Wei. H:
Sun. H; Qin. Z. (2010). Part 5. Public health and air pollution in Asia (PAPA): A combined analysis of
four studies of air pollution and mortality. In Public Health and Air Pollution in Asia (PAPA):
Coordinated Studies of Short-Term Exposure to Air Pollution and Daily Mortality in Four Cities (pp.
377-418). Boston, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=348
November 2013 1 -62 DRAFT: Do Not Cite or Quote
-------
CHAPTER 2 ATMOSPHERIC CHEMISTRY AND
EXPOSURE TO OXIDES OF NITROGEN
2.1 Introduction
1 This chapter presents concepts and findings relating to emissions sources, atmospheric
2 science, and human exposure assessment. It is intended as a prologue for detailed
3 discussions on the evidence for health effects that follow in the subsequent chapters, and
4 as a source of information to help interpret those effects in the context of data about
5 atmospheric concentrations and exposures.
6 In the ISA, "oxides of nitrogen" (NOY) refer to all forms of oxidized nitrogen (N)
7 compounds, including NO, NO2, and all other oxidized N-containing compounds formed
8 from NO and NO2. NO and NO2, along with volatile organic compounds (VOCs), are
9 precursors in the formation of ozone (O3) and photochemical smog. NO2 is an oxidant
10 and can react to form other photochemical oxidants, including organic nitrates (RONO2)
11 such as the peroxyacyl nitrates (PANs). NO2 can also react with a variety of atmospheric
12 species, resulting in organic and inorganic nitrates, and making substantial contributions
13 to the mass of atmospheric particulate matter (PM) and the acidity of cloud, fog, and
14 rainwater, as well as PM. The abbreviation NOX refers specifically to the sum of NO and
15 NO2. This chapter describes origins, distribution, fate, and exposure of gaseous oxides of
16 nitrogen, while aspects of particulate nitrogen species (such as pNO3) were addressed in
17 the 2009 PM ISA (U.S. EPA. 2009a).
2.2 Atmospheric Chemistry and Fate
18 The chemistry of oxidized nitrogen compounds in the atmosphere was reviewed in the
19 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). The role of NOX in O3 formation
20 was reviewed in Chapter 3 of the 2013 ISA for Ozone and other Photochemical Oxidants
21 (U.S. EPA. 2013b) and has been presented in numerous texts (e.g., Jacobson. 2002;
22 Jacob. 1999; Seinfeld and Pandis. 1998). The main points from the 2008 ISA for Oxides
23 of Nitrogen will be presented here along with updates based on recent material.
November 2013 2-1 DRAFT: Do Not Cite or Quote
-------
— -IMOY
Long range transport to remote
regions at low temperatures
~HN
-------
1 Figure 2-1). All the other species mentioned above in the definition of NOY are products
2 of reactions of NO or NO2 and collectively are referred to as NOZ (shown in the outer
3 box in Figure 2-1). The totality of all the species shown in Figure 2-1 is referred to as
4 NOY, such that NOY = NOX + NOZ. Inorganic NOZ species are shown on the left side of
5 the outer box and organic species are shown on the right side of the outer box in Figure
6 2-1. Ammonium nitrate and other inorganic particulate species (e.g., Na+, Ca2+ nitrates)
7 are formed from species shown on the left side of the figure; organic nitrates are formed
8 from species shown on the right side of Figure 2-1.
9 Inorganic NOZ species include HONO, HNO3, HNO4, and pNO3. Mollner et al. (2010)
10 identified pernitrous acid (HOONO), an unstable isomer of nitric acid, as a product of the
11 major gas phase reaction forming HNO3. However, since pernitrous acid is unstable, it is
12 not a substantial reservoir for NOX. While a broad range of organic nitrogen compounds
13 are emitted by combustion sources (e.g., nitrosamines and nitro-PAHs), they are also
14 formed in the atmosphere from reactions of NO, NO2, and NO3. These include
15 peroxyacyl and alkyl nitrates, other nitro-PAHs, and the more recently identified nitrated
16 organic compounds in the quinone family. Most of the mass of products shown in the
17 outer box of Figure 2-1 is in the form of PAN and HNO3, although other organic nitrates
18 (e-g-, isoprene nitrates) can be important at locations closer to biogenic sources (Horowitz
19 et al.. 2007; Singh et al., 2007). The concentrations and atmospheric lifetimes of
20 inorganic and organic products from reactions of NO and NO2 vary widely in space and
21 time.
22 Sources of NOX include naturally occurring processes associated with fires, lightning and
23 microbial processes occurring in soils. Anthropogenic sources are dominated by
24 combustion processes from electricity generating units and wide spread transportation
25 sources. Sources are distributed with height with some occurring at or near ground level
26 and others aloft as indicated in Figure 2-1. Because the prevailing winds aloft are
27 generally stronger than those at the surface, emissions from elevated sources (e.g., the
28 stacks of electrical utilities) can be distributed over a wider area than those emitted at the
29 surface (e.g., motor vehicles).
30 Oxidized nitrogen compounds are ultimately lost from the atmosphere by wet and dry
31 deposition to the Earth's surface. Soluble species are taken up by aqueous aerosols and
32 cloud droplets and are removed by wet deposition by rainout (i.e., incorporation into
33 cloud droplets that eventually coagulate into falling rain drops). Both soluble and
34 insoluble species are removed by washout ([i.e., impaction with falling rain drops],
35 another form of wet deposition), and by dry deposition (i.e., impaction with the surface
36 and gas exchange with plants). NO and NO2 are not very soluble and therefore wet
November 2013 2-3 DRAFT: Do Not Cite or Quote
-------
1 deposition is not a major removal process for them. However, a major NOX reservoir
2 species, HNO3, is extremely soluble and its deposition represents a major sink for NOY.
3 Many species, including particulate nitrate and gas phase HONO, are formed by
4 multiphase processes. Data collected in Houston as part of TexAQS-II summarized by
5 Olaguer et al. (2009) indicate that concentrations of HONO are much higher than can be
6 explained by gas phase chemistry and by tailpipe emissions. The uptake of N2O5 by
7 atmospheric aerosols or cloud droplets leads to the loss of O3 and NOX and the
8 production of aqueous phase nitric acid, aerosol nitrate, and gaseous halogen nitrites.
9 N2O5 is the acid anhydride of HNO3, and its uptake on aqueous aerosol represents a
10 major sink for NOX. Macintyre and Evans (2010) showed that the sensitivity of key
11 tropospheric species such as O3 varies from very small to significant over the range of
12 uptake coefficients (y) for N2O5 obtained in laboratory studies. For example, global O3
13 loss ranges from 0 to over 10%, with large regional variability over the range of reported
14 N2O5 uptake coefficients. However, uptake coefficients forN2O5, ory(N2O5), on
15 atmospheric particles are not well defined, in large part because of uncertainty and
16 variability in aerosol composition. As noted by Brown and Stutz (2012). y(N2O5) is
17 largest (-0.02) for aqueous inorganic aerosols and water droplets, except for nitrate in
18 aerosol, which can reduce y(N2O5) by up to an order of magnitude. Organic aerosol and
19 soot can reduce y(N2O5) by two orders of magnitude or more. The uptake of N2O5 by
20 aqueous aerosols containing chloride (Cl~) and bromide (Br~) has also been associated
21 with the release of gaseous nitryl chloride (C1NO2) from marine (sea-spray) aerosol
22 (Osthoffetal. 2008). C1NO2 readily photolyzes to yield Cl and NO2. Although gas
23 phase C1NO2 can be a major sources of reactive Cl, capable of initiating the oxidation of
24 hydrocarbons (as do OH radicals), C1NO2 causes only modest ozone increases (e.g., ~1 to
25 1.5 ppb for nominal O3 concentrations between 60 and 85 ppb in the Houston airshed)
26 (Simon et al.. 2009). Nitryl chloride is found not only in coastal and marine
27 environments. For example, Thornton et al. (2010) found production rates of gaseous
28 C1NO2 near Boulder, CO from reaction of N2O5 with particulate Cl", at levels similar to
29 those found in coastal and marine environments. They also found that substantial
30 quantities of N2O5 are recycled through C1NO2 back into NOX instead of forming HNO3.
31 The lifetimes of PANs are strongly temperature dependent but they are stable enough at
32 low temperatures to be transported long distances before decomposing to release NO2,
33 which can then participate in O3 formation in regions remote from the original NOX
34 source. Nitric acid acts similarly to some extent, but its high solubility and high
35 deposition rate imply that it is removed from the gas phase faster than PAN and thus
36 would not be as important as a source of NOX in remote regions.
November 2013 2-4 DRAFT: Do Not Cite or Quote
-------
1 The time scale for reactions of NOX to form products shown in the outer box of Figure
2 2-1 typically ranges from a few hours during summer to about a day during winter. As a
3 result, NOX emitted during morning rush hour by vehicles can be converted almost
4 completely to products by late afternoon during warm, sunny conditions. However, the
5 conversion of NO2 to HNO3 and hence the atmospheric lifetime of NOX depends on the
6 concentration of OH radicals, which in turn depends on the concentration of NO2 (e.g.,
7 Valin et al., 2013; Hameedetal.. 1979). Because the time required for mixing of
8 emissions to the surface is similar or longer than the time for oxidation of NOX,
9 emissions of NOX from elevated sources tend to be transformed to more oxidized NOZ
10 products (such as particulate nitrate and HNO3) before they reach the surface. It should
11 be noted that O3 can still be formed aloft in the remnants of power plant plumes.
12 However, because people live in closer proximity to surface sources such as motor
13 vehicles, they are more likely to be exposed to NO and NO2 from these sources. Thus,
14 atmospheric chemical reactions determine the partitioning of a person's exposure to NO2
15 and its reaction products from different sources; and sources of a person's exposure
16 cannot be judged solely by the source strengths given in the national emissions
17 inventories.
18 The oxidation of many species is initiated by OH radicals during the day. During the
19 night, NO3 radicals formed from the reaction of NO2 and O3, assume the role of
20 dominant oxidant for many species such as biogenic and anthropogenic alkenes; for some
21 species (e.g., dimethyl sulfide), it is the overall dominant oxidant (see, e.g.. Brown and
22 Stutz. 2012). The reaction of NO3 with alkenes results in the production of gas phase
23 organic nitrates and secondary organic aerosol formation. Many of the reactions shown in
24 Figure 2-1 occur mainly during the night, when NO3 radicals are most abundant. For
25 example, the formation of N2O5, which has a short lifetime with respect to photolysis and
26 thermal decomposition, is favored at low temperatures during the night. Many of the
27 reactions of NO3 in addition to those of O3 with alkenes also result in the production of
28 OH and HO2 radicals during the night.
29 Isoprene nitrates (INs) and their reaction products could be important for controlling the
30 abundance of NOX and hence the abundance of O3 over the eastern U.S. (e.g.. Perring et
31 al.. 2009). INs and their reaction products could also be important for exporting reactive
32 nitrogen species to remote areas. Yields for IN formation from isoprene oxidation have
33 been estimated to range from 4% (Horowitz et al.. 2007) to 6% to 12% (Xie et al.. 2013)
34 based on model simulations of data collected during the ICARTT (International
35 Consortium for atmospheric research on Transport and Transformation) campaign in
36 2004 and from 7% to 12% in laboratory studies (Lockwood et al., 2010; Paulot et al..
37 2009; Perring et al.. 2009; Horowitz et al.. 2007; von Kuhlmann et al.. 2004). The initial
38 step in the production of INs involves the reaction of isoprene with OH radicals to
November 2013 2-5 DRAFT: Do Not Cite or Quote
-------
1 produce isoprene peroxy radicals. Under low NOX conditions, these radicals favor
2 reaction with HO2 radicals to produce mainly organic peroxides, with smaller amounts of
3 methacrolein, methyl vinyl ketone, and formaldehyde. Under higher NOX conditions,
4 isoprene peroxy radicals can also react with NO resulting in the production of many of
5 the same or similar compounds such as methacrolein and methyl vinyl ketone as
6 well as 'first generation' isoprene nitrates (INs). Lifetimes of the order of one to a few
7 hours can be estimated for these first generation INs based on their reactions with OH
8 radicals and O3 (Lockwood et al., 2010; Paulot et al., 2009). The first generation INs can
9 undergo reactions with OH radicals and O2 and the reaction products can further react
10 with NO (after internal rearrangement) to form secondary organic nitrates such as ethanal
11 nitrate, methacrolein nitrate, propanone nitrate, and methylvinylketone nitrate. The
12 second generation organic nitrates are more stable than the first generation INs because
13 they lack a double C=C bond. Paulot et al. (2009) estimated the yield of NOX from the
14 destruction of second-generation nitrates to be -55%. Obviously, the relative importance
15 of pathways forming nitrates or other products depends on the ambient concentrations of
16 NO and other oxides of nitrogen for which key experimental details are still lacking.
17 In addition to oxidation initiated by OH radicals, isoprene is also oxidized by NO3
18 radicals. Rollins et al. (2009) determined a yield of 70% yield of first generation carbonyl
19 nitrates based on experiments in large reaction chambers. These first generation nitrates
20 can be further oxidized by NO leading to the production of second generation organic
21 (alkyl) nitrates. Mao et al. (2013) estimated that the global mean lifetime is ~5 days for
22 these organic nitrates. They also suggested that the export of INs and other organic
23 nitrates followed by their decomposition is potentially a larger source of NOX to the
24 boundary layer of the western North Atlantic Ocean compared to the export of PANs. It
25 should also be noted that some isoprene nitrates are low enough in volatility that they can
26 partition to the aerosol phase and form PM (e.g.. Rollins et al., 2009).
27 Describing O3 formation accurately requires detailed knowledge of the chemistry of
28 isoprene nitrates (INs). Regional or global models that use a lower yield for forming
29 these nitrates and a higher yield for recycling NOX tend to over-predict O3 concentrations
30 in areas with high isoprene emissions, such as the Southeast compared to those that have
31 a higher yield for the formation of these nitrates and/or a lower yield for their recycling
32 back to NOX (U.S. EPA. 2013b). The formation rates and the rates that are used to
33 recycle INs and other organic nitrates back to NOX also have implications for calculating
34 the yield of O3 from isoprene emissions. For example, Fiore et al. (2005) found a
35 negative dependence of O3 production on isoprene emissions in the eastern U.S. in
36 summer, whereas Mao etal. (2013) found a positive yield for O3 from isoprene
37 emissions. Xie etal. (2013) determined that the uncertainties in the isoprene nitrates
38 could affect ozone production by 10% over the U.S. and that uncertainties in the NOX
November 2013 2-6 DRAFT: Do Not Cite or Quote
-------
1 recycling efficiency had a larger affect than the isoprene nitrate yield. These
2 considerations underlie the importance of further laboratory and field studies to more
3 quantitatively determine the response of O3 to changes in isoprene emissions at different
4 NOX levels.
5 As mentioned earlier, NO and NO2 are important precursors of O3 formation. However,
6 because O3 changes in a nonlinear way with the concentrations of its precursors (NOX
7 and VOCs), it is unlike many other atmospheric species whose rates of formation vary
8 directly with emissions of their precursors. At the low NOX concentrations found in
9 environments ranging from remote continental areas to rural and suburban areas
10 downwind of urban centers (c.f., Figure 2-12. low-NOx regime), the net production of O3
11 typically increases with increasing NOX. In this low-NOx regime, the overall effect of
12 the oxidation of VOCs is to generate (or at least not consume) radicals, and O3
13 production varies directly with NOX. In a high-NOx regime, NO2 reacts with OH
14 radicals to form HNO3 (e.g.. Hameed et al.. 1979). Otherwise, these OH radicals would
15 oxidize VOCs to produce peroxy radicals, which in turn would oxidize NO to NO2. In
16 this regime, O3 production is limited by the availability of radicals (Tonnesen and
17 Jeffries. 1994) and O3 shows only a weak dependence on NOX concentrations. Reaction
18 of O3 with NO in fresh motor vehicle exhaust depletes O3 in urban cores, but O3 can be
19 regenerated during transport downwind of urban source areas and additional chemical
20 production of O3 can occur, resulting in higher ozone concentrations than found upwind
21 of the urban center. Similar depletion of O3 can occur in power plant plumes with
22 subsequent O3 regeneration downwind.
23 Brown et al. (2012) conducted a field study comparing nighttime chemistry in the plumes
24 of power plants, one with selective catalytic reduction (SCR) NOX emissions controls
25 and one without these controls, in Texas. They noted that the plume from the power plant
26 with SCR controls did not have enough NOX to deplete all of the O3 present in
27 background air. As a result, almost all of the NOX in the plume was oxidized to NOZ
28 species, and so the NOX that was oxidized was not available to participate in O3
29 production the next day. This situation contrasts with that in the plume from the power
30 plant without controls. In that plume, there was minimal formation of NOZ species.
31 Instead, NOX was more nearly conserved and the NO2 that was formed from the reaction
32 of emitted NO with O3 photolyzed the following morning, leading to higher O3
33 formation rates compared to plumes from the plant with controls.
November 2013 2-7 DRAFT: Do Not Cite or Quote
-------
2.3 Sources
2.3.1 Overview
1 The major sources of NOX in the U.S. identified from the 2008 National Emission
2 Inventory U.S. EPA (2011) are described in Figure 2-2.: The values shown are
3 nationwide averages and may not reflect an individual person's exposures to NO2.
4 Highway vehicles are the largest source of NOX, contributing 39% of total emissions.
5 Off-highway vehicles account for 19% of emissions, fuel combustion by electric utilities
6 makes up 17% of emissions, and industrial fuel combustion accounts for 8% of
7 emissions. Other sources listed in Figure 2-2 account for less than 5% of national
8 emissions each. Sources that are not listed in Figure 2-2 can still be important for
9 exposure. For example, intense industrial operations including cement plants are not
10 listed as nationally important sources, but they are subject to variable emissions with high
11 peaks (Walters et al., 1999). Note that lightning emissions of NO are not in included in
12 Figure 2-2. Estimates of emissions of NO from lightning found in the literature are given
13 in Section 2.3.9.
1 This section currently discusses data from the 2008 National Emissions Inventory, version 3 U.S. EPA
(2011). The 2011 National Emissions Inventory became available to the public in November 2013 and
will be incorporated into the Second External Release Draft of the Integrated Science Assessment for
Oxides of Nitrogen.
November 2013 2-8 DRAFT: Do Not Cite or Quote
-------
HIGHWAY VEHICLES
OFF-HIGHWAY
FUEL COMB. ELEC. UTIL
FUEL COMB. INDUSTRIAL
BIOGENICS
FUEL COMB. OTHER
PETROLEUM & RELATED INDUSTRIES
OTHER INDUSTRIAL PROCESSES
MISCELLANEOUS
WASTE DISPOSAL & RECYCLING
METALS PROCESSING
CHEMICAL & ALLIED PRODUCT MFG
STORAGE & TRANSPORT
SOLVENT UTILIZATION
f
Note: Lightning is not included.
Source: U.S. EPA (2011)
| I I I | I I I | I I I I | I I I I | I I I I | I I I | I I I I | I I I
0 2000000 4000000 6000000 8000000
Emissions (tons/year)
Figure 2-2 Major sources of NOx averaged over the United States, 2008.
i
2
3
4
5
6
7
Figure 2-3 describes the decrease in NOX emissions in the U.S. for two major sources
over the period from 1990 to 2012. Overall emissions decreased by more than 50% over
this period, with substantial declines for both on-road vehicles and fuel combustion. It is
possible to describe emissions through 2012 using the 2008 National Emissions
Inventory because estimates later than 2008 are posted as updates, with methods for
estimating sector emissions in these later years clearly described in the inventory
documentation. These updates are still considered as versions of the 2008 inventory.
November 2013
2-9
DRAFT: Do Not Cite or Quote
-------
to
o
to
c
o
LJJ
25x10 -
20-
15-
10-
5-
0J
Source: U.S. EPA (2011)
I I I
CD O O
CD O O
O O ->•
I I
IO NJ
O O
O O
ro w
i
hO
o
o
I
I\J
o
o
Cn
o
o
01
o
o
I I
to 10
o o
o o
00 CD
I
O
o
I
M
o
I
ro
o
Year
Figure 2-3 National average NOx emissions attributed to on-road vehicles
and fuel combustion, from 1990 to 2012.
i
2
3
4
5
6
7
8
9
10
11
12
13
14
15
The composition of NOX, specifically the ratio of NO2 to total NOX, can vary
considerably among source types or between sources of the same type. Emission height
can also have a substantial influence on resulting NOX concentrations at ground level
where most exposure occurs, with some sources emitted at or near ground level (e.g.,
highway vehicles) and others aloft from tall stacks (e.g., electric utility fuel combustion).
These considerations were described in detail in the 2008 ISA for Oxides of Nitrogen.
There have been some new developments with regard to composition as new control
technologies have become available for mobile sources. As described in the 2008 ISA for
Oxides of Nitrogen, the fraction of NO2 in total NOX emissions from the exhaust of
gasoline vehicles has generally been reported as only a few percent, but catalyzed diesel
particle filters (CDPFs) can increase the NO2 fraction to 30-70% (U.S. EPA. 2008c).
Improvement of NOX emission control technology is an active area of research. A
number of new advances have been reported for electric utility and motor vehicle engine
emission abatement. For electric utilities, emission control strategies fall into three broad
categories: (1) pre-combustion modification through fuel purification or fuel choice to
November 2013
2-10
DRAFT: Do Not Cite or Quote
-------
1 reduce the amount of nitrogen introduced, (2) combustion modification by reducing
2 combustion temperature, creating oxygen deficient conditions, or varying residence time
3 within different parts of the combustion zone; or (3) post-combustion treatment. Emission
4 control technologies were thoroughly reviewed recently (Skalska et al.. 2010). The most
5 widely used control technology is selective catalytic reduction (SCR) with ammonia in
6 the presence of oxygen (Bruggemann and Keil. 2008).
7 For gasoline powered vehicles, emission control is usually achieved with a three-way
8 catalyst for simultaneous control of NOX, CO, and hydrocarbons (Heeb et al., 2008). in
9 which NO is reduced to N2 by CO (Roy and Baiker. 2009). However, this approach is not
10 as effective for diesel or lean burning gasoline engines because oxygen levels are too
11 high (Brandenberger et al.. 2008; Takahashi et al.. 2007).
12 New technologies developed for diesel are NOX storage reduction (NSR) and selective
13 NOX recirculation (SNR) (Rov and Baiker. 2009). These and other methods of NOX
14 emission control applied to diesel engines before 2010 were recently reviewed by Skalska
15 et al. (2010). Since 2010, SCR, which was originally developed for electric utility
16 emission controls, has been applied to diesel emission control to achieve even lower
17 emissions. SCR with ammonia proved unfeasible for diesel emission control because of
18 slip, manipulation, storage, and corrosion problems (Skalska et al., 2010). but replacing
19 ammonia with urea led to successful application of SCR control technology to diesel
20 emissions control (Johnson et al.. 2009).
2.3.2 Highway Vehicles
21 Highway vehicles account for a large fraction of NOX emissions in high traffic areas. For
22 example, on-road vehicles were estimated to account for about 80% of anthropogenic
23 NOX concentrations in the Los Angeles area (Mcdonald et al.. 2012) and 72% in the
24 Atlanta area (Pachon et al.. 2012).
25 The relative importance of diesel and gasoline engine related NOX emissions varies
26 considerably among airsheds. Mcdonald et al. (2012) estimated that diesel engines were
27 the dominant on-road NOX sources in the San Joaquin Valley, accounting for up 70% of
28 NOX emissions. In contrast in Fulton County, Georgia it was estimated that 60% of on-
29 road NOX emissions were from gasoline vehicles and 40% from diesel (Pachon et al..
30 2012). Mcdonald et al. (2012) estimated that in California, gasoline engine-related NOX
31 emissions steadily decreased by 65% over the period from 1990 to 2010. They also found
32 that the ratio of NOX emission factors for heavy-duty diesel versus light-duty gasoline
33 engines grew from ~3 to ~8 between 1990 and 2010 due to improved effectiveness of
34 catalytic converters on gasoline engines.
November 2013 2-11 DRAFT: Do Not Cite or Quote
-------
1 NOX emissions from on-road diesel engines in the U.S. have decreased substantially as
2 the result of stricter emission standards. Emission standards for heavy duty diesel trucks
3 were first established at 10.7 g/bhp-h in 1988 and decreased to 2.0 g/bhp-h for the 2004
4 model year and after (66 FR 5002). Standards were achieved mainly through
5 development of selective catalytic reduction (SCR) with urea (Johnson et al. 2009). A
6 NOX emission standard of 0.20 g/bhp-h was gradually phased in for model years 2007
7 through 2010 (66 FR 5002), so that emission standards from heavy duty diesel trucks
8 have been reduced by more than a factor of 50 between 1988 and 2010.
9 As discussed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). increases in the
10 NO2/total NOX emission ratio have been observed in exhaust from diesel engines
11 equipped with catalyzed diesel particulate filters (CDPFs) and diesel oxidation catalysts
12 (DOCs). Both CDPF and DOC control techniques involve oxidation of exhaust pollutants
13 through catalysis reactions for regeneration of filter media and effective removal of
14 targeted species, respectively. During catalysis, combustion-derived NO may be oxidized
15 to NO2, potentially increasing the amount of NOX emitted as NO2. The SCR control
16 process also intentionally generates NO2 to manipulate NO2/NOX ratios in order to
17 optimize performance, and integrated NOX/PM control systems are currently being
18 developed (Johnson et al.. 2009). However, the SCR control system is also designed to
19 eliminate the excess NO2 generated.
20 The HEI ACES study sampled NOX emissions from a variety of engines equipped with
21 exhaust gas recirculation (EGR) and DOC/CDPF, which represent the typical control
22 device configuration used in 2007-2009 EPA compliant diesel engines (Khalek et al..
23 2011). NOX emissions averaged 1.09 ± 0.15 g/bhp-h, which is 73% and 9% lower than
24 the 1998 and 2007 EPA standards, respectively. However, average NO2 emissions from
25 2007-2009 compliant engines were 0.73 ± 0.12 g/bhp-h, indicating that NO2 from newer,
26 2007 engines is 33% higher than emissions from 1998 EPA compliant engines. NOX
27 emissions from 2010 EPA compliant engines designed to meet the 0.20 g/bhp-h standard
28 are also being evaluated in the HEI ACES study, but results are not yet published.
2.3.3 Off-Highway Vehicles
29 Off-highway sources include aviation, marine, and railroad engines as well as agricultural
30 and industrial equipment, all of which emit NOX through combustion processes.
31 Aviation, marine, and railroad engines are treated in the following section. In this section,
32 other off-highway engines are considered. A few examples of these engines include farm
33 tractors, excavators, bulldozers, and wheel loaders. Emissions from the nonroad source
34 sector can also significantly contribute to local and national air quality. On a national
November 2013 2-12 DRAFT: Do Not Cite or Quote
-------
1 scale, Zhu etal. (2011) estimated that nonroad diesel engines contribute 12% of total
2 NOX emissions from mobile sources.
3 Though limited measurement data exists on non-road diesel engine emissions, studies
4 show that these emissions are widely variable across engine type, engine operating mode,
5 engine age, and ambient temperature, with most studies focusing on differences in
6 emissions during different operation modes (Fuet al.. 2012b; Zhu etal.. 2011;
7 Abolhasani et al.. 2008). Fu etal. (2012b) studied NOX emissions from a variety of
8 construction equipment in Beijing, China, including 12 excavators and 8 wheel loaders.
9 They found that NOX emissions factors during working mode were 3.66 and 1.36 times
10 higher than emission factors during idling mode and moving mode, respectively. Similar
11 to Fuetal. (2012b); Abolhasani et al. (2008) observed large differences in NOX
12 emissions from three different excavators during various operating modes. These
13 differences in emissions were larger than the variation in emissions among different
14 engines. Together these studies emphasize the importance of considering intercycle
15 operating modes when estimating non-road diesel engine emissions using modeling
16 approaches.
17 EPA has set a series of standards to reduce non-road diesel NOX emissions, referred to as
18 Tier 1-4 standards. The most recent standard, Tier 4, was introduced in May 2004, and
19 the fleet turnover is currently underway, covering a time period between 2008 and 2015.
20 In most cases, advanced diesel engine design has been used to comply with these
21 standards.
2.3.4 Aviation Emissions
22 Airport-related NOX emissions can significantly impact local and regional air quality. In
23 the U.K., within a 2-3 km radius of London Heathrow Airport, Carslaw et al. (2006)
24 reported that airport emissions can comprise up to 15% of total NOX in the background
25 air. In Atlanta, GA, Unal et al. (2005) showed that roughly 2.6% of regional NOX
26 concentrations can be attributed to emissions from activities at Hartfield-Jackson
27 International airport. Compared to airport-related emissions of other gaseous pollutants
28 (e-g-, NH3, CO, SO2, VOC), airport NOX emissions had the largest contribution to
29 regional air quality in Atlanta, GA.
30 Different aircraft operations and ground-level airport activities impact NOX emissions.
31 Higher NOX emissions are typically observed during more power-intensive operations
32 such as landing and take-off cycles (LTO) compared to taxiing and idling (Klapmeyer
3 3 and Marr. 2012: Mazaheri etal.. 2011. 2009). Klapmever and Marr (2012) showed that
34 emission indices (mass of pollutant per mass of fuel used) for LTO cycles can be 3 to 7
November 2013 2-13 DRAFT: Do Not Cite or Quote
-------
1 times higher than taxiing operations at a regional airport in Virginia. Engine type and
2 aircraft/cargo weight also impact the amount of NOX emitted from an aircraft (Carslaw et
3 al., 2008). Ground-level support activities and auxiliary power units are also important
4 sources of NOX emissions at airports. In some cases, these NOX emissions from ground
5 level activities can be comparable to, or more than, NOX emissions from aircraft
6 operations (LTO cycles) (Klapmever and Marr. 2012).
2.3.5 Shipping Emissions
7 Globally, shipping emissions are a significant source of nitrogen emissions, accounting
8 for more than 14% of all global nitrogen emissions from fossil fuel combustion (mostly
9 NOX) (Corbett et al.. 1999). On a regional scale, the contribution of shipping emissions
10 to total NOX emissions is variable and can be a substantial fraction near port cities (Kim
11 etal.. 20lib: Williams et al.. 2009: Vutukuru and Dabdub. 2008). In Los Angeles, CA
12 Vutukuru and Dabdub (2008) estimated that commercial shipping contributed
13 approximately 4.2% to total NOX emissions in 2002. Using the NEI-05, Kim et al.
14 (20 lib) estimated that roughly 50% of NOX concentration near the Houston Shipping
15 Channel is associated with commercial shipping emissions. However, this estimation is
16 much higher than observed in satellite and aircraft measurements.
17 NOX emissions vary among different ship types and operation modes. NOX emissions are
18 typically lower in docked ships compared to underway ships. In the Houston Shipping
19 Channel, Williams et al. (2009) calculated NO2 emission factors, and they found that
20 they were highest for bulk freight carriers (87.0 ± 29.6 g NO2/kg fuel), followed by
21 underway tanker ships (79 ± 23 g NO2/kg fuel) and other smaller underway vessels
22 (stationary vessels) (60 g NO2/kg fuel). Despite the large variability in shipping
23 emissions, no clear trend in emissions could be drawn from differences in engine load
24 and speed. Though these studies report unique findings, uncertainties exist in current
25 shipping emission inventories, making it challenging to accurately estimate air quality
26 impacts of shipping emissions (Kim etal.. 201 lb: Williams et al.. 2009: Corbett et al..
27 1999).
2.3.6 Locomotive Emissions
28 Locomotives powered by diesel engines are a source of NOX emissions. Using a fuel-
29 based approach to quantify emissions, Dallmann and Harley (2010) estimated that diesel
30 locomotives emitted on average 50% of total NOX from all non-road mobile sources and
31 roughly 10% of total NOX from all mobile sources in the U.S. from 1996-2006
November 2013 2-14 DRAFT: Do Not Cite or Quote
-------
1 (Dallmann and Harley. 2010). Locomotives can comprise a much larger fraction of NOX
2 emissions for areas in or near large rail yard facilities (>90% of emissions), including
3 NO2 non-attainment areas (U.S. EPA, 2010a). In a year-long study at the Rougemere Rail
4 Yard facility near Dearborn, MI, 98% of NOX emissions was attributed to locomotive
5 operation, with only minimal impacts from other sources such as on-road mobile sources
6 and stationary sources (U.S. EPA. 2009a). Cahill et al. (2011) measured gaseous and PM
7 pollutants during a two-week period near the Roseville Rail Yard in Placer County,
8 California. They observed several transient NOX emission events, where NO levels of
9 100s of ppb were observed downwind of the Rail Yard, which was roughly 7 times larger
10 than the observed urban background NO.
11 Limited measurement data exist on emissions from locomotives. Among these studies,
12 there is evidence that NOX levels from locomotives vary spatially and temporally
13 depending upon train activity. On a per fuel basis, engine idling is associated with the
14 highest NOX emission factor compared to activities at higher engine loads (or notches)
15 (Sawant et al.. 2007). Sawant et al. (2007) emphasized the importance of this trend since
16 many types of locomotives (switching yard locomotives) spend a significant amount of
17 time idling (U.S. EPA. 2010a; Sawant et al.. 2007). However, most of these studies
18 measured emissions from locomotives equipped with older emission control technology
19 and may not entirely reflect emissions from locomotives that are currently operating.
2.3.7 Fuel Combustion for Electrical Utilities and Industrial Use
20 NOX from coal combustion is emitted primarily as NO, increases with decreasing
21 temperature, and also increases with increasing nitrogen content of the coal (Kim et al..
22 201 la). Emission control strategies, including widely used SCR controls described in
23 Section 2.3.1. have been thoroughly reviewed by Skalska et al. (2010). Decreased NOX
24 emissions as well as decreases in predicted ozone concentrations have been attributed to
25 power generation controls Gego et al. (2008). Satellite data consistent with decreasing
26 power plant emissions over time are presented in detail in Section 2.4.3.
27 NOX emissions from electric utility power plants have decreased considerably since the
28 Clean Air Act Amendments of 1990 (CAA. 1990b). The Acid Rain Program (described
29 by Title IV of the Amendments) targeted SO2 and NOX emissions from coal-fired power
30 plants and other major stationary sources. Title I addressed regional transport of ground
31 level O3, aiming to reduce transport of O3 across state boundaries in the eastern U.S.
32 Title I led to the formation of the Ozone Transport Assessment Group (OTAG), a
33 partnership between the U.S. EPA, the Environmental Council of the States and various
34 industry and environmental groups to assess long-range transport of O3 and O3
November 2013 2-15 DRAFT: Do Not Cite or Quote
-------
1 precursors, and their efforts eventually resulted in the U.S. EPA's publication of the NOX
2 SIP Call (63 FR 57356) to control NOX concentrations in order to reduce O3
3 concentrations.
4 Title IV set NOX emission limitations for coal-fired electric utility boilers to be
5 implemented by 1995. The first phase set targets for 239 older coal-fired generating units.
6 Emission targets were achieved mainly by retrofitting with low-NO burners or similar
7 modifications that control fuel and air mixing to limit NO formation. Average NOX
8 emission rates from these units decreased by 40% between 1990 and 1996.
9 This first phase was followed in 1998 by the NOX SIP Call which required 22 eastern
10 states (later amended to 20 states) and the District of Columbia to set statewide
11 O 3-season NOX budgets, with emission reduction measures to be in place by 2003. As a
12 result, summertime NOX emission rates from the electric power generating units, which
13 were subject to control, decreased by approximately 50% between 1999 and 2003, as
14 measured by Continuous Emission Monitoring System (CEMS) measurements at a subset
15 of these power generating units (Frost et al.. 2006). The NOX SIP Call target emission
16 rates were considerably lower than those established by the Title IV Acid Rain Program,
17 and were achieved primarily through more advanced controls, such as selective catalytic
18 reduction (SCR) and selective non-catalytic reduction (SNCR).
19 Satellite based observations confirm these reductions in power plant NOX emissions, van
20 der Aetal. (2008) found a consistent decline of NO2 concentrations at 7% per year for
21 the period from 1996-2006 over the eastern United States. Stavrakou et al. (2008) found
22 a decrease in emissions of 4.3% per year over the Ohio River Valley from 1997 to 2006
23 from the tropospheric NO2 column based on decreases in NO2 column measurements in
24 locations where electric utility power plant emissions predominate. National scale spatial
25 variability is further explored in Section 2.5.1.
2.3.8 Biogenics and Wildfires
26 Major biogenic sources of NOX in the U.S. include controlled biomass burning,
27 vegetation, and soil. Uncertainties in natural NOX emissions are much larger than for
28 anthropogenic NOX emissions.
29 Emissions from wildfires can produce enough NOX to cause local and regional
30 degradation of air quality in some regions (Pfister et al.. 2008). Roughly 15% of global
31 NOX emissions are from biomass burning (Penman et al.. 2007). Burling et al. (2010)
32 reported that NOx emissions from southwest U.S. vegetation ranging from 2.3 to 5.1
33 g/kg, with the majority of the NOX present as NO. Emissions vary considerably among
November 2013 2-16 DRAFT: Do Not Cite or Quote
-------
1 different species of biota, making it difficult to estimate emissions for key ecosystems,
2 such as extratropical forests (McMeeking et al.. 2009). Emissions can be more than
3 double per amount of energy consumed for forests than for shrubs (Mebust et al.. 2011).
4 NOX emissions increase with increasing fuel nitrogen content (Burling et al., 2010).
5 Burning conditions also play an important role, with biomass emissions higher during
6 flaming than during smoldering conditions (Burling et al.. 2010). but fuel nitrogen
7 content is more important than burning conditions (Burling et al.. 2010; McMeeking et
8 al.. 2009). Biomass burning also produces HONO in both laboratory (Roberts et al..
9 2010; Keene et al.. 2006) and field conditions (Yokelson et al.. 2009; Yokelson et al..
10 2007).
11 NOX concentrations in plumes from boreal forest fires as well as tropical biomass
12 burning for agricultural purposes decay with time due to the formation of HNO3, and
13 NOX reservoir species such as PAN (e.g., Alvarado et al.. 2010; Leung et al.. 2007; Real
14 et al.. 2007; Mauzerall et al.. 1998; Jacob etal. 1992). Rapid conversions of NOX to
15 PAN and pNO3 have been observed in wildfire plumes (Akagi et al.. 2012).
16 Both nitrifying and denitrifying organisms in the soil can produce NOX, mainly in the
17 form of NO. Emission rates depend mainly on the amount of applied fertilizer, soil
18 temperature, and soil moisture. Nationwide, about 60% of the total NOX emitted by soils
19 is estimated to occur in the central corn belt of the United.States. Spatial and temporal
20 variability in NOX emissions from soil leads to considerable variability in emission
21 estimates. However, these emissions are relatively low, comprising only about 6% of
22 total anthropogenic NOX emissions. Additionally, these emissions occur mainly during
23 summer when O3 concentrations are highest across the entire country, including areas
24 where anthropogenic emissions are low.
2.3.9 Lightning
25 Lightning is not included in the emission inventory of Figure 2-2. but it is an important
26 source that varies with season, region, and altitude. It has been well established that
27 lightning produces NO, which influences atmospheric chemistry (Chameides and Walker.
28 1973; Crutzen. 1973; Crutzen. 1970). For example. Noxon (1976) and Noxon (1978) first
29 reported direct observations of NO2 concentrations up to 100 ppb near lightning flashes.
30 Lightning is usually a minor contributor to urban ground-level NOX concentration, but it
31 can be important on a regional or national scale. For example, lightning has been widely
32 recognized as a particularly important source over the U.S. in the summer, especially in
33 the Southeast (Hudman et al.. 2007; Bond etal.. 2001; Biazar and McNider. 1995).
34 Kaynak et al. (2008) estimated that nearly 30% of all U.S.-wide NOX generation during
November 2013 2-17 DRAFT: Do Not Cite or Quote
-------
1 July could be attributed to lightning. However, lightning-generated NOX generally
2 deposits in the free troposphere Pickering et al. (1998). and several recent studies
3 concluded that an even greater fraction ends up in the free troposphere than previously
4 thought (Fang etal.. 2010: Ottetal.. 2010: Kavnak et al.. 2008). NOX depositing in the
5 free troposhere has been observed to occur in part because intracloud flashes are more
6 productive relative to cloud-to-ground flashes than previously considered (Ott et al..
7 2007: DeCaria et al.. 2005). For example, Fang etal. (2010) estimated that lightning-
8 generated NOX over the U.S. for July 2004 was -40% of the anthropogenic emissions for
9 the same period, but the authors estimated that -98% is formed in the free troposphere;
10 therefore, contributions to the ground-level NOX burden are low because most of this
11 NOX is oxidized to NO3" -containing species during downward transport into the
12 planetary boundary layer. The remaining 2% is formed within the planetary boundary
13 layer itself.
14 There is greater uncertainty in NOX production from lightning than from other sources,
15 with recent global estimates ranging from 2 to 8 Tg/year (Schumann and Huntrieser.
16 2007). Recent research has advanced our understanding of NOX production from
17 lightning, with results generally suggesting that lightning is a somewhat more important
18 source than previously thought. Recent research suggests that the amount of NOX
19 produced per flash of lightning (a widely used modeling parameter) was previously
20 substantially underestimated (Jourdain etal.. 2010). Ott et al. (2007) and Ottetal. (2010)
21 recommended a mean value of 500 moles NOX per flash for both cloud-to-ground and
22 intracloud flashes based on field observations. Peterson and Beasley (2011) investigated
23 ice crystals as a catalyst for production of NOX by lightning, and they estimated 2.7 times
24 more NO is produced when ice crystals are present. Finally, enhancement of lightning by
25 urban aerosols, first suggested by Westcott (1995). has been further documented (Kar et
26 al.. 2009). and its impact on increasing lightning generated NOX is an active area of
27 research (Yuan etal.. 2012: Wang etal.. 2011).
2.3.10 Oil and Gas Development
28 The oil and gas production sector is an increasing source of NOX, with 2008 emission
29 estimates of 400,000 tons nationally. A number of operational activities contribute to
30 emissions from oil and gas production facilities. Pacsi etal. (2013) estimated that routine
31 operating activities from the Barnett Shale production facility near Dallas, Texas can emit
32 roughly 46 to 30 tons of NOx/day, depending on the demand for natural gas electricity
33 generation. Non-routine gas flares can also result in episodic peaks of large NOX
34 emissions, affecting local air quality (Olaguer. 2012). While a majority of production
35 facilities are located in remote areas, emissions can impact regional air quality, including
November 2013 2-18 DRAFT: Do Not Cite or Quote
-------
1 major urban centers (Olaguer. 2012) and national parks (Rodriguez et al., 2009)
2 downwind of these facilities.
2.4 Measurement Methods
2.4.1 Federal Reference and Equivalent Methods
3 Nitric oxide (NO) is routinely measured using the chemiluminescence induced by its
4 reaction with O3 at low pressure. The Federal Reference Method (FRM) for NO2 makes
5 use of this technique of NO detection with a prerequisite step that is meant to reduce NO2
6 to NO on the surface of a molybdenum oxide (MoOx) substrate heated to between 300
7 and 400 °C. On June 1, 2012, an automated Federal Equivalent Method (FEM) for
8 measuring NO2 using a photolytic converter to reduce NO2 to NO met the equivalency
9 specifications outlined in 40 CFR Part 53 and was approved by the U.S. EPA. Although
10 photolytic converters have lower conversion efficiencies than FRM based analyzers, they
11 have been found to be stable over a period of at least two months (Pollack et al., 2011).
12 Because the FRM monitor cannot detect NO2 specifically, the concentration of NO2 is
13 determined as the difference between the NO in the air stream passed over the heated
14 MoOx substrate (measuring total oxides of nitrogen) and the NO in the air stream that
15 has not passed over the substrate.
16 However, the reduction of NO2 to NO on the MoOx catalyst substrate also reduces other
17 oxidized nitrogen compounds (i.e., NOZ compounds shown in the outer box of Figure
18 2-1) to NO. Hence, the chemiluminescence analyzers could be subject to unknown and
19 varying interference. This interference by NOZ compounds has long been recognized
20 based on intercomparisons of measurements using the FRM and other techniques for
21 measuring NO2 (Dunlea et al. 2007; Steinbacher et al.. 2007; U.S. EPA. 2006;
22 McClenny et al.. 2002; Parrish and Fehsenfeld. 2000; Nunnermacker et al.. 1998;
23 Croslev. 1996; U.S. EPA. 1993; Rodgers and Davis. 1989; Fehsenfeld et al.. 1987). The
24 sensitivity of the FRM to potential interference by individual NOZ compounds was found
25 to be variable, depending on characteristics of individual monitors, such as the design of
26 the instrument inlet, the temperature and composition of the reducing substrate, and on
27 the interactions of atmospheric species with the reducing substrate.
28 Only recently have attempts been made to systematically quantify the magnitude and
29 variability of the interference by NOZ species in ambient measurements of NO2. Dunlea
30 et al. (2007) found an average of about 22% of ambient NO2 (~9 to 50 ppb) measured in
31 Mexico City over a five week period during the spring of 2004 was due to interference
November 2013 2-19 DRAFT: Do Not Cite or Quote
-------
1 from NOZ compounds. Comparable levels of NO2 are found in many locations in the
2 U.S. However, similar comparisons have not been carried out under conditions typical for
3 AQS monitoring sites in the United States. Dunlea et al. (2007) compared NO2 measured
4 using the conventional chemiluminescent instrument with other (optical) techniques. The
5 main sources of interference were HNO3 and various organic nitrates. Efficiency of
6 conversion was estimated to be -38% for HNO3 and -95% for PAN and other organic
7 nitrates. Peak interference of up to 50% was found during afternoon hours and was
8 associated with O3 and NOZ compounds such as HNO3 and the alkyl and multifunctional
9 alkyl nitrates.
10 Lamsal et al. (2008) used data for the efficiency of reduction of NOZ species on the
11 MoOx catalytic converters to estimate seasonal correction factors for NO2 measurements
12 across the U.S. These factors range from <10 % in winter to >80% with the highest
13 values found during summer in relatively unpopulated areas. In general, interference by
14 NOZ species in the measurement of NO2 is expected to be larger downwind of urban
15 source areas and in relatively remote areas because of the conversion of NO2 to NOZ
16 during transport downwind of source areas.
17 In a rural study in Switzerland, Steinbacher et al. (2007) compared continuous
18 measurements of NO2 from a chemiluminescence analyzer with a MoOx catalytic
19 converter (CL/MC) with measurements from a photolytic converter (CL/PC) that reduces
20 NO2 to NO. They found the conventional technique using catalytic reduction (as in the
21 FRM) overestimated the measured NO2 compared to the photolytic technique, on average
22 by 10% during winter and 50% during summer.
23 Villenaetal. (2012) and Kleffmann et al. (2013) suggested that negative interference in
24 the chemiluminescent method using the photolytic converter could occur by the
25 production of HO2 and RO2 radicals by the photolysis of VOCs, e.g., glyoxal, in the
26 photolytic converter. Subsequent to photolysis and prior to detection, these radicals react
27 with NO that is either produced by the photolytic converter or already in the sampling
28 stream. Since the chemiluminescent techniques rely on detection of NO, a negative
29 artifact results. The most direct evidence for this artifact was found at high concentrations
30 in a smog chamber containing 1 ppm glyoxal, a concentration more than a thousand times
31 higher than typically found in ambient air. Similar indications were also found by
32 Kleffmann et al. (2013) in a street canyon (in Wuppertal, Germany) and in an urban
33 background environment (University of Santiago, Chile). However, Kleffmann et al.
34 (2013) also found that the magnitude of the negative artifact is smaller when a light
35 source with a smaller spectral range is used and that this artifact is expected to be most
36 apparent under high VOC conditions, such as in street canyons.
November 2013 2-20 DRAFT: Do Not Cite or Quote
-------
1 To summarize the discussion of NO2 measurements by the FRM: the current FRM for
2 determining ambient NOX concentrations and then reporting NO2 concentrations by
3 subtraction of NO is subject to a consistently positive interference by NOX oxidation
4 products, chiefly HNO3 and PAN as well as other oxidized TV-containing compounds.
5 Note, though, the magnitude of this positive bias is largely unknown and can change
6 rapidly. Measurements of these oxidation products in urban areas are sparse.
7 Concentrations of these oxidation products are expected to peak in the afternoon because
8 of the continued oxidation of NO2 emitted during the morning rush hours during
9 conditions conducive to photochemistry in areas well downwind of sources, particularly
10 during summer.
11 Within the urban core of metropolitan areas, where many of the ambient monitors are
12 sited close to strong NOX sources such as motor vehicles on busy streets and highways
13 (i.e., where NO2 concentrations are highest), the positive artifacts due to the NO2
14 oxidation products are much smaller on a relative basis, typically <~10%. Conversely,
15 the positive artifacts are larger in locations more distant from NOX sources (i.e., where
16 NO2 concentrations are lowest) and could exceed 50%.
2.4.2 Other Methods for Measuring NO2
17 Optical methods such as those using differential optical absorption spectroscopy (DOAS)
18 or laser induced fluorescence (LIF) are also available. However, these particular methods
19 are even more expensive than either the FRM monitors or photolytic reduction technique
20 and require specialized expertise to operate as well; moreover, the DOAS obtains a path-
21 integrated rather than a point measurement. Cavity attenuated phase shift (CAPS)
22 monitors are an alternative optical approach requiring much less user intervention and
23 expense than either DOAS or LIF (Kebabian et al.. 2008). At first glance, it might appear
24 that this technique is not highly specific to NO2, as it is subject to interference by species
25 that absorb at 440 nm such as 1,2-dicarbonyl compounds. However, this source of
26 interference is expected to be small (-1%), and if necessary, the extent of this
27 interference can be limited by shifting the detection to longer wavelengths and adjusting
28 the lower edge of the detection band to 455 nm. In principle, detection limits could be
29 <30 ppt for a 60s time scale.
30 Lee et al. (2011 a) describe the development of a dual continuous - wave mode quantum
31 cascade - tunable infrared laser differential absorption spectrometer or QC-TILDAS to
32 measure NO2 and HONO simultaneously. The one-second detection limit (S/N = 3) is
33 30 ppt. Field comparisons of measurements of NO2 by CAPS and QC-TILDAS to NO2
34 measured by chemiluminescence monitors with MoOx converters (CL/MC) in Houston,
November 2013 2-21 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
TX, are shown in Figure 2-4 and Figure 2-5. Both figures show very high R2 and close
agreement over concentrations ranging from <1 ppb to >30 ppb and both comparisons are
characterized by small non-zero intercepts. For the CAPS instrument (see Figure 2-4).
slightly higher values than those reported by the CL/MC monitor are seen at
concentrations < ~2 ppb. The CAPS - CL/MC (Thermo Electron 421) data were obtained
over 4 days in a parking lot located -200 meters from a major arterial highway (Route 3
in Billerica, MA). Figure 2-5 shows that the QC-TILDAS obtains slightly lower
concentrations than reported by CL/MC at concentrations < ~1 ppb. The measurements
shown in Figure 2-5 were made under rather highly polluted conditions in Houston, TX,
over a period of 4 weeks during the SHARP (Study of Houston Atmospheric Radical
Precursors) campaign (Olaguer et al.. In Press). Under polluted conditions such as these,
the possibility of interference by NOZ species in the measurements by CL/MC should be
considered. Interference caused by HNO3 and PAN is estimated to be <1 ppb using the
conversion efficiencies obtained by Dunlea et al. (2007) and concentrations of HNO3 and
PAN obtained during SHARP.
0,1
10 :
_Q
Q.
O.
l/l
CL
<
U
0.1
CAPS vs CL
1.007x-0.0357
R2 = 0.9939
10
Chemiluminescence NO2 ppb
Source: NCEA, using data from Kebabian et al. (2008)
Figure 2-4 Comparison of NO2 measured by CAPS (Cavity Attenuated Phase
Shift) spectroscopy to NO2 measured by Chemiluminescence.
November 2013
2-22
DRAFT: Do Not Cite or Quote
-------
o.i
QC-TILDASvsCL
3 10 :
1
0.1 J
= 1.0011x-0.1182
R2 = 0.9924
1 10
Chemiluminescence NOZ ppb
Source: NCEA, using data from Lee et al. (2013)
Figure 2-5 Comparison of NO2 measured by QC-TILDAS (Quantum Cascade-
Tunable Infrared Differential Absorption Spectroscopy) to
measured by Chemiluminescence with photolytic converter.
i
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Villenaetal. (2011) describe the development of a long path absorption photometer, or
LOPAP, to measure NO2. In this technique, NO2 is sampled in a stripping coil using a
modified Griess-Saltzman reagent with the production of an azodye whose absorption in
the visible is measured by long path photometry. This reaction was the basis for a much
earlier manual method for measuring NO (Saltzman. 1954). Interference, which can be
minimized by additional stripping coils, could be caused by HONO, O3, and PAN. In an
intercomparison with a Chemiluminescence monitor with photolytic converter (CL/PC)
conducted on the 5th floor balcony of a building at the University of Wuppertal in
Germany, very good agreement (mean deviation of 2%) was obtained. Interestingly, in
the entire range of measurements (-0.5 ppb to -40 ppb) the relation between LOPAP and
CL/PC can be characterized by LOPAP (ppb) = 0.984*CL/PC - 0.42 (ppb); but if the
range <6 ppb only is considered, the relation becomes LOPAP (ppb) =
0.998*CL/PC +0.19 (ppb).
Diode laser based cavity ring down spectroscopy (CRDS) has also been used to detect
NO2. Fuchs et al. (2009) developed a portable instrument that relies on NO2 absorption at
404 nm, with 22 ppt detection limit at 1 second (S/N = 2). As opposed to
November 2013
2-23
DRAFT: Do Not Cite or Quote
-------
1 chemiluminescence monitors that measure NO2 indirectly based on direct measurement
2 of NO, NO2 (formed by reaction of NO with excess O3) is directly measured in CRDS.
3 NO is then determined by subtracting NO2 measured in the first cavity from the sum of
4 NO2 and NO (i.e., NOX) measured in the second cavity. The O3 is generated by
5 photolysis of O2 in the Schumann-Runge bands at 185 nm. This conversion should be
6 much more quantitative than relying on the reduction of NO2 and NOZ species with
7 variable efficiency on a Mo converter. Again, it should be noted that the optical methods
8 relying on NO2 absorption at -400 nm described above (i.e., CAPS, CRDS), might be
9 subject to positive interference from absorption by trace components (e.g., glyoxal and
10 methyl glyoxal). However, absorption cross sections for these dicarbonyls are much
11 lower than for NO2 at this wavelength, and in general, concentrations for these
12 potentially interfering species are generally lower than for NO2. Note that it is possible
13 that thermal decomposition of NOZ species, such as PAN, in inlets or their reduction on
14 inlet surfaces or in optical cavities can be a source of NO2 in these or other instruments
15 requiring an inlet.
16 This discussion focuses on current methods and on promising new technologies, but no
17 attempt is made here to cover in detail the development of these methods, or of methods
18 such as wet chemical techniques, which are no longer in use. More detailed discussions
19 of the histories of these methods can be found elsewhere (U.S. EPA. 1996. 1993).
2.4.3 Satellite Measurements of NO2
20 Remote sensing by satellites is an approach that could be especially useful in areas where
21 surface monitors are sparse. The retrieval involves three steps: (1) determining the total
22 NO2 integrated line-of-sight (slant) abundance by spectral fitting of solar backscatter
23 measurements, (2) removing the stratospheric contribution by using data from remote
24 regions where the tropospheric column abundance1 is small, and (3) applying an air mass
25 factor (AMF) for the scattering atmosphere to convert tropospheric slant columns into
26 vertical columns. The retrieval uncertainty is largely determined by steps 1 and 2 over
27 remote regions where there is little tropospheric NO2, and by step 3, over regions of
28 elevated tropospheric NO2 (Boersma et al., 2004; Martin et al., 2002). Retrievals are
29 largely limited to cloud fractions <20%. The current algorithm used to derive the
30 tropospheric column of NO2 is given in Bucselaet al. (2013). This algorithm was used to
31 generate the maps in Figure 2-6 for 2005 to 2007 and in Figure 2-7 for 2010 to 2012
32 showing seasonal average NO2 columns obtained by the Ozone Monitoring Instrument
3 3 (OMI) on the AURA satellite.
1 Column refers to the integrated line-of-sight abundance in a unit cross section, such that its units are molecules/
cm2.
November 2013 2-24 DRAFT: Do Not Cite or Quote
-------
OMI Tropospheric NO2 (1O15 molec. cm
Note: Images shown were constructed by Dr. Lok Lamsal of Universities Space Research Association from data obtained by the
OMI instrument on the AURA satellite (http://aura.gsfc.nasa.gov/instruments/omi.html) using the algorithm described in Bucsela et
al. (2013). Top panel (winter; DJF: December, January, February). Lower panel (summer; JJA: June, July, August).
Figure 2-6 Seasonal average tropospheric column abundances for NO2
(1015 molecules/cm2) derived by OMI for winter (upper panel) and
summer (lower panel), for 2005 to 2007.
November 2013
2-25
DRAFT: Do Not Cite or Quote
-------
OMI Tropospheric NO2 (1Q15 molec. cm2)
10
8
6
Note: Images shown were constructed by Dr. Lok Lamsal of Universities Space Research Association from data obtained by the
OMI instrument on the AURA satellite (http://aura.gsfc.nasa.gov/instruments/omi.html) using the algorithm described in Bucsela et
al. (2013). Top panel (winter; DJF: December, January, February). Lower panel (summer; JJA: June, July, August).
Figure 2-7 Seasonal average tropospheric column abundances for NO2
(1015 molecules/cm2) derived by OMI for winter (upper panel) and
summer (lower panel), for 2010 to 2012.
3
4
5
6
7
8
9
Areas of high column NO2 abundance are found over major source areas during both
periods shown in Figure 2-6 and Figure 2-7. High column abundances are found over
many major urban areas, such as Los Angeles, CA; Houston, TX; Chicago, IL; and New
York City, NY; and over major power plant complexes such as the Four Corners and the
Ohio River Valley. A diffuse area with column abundances above background is found
over the Bakken Shale fields in northwestern North Dakota in winter. However, in
general, the area of very high column abundance of NO2 (shown in red) is smaller in the
2010 to 2012 composite than from 2005 to 2007. The photochemical lifetime of NO2 is
longer in winter than in summer and since NO2 is mainly a near surface pollutant, its
November 2013
2-26
DRAFT: Do Not Cite or Quote
-------
1 concentration will be sensitive to mixing layer heights, which are lower in winter than in
2 summer. As a result, column abundances of NO2 are lower in summer than during winter
3 during both periods shown in Figure 2-6 and Figure 2-7.
4 However, since satellite instruments do not return surface concentrations directly,
5 information on NO2 surface concentrations must be inferred from the column
6 measurements. Lamsal et al. (2008) and Lamsal etal. (2010) used satellite data for
7 column NO2 from OMI combined with results from the GEOS-Chem global scale,
8 chemistry-transport model to derive surface concentrations-to-NO2 columns and by a
9 factor accounting for feedbacks of NO2 on its lifetime calculated by the GEOS-Chem
10 global scale. GEOS-Chem is a chemical transport model to derive surface NO2
11 concentrations (see Figure 2-12 for an example of seasonally averaged surface NO2
12 concentrations derived by this method). Note though, this approach is based on data
13 collected during the daily satellite overpass in early afternoon and thus is applicable only
14 for time of satellite overpass in early afternoon.
15 Over the past decade, satellite measurements have shown appreciable reductions in NOX
16 power plant emissions across the U.S. as a result of emission abatement strategies
17 (Stavrakou et al.. 2008; Kim et al.. 2006b). For instance, Kim et al. (2006b) observed a
18 34% reduction in NOX emission over the Ohio River Valley from 1999-2006 due to such
19 strategies. Based on these results, less than 25% of anthropogenic NOX emissions were
20 expected to originate from power plants in this region. Uncertainty in NOX satellite
21 measurements are impacted by several factors, such as cloud and aerosol properties,
22 surface albedo, stratospheric NOX concentration, and solar zenith angle. Boersmaetal.
23 (2004) estimated an overall uncertainty between 35-60% for satellite-retrieved NOX
24 measurements in urban, polluted regions. Although trends in satellite-retrieved NOX
25 power plant emissions reported by Kim et al. (2006b) are uncertain to some extent,
26 similar reductions were reported by region-wide power plant measurements
27 (e-g-, Continuous Emission Monitoring System observations, CEMS).
2.4.4 Measurements of Total Oxidized Nitrogen Compounds (NOY) in the
Atmosphere
28 Commercially available NOX monitors have been converted to NOY monitors by moving
29 the MoOx converter to interface directly with the sample inlet. Because of losses on inlet
30 surfaces and differences in the efficiency of reduction of NOZ compounds on the heated
31 MoOx substrate, NOX concentrations cannot be considered as a universal surrogate for
32 NOY. However, close to sources of fresh combustion emissions, such as highways during
33 rush hour, most of the NO Y is present as NOX. To the extent that all the major oxidized
November 2013 2-27 DRAFT: Do Not Cite or Quote
-------
1 nitrogen species can be reduced quantitatively to NO, measurements of NOY
2 concentrations should be more reliable than those for NOX concentrations, particularly at
3 typical ambient levels of NO2. However, it is worth reiterating that the direct
4 measurements of NO are still the most reliable of all. Reliable measurements of NOY and
5 NO 2 concentrations, especially at the low concentrations observed in many areas remote
6 from sources, are also crucial for evaluating the performance of three-dimensional,
7 chemical transport models of oxidant and acid production in the atmosphere.
2.4.5 Ambient Sampling Network Design
8 Figure 2-8 shows routinely operating monitoring sites for approximately 500 ambient air
9 oxidized nitrogen measurement sites across the U.S. Four networks are highlighted:
10 (1) regulatory based State and Local Air Monitoring Stations (SLAMS) designed to
11 determine NAAQS compliance, (2) National Core (NCORE) network of approximately
12 70 stations designed to capture area representative multiple-pollutant concentrations that
13 provides routinely measured NOY, (3) the Southeast Aerosol Research Characterization
14 (SEARCH), a privately funded network of 6-10 sites that provides direct measurements
15 of true NO2 as well as NOY and other nitrogen species (oxidized and reduced forms), and
16 (4) Clean Air Status and Trends Network (CASTNET) which provides weekly averaged
17 values of total nitrate (HNO3 and pNO3) in rural locations.
18 With the exception of 4-6 sites in the Southeast Aerosol Research Characterization
19 (SEARCH), direct or true NO2 is not measured routinely (Hansen et al.. 2003). The
20 regulatory networks rely on chemiluminescence difference techniques that provide NO
21 concentration directly and report a calculated NO2 concentration as the difference
22 between NOX concentration and NO concentration as discussed above. Criteria for siting
23 ambient monitors are given in the SLAMS/NAMS/PAMS Network Review Guidance
24 (U.S. EPA. 1998b). NO2 monitors are meant to be representative of several scales:
25 microscale (in close proximity, up to 100 meters from the source), middle (several city
26 blocks, 100 to 500 meters), neighborhood (0.5 to 4 km), and urban (4 to 50 km) (40 CFR
27 Park 58, Appendix D). Micro-scale to neighborhood-scale monitors are used to determine
28 highest concentrations and source impacts, while neighborhood- and urban-scale
29 monitors are used for monitoring population exposures.
30 In recognition that roadway-associated exposures account for a majority of ambient
31 exposures to peak NO2 concentrations, EPA promulgated new minimum monitoring
32 requirements in February of 2010 for state and local air monitoring agencies to install
33 near-road NO2 monitoring stations at locations where peak hourly NO2 concentrations
34 are expected to occur within the near-road environment in larger urban areas. Under these
November 2013 2-28 DRAFT: Do Not Cite or Quote
-------
1 new requirements state and local air agencies will operate one near-road NO2 monitor in
2 any Core Based Statistical Area (CBSA) with a population of 500,000 or more persons,
3 and two near-road NO2 monitors in CBSAs with 2,500,000 or more persons or roadway
4 segments carrying traffic volumes of 250,000 or more vehicles. These monitoring data
5 are intended to represent the highest population exposures that may be occurring in the
6 near-road environment throughout an urban area over the averaging times of interest. The
7 near-road NO2 network is intended to focus monitoring resources on near-road locations
8 where peak, ambient NO2 concentrations are expected to occur as a result of on-road
9 mobile source emissions and to provide a clear means to determine whether the NAAQS
10 is being met within the near-road environment throughout a particular urban area. The
11 network is now being phased in, with the first of three phases scheduled to be operational
12 in January of 2014.
November 2013 2-29 DRAFT: Do Not Cite or Quote
-------
• SLAMS
A CASTNET
D NCORE_
« SEARCH
Source: U.S. EPA (2013p)
Figure 2-8 Map of oxides of nitrogen monitoring sites in the U.S., from four
networks (SLAMS, Castnet, NCore, and SEARCH).
2.5 Ambient Concentrations of Oxides of Nitrogen
1 This section provides a brief overview of ambient concentrations of NO2 and associated
2 oxidized TV compounds in the U.S.; it also provides estimates of background
3 concentrations used to inform risk and policy assessments for the review of the NAAQS.
4 The 2008 ISA for Oxides of Nitrogen summarized NO2 concentrations by explaining that
5 the annual avg NO2 concentrations of-15 ppb reported by the regulatory monitoring
6 networks are well below the level of the current NAAQS (53 ppb), but that the daily
November 2013
2-30
DRAFT: Do Not Cite or Quote
-------
1 maximum 1-h avg concentrations can be greater than 100 ppb in some locations,
2 especially in areas with heavy traffic (U.S. EPA. 2008c).
2.5.1 National Scale Spatial Variability
3 In the 2008 ISA for Oxides of Nitrogen, urban data were analyzed in several
4 Consolidated Metropolitan Statistical Areas (CMSAs), with NO2 measured at all
5 monitoring sites located within MSAs or urbanized areas in the U.S. (U.S. EPA. 2008C).
6 NO 2 concentrations were -15 ppb for averaging periods ranging from a day to a year and
7 average daily 1-h max NO2 concentration was ~30 ppb, about twice as high as the
8 24-h avg. The highest maximum hourly concentration (-200 ppb) found was more than a
9 factor often greater than the overall mean 24-h concentrations. Data on NOZ
10 concentrations were very limited in the 2008 ISA for Oxides of Nitrogen, with HNO3 and
11 HONO concentrations indicating that they were considerably lower than NO2
12 concentrations. HNO3 concentrations in one study ranged from <1 to >10 ppb and
13 HONO concentrations were reported as <1 ppb even under heavily polluted conditions.
14 HNO3 concentrations were highest downwind of an urban center and HONO
15 concentrations were several percent of those of NO2 in traffic (U.S. EPA, 2008c). Field
16 study results indicating much higher NOZ concentrations than NOX concentrations in
17 relatively unpolluted rural air were also described (U.S. EPA. 2008c).
18 Figure 2-9. Figure 2-10. Figure 2-11. Table 2-1. and Table 2-2 present monitoring data
19 for 2009-2011. Note that several mid-sized urban areas do not have data records
20 capturing this three-year period. Table 2-1 and Table 2-2 present data from selected urban
21 areas that are examined in recent epidemiological studies on the health effects of NO2
22 (Chapter 4 and Chapter 5). The highest concentrations are evident in the Northeast
23 Corridor and other urbanized regions, and the lowest concentrations are in sparsely
24 populated regions, notably in the west. These observations are consistent with those
25 described in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
November 2013 2-31 DRAFT: Do Not Cite or Quote
-------
2009 - 2011 annual NO mean ppb
» 0-3
• t-J
O B-12
» 13-21
Source: OAQPS and NCEA analysis of AQS network data.
Figure 2-9 Annual average ambient NO concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors.
November 2013
2-32
DRAFT: Do Not Cite or Quote
-------
Source: OAQPS and NCEA analysis of AQS network data.
Figure 2-10 Annual average ambient NO2 concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors.
November 2013
2-33
DRAFT: Do Not Cite or Quote
-------
2049-2011 annual NO x
• 0-7
B-15
O 16 • 2*
• 25-39
• 40 65
Source: OAQPS and NCEA analysis of AQS network data.
Figure 2-11 Annual average ambient NOx concentrations for 2009-2011 at the
site level for the SLAMS regulatory monitors.
November 2013
2-34
DRAFT: Do Not Cite or Quote
-------
Table 2-1 Summary statistics for 1-hour daily maximum NO2 concentrations based on 295 SLAMS monitoring
sites (ppb).
Year
N
Mean Min
10
25
50
75
90
95
99
Max
Pollutant
NO2
NO2
NO2
NO2
NO2
NO2
NO2
NO2
2009-2011
2009
2010
2011
1st Quarter
2nd Quarter
3rd Quarter
4th Quarter
389,946
130,681
131,308
127,957
95,266
97,947
99,500
97,223
20
20
20
19
23
17
17
22
0.1 1
0.1 1
0.1 1
0.1 1
0.1 1
0.1 1
0.1 1
0.1 1
2
3
2
2
3
2
2
3
4
4
4
4
5
3
3
5
8
9
8
8
11
7
7
10
17
18
17
16
22
14
14
21
29
29
28
28
34
24
23
23
39
40
39
39
43
35
34
41
45
46
45
45
49
41
41
47
57
58
57
57
60
53
54
60
360
215
141
360
215
360
229
137
City
Atlanta3
Atlanta-all"
2009-2011
2009-2011
3,237
3,354
14
15
1 1
1 1
2
2
3
3
4
4
8
9
20
22
36
38
44
44
54
55
77
77
Boston3
Boston-all13
Denver3
Denver-all13
Houston3
Houston-all13
Los Angeles3
Los Angeles-allb
New York3
New York-all13
Seattle3
Seattle-all13
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
2009-2011
7,210
11,150
1,739
1,739
11,932
17,148
14,850
34,266
7,598
14,279
804
1,333
23
20
39
39
20
19
28
30
30
28
11
12
0
0
0
0
0
0
0
0
0
0
3
0
4
1
5
5
1
1
4
4
3
2
4
3
7
3
14
14
3
3
7
8
6
5
5
4
9
5
21
21
5
5
10
11
9
8
6
6
13
10
30
30
10
8
17
19
18
16
8
8
21
18
39
39
17
15
28
30
30
28
10
10
30
28
47
47
29
26
39
40
40
38
13
14
36
38
55
55
36
38
47
49
49
48
19
19
44
43
61
61
45
43
53
54
55
54
23
23
53
52
72
72
55
55
64
67
69
69
30
42
197
197
94
94
82
118
137
137
108
153
54
88
City name only rows meet 75% completeness criteria.
bCity-all rows report data regardless of whether completeness criteria are met.
Source: OAQPS and NCEA analysis of AQS network data.
November 2013
2-35
DRAFT: Do Not Cite or Quote
-------
Table 2-2 Summary statistics for
monitoring sites (ppb).
Pollutant
Year
N
NO2, NO and
Mean
Min
NOX annual
1
5
average concentrations based on 295 SLAMS
10
25
50
75
90
95
99
Max
NO2
NO2
NO2
NO2
NO2
2009-2011
2009
2010
2011
295
295
295
295
9.42
9.75
9.33
9.19
0.25
0.23
0.24
0.19
0.63
0.50
0.51
0.60
1.83
1.82
1.83
2.02
2.71
2.64
2.76
2.68
5.00
5.06
4.92
5.05
8.63
8.65
8.56
8.35
12.95
13.42
12.87
12.67
17.17
18.18
17.14
16.78
20.62
21.51
20.02
20.23
24.56
27.35
24.87
23.73
27.66
30.84
27.70
24.66
NO
NO
NO
NO
NO
2009-2011
2009
2010
2011
295
295
295
295
5.78
6.41
5.42
5.52
0.02
0.02
0.00
0.01
0.04
0.03
0.03
0.03
0.23
0.19
0.20
0.23
0.42
0.44
0.38
0.45
1.34
1.34
1.24
1.27
3.68
3.87
3.51
3.41
8.37
8.85
7.69
8.13
14.06
14.58
13.86
13.51
15.98
18.78
16.12
15.86
30.76
35.93
27.13
27.02
42.35
49.53
37.60
50.94
NOX
NOX
NOX
NOX
NOX
2009-2011
2009
2010
2011
295
295
295
295
15.20
16.15
14.75
14.71
0.27
0.27
0.27
0.20
0.66
0.54
0.52
0.70
2.15
2.05
2.11
2.38
3.3
3.26
3.31
3.45
6.28
6.52
6.36
6.15
12.67
12.39
12.37
11.83
20.81
21.78
20.42
19.72
30.46
33.05
29.18
30.04
34.25
38.22
35.99
33.80
54.29
59.91
50.22
50.46
65.16
74.65
58.29
75.59
Source: OAQPS and NCEA analysis of AQS network data.
November 2013 2-36 DRAFT: Do Not Cite or Quote
-------
1 Because of the short lifetime of NOX due to oxidation from PANs and HNO3, NOX
2 concentrations are highly spatially and temporally variable. Average concentrations range
3 from tens of ppt in remote areas of the globe to tens of ppb in urban cores, i.e., by three
4 orders of magnitude. Because ambient NO2 monitoring data are so sparse across the U.S.,
5 especially in rural areas (see Figure 2-8 for location of monitoring sites), measurement
6 data on NOX concentrations across the continental U.S. are incomplete. The short
7 lifetime of NO2 with respect to conversion to NOZ species and the concentrated nature of
8 NO2 emissions result in steep gradients and low concentrations away from major sources
9 that are not adequately captured by the existing monitoring networks. Satellite data
10 coupled with model simulations might be more useful for showing large-scale features in
11 the distribution of NO2. Winter and summer seasonal average NO2 concentrations for
12 2009-2011 derived from the OMI instrument on the AURA satellite and the GEOS-Chem
13 global, three-dimensional chemistry-transport model are shown in Figure 2-12. In this
14 method, integrated vertical column abundances of NO2 derived from the OMI instrument
15 are scaled to surface mixing ratios using scaling factors derived from GEOS-Chem [see
16 Lamsal et al. (2010): Lamsal et al. (2008): also see Section 2.4 for more complete
17 descriptions of the method]. A nested version of GEOS-Chem at 50 km x 50 km
18 horizontal resolution is used in this method. A description of the capabilities of GEOS-
19 Chem and other three-dimensional CTMs is given in the O3 ISA (U.S. EPA. 2013b).
20 Large variability in NO2 concentrations is apparent in Figure 2-12. As expected, the
21 highest NO2 concentrations are seen in large urban regions, such as the Northeast
22 Corridor, and lowest values are found in sparsely populated regions located mainly in the
23 West. As can be seen, minimum hourly values are of the order of ~10 ppt, leading to a
24 range between maximum and minimum concentrations of over a factor of a thousand.
25 NO2 concentrations tend to be higher in January than in July reflecting lower planetary
26 boundary layer heights in winter and more widespread emissions from residential heating
27 during winter. Such seasonal variability is also evident on a local scale, as measured by
28 surface monitors. For example, in Atlanta, GA, NOX measurements also exhibited higher
29 concentrations in winter and lower concentrations in summer, when NOX is more rapidly
30 removed by photochemical reactions (Pachon et al., 2012).
November 2013 2-37 DRAFT: Do Not Cite or Quote
-------
OMI-derived surface NO2 (ppb)
I
I
7.00
6.22
5.44
4.67
3.89
3.11
2.34
1.56
0.78
0.01
Note: Images shown were constructed by Dr. Lok Lamsal of Universities Space Research Association from data obtained by the
OMI instrument on the AURA satellite (http://aura.asfc.nasa.gov/instruments/omi.html) using the algorithm described in Bucsela et
al. (2013). Output from the GEOS-Chem, global-scale, three- dimensional, chemistry-transport model to derive surface
concentration fields from the satellite data as described in Lamsal et al. (2008) and Lamsal et al. (2010).
Top panel (winter; DJF: December, January, February). Lower panel (summer; JJA: June, July, August).
Figure 2-12 Seasonal average surface NO2 concentrations in ppb for winter
(upper panel) and summer (lower panel) derived by OMI/GEOS-
Chem, for 2009-2011.
2.5.2 Urban Scale Spatial Variability
1 In the past decade considerable urban scale NOX spatial variability was observed in
2 several studies (Monn. 2001; Fischer et al., 2000; Kingham et al., 2000; Lebret et al.,
3 2000). More recently, spatial variability has been further characterized, especially
4 through determining factors influencing NO2 spatial variations and refining methods for
5 estimating urban scale concentrations.
November 2013
2-38
DRAFT: Do Not Cite or Quote
-------
1 This high spatial variability makes it impractical to rely solely on measurements to obtain
2 urban scale concentrations. Instead, interpolation and modeling methods are often used,
3 and these were recently reviewed (Briggs. 2005; Jerrett and Finkelstein, 2005). Of these
4 methods, land use regression has emerged as a widely used tool for estimating urban
5 scale air pollutant concentrations, particularly for NO2. Land-use regression combines
6 monitoring of air pollution at a small number of locations and development of stochastic
7 models using predictor variables usually obtained through geographic information
8 systems (GIS). A critical review of 25 recent applications of land use regression, many of
9 them specifically targeting NO2, concluded that land use regression models have
10 generally been applied successfully in a variety of North American and European cities,
11 and that its performance in predicting measured concentrations in urban areas is typically
12 better or equivalent to geostatistical methods such as kriging and dispersion models
13 (Hoek et al.. 2008).
14 Jerrett et al. (2007) pioneered the use of land use regression modeling in North America
15 by applying it to NO2 in Toronto. They noted elevated NO2 concentrations within
16 1,500 meters of roadways and concluded that small area variations probably due to traffic
17 were captured by land use regression methods. In a highly industrialized urban area in
18 Sarnia, Ontario, land use regression methods were used to determine that the factors
19 responsible for elevated concentrations included proximity to the region's industrial core,
20 placement within 1,600 meters of industrial areas, placement 400 meters from highways,
21 and dwelling counts within 2,400 meters (Atari et al.. 2008).
22 Hart et al. (2009) developed generalized additive models to spatially model NO2
23 concentrations in the continental U.S., and concluded that distance to road, population
24 density, elevation, land use, and distance to emissions of the nearest oxide of nitrogen-
25 emitting power plant were all statistically significant predictors of measured NO2.
26 Applications of land use regression for assessing exposure to NO2 are described in detail
27 in Section 2.6.2.3.
2.5.3 Micro-to-neighborhood Scale Spatial Variability, Including Near Roads
28 Complex spatial gradients in NO, NO2, and NOX concentrations have been observed in
29 near road environments. Several factors that impact near-road NOX concentrations, such
30 as distance from roadway, traffic volume, and season, were discussed in the 2008 ISA for
31 Oxides of Nitrogen (U.S. EPA. 2008c). These studies reported a sharp decline in NOX
32 concentration within 350 meters downwind of roadways (Gilbert et al.. 2007; Singer et
33 al.. 2004) and several meters above street canyons (Restrepo et al.. 2004). Another study
34 by Monn (2001) reported the influence of meteorology and enhanced photochemical
November 2013 2-39 DRAFT: Do Not Cite or Quote
-------
1 activity in the spring and summer on NO2 spatial gradients was also discussed (Monn.
2 2001). This section discusses the characteristics and factors impacting near-road spatial
3 gradients in NOX concentrations.
4 Concentrations of NO, NO2, and NOX measured on or near roads have varied
5 substantially depending on measurement location, fleet mix, and time of day. For
6 example, Zhu et al. (2008) used a mobile monitoring station to measure NOX on two
7 freeways in Los Angeles, CA: 1-405, which predominantly carries automobile traffic, and
8 1-710, which carries the majority of diesel truck traffic for the area (-25,000 trucks per
9 day) (Fruinetal.. 2008). On 1-405, average NOX concentrations of unfiltered air
10 measured 267 ppb with a maximum concentration around 850 ppb, while on 1-710
11 unfiltered NOX concentrations averaged 432 ppb with a maximum concentration around
12 950 ppb. Similarly, Fruin et al. (2008) measured on-road NO concentrations on three Los
13 Angeles highways and five arterial roads. They observed average NO concentrations
14 ranging from 170-390 ppb on the freeways, with the highest average concentration
15 corresponding to the highest diesel truck traffic on 1-710. On-road concentrations on
16 arterial roads averaged 17-79 ppb. Baldauf etal. (2008a) presented time-series of near
17 road pollutants measured 5 meters from 1-40 in Raleigh, NC, and reported that NO
18 concentrations reached near 250 ppb between 8:00 a.m. and 9:00 a.m., with minimum
19 NO concentrations around 50 ppb during that time period. The rush hour period spanned
20 6:00 a.m. - 9:00 a.m., and exhibited a distinct peak during that time period. During the
21 evening rush hour between 4:00 p.m. and 7:00 p.m., NO concentrations fluctuated
22 between 20-150 ppb. Clements et al. (2009) measured concentrations of NO, NO2, and
23 NOX, 5 meters downwind from a state road in Austin, TX, and observed NOX
24 concentrations of approximately 40-50 ppb, NO concentrations of approximately 15-40
25 ppb, and NO2 concentrations of approximately 5-15 ppb under downwind conditions.
26 Taken together, these results suggest that NO2 represented 10-38% of freshly emitted
27 NOX.
28 More recent studies show similar spatial gradients in NO, NO2, and NOX concentrations
29 near roadways and also provide more fine-scale temporal and spatial information. In a
30 pilot study to better understand issues involved in meeting new monitoring requirements
31 for the 1-hour NO2 NAAQS promulgated in 2010, state and local air agencies collected
32 NO2 and NOX concentration data with passive sampling devices (PSDs) near heavily
33 trafficked roads within five Core Based Statistical Areas (CBSAs): Albuquerque, New
34 Mexico; Baltimore, Maryland; Boise, Idaho; and Miami-Broward County and Tampa-
35 Hillsborough County in Florida (STi. 2011). The study confirmed that near-road NO2
36 concentrations were generally highest at locations nearest the roadway and near those
37 roads with the highest daily traffic. Deviations from this pattern could be explained by
38 roadway configuration or other considerations (e.g., higher NO2 observed because of
November 2013 2-40 DRAFT: Do Not Cite or Quote
-------
1 accelerating truck traffic on an on-ramp, site placement on a toll booth near a port, tunnel
2 entrance/exit, and rail activities). Horizontal concentration gradients were relatively
3 gradual, with concentrations decreasing slightly from 7 to 45 meters. Vertical
4 concentration gradients were also evaluated, and the highest concentrations were
5 typically at the sampling height closest to the roadway (typically closest to ground level),
6 although concentration differences were relatively small.
7 Additional studies show that NO, NO2, and NOX concentrations decrease exponentially
8 with greater horizontal distance from the roadway. Average concentrations near the
9 roadway are typically 30% to 200% of urban background concentrations, and average
10 concentrations fall to background levels within 100-500 meters of the roadway (Polidori
11 and Fine. 2012; Karner etal. 2010; Beckerman et al.. 2008; Zhou and Levy. 2007). This
12 behavior is illustrated in Figure 2-13. which describes measurements from two
13 monitoring stations in southern California, located 15 meters and 80 meters east and
14 downwind of the 1-710 freeway ("near" and "far" site, respectively), nearthe intersection
15 with North Long Beach Boulevard. It also includes results from a monitoring site far
16 from the influence of the 1-710 and representative of background conditions was operated
17 in Carson, CA, next to Del Amo Elementary School (labeled 'Del Amo' in Figure 2-13).
November 2013 2-41 DRAFT: Do Not Cite or Quote
-------
WINTER
SUMMER
1 2
c.
O 1
u
0
300
g; 200
O
0
OA
"I 6°
O
~ 4°
* 20
0
~ 300
c.
3- 200
X
O 100
;z
0
*
T(X58 T0
+ -*-
T71.0 T
1 V
1 4t
T37.2 • -
1 T 3
-T--^
•
T108 T
\&(
4- *—r~^
52 *
T 0.29
' — • — '
!•
.H
T Tg 9
1 . 3"
2.3
^ "'I
4-
.7
T 4S 1
j — 1 — i
*
V8 °^3 0,19
"
29.6
I T i 8.14 4.04
~~U^^J * '
J28.0
[— *— j ^. 10.9
* NF=I
• ^
T 50.4
4. 26.U |^n
_ -*" ^^
SCAB Average
-i Winter: 0.36 ppm
Summer: 0.06 ppm
SCAB Average
Winter: 2 1.4 ppb
Summer: 4.j6 ppb
SCAB Average
Winter. 20.5 ppb
Summer: 15.3 ppb
SCAB Average
Winter: 4 1.8 ppb
Summer: 19.2 nnb
-------
1 While levels of all NO, NO2, and NOX concentrations decrease from the roadway,
2 different species have different decay profiles which are influenced by chemistry and
3 atmospheric factors. NO typically decays more rapidly than NO2 or CO, typically
4 reaching background levels within the first 100 meters of the roadway. This trend
5 suggests that chemical processing by O3 as well as dilution is an important factor in
6 determining NO concentrations in near road environments (Clements et al.. 2009). This
7 titration mechanism likely explains the more pronounced spatial gradient observed in NO
8 during the summer when O3 is prevalent due to enhanced photochemical activity
9 (Polidori and Fine. 2012; Monn. 2001). This is consistent with results by Polidori and
10 Fine (2012) (from a near-road field campaign in Los Angeles, CA) showing that roadside
11 NO was 7 times higher than the background during the summer compared to 3 times
12 higher during the winter.
13 The decrease in NO2 concentrations from the road is more gradual than that of most other
14 traffic related air pollutants (including NO) (Gordon et al.. 2012; Beckerman et al.. 2008).
15 or does not even decrease with distance (Massoli et al.. 2012). Quantitative reviews of
16 near-roadway pollutant impacts demonstrate that the NO2 spatial gradient extends further
17 from the roadway (200-500 meters) than pollutants rapidly removed by chemical reaction
18 (Karneretal. 2010; Zhou and Levy. 2007). Massoli et al. (2012) described another
19 characteristic of NO2 concentrations near roads, an abrupt increase after sunrise due to
20 conversion of abundant NO near roads to quickly reach a NO2 concentration "ceiling"
21 that is limited by background O3 levels. They concluded that over short temporal scales
22 (shorter than the time scale for replenishment of background O3) less than 500 meters
23 from a source, NO2 alone was not a good indicator of traffic related emissions.
24 Production of other traffic related pollutants is not limited by available background O3
25 concentrations.
26 In addition, near road profiles of NOX concentration are largely influenced by
27 meteorological parameters and atmospheric stability. Wind direction dictates the spatial
28 profile of NO, NO2, and NOX concentrations (as well as other traffic pollutants) with
29 more gradual gradients on the downwind side of the roadway (Durant et al.. 2010;
30 Clements et al.. 2009; Hu et al.. 2009; Beckerman et al.. 2008). Wind speed and
31 atmospheric stability also impact roadway NOX concentrations. Peak roadway
32 concentrations are often observed during pre-sunrise hours when winds are weak and
33 atmospheric inversions are present (Gordon etal.. 2012; Durant etal.. 2010; Hu et al..
34 2009). During these pre-sunrise hours, the spatial impact of roadway NOX concentrations
35 may also extend, resulting in a more gradual decay from the roadway. Hu et al. (2009)
36 observed this effect during a near-road field campaign in Santa Monica, CA. They
November 2013 2-43 DRAFT: Do Not Cite or Quote
-------
1 observed elevated NO concentrations (90-160 ppb) as far as 1,200 meters downwind of
2 the roadway during pre-sunrise hours, which is much larger than the expected spatial
3 extent of NO (100-300 meters) (Karner et al.. 2010; Zhou and Levy. 2007). NOX
4 concentration gradients continue to change throughout the day as atmospheric stability
5 evolves. After sunrise, near-road NOX concentrations drop as vertical mixing increases
6 (Gordon et al.. 2012; Durant et al.. 2010) until concentrations reach near-background
7 levels during the late afternoon (Gordon et al.. 2012). In some studies, no clear gradient is
8 observed in NOX concentrations (or other traffic related species) during mid-morning or
9 early evening hours because near-roadway concentrations are as low as background
10 levels due to dilution (Gordon et al.. 2012; Durant et al.. 2010). However, the exact
11 response of the concentration gradient to boundary layer expansion is unresolved to some
12 extent.
13 Dispersion of NOX in the near road environment is influenced by several factors:
14 atmospheric turbulence, vehicle-induced turbulence, and roadway-induced turbulence
15 (Baldauf et al.. 2009; Wang and Zhang. 2009). Atmospheric turbulence occurs as a result
16 of meteorological factors within the urban boundary layer. Vehicle-induced turbulence
17 results from the air disturbances caused by the direction and speed of vehicle motion.
18 Roadway-induced turbulence happens when wind-driven air masses undergo separation
19 following impact with a roadway structure in the built environment. These sources of
20 turbulence interact with each other to create complex, unique dispersion profiles at a
21 given road segment to influence NOX concentrations. This discussion addresses the
22 physical factors influencing dispersion of NOX.
23 Several atmospheric conditions affect regional or urban airflow profiles and potentially
24 may impact the dispersion profile of NOX even in the absence of adjacent buildings,
25 roadway structures, or traffic-related turbulence. In urban areas, effects of the built
26 environment can be seen at regional, urban, neighborhood, and street-level scales
27 (Fernando. 2010; Britter and Hanna. 2003). Roughness created by upstream buildings
28 contributes to local turbulence levels, even in the absence of adjacent buildings. Land
29 forms such as slopes and valleys can also affect the atmospheric turbulence level, because
30 they interact with atmospheric stability conditions to restrict air movement. Finn et al.
31 (2010) observed that tracer gas concentration increased with increasing atmospheric
32 stability. This finding is consistent with results with other studies (Gordon et al.. 2012;
33 Durant etal. 2010; Hu et al.. 2009) that observed the highest concentrations of NO, NO2,
34 and NOX concentrations before sunrise when traffic levels and atmospheric stability are
35 high. Hu et al. (2009) also argued that atmospheric stability potentially extends the decay
36 profile of near roadway pollutants. Under stable atmospheric conditions, Hu et al. (2009)
37 reported elevated NO extended as far as 1,200 meters downwind of a roadway, which is
38 much further than the expected spatial extent of NO in the daytime with lower
November 2013 2-44 DRAFT: Do Not Cite or Quote
-------
1 atmospheric stability (Karneretal.. 2010). Additionally, the presence of slopes and
2 valleys can cause spots where airflow converges or diverges (Fernando. 2010). Heat flux
3 can be sizeable in urban areas where the "heat island" effect from roadways and buildings
4 can raise local temperatures by several degrees (Britter and Hanna. 2003): heat flux
5 potentially contributes to convection near roadways and other structures in the built
6 environment. Underscoring the dominant role of local turbulence on dispersion patterns,
7 Venkatram et al. (2007) measured meteorological factors potentially affecting NO
8 concentrations near a road segment in Raleigh, NC and found that, among meteorological
9 variables, vertical velocity fluctuations had the largest effect on NO concentration.
10 Vehicle motion creating high levels of turbulence on and near roads can contribute to the
11 dispersion of traffic-related air pollution in the vicinity of a roadway (Baldauf et al.,
12 2008a). An early description of this was provided by Sedefian et al. (1981) for the
13 General Motors experiments, in which groups of vehicles were driven along a test track
14 while towers with mounted anemometers measured mean and fluctuating velocities. It
15 was observed that vehicle-induced turbulence dissipates slowly under low mean wind
16 conditions and vice versa. Vehicle-induced turbulence was found in that study to
17 contribute to vertical dispersion of emitted pollutants. Computational fluid dynamics
18 (CFD) simulations by Wang and Zhang (2009) also found that vehicle-induced
19 turbulence contributed to vertical dispersion. Rao et al. (2002) also observed large
20 measurements of turbulence kinetic energy in the wake of a vehicle outfitted with a trailer
21 carrying sonic anemometers driving along a runway. Sedefian et al. (1981) also found
22 that advection of vehicle-induced turbulence away from the roadway was related to the
23 speed and direction of mean winds, di Sabatino et al. (2003) showed that vehicle-induced
24 turbulence is related to traffic levels. In light traffic, the wake behind a vehicle is isolated,
25 but for increasing traffic, the wakes interact and turbulence is a function of the number of
26 vehicles and vehicle length scale. At congested traffic levels, the vehicle-induced
27 turbulence becomes independent of the number of vehicles. For street canyon simulations
28 and measurements, Kastner-Klein et al. (2003) observed that predictions of tracer
29 concentrations were overestimated when vehicle-induced turbulence was not considered;
30 this implies additional dispersion related to vehicle-induced turbulence. Traffic
31 directionality was investigated by He and Dhaniyala (2011) and Kastner-Klein et al.
32 (2001). He and Dhaniyala (2011) observed that turbulence kinetic energy from two-way
33 traffic was roughly 20% higher than for one-way traffic, and they found that the
34 turbulence kinetic energy increased with decreasing distance between the traffic lanes.
35 Kastner-Klein et al. (2001) observed that two-way traffic suppresses the mean flow of
36 vehicle-induced air motion along a street canyon, whereas one-way traffic produces a
37 piston-like effect (note that the Kastner-Klein et al. (2001) study was for the geometrical
38 case of a street canyon). Substantially higher turbulence levels were produced with two-
November 2013 2-45 DRAFT: Do Not Cite or Quote
-------
1 way traffic compared with one-way traffic for the Kastner-Klein et al. (2001) study as
2 well.
3 The presence of near-road structures results in recirculating airflow regions that may trap
4 air pollutants on one side and disperse them on another side, depending on wind
5 conditions (Baldauf et al.. 2008b). Finn etal. (2010) simulated transport from a roadway
6 using a point source tracer gas with barrier and open terrain conditions. With airflow
7 from the simulated roadway and high atmospheric stability, high concentrations were
8 trapped in the roadway region with a negligible tracer gas in the wake downstream of the
9 barrier with considerable lateral and vertical plume dispersion. For open terrain, transport
10 of the tracer was characterized by a narrow plume. Hagler et al. (2011) used CFD to
11 model airflow and concentrations around barriers of different heights and similarly found
12 reductions in inert tracer concentration downwind of the barrier compared with the open
13 terrain case with trapping of air pollutants upstream of the barrier. With the barrier in
14 place, downwind tracer concentrations were observed at elevations of twice the barrier
15 height. Mean airflow vectors also illustrate a wind disturbance at elevations of twice the
16 barrier height. Even for the open terrain case, Gaussian dispersion caused vertical
17 dispersion with the plume spread. In additional simulations involving a service readjust
18 downstream of the barrier, Hagler etal. (2011) observed entrainment of tracer in the
19 wake downstream of the barrier. Tokairin and Kitada (2005) used CFD to investigate the
20 effect of porous fences on contaminant transport near roads and observed tracer gas
21 retention and airflow recirculation when the fences were designed with less than 40-50%
22 porosity. Heist et al. (2009b) investigated the effect of geometry of road cuts and noise
23 barriers in wind tunnel tracer gas experiments. They observed that elevated roadways,
24 depressed roadways, and noise barriers all resulted in lower downwind concentrations
25 compared with the open terrain case with elevated roadways producing the least
26 reduction in concentration. As in Hagler et al. (2011). Heist et al. (2009b) observed
27 measurable concentrations at elevations that resulted from Gaussian dispersion for all
28 geometries of the road cut or barrier, but vertical dispersion was enhanced or dampened
29 depending on the specific geometry. Similarly, for wind tunnel simulations of a single
30 tower above a matrix of street canyons, the tower was shown to induce both airflow and
31 tracer concentration along the leeward edge of the building to a height exceeding the
32 tower height (Brixev et al.. 2009; Heist et al.. 2009a).
33 For the special case of street canyons, retention time for traffic-based pollution increases
34 on the roadway with increasing building height-to-road width ratio because recirculating
35 airflow forms closed streamlines within the canyon (Li etal.. 2005; Liu et al.. 2005). For
36 wind tunnel simulations of tracer emission at street level with and without traffic,
37 Kastner-Klein et al. (2001) observed measurable tracer concentrations near the top of the
38 street canyon but with some dispersion from maximum tracer levels at the canyon floor.
November 2013 2-46 DRAFT: Do Not Cite or Quote
-------
1 Dilution of NOX concentrations through these recirculating air structures leads to a steep
2 decrease in concentration with increasing distance from the ground (Lee et al.. 2012a).
3 For low aspect ratio street canyons, secondary recirculating structures can arise; while
4 contaminant retention still occurs in this case, ventilation occurs more readily than for the
5 high aspect ratio case (Simoens and Wallace. 2008; Simoens et al., 2007). Cheng et al.
6 (2008) used CFD to evaluate factors leading to contaminant retention in street canyons
7 and observed that the exchange rate for air and a tracer gas was driven by the turbulent
8 component of airflow at the roof-level interface of the street canyon. Subsequent
9 simulations showed that exchange rate was also aided by unstable atmospheric conditions
10 (Cheng et al.. 2009b). CFD simulations by Gu et al. (2010) of transport within a street
11 canyon with and without vegetation suggested that the recirculating flow is dampened by
12 the presence of vegetation.
2.5.4 Seasonal, Weekday/Weekend and Diurnal Trends
13 Month-to-month variability in 24-h avg NO2 concentrations was described in the 2008
14 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). Strong seasonal variability in NO2 was
15 reported with higher concentrations in winter and lower concentrations in summer.
16 Monthly maxima varied regionally. Day-to-day variability in NO2 concentration was
17 generally larger during the winter. Differences between weekdays and weekends were
18 also noted, with lower concentrations and a more compressed cycle on weekends, with
19 more pronounced differences at sites more influenced by traffic. Observations of lower
20 NO2 concentrations on weekends are consistent with other recent results. Summer
21 satellite data indicated higher concentrations on weekdays than on weekends regardless
22 of land coverage, for urban, forest, and other regions (Choi et al.. 2012). In southern
23 California, NOX concentrations were an average of 46% lower in ground-based
24 measurements, and 34% lower in airborne measurements (Pollack et al.. 2012). In
25 Atlanta, NOX concentrations were 24% higher on weekdays than on weekends (Pachon et
26 al..2Q12).
27 Recent data presented in Table 2-1 continue to show similar seasonal trends for average
28 seasonal concentrations across 3 years. Mean concentrations are highest in the first and
29 fourth quarters, but maximum concentrations are highest in the second and third quarters.
30 Recent data also indicate concentration patterns of NO and NO2 are affected strongly by
31 emissions and meteorology as concentrations peak during early morning hours and
32 seasonally (higher in winter) when planetary boundary layer heights are lowest (Figure
33 2-14). NO2 exhibits flatter profiles relative to NO as secondary formation processes
34 influence concentration patterns.
November 2013 2-47 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
Figure 2-15 shows a typical diurnal cycle for NO and NO2. As described in the 2008 ISA
for Oxides of Nitrogen (U.S. EPA. 2008c). the NO2 typically exhibits a daily maximum
during morning rush hour, although it can occur at other times of day. This pattern is
shown for Atlanta, GA, in Figure 2-15. but it is also typical for other urban sites.
- January 130890002 NO
.January 130890002 N 02
•July 130890002 NO
July 130890002 N 02
12345678 9101112131415161718192021222324
Source: EPA Analysis of AQS Data.
Figure 2-14 January and July hourly profiles of NO and NO2 (ppb) for Atlanta,
Georgia (site with maximum NO2 levels).
•130890002 HO Weekday
•130890002 HO Weekend
130890002 H02 Weekday
•130890002 H02 Weekend
12345678 9101112131415161718192021222324
Source: EPA Analysis of AQS Data.
Figure 2-15 Weekend/Weekday hourly profiles of NO and NO2 (ppb) for
Atlanta, Georgia (site with maximum NO2 levels).
November 2013
2-48
DRAFT: Do Not Cite or Quote
-------
1 Differences between weekdays and weekends are also observed. Typically, weekday
2 levels of NOX, particularly NO, exceed weekend levels and diurnal cycles are more
3 compressed on weekends. The weekend effect for NO was first observed by Cleveland et
4 al. (1974) and it is a general characteristic of urban NO and NOX levels observed in many
5 locations (Tonse et al., 2008; Pun etal., 2003; Marr and Harley. 2002). The lower
6 concentrations of NOX on weekends can have a profound but complex effect on urban
7 atmospheric chemistry. The lower concentrations of NOX on weekends can also
8 substantially increase or decrease O3 concentrations, depending on whether O3
9 production is limited by VOC or NOX concentrations (Tonse et al., 2008; Heuss et al.,
10 2003). It can also lead to differences the composition of NOZ between weekdays and
11 weekends. For example, Pollack etal. (2012) observed greater production of HNO3, a
12 radical termination product, on weekdays and greater production of PAN, a VOC-NOX
13 oxidation product, on weekends in southern California.
2.5.5 Multi-year Trends in Oxides of Nitrogen
14 The annual average NO2 concentration across the U.S. decreased by 48% from 1990 to
15 2012, as shown in Figure 2-16. Information on trends on a regional basis and at
16 individual, local air monitoring sites can be found at
17 http://www.epa.gov/air/airtrends/nitrogen.html; National Trends in Nitrogen Dioxide
18 Levels. The steady decline in NO2 concentrations over the years can be attributed mainly
19 to reductions in emissions from mobile and stationary sources (cf Figure 2-3).
20 In Atlanta, NOX concentrations decreased from 1999 to 2001, increased during 2002 and
21 2003, and decreased again until 2007. The decrease from 1999 to 2001 was attributed to
22 the implementation of EPA's acid rain program and the decrease from 2002 to 2007 to
23 decreases in on-road NOX emissions (Faction et al.. 2012).
November 2013 2-49 DRAFT: Do Not Cite or Quote
-------
NO2 Air Quality, 1990 - 2012
(Annual Arithmetic Average)
National Trend based on 135 Sites
70
60 H
.Q
CL
°-50l
s
£ 40 H
ro
c 30
£ 20
National Standard
U
10
T
T
I I
T
T
I I
T
T
T
T
T
1111111111
9999999999
9999999999
01 23456789
2222222222222
000
1 1 1
0000000000
0000000000
01 2345678901
1990 to 2012 : 48% decrease in National Average
Source: U.S. EPA (2013)
Figure 2-16 National annual average ambient NO2 concentration trends,
1990-2012.
2.5.6 Background Concentrations
1 In the context of a review of the NAAQS, EPA generally defines "background
2 concentrations" in a way that distinguishes between concentrations that result from
3 precursor emissions that are relatively less controllable from those that are relatively
4 more controllable through U.S. policies or through international agreements. The most
5 commonly used form in the past is North American Background (NAB), which refers to
6 simulated NO2 concentrations that would exist in the absence of anthropogenic emissions
7 from the U.S., Canada, and Mexico. This definition of background includes contributions
November 2013
2-50
DRAFT: Do Not Cite or Quote
-------
1 resulting from emissions from natural sources (e.g., soils, wildfires, lightning) around the
2 world. Other definitions can also be used. For example in the 2013 ISA for Ozone (U.S.
3 EPA, 2013b). a U.S. background, which includes emissions from Canada and Mexico in
4 addition to those in the definition of a North American Background, and a natural
5 background, which includes only emissions from natural sources globally, were used.
6 Background is used to inform policy considerations regarding the current or potential
7 alternative standards.
8 As can be seen from Figure 2-12. maximum concentrations of NO2 occur along the
9 Northeast Corridor, the Ohio River Valley and in the Los Angeles basin. While NO2
10 concentrations are often above 5 ppb, background is less than 300 ppt over most of the
11 continental U.S., and less than 100 ppt in the eastern U.S., as shown in the 2008 ISA for
12 Oxides of Nitrogen (U.S. EPA. 2008c). The distribution of background concentrations
13 largely reflects the distribution of soil NO emissions, with some local enhancements due
14 to biomass burning, mainly in the western U.S. In the northeastern U.S., where present-
15 day NO2 concentrations are highest, NAB contributes <1% to the total.
16 In addition to U.S. and other North American sources, various NOY species from sources
17 outside North America have long enough residence times in the atmosphere so they can
18 be transported to the U.S. Figure 2-17 shows monthly average contributions for April
19 2010 to concentrations of PAN, and NOX, in addition to CO and O3, resulting from the
20 transport of Asian anthropogenic emissions. The contribution from Asian anthropogenic
21 emissions was estimated by taking the difference between two model simulations, one
22 with emissions from all continents globally (see Figure 2-18) and the other omitting
23 Asian emissions. The results from the second simulation were subtracted from the first
24 and the contribution from Asia was taken to be the difference between the two
25 simulations. As opposed to simply zeroing out anthropogenic emissions outside of Asia,
26 this approach allows for the effects of global scale chemistry to be felt on emissions from
27 Asia. These simulations were carried out using the AM3, global scale, three-dimensional
28 chemical tracer model described by Lin et al. (2012). As noted in the O3 ISA (U.S. EPA,
29 2013b). spring is the dominant season for effects of intercontinental transport of pollution
30 to be detected in the United States. Results for April are shown because it is the month in
31 which maximum springtime contributions are found. As can be seen from Figure 2-17.
32 transported PAN concentrations over the western U.S. can be larger than those for NOX
33 by a factor often or more. These values refer to the 800 hPa level and are likely to be
34 lower at the surface because of dilution and deposition. Corresponding model
35 calculations of total monthly average concentrations for the sum of PAN and NOX
36 concentrations over the western U.S. are less than ~1 ppb (see Figure 2-18). These values
37 are broadly consistent with those given in Figure 2-12. which shows seasonal average
38 NO2 concentrations typically less than 1 ppb across broad areas of the U.S., and with
November 2013 2-51 DRAFT: Do Not Cite or Quote
-------
1
2
3
modeling results and data from field studies given in the 2008 ISA for Oxides of
Nitrogen (U.S. EPA. 2008c). All of these results indicate that background levels of NO2
are well beneath the level of the current NO2 NAAQS.
8 12 16 20 24
Note: The contribution from Asian anthropogenic emissions was estimated by taking the difference between two model simulations
using the modeling approaches described in Lin et al. (2012). One simulation included emissions from all continents globally (see
Figure 2-18) and the other omitted Asian emissions. The results from the second simulation were subtracted from the first and the
contribution from Asia was taken to be the difference between the two simulations.
Figure 2-17 Simulated Asian contributions to CO, Oz, PAN, and NOx
concentrations at 800 hPa in April, 2010.
November 2013
2-52
DRAFT: Do Not Cite or Quote
-------
[ppbvj 03
[ppbv]
70 90 110 130 150 170 190 210 40 50
PAN [Pptv] NOx
20 60 100 140 180 220 260 300
45
65
85
105
Figure 2-18 Simulated total CO, O3, PAN, and NOX concentrations at 800 hPa
in April, 2010.
2.6 Exposure Assessment
2.6.1 Conceptual Model
1 Personal exposure to an ambient air pollutant, such as NO2, is the concentration of the air
2 pollutant encountered by an individual over a given time. In addition to time-activity (i.e.,
3 time spent in different microenvironments), personal exposure to ambient NO2 is
4 associated with climate (including weather and season), housing characteristics (e.g.,
5 window openings, draftiness, air conditioning), and microenvironmental sources (e.g.,
6 roadways, construction equipment, indoor sources).
November 2013
2-53
DRAFT: Do Not Cite or Quote
-------
1 Total personal exposure, ET, integrates the product of microenvironmental concentration,
2 C, and fraction of time spent in a microenvironment across an individual's
3 microenvironmental exposures, t:
ET =
Equation 2-1
4 where Q = average NO2 concentration in the rth microenvironment, t; = fraction of total
5 time spent in the rth microenvironment, and n = total number of microenvironments
6 which the individual has encountered (U.S. EPA. 2008c; Klepeis et al., 2001). Hence,
7 both the microenvironmental NO2 concentration and time activity aspects of total
8 exposure must be considered. Note that if NOX concentration is measured, then it can be
9 used in lieu of NO2 concentration.
10 Alternatively, based on the principle of mass balance, an individual's total NO2 exposure
1 1 can be expressed as the sum of its ambient NO2 exposure, Ea, and non-ambient NO2
12 exposure, Ena, components (U.S. EPA. 2008c; Wilson and Brauer. 2006):
Equation 2-2
13 Ea represents the amount of NO2 exposure derived from outdoor sources, and Ena
14 represents the amount of NO2 exposure from indoor sources. The microenvironmental
15 formulation presented in Equation 2-1 and the component formulation presented in
16 Equation 2-2 can be rectified by recognizing that Ea and Ena can both be expressed in
17 terms of microenvironmental concentrations and time spent in outdoor and indoor
18 microenvironments. During the fraction of a day spent outdoors, y0, it is often presumed
19 that an individual is exposed to the ambient concentration of NO2.
November 2013 2-54 DRAFT: Do Not Cite or Quote
-------
1 Outdoor microenvironmental NO2 exposures, E0, can then be expressed simply as the
2 product of the fraction of the time spent outdoors, y0 and ambient NO2 concentration, Ca:
= y0ca
Equation 2-3
3 Indoor NO2 exposures in the ith microenvironment, El5 are more complicated, because
4 some part of indoor exposure emanates from nonambient sources, and some part of
5 indoor exposure infiltrates from outdoors. Indoor exposures from nonambient sources are
6 already given as Ena. Indoor exposures from ambient sources are also influenced by
7 infiltration of outdoor NO2, INF, time spent indoors, yi; and Ca:
Ca + Ena
Equation 2-4
8 Infiltration is a function of the ith microenvironment' s air exchange rate, al5 air pollutant
9 penetration, P15 and decay rate, kji
Equation 2-5
10 Hence, indoor NO2 exposure for microenvironment i is the sum of an ambient and a
1 1 nonambient component:
Ei=yi[Piai/(ai+ki)]Ca + Ean
Equation 2-6
November 2013 2-55 DRAFT: Do Not Cite or Quote
-------
1 Finally, Ea can be described as the sum of the outdoor NO2 exposure and the ambient
2 component of the indoor NO2 exposure, summed over /' indoor microenvironments (U.S.
3 EPA. 2008c: Wilson and Brauer. 2006; Wilson et al. 2000):
a = y0 + yi[Piai/(ai + fc<)] CB
Equation 2-7
4 If it is further assumed that the individual occupies only one indoor and one outdoor
5 microenvironment, then the infiltration term simplifies to y; [Pa/(a + k)], and since y0 + y;
6 =1, then an exposure factor, a, can be defined to express the influence of time-weighting
7 and infiltration on NO2 exposure:
Equation 2-8
Last, an approximate expression for total personal exposure is obtained:
Equation 2-9
9 Comparison of Equation 2-2. Equation 2-7. and Equation 2-9 reveals that a can also be
10 defined as the ratio Ea/Ca. This assessment focuses on the ambient component of NO2
1 1 exposure, Ea, because this is more relevant to the review of the NAAQS compared with
12 Ena. As such, subsequent sections examine how Ea and the factors a and Ca of the Ea
13 term are modeled or measured.
2.6.2 Spatially Resolved Models for Use in Exposure Assessment
14 Computational models are employed to provide estimates of exposure when
15 measurements are not available at locations and/or times needed to estimate spatial and
16 temporal variability in concentration within communities. These methods can sometimes
17 account for complex urban morphometry and meteorology, which can interact to cause
November 2013 2-56 DRAFT: Do Not Cite or Quote
-------
1 turbulence that may affect pollutant residence times (Fernando. 2010) or incorporate
2 localized sources that might not otherwise be detected by central site monitoring
3 (Goldman et al., 2012). Such estimates can then be used as inputs to exposure models
4 described in Section 2.6.2.4. These modeling approaches produce data at times and/or
5 locations where exposures are uncharacterized, but each method carries its own
6 uncertainty (Tuentes. 2009). Detailed descriptions of computational models used for
7 predicting spatially resolved concentration profiles for exposure assessment have been
8 provided in Section AX 3.6 of the 2008 ISA for Oxides of Nitrogen Annex (U.S. EPA.
9 2008a) and Section 3.8 of the 2009 ISA for PM (U.S. EPA. 2009a). Methods include
10 chemical transport models (CTM), land use regression (LUR) models, spatial
11 interpolation through statistical techniques, and dispersion models.
2.6.2.1 Chemical Transport Models
12 CTMs can be used to develop estimates of human exposure to NOX, as tested in Marshall
13 et al. (2008). CTMs are used to compute interactions among atmospheric pollutants and
14 their transformation products, the production of secondary aerosols, the evolution of
15 particle size distribution, and transport and deposition of pollutants. CTMs are driven by
16 emissions inventories for primary species such as NO2, SO2, NH3, VOCs, and primary
17 PM, and by meteorological fields produced by other numerical prediction models. Values
18 for meteorological state variables such as winds and temperatures are taken from
19 operational analyses, re-analyses, or weather circulation models. In most cases, these are
20 off-line meteorological analyses, meaning that they are not modified by radiatively active
21 species generated by the air quality model (AQM). Work to integrate meteorology and
22 chemistry was done in the mid-1990s by Lu et al. (1997a. b) and references therein,
23 although limits to computing power prevented their wide-spread application. More
24 recently, new, integrated models of meteorology and chemistry are now available as well;
25 see, for example, Binkowski et al. (2007) and the Weather Research and Forecast model
26 with chemistry (WRF-Chem) (http://ruc.noaa.gov/wrfAVGl I/).
27 CTMs provide regional concentration estimates, and they are typically run with surface
28 grid resolutions of 4 km, 12 km, or 36 km. Temporal resolution of CTMs can be as fine
29 as one hour, although larger temporal aggregation often occurs for the purpose of
30 maintaining reasonable data file size. Hence, substantial uncertainties at the subgrid scale
31 remain (U.S. EPA. 2008a). In densely populated regions of the country, monitor density
32 may be finer than CTM surface grid resolution. Moreover, Community Multiscale Air
33 Quality (CMAQ) and other CTMs suffer from pollutant-specific concentration biases,
34 such as underestimation of total nitrate, that require correction (Fuentes and Raftery.
35 2005) prior to interpretation for exposure assessment. Bayesian combination (Fuentes and
November 2013 2-57 DRAFT: Do Not Cite or Quote
-------
1 Raftery. 2005) and downscaling (Berrocal et al.. 2010a. b) have recently been developed
2 to improve spatial resolution and provide bias correction. Isakov et al. (2009) developed a
3 methodology to model subgrid spatial variability within CMAQ using the American
4 Meteorological Society/Environmental Protection Agency Regulatory Model
5 (AERMOD) dispersion model prior to linking the modeled results with stochastic
6 population exposure models to predict annual and seasonal variation in urban population
7 exposure within urban microenvironments. In each case, these papers have referred to
8 other air pollutants, but the methodology is still applicable to NO2 exposure assessment.
2.6.2.2 Dispersion Models
9 Dispersion models, or Gaussian plume models, predict the transport and dispersion of
10 ambient air pollutants emanating from a point or line source through solution of an
11 equation that estimates the spread of the pollutant to follow a Gaussian curve that is a
12 function of distance from the source. Several studies of health effects related to NOX
13 exposure employ dispersion models to estimate NOX concentrations (e.g., Gruzieva et al..
14 2013; McConnell et al.. 2010; Oftedal et al.. 2009) because NO2 has high local spatial
15 variability (Section 2.5.3). The grid spacing in regional CTMs, usually between 1 and 12
16 km2, is usually too coarse to resolve spatial variations on the neighborhood scale. More
17 finely resolved spatial scales that better represent human exposure scales are provided by
18 smaller scale dispersion models. Several models could be used to simulate concentration
19 fields near roads, each with its own set of strengths and weaknesses. The California
20 Department of Transportation's most recent line dispersion model is CALINE4; see
21 http://www.dot.ca.gov/hq/env/air/software/caline4/calinesw.htm. The CALINE family of
22 models is not supported by the California Department of Transportation for modeling of
23 highway source NO2 and does not include NOX transformation chemistry.
24 In addition, AERMOD (http://www.epa.gov/scramOOl/dispersion_prefrec.htm) is a
25 steady state plume model formulated as a replacement to the ISC3 dispersion model. In
26 the stable boundary layer (SBL), it assumes the concentration distribution to be Gaussian
27 in both the vertical and horizontal dimensions. In the convective boundary layer (CBL),
28 the horizontal distribution is also assumed to be Gaussian, but the vertical distribution is
29 described with a bi-Gaussian probability density function. AERMOD has provisions that
30 can be applied to flat and complex terrain and multiple source types (including point,
31 area, and volume sources) in both urban and rural areas. It incorporates air dispersion
32 based on the structure of turbulence in the planetary boundary layer (PEL) and scaling
33 concepts and is meant to treat surface and elevated sources, in both simple and complex
34 terrain in rural and urban areas. The dispersion of emissions from line sources like
35 highways in AERMOD is handled as a source with dimensions set using an area or
November 2013 2-58 DRAFT: Do Not Cite or Quote
-------
1 volume source algorithm in the model; however, actual emissions are usually not in
2 steady state. Moreover, most simple dispersion models including AERMOD are designed
3 without explicit chemical mechanisms but do have non-default options to estimate
4 conversion of NO to NO2 based on a NOX/O3 titration model.
5 There are also non-steady state models for different types of sources. For example,
6 CALPUFF (http://www.src.com/calpuff/calpuffl .htm). which is EPA's recommended
7 dispersion model for transport in ranges >50 km, is a non-steady-state puff dispersion
8 model that simulates the effects of time- and space-varying meteorological conditions on
9 pollution transport, transformation, and removal and has provisions for calculating
10 dispersion from surface sources. However, CALPUFF was not designed to treat the
11 dispersion of emissions from roads, and like AERMOD has some limited chemistry
12 options to estimate production of secondary pollutants. The distinction between a steady-
13 state and time varying model could be unimportant for studying health effects where long
14 exposure time scales are relevant; however, when short exposure time scales are of
15 interest, it may be more important to capture the temporal variability in emissions.
2.6.2.3 Land-use Regression Models
16 Empirical LUR modeling has been applied extensively to estimate the spatial distribution
17 of ambient NO2 or NO for exposure assessment on a neighborhood or urban scale
18 (Hatzopoulou et al.. 2013; Cesaroni etal.. 2012; Gonzales et al.. 2012; Mukerjee et al..
19 2012a: Mukerjee et al.. 2012b: Oiamo et al.. 2012; Esplugues et al.. 2011; Fernandez-
20 Somoano et al.. 2011; Hystad et al.. 2011; Oiamo etal.. 2011; Rose etal. 2011; Smith et
21 al.. 2011; Szpiro et al.. 2011; Adamkiewicz et al.. 2010; Aguilera et al.. 2009; Cohen et
22 al.. 2009; Hart et al.. 2009; Iniguez et al.. 2009; Karr et al.. 2009; Mukeriee et al.. 2009;
23 Su et al.. 2009b: Aguilera etal.. 2008: Atari et al.. 2008: Cesaroni et al.. 2008: Rosenlund
24 et al.. 2008a). LUR fits a statistical model of concentration based on land use data and
25 then applies that model to locations without monitors to improve the spatial resolution of
26 the concentration field. LUR methods are used frequently, because they offer improved
27 spatial variability over other methods, with spatial resolutions of 300 meters for NO and
28 1 km for NO2 (Marshall et al.. 2008). Recently, nationwide LUR has been implemented
29 to examine local-scale estimates across a nation (Hystad etal.. 2011: Novotny et al..
30 2011; Hart et al.. 2009). Models are typically calibrated using data from NO2 or NO from
31 passive sampler measurements and several predictor variables, such as land use, road
32 length, population density, and proximity to areas of high concentrations (city center,
33 major road and/or highway, and point sources). Given that most passive measurement
34 methods are not designed for short-term sampling, LUR models are typically based on
35 several days or weeks of data and hence do not account for temporal variability well.
November 2013 2-59 DRAFT: Do Not Cite or Quote
-------
1 Several methodological issues must be considered when interpreting LUR model results;
2 these issues include number of measurement sites used to fit the statistical model,
3 predictor variable selection, and comparison of LUR performance among LUR model
4 formulations and with other models. These issues affect how well the spatial variability
5 of NOX concentration in a city is represented.
6 More finely resolved spatial resolution of calibration points can improve goodness of fit
7 of the model for the city in which it was fit, although generalizability of LUR results to
8 other cities may be independent of these factors. Allen et al. (2011) developed separate
9 LUR models for two Canadian cities (Winnipeg, Manitoba and Edmonton, Alberta) with
10 50 calibration points each and then applied the models to the other city to compare
11 performance. As anticipated, locally generated model performance (NO2: R2 = 0.81-0.84;
12 NO: R2 = 0.55-0.56) was superior to performance of the model fit for the other city (NO2:
13 R2 = 0.37-0.52; NO: R2 = 0.24-0.41) and to bivariate local models using only road
14 proximity (R2 < 0.19). NO2 models consistently performed better than NO models.
15 Parenteau and Sawada (2012) examined LUR model performance when basing the model
16 on successively finer spatial resolution from 2 km down to 50 meters, with the
17 geographic borders of the finely resolved regions tied to population groupings based on
18 population density mapping. The two finer resolution approaches yielded better
19 agreement with measured NO2 data (R2 = 0.80-0.81) than the less spatially resolved
20 approach (R2 = 0.70). Likewise, Dijkemaetal. (2011) compared LUR based on spatial
21 resolution and observed better agreement with NO2 observations for neighborhood-level
22 simulations (R2 = 0.57) compared with whole-city simulations (R2 = 0.47). Janssen et al.
23 (2012) proposed using LUR to improve validation of a CTM by downscaling the CTM to
24 the LUR. Downscaling entails a redistribution of the CTM-modeled concentrations
25 through a statistical model to conform to measured concentrations using the LUR-derived
26 regression parameters. Janssen et al. (2012) found that the spatial representativeness of
27 the CTM for NO2 improved by roughly 20% when incorporating the LUR downscaler.
28 Studies have evaluated LUR model performance when the LUR was fit with different
29 numbers of NO2 measurement sites and observed that the number of measurement sites
30 needed is sensitive to the LUR model design. Basagana et al. (2012) evaluated LUR
31 model for 24-120 NO2 measurement sites in Girona, Spain and different numbers of
32 predictor variables, starting with 106 prediction variables related to land use and then
33 reducing the set to 18 components through principal component analysis (PCA). Johnson
34 et al. (2010) evaluated LUR performance in New Haven, CT when the LUR model was
35 fit with NO2 data from 25-285 measurement sites. Wang et al. (2012) also evaluated
36 LUR performance when fit with 24-120 NO2 monitors in the Netherlands. These studies
37 (Basagana et al.. 2012; Wang etal.. 2012; Johnson etal.. 2010) observed that, when a
38 large number of prediction covariates were used, the model performed better (higher
November 2013 2-60 DRAFT: Do Not Cite or Quote
-------
1 adjusted R2 and R2 for cross-validation) for a smaller number of NO2 measurement sites
2 compared with the model using a larger number of NO2 sites, but when the number of
3 prediction covariates was reduced through PCA, then a larger number of NO2
4 measurement sites was needed.
5 Selection of predictor variables, such as meteorology, traffic, land use, and population
6 density, influences the ability of the LUR model to predict NOX concentrations and
7 depends on the specific city for which the model is fit. Su et al. (2008a) and Ainslie et al.
8 (2008) developed the Source Area-LUR (SA-LUR) to incorporate the effects of
9 meteorology on the model results. The SA-LUR integrates data for wind speed, wind
10 direction, and cloud cover variables in estimates for NO and NO2 and was found to
11 perform better when seasonal variability in concentrations was high. Su et al. (2008b)
12 included street canyon aspect ratio as an LUR predictor variable to account for retention
13 of pollutants in street canyons. They observed that, upon adding aspect ratio to the LUR
14 model, R2 increased from 0.56 to 0.67 for NO2 and from 0.72 to 0.85 for NO. Franklin et
15 al. (2012) explored bivariate correlations between NO2, NO, and NOX concentrations and
16 several predictors reflecting traffic, population, elevation, and land use in twelve southern
17 California communities. In this study, statistically significant correlations (p <0.005)
18 were observed between NO2, NO, and NOX concentrations and distance to road, traffic
19 volume, concentrations from dispersion models, population density, elevation of the
20 neighborhood relative to the community, local standard deviation of the elevation,
21 transportation land use, commercial land use, and residential land use. Su et al. (2009a)
22 developed a method to optimize the LUR variable selection process in which correlations
23 between several land use variables and NO2 concentrations were computed across a 3 km
24 buffer of the NO2 measurement (1.5 km buffer for traffic-related variables), and the data
25 for correlation versus distance were fit to a curve describing that relationship. The
26 variable with highest correlation at the optimum buffer distance is added to the model if
27 its addition produces a statistically significant change (p <0.1) in the model R2. Su et al.
28 (2009a) found the important variables to be distance from monitor, 24-h traffic levels,
29 expressway casement, open land use, railway, major road, land grade, population density,
30 and distance to coast. It can be anticipated that the important variables might be different
31 depending on city-specific factors.
32 LUR models applied several years after model development have demonstrated
33 moderate-to-good predictive ability in a few studies. Eeftens et al. (2011) compared LUR
34 obtained from NO2 measurements at 35 locations in the Netherlands over the years
35 1999-2000 with LUR developed from NO2 measurements at 144 locations in the
36 Netherlands during 2007. Both the NO2 measurements and the LUR models agreed well
37 for the two time periods studied ((3 = 0.9998; R2 = 0.89). Similarly, Wang etal. (2013)
38 tested stability of an LUR model for Vancouver, Canada between 2003 (based on 116
November 2013 2-61 DRAFT: Do Not Cite or Quote
-------
1 sites) and 2010 (based on 116 sites, with 73 from the 2003 study). Wang etal. (2013)
2 evaluated the model by testing how much variability in the measurements was predicted
3 by models from the other year with moderate results. Linear regression for comparison of
4 the 2003 model with 2010 measurements produced R2 = 0.58-0.60 for NO and R2 =
5 0.52-0.61 forNO2, while comparison of the 2010 model with 2003 measurements
6 produced R2 = 0.50-0.55 for NO and R2 = 0.44-0.49 for NO2.
7 LUR comparison with other models varies substantially among studies and depends on
8 the validation algorithm as well as the model conditions. A recent study of LUR
9 application in twenty European study areas, in which Wang et al. (In Press) found that
10 leave one out cross-validation (LOOCV), typically used to validate LUR, produced
11 higher R2 for NO2 compared with hold-out evaluation (HEV) (LOOCV: R2 = 0.83, HEV:
12 R2 = 0.52). LOOCV entails repeatedly withholding a fraction of the monitoring sites from
13 the fitting process for validation and then computing an ensemble R2, whereas HEV
14 entails prediction with the LUR at locations not fit by the model. Mercer et al. (2011)
15 compared ten-fold cross-validated LUR with universal kriging (UK), in which a surface
16 of concentrations was built based on measured values, for three seasons in Los Angeles
17 with roughly 150 measurement sites. UK performance was slightly better than LUR for
18 all seasons (UK: R2 = 0.75, 0.72, and 0.74; LUR: R2 = 0.74, 0.60, 0.67). Li etal. (2012b)
19 developed a new formulation for LUR using generalized additive models (GAM) and
20 cokriging to boost the performance of LUR over LUR methods using linear models and
21 evaluated it for Los Angeles, CA. GAM enables incorporation of localized nonlinear
22 effects among the prediction covariates, while cokriging is intended to improve spatial
23 smoothing. The LUR using GAM and cokriging, had the highest cross-validation (R2 =
24 0.88-92), compared with universal kriging (R2 = 0.68-0.75) and multiple linear LUR (R2
25 = 0.42-0.64). Beelen etal. (2010) compared LUR with a dispersion model incorporating a
26 near road module for modeling NO2 concentrations in a Rotterdam, the Netherlands
27 neighborhood. The dispersion model agreed better (R = 0.77) compared with LUR (R =
28 0.47) with NO2 measurements from 18 validation sites. Dijkema et al. (2011) also
29 compared LUR for the city of Amsterdam and neighborhoods therein with a dispersion
30 model and found better agreement of the dispersion models with observations for the
31 city-wide model than for LUR (dispersion: R2 = 0.74; LUR: R2 = 0.47) although
32 agreement was comparable for the neighborhood specific model (R2 = 0.57 for both
33 models). Molter etal. (2010a) used dispersion modeling data in lieu of measurement data
34 in an LUR for Greater Manchester, U.K. and found reasonable agreement of NO2
35 predictions with validation monitoring data (R2 = 0.86). Marshall et al. (2008) compared
36 LUR with inverse distance-weighted spatial interpolation of NO and NO2 measurements,
37 nearest NO and NO2 measurements, and a CMAQ model run for Vancouver, Canada.
38 The LUR location was matched to each CMAQ grid cell centroid and compared with the
39 grid cell concentration. LUR and CMAQ produced similar average absolute bias in the
November 2013 2-62 DRAFT: Do Not Cite or Quote
-------
1 concentration compared with measured concentrations for NO (LUR: 42%, CMAQ:
2 47%) and NO2 (LUR: 17%, CMAQ: 17%), while nearest monitor and spatial
3 interpolation methods produced less than 5% bias for both pollutants and methods.
2.6.2.4 Stochastic Population Exposure Models
4 Stochastic population exposure models combine measured or modeled ambient NO2
5 concentration data with population-level statistical distributions of time-activity data, air
6 exchange rate for residences and other buildings, meteorology, physiological parameters,
7 and other relevant data to simulate microenvironmental NO2 concentrations and
8 individuals' exposures to NO2, which is then summarized through descriptive statistics
9 for the simulated population. The state of the science for stochastic population exposure
10 models has not changed substantially since the 2008 ISA for Oxides of Nitrogen, as
11 described in detail in 2008 Annex 3.6 (U.S. EPA. 2008c). Examples of stochastic
12 population exposure models include the Air Pollution Exposure (APEX), Stochastic
13 Human Exposure and Dose Simulation (SHEDS), and EXPOLIS (exposure in polis, or
14 cities) models, which involve stochastic treatment of the model input factors (Kruize et
15 al., 2003; Burke et al., 2001). Advancement in exposure modeling has come from its
16 integration with chemical transport models of outdoor air quality through a hybrid
17 approach (Isakov et al., 2009) and characterization of the uncertainty in these models
18 (Ozkavnak et al.. 2009: Zidek et al. 2007).
19 Hybrid exposure modeling uses ambient air quality input from grid-based models rather
20 than from central site monitoring data, as is typically done (Isakov et al., 2009). In the
21 hybrid version, the CMAQ model is used to simulate concentrations for a coarse discrete
22 grid, e.g., 12 km x 12 km. Next, local scale concentrations from point and mobile sources
23 are estimated using Gaussian dispersion modeling through AERMOD. In combination,
24 these models produce an ambient air quality estimate at the location of the receptor that is
25 then input into APEX or SHEDS to estimate total human exposure. Isakov et al. (2009)
26 observed that the omission of specific point and traffic sources led to an underestimate in
27 median concentration by up to a factor of two, depending on location; these simulations
28 were for benzene and PM2 5. NO2 tends to be comparable in spatial variability with
29 benzene and more spatially variable compared with PM25 (Beckerman et al.. 2008).
30 Recent studies have considered the variability and uncertainty associated with exposure
31 modeling. Ozkaynak et al. (2009) considered uncertainty and variability in simulations
32 involving estimation of concentration, exposure, and dose in separate compartments of a
33 model. They found that uncertainty and variability propagated from one compartment to
34 the next. Zidek et al. (2007) addressed uncertainty and variability in exposure modeling
November 2013 2-63 DRAFT: Do Not Cite or Quote
-------
1 by using distributions of input parameters in the exposure model framework rather than
2 point estimates. Zidek et al. (2007) examined uncertainty and variability in ambient and
3 microenvironmental concentrations, time-activity data, intrinsic variables (e.g., age, sex,
4 exercise), meteorology, and geographical position. These models estimate time-weighted
5 exposure for modeled individuals by summing exposure in each microenvironment
6 visited during the exposure period. Zidek et al. (2007) found that use of distributions of
7 input parameters enabled examination of cases for potential subpopulations with common
8 characteristics. Note that both of these studies model PM, but the findings are applicable
9 toNO2.
10 Sarnat et al. (2013b) recently compared risks of cardiovascular and respiratory morbidity
11 with 24-h average NOX concentration and other primary and secondary air pollutants in
12 Atlanta using various exposure metrics. Epidemiologic results based on the mean,
13 median, and 95th percentile of the exposure distributions from APEX were compared
14 with measures from a central site monitor, regional background, AERMOD, and a hybrid
15 model merging AERMOD output with regional background data. NOX concentrations
16 modeled with APEX were generally higher than those obtained with the hybrid model,
17 likely because the APEX model incorporates road activity levels in its exposure
18 estimates. Epidemiologic analyses for emergency department admission for
19 asthma/wheeze produced statistically significantly higher risk ratios for the APEX mean,
20 median, and 95th percentile compared with the hybrid model and central site and
21 background metrics but negligible difference among the APEX and hybrid results for
22 emergency department admission for all respiratory or cardiovascular diseases.
2.6.3 Personal Sampling Considerations
23 The following sections outline personal NO2 and NOX exposure sampling techniques,
24 penetration, and indoor sources, sinks, and chemistry. This information is provided for
25 context about exposure to oxides of nitrogen infiltrating indoors, since total personal
26 exposure consists of ambient and nonambient oxides of nitrogen.
2.6.3.1 Personal Sampling Techniques
27 Personal sampling for NO2 was described in detail in Annex 3.3 to the 2008 ISA for
28 Oxides of Nitrogen (U.S. EPA. 2008c). Active sampling systems typically involve air
29 pumped past a chemiluminescent device; they enable measurements of NO2 over short
30 time periods to produce near real-time data. Given the weight of active sampling systems,
31 they are not used extensively for personal sampling. Passive samplers based on Pick's
November 2013 2-64 DRAFT: Do Not Cite or Quote
-------
1 first law of diffusion are more commonly deployed for personal NO2 sampling in a
2 badge, tube, or radial manifold. Passive sampling results are integrated over the time
3 period during which the sorbent material is exposed. The 2008 ISA for Oxides of
4 Nitrogen (U.S. EPA. 2008c) reported that, depending on the sorbent material, personal
5 NO2 samplers may be subject to biases related to interferences from HONO, PAN, HNO3
6 (Gair et al.. 1991). and high relative humidity (Centre di Ricerche Ambientali. 2006).
7 Recent work has been performed to evaluate passive sampling device performance.
8 Sather et al. (2007) compared Ogawa passive samplers with an NO2 FRM monitor over a
9 four-week field study in El Paso, TX and observed good agreement, with an average
10 absolute difference of 1.2 ppb with R2 = 0.95. For measurements in Umea, Sweden,
11 Hagenbjork-Gustafsson et al. (2009) observed that, when using the manufacturer's
12 recommended uptake rates to calculate concentration, NO2 measurements were
13 negatively biased by 9.1%, and NOX concentration measurements were positively biased
14 by 15% compared with an FRM. When uptake rates were derived in the field based on
15 the chemiluminescent FRM, NO2 measurements were positively biased by 2%, and NOX
16 concentration measurements were unbiased. These results suggest that deviation from
17 temperature conditions under which the samplers were laboratory tested may lead to
18 biased results. Jimenez et al. (2011) used Palmes-type passive diffusion tubes to measure
19 both NO2 and NOX concentrations and investigated specific sources of biases in their
20 measurements. They found that, within the passive diffusion tubes, NO and O3 were
21 reacting to form NO2, causing NO measurements to be negatively biased while NO2
22 measurements were positively biased. Wind was also a source of positive bias in the NO2
23 and NOX concentration measurements because increased airflow effectively reduced the
24 diffusion lengths of the gas collection tubes. In laboratory and field evaluation of NO2
25 passive diffusion tubes, Buzica et al. (2008) observed negligible difference between the
26 diffusion tubes and FRM measurements; however, uncertainty increased with decreasing
27 concentration.
28 Triethanolamine (TEA) is used as an alternative to activated charcoal sorbent material,
29 because it can be applied in an even coating. However, sampling efficiency is sensitive to
30 sampler flow rate (Vichi and De Santis. 2012) and relative humidity (Poddubny and
31 Yushketova. 2013: Sereviciene and Paliulis. 2012: Vardoulakis et al.. 2009). Heal (2008)
32 found that NO2 bias was sensitive to method of application of the TEA to the substrate.
33 Sekine et al. (2008) and Nishikawa et al. (2009) experimented with size and number of
34 filters, respectively, in a passive sampler and found minimal effect on NO2 or NOX
35 concentration. However, Ozden and Dogeroglu (2008) observed that TEA-complexed
36 NO2 was sensitive to photodegradation if not stored in a dark glass tube.
November 2013 2-65 DRAFT: Do Not Cite or Quote
-------
2.6.3.2 Sources, Sinks, and Penetration
1 The general understanding production of oxide of nitrogen indoors has not changed since
2 the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). Indoor sources of oxides of
3 nitrogen are combustion-based, including gas stoves, gas heating, oil furnaces, coal
4 stoves, wood burning stoves, kerosene heaters, smoking, and to a lesser extent, electric
5 cooking. The magnitude of indoor oxides of nitrogen depends on ventilation of the indoor
6 space and appliances as well as source strength. Recent studies show associations
7 between indoor NO2 levels and indoor combustion (Vrijheid et al.. 2012; Kornartit et al..
8 2010; Park et al., 2008). HONO can also be emitted directly during combustion or
9 through surface reactions, as described in Section 2.6.3.3. Park et al. (2008) measured
10 HONO and NO2 during combustion and compared their results with older studies in the
11 peer-reviewed literature, as shown in Table 2-3. This review generally found higher
12 HONO concentrations in the presence of indoor combustion sources. Oxides of nitrogen
13 can be lost through indoor deposition and ventilation (U.S. EPA. 2008c). Sarwar et al.
14 (2002) reported deposition velocities of 6-7 x 10'5 m/sec for NO2, HONO, HNO3,
15 HO2NO2, NO3, and N2O5. Much lower deposition velocities (N.D. - 2 x 10"6 m/sec)
16 were reported for NO, PAN, and organic NO3 species.
November 2013 2-66 DRAFT: Do Not Cite or Quote
-------
Table 2-3 Indoor NO2 and MONO concentrations in the presence and absence
of combustion.
Study
Braueretal. (1990)3
Braueretal. (1990)b
Braueretal. (1991)°
Spenqleretal. (1993)d
Simon and Dasqupta
(1995)
Leaderer et al. (1999)'
Khoder (2002)g
Lee et al. (2002)h
Jarvis et al. (2005)1
Hong et al. (2007V
Park et al. (2008)
Combustion Source
No source (background)
Gas range3
Convective space heater3
No source
Gas rangeb
Unknown
Gas range, stove, furnace
Kerosene heater
No source'
Gas stoves'
Kerosene heaters'
No source'
Gas stoves'
Gas appliances (summer)
Gas appliances (winter)
Gas range, etc.
Gas hob
Gas oven
Gas range
Gas range
Measurement
Frequency
15 min
15 min
15 min
15 min
15 min
15 min
24-h
8 min
24-h
24-h
24-h
24-h
24-h
24-h
24-h
6-day
4 min
4 min
N02 (ppb)
Peak 24-h avg
29 17
157 36
955 209
5.0 1.8
37 8
-
60
(24-115)
-
-
-
39
(20-73)
65
(27-120)
28
(4.3-52.0)
12.8
12.8
81.1
189.3 19.4
MONO (ppb)
Peak 24-avg
8 5
35 13
106 42
3.5 3.4
31 9.6
1-12
4.7 (2-8)
5-10
0.8
(0.0-2.9)
4.0
(0.0-11.3)
6.8
(0.2-35.9)
2.4
(0.1-20.1)
5.5
(0.4-20.1)
3.7
(1.3-7.3)
6 8
(1.6-12.5)
4.6
(0.1-21.1)
4.1
5.0
9.3
15.2 2.1
"Location: Chicago, IL, research home, unvented combustion condition; Gas range operation hours: 1 h (with one burner and
2,320 kcal/h); Convective space heater operation hours: 4 h (with one burner and 2785 kcal/h).
bLocation: Maryland research home, unvented combustion condition; Gas range operation hours: 1 h (with one burner and
2320 kcal/h).
location: 11 Boston, MA, homes (winter).
dLocation: 10 homes in Albuquerque, NM (winter).
eLocation: four different home environments with small kerosene heater (2270 kcal/h).
'location: 58 homes (summer) and 223 homes (winter) in southwest Virginia and Connecticut, U.S.; 39 inside homes without gas
stoves (summer); 19 inside homes with gas stoves (summer); 74 inside kerosene-heater homes (winter); 96 inside homes without
kerosene heaters and gas stoves (winter); 52 inside homes without kerosene heaters and with gas stoves (winter).
9Location: Four homes in suburban residential areas in Greater Cairo, Egypt.
Yocation: 119 homes in southern California (spring).
'Location: Homes in European community.
'Location: Living room of an apartment in Gwangju, Korea (May 2006).
Source: Reprinted with permission of Elsevier, Parket al. (2008).
November 2013
2-67
DRAFT: Do Not Cite or Quote
-------
2.6.3.3 Indoor Chemistry
1 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) described well-established
2 reactions involving oxides of nitrogen and other indoor air pollutants for gas-phase and
3 surface chemistry that serves as both a source and sink for oxides of nitrogen. Knowledge
4 of indoor chemistry helps identification of potential sources of uncertainty in estimates of
5 indoor exposure to ambient oxides of nitrogen. For gas phase reactions, indoor NO can be
6 oxidized to NO2 via reaction with O3 or HO2 radicals generated by indoor O3 chemistry
7 or volatile organic compounds found in household products. NO2 can react with O3 to
8 form NO3 radicals that may subsequently oxidize organic compounds. NO2 also reacts
9 with free radicals to produce organic nitrates (RONO2) and peroxynitrates
10 (RC(=O)OONO2). NO2 removed through surface reactions was known to contribute to
11 NO levels indoors either by surface reduction of NO2 or by reaction of NO2 with aqueous
12 HONO on indoor surfaces (Spicer et al., 1989). Conversion of NO2 to HONO occurs
13 through a number of indoor surface reactions, and reaction increases with increased
14 relative humidity. Surface reactions of NO and OH radicals may also produce HONO,
15 but the reaction rate is slower than for NO2.
16 Indoor combustion can lead to direct emission of NO and HONO, and conversion of NO
17 to NO2 can lead to secondary HONO production from heterogeneous reactions involving
18 NO2 on indoor surfaces. Park et al. (2008) observed HONO to be correlated with both
19 NO (r = 0.64) and NO2 (r = 0.68) during combustion. They noted that HONO
20 concentrations were 4-8% of NO2 concentrations during gas range operations but rose to
21 -25% of NO2 concentrations after combustion ceased, which underscores the role of
22 surface reaction as the major source of HONO production. In a model of combustion
23 products for oxides of nitrogen during candle and incense burning, Loupa and
24 Rapsomanikis (2008) observed simultaneous NO and HONO production, the latter of
25 which were in agreement with older test chamber results of HONO production during
26 combustion (De Santis et al., 1996).
27 Recent gas-phase indoor chemistry work has shed light on processes involving organic
28 compounds and/or secondary organic aerosols (SOA). Carslawetal. (2012) modeled
29 indoor reactions forming SOA and observed that for their base case simulation, organic
30 nitrates constituted 64% of the overall SOA, while PANs constituted an additional 21%.
31 In sensitivity tests varying ambient concentrations and meteorological conditions, organic
32 nitrates varied from 23-76% of the SOA, and PAN varied from 6-42%. Najgaard et al.
33 (2006) investigated the interference of NO2 in ozonolysis of monoterpenes in a
34 simulation of indoor air chemistry and observed that NO2 reacted with O3 and hence
35 reduced SOA formation from ozonolysis of alkenes a-pinene and /?-pinene while
36 increasing the mode of the SOA size distribution. However, the presence of NO2 had less
November 2013 2-68 DRAFT: Do Not Cite or Quote
-------
1 effect on ozonolysis of d-limonene, and this is thought to occur because the ozonolysis
2 reaction rate is faster. In chamber experiments and computational chemistry models, Cao
3 and Jang (2008) and Cao and Jang (2010) tested toluene SOA formation in the presence
4 of low (< 3 ppb), medium (90-135 ppb), and high (280-315 ppb) NOX concentrations and
5 found that the organic matter component of the toluene SOA yield generally decreased
6 with increasing NOX concentrations, especially when high NO levels (-222-242 ppb)
7 were present. Ji et al. (2012) explored rate constants of NO2 reactions with various low
8 molecular weight aldehydes found indoors and observed that the reaction rates, k,
9 increased in the following order: kformaldehyde <£acetaidehyde <£Propanai <£butanai. Ji et al. (2012)
10 concluded from this observation that NO2 reacts more with longer chain, low molecular
11 weight aldehydes compared with shorter chain, low molecular weight aldehydes.
12 RC(=O)- radicals and HONO were both observed to be products of these reactions.
13 Reactions involving N2O5 (formed by reaction of NO2 and NO3 in the presence of
14 another molecule) in an indoor context have been studied in recent years. In an
15 examination of NO3 and N2O5 (measured as the sum of those two species) in an office
16 building, N0jgaard (2010) observed that alkenes remove more indoor NO3 and N2O5
17 than either ventilation or surface deposition. Griffiths et al. (2009) studied N2O5 uptake
18 by organic aerosols in a reaction cell and large (260 m3) chamber and observed little
19 N2O5 uptake by solid organic aerosols, more efficient uptake by liquid aerosols, and
20 uptake that increased with increasing relative humidity (RH). N2O5 uptake by
21 dicarboxylic acids (oxalic acid, malonic acid, succinic acid, and glutaric acid) was
22 30-90% of that by (NH4)2SO4 and (NH4)2SO4-mixed dicarboxylic acid aerosols at
23 similar RH. N2O5 uptake by malonic or azelaic acid in the presence of higher RH is
24 consistent with findings of Thornton et al. (2003) for experiments conducted in a reaction
25 cell. Raff etal. (2009) suggested that N2O5 autoionizes to NO2+NO3" and then reacts
26 quickly with water to form HNO3; it is possible that HNO3 might then participate in the
27 liquid aerosol reactions described by Griffiths et al. (2009) and Thornton et al. (2003).
28 Raff etal. (2009) also proposed autoionization of N2O5 as a likely mechanism for
29 reaction with HC1, which would result in C1NO and HNO3 formation while NO2 and
30 water vapor experienced an intermediate surface reaction to form HONO, which would
31 react with HC1.
2.6.4 Oxides of Nitrogen in a MuItipollutant Context
32 Correlations between ambient or personal NO2 and other copollutants can help reveal
33 information on source emissions, exposures, and health outcomes of NO2 Studies and
34 analyses reported in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c)
35 demonstrated that ambient NO2 was moderately correlated with several traffic-related
November 2013 2-69 DRAFT: Do Not Cite or Quote
-------
1 pollutants (e.g., PM25, CO, and EC) in urban and suburban areas, suggesting that in some
2 cases NO2 can be a surrogate for traffic pollution. Compared to other traffic-related
3 pollutants, EC generally had the strongest correlations with NO2. In contrast, O3 was
4 generally poorly or negatively correlated with NO2. A limited number of studies reported
5 in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) investigated the relationship
6 between personal NO2 and personal or ambient measurements of other pollutants (e.g.,
7 PM2 5, EC, CO, volatile organic compounds, and HONO). In most cases, personal NO2
8 was moderately correlated with these pollutants. More recent studies expand upon these
9 findings and are discussed below.
2.6.4.1 Ambient Relationships between NO2 and Copollutants
10 Numerous air quality, exposure and epidemiologic studies have evaluated associations
11 between concentrations of ambient NO2 and those of other pollutants. Many of these
12 studies report Pearson or Spearman correlations of ambient NO2 with other NAAQS
13 pollutants, mainly focusing on those related to traffic sources (PM25, CO, PMi0). A few
14 studies have explored associations between NO2 and other traffic-related pollutants, such
15 as EC, ultrafine particulate matter (UFP), and volatile organic compounds (VOCs). Data
16 for criteria pollutants are summarized in Table 2-4; VOCs are not included in this table
17 due to a limited amount of data.
18 Figure 2-19 shows the range of NO2 copollutant correlation coefficients among the
19 studies in Table 2-4. Existing studies indicate that NO2 is, in general, moderately
20 correlated with other NAAQS and traffic-related pollutants. Similar to findings in the
21 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). the strongest correlations are
22 typically observed for NO2 with primary traffic-related pollutants, such as CO, BC and
23 UFP. Correlations between NO2 and CO are highest on average, with most correlations
24 exceeding 0.6. A wide range of correlations is observed for NO2 with PM2 5, PMi0, and
25 SO2. The lowest correlations are typically observed between NO2 and O3, with
26 correlations typically being negative or very low (r = -0.71-0.66, median r = 0.15).
27 Fewer studies have explored seasonal correlations between NO2 and copollutants.
28 Among these, a majority of studies report correlations of NO2 with PM2 5 and PMi0. In
29 general, studies show stronger correlations of NO2 with PM25 and PMi0 during cooler
30 seasons. Connell et al. (2005) investigated associations between PM2 5 and gaseous
31 copollutants in Steubenville, OH using linear regression. NO2 was more strongly
32 correlated with PM25 during the fall (R2 = 0.53) and winter (R2 = 0.53) seasons compared
33 with the spring (R2 = 0.27) and summer (R2 = 0.086) seasons. Similarly, Sarnat et al.
34 (2005) found positive associations between PM2 5 and NO2 during both seasons
November 2013 2-70 DRAFT: Do Not Cite or Quote
-------
1 (summer: (3 = 0.44; winter: (3 = 0.64), with stronger associations in the winter in
2 Baltimore, MD. Arhami et al. (2009) evaluated relationships between personal, indoor,
3 and ambient copollutants at two sites in southern California (San Gabriel Valley and
4 Riverside) for warmer and cooler seasons. During the warm season, the Spearman
5 correlation coefficient (average among sites) was small (r = 0.09) between NO2 and
6 PM25, whereas during the winter the correlation was slightly stronger (r = 0.50).
7 However, they did not observe a consistent seasonal trend between NO2 and PM10. While
8 associations between NO2 and PMi0 were substantially lower during the summer (r =
9 0.21) at the Riverside site, correlations were relatively similar during both seasons at the
10 San Gabriel Valley site (summer PM10: r = 0.31; winter PM10: r = 0.34).
11 The correlation between NO2 and O3 may also have seasonal patterns, although limited
12 seasonal data exists between these two pollutants. In the 2008 ISA for Oxides of Nitrogen
13 (U.S. EPA. 2008c). ambient concentrations of NO2 and O3 from several sites across Los
14 Angeles, CA were compared during a multi-year period. Slightly positive correlations
15 between these two pollutants were observed during the summer (r = 0 to 0.4), while
16 negative correlations were observed during the winter (r = -0.5 to -0.8). The slightly
17 positive correlations during the summer can be attributed in part to increased
18 photochemical activity, resulting in enhanced O3 formation. Higher O3 concentrations
19 increase the ratio of NO2 to NO due to enhanced oxidation, thereby resulting in a
20 stronger correspondence between NO2 and O3 during the summer. Only one study in
21 Table 2-4 reported seasonal differences in the correlation between NO2 and O3. Sarnat et
22 al. (2001) measured daily concentrations of gaseous and PM pollutants during different
23 seasons in Baltimore, MD. Similar to the trends reported in the 2008 ISA for Oxides of
24 Nitrogen, they observed a negative correlation between NO2 and O3 during the winter (r
25 = -0.71) and a near-zero correlation during the summer (r = 0.02). However, because
26 there is a lack of studies reporting such correlations, it is uncertain whether or not this
27 seasonal trend exists between the two pollutants in different locations.
28 Recent studies have also compared NO2 copollutant correlations across different regions
29 in the U.S. Baxter et al. (2013) studied differences in air pollution for the Northeast
30 (Boston, MA; Pittsburgh, PA), South (Memphis, TN; Birmingham, AL), Midwest
31 (Milwaukee, WI; Detroit, MI), and West (San Diego, CA; Riverside, CA). Average
32 Spearman correlation coefficients between PM2 5 and NO2 for each region were different
33 (Northeast: r = 0.44; South (data available for one city only): r = 0.27; Midwest: r = 0.57;
34 West: r = 0.47). Schildcrout et al. (2006) compared a number of gaseous and particulate
35 pollutants in different cities across the U.S., including Albuquerque, NM; Baltimore,
36 MD; Boston, MA; and Denver, CO. While correlations between ambient NO2 and CO
37 were relatively similar in all four locations, larger differences were observed between
38 NO2 and PM10 correlations, ranging from a moderate correlation in Denver (r = 0.64) to
November 2013 2-71 DRAFT: Do Not Cite or Quote
-------
1 low correlations in Baltimore and Boston (r = 0.26 for both cities). Other multicity
2 studies conducted outside of the U.S. show that NO2 copollutant correlations are widely
3 variable (Faustini etal.. 2011; Dales etal.. 2010. 2009b; Stieb et al.. 2008; Timonen et
4 al.. 2006).
5 A small subset of studies investigated correlations between NO2 and traffic-related VOCs
6 (volatile organic compounds), such as benzene, toluene, ethene, and xylene (BTEX). In
7 these studies, correlations between NO2 and VOCs are variable. Beckerman et al. (2008)
8 observed a strong correlation between NO2 and BTEX in a near-road field campaign. In a
9 panel study, Greenwald et al. (2013) compared ambient concentrations of traffic
10 pollutants monitored at two schools in El Paso, Texas, including one school within close
11 proximity to a major roadway with heavy diesel truck traffic. A moderately strong
12 correlation (r = 0.77) was observed between NO2 and BTEX (presented as the sum of
13 benzene, toluene, ethene, and xylene), suggesting that both pollutants are related to traffic
14 sources. Another study by Martins et al. (2012) estimated personal NO2 and BTEX
15 exposure during four one-week periods using a microenvironment approach that
16 combined outdoor and indoor concentrations with time activity patterns. In contrast to
17 Beckerman et al. (2008) and Greenwald et al. (2013). Martins etal. (2012) consistently
18 observed poor correlations (r=-0.423-0.138) between NO2 and BTEX during different
19 periods. The lack of correlation between these pollutants can be attributed in part to
20 differences in sources between indoor and outdoor microenvironments. While exposure
21 to VOCs, namely benzene, was attributed mainly to indoor sources, NO2 was largely
22 associated with traffic sources. These studies emphasize that proximity to roadway and
23 time spent in various indoor and outdoor microenvironments can impact the relationship
24 between NO 2 and traffic-related VOCs.
November 2013 2-72 DRAFT: Do Not Cite or Quote
-------
Table 2-4 Synthesis of NO2 ambient-ambient copollutant correlations reported in the literature.
Study
Faustini et al. (2011)
Samolietal. (2011)
Ko et al. (2007a)
Mehta et al. (2013)
Andersen et al. (2008a)
Mannes et al. (2005)
Schildcrout et al. (2006)
Liu et al. (2009b)
Straketal. (2013)
O'Connor et al. (2008)
Timonen et al. (2006)
Guo et al. (2009)
Time
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
Location
6 Italian Cities
Athens, Greece
Hong Kong
Ho Chi Minh City, Vietnam
(dry season)
Ho Chi Minh City, Vietnam
(wet season)
Copenhagen, Denmark
Sydney, Australia
Albuquerque, NM
Baltimore, MD
Boston, MA
Denver, CO
San Diego, CA
St. Louis, MO
Toronto, Canada
Ontario, Canada
Locations across
the Netherlands
Inner-cities across the U.S.
Amsterdam, the Netherlands
Erfurt, Germany
Helsinki, Finland
Beijing, China
Correlation
Measure
Pearson
NR
Pearson
NR
NR
Spearman
Pearson
NR
NR
NR
NR
NR
NR
NR
Spearman
Spearman
NR
Spearman
Spearman
Spearman
Pearson
Pollutant
CO
NR
NR
NR
NR
NR
NR
0.57
0.76
0.69
0.80
0.85
0.92
0.71
0.63
NR
NR
0.54
0.76
0.86
0.32
NR
03
NR
NR
0.34
0.44
0.17
-0.58
0.29
0.04
0.44
0.47
0.24
0.39
0.42
0.40
-0.51
-0.62
-0.31
NR
NR
NR
NR
SO2
NR
0.55
0.66
0.29
0.01
NR
NR
NR
0.49
0.68
0.56
0.23
0.58
0.63
0.18
NR
0.59
NR
NR
NR
0.53
PM2.5
NR
NR
0.44
NR
NR
0.41 (PM2.s)
0.67 (UFP)
0.66
NR
NR
NR
NR
NR
NR
NR
0.71
0.45(PM2.s)
0.56 (PNC)
0.59
0.49
0.82
0.35
0.67
PM10
0.19-
0.79
NR
0.4
0.78
0.18
0.43
0.47
0.26
0.62
0.48
0.64
0.55
0.45
0.64
NR
0.49
NR
NR
NR
NR
NR
November 201:
2-73
DRAFT: Do Not Cite or Quote
-------
Table 2-4 (Continued): Synthesis of NO2 ambient-ambient copollutant correlations reported in the literature.
Study
Dales et al.
Dales et al.
(2010)
(2009b)
Rojas-Martinez et al. (2007a)
Sarnat et al
Sarnat et al
. (2001)
. (2005)
Kim et al. (2006a)
Roberts and Martin (2006)
Andersen et al. (2007)
Chen et al.
(2008)
Arhami et al. (2009)
Baxter et al
Williams et
. (2013)
al. (2012a)
Time
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
Location
Santiago Province, Chile
7 Chilean Cities
Mexico City, Mexico
Baltimore, MD (summer)
Baltimore, MD (winter)
Baltimore, MD (summer)
Baltimore, MD (winter)
Toronto, Canada
Cleveland, OH
Nashville, TN
Copenhagen, Denmark
Shanghai, China
San Gabriel Valley, CA
(Summer and Fall)
San Gabriel Valley, CA
(Fall and Winter)
Riverside, CA
(Summer and Fall)
Riverside, CA
(Fall and Winter)
Boston, MA
Pittsburgh, PA
Memphis, TN
Detroit, Ml
Milwaukee, Wl
San Diego, CA
Riverside, CA
Research Triangle Park, NC
Correlation
Measure
NR
Pearson
Pearson
Spearman
Spearman
Spearman
Spearman
Spearman
NR- Pairwise
NR-Pairwise
Spearman
NR
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Spearman
Pollutant
CO
NR
0.79-
0.84
NR
0.75
0.76
NR
NR
0.72
0.67
0.36
0.74
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
03
NR
-0.34 -
-0.09
0.17
0.02
-0.71
NR
NR
NR
0.36
0.26
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
-0.12
SO2
NR
0.42-0.80
NR
NR
-0.17
NR
NR
NR
0.56
0.08
NR
0.73
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
PM2.5
0.73-0.92
. 12. — U.OZ
NR
0.37
0.75
0.44
0.64
0.44
NR
NR
NR
NR
0.10
0.44
0.07
0.56
0.41
0.46
0.27
0.59
0.55
0.57
0.37
0.03
0.25 (BC)
PM10
NR
0.61 -0.79
0.25
NR
NR
NR
NR
NR
0.63
0.44
0.42
0.71
0.31
0.34
0.21
0.64
NR
NR
NR
NR
NR
NR
NR
NR
November 2013
2-74
DRAFT: Do Not Cite or Quote
-------
Table 2-4 (Continued): Synthesis of NO2 ambient-ambient copollutant correlations reported in the literature.
Study
Delfino et al. (2008a)
Suh and Zanobetti (201 Ob)
Schembari et al. (2013)
Chuang et al. (2008)
Strickland et al. (2010)
Villeneuve et al. (2007)
Jalaludin et al. (2007)
Mortimer et al. (2002)
Burnett et al. (2000)
Maretal. (2000)
Tolbert et al. (2007)
Moshammer et al. (2006)
Time
24-h avg
24-h avg
24-h avg
Hourly
Daily 1-h max
Daily 1-h max
Daily 1-h max
Daily 1-h max
Daily 1-h max
Daily 1-h max
Daily 1-h max
8-h avg
Location
Los Angeles, CA
Atlanta, GA
Barcelona, Spain
Boston, MA
Atlanta, GA
(Cold Season)
Atlanta, GA
(Warm Season)
Edmonton, Canada
Sydney, Australia
8 US Cities
8 Canadian Cities
Phoenix, AZ
Atlanta, GA
Linz, Austria
Correlation
Measure
Spearman
Spearman
Spearman
Pearson
Spearman
Spearman
Pearson
NR
NR
NR
NR
Spearman
Pearson
Pollutant
CO
NR
NR
NR
NR
0.59
0.54
0.74
0.60
NR
0.65
0.87
0.7
NR
03
NR
NR
NR
NR
0.11
0.42
NR
0.25
0.27
0.12
NR
0.44
NR
SO2
NR
NR
NR
NR
0.36
0.37
NR
0.46
NR
0.49
0.57
0.36
NR
PM2.5
0.36
(PM2.5)
0.61 (EC)
0.47(PM2.5)
0.58 (EC)
0.41 (PM2.s)
0.6 (EC)
0.38
0.37
0.36
NR
0.65
NR
0.53
0.77
0.47
(PM2.5)
0.64 (EC)
0.62 (OC)
0.54
PM10
NR
NR
NR
0.33
0.46
0.44
NR
0.48
NR
0.53
0.53
0.53
0.62
November 2013
2-75
DRAFT: Do Not Cite or Quote
-------
Table 2-4 (Continued): Synthesis of NO2 ambient-ambient copollutant correlations reported in the literature.
„ . .. Pollutant
Oorrplation
Study
Sarnatetal. (2012)
Greenwald et al. (2013)
Katanoda et al. (2011)
Dongetal. (2011)
Hwang and Lee (2010)
McConnell et al. (2003)
Time Location
96-h avg El Paso, TX
(Site A)
El Paso, TX
(Site B)
Ciudad Juarez, Mexico
(Site A)
Ciudad Juarez, Mexico
(Site B)
96-h avg 2 sites in El Paso, TX
1-yrmean Japanese Cities
1-yr avg 7 Cities across China
1-yravg 14 Taiwanese Communities
4-yr avg 12 communities in
southern California
Measure CO O3 SO2
Spearman NR NR NR
Spearman NR NR NR
Spearman NR NR NR
Spearman NR NR NR
Pearson NR NR NR
Pearson NR NR 0.76
NR 0.23 0.66 0.52
NR 0.86 -0.07 0.55
Pearson NR 0.59 NR
PM2.5
-0.39
(PM2.5)
-0.24
(PM10-2.5)
-0.28
(PM2.5)
0.04
(PM-,0-2.5)
-0.28
(PM2.5)
0.04
(PM-,0-2.5)
0
(PM2.5)
0.34
(PM-,0-2.5)
0.05
(PM2.5)
0.02
(PM-,0-2.5)
NR
NR
0.37
0.54
PM10
-0.3
-0.1
-0.1
0.11
0.01
NR
0.70
NR
0.20
BC = Black carbon; EC = Elemental carbon; OC = Organic level; TC = Total carbon; UFP = Ultrafine particles; PNC = Particle Number Concentration
November 2013
2-76
DRAFT: Do Not Cite or Quote
-------
c
03
o
CL
CO
O3
SO2
PM2.5
PM2.5-10
PM10
BC
UFP
n ^ n nn rTfi'ff^ifflm
u w u uu (_ jja 2iwJ m^j
^ /~,^i ^ ,-Y-V-, r.#^^ ^ ^,^^ rA?7v> ^-x^ ^7-^^^f^ r-, ,-%
D ^ (T5TD (H) no ^K-^ •yywwwc" npOTD ^
U v^ UiUJ UJ U (TO
^ v3i^uL-^ vjWaA viaa) \^ u/
/^P7\ r^~\ f~, ~\f~\ ~~\ /T\
() C) ()
-1.0
T
-0.5 0.0
Correlation Coefficient
0.5
1.0
Note: Boxes represent the interquartile range of the data with the median line plotted, and 90th and 10th percentile of the data are
plotted as the whiskers. Original data are plotted as red markers.
Source: NCEA analysis of data from studies referenced in Table 2-4.
Figure 2-19 Summary of copollutant correlation coefficients reported in
studies in Table 2-4.
i
2
o
J
4
5
6
2.6.4.2 Personal and Indoor Relationships between NO2 and
Copollutants
Many studies have investigated the relationship between personal and ambient
measurements of NO2 and other pollutants to evaluate the use of central site
measurements as a proxy for personal exposure to pollution. Other studies have explored
relationships between indoor NO2 and copollutants to understand sources and personal
exposure in an indoor environment. Table 2-5. Table 2-6. Table 2-7. and Table 2-8
present correlations of ambient, personal, or indoor NO2 with similar measurements of
November 2013
2-77
DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
copollutants. Similar to the results in the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
2008c). moderate correlations were generally observed between personal NO2 and
personal or ambient measurements of other regional (PM2 5) and traffic-related pollutants
(e.g., EC, OC). Additionally, O3 consistently showed a negative or no correlation with
NO2 due to complex chemistry. More recent studies report indoor NO2 copollutant
correlations and observe moderate correlations between NO2 and EC.
Table
Study
2-5 Pearson correlation
copollutants.
Delfino et al. (2008a)
Suh and
Williams
Zanobetti(2010b)
etal. (2012a)
Schembari et al. (2013)
Location
Los Angeles, CA
Atlanta, GA
Chapel Hill, NC
coefficients
N
<170
<277
<357
Barcelona, Spain < 65
between ambient NO2 and personal
Averaging times
All: 24-h
All: 24-h
All: 24-h
NO2: 7-days;
PM2.5/EC: 2-days
PM2.5
0.32
0.25
-0.19
0.21
EC
0.2
0.33
-0.17
0.44
OC O3
0.16
-0.09
-0.01
-
Table
Study
2-6 Pearson correlation
copollutants.
Delfino et al. (2008a)
Suh and
Williams
Zanobetti(2010b)
etal. (2012a)
Schembari et al. (2013)
Location
Los Angeles, CA
Atlanta, GA
Chapel Hill, NC
Barcelona, Spain
coefficients
N
<170
<277
<326
<65
between personal NO
Averaging times
All: 24-h
All: 24-h
All: 24-h
NO2: 7-days;
PM2.5/EC: 2-days
PM2.5
0.21
0.2
0.33
0.28
2 and
EC
0.2
0.22
-0.3
0.22
ambient
OC O3
0.18
-
-0.26
-
Table
Study
2-7 Pearson correlation
copollutants.
Delfino et al. (2008a)
Suh and
Williams
Zanobetti(2010b)
etal. (2012a)
Schembari et al. (2013)
Location
Los Angeles, CA
Atlanta, GA
Chapel Hill, NC
Barcelona, Spain
coefficients
between personal NO
N Averaging times
<486
<277
<326
All: 24-h
All: 24-h
All: 24-h
N02: 7-days;
PM2.5/EC: 2-days
PM2.5
0.38
0.29
0.06
0.11
2 and
EC
0.22
0.49
0.33
0.3
personal
OC 03
0.2
-0.03
-0.11
-
November 2013
2-78
DRAFT: Do Not Cite or Quote
-------
Table 2-8 Correlation coefficients between indoor NO2 and indoor
copollutants.
Study
Sarnatetal. (2012)a
Location
El Paso, TX
(Site A)
El Paso, TX
(Site B)
Averaging
N times
.f. NC>2: 4-days;
PM2.5/EC:2-days
1E. NO2: 4-days;
PM2.5/EC:2-days
PM2.5 EC OC O3
-0.35(PM2.5)
-0.26(PMio-2.s) 0.58
-0.19 (PMio)
0.06 (PM2.s)
0.28(PM10-2.5) -0.37
0.12 (PM10)
Ciudad Juarez, Mexico .,. NO2:4-days;
(Site A) PM2.5/EC:2-days
-0.29 (PM2.s)
-0.58(PMio-2.s) 0.66
-0.5 (PMio)
Ciudad Juarez, Mexico 1C- NO2:4-days;
(SiteB) PM2.5/EC:2-days
-0.04(PM2.5)
-0.5 (PM10-2.s) 0.45
-0.34 (PMio)
Greenwald et al. (2013)b 2 sites in El Paso, TX 18-26 All: 4-days
0.76(PM2.5)
0.83 (PMio)
0.45
aSpearman correlation
bPearson correlation
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
In addition to these findings, higher correlations were typically observed between
ambient measurements of NO2 and other traffic-related pollutants (see Section 2.6.4.1)
compared to personal measurements (e.g., correlations among personal exposure
measurements in Table 2-7) (Schembari et al., 2013; Williams et al., 2012; Suh and
Zanobetti. 201 Ob; Delfino et al.. 2008a). For example, Suh and Zanobetti (201 Ob)
observed a stronger relationship between ambient NO2-EC (r = 0.61) and ambient
NO2-PM2 5 (r = 0.47) compared to personal NO2-EC (r = 0.49) and personal NO2-PM2 5
(r = 0.29). Delfino et al. (2008a) observed similar results in the NO2-EC relationship in a
health study investigating the relationship between traffic-related pollution and lung
function decrements in Los Angeles, CA. While the ambient NO2-EC correlation was
moderate (r = 0.61), lower correlations were observed for personal NO2-EC (r = 0.22).
Weaker correlations observed between personal measurements of NO2 and other traffic-
related pollutants (compared to ambient measurement correlations) suggest that personal
exposure to NO2 may include a number of outdoor and indoor sources comprising traffic
and non-traffic emissions (e.g., gas stoves, residential wood burning, biomass burning).
Additionally, personal exposures are influenced by building air exchange rate and time-
activity patterns that differ among study participants. This is in contrast to ambient NO2
November 2013
2-79
DRAFT: Do Not Cite or Quote
-------
1 concentrations, which appear to be largely driven by variability in traffic pollution in
2 many areas. This type of exposure error is discussed in more detail in Section 2.6.5.
3 Few studies have reported indoor NO2 copollutant correlations, focusing on correlations
4 between NO2 and PM in different size fractions as well as NO2 and BC. In these studies,
5 moderate correlations are typically observed between indoor NO2 and BC; however, less
6 consistent correlations are observed for indoor NO2 and PM. Sarnat etal. (2012)
7 measured indoor concentrations of NO2, BC, PM2 5, PMi0.2 5, and PMi0 at four
8 elementary schools in two cities near the US-Mexico border: El Paso, TX and Ciudad
9 Juarez, Mexico. While correlations between NO2 and BC were generally moderate (r =
10 -0.37-0.66), NO2 and PM showed weaker, inverse correlations at all four elementary
11 schools (r = -0.58-0.12). Greenwald et al. (2013) later conducted a follow-up study to
12 Sarnat etal. (2012) and measured similar pollutants at the same schools in El Paso Texas.
13 Although Greenwald et al. (2013) reported similar NO2/BC correlations to those reported
14 in Sarnat et al. (2012). stronger correlations were observed between NO2 and PM2 5
15 (r = 0.76) and NO2 and PM10 (r = 0.83). Differences in the NO2 and PM correlations
16 between these two studies reflect that NO2 and PM can have many different sources in
17 indoor environments, which impact their temporal and spatial patterns.
18 A small number of studies have used NO2 in receptor models to relate health effects to
19 sources/factors. Mar etal. (2000) used factor analysis to apportion PM mass collected in
20 Phoenix, AZ and found high NO2 loadings on the motor vehicle exhaust factor. Cakmak
21 et al. (2009) applied factor analysis to apportion PM2 5 in Santiago, Chile and found a
22 motor vehicle exhaust factor with high loadings of NO2. Halonen et al. (2009) applied the
23 EPA positive matrix factorization (PMF) method
24 (http://intranet.epa.gov/heasd/products/pmf/pmf.htm) on PM2 5 data from Helsinki,
25 Finland and reported a traffic emissions factors with high loadings of NO2. Similarly,
26 Baxter etal. (2013) conducted PCA analysis using PM2 5 data from eight U.S. cities and
27 found NO2 associated with a traffic related factor in Boston, MA only.
2.6.4.3 NO2 Concentration as an Indicator of Source-Based
Mixtures
28 Health studies often use NO2 concentration as a surrogate for exposure to traffic pollution
29 mixtures when measurements of other pollutants are not available, because NO2
30 concentration is routinely measured at sampling sites nationwide and is a prevalent
31 component of vehicle exhaust. Section 2.5.2 concluded that NO2 generally correlates
32 spatially with other traffic-related pollutants in urban areas.
November 2013 2-80 DRAFT: Do Not Cite or Quote
-------
1 Several recent studies have evaluated the use of central-site NOX or NO2 concentration
2 as a surrogate for personal exposure to traffic pollution mixtures. In a near-road
3 environment, NOX concentration can be correlated with pollutants that are also associated
4 with health effects, including UFP and water soluble metals (Sanchez Jimenez et al.
5 2012). polycyclic aromatic hydrocarbons (PAHs) (Brook et al.. 2007). BTEX
6 (Beckerman et al.. 2008). and EC (Minguillon et al.. 2012). The moderate-strong
7 correlation between CO, NOX, and EC concentrations forms the basis for a proposed
8 multipollutant mobile source indicator that combines these three species into an
9 Integrated Mobile Source Indicator (IMSI) for traffic related air pollution as a weighted
10 average of their concentrations by the ratio of mobile source to total emissions for each
11 pollutant (Pachon et al.. 2012):
IMSIEB
EmissionECmobile -, EmissionNOxmobile EmissionCOmobile
EmissionECtota[ EC EmissionNOxtota[ N0x Emissioncotota[ c°
EmissionECmohile EmissionNOxmohile EmissionCOmohile
EmissionECtota[ EmissionNOxtota[ Emissioncotota[
Equation 2-10
12 Compared to other common traffic surrogates (PM2 5 and CO), NOX concentration may
13 better capture spatial and temporal trends of traffic pollution. Wheeler et al. (2008).
14 Beckerman et al. (2008). and Karner et al. (2010) reported strong correlations among
15 NO 2 and several traffic-related pollutants, including benzene and toluene, at various
16 distances from a roadway. These studies concluded that gradients in NO2 concentrations
17 were spatially correlated with gradients in traffic-related pollution. Brook et al. (2007)
18 demonstrated that benzo(e)pyrene and hopanes, specific mobile source tracers, were more
19 strongly correlated with NO2 (r = 0.27-0.80) compared to PM2 5 (r = 0.26-0.62) at several
20 urban sites in Canada. Sanchez Jimenez et al. (2012) observed similar findings in a
21 spatial variability study in Glasgow and London, which included measurements of
22 several criteria pollutant traffic surrogates (e.g., CO, PMi0, PM2 5) at roadside and
23 background sites. Of all the surrogates, NOX and NO2 concentrations showed the
24 strongest intra-site and inter-site (roadside site versus background site) correlations with
25 particle number count and some water-soluble metal species (Cu and Ni).
26 Although NO2 tends to correlate with most roadway pollutants in a near-road
27 environment, the NO2 concentration gradient tends to be shallower than gradients for
28 other primary traffic-related pollutants (e.g., CO, UFP). For example in Beckerman et al.
29 (2008). peak near-road NO2 concentrations were only 2 times higher than the urban
30 background concentration, defined here as the lowest concentration measured upwind of
31 the road. In contrast, peak near-road UFP counts were 23 times higher than the urban
November 2013 2-81 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
background concentration. These results suggest that, although NO2 may capture many
aspects of pollutant gradients from the roadway, NO2 concentration used as a marker for
traffic may underestimate the magnitude of the concentration gradient for other near-road
pollutants, such as UFP and CO. Figure 2-20 presents the spatial variability of NO2 and
copollutants at various gradients from the roadway reported in Karner etal. (2010) to
compare the spatial near-road gradient of NO2, NO, and NOX concentrations with those
of other traffic-related pollutants (Beckerman et al.. 2008).
s
O)
o
ro
J2
TJ
I
a
E
o
c
e
o
c
0)
o
c
o
o
c
I
"5
0.
100
200
300
400
6 -
5 -
4 -
3 -
2 -
0 -
Rapid: >50% drop by 150 m
CO (11)
— — — Metal deposition (15)
UF1 particle no. (44)
X
s
N .
N
- -_ _ —
Less rapid or gradual decay/change
Benzene (21)
EC (51)
HO (20)
\
\
\
' '\ ' • "~
-S^,
N0x(15)
Ozone (20)
• PM,0(39)
UF2no. (71)
VOC1 (64)
'•' •-."^"^--~-
Mo trend
Fine particle no, (16)
PM25(49)
VOC2 (24)
0 100 200 300 400
100
200
300
400
Distance from edge (m)
Note: NO2, NO, and NOX concentration gradients are presented in the center panel.
Data presented from Karner etal. (2010) were synthesized from 41 peer reviewed references.
Source: Reprinted with permission of the American Chemical Society, Karner et al. (2010).
Figure 2-20 Spatial variability in concentrations of near-road pollutants,
including NO2, NOX, CO, PM2.5, and UFP. NO2, NO, and NOX
concentration gradients are presented in the center panel.
9
10
11
12
13
14
15
16
As described in Section 2.3. other sources contributing to ambient NOX concentrations
include non-road mobile sources, electric generating units, industrial sources, and
wildfires. Non-road mobile sources, such as airports, shipping ports, and rail yards, can
contribute substantially to local and regional ambient NOX concentrations (Kim et al..
201 Ib: Williams et al.. 2009: Vutukuru and Dabdub. 2008: Carslaw et al.. 2006: Unal et
al.. 2005). At shipping ports and airports, traffic from ground-level support activities can
also contribute a large portion to NOX emissions from these sources (Klapmeyer and
Marr. 2012: Kim et al.. 20 lib). Outside of urban centers where traffic is not a dominant
source, other sources of NOX may include wildfires and residential wood-burning. As
November 2013
2-82
DRAFT: Do Not Cite or Quote
-------
1 such, NOX concentration may not always be a reliable proxy for traffic pollution. Section
2 2.3 discusses different sources of NOX in more detail.
2.6.5 Considerations for Use of Exposure Metrics in Epidemiology
3 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) examined several factors
4 influencing exposure to ambient oxides of nitrogen and measurements used to represent
5 exposures. These include high spatial and temporal variability of NO2 concentrations in
6 urban areas and near roads, location of NO2 samplers, and ventilation of indoor
7 microenvironments. The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) concluded
8 that errors associated with the use of NO2 concentrations measured at central site
9 monitors as exposure metrics for epidemiology studies tended to bias the health effect
10 estimate towards the null. The following sections provide evidence from recent studies to
11 support that conclusion.
2.6.5.1 Personal-Ambient Relationships
12 In most epidemiologic studies of the health effects of NO2, the health effect endpoint is
13 modeled as a function of ambient exposure, Ea. Equation 2-2. Equation 2-3. Equation
14 2-4. Equation 2-5. and Equation 2-6 define Ea as the product of ambient concentration,
15 Ca, and a, a term encompassing time-weighted averaging and infiltration of NO2 (Section
16 2.6.1). Community time-series epidemiologic studies capturing the exposures and health
17 outcomes of a large cohort frequently use the concentration measured at a central site
18 monitor, Ca!Csm as a surrogate for Ea in an epidemiologic model (Wilson et al.. 2000).
19 When averaging across individuals, a can quantify the bias introduced by substituting
20 Ca)Csm for the average exposure to ambient NO2, Ea. Personal measurements typically
21 capture both ambient and nonambient exposure contributions; for the purpose of this
22 document, these are referred to as "total personal exposure" measurements, even though
23 time activity data is not always incorporated into their computation. The 2008 ISA for
24 Oxides of Nitrogen (U.S. EPA. 2008c) concluded that literature relating ambient NO2
25 concentrations measured by a central site monitor to personal NO2 exposures was mixed,
26 with some studies finding statistically significant associations and other studies finding
27 no statistically significant association. These inconsistencies reflected various factors that
28 influence exposure in respective studies, including proximity and strength of sources of
29 NOX, spatiotemporal variability of NO2 concentrations, and time-activity behavior of the
30 exposed sample population. See Table 2-9 and Table 2-10 for data from recent studies
31 including ambient, outdoor, and indoor NO 2 concentration data; total personal NO2
32 exposure concentration data; and correlations among measurements. In some cases,
November 2013 2-83 DRAFT: Do Not Cite or Quote
-------
1 median or average personal exposure measures were comparable to median or average
2 central site monitor concentrations, but in other cases, median or average total personal
3 NO2 exposures and median or average ambient NO2 concentrations varied considerably.
4 Personal NO2 concentration measurements tended to be more highly correlated with
5 indoor concentrations compared with outdoor or ambient concentrations. Personal-
6 outdoor correlations were also higher for summer compared with winter; this is not
7 surprising, because open windows during summer likely increase exposure to outdoor
8 NO 2. Implications for use of central site monitoring data in epidemiologic studies are
9 discussed in Section 2.6.5.3.
10 Even when the median or average total personal NO2 exposures and ambient
11 concentrations were comparable, the total personal exposure measurements and central
12 site monitor concentrations might not have always been correlated. For example,
13 Williams et al. (2012a) measured total personal NO2 exposures for the Detroit Exposure
14 and Aerosol Research Study (DEARS) population of non-smoking adults and found that
15 total personal NO2 exposure was not statistically significantly associated with NO2
16 measured at central site monitors. Likewise, Suh and Zanobetti (201 Ob) measured low
17 correlation between total personal exposure and central site NO2 measurements among an
18 Atlanta panel of 30 adults. Together, these results indicate that most of the total personal
19 NO2 exposure measurements for these studies were influenced by either spatially variable
20 NO2 not well detected by the central site monitor or by nonambient sources.
November 2013 2-84 DRAFT: Do Not Cite or Quote
-------
Table 2-9 Ambient, outdoor, transport, indoor, and personal NO2 measurements (ppb) across studies.
Study
Sarnatetal. (2012)
Williams et al. (2012b):
Menq et al. (2012a)
Suh and Zanobetti
(2010b)
Brown et al. (2009)
Delfino et al. (2008a)
Location
El Paso, TX
(large city)
Ciudad Juarez,
Mexico (large city)
Detroit, Ml
(large city)
Metropolitan
Atlanta, GA
(large city)
Metropolitan
Boston, MA
(large city)
Riverside, CA;
Whittier, CA
(SoCAB)
(large city)
Sampling
Time Period Interval
January-May, 96-h
2008
Summer, 24-h
2004-2007
Winter,
2004-2007
Fall, 1999- 24-h
Spring, 2000
November, 24-h
1999-January,
2000
June-July,
2000
July-Dec, 2003 24-h
(Riverside);
July-Dec, 2004
(Whittier)
Ambient Outdoor Transport
14.0-20.6b 4.5-14.2b
18.7-27.2b
Williams:
22. Oa;
Meng:
22.0a; 22.7b
24.0a; 23.9b
17.96a;
17.13b
25.8C; 26.8b
22. Oc; 22. 8b
25.3a; 25.0b
Indoor Personal
4.0-8. 1b
23.1-120.8b
Total:
Williams:
25.5b;
Meng: 25.4b
Ambient:
16.0a;21.0b
Total:
24.0a; 35.6b
Ambient:
18.0a;20.4b
8.08a; 11.60b
10.4C; 12.9b
13.9C; 17.4b
26.7a; 28.6b
November 201:
2-85
DRAFT: Do Not Cite or Quote
-------
Table 2-9 (Continued): Ambient, outdoor, transport, indoor, and personal NO2 measurements (ppb) across studies.
Study
Delqado-Saborit (2012)
Kornartit et al. (2010)
Location
Birmingham, U.K.
(large city)
Hertfordshire, U.K.
(greater London
area) (large city)
Time Period
July-October,
2011
Winter, 2000
Sampling
Interval Ambient
5-min 47b
7-days
Outdoor Transport
64b Car: 40b
Bus: 71b
Bike: 125b
Train: 58b
~
Indoor
Office: 14b
Home: 17b
Electric
oven:
Bedroom:
Personal
All: 23b
Gas oven:
31b
Electric
oven: 19b
Electric
oven: 8.1b
Gas oven:
Summer, 2001
7.8b
Living room:
7.9b
Kitchen: 7.1b
Gas oven:
Bedroom:
10.8b
Living room:
13.7b
Kitchen:
20.6b
11.2b
Electric
oven:
Bedroom:
12.7b
Living room:
13.1b
Kitchen:
11.Ob
Gas oven:
Bedroom:
14.3b
Living room:
14.7b
Kitchen:
14.2b
Electric
oven: 13.3b
Gas oven:
14.6b
November 2013
2-86
DRAFT: Do Not Cite or Quote
-------
Table 2-9 (Continued): Ambient, outdoor, transport, indoor, and personal NO2 measurements (ppb) across studies.
Sampling
Study Location Time Period Interval
Leeetal. (2013) Seoul, Korea July, 2008 NR
(large city)
January, 2009 NR
Daegu, Korea July, 2008 NR
(mid-sized city)
January, 2009 NR
Asan, Korea July, 2008 NR
(small city)
January, 2009 NR
Ambient Outdoor Transport Indoor
29.5C; 30.7b - - Home:
24.4C; 25.7b
Work: 19.2C;
21. 5b
29.5c;31.1b Home:
20.9C; 24.9b
Work: 27. 9C;
29.9b
19.9c;21.1b - - Home:
19.3c;20.3b
Work: 21. 3C;
22.8b
23.0C; 24.3b Home:
23.3c;25.1b
Work: 20. 3C;
22.9b
26.0C; 27.9b - - Home:
23.8C; 24.9b
Work: 21. 1C;
25.6b
21.6c;23.9b Home:
20.3C; 22.9b
Work: 13.0C;
18.6b
Personal
25. 3C;
27b
22.5C; 24.2b
21.4c;22.6b
20.3c;21.7b
22.6C; 24.3b
19.9c;22.3b
November 2013
2-87
DRAFT: Do Not Cite or Quote
-------
Table 2-9 (Continued): Ambient, outdoor, transport, indoor, and personal NO2 measurements (ppb) across studies.
Study
Leeetal. (2013)
(Continued)
Phvsicketal. (2011)
Sampling
Location Time Period Interval Ambient Outdoor
Suncheon, Korea July, 2008 NR 15.0C; 15.9b
(rural)
January, 2009 NR 12.5C; 15.2b
Total July, 2008 NR 21. 7C; 23.7b
January, 2009 NR 20.6C; 23.6b
Melbourne, May, 2006; Ambient: 6:00 p.m.to
Australia June, 2006; 1 h; Personal: 8:00 a.m:
(large city) April, 2007; Participants 19. 8a; 18. 7b
May 2007 wore two sets 8'00 a m to
of passive 6:00 pm:
samplers. 20 3a' 21 2b
One was
worn for 48 h.
One was
worn only
during the
hours spent
at home, at
work, in
transit, or
while
performing
other
activities.
Transport Indoor Personal
Home: 14.0C; 16.3b
13.0C; 14.3b
Work: 12.0C;
14.5b
Home: 12.9C; 15.7b
15.9c;20.4b
Work: 9.3C;
12.9b
Home: 20.5C; 22.6b
19.5c;21.2b
Work: 18.4C;
21. 4b
Home: 18.6c;21.0b
19.9c;23.3b
Work: 16.4C;
21. 1b
Home: Total: 12.2f
17.2a;16.8b Home:8.2f
Work: 21. 6a; Work: 14./
21 7
Transit: 23.4f
Other: 17.4f
November 2013
2-8
DRAFT: Do Not Cite or Quote
-------
Table 2-9 (Continued): Ambient, outdoor, transport, indoor, and personal NO2 measurements (ppb) across studies.
Study
Sahsuvaroqlu et al.
(2009)
Schembari et al. (2013)
Mollov et al. (2012)
Peqasetal. (2012)
Location
Hamilton, Canada
(mid-sized city)
Barcelona, Spain
(large city)
Melbourne,
Australia
(large city)
Aveiro, Portugal
(small city center,
suburb)
Sampling
Time Period Interval Ambient
October, 2003 72-h
May, 2004
August, 2004
Total
November, 7-day
2008 and
October, 2009
August, 2008- 7-day
December,
2008; January,
2009-April,
2009
April-June, 7-day
2010
Outdoor Transport
All: 32.0b
Non-ETS:
31. 7b
All: 17.6b
Non-ETS:
16.8b
All: 9.7b
Non-ETS:
9.6b
All: 19.3b
Non-ETS:
18.9b
18.7c'e;
1g4b,e
9.5a; 10.0b
City center:
10.5b'e;
Suburb:
mlb.e
Indoor
All: 22.4b
Non-ETS:
21. 9b
All: 13.5b
Non-ETS:
12.3b
All: 8.2b
Non-ETS:
7.4b
All: 14.4b
Non-ETS:
13.6b
19.2c'e;
206b,e
7.9a; 8.4b
City center:
7_41>.d.e;
Suburb:
6_gb.d.e
Personal
All: 23.3b
Non-ETS:
22.4b
All: 14.4b
Non-ETS:
14.0b
All: 8.8b
Non-ETS:
8.2b
All: 15.2b
Non-ETS:
14.6b
17.7c'e;
18.6b'e
amed
bavg
°geo. mean
dAveraged over 4 classrooms and 2 weeks.
eReported in |jg/m3 and converted to ppb assuming 25 °C and 760 mmHg.
'Data provided by the authors for Figure 1 of Physick et al. (2011).
November 2013
2-89
DRAFT: Do Not Cite or Quote
-------
Table 2-10 Correlations between measured
monitors.
Study
Sarnatetal. (2012)a
Williams et al. (2012a)a
Suh and Zanobetti (2010b)a
Brown et al. (2009)
Delfino et al. (2008a)
Delqado-Saborit (2012)
Leeetal. (201 3)b
Ph\/cii~k at al fOITI "H
Location
Ciudad Juarez, Mexico
and
El Paso, Texas
Wayne County,
Michigan
Atlanta, GA
Boston, MA
2 Southern California Cities
Birmingham, U.K.
Seoul, South Korea
Daegu, South Korea
Asan, South Korea
Suncheon, South Korea
All 4 Cities
NO2 concentrations from personal, indoor, outdoor,
Personal-Ambient
—
All Subjects: 0.11
Vest-Compliant
(>60%)c: 0.14
0.12
Winter: 0.00
Summer: 0.03
0.43
1-h NO2: 0.024
Sampling event NC>2:
0.15
—
—
—
—
Outdoor-Personal Indoor-Personal
—
—
—
Summer: 0.39 Summer: 0.50
Winter: 0.47 Winter: 0.55
Summer: 0.43 Summer: 0.32
Winter: 0.47 Winter: 0.59
Summer: 0.62 Summer: 0.63
Winter: 0.11 Winter: 0.37
Summer: 0.46 Summer: 0.46
Winter: 0.56 Winter: 0.60
Summer: 0.58 Summer: 0.60
Winter: 0.53 Winter: 0.55
and ambient
Indoor-Outdoor
CJ-A: 0.36
CJ-B: 0.92
EP-A: 0.66
EP-B: 0.01
—
—
—
Summer: 0.71
Winter: 0.22
Summer: 0.65
Winter: 0.57
Summer: 0.67
Winter: 0.37
Summer: 0.77
Winter: 0.80
Summer: 0.78
Winter: 0.55
November 201:
2-90
DRAFT: Do Not Cite or Quote
-------
Table 2-10 (Continued): Correlations between measured NO2 concentrations from personal, indoor, outdoor, and ambient monitors.
Study
Location
Personal-Ambient
Outdoor-Personal
Indoor-Personal
Indoor-Outdoor
Sahsuvaroqlu et al. (2009)
Lake Ontario, Canada
(Winter)
Lake Ontario, Canada
(Spring)
Lake Ontario, Canada
(Summer)
Lake Ontario, Canada
(All Seasons)
All Subjects: 0.002
Non-ETS: 0.020
All Subjects: 0.430
Non-ETS: 0.283
All Subjects: 0.233
Non-ETS: 0.187
All Subjects: 0.589
Non-ETS: 0.599
All Subjects: 0.067
Non-ETS: 0.011
All Subjects: 0.822
Non-ETS: 0.783
All Subjects: 0.517
Non-ETS: 0.540
All Subjects: 0.729
Non-ETS: 0.693
Schembari et al. (2013)a
Barcelona, Spain
0.58
0.78
0.53
"Spearman coefficient.
bPearson coefficient.
""Subjects wore the sampling vests at least 60% of the sampling period.
November 2013
2-91
DRAFT: Do Not Cite or Quote
-------
2.6.5.2 Factors Influencing Exposure Measurement Error
1 Estimates of NO2 exposures are subject to errors that can vary in nature. Classical error is
2 defined as error scattered around the true personal exposure and independent of the
3 measurement. Classical error results in bias of the true Et. Berkson error is defined as
4 error scattered around the exposure surrogate (i.e., central site monitor measurement) and
5 independent of the true value (Goldman et al.. 2011; Reeves etal. 1998).When an
6 epidemiology study is performed, nonambient contributions are thought to introduce
7 Berkson error into the Et term that does not bias epidemiologic effect estimates for
8 ambient NO2 assuming that nonambient NO2 sources are independent of ambient sources
9 but does cause the confidence intervals around effect estimates to widen (Sheppard.
10 2005: Wilson etal. 2000).
11 Recent studies suggest that exposure error is a combination of Berkson-like and classical-
12 like errors and depend on how exposure metrics are averaged across space. Goldman et
13 al. (2011) simulated the effect of classical-like and Berkson-like errors due to
14 spatiotemporal variability among ambient or outdoor air pollutant concentrations over a
15 large urban area on estimates of ED visits for cardiovascular disease. The relative risk
16 (RR) per ppm was negatively biased in the case of classical-like error (1-h max NO2:
17 -1.3%; 1-h max NOX: 1.1%) and negligibly positively biased in the case of Berkson-like
18 error (1-h max NO2: 0.0042%; 1-h max NOX: 0.0030%). Conversely, the 95%
19 confidence interval range for RR per ppm was wider for Berkson-like error (1-h max
20 NO2: 0.028; 1-h max NOX: 0.023) compared with classical-like error (1-h max NO2:
21 0.0025; 1-h max NOX: 0.0043). This is in agreement with previous findings
22 (e.g..Sheppard et al.. 2005; Zeger et al.. 2000). that Berkson error tends to widen the
23 confidence interval of the effect estimate, while classical error tends to bias the health
24 effect estimate.
25 Several studies have investigated factors that influence the relationship between personal
26 exposure measurements and ambient concentrations. Meng et al. (2012b) performed a
27 random effects meta-analysis of 15 studies that calculated slopes and correlations
28 between personal NO2 measurements of ET and ambient NO2 concentrations for 32
29 sample populations, of which 17 were from pooled analyses, 8 were from longitudinal
30 analyses, and 7 were from daily average analyses. Meta-regression results are shown in
31 Table 2-11 and were reported to be statistically significant. Meng et al. (2012b) found
32 that the associations depended on several factors, including season, age, pre-existing
33 disease, and potentially, sampling artifacts and local sources. Bellander et al. (2012)
34 measured personal NO2 exposure and modeled it as a function of NO2 concentrations
35 measured at an urban area, at a rural area, at a roadside, and outside of the participants'
November 2013 2-92 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
homes and places of work in Stockholm County, Sweden. They observed slopes ranging
from 0.25-0.37 (R2 = 0.01-0.20) with all models being statistically significant except that
for rural NO2 (p = 0.18). Factors influencing the association between Cacsm and Ea
include spatiotemporal gradients in ambient NO2 concentrations, instrument error,
housing characteristics such as air conditioning usage (Sheppard et al.. 2005). and
uncertainty in time-activity data (Isaacs et al.. 2013). These factors are described in the
following subsections.
Table 2-11 Meta-regression results from 15 studies examining the relationship
between personal exposure measurements and ambient
concentrations.
Slope
Correlation
Slope
Correlation
Study design
Based on original studies
Corrected for publication bias
Pooled3
0.40
0.42
0.30
0.37
Longitudinal
0.14
0.16
0.14
0.16
Daily average0
0.29
0.72
0.20
0.45
"Pooled analyses: Piechocki-Minguv et al. (2006). Linn et al. (1996). Liardetal. (1999).Gauvin et al. (2001) , Aim et al. (1998).
Brown et al. (2009). Sarnat et al. (2006). Delfino et al. (2008a)
""Longitudinal analyses: Sarnat et al. (2005). Sarnat et al. (2001). Sarnat et al. (2000). Linaker et al. (2000). Kim et al. (2006a).
Koutrakis et al. (2005)
°Daily average analyses: Mukala et al. (2000). Liard et al. (1999). and Aim et al. (1998)
Source: Meng et al. (2012b).
9
10
11
12
13
14
15
16
17
18
19
20
Spatial Variability of Ambient NO2 Concentrations
Recent studies have explored the effect of spatial exposure measurement error on health
effect estimates. Goldman et al. (2010) simulated spatial exposure measurement error
based on a semivariogram function across monitor sites with and without temporal
autocorrelation at one- and two-day lags incorporated into the analysis to analyze the
influence of spatiotemporal variability among ambient or outdoor concentrations over a
large urban area on ED visits for cardiovascular disease. A random error term was
calculated from the semivariogram and added to a base case time series to simulate
estimates of population exposure to NO2 concentrations subject to spatial error in 1,000
Monte Carlo simulations. For the analysis with autocorrelation considered, RRper ppm
for 1-h max NO2 dropped to 1.0046 (p = 0.10), and RR per ppm for 1-h max NOX
dropped to 1.0079 (p = 0.026) in comparison with the base case RRper ppm = 1.0139
(p = 9 x 10"6). When autocorrelation was not considered, RR per ppm dropped to 1.0044
(p = 0.12) for l-hmaxNO2 and 1.0074 (p = 0.032) for l-hmaxNOx. Goldman et al.
November 2013
2-93
DRAFT: Do Not Cite or Quote
-------
1 (2010) results suggest that spatial exposure measurement error results in biasing the
2 health effect estimate towards the null. Sarnat et al. (2010) studied the spatial variability
3 of concentrations of NO2, along with CO, O3, and PM2 5, in the Atlanta, GA,
4 metropolitan area and how spatial variability affects interpretation of epidemiologic
5 results, using time-series data for circulatory disease ED visits. Sensitivity to spatial
6 variability was examined at slightly greater than neighborhood scale (8 km) in this study.
7 Interestingly, Sarnat et al. (2010) found that relative risk varied with distance between the
8 monitor and study population when comparing urban to rural locations, but distance of
9 the study population to the monitor was not an important factor when comparing urban
10 population groups. This suggests that, even for spatially heterogeneous NO2, urban scale
11 concentration measures may produce results comparable to neighborhood-scale
12 concentration measures if the sites were comparable throughout the city, for example, as
13 a result of similar traffic patterns. However, Sarnat et al. (2010) caution that, because
14 their study was limited to 8 km radii, it is not possible to interpret this work with respect
15 to near-road and on-road microscale concentrations. In a study of the effect of
16 concentration metric choice (central site, arithmetic average across space, or population-
17 weighted average) used to represent exposure in a time-series epidemiologic model,
18 Strickland et al. (2011) found that choice of the concentration metric resulted in large
19 differences in the observed associations between ED visits for pediatric asthma and
20 exposure for spatially heterogeneous NO2 but not for spatially homogeneous PM2 5.
21 Spatial resolution of the personal exposure estimates has been evaluated in recent studies.
22 Szpiro et al. (2011) explored the effect of specification of spatial conditions in a health
23 model by comparing bias and prediction accuracy for the health effect estimate using
24 correctly specified and misspecified exposure simulation conditions. The Szpiro et al.
25 (2011) simulations were for a generic air pollutant, but the results are relevant to NO2
26 exposure. Land use regression (LUR) calculations were used to simulate exposure, and
27 correct specification was considered when three spatial covariates were included in the
28 model; the misspecified model omitted a covariate. Although prediction accuracy was
29 higher for the correctly specified model (R2 = 0.73 to 0.74 versus R2 = 0.49 to 0.50 for
30 the misspecified model), magnitude of bias in the effects estimate was also slightly higher
31 for the correctly specified model (bias = -0.007 to -0.035 compared with bias = -0.001 to
32 0.001 for the misspecified model). The results of Szpiro etal. (2011) suggested that use
33 of more spatially resolved personal exposure metrics does not necessarily decrease the
34 magnitude of bias in effect estimates. However, given the small magnitude of bias noted
35 in either case, it is possible that the spatial error was primarily Berkson in nature.
36 Goldman et al. (2012) also studied the effect of different types of spatial averaging on
37 bias in the health effect risk ratio and the effect of correlation between measured and
38 "true" ambient concentrations of NO2, NOX, and other air pollutant measures to analyze
November 2013 2-94 DRAFT: Do Not Cite or Quote
-------
1 the influence of spatiotemporal variability among ambient or outdoor concentrations over
2 a large urban area on health effect estimates; see Table 2-12. Specifically, Goldman et al.
3 (2012) examined the correlations between exposure measurement errors with measured
4 and true exposures, where true exposures were estimated from chemical transport
5 modeling (CTM) for a cohort living in the 20-county Atlanta metropolitan area; see Table
6 2-12. Here, they compared exposure measurement error metrics among exposure
7 estimates obtained from using a central site monitor, an average of monitors distributed
8 throughout the region where the study population lives, a population-weighted average,
9 an area-weighted average, and population-weighted average computed with
10 concentrations modeled using CTM, at 5 km resolution.
11 Exposure error was simulated in Goldman et al. (2010). They observed that the exposure
12 measurement error was somewhat correlated with both the measured and true values,
13 reflecting both Berkson-like and classical-like error components. For the central site
14 monitor, the exposure measurement errors were somewhat anti-correlated with the true
15 value but had relatively higher positive correlation with the measured value. For the other
16 sites, the exposure measurement errors were anti-correlated with the true value, while
17 they had positive but lower magnitude correlation with the measured value. At the same
18 time, the exposure measurement bias, given by the ratio of the exposure measurement
19 error to the measured value, was much higher in magnitude at the central site monitor
20 than for the other measurement methods for NO2 and for NOX concentrations with the
21 exception of the area-weighted average, which produced a large negative exposure
22 measurement bias. These findings suggest more Berkson-like error in the more spatially
23 resolved exposure metrics and more classical-like error in the central site monitor;
24 accounting for spatial variation in NOX concentrations across the population likely
25 resulted in improved accuracy of the exposure metric. Correlations between measured
26 values at the central site monitor and the CTM were typically lower than for the
27 monitoring average techniques.
November 2013 2-95 DRAFT: Do Not Cite or Quote
-------
Table 2-12 Exposure measurement error metrics for comparing central site
monitoring data and various monitor averages compared with values
computed from a CTM.
Pollutant
Bias[(Z - Z*)/Z]a
R2(Z, Z*)b
R[(Z-Z*),Z*]
R[(Z-Z*),Z]
NO2
Central site monitor
Unweighted average
Population-weighted average
Area-weighted average
Chemical transport model -
population weighted average
0.62
0.25
0.18
-0.07
N/A
0.24
0.38
0.38
0.38
0.45
-0.46
-0.73
-0.78
-0.87
-0.82
0.61
0.20
0.14
-0.04
0.0017
NOX
Central site monitor
Unweighted average
Population-weighted average
Area-weighted average
Chemical transport model -
population-weighted average
0.71
0.31
0.03
-0.88
N/A
0.33
0.45
0.46
0.47
0.52
-0.11
-0.63
-0.81
-0.96
-0.80
0.81
0.29
0.02
-0.31
-0.00042
More: Z denotes the measured concentration, and Z* denotes the true concentration, considered here to be from the CTM. Bias in
the exposure metric is given as the proportion of error between the measurement and true value to the measurement.
aData provided by the authors for Figure 5 of Goldman et al. (2012).
bData provided by the authors of Figure 4 of Goldman et al. (2012).
Source: Goldman et al. (2012).
Uncertainty in Time-Activity Data
1 Large-scale human activity databases, such as those developed for the Consolidated
2 Human Activity Database (CHAD) including the National Human Activity Pattern
3 Survey (NHAPS), have been designed to characterize exposure patterns among much
4 larger population subsets than can be examined during individual panel studies (Klepeis
5 et al.. 2001). CHAD consists of a consolidation of human activity data obtained during
6 several panel studies in which diary or retrospective activity data were obtained for
7 metropolitan, state-wide, or nationwide samples (Graham and McCurdy. 2004). CHAD is
8 intended to provide data to reproduce particular study conditions but does not contain a
9 distribution that mimics the U.S. population given that it contains multiple location-
10 specific panel studies; NHAPS by itself does contain a sufficient data distribution to
11 provide for national inference (Graham and McCurdv. 2004).
12 The complex human activity patterns across the population (all ages) for NHAPS are
13 illustrated in Figure 2-21 (Klepeis etal. 2001). This figure is presented to illustrate the
November 2013 2-96 DRAFT: Do Not Cite or Quote
-------
1 diversity of daily activities among the entire population as well as the proportion of time
2 spent in each microenvironment. Different time-activity patterns have been found when
3 analyzing activity patterns for different populations or life stages. For example, Wu et al.
4 (2010) observed activity patterns for a panel of adults and children from communities
5 with larger percentages of non-whites (85%) and those below the poverty line (33%)
6 compared with NHAPS. The study participants spent more time outdoors compared with
7 the nationwide cohort (3.8 hours versus 1.8 hours nationally); note that Wu et al. (2010)
8 undersampled participants ages 65 + years, and the median age of the population studied
9 in Wu etal. (2010) was 27 years compared with 35 years nationwide. Other recent time-
10 activity studies have included working adults (Bellander et al., 2012; Kornartit et al..
11 2010). pregnant women (Iniguez et al.. 2009). adolescents (deCastro et al.. 2007). and
12 children (Molter et al.. 2012; Xue et al.. 2004). In many cases, the time activity data were
13 limited to residential, occupational, and outdoor location categories to simplify
14 assignment of concentrations to which the subjects were exposed in each
15 microenvironment.
16 Recently, Kornartit et al. (2010) tested the associations between time-weighted exposure
17 estimates from area samples with personal sampling measurements for a London, U.K.
18 panel study. Kornartit et al. (2010) measured NO2 concentration in several outdoor and
19 indoor microenvironments for 55 subjects aged 21-60 years and correlated a time-
20 weighted average of those microenvironmental NO2 concentration measurements with
21 personal NO2 concentration measurements. They observed a slope of 0.94 for the
22 relationship between time-weighted average and personal NO2 concentrations (R2 = 0.85)
23 in winter and a slope of 0.59 (R2 = 0.65) in summer. Higher levels of NO2 were observed
24 for both time-weighted average and personal concentrations in summer compared with
25 winter. The authors concluded that the time-weighting approach provided a reasonable
26 approximation of personal exposure but sometimes underestimated it.
November 2013 2-97 DRAFT: Do Not Cite or Quote
-------
>
13
s
o
p.
a;
a>
CU
DH
a S R
73 TO CS
ooooooooooooooooooooooo
OOOOOCpsOOOOCDOOOOOOOOOOOO
i/~; MD C-- 00 C\ O —< CN
Ol ^H
to
t-
CO O1.
Time of Day
Source:Reprinted with permission of Nature Publishing Group, Klepeiset al. (2001).
Figure 2-21 Distribution of time sample population spends in various
environments, from the U.S. National Human Activity Pattern
Survey (all ages).
i
2
o
J
4
5
6
Variability in time-activity data also presents a source of uncertainty in the exposure
model, particularly for the effect of exposure misclassification on the a term. Isaacs et al.
(2013) performed a time-activity study of a panel of eight adults (four men and four
women living around Research Triangle Park, NC) of similar demographic and
socioeconomic groups and observed statistically significant inter- and intra-individual
variability that could potentially add uncertainty to exposure predictions.
9
10
11
12
Error and Uncertainty Related to Infiltration
Given that people spend the majority of their time indoors, building air exchange rates
influence exposure to ambient NO2. In an analysis of NO2 data from the Detroit
Exposure and Aerosol Research Study (DEARS), Meng et al. (2012a) observed seasonal
differences, with statistically significant slopes of 0.24 (p <0.001) for ET versus CajCsm
and of 0.13 (p = 0.033) for Ea versus CajCsm for summer measurements. For winter
measurements, the slopes were not statistically significant (ET versus CajCsm: slope = 0.08,
November 2013
2-98
DRAFT: Do Not Cite or Quote
-------
1 p = 0.10; Ea versus Cacsm: slope = 0.07, p = 0.33). Meng etal. (2012a) found that high air
2 exchange rate (>1.3 air changes per hour), no central air conditioning, use and non-use of
3 window fans, and presence of old carpeting were statistically significant determinants of
4 a forNO2 (p <0.05) in summer; none of these factors were statistically significant
5 determinants of a for NO2 in winter. Molter etal. (2012) calculated associations with
6 time spent in several home, transit, and school microenvironments for a cohort of 12-13
7 year-old children from Greater Manchester, U.K. and observed that time spent in transit
8 was positively and statistically significantly associated with prediction error of a
9 microenvironmental model of personal NO2 exposure (p = 0.01). In Molter et al. (2012).
10 outdoor exposures were calculated with LUR, while indoor exposures were calculated
11 using the INDAIR model that accounts both for infiltration due to home ventilation
12 characteristics and indoor sources. Sensitivity to air exchange rate of INDAIR predictions
13 of indoor NO2 in the absence of indoor sources underscores potential for bias and
14 uncertainty in a, which depends on air exchange rate, penetration, and indoor deposition
15 (Dimitroulopoulou et al.. 2006). Sarnatet al. (2013a) tested if air exchange rate acted as
16 an effect modifier of NOX exposure on asthma emergency department visits and observed
17 an effect in both interaction and stratified models. Because the Sarnat et al. (2013a) paper
18 treated air exchange as an effect modifier rather than a source of error, it is discussed
19 further in Section 2.6.5.3.
Instrument Error
20 Exposure measurement error related to instrument precision has a smaller effect
21 compared with error related to spatial gradients in the concentration. Goldman et al.
22 (2010) investigated instrument precision error at locations where ambient monitors were
23 co-located. Instrument precision error increased with increasing concentration and was
24 observed to exhibit some autocorrelation at one- and two-day lags. A random error term
25 based on observations from co-located monitors was added to a base case time series to
26 simulate population estimates for ambient air concentrations subject to instrument
27 precision error in 1,000 Monte Carlo simulations. Very little change in risk ratios and
28 significance levels was observed for 1-h max NO2 and 1-h max NOX concentrations. For
29 1-h max NO2 concentration, the RR per ppm of NO2 concentration with simulated
30 instrument precision error was 1.0133 (p = 2.1 * 10"5) compared with RR per ppm =
31 1.0139(p = 9x 10"6) for the base case. For 1-h max NOX concentration with simulated
32 instrument precision error, RR per ppm = 1.0132 (p =1.8 x 10"5) compared with the base
33 case of 1.0139 (p = 9.0 x 10"6). Although statistically significant, the amount of exposure
34 measurement bias related to instrument precision was very small.
November 2013 2-99 DRAFT: Do Not Cite or Quote
-------
2.6.5.3 Implications for Epidemiology
The model of human health effects related to ambient NO2 exposure is of the form:
Equation 2-11
Where (30 = model intercept, (3i = effect estimate for the ambient exposure, Ea = ambient
exposure, Z = covariate vector, (3Z = vector of slope related to each covariate, and
8 = random error
5 Recognizing the relationship between Ea and Ca described in Section 2.6.1. Equation
6 2-11 can also be written as:
Y =
Equation 2-12
7 Here, a = exposure factor, as described in Section 2.6.1 and Ca]Csm = ambient
8 concentration measured at a central site monitor, as defined in Section 2.6.5.1. Hence,
9 two metrics of interest for exposure assessment studies can be considered: Ea, personal
10 exposure to ambient NO2, and Ca>csm, NO2 concentration measured at a central site
1 1 monitor to represent Ea .
12 For long-term epidemiologic studies of human exposure to NO2, where the magnitude of
13 the concentration is of most interest, Ea may be the most appropriate metric. If CajCsm is
14 then used as a surrogate for Ea, then a can be considered to encompass the exposure
15 measurement error related to uncertainties in the spatial distribution of NO2, time activity
16 data, air exchange rate, or instrument precision, as described in detail in Section 2.6.5.2.
17 Ca>csm may be an acceptable (i.e., minimally biased) surrogate for Ea if the central site
18 monitor is located in close proximity to the entire study population (e.g., in a dense urban
19 setting) or if there are either few or well-dispersed localized NO2 sources such that the
20 influence of spatial variability is minimized. There is limited information regarding the
21 influence of near-road exposures on exposure measurement error and if CajCsm may be a
22 biased exposure surrogate when representing people's exposure in the near-road
23 environment for epidemiologic studies of long-term exposure.
24
November 2013 2-100 DRAFT: Do Not Cite or Quote
-------
1 For community time-series epidemiology studies of short-term exposure, the temporal
2 variability in concentration is of primary importance to relate to variability in the health
3 effect estimate (Zeger et al., 2000). Magnitude of the concentration is less important in
4 this case, so Ca>csm can be an acceptable surrogate if the central site monitor captures the
5 temporal variability of the true air pollutant concentration even if the magnitude of
6 concentration is biased. Additionally, for studies involving thousands of participants, it is
7 not feasible to measure personal exposures. For this reason, Cacsm is typically employed
8 to represent the community average concentration as measured at the central site monitor.
9 Typically, it is assumed that Cacsm is a surrogate for Ea; see U.S. EPA (2008c) and
10 studies cited therein as well as studies cited throughout this section.
11 Consideration of errors in use of CajCsm from central site monitoring data as a surrogate
12 for Ea is for the purpose of assessing the impact of this substitution on health effect
13 estimates in community time-series studies of short-term exposure. Recently, Setton et al.
14 (2011) investigated how spatial variability and unaccounted study participant mobility
15 bias effect estimates in short-term epidemiologic models of NO2 exposure in southern
16 California and Vancouver, British Columbia. In this case, a monitor was placed at each
17 participant's home, and bias increased in magnitude towards the null with distance from
18 home and time spent away from home. Moreover, when spatial variability increased
19 (through comparison of spatially variable LUR-derived NO2 concentrations with a
20 smoother monitor-based approach for mapping NO2), the effect estimate in the monitor-
21 based approach was more biased towards the null. Similarly, Van Roosbroeck et al.
22 (2008) evaluated effect estimates for the influence of NO2 on four respiratory outcomes
23 among children obtained with the NO2 data from a single monitor located at the
24 children's school in an epidemiologic study of short-term exposure. The effect estimates
25 were compared with those obtained from personal NO2 monitoring to capture spatial
26 variability in NO2 concentrations and time-activity data. Van Roosbroeck et al. (2008)
27 observed that effect estimates were biased towards the null by one-third to one-half when
28 using a single monitor. The results of Setton etal. (2011) and Van Roosbroeck et al.
29 (2008) imply that failure to capture spatial variability in ambient NO2 exposures can lead
30 to biasing the effect estimate towards the null in time-series epidemiologic studies of
31 short-term exposure. This is in agreement with the primary conclusion of the 2008 ISA
32 for Oxides of Nitrogen U.S. EPA (2008c).
33 In the model formulation presented in Equation 2-12. a may be considered an effect
34 modifier that interacts with Ca csm rather than contributing error to Ea. For example,
35 Sarnat et al. (2013a) developed an effect modification model that used al5 air exchange
36 rate, as the effect modifier of exposure to NOX, where a; is defined in Section 2.6.1. The
37 effect estimate was positive ((3 = 1.9) and statistically significant forNOx (p = 0.04) for
38 the model interaction term but not for the linear concentration and a; terms, suggesting
November 2013 2-101 DRAFT: Do Not Cite or Quote
-------
1 effect modification by a;. Further evidence of effect modification comes from an analysis
2 in which the model (Equation 2-11) was stratified by low and high a^ positive
3 associations different from the linear estimate were observed for both low and high al5
4 with a stronger and statistically significant association for the high al term. Given that
5 most people spend upwards of 90% of their time indoors (Sarnat et al., 2013a). a; is a
6 major determinant of an individual's a. Exposure measurement errors affecting a, as
7 described in Section 2.6.5.2. would add error to the effect modification term but not to
8 the linear concentration term, as would be the case for Equation 2-11. The Sarnat et al.
9 (2013a) paper was the first to apply the concept of effect modification to study health
10 effects associated with NOX exposure.
2.7 Summary and Conclusions
11 NOX concentrations have generally decreased over the past 20 years and this trend
12 continues. However, new diesel control technologies cause a greater proportion of NOX
13 to be present as NO2 rather than NO. Moreover, diesel vehicles have become a relatively
14 more important source as emissions from other major sources have decreased. Annual
15 average NO2 concentrations remain well below NAAQS levels, but 1-hour daily
16 maximum levels appear to be exceeded in several near road studies. Background levels
17 are much lower than ambient concentrations. In urban areas, NOX and NO2
18 concentrations exhibit a high degree of spatial variability, which presents difficulties for
19 estimating exposure. Measurement of NO and NO2 concentrations is also a challenge,
20 and methods are under development for measuring true NO2 concentration that improve
21 interference problems associated with the current federal reference method.
22 Concentrations are especially high near local sources, especially roadways with heavy
23 traffic. Influence of distance from roadways is especially pronounced for NO, with a
24 weaker influence of NO2 concentrations.
25 Although total personal exposure to NO2 includes ambient and nonambient components,
26 this assessment is focused on the ambient component of personal NO2 exposure, because
27 it is relevant to review of the NAAQS. Personal exposure to ambient NO2 can be
28 estimated by a variety of techniques. These include models (e.g., dispersion models, land
29 use regression models, and stochastic population exposure models), personal exposure
30 measurements, and use of central site NO2 concentration measurements as exposure
31 surrogates. NO2 exposure estimates are subject to error. These errors are influenced by
32 spatial NO2 concentration variability, time-activity data, air exchange characteristics of
33 microenvironments, and accuracy and precision of instrumentation. Central site NO2
34 concentration may be acceptable (i.e., minimally biased) for use as a surrogate for
35 ambient NO2 exposure in epidemiology studies of long-term exposure if the central site
November 2013 2-102 DRAFT: Do Not Cite or Quote
-------
1 monitor is located in close proximity to the entire study population (e.g., in a dense urban
2 setting) or if there are either few or well-dispersed localized NO2 sources such that the
3 influence of spatial variability is minimized. Central site NO2 measurements may be an
4 acceptable NO2 exposure surrogate for community time-series epidemiology studies if
5 the central site monitor concentration captures the temporal variability of the true
6 personal exposure to ambient NO2. Recent time-series epidemiology studies of short-
7 term exposure evaluating the effect of using a single monitor to represent exposure to
8 ambient NO2 demonstrate that use of a single monitor results in health effect estimates
9 that are biased towards the null. This is in agreement with the findings of the 2008 ISA
10 for Oxides of Nitrogen (U.S. EPA. 2008c). The effect of near-road exposures on
11 adequacy of exposure estimates from central site monitors is not yet well characterized
12 for epidemiologic studies.
November 2013 2-103 DRAFT: Do Not Cite or Quote
-------
References for Chapter 2
Abolhasani. S: Frev. HC: Kim. K; Rasdorf. W: Lewis. P; Pang. SH. (2008). Real-world in-use activity, fuel
use, and emissions for nonroad construction vehicles: A case study for excavators. J Air Waste Manag
Assoc 58: 1033-1046. http://dx.doi.Org/10.3155/1047-3289.58.8.1033
Adamkiewicz. G: Hsu. HH; Vallarino. J; Melly. SJ: Spengler. JD; Lew. JI. (2010). Nitrogen dioxide
concentrations in neighborhoods adjacent to a commercial airport: A land use regression modeling
study. Environ Health Global Access Sci Source 9: 73. http://dx.doi.org/10.1186/1476-069X-9-73
Aguilera. I; Guxens. M; Garcia-Esteban. R; Corbella. T; Nieuwenhuijsen. MJ; Foradada. CM; Sunyer. J.
(2009). Association between GIS-based exposure to urban air pollution during pregnancy and birth
weight in the INMA Sabadell cohort. Environ Health Perspect 117: 1322-1327.
Aguilera. I; Sunyer. J: Fernandez-Patier. R; Hoek. G: Aguirre-Alfaro. A; Meliefste. K; Bomboi-Mingarro.
MT; Nieuwenhuijsen. MJ: Herce-Garraleta. D; Brunekreef. B. (2008). Estimation of outdoor NOx,
NO2, and BTEX exposure in a cohort of pregnant women using land use regression modeling. Environ
Sci Technol 42: 815-821. http://dx.doi.org/10.1021/es0715492
Ainslie. B; Steyn. DG: Su. J: Buzzelli. M; Brauer. M; Larson. T; Rucker. M. (2008). A source area model
incorporating simplified atmospheric dispersion and advection at fine scale for population air pollutant
exposure assessment. Atmos Environ 42: 2394-2404. http://dx.doi.Org/10.1016/j.atmosenv.2007.12.021
Akagi. SK; Craven. JS: Taylor. JW: McMeeking. GR; Yokelson. RJ: Burling. IR; Urbanski. SP; Wold. CE;
Seinfeld. JH; Coe. H; Alvarado. MJ: Weise. DR. (2012). Evolution of trace gases and particles emitted
by a chaparral fire in California. Atmos Chem Phys 12: 1397-1421. http://dx.doi.org/10.5194/acp-12-
1397-2012
Allen. RW: Amram. O; Wheeler. AJ; Brauer. M. (2011). The transferability of NO and NO2 land use
regression models between cities and pollutants. Atmos Environ 45: 369-378.
http://dx.doi.0rg/10.1016/i.atmosenv.2010.10.002
Aim. S: Mukala. K; Pasanen. P; Tiittanen. P; Ruuskanen. J: Tuomisto. J: Jantunen. MJ. (1998). Personal
NO2 exposures of preschool children in Helsinki. J Expo Anal Environ Epidemiol 8: 79-100.
Alvarado. MJ: Logan. JA; Mao. J: Apel. E; Riemer. D; Blake. D; Cohen. RC: Min. KE; Perring. AE;
Browne. EC: Wooldridge. PJ: Diskin. GS: Sachse. GW: Fuelberg. H; Sessions. WR; Harrigan. PL;
Huev. G: Liao. J: Case-Hanks. A; Jimenez. JL; Cubison. MJ: Vav. SA; Weinheimer. AJ: Knapp. DJ:
Montzka. DP; Flocke. FM; Pollack. IB; Wennberg. PO; Kurten. A; Crounse. J: St. Clair. JM; Wisthaler.
A; Mikoviny. T; Yantosca. RM; Carouge. CC: Le Sager. P. (2010). Nitrogen oxides and PAN in plumes
from boreal fires during ARCTAS-B and their impact on ozone: An integrated analysis of aircraft and
satellite observations. Atmos Chem Phys 10: 9739-9760. http://dx.doi.org/10.5194/acp-10-9739-2010
Andersen. ZJ: Loft. S: Ketzel. M; Stage. M; Scheike. T; Hermansen. MN; Bisgaard. H. (2008a). Ambient
air pollution triggers wheezing symptoms in infants. Thorax 63: 710-716.
http://dx.doi.org/10.1136/tfax.2007.085480
Andersen. ZJ: Wahlin. P: Raaschou-Nielsen. O; Scheike. T; Loft. S. (2007). Ambient particle source
apportionment and daily hospital admissions among children and elderly in Copenhagen. J Expo Sci
Environ Epidemiol 17: 625-636. http://dx.doi.org/10.1038/sj.jes.7500546
Arhami. M; Polidori. A; Delfino. RJ: Tjoa. T; Sioutas. C. (2009). Associations between Personal, Indoor,
and Residential Outdoor Pollutant Concentrations: Implications for Exposure Assessment to Size-
Fractionated Paniculate Matter. J Air Waste Manag Assoc 59: 392-404. http://dx.doi.org/10.3155/1047-
3289.59.4.392
November 2013 2-104 DRAFT: Do Not Cite or Quote
-------
Atari. DO; Luginaah. I; Xu. X; Fung. K. (2008). Spatial variability of ambient nitrogen dioxide and sulfur
dioxide in Sarnia, "Chemical Valley," Ontario, Canada. J Toxicol Environ Health A 71:
1572-1581. http://dx.doi.org/10.1080/15287390802414158
Baldauf. R; Thoma. E; Hays. M; Shores. R; Kinsev. J: Gullett. B; Kimbrough. S: Isakov. V; Long. T; Snow.
R; Khlystov. A; Weinstein. J; Chen. FL; Seila. R; Olson. D; Gilmour. I; Cho. SH; Watkins. N; Rowley.
P; Bang. J. (2008a). Traffic and meteorological impacts on near-road air quality: Summary of methods
and trends from the Raleigh near-road study. J Air Waste Manag Assoc 58: 865-878.
http://dx.doi.Org/10.3155/1047-3289.58.7.865
Baldauf. R: Thoma. E; Khlystov. A; Isakov. V; Bowker. G: Long. T; Snow. R. (2008b). Impacts of noise
barriers on near-road air quality. Atmos Environ 42: 7502-7507.
http://dx.doi.0rg/10.1016/i.atmosenv.2008.05.051
Baldauf. R: Watkins. N; Heist. D; Bailey. C: Rowley. P; Shores. R. (2009). Near-road air quality
monitoring: Factors affecting network design and interpretation of data. Air Qual Atmos Health 2: 1-9.
http://dx.doi.org/10.1007/sll869-009-0028-0
Basagana. X; Rivera. M; Aguilera. I; Agis. D: Bouso. L; Elosua. R: Foraster. M; de Nazelle. A;
Nieuwenhuijsen. M; Vila. J: Kuenzli. N. (2012). Effect of the number of measurement sites on land use
regression models in estimating local air pollution. Atmos Environ 54: 634-642.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.01.064
Baxter. LK; Duvall. RM; Sacks. J. (2013). Examining the effects of air pollution composition on within
region differences in PM(2.5) mortality risk estimates. J Expo Sci Environ Epidemiol 23: 457-465.
http://dx.doi.org/10.1038/ies.2012.114
Beckerman. B; Jerrett. M; Brook. JR; Verma. DK; Arain. MA; Finkelstein. MM. (2008). Correlation of
nitrogen dioxide with other traffic pollutants near a major expressway. Atmos Environ 42: 275-290.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.09.042
Beelen. R; Voogt. M; Duyzer. J: Zandveld. P: Hoek. G. (2010). Comparison of the performances of land
use regression modelling and dispersion modelling in estimating small-scale variations in long-term air
pollution concentrations in a Dutch urban area. Atmos Environ 44: 4614-4621.
http://dx.doi.0rg/10.1016/i.atmosenv.2010.08.005
Bellander. T; Wichmann. J: Lind. T. (2012). Individual exposure to NO2 in relation to spatial and temporal
exposure indices in Stockholm, Sweden: The INDEX study. PLoS ONE 7: e39536.
http://dx.doi.org/10.1371/iournal.pone.0039536
Berrocal. VJ: Gelfand. AE; Holland. DM. (2010a). Abivariate space-time downscaler under space and time
misalignment. Ann Appl Stat 4: 1942-1975. http://dx.doi.org/10.1214/10-AOAS351
Berrocal. VJ: Gelfand. AE; Holland. DM. (201 Ob). A spatio-temporal downscaler for output from numerical
models. Journal of Agricultural, Biological, and Environmental Statistics 15: 176-197.
http://dx.doi.org/10.1007/sl3253-009-0004-z
Biazar. AP; McNider. RT. (1995). Regional estimates of lightning production of nitrogen oxides. J Geophys
Res Atmos 100: 22861-22874. http://dx.doi.org/10.1029/95JD01735
Binkowski. FS: Arunachalam. S: Adelman. Z; Pinto. JP. (2007). Examining photolysis rates with a
prototype online photolysis module in CMAQ. J Appl Meteor Climatol 46: 1252-1256.
http://dx.doi.Org/10.1175/JAM2531.l
Boersma. KF; Eskes. HJ; Brinksma. EJ. (2004). Error analysis for tropospheric NO2 retrieval from space. J
Geophys Res 109: D04311. http://dx.doi.org/10.1029/2003JD003962
Bond. DW: Zhang. RY; Tie. XX; Brasseur. G: Huffines. G: Orville. RE; Boccippio. DJ. (2001). NOx
production by lightning over the continental United States. J Geophys Res Atmos 106: 27701-27710.
Brandenberger. S: Krocher. O; Tissler. A; Althoff. R. (2008). The state of the art in selective catalytic
reduction of NOx by ammonia using metalexchanged zeolite catalysts. Catalysis Reviews: Science and
Engineering 50: 492-531. http://dx.doi.org/10.1080/01614940802480122
November 2013 2-105 DRAFT: Do Not Cite or Quote
-------
Brauer. M; Koutrakis. P; Keeler. GJ: Spengler. JD. (1991). Indoor and outdoor concentrations of inorganic
acidic aerosols and gases. JAir Waste Manag Assoc 41: 171-181.
http://dx.doi.org/10.1080/10473289.1991.10466834
Brauer. M; Ryan. PB; Suh. HH; Koutrakis. P; Spengler. JD; Leslie. NP; Billick. IH. (1990). Measurements
of nitrous acid inside two research houses. Environ Sci Technol 24: 1521-1527.
http://dx.doi.org/10.1021/es00080a011
Briggs. D. (2005). The role of GIS: coping with space (and time) in air pollution exposure assessment. J
Toxicol Environ Health A 68: 1243-1261. http://dx.doi.org/10.1080/15287390590936094
Britter. RE; Hanna. SR. (2003). Flow and dispersion in urban areas. Annu Rev Fluid Mech 35: 469-496.
http://dx.doi.org/10.1146/annurev.fluid.35.101101.161147
Brixev. LA; Heist. DK; Richmond-Bryant. J: Bowker. GE; Perry. SG: Wiener. RW. (2009). The effect of a
tall tower on flow and dispersion through a model urban neighborhood: Part 2: Pollutant dispersion. J
Environ Monit 11: 2171-2179. http://dx.doi.org/10.1039/B907137g
Brook. JR; Burnett. RT; Dann. TF; Cakmak. S: Goldberg. MS: Fan. X; Wheeler. AJ. (2007). Further
interpretation of the acute effect of nitrogen dioxide observed in Canadian time-series studies. J Expo
Sci Environ Epidemiol 17: S36-S44. http://dx.doi.org/10.1038/sj.jes.7500626
Brown. KW: Sarnat. JA; Suh. HH; Coull. BA; Koutrakis. P. (2009). Factors influencing relationships
between personal and ambient concentrations of gaseous and paniculate pollutants. Sci Total Environ
407: 3754-3765. http://dx.doi.Org/10.1016/i.scitotenv.2009.02.016
Brown. SS: Dube. WP; Karamchandani. P; Yarwood. G: Peischl. J: Ryerson. TB; Neuman. JA; Nowak. JB;
Hollowav. JS: Washenfelder. RA; Brock. CA; Frost. GJ: Trainer. M; Parrish. DP; Fehsenfeld. FC:
Ravishankara. AR. (2012). Effects of NO x control and plume mixing on nighttime chemical processing
of plumes from coal-fired power plants. J Geophys Res 117: D07304.
http://dx.doi.org/10.1029/2011JD016954
Brown. SS: Stutz. J. (2012). Nighttime radical observations and chemistry. Chem Soc Rev 41: 6405-6447.
http://dx.doi.org/10.1039/c2cs35181a
Bruggemann. TC: Keil. FJ. (2008). Theoretical investigation of the mechanism of the selective catalytic
reduction of nitric oxide with ammonia on h-form zeolites. J Phys Chem C 112: 17378-17387.
http://dx.doi.org/10.1021/ip806674d
Bucsela. EJ; Krotkov. NA; Celarier. EA; Lamsal. LN; Swartz. WH; Bhartia. PK; Boersma. KF; Veefkind.
JP; Gleason. JF; Pickering. KE. (2013). Anew stratospheric and tropospheric NO2 retrieval algorithm
for nadir-viewing satellite instruments: Applications to OMI. Atmospheric Measurement Techniques
Discussions 6: 1361-1407. http://dx.doi.org/10.5194/amtd-6-1361-2013
Burke. JM; Zufall. MJ; Ozkaynak. H. (2001). A population exposure model for paniculate matter: Case
study results for PM2.5 in Philadelphia, PA. J Expo Anal Environ Epidemiol 11: 470-489.
http://dx.doi.org/10.1038/sj.jea.7500188
Burling. IR; Yokelson. RJ; Griffith. DWT; Johnson. TJ; Veres. P: Roberts. JM; Warneke. C: Urbanski. SP;
Reardon. J: Weise. PR; Hao. WM; De Gouw. J. (2010). Laboratory measurements of trace gas
emissions from biomass burning of fuel types from the southeastern and southwestern United States.
Atmos Chem Phys 10: 11115-11130. http://dx.doi.org/10.5194/acp-10-11115-2010
Burnett. RT; Brook. J: Dann. T; Delocla. C: Philips. O: Cakmak. S: Vincent. R; Goldberg. MS: Krewski. D.
(2000). Association between paniculate- and gas-phase components of urban air pollution and daily
mortality in eight Canadian cities. Inhal Toxicol 12: 15-39.
http://dx.doi.org/10.1080/08958370050164851
Buzica. D: Gerboles. M; Plaisance. H. (2008). The equivalence of diffusive samplers to reference methods
for monitoring O3, benzene and NO2 in ambient air. J Environ Monit 10: 1052-1059.
http://dx.doi.org/10.1039/b802260g
November 2013 2-106 DRAFT: Do Not Cite or Quote
-------
CAA. Clean Air Act, as amended by Pub. L. No. 101-549. section 109: National primary and secondary
ambient air quality standards. 42 USC § 7409 (1990b). http://www.epa.gov/air/caa/title 1 .html#ia
Cahill. TA; Cahill. TM; Barnes. DE; Spada. NJ; Miller. R. (2011). Inorganic and organic aerosols
downwind of California's Roseville Railyard. Aerosol Sci Technol 45: 1049-1059.
http://dx.doi.org/10.1080/02786826.2011.580796
Cakmak. S: Dales. RE; Vida. CB. (2009). Components of particulate air pollution and mortality in Chile.
Int J Occup Environ Health 15: 152-158. http://dx.doi.org/10.1179/107735209799195844
Cao. G: Jang. M. (2008). Secondary organic aerosol formation from toluene photooxidation under various
NOx conditions and particle acidity. Atmos Chem Phys Discuss 8: 14467-14495.
http://dx.doi.org/10.5194/acpd-8-14467-2008
Cao. G: Jang. M. (2010). An SOA model for toluene oxidation in the presence of inorganic aerosols.
Environ Sci Technol 44: 727-733. http://dx.doi.org/10.1021/es901682r
Carslaw. DC: Beevers. SD; Ropkins. K; Bell. MC. (2006). Detecting and quantifying aircraft and other on-
airport contributions to ambient nitrogen oxides in the vicinity of a large international airport. Atmos
Environ 40: 5424-5434. http://dx.doi.Org/10.1016/i.atmosenv.2006.04.062
Carslaw. DC: Ropkins. K; Laxen. D; Moorcroft. S: Marner. B; Williams. ML. (2008). Near-field
commercial aircraft contribution to nitrogen oxides by engine, aircraft type, and airline by individual
plume sampling. Environ Sci Technol 42: 1871-1876. http://dx.doi.org/10.1021/es071926a
Carslaw. N; Mota. T; Jenkin. ME; Barley. MH; Mcfiggans. G. (2012). A significant role for nitrate and
peroxide groups on indoor secondary organic aerosol. Environ Sci Technol 46: 9290-9298.
http://dx.doi.org/10.1021/es301350x
Centro di Ricerche Ambientali (Centra di Ricerche Ambientali di Padova). (2006). Radiello(R) manual.
Padova, Italy: Fondazione Salvatore Maugeri Clinica del Lavoro e Delia Riabilitazione I.R.C.C.S.,
Centro di Ricerche Ambientali. http://www.radiello.com/english/Radiello%27s%20manual%2001 -
06.pdf
Cesaroni. G: Badaloni. C: Porta. D; Forastiere. F; Perucci. CA. (2008). Comparison between various
indices of exposure to traffic-related air pollution and their impact on respiratory health in adults. Occup
Environ Med 65: 683-690. http://dx.doi.org/10.1136/oem.2007.037846
Cesaroni. G: Porta. D: Badaloni. C: Stafoggia. M; Eeftens. M; Meliefste. K; Forastiere. F. (2012). Nitrogen
dioxide levels estimated from land use regression models several years apart and association with
mortality in a large cohort study. Environ Health Global Access Sci Source 11: 48.
http://dx.doi.org/10.1186/1476-069X-ll-48
Chameides. W: Walker. JCG. (1973). A photochemical theory of tropospheric ozone. J Geophys Res 78:
8751-8760. http://dx.doi.org/10.1029/JC078i036p08751
Chen. G: Song. G: Jiang. L; Zhang. Y; Zhao. N; Chen. B; Kan. H. (2008). Short-term effects of ambient
gaseous pollutants and particulate matter on daily mortality in Shanghai, China. J Occup Health 50:41-
47. http://dx.doi.org/10.1539/ioh.50.41
Cheng. WC: Liu. CH; Leung. DYC. (2008). Computational formulation for the evaluation of street canyon
ventilation and pollutant removal performance. Atmos Environ 42: 9041-9051.
http://dx.doi.0rg/10.1016/i.atmosenv.2008.09.045
Cheng. WC: Liu. CH; Leung. DYC. (2009b). On the correlation of air and pollutant exchange for street
canyons in combined wind-buoyancy-driven flow. Atmos Environ 43: 3682-3690.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.04.054
Choi. Y; Kim. H; Tong. D; Lee. P. (2012). Summertime weekly cycles of observed and modeled NOx and
O3 concentrations as a function of satellite-derived ozone production sensitivity and land use types over
the Continental United States. Atmos Chem Phys 12: 6291-6307. http://dx.doi.org/10.5194/acp-12-
6291-2012
November 2013 2-107 DRAFT: Do Not Cite or Quote
-------
Chuang. KJ; Coull. BA; Zanobetti. A; Suh. H; Schwartz. J; Stone. PH; Litonjua. A; Speizer. FE; Gold. DR.
(2008). Paniculate Air Pollution as a Risk Factor for ST-Segment Depression in Patients With Coronary
Artery Disease. Circulation 118: 1314-1320.
http://dx.doi.org/10.1161/CIRCULATIONAHA.108.765669
Clements. AL; Jia. Y; Denbleyker. A; Mcdonald-Buller. E; Fraser. MP; Allen. DT; Collins. PR; Michel. E;
Pudota. J: Sullivan. D; Zhu. Y. (2009). Air pollutant concentrations near three Texas roadways, part II:
Chemical characterization and transformation of pollutants. Atmos Environ 43: 4523-4534.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.06.044
Cleveland. WS: Graedel. TE; Kleiner. B; Warner. JL. (1974). Sunday and workday variations in
photochemical air pollutants in New Jersey and New York. Science 186: 1037-1038.
http://dx.doi.org/10.1126/science.186.4168.1037
Cohen. MA; Adar. SD; Allen. RW: Avol. E; Curl. CL; Gould. T; Hardie. D; Ho. A; Kinnev. P; Larson. TV;
Sampson. P; Sheppard. L; Stukovsky. KD; Swan. SS: Liu. LJ; Kaufman. JD. (2009). Approach to
estimating participant pollutant exposures in the multi-ethnic study of atherosclerosis and air pollution
(MESA Air). Environ Sci Technol 43: 4687-4693. http://dx.doi.org/10.1021/es8030837
Connell. DP; Withum. JA; Winter. SE; Statnick. RM; Bilonick. RA. (2005). The Steubenville
Comprehensive Air Monitoring Program (SCAMP): Associations among fine paniculate matter, co-
pollutants, and meteorological conditions. J Air Waste Manag Assoc 55: 481-496.
http://dx.doi.org/10.1080/10473289.2005.10464631
Corbett. JJ: Fischbeck. PS: Pandis. SN. (1999). Global nitrogen and sulfur inventories for oceangoing
ships. J Geophys Res Atmos 104: 34573470. http://dx.doi.org/10.1029/1998JD100040
Croslev. DR. (1996). NOy blue ribbon panel. J Geophys Res 101: 2049-2052.
http://dx.doi.org/10.1029/95JD02276
Crutzen. P. (1973). A discussion of the chemistry of some minor constituents in the stratosphere and
troposphere. Pure Appl Geophys 106-108: 1385-1399. http://dx.doi.org/10.1007/BF00881092
Crutzen. PJ. (1970). The influence of nitrogen oxides on the atmospheric ozone content. Q J Roy Meteorol
Soc 96: 320-325. http://dx.doi.org/10.1002/qi.49709640815
Dales. RE; Cakmak. S: Vidal. CB. (2009b). Air pollution and hospitalization for headache in Chile. Am J
Epidemiol 170: 1057-1066. http://dx.doi.org/10.1093/aje/kwp217
Dales. RE; Cakmak. S: Vidal. CB. (2010). Air Pollution and hospitalization for venous thromboembolic
disease in Chile. J Thromb Haemost 8: 669-674. http://dx.doi.Org/10.llll/i.1538-7836.2010.03760.x
Dallmann. TR; Harlev. RA. (2010). Evaluation of mobile source emission trends in the United States. J
Geophys Res 115: D14305. http://dx.doi.org/10.1029/2010JD013862
De Santis. D: Allegrini. I; Fazio. MC: Pasella. D. (1996). Characterization of Indoor Air Quality in the
Church of San luigi Dei Francesi, Rome, Italy. Int J Environ Anal Chem 64: 71-81.
http://dx.doi.org/10.1080/03067319608028336
DeCaria. AJ; Pickering. KE; Stenchikov. GL; Ott. LE. (2005). Lightning-generated NOx and its impact on
tropospheric ozone production: A three-dimensional modeling study of a stratosphere-troposphere
experiment: Radiation, aerosols and ozone (STERAO-A) thunderstorm. J Geophys Res Atmos 110:
D14303. http://dx.doi.org/10.1029/2004JD005556
deCastro. BR; Sax. AN; Chillrud. SN; Kinnev. PL; Spengler. JD. (2007). Modeling time-location patterns
of inner-city high school students in New York and Los Angeles using a longitudinal approach with
generalized estimating equations. J Expo Sci Environ Epidemiol 17: 233-247.
http://dx.doi.org/10.1038/sj.jes.7500499
Delfino. RJ; Staimer. N; Tjoa. T; Gillen. D; Kleinman. MT; Sioutas. C: Cooper. D. (2008a). Personal and
ambient air pollution exposures and lung function decrements in children with asthma. Environ Health
Perspect 116: 550-558. http://dx.doi.org/10.1289/ehp.10911
November 2013 2-108 DRAFT: Do Not Cite or Quote
-------
Delgado-Saborit JM. (2012). Use of real-time sensors to characterise human exposures to combustion
related pollutants. J Environ Monit 14: 1824-1837. http://dx.doi.org/10.1039/c2eml0996d
Penman. KL; Brasseur. G: Chidthaisong. A; Ciais. P; Cox. PM; Dickinson. RE; Hauglustaine. D; Heinze.
C: Holland. E; Jacob. D; Lohmann. U; Ramachandran. S: da Silva Dias. PL; Wofsv. SC: Zhang. X.
(2007). Couplings between changes in the climate system and biogeochemistry. In Climate change
2007: The physical science basis (pp. 499-587). New York, NY: Cambridge University Press.
http://www.ipcc.ch/publications and data/publications ipcc fourth assessment report wgl report the
_physical science basis.htm
di Sabatino. S: Kastner-Klein. P; Berkowicz. R; Britter. RE; Fedorovich. (2003). The modelling of
turbulence from traffic in urban dispersion models Part I: Theoretical considerations. Environ Fluid
Mech3: 129-143.
Dijkema. MB; Gehring. U; van Strien. RT; van der Zee. SC: Fischer. P; Hoek. G: Brunekreef. B. (2011). A
comparison of different approaches to estimate small-scale spatial variation in outdoor NO
concentrations. Environ Health Perspect 119: 670-675. http://dx.doi.org/10.1289/ehp.0901818
Dimitroulopoulou. C: Ashmore. MR; Hill. MTR; Byrne. MA; Kinnerslev. R. (2006). INDAIR: a
probabilistic model of indoor air pollution in UK homes. Atmos Environ 40: 6362-6379.
http://dx.doi.0rg/10.1016/i.atmosenv.2006.05.047
Dong. GH; Chen. T; Liu. MM; Wang. D; Ma. YN; Ren. WH; Lee. YL; Zhao. YD; He. QC. (2011). Gender
differences and effect of air pollution on asthma in children with and without allergic predisposition:
Northeast Chinese children health study. PLoS ONE 6: e22470.
http://dx.doi.org/10.1371/iournal.pone.0022470
Dunlea. EJ; Herndon. SC: Nelson. DP; Volkamer. RM; San Martini. F; Sheehy. PM; Zahniser. MS: Shorter.
JH; Wormhoudt. JC: Lamb. BK; Allwine. EJ; Gaffnev. JS: Marlev. NA; Grutter. M; Marquez. C:
Blanco. S: Cardenas. B; Retama. A; Ramon Villegas. CR; Kolb. CE; Molina. IT; Molina. MJ. (2007).
Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban environment. Atmos
Chem Phys 7: 2691-2704.
Durant. JL; Ash. CA; Wood. EC: Herndon. SC: Jayne. JT; Knighton. WB; Canagaratna. MR; Trull. JB;
Brugge. D: Zamore. W: Kolb. CE. (2010). Short-term variation in near-highway air pollutant gradients
on a winter morning. Atmos Chem Phys 10: 8341-8352. http://dx.doi.org/10.5194/acp-10-8341-2010
Eeftens. M; Beelen. R; Fischer. P; Brunekreef. B; Meliefste. K; Hoek. G. (2011). Stability of measured and
modelled spatial contrasts in NO2 over time. Occup Environ Med 68: 765-770.
http://dx.doi.org/10.1136/oem.2010.061135
Esplugues. A; Ballester. F; Estarlich. M; Llop. S: Fuentes-Leonarte. V: Mantilla. E; Vioque. J: Iniguez. C.
(2011). Outdoor, but not indoor, nitrogen dioxide exposure is associated with persistent cough during
the first year of life. Sci Total Environ 409: 4667-4673.
http://dx.doi.0rg/10.1016/i.scitotenv.2011.08.007
Fang. Y; Fiore. AM; Horowitz. LW: Lew II. H; Hu. Y; Russell. AG. (2010). Sensitivity of the NOy budget
over the United States to anthropogenic and lightning NOx in summer. J Geophys Res 115: D18312.
http://dx.doi.org/10.1029/2010JD014079
Faustini. A; Stafoggia. M; Berti. G: Bisanti. L; Chiusolo. M; Cernigliaro. A; Mallone. S: Primerano. R;
Scarnato. C: Simonato. L; Vigotti. MA; Forastiere. F. (2011). The relationship between ambient
paniculate matter and respiratory mortality: A multi-city study in Italy. Eur Respir J 38: 538-547.
http://dx.doi.org/10.1183/09031936.00093710
Fehsenfeld. FC: Dickerson. RR; Hubler. G: Luke. WT; Nunnermacker. LJ; Williams. EJ: Roberts. JM;
Calvert JG: Curran. CM; Delany. AC: Eubank. CS: Fahev. DW: Fried. A; Gandrud. BW: Langford.
AO; Murphy. PC: Norton. RB; Pickering. KE; Ridley. BA. (1987). Aground-based intercomparison of
NO, NOx, and NOy measurement techniques. J Geophys Res 92: 14,710-714,722.
November 2013 2-109 DRAFT: Do Not Cite or Quote
-------
Fernandez-Somoano. A; Estarlich. M; Ballester. F; Fernandez-Patier. R; Aguirre-Alfaro. A; Herce-
Garraleta. MD; Tardona. A. (2011). Outdoor NO(2) and benzene exposure in the INMA (Environment
and Childhood) Asturias cohort (Spain). Atmos Environ 45: 5240-5246.
http://dx.doi.0rg/10.1016/i.atmosenv.2011.02.010
Fernando. HJS. (2010). Fluid dynamics of urban atmospheres in complex terrain. Annu Rev Fluid Mech 42:
365-389. http://dx.doi.org/10.1146/annurev-fluid-121108-145459
Finn. D; Clawson. KL; Carter. RG: Rich. JD; Eckman. RM; Perry. SG: Isakov. V; Heist. DK. (2010).
Tracer studies to characterize the effects of roadside noise barriers on near-road pollutant dispersion
under varying atmospheric stability conditions. Atmos Environ 44: 204-214.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.10.012
Fiore. AM; Horowitz. LW: Purves. DW: Lew. H. II; Evans. MJ; Wang. Y; Li. Q; Yantosca. RM. (2005).
Evaluating the contribution of changes in isoprene emissions to surface ozone trends over the eastern
United States. J Geophys Res 110: D12303. http://dx.doi.org/10.1029/2004JD005485
Fischer. PH; Hoek. G: Van Reeuwijk. H; Briggs. DJ: Lebret. E; Van Wijnen. JH; Kingham. S: Elliott. PE.
(2000). Traffic-related differences in outdoor and indoor concentrations of particles and volatile organic
compounds in Amsterdam. Atmos Environ 34: 3713-3722. http://dx.doi.org/10.1016/S1352-
2310(00)00067-4
Franklin. M; Vora. H; Avol. E; Mcconnell. R. ob: Lurmann. F; Liu. F; Penfold. B; Berhane. K; Gilliland. F;
Gauderman. WJ. (2012). Predictors of intra-community variation in air quality. J Expo Sci Environ
Epidemiol22: 135-147. http://dx.doi.org/10.1038/jes.2011.45
Frost. GJ: Mckeen. SA; Trainer. M; Ryerson. TB; Neuman. JA; Roberts. JM; Swanson. A; Hollowav. JS:
Sueper. DT; Fortin. T; Parrish. DP; Fehsenfeld. FC: Flocke. F; Peckham. SE; Grell. GA; Kowal. D;
Cartwright J: Auerbach. N; Habermann. T. (2006). Effects of changing power plant NO emissions on
ozone in the eastern United States: Proof of concept. J Geophys Res 111: D12306.
http://dx.doi.org/10.1029/2005JD006354
Fruin. S: Westerdahl. D; Sax. T; Sioutas. C: Fine. PM. (2008). Measurements and predictors of on-road
ultrafine particle concentrations and associated pollutants in Los Angeles. Atmos Environ 42: 207-219.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.09.057
Fu. M; Ge. Y; Tan. J: Zeng. T; Liang. B. (2012b). Characteristics of typical non-road machinery emissions
in China by using portable emission measurement system. Sci Total Environ 437: 255-261.
http://dx.doi.0rg/10.1016/i.scitotenv.2012.07.095
Fuchs. H; Dube. WP; Lerner. BM; Wagner. NL; Williams. EJ; Brown. SS. (2009). A sensitive and versatile
detector for atmospheric NO2 and NOx based on blue diode laser cavity ring-down spectroscopy.
Environ Sci Technol 43: 7831-7836. http://dx.doi.org/10.1021/es902067h
Fuentes. M. (2009). Statistical issues in health impact assessment at the state and local levels. Air Quality,
Atmosphere and Health 2: 47-55. http://dx.doi.org/10.1007/sll869-009-0033-3
Fuentes. M; Raftery. AE. (2005). Model evaluation and spatial interpolation by Bayesian combination of
observations with outputs from numerical models. Biometrics 61: 36-45.
http://dx.doi.Org/10.llll/i.0006-341X.2005.030821.x
Gair. AJ; Penkett. SA; Oyola. P. (1991). Development of a simple passive technique for the determination
of nitrogen-dioxide in remote continental locations. Atmos Environ 25A: 1927-1939.
Gauvin. S: Le Moullec. Y; Bremont. F; Momas. I; Balducci. F; Ciognard. F; Poilve. MP; Zmirou. D;
Investigators. V. (2001). Relationships between nitrogen dioxide personal exposure and ambient air
monitoring measurements among children in three French metropolitan areas: VESTA study. Arch
Environ Occup Health 56: 336-341. http://dx.doi.org/10.1080/00039890109604465
Gego. E; Gilliland. A; Godowitch. J: Trivikrama Rao. S: Porter. PS: Hogrefe. C. (2008). Modeling analyses
of the effects of changes in nitrogen oxides emissions from the electric power sector on ozone levels in
the eastern United States. J Air Waste Manag Assoc 58: 580-588. http://dx.doi.org/10.3155/1047-
3289.58.4.580
November 2013 2-110 DRAFT: Do Not Cite or Quote
-------
Gilbert. NL; Goldberg. MS: Brook. JR; Jerrett. M. (2007). The influence of highway traffic on ambient
nitrogen dioxide concentrations beyond the immediate vicinity of highways. Atmos Environ 41: 2670-
2673. http://dx.doi.0rg/10.1016/i.atmosenv.2006.12.007
Goldman. GT; Mulholland. JA; Russell. AG: Gass. K; Strickland. MJ; Tolbert. PE. (2012). Characterization
of ambient air pollution measurement error in a time-series health study using a geostatistical simulation
approach. Atmos Environ 57: 101-108. http://dx.doi.Org/10.1016/j.atmosenv.2012.04.045
Goldman. GT; Mulholland. JA; Russell. AG: Srivastava. A; Strickland. MJ: Klein. M; Waller. LA; Tolbert.
PE; Edgerton. ES. (2010). Ambient air pollutant measurement error: characterization and impacts in a
time-series epidemiologic study in Atlanta. Environ Sci Technol 44: 7692-7698.
http://dx.doi.org/10.1021/esl01386r
Goldman. GT; Mulholland. JA: Russell. AG: Strickland. MJ: Klein. M; Waller. LA; Tolbert. PE. (2011).
Impact of exposure measurement error in air pollution epidemiology: Effect of error type in time-series
studies. Environ Health Global Access Sci Source 10: 61. http://dx.doi.org/10.1186/1476-069X-10-61
Gonzales. M; Myers. O: Smith. L; Olvera. HA; Mukerjee. S: Li. WW: Pingitore. N; Amava. M; Burchiel.
S: Berwick. M. (2012). Evaluation of land use regression models for NO2 in El Paso, Texas, USA. Sci
Total Environ 432: 135-142. http://dx.doi.Org/10.1016/i.scitotenv.2012.05.062
Gordon. M; Staebler. RM; Liggio. J; Li. SM; Wentzell. J: Lu. G: Lee. P; Brook. J. R. (2012). Measured and
modeled variation in pollutant concentration near roadways. Atmos Environ 57: 138-145.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.04.022
Graham. SE; McCurdv. T. (2004). Developing meaningful cohorts for human exposure models. J Expo
Anal Environ Epidemiol 14: 23-43. http://dx.doi.org/10.1038/sj.jea.7500293
Greenwald. R; Sarnat. SB; Ravsoni. AU; Li. WW: Johnson. BA; Stock. TH; Holguin. F; Sosa. T; Sarnat.
JA. (2013). Associations between source-indicative pollution metrics and increases in pulmonary
inflammation and reduced lung function in a panel of asthmatic children. Air Qua! Atmos Health 6: 487-
499. http://dx.doi.org/10.1007/sll869-012-0186-3
Griffiths. PT; Badger. CL; Cox. RA; Folkers. M; Henk. HH; Mentel. TF. (2009). Reactive uptake of N2O5
by aerosols containing dicarboxylic acids. Effect of particle phase, composition, and nitrate content. J
Phys Chem A 113: 5082-5090. http://dx.doi.org/10.1021/ip8096814
Gruzieva. O: Bergstrom. A; Hulchiy. O; Kull. I; Lind. T; Melen. E; Moskalenko. V; Pershagen. G:
Bellander. T. (2013). Exposure to air pollution from traffic and childhood asthma until 12 years of age.
Epidemiology 24: 54-61. http://dx.doi.org/10.1097/EDE.Ob013e318276clea
Gu. Z; Zhang. Y; Lei. K. (2010). Large eddy simulation of flow in a street canyon with tree planting under
various atmospheric instability conditions. Science China Technological Sciences 53: 1928-1937.
httrj://dx.doi.org/10.1007/sll431-010-3243-x
Guo. Y; Jia. Y; Pan. X; Liu. L; Wichmann. HE. (2009). The association between fine particulate air
pollution and hospital emergency room visits for cardiovascular diseases in Beijing, China. Sci Total
Environ 407: 4826-4830. http://dx.doi.Org/10.1016/i.scitotenv.2009.05.022
Hagenbjork-Gustafsson. A; Eriksson. K; Forsberg. B. (2009). Field Evaluation of the Ogawa Diffusive
Sampler for NO(2)/NO(x) in a Cold Climate. Epidemiology 20: S161-S161.
http://dx.doi.org/10.1097/01.ede.0000362548.83420.d7
Hagler. GSW: Tang. W: Freeman. MJ: Heist. DK; Perry. SG: Vette. AF. (2011). Model evaluation of
roadside barrier impact on near-road air pollution. Atmos Environ 45: 2522-2530.
http://dx.doi.0rg/10.1016/i.atmosenv.2011.02.030
Halonen. JI; Lanki. T; Yli-Tuomi. T; Tiittanen. P; Kulmala. M; Pekkanen. (2009). Particulate air pollution
and acute cardiorespiratory hospital admissions and mortality among the elderly. Epidemiology 20: 143-
153. http://dx.doi.org/10.1097/EDE.Ob013e31818c7237
November 2013 2-111 DRAFT: Do Not Cite or Quote
-------
Hameed. S: Pinto. JP; Stewart. RW. (1979). Sensitivity of the predicted CO-OH-CH4 perturbation to
tropospheric NOx concentrations. J Geophys Res Oceans 84: 763-768.
http://dx.doi.org/10.1029/JC084iC02p00763
Hansen. DA; Edgerton. ES: Hartsell. BE; Jansen. JJ; Kandasamy. N; Hidv. GM; Blanchard. CL. (2003).
The Southeastern Aerosol Research and Characterization Study: Part I - overview. J Air Waste Manag
Assoc 53: 1460-1471. http://dx.doi.org/10.1080/10473289.2003.10466318
Hart. J: Yanosky. J: Puett R; Ryan. L; Dockery. D; Smith. T; Garshick. E; Laden. F. (2009). Spatial
modeling of PM10 andNO2 in the continental United States, 1985-2000. Environ Health Perspect 117:
1690-1696.
Hatzopoulou. M; Weichenthal S: Barreau. G: Goldberg. M; Farrell. W: Grouse. D; Ross. N. (2013). A web-
based route planning tool to reduce cyclists' exposures to traffic pollution: A case study in Montreal,
Canada. Environ Res 123: 58-61. http://dx.doi.Org/10.1016/i.envres.2013.03.004
He. M; Dhaniyala. S. (2011). A dispersion model for traffic produced turbulence in a two-way traffic
scenario. Environmental Fluid Mechanics (Dordrecht, 2001) 11: 627-640.
http://dx.doi.org/10.1007/sl0652-011-9215-2
Heal. MR. (2008). The effect of absorbent grid preparation method on precision and accuracy of ambient
nitrogen dioxide measurements using Palmes passive diffusion tubes. J Environ Monit 10: 1363-1369.
http://dx.doi.org/10.1039/b811230d
Heeb. NV; Saxer. CJ: Forss. AM; Briihlmann. S. (2008). Trends of NO-, NO2-, and NFC-emissions from
gasoline-fueled Euro-3- to Euro-4-passenger cars. Atmos Environ 42: 2543-2554.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.12.008
Heist. DK; Brixev. LA; Richmond-Bryant. J: Bowker. GE; Perry. SG: Wiener. RW. (2009a). The effect of a
tall tower on flow and dispersion through a model urban neighborhood. J Environ Monit 11: 2163-2170.
http://dx.doi.org/10.1039/B907135k
Heist. DK; Perry. SG: Brixev. LA. (2009b). A wind tunnel study of the effect of roadway configurations on
the dispersion of traffic-related pollution. Atmos Environ 43: 5101-5111.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.06.034
Heuss. JM; Kahlbaum. DF; Wolff. GT. (2003). Weekday/weekend ozone differences: What can we learn
from them? J Air Waste Manag Assoc 53: 772-788.
Hoek. G: Beelen. R; de Hoogh. K; Vienneau. D: Gulliver. J: Fischer. P; Briggs. D. (2008). A review of
land-use regression models to assess spatial variation of outdoor air pollution [Review]. Atmos Environ
42: 7561-7578. http://dx.doi.Org/10.1016/i.atmosenv.2008.05.057
Hong. JH; Lee. JH; Park. SS. (2007). Application of an in-situ measurement system to determine HONO
levels in an indoor environment. Journal of Korean Society for Atmospheric Environment 23: 191-202.
http://dx.doi.Org/10.5572/KOSAE.2007.23.2.191
Horowitz. LW: Fiore. AM; Milly. GP; Cohen. RC: Perring. A; Wooldridge. PJ: Hess. PG: Emmons. LK;
Lamarque. JF. (2007). Observational constraints on the chemistry of isoprene nitrates over the eastern
United States. J Geophys Res 112: D12S08. http://dx.doi.org/10.1029/2006JD007747
Hu. S: Fruin. S: Kozawa. K; Mara. S: Paulson. SE; Winer. AM. (2009). A wide area of air pollutant impact
downwind of a freeway during pre-sunrise hours. Atmos Environ 43: 2541-2549.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.033
Hudman. RC: Jacob. DJ: Turquetv. S: Leibensperger. EM; Murray. IT; Wu. S: Gilliland. AB; Avery. M;
Bertram. TH; Brune. W: Cohen. RC: Dibb. JE; Flocke. FM; Fried. A; Hollowav. J: Neuman. JA;
Orville. R: Perring. A; Ren. X; Sachse. GW: Singh. HB; Swanson. A; Wooldridge. PJ. (2007). Surface
and lightning sources of nitrogen oxides over the United States: Magnitudes, chemical evolution, and
outflow. J Geophys Res Atmos 112: D12S05. http://dx.doi.org/10.1029/2006JD007912
Hwang. BF; Lee. YL. (2010). Air pollution and prevalence of bronchitic symptoms among children in
Taiwan. Chest 138: 956-964. http://dx.doi.org/10.1378/chest.09-2600
November 2013 2-112 DRAFT: Do Not Cite or Quote
-------
Hystad. P; Setton. E; Cervantes. A; Poplawski. K; Deschenes. S: Brauer. M; van Donkelaar. A; Lamsal. L;
Martin. R; Jerrett. M; Demers. P. (2011). Creating national air pollution models for population exposure
assessment in Canada. Environ Health Perspect 119: 1123-1129. http://dx.doi.org/10.1289/ehp.1002976
Iniguez. C: Ballester. F; Amoros. R; Estarlich. M; Murcia. M; Esplugues. A; Rebagliato. M. (2009). The
impact of outdoor NO2 exposure on fetal growth assessed by ultrasounds during pregnancy [Abstract].
Epidemiology 20: S78. http://dx.doi.org/10.1097/01.ede.0000362944.07976.ba
Iniguez. C: Ballester. F; Estarlich. M; Llop. S: Fernandez-Patier. R; Aguirre-Alfaro. A; Esplugues. A; Inma
Study group. V. (2009). Estimation of personal NO2 exposure in a cohort of pregnant women. Sci Total
Environ 407: 6093-6099. http://dx.doi.Org/10.1016/i.scitotenv.2009.08.006
Isaacs. K; McCurdv. T; Glen. G: Nysewander. M; Errickson. A; Forbes. S: Graham. S: McCurdv. L; Smith.
L; Tulve. N; Vallero. D. (2013). Statistical properties of longitudinal time-activity data for use in human
exposure modeling. J Expo Sci Environ Epidemiol 23: 328-336. http://dx.doi.org/10.1038/ies.2012.94
Isakov. V; Touma. JS: Burke. J: Lobdell. DT; Palma. T; Rosenbaum. A; Ozkaynak. H. (2009). Combining
regional- and local-scale air quality models with exposure models for use in environmental health
studies. J Air Waste Manag Assoc 59: 461-472.
Jacob. DJ. (1999). Introduction to atmospheric chemistry. New Jersey: Princeton University Press.
Jacob. DJ: Wofsv. SC: Bakwin. PS: Fan. SM; Harriss. RC: Talbot. RW: Bradshaw. JD; Sandholm. ST;
Singh. HB: Browell. EV: Gregory. GL: Sachse. GW: Shipham. MC: Blake. PR: Fitziarrald. DR. (1992).
Summertime photochemistry of the troposphere at high northern latitudes. J Geophys Res 97: 16421-
16431. http://dx.doi.org/10.1029/91JD01968
Jacobson. MZ. (2002). Atmospheric pollution: History, science, and regulation. New York: Cambridge
University Press.
Jalaludin. B; Mamies. T; Morgan. G: Lincoln. D; Sheppeard. V: Corbett. S. (2007). Impact of ambient air
pollution on gestational age is modified by season in Sydney, Australia. Environ Health 6: 16.
http://dx.doi.org/10.1186/1476-069X-6-16
Janssen. S: Dumont. G: Fierens. F; Deutsch. F; Maiheu. B; Celis. D: Trimpeneers. E; Mensink. C. (2012).
Land use to characterize spatial representativeness of air quality monitoring stations and its relevance
for model validation. Atmos Environ 59: 492-500. http://dx.doi.Org/10.1016/i.atmosenv.2012.05.028
Jarvis. PL; Leaderer. BP; Chinn. S: Burnev. PG. (2005). Indoor nitrous acid and respiratory symptoms and
lung function in adults. Thorax 60: 474-479. http://dx.doi.org/10.1136/thx.2004.032177
Jerrett. M; Arain. MA; Kanaroglou. P; Beckerman. B; Grouse. D: Gilbert. NL; Brook. JR; Finkelstein. N;
Finkelstein. MM. (2007). Modeling the intraurban variability of ambient traffic pollution in Toronto,
Canada. J Toxicol Environ Health A 70: 200-212. httrj://dx.doi.org/10.1080/15287390600883018
Jerrett. M; Finkelstein. M. (2005). Geographies of risk in studies linking chronic air pollution exposure to
health outcomes. J Toxicol Environ Health A 68: 1207-1242.
http://dx.doi.org/10.1080/15287390590936085
Ji. Y; Gao. Y; Li. G: An. T. (2012). Theoretical study of the reaction mechanism and kinetics of low-
molecular-weight atmospheric aldehydes (C1C4) with NO2. Atmos Environ 54: 288-295.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.02.040
Jimenez. AS: Heal. MR; Beverland. IJ. (2011). Intercomparison study of NO(x) passive diffusion tubes
with chemiluminescence analysers and evaluation of bias factors. Atmos Environ 45: 3062-3068.
http://dx.doi.0rg/10.1016/i.atmosenv.2011.03.011
Johnson. PR; Bedick. CR; Clark. NN; McKain. PL. Jr. (2009). Design and testing of an independently
controlled urea SCR retrofit system for the reduction of NOx emissions from marine diesels. Environ
Sci Technol 43: 3959-3963. http://dx.doi.org/10.1021/es900269p
Johnson. M; Isakov. V; Touma. JS: Mukerjee. S: Ozkaynak. H. (2010). Evaluation of land-use regression
models used to predict air quality concentrations in an urban area. Atmos Environ 44: 3660-3668.
http://dx.doi.0rg/10.1016/i.atmosenv.2010.06.041
November 2013 2-113 DRAFT: Do Not Cite or Quote
-------
Jourdain. L; Kulawik. SS: Worden. HM; Pickering. KE; Worden. J; Thompson. AM. (2010). Lightning
NOx emissions over the USA constrained by TES ozone observations and the GEOS-Chem model.
Atmos Chem Phys 10: 107-119. http://dx.doi.org/10.5194/acp-10-107-2010
Kar. P; Pradhan. NC: Adhikari. B. (2009). Application of sulfuric acid doped poly (m-aminophenol) as
aliphatic alcohol vapor sensor material. Sens Actuators B 140: 525-531.
http://dx.doi.0rg/10.1016/i.snb.2009.05.013
Karner. AA; Eisinger. DS: Niemeier. DA. (2010). Near-roadway air quality: Synthesizing the findings from
real-world data. Environ Sci Technol 44: 5334-5344. http://dx.doi.org/10.1021/esl00008x
Karr. CJ: Rudra. CB; Miller. KA; Gould. TR; Larson. T; Sathyanaravana. S: Koenig. JQ. (2009). Infant
exposure to fine paniculate matter and traffic and risk of hospitalization for RSV bronchiolitis in a
region with lower ambient air pollution. Environ Res 109: 321-327.
http://dx.doi.0rg/10.1016/i.envres.2008.ll.006
Kastner-Klein. P; Fedorovich. E; Ketzel. M; R. B; Britter. R. (2003). The modelling of turbulence from
traffic in urban dispersion models part II: Evaluation against laboratory and full-scale concentration
measurements in street canyons. Environ Fluid Mech 3: 145-172.
http://dx.doi.org/10.1023/A: 1022049224166
Kastner-Klein. P; Fedorovich. E; Rotach. MW. (2001). A wind tunnel study of organised and turbulent air
motions in urban street canyons. J Wind Eng Ind Aerod 89: 849-861. http://dx.doi.org/10.1016/S0167-
6105(01)00074-5
Katanoda. K; Sobue. T; Satoh. H; Taiima. K; Suzuki. T; Nakatsuka. H; Takezaki. T; Nakavama. T; Nitta. H;
Tanabe. K; Tominaga. S. (2011). An association between long-term exposure to ambient air pollution
and mortality from lung cancer and respiratory diseases in Japan. J Epidemiol 21: 132-143.
http://dx.doi.org/10.2188/iea.JE20100098
Kavnak. B: Hu. Y: Martin. RV: Russell. AG: Choi. Y: Wang. Y. (2008). The effect of lightning NOx
production on surface ozone in the continental United States. Atmos Chem Phys 8: 5151-5159.
http://dx.doi.org/10.5194/acp-8-5151-2008
Kebabian. PL; Wood. EC: Herndon. SC: Freedman. A. (2008). A practical alternative to
chemiluminescence-based detection of nitrogen dioxide: Cavity attenuated phase shift spectroscopy.
Environ Sci Technol 42: 6040-6045. http://dx.doi.org/10.1021/es703204j
Keene. WC: Lobert. RM; Crutzen. PJ; Maben. JR; Scharffe. DH; Landmann. T; Hely. C: Brain. C. (2006).
Emissions of major gaseous and paniculate species during experimental burns of southern African
biomass. J Geophys Res 111: D04301. http://dx.doi.org/10.1029/2005JD006319
Khalek. IA; Bougher. TL; Merritt. PM; Zielinska. B. (2011). Regulated and unregulated emissions from
highway heavy-duty diesel engines complying with U.S. Environmental Protection Agency 2007
emissions standards. J Air Waste Manag Assoc 61: 427-442. http://dx.doi.org/10.3155/1047-
3289.61.4.427
Khoder. MI. (2002). Nitrous acid concentrations in homes and offices in residential areas in Greater Cairo.
J Environ Monit 4: 573-578. http://dx.doi.org/10.1039/b201818g
Kim. D; Sass-Kortsak. A; Purdham. JT; Dales. RE; Brook. JR. (2006a). Associations between personal
exposures and fixed-site ambient measurements of fine paniculate matter, nitrogen dioxide, and carbon
monoxide in Toronto, Canada. J Expo Sci Environ Epidemiol 16: 172-183.
http://dx.doi.org/10.1038/sj.jea.7500446
Kim. S. -W: Heckel. A; Mckeen. SA; Frost. GJ: Hsie. E. -Y; Trainer. MK; Richter. A; Burrows. JP;
Peckham. SE; Grell. GA. (2006b). Satellite-observed U.S. power plant NOx emission reductions and
their impact on air quality. Geophys Res Lett 33: L22812. http://dx.doi.org/10.1029/2006GL027749
Kim. S: Kang. Y; Lee. H; Kim. J: Hong. S. (2011 a). Characteristics of fundamental combustion and NOx
emission using various rank coals. J Air Waste Manag Assoc 61: 254-259.
November 2013 2-114 DRAFT: Do Not Cite or Quote
-------
Kim. SW: McKeen. SA; Frost. GJ: Lee. SH; Trainer. M; Richter. A; Angevine. WM; Atlas. E; Bianco. L;
Boersma. KF; Brioude. J: Burrows. JP; de Gouw. J: Fried. A; Gleason. J; Hilboll. A; Mellqvist. J:
Peischl. J; Richter. D; Rivera. C: Ryerson. T; te Lintel Hekkert. S: Walega. J: Warneke. C: Weibring. P;
Williams. E. (20lib). Evaluations of NOx and highly reactive VOC emission inventories in Texas and
their implications for ozone plume simulations during the Texas Air Quality Study 2006. Atmos Chem
Phys 11: 11361-11386. http://dx.doi.org/10.5194/acp-ll-11361-2011
Kingham. S: Briggs. D; Elliott. P; Fischer. P; Lebret. E. (2000). Spatial variations in the concentrations of
traffic-related pollutants in indoor and outdoor air in Huddersfield, England. Atmos Environ 34: 905-
916. http://dx.doi.org/10.1016/81352-2310(99)00321-0
Klapmever. ME; Marr. LC. (2012). CO2, NOx, and particle emissions from aircraft and support activities at
a regional airport. Environ Sci Technol 46: 10974-10981. http://dx.doi.org/10.1021/es302346x
Kleffmann. J: Tapia. GV: Bejan. I; Kurtenbach. R; Wiesen. P. (2013). NO2 measurement techniques:
Pitfalls and new developments. In I Barnes; KJ Rudzin'ski (Eds.), Disposal of dangerous chemicals in
urban areas and mega cities. New York, NY: Springer, http://dx.doi.org/10.1007/978-94-007-5034-0 2
Klepeis. NE; Nelson. WC: Ott. WR; Robinson. JP; Tsang. AM; Switzer. P; Behar. JV: Hern. SC:
Engelmann. WH. (2001). The National Human Activity Pattern Survey (NHAPS): A resource for
assessing exposure to environmental pollutants. J Expo Anal Environ Epidemiol 11: 231-252.
http://dx.doi.org/10.1038/sj.jea.7500165
Ko. FW: Tarn. W: Wong. TW: Chan. DP; Tung. AH; Lai. CK; PS. H. (2007a). Temporal relationship
between air pollutants and hospital admissions for chronic obstructive pulmonary disease in Hong
Kong. Thorax 62: 780-785. http://dx.doi.org/10.1136/thx.2006.076166
Kornartit C: Sokhi. RS: Burton. MA; Ravindra. K. (2010). Activity pattern and personal exposure to
nitrogen dioxide in indoor and outdoor microenvironments. Environ Int 36: 36-45.
http://dx.doi.0rg/10.1016/i.envint.2009.09.004
Koutrakis. P: Suh. HH; Sarnat. JA; Brown. KW: Coull. BA; Schwartz. J. (2005). Characterization of
paniculate and gas exposures of sensitive subpopulations living in Baltimore and Boston (pp. 1-65;
discussion 67-75). (ISSN 1041-5505131). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=91
Kruize. H; Hanninen. O; Breugelmans. O; Lebret. E; Jantunen. M. (2003). Description and demonstration
of the EXPOLIS simulation model: two examples of modeling population exposure to paniculate
matter. J Expo Sci Environ Epidemiol 13: 87-99. http://dx.doi.org/10.1038/sj.jea.7500258
Lamsal. LN; Martin. RV; Van Donkelaar. A; Celarier. EA; Bucsela. EJ; Boersma. KF; Dirksen. R; Luo. C;
Wang. Y (2010). Indirect validation of tropospheric nitrogen dioxide retrieved from the OMI satellite
instrument: Insight into the seasonal variation of nitrogen oxides at northern midlatitudes. J Geophys
Res 115: D05302. http://dx.doi.org/10.1029/2009JD013351
Lamsal. LN; Martin. RV; Van Donkelaar. A; Steinbacher. M; Celarier. EA; Bucsela. E; Dunlea. EJ; Pinto.
JP. (2008). Ground-level nitrogen dioxide concentrations inferred from the satellite-borne ozone
monitoring instrument. J Geophys Res 113: D16308. http://dx.doi.org/10.1029/2007JD009235
Leaderer. BP; Naeher. L; Jankun. T; Balenger. K; Holford. TR; Torn. C; Sullivan. J; Wolfson. JM;
Koutrakis. P. (1999). Indoor, outdoor, and regional summer and winter concentrations of PM10, PM25,
SO42-, H+, NH4+, NO3-, NH3, and nitrous acid in homes with and without kerosene space heaters.
Environ Health Perspect 107: 223-231.
Lebret. E; Briggs. D; van Reeuwijk. H; Fischer. P; Smallbone. K; Harssema. H; Kriz. B; Gorvnski. P;
Elliot. P. (2000). Small area variations in ambient NO2 concentrations in four European areas. Atmos
Environ 34: 177-185. http://dx.doi.org/10.1016/81352-2310(99)00292-7
Lee. BH: Wood. EC: Zahniser. MS: Mcmanus. JB: Nelson. DP: Herndon. SC: Santoni. GW: Wofsv. SC:
Munger. JW. (2011 a). Simultaneous measurements of atmospheric HONO and NO2 viaabsorption
spectroscopy using tunable mid-infrared continuous-wave quantum cascade lasers. Appl Phys B 102:
417-423. http://dx.doi.org/10.1007/s00340-010-4266-5
November 2013 2-115 DRAFT: Do Not Cite or Quote
-------
Lee. H; Gu. M; Kim. YJ; Hwang. J; Jung. U. (2012a). First-time remote sensing of NO2 vertical
distributions in an urban street canyon using Topographic Target Light scattering Differential Optical
Absorption Spectroscopy (ToTaL-DOAS). Atmos Environ 54: 519-528.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.02.065
Lee. K; Xue. J: Geyh. AS: Ozkaynak. H; Leaderer. BP; Weschler. CJ: Spengler. JD. (2002). Nitrous acid,
nitrogen dioxide, and ozone concentrations in residential environments. Environ Health Perspect 110:
145-150.
Lee. K; Yeom. J: Yoon. C: Yang. W: Son. BS: Jeon. JM; Yu. SD; Park. CH. (2013). Seasonal and
geographic effects on predicting personal exposure to nitrogen dioxide by time-weighted
microenvironmental model. Atmos Environ 67: 143-148.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.ll.010
Leung. FYT; Logan. JA; Park. R; Hyer. E; Kasischke. E; Streets. D; Yurganov. L. (2007). Impacts of
enhanced biomass burning in the boreal forests in 1998 on tropospheric chemistry and the sensitivity of
model results to the injection height of emissions. J Geophys Res Atmos 112: D10313.
http://dx.doi.org/10.1029/2006jd008132
Li. L; Wu. J: Wilhelm. M; Ritz. B. (2012b). Use of generalized additive models and cokriging of spatial
residuals to improve land-use regression estimates of nitrogen oxides in Southern California. Atmos
Environ 55: 220-228. http://dx.doi.Org/10.1016/i.atmosenv.2012.03.035
Li. X; Liu. C: Leung. D. (2005). Development of a model for the determination of air exchange rates for
street canyons. Atmos Environ 39: 7285-7296. http://dx.doi.Org/10.1016/i.atmosenv.2005.09.007
Liard. R; Zureik. M; Le Moullec. Y; Soussan. D; Glorian. M; Grimfeld. A; Neukirch. F. (1999). Use of
personal passive samplers for measurement of NO2, NO, and O3 levels in panel studies. Environ Res
81: 339-348. http://dx.doi.org/10.1006/enrs.1999.3993
Lin. M; Fiore. AM; Horowitz. LW: Cooper. OR; Naik. V; Hollowav. J: Johnson. BJ; Middlebrook. AM;
Oltmans. SJ: Pollack. IB; Ryerson. TB; Warner. JX; Wiedinmyer. C: Wilson. J: Wyman. B. (2012).
Transport of Asian ozone pollution into surface air over the western United States in spring. J Geophys
Res 117: DOOV07. http://dx.doi.org/10.1029/2011JD016961
Linaker. CH; Chauhan. AJ; Inskip. HM; Holgate. ST; Coggon. D. (2000). Personal exposures of children to
nitrogen dioxide relative to concentrations in outdoor air. Occup Environ Med 57: 472-476.
Linn. WS: Shamoo. DA; Anderson. KR; Peng. RC: Avol. EL; Hackney. JD: Gong. H. Jr. (1996). Short-term
air pollution exposures and responses in Los Angeles area schoolchildren. J Expo Sci Environ
Epidemiol 6: 449-472.
Liu. CH; Leung. DYC: Barth. MC. (2005). On the prediction of air and pollutant exchange rates in street
canyons of different aspect ratios using large-eddy simulation. Atmos Environ 39: 1567-1574.
http://dx.doi.0rg/10.1016/i.atmosenv.2004.08.036
Liu. L; Poon. R; Chen. L; Frescura. AM; Montuschi. P; Ciabattoni. G: Wheeler. A; Dales. R. (2009b).
Acute effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in
asthmatic children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehpll813
Lockwood. AL; Shepson. PB; Fiddler. MN; Alaghmand. M. (2010). Isoprene nitrates: preparation,
separation, identification, yields, and atmospheric chemistry. Atmos Chem Phys 10: 6169-6178.
http://dx.doi.org/10.5194/acp-10-6169-2010
Loupa. G: Rapsomanikis. S. (2008). Air pollutant emission rates and concentrations in medieval churches. J
Atmos Chem 60: 169-187. http://dx.doi.org/10.1007/sl0874-008-9116-z
Lu. R; Turco. RP; Jacobson. MZ. (1997a). An integrated air pollution modeling system for urban and
regional scales: 1. Structure and performance. J Geophys Res 102: 6063-6079.
http://dx.doi.org/10.1029/96JD03501
November 2013 2-116 DRAFT: Do Not Cite or Quote
-------
Lu. R; Turco. RP; Jacobson. MZ. (1997b). An integrated air pollution modeling system for urban and
regional scales: 2. Simulations for SCAQS 1987. J Geophys Res Atmos 102: 6081-6098.
http://dx.doi.org/10.1029/96JD03502
Macintyre. HL; Evans. MJ. (2010). Sensitivity of a global model to the uptake of N2O5 by tropospheric
aerosol. Atmos Chem Phys 10: 7409-7414. http://dx.doi.org/10.5194/acp-10-7409-2010
Mannes. T; Jalaludin. B; Morgan. G: Lincoln. D; Sheppeard. V: Corbett. S. (2005). Impact of ambient air
pollution on birth weight in Sydney, Australia. Occup Environ Med 62: 524-530.
http://dx.doi.org/10.1136/oem.2004.014282
Mao. J: Paulot F: Jacob. DJ: Cohen. RC: Crounse. JD: Wennberg. PO: Keller. CA: Hudman. RC: Barklev. MP:
Horowitz. LW. (2013). Ozone and organic nitrates over the eastern United States: Sensitivity to isoprene
chemistry. J Geophys Res Atmos 118: 11256-11268. http://dx.doi.org/10.1002/jgrd.50817
Mar. TF; Norris. GA; Koenig. JQ; Larson. TV. (2000). Associations between air pollution and mortality in
Phoenix, 1995-1997. Environ Health Perspect 108: 347-353.
Marr. LC: Harlev. RA. (2002). Modeling the effect of weekday-weekend differences in motor vehicle
emissions on photochemical air pollution in central California. Environ Sci Technol 36: 4099-4106.
http://dx.doi.org/10.1021/es020629x
Marshall. JD: Nethery. E; Brauer. M. (2008). Within-urban variability in ambient air pollution: Comparison
of estimation methods. Atmos Environ 42: 1359-1369.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.012
Martin. RV; Chance. K; Jacob. DJ: Kurosu. TP; Spurr. RJD; Bucsela. E; Gleason. JF; Palmer. PI; Bey. I;
Fiore. AM; Li. Q; Yantosca. RM; Koelemeijer. RBA. (2002). An improved retrieval of tropospheric
nitrogen dioxide from GOME. J Geophys Res 107: 4437. http://dx.doi.org/10.1029/2001JD001027
Martins. PC: Valente. J: Papoila. AL; Caires. I; Araujo-Martins. J: Mata. P; Lopes. M; Torres. S: Rosado-
Pinto. J: Borrego. C: Annesi-Maesano. I; Neupartfa. N. (2012). Airways changes related to air pollution
exposure in wheezing children. Eur Respir J 39: 246-253. http://dx.doi.org/10.1183/09031936.00025111
Massoli. P: Former. EC: Canagaratna. MR; Williams. LR; Zhang. Q; Sun. Y; Schwab. JJ: Trimborn. A;
Onasch. TB; Demerjian. KL; Kolb. CE; Worsnop. PR; Jayne. JT. (2012). Pollution gradients and
chemical characterization ofparticulatematter from vehicular traffic near major roadways: Results from
the 2009 Queens College air quality study in NYC. Aerosol Sci Technol 46: 1201-1218.
htrp://dx.doi.org/10.1080/02786826.2012.701784
Mauzerall. PL; Logan. JA; Jacob. DJ: Anderson. BE; Blake. PR; Bradshaw. JD: Heikes. B; Sachse. GW:
Singh. H; Talbot. B. (1998). Photochemistry in biomass burning plumes and implications for
tropospheric ozone over the tropical South Atlantic. J Geophys Res 103: 8401.
http://dx.doi.org/10.1029/97JD02612
Mazaheri. M; Johnson. GR; Morawska. L. (2009). Particle and gaseous emissions from commercial aircraft
at each stage of the landing and takeoff cycle. Environ Sci Technol 43: 441-446.
http://dx.doi.org/10.1021/es8013985
Mazaheri. M; Johnson. GR: Morawska. L. (2011). An inventory of particle and gaseous emissions from
large aircraft thrust engine operations at an airport. Atmos Environ 45: 3500-3507.
http://dx.doi.0rg/10.1016/i.atmosenv.2010.12.012
McClenny. WA; Williams. EJ; Cohen. RC: Stutz. J. (2002). Preparing to measure the effects of the NOx
SIP call-methods for ambient air monitoring of NO, NO2, NOy, and individual NOz species. J Air
Waste Manag Assoc 52: 542-562. http://dx.doi.org/10.1080/10473289.2002.10470801
McConnell. R; Berhane. K; Gilliland. F; Molitor. J: Thomas. D; Lurmann. F; Avol. E; Gauderman. WJ:
Peters. JM. (2003). Prospective study of air pollution and bronchitic symptoms in children with asthma.
Am J Respir Crit Care Med 168: 790-797. http://dx.doi.org/10.1164/rccm.200304-466OC
November 2013 2-117 DRAFT: Do Not Cite or Quote
-------
McConnell. R: Islam. T: Shankardass. K: Jerrett M: Lurmann. F: Gilliland. F: Gauderman. J: Avol. E: Kuenzlj
N: Yao. L: Peters. J: Berhane. K. (2010). Childhood incident asthma and traffic-related air pollution at home
and school. Environ Health Perspect 118: 1021-1026. http://dx.doi.org/10.1289/ehp.0901232
Mcdonald. BC: Dallmann. TR; Martin. EW: Harlev. RA. (2012). Long-term trends in nitrogen oxide
emissions from motor vehicles at national, state, and air basin scales. J Geophys Res 117: DOOV18.
http://dx.doi.org/10.1029/2012JD018304
McMeeking. GR; Kreidenweis. SM; Baker. S: Carrico. CM; Chow. JC: Collett. JL. Jr; Hao. WM; Holden.
AS: Kirchstetter. TW: Malm. WC: Moosmiiller. H; Sullivan. AP; Wold. CE. (2009). Emissions of trace
gases and aerosols during the open combustion of biomass in the laboratory. J Geophys Res Atmos 114:
D19210. http://dx.doi.org/10.1029/2009jd011836
Mebust AK; Russell. AR; Hudman. RC: Valin. LC: Cohen. RC. (2011). Characterization of wildfire NOx
emissions using MODIS fire radiative power and OMI tropospheric NO2 columns. Atmos Chem Phys
11: 5839-5851. http://dx.doi.org/10.5194/acp-ll-5839-2011
Mehta. S: Ngo. LH; Dzung. DV; Cohen. A; Thach. TQ; Dan. VX; Tuan. ND; Giang. LT. (2013). Air
pollution and admissions for acute lower respiratory infections in young children of Ho Chi Minh City.
Air Qual Atmos Health 6: 167-179. http://dx.doi.org/10.1007/sll869-011-0158-z
Meng. Q; Williams. R; Pinto. JP (2012a). Determinants of the associations between ambient concentrations
and personal exposures to ambient PM2.5, NO2, and O3 during DEARS. Atmos Environ 63: 109-116.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.09.019
Meng. QY; Svendsgaard. D; Kotchmar. DJ; Pinto. JP. (2012b). Associations between personal exposures
and ambient concentrations of nitrogen dioxide: A quantitative research synthesis. Atmos Environ 57:
322-329. http://dx.doi.0rg/10.1016/i.atmosenv.2012.04.035
Mercer. LD; Szpiro. AA; Sheppard. L; Lindstrom. J: Adar. SD; Allen. RW: Avol. EL; Oron. AP; Larson. T;
Liu. LJ; Kaufman. JD. (2011). Comparing universal kriging and land-use regression for predicting
concentrations of gaseous oxides of nitrogen (NOx) for the Multi-Ethnic Study of Atherosclerosis and
Air Pollution (MESA Air). Atmos Environ 45: 4412-4420.
http://dx.doi.0rg/10.1016/i.atmosenv.2011.05.043
Minguillon. MC: Rivas. I; Aguilera. I; Alastuev. A; Moreno. T; Amato. F; Sunyer. J: Querol. X. (2012).
Within-city contrasts in PM composition and sources and their relationship with nitrogen oxides. J
Environ Monit 14: 2718-2728. http://dx.doi.org/10.1039/c2em30469d
Mollner. AK; Valluvadasan. S: Feng. L; Sprague. MK; Okumura. M; Milligan. DB; Bloss. WJ: Sander. SP;
Martien. PT; Harlev. RA. (2010). Rate of gas phase association of hydroxyl radical and nitrogen
dioxide. Science 330: 646-649. http://dx.doi.org/10.1126/science.1193030
Molloy. SB; Cheng. M; Galbally. IE; Kevwood. MD; Lawson. SJ: Powell. JC: Gillett. R: Dunne. E;
Selleck. PW. (2012). Indoor air quality in typical temperate zone Australian dwellings. Atmos Environ
54: 400-407. http://dx.doi.Org/10.1016/i.atmosenv.2012.02.031
Molter. A; Lindlev. S: de Vocht. F; Agius. R: Kerry. G: Johnson. K; Ashmore. M; Terry. A;
Dimitroulopoulou. S: Simpson. A. (2012). Performance of a microenviromental model for estimating
personal NO2 exposure in children. Atmos Environ 51: 225-233.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.01.030
Molter. A; Lindlev. S: de Vocht. F; Simpson. A; Agius. R. (2010a). Modelling air pollution for
epidemiologic research-Part I: A novel approach combining land use regression and air dispersion. Sci
Total Environ 408: 5862-5869. http://dx.doi.Org/10.1016/i.scitotenv.2010.08.027
Monn. C. (2001). Exposure assessment of air pollutants: A review on spatial heterogeneity and
indoor/outdoor/personal exposure to suspended paniculate matter, nitrogen dioxide and ozone [Review].
Atmos Environ 35: 1-32. http://dx.doi.org/10.1016/sl352-2310(00)00330-7
Mortimer. KM; Neas. LM; Dockery. DW: Redline. S: Tager. IB. (2002). The effect of air pollution on
inner-city children with asthma. Eur Respir J 19: 699-705.
http://dx.doi.org/10.1183/09031936.02.00247102
November 2013 2-118 DRAFT: Do Not Cite or Quote
-------
Moshammer. H; Hutter. HP; Hauck. H; Neuberger. M. (2006). Low levels of air pollution induce changes
of lung function in a panel of schoolchildren. Eur Respir J 27: 1138-1143.
http://dx.doi.org/10.1183/09031936.06.00089605
Mukala. K; Aim. S: Tiittanen. P; Salonen. RO; Jantunen. M; Pekkanen. J. (2000). Nitrogen dioxide
exposure assessment and cough among preschool children. Arch Environ Health 55: 431-438.
Mukerjee. S: Smith. L; Neas. L; Norris. G. (2012a). Evaluation of land use regression models for nitrogen
dioxide and benzene in Four US cities. ScientificWorldJournal 2012: 865150.
http://dx.doi.org/10.1100/2012/865150
Mukerjee. S: Smith. LA; Johnson. MM; Neas. LM; Stallings. CA. (2009). Spatial analysis and land use
regression of VOCs and NO(2) from school-based urban air monitoring in Detroit/Dearborn, USA. Sci
Total Environ 407: 4642-4651. http://dx.doi.Org/10.1016/i.scitotenv.2009.04.030
Mukerjee. S: Willis. RD; Walker. JT; Hammond. D; Norris. GA; Smith. LA; Welch. DP; Peters. TM.
(2012b). Seasonal effects in land use regression models for nitrogen dioxide, coarse paniculate matter,
and gaseous ammonia in Cleveland, Ohio. Atmospheric Pollution Research 3: 352-361.
http://dx.doi.org/10.5094/APR.2012.039
Nishikawa. Y; Murano. K; Mukai. H. (2009). Comparison of sampling resistance for one to three sheets of
membrane type passive sampler. Water Air Soil Pollut 197: 241-247. http://dx.doi.org/10.1007/sll270-
008-9807-8
Nengaard. JK. (2010). Indoor measurements of the sum of the nitrate radical, NO3, and nitrogen pentoxide,
N2O5 in Denmark. Chemosphere 79: 898-904. http://dx.doi.Org/10.1016/j.chemosphere.2010.02.025
N0jgaard. JK; Bilde. M; Stenbv. C: Nielsen. OJ; Wolkoff. P. (2006). The effect of nitrogen dioxide on
particle formation during ozonolysis of two abundant monoterpenes indoors. Atmos Environ 40: 1030-
1042. http://dx.doi.0rg/10.1016/i.atmosenv.2005.ll.029
Novotny. EV; Bechle. MJ; Millet. DB; Marshall. JD. (2011). National satellite-based land-use regression:
NO2 in the United States. Environ Sci Technol 45: 4407-4414. http://dx.doi.org/10.1021/esl03578x
Noxon. JF. (1976). Atmospheric nitrogen fixation by lightning. Geophys Res Lett 3: 463-465.
http://dx.doi.org/10.1029/GL003i008pQ0463
Noxon. JF. (1978). Tropospheric NO2. J Geophys Res Oceans 83: 3051-3057.
http://dx.doi.org/10.1029/JC083iC06p03051
Nunnermacker. LJ; Imre. D; Daum. PH; Kleinman. L; Lee. YN; Lee. JH; Springston. SR; Newman. L;
Weinstein-Llovd. J: Luke. WT; Banta. R; Alvarez. R; Senff. C: Sillman. S: Holdren. M; Keiglev. GW:
Zhou. X. (1998). Characterization of the Nashville urban plume on July 3 and July 18, 1995. J Geophys
Res 103: 28129-28148.
O'Connor. GT; Neas. L; Vaughn. B; Kattan. M; Mitchell. H; Grain. EF; Evans. R. Ill; Gruchalla. R:
Morgan. W: Stout. J: Adams. GK; Lippmann. M. (2008). Acute respiratory health effects of air pollution
on children with asthma in US inner cities. J Allergy Clin Immunol 121: 1133-1139.
http://dx.doi.0rg/10.1016/i.iaci.2008.02.020
Oftedal. B; Nystad. W: Brunekreef. B; Nafstad. P. (2009). Long-term traffic-related exposures and asthma
onset in schoolchildren in Oslo, Norway. Environ Health Perspect 117: 839-844.
http://dx.doi.org/10.1289/ehp.11491
Oiamo. TH; Luginaah. IN; Atari. DO; Gorev. KM. (2011). Air pollution and general practitioner access and
utilization: a population based study in Sarnia, 'Chemical Valley,' Ontario. Environ Health Global
Access Sci Source 10: 71. http://dx.doi.org/10.1186/1476-069X-10-71
Oiamo. TH; Luginaah. IN; Buzzelli. M; Tang. K; Xu. X; Brook. JR; Johnson. M. (2012). Assessing the
spatial distribution of nitrogen dioxide in London, Ontario. J Air Waste Manag Assoc 62: 1335-1345.
http://dx.doi.org/10.1080/10962247.2012.715114
Olaguer. EP (2012). The potential near-source ozone impacts of upstream oil and gas industry emissions. J
Air Waste Manag Assoc 62: 966-977. http://dx.doi.org/10.1080/10962247.2012.688923
November 2013 2-119 DRAFT: Do Not Cite or Quote
-------
Olaguer. EP; Kolb. CE; Lefer. B; Rappengliick. B; Zhang. RY; Pinto. JP. (In Press) Overview of the
SHARP campaign: Motivation, design, and major outcomes. J Geophys Res.
Olaguer. EP; Rappengliick. B; Lefer. B; Stutz. J; Dibb. J: Griffin. R; Brune. WH; Shauck. M; Buhr. M;
Jeffries. H; Vizuete. W: Pinto. JP. (2009). Deciphering the role of radical precursors during the Second
Texas Air Quality Study. J Air Waste Manag Assoc 59: 1258-1277. http://dx.doi.org/10.3155/1047-
3289.59.11.1258
Osthoff. HP; Roberts. JM; Ravishankara. AR; Williams. EJ; Lerner. BM; Sommariva. R; Bates. TS:
Coffman. D; Quinn. PK; Dibb. JE; Stark. H; Burkholder. JB; Talukdar. RK; Meagher. J: Fehsenfeld. FC:
Brown. SS. (2008). High levels of nitryl chloride in the polluted subtropical marine boundary layer. Nat
Geosci 1: 324-328. http://dx.doi.org/10.1038/ngeol77
Ott. LE; Pickering. KE; Stenchikov. GL; Allen. DJ; Decaria. AJ; Ridley. B; Lin. R. -F; Lang. S: Tao. W. -K.
(2010). Production of lightning NO x and its vertical distribution calculated from three-dimensional
cloud-scale chemical transport model simulations. J Geophys Res 115: D04301.
http://dx.doi.org/10.1029/2009JD011880
Ott. LE; Pickering. KE; Stenchikov. GL; Huntrieser. H; Schumann. U. (2007). Effects of lightning NOx
production during the July 21 European Lightning Nitrogen Oxides Project storm studied with a three-
dimensional cloud-scale chemical transport model. J Geophys Res Atmos 112: D05307.
http://dx.doi.org/10.1029/2006JD007365
Ozden. O; Dogeroglu. T. (2008). Field evaluation of a tailor-made new passive sampler for the
determination of NO2 levels in ambient air. Environ Monit Assess 142: 243-253.
http://dx.doi.org/10.1007/sl0661-007-9921-x
Ozkaynak. H; Frev. HC: Burke. J: Finder. RW. (2009). Analysis of coupled model uncertainties in source-
to-dose modeling of human exposures to ambient air pollution: A PM2.5 case study. Atmos Environ 43:
1641-1649. http://dx.doi.0rg/10.1016/i.atmosenv.2008.12.008
Pachon. JE; Balachandran. S: Hu. Y; Mulholland. JA; Darrow. LA; Sarnat. JA; Tolbert. PE; Russell. AG.
(2012). Development of outcome-based, multipollutant mobile source indicators. J Air Waste Manag
Assoc 62: 431-442. http://dx.doi.org/10.1080/10473289.2012.656218
Pacsi. AP; Alhajeri. NS: Zavala-Araiza. D; Webster. MD; Allen. DT. (2013). Regional air quality impacts of
increased natural gas production and use in Texas. Environ Sci Technol 47: 3521-3527.
http://dx.doi.org/10.1021/es3044714
Parenteau. MP; Sawada. MC. (2012). The role of spatial representation in the development of a LUR model
for Ottawa, Canada. Air Qual Atmos Health 5: 311-323. http://dx.doi.org/10.1007/sll869-010-0094-3
Park. SS: Hong. JH; Lee. JH; Kim. YJ; Cho. SY; Kim. SJ. (2008). Investigation of nitrous acid
concentration in an indoor environment using an in-situ monitoring system. Atmos Environ 42: 6586-
6596. http://dx.doi.0rg/10.1016/i.atmosenv.2008.05.023
Parrish. DP; Fehsenfeld. FC. (2000). Methods for gas-phase measurements of ozone, ozone precursors and
aerosol precursors. Atmos Environ 34: 1921-1957. http://dx.doi.org/10.1016/81352-2310(99)00454-9
Paulot. F; Crounse. JD; Kjaergaard. HG: Kroll. JH; Seinfeld. JH; Wennberg. PP. (2009). Isoprene
photooxidation: New insights into the production of acids and organic nitrates. Atmos Chem Phys 9:
1479-1501. http://dx.doi.org/10.5194/acp-9-1479-2009
Pegas. PN; Nunes. T; Alves. CA; Silva. JR; Vieira. SLA; Caseiro. A; Pio. CA. (2012). Indoor and outdoor
characterisation of organic and inorganic compounds in city centre and suburban elementary schools of
Aveiro, Portugal. Atmos Environ 55: 80-89. http://dx.doi.Org/10.1016/i.atmosenv.2012.03.059
Perring. AE; Bertram. TH; Wooldridge. PJ: Fried. A; Heikes. BG: Dibb. J: Crounse. JD: Wennberg. PO;
Blake. NJ; Blake. PR; Brune. WH; Singh. HB; Cohen. RC. (2009). Airborne observations of total
RONO2: New constraints on the yield and lifetime of isoprene nitrates. Atmos Chem Phys 9: 1451-
1463. http://dx.doi.org/10.5194/acp-9-1451-2009
November 2013 2-120 DRAFT: Do Not Cite or Quote
-------
Peterson. HS: Beaslev. WH. (2011). Possible catalytic effects of ice particles on the production of NOx by
lightning discharges. Atmos Chem Phys 11: 10259-10268. http://dx.doi.org/10.5194/acp-ll-10259-2011
Pfister. GG: Wiedinmyer. C: Emmons. LK. (2008). Impacts of the fall 2007 California wildfires on surface
ozone: Integrating local observations with global model simulations. Geophys Res Lett 35: L19814.
http://dx.doi.org/10.1029/2008gl034747
Physick. W: Powell. J: Cope. M; Boast K; Lee. S. (2011). Measurements of personal exposure to NO2 and
modelling using ambient concentrations and activity data. Atmos Environ 45: 2095-2102.
http://dx.doi.0rg/10.1016/i.atmosenv.2011.01.063
Pickering. KB; Wang. Y; Tao. WK; Price. C: Muller. JF. (1998). Vertical distributions of lightning NOx for
use in regional and global chemical transport models. J Geophys Res 103: 31203-31216.
Piechocki-Minguy. A; Plaisance. H; Schadkowski. C: Sagnier. I; Saison. JY; Galloo. JC: Guillermo. R.
(2006). A case study of personal exposure to nitrogen dioxide using a new high sensitive diffusive
sampler. Sci Total Environ 366: 55-64. http://dx.doi.Org/10.1016/i.scitotenv.2005.08.009
Poddubny. VA; Yushketova. NA. (2013). A physicochemical model of sorption processes in NO(2) passive
sampling with air humidity effects. Environ Monit Assess 185: 3819-3829.
http://dx.doi.org/10.1007/sl0661-012-2830-7
Polidori. A; Fine. PM. (2012). Ambient concentrations of criteria and air toxic pollutants in close proximity
to a freeway with heavy-duty diesel traffic. Final report. South Coast Air Quality Management District.
http ://www. aqmd.gov/tao/AQ-Reports/I71 OFwv Studv.pdf
Pollack. IB; Lerner. BM; Ryerson. TB. (2011). Evaluation of ultraviolet light-emitting diodes for detection
of atmospheric NO2 by photolysis - chemiluminescence. J Atmos Chem 65: 111-125.
http://dx.doi.org/10.1007/sl0874-011-9184-3
Pollack. IB; Ryerson. TB; Trainer. M; Parrish. DP; Andrews. AE; Atlas. EL; Blake. PR; Brown. SS:
Commane. R; Daube. BC: De Gouw. JA; Dube. WP; Flynn. J: Frost. GJ: Gilman. JB; Grossberg. N;
Hollowav. JS: Kofler. J: Kort. EA; Kuster. WC: Lang. PM; Lefer. B; Lueb. RA; Neuman. JA; Nowak.
JB; Novelli. PC: Peischl. J: Perring. AE: Roberts. JM; Santoni. G: Schwarz. JP; Spackman. J. R.:
Wagner. NL; Warneke. C: Washenfelder. RA; Wofsv. SC: Xiang. B. (2012). Airborne and ground-based
observations of a weekend effect in ozone, precursors, and oxidation products in the California South
Coast Air Basin. J Geophys Res 117: DOOV05. http://dx.doi.org/10.1029/2011JD016772
Pun. BK; Seigneur. C: White. W. (2003). Day-of-week behavior of atmospheric ozone in three US cities. J
Air Waste Manag Assoc 53: 789-801.
Raff. JD: Niegic. B; Chang. WL; Gordon. MS: Dabdub. D; Gerber. RB; Finlavson-Pitts. BJ. (2009).
Chlorine activation indoors and outdoors via surface-mediated reactions of nitrogen oxides with
hydrogen chloride. PNAS 106: 13647-13654. http://dx.doi.org/10.1073/pnas.0904195106
Rao. KS: Gunter. RL; White. JR; Hosker. RP. (2002). Turbulence and dispersion modeling near highways.
Atmos Environ 36: 4337-4346. http://dx.doi.org/10.1016/81352-2310(02)00353-9
Real. E; Law. KS: Weinzierl. B; Fiebig. M; Petzold. A; Wild. O: Methven. J: Arnold. S: Stohl. A;
Huntrieser. H; Roiger. A; Schlager. H; Stewart. D; Avery. M; Sachse. G: Browell. E; Ferrare. R: Blake.
EX (2007). Processes influencing ozone levels in Alaskan forest fire plumes during long-range transport
over the North Atlantic. J Geophys Res 112: D10S41. http://dx.doi.org/10.1029/2006jd007576
Reeves. GK; Cox. PR; Darby. SC: Whitlev. E. (1998). Some aspects of measurement error in explanatory
variables for continuous and binary regression models. Stat Med 17: 2157-2177.
http://dx.doi.org/10.1002/(SICI)1097-0258(19981015)17:19<2157::AID-SIM916>3.0.CO:2-F
Restrepo. C: Zimmerman. R; Thurston. G: Clemente. J: Gorczynski. J: Zhong. M; Blaustein. M; Chen. LC.
(2004). A comparison of ground-level air quality data with New York State Department of
Environmental Conservation monitoring stations data in South Bronx, New York. Atmos Environ 38:
5295-5304. http://dx.doi.Org/10.1016/i.atmosenv.2004.06.004
November 2013 2-121 DRAFT: Do Not Cite or Quote
-------
Roberts. JM; Veres. P; Warneke. C: Neuman. JA; Washenfelder. RA; Brown. SS: Baasandorj. M;
Burkholder. JB; Burling. IR; Johnson. TJ; Yokelson. RJ; de Gouw. J. (2010). Measurement of HONO,
HNCO, and other inorganic acids by negative-ion proton-transfer chemical-ionization mass
spectrometry (NI-PT-CIMS): Application to biomass burning emissions. Atmos Meas Tech 3: 981-990.
http://dx.doi.org/10.5194/amt-3-981-2010
Roberts. S: Martin. MA. (2006). Using supervised principal components analysis to assess multiple
pollutant effects. Environ Health Perspect 114: 1877-1882. http://dx.doi.org/10.1289/ehp.9226
Rodgers. MO; Davis. DP. (1989). A UV-photofragmentation/laser-induced fluorescence sensor for the
atmospheric detection of HONO. Environ Sci Technol 23: 1106-1112.
Rodriguez. MA; Barna. MG: Moore. T. (2009). Regional impacts of oil and gas development on ozone
formation in the western United States. J Air Waste Manag Assoc 59: 1111-1118.
http://dx.doi.0rg/10.3155/1047-3289.59.9.llll
Rojas-Martinez. R; Perez-Padilla. R; Olaiz-Fernandez. G: Mendoza-Alvarado. L; Moreno-Macias. H;
Fortoul. T; McDonnell. W: Loomis. D; Romieu. I. (2007a). Lung function growth in children with long-
term exposure to air pollutants in Mexico City. Am J Respir Crit Care Med 176: 377-384.
http://dx.doi.org/10.1164/rccm.200510-1678OC
Rollins. AW: Kiendler-Scharr. A; Fry. JL; Brauers. T; Brown. SS: Porn. HP; Dube. WP; Fuchs. H; Mensah.
A; Mentel. TF; Rohrer. F; Tillmann. R; Wegener. R; Wooldridge. PJ: Cohen. RC. (2009). Isoprene
oxidation by nitrate radical: alkyl nitrate and secondary organic aerosol yields. Atmos Chem Phys 9:
6685-6703. http://dx.doi.org/10.5194/acp-9-6685-2009
Rose. N; Cowie. C: Gillett R: Marks. GB. (2011). Validation of a spatiotemporal land use regression model
incorporating fixed site monitors. Environ Sci Technol 45: 294-299. http://dx.doi.org/10.1021/esl00683t
Rosenlund. M; Forastiere. F; Stafoggia. M; Porta. D; Perucci. M; Ranzi. A; Nussio. F; Perucci. C. (2008a).
Comparison of regression models with land-use and emissions data to predict the spatial distribution of
traffic-related air pollution in Rome. J Expo Sci Environ Epidemiol 18: 192-199.
http://dx.doi.org/10.1038/si.ies.7500571
Roy. S: Baiker. A. (2009). NOx storage-reduction catalysis: from mechanism and materials properties to
storage-reduction performance [Review]. Chem Rev 109: 4054-4091.
http://dx.doi.org/10.1021/cr800496f
Sahsuvaroglu. T; Su. JG: Brook. J: Burnett. R: Loeb. M; Jerrett. M. (2009). Predicting personal nitrogen
dioxide exposure in an elderly population: Integrating residential indoor and outdoor measurements,
fixed-site ambient pollution concentrations, modeled pollutant levels, and time-activity patterns. J
Toxicol Environ Health A 72: 1520-1533. http://dx.doi.org/10.1080/15287390903129408
Saltzman. BE. (1954). Colorimetric microdetermination of nitrogen dioxide in the atmosphere. Anal Chem
26: 1949-1955.
Samoli. E; Nastos. PT; Paliatsos. AG: Katsouvanni. K; Priftis. KN. (2011). Acute effects of air pollution on
pediatric asthma exacerbation: Evidence of association and effect modification. Environ Res 111: 418-
424. http://dx.doi.0rg/10.1016/j.envres.2011.01.014
Sanchez Jimenez. A; Heal. MR; Beverland. IJ. (2012). Correlations of particle number concentrations and
metals with nitrogen oxides and other traffic-related air pollutants in Glasgow and London. Atmos
Environ 54: 667-678. http://dx.doi.Org/10.1016/i.atmosenv.2012.01.047
Sarnat. JA; Brown. KW: Schwartz. J: Coull. BA; Koutrakis. P. (2005). Ambient gas concentrations and
personal paniculate matter exposures: Implications for studying the health effects of particles.
Epidemiology 16: 385-395. http://dx.doi.org/10.1097/01.ede.0000155505.04775.33
Sarnat. JA: Koutrakis. P; Suh. HH. (2000). Assessing the relationship between personal paniculate and
gaseous exposures of senior citizens living in Baltimore, MD. J Air Waste Manag Assoc 50: 1184-1198.
http://dx.doi.org/10.1080/10473289.2000.10464165
November 2013 2-122 DRAFT: Do Not Cite or Quote
-------
Sarnat. JA; Sarnat SE: Flanders. WD; Chang. HH; Mulholland. J: Baxter. L; Isakov. V; Ozkaynak. H.
(2013a). Spatiotemporally resolved air exchange rate as a modifier of acute air pollution-related
morbidity in Atlanta. J Expo Sci Environ Epidemiol 23: 606-615. http://dx.doi.org/10.1038/ies.2013.32
Sarnat. JA; Schwartz. J; Catalano. PJ; Suh. HH. (2001). Gaseous pollutants in paniculate matter
epidemiology: Confounders or surrogates? Environ Health Perspect 109: 1053-1061.
Sarnat. SE; Coull. BA; Ruiz. PA; Koutrakis. P; Suh. HH. (2006). The influences of ambient particle
composition and size on particle infiltration in Los Angeles, CA, residences. J Air Waste Manag Assoc
56: 186-196. http://dx.doi.org/10.1080/10473289.2006.10464449
Sarnat. SE: Klein. M; Sarnat. JA: Flanders. WD: Waller. LA; Mulholland. JA: Russell. AG: Tolbert. PE.
(2010). An examination of exposure measurement error from air pollutant spatial variability in time-
series studies. J Expo Sci Environ Epidemiol 20: 135-146. http://dx.doi.org/10.1038/ies.2009.10
Sarnat. SE: Ravsoni. AU; Li. WW: Holguin. F; Johnson. BA; Flores Luevano. S: Garcia. JH; Sarnat. JA.
(2012). Air pollution and acute respiratory response in a panel of asthmatic children along the U.S.-
Mexico border. Environ Health Perspect 120: 437444. http://dx.doi.org/10.1289/ehp.1003169
Sarnat. SE: Sarnat. JA: Mulholland. J: Isakov. V; Ozkaynak. H; Chang. HH: Klein. M; Tolbert. PE.
(2013b). Application of alternative spatiotemporal metrics of ambient air pollution exposure in a time-
series epidemiological study in Atlanta. J Expo Sci Environ Epidemiol 23: 593-605.
http://dx.doi.org/10.1038/ies.2013.41
Sarwar. G: Corsi. R; Kumura. Y; Allen. D; Weschler. CJ. (2002). Hydroxyl radicals in indoor environments.
Atmos Environ 36: 3973-3988.
Sather. ME; Slonecker. ET; Mathew. J: Daughtrev. H; Williams. DP. (2007). Evaluation of Ogawa Passive
Sampling Devices as an alternative measurement method for the Nitrogen Dioxide annual standard in El
Paso, Texas. Environ Monit Assess 124: 211-221. http://dx.doi.org/10.1007/sl0661-006-9219-4
Sawant. AA; Nigam. A; Miller. JW: Johnson. KG: Cocker. DR. (2007). Regulated and non-regulated
emissions from in-use diesel-electric switching locomotives. Environ Sci Technol 41: 6074-6083.
http://dx.doi.org/10.1021/es061672d
Schembari. A; Triguero-Mas. M; De Nazelle. A; Dadvand. P; Vrijheid. M; Cirach. M; Martinez. D;
Figueras. F; Querol. X; Basagafia. X; Eeftens. M; Meliefste. K; Nieuwenhuijsen. MJ. (2013). Personal,
indoor and outdoor air pollution levels among pregnant women. Atmos Environ 64: 287-295.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.09.053
Schildcrout. JS: Sheppard. L; Lumlev. T; Slaughter. JC: Koenig. JQ; Shapiro. GG. (2006). Ambient air
pollution and asthma exacerbations in children: An eight-city analysis. Am J Epidemiol 164: 505-517.
http://dx.doi.org/10.1093/aie/kwj225
Schumann. U; Huntrieser. H. (2007). The global lightning-induced nitrogen oxides source. Atmos Chem
Phys 7: 3823-3907.
Sedefian. L; Rao. ST; Czapski. U. (1981). Effects of traffic-generated turbulence on near-field dispersion.
Atmos Environ (1967) 15: 527-536. htrp://dx.doi.org/10.1016/0004-6981(81)90182-7
Seinfeld. JH; Pandis. SN. (1998). Atmospheric chemistry and physics: From air pollution to climate
change. New York: John Wiley & Sons. http://dx.doi.Org/10.1063/l.882420
Sekine. Y; Watts. SF; Rendell. A; Butsugan. M. (2008). Development of highly Sensitive passive sampler
for nitrogen dioxide using porous polyethylene membrane filter as turbulence limiting diffuser. Atmos
Environ 42: 4079-4088. http://dx.doi.Org/10.1016/i.atmosenv.2008.01.029
Sereviciene. V: Paliulis. D. (2012). Influence of temperature and relative humidity on the performance of
nitrogen dioxide diffusive sampler. Res J of Chem Sci 2: 89-92.
Setton. E; Marshall. JD; Brauer. M; Lundquist. KR; Hystad. P: Keller. P; Cloutier-Fisher. D. (2011). The
impact of daily mobility on exposure to traffic-related air pollution and health effect estimates. J Expo
Sci Environ Epidemiol 21: 42-48. http://dx.doi.org/10.1038/jes.2010.14
November 2013 2-123 DRAFT: Do Not Cite or Quote
-------
Sheppard. L. (2005). Acute air pollution effects: Consequences of exposure distribution and measurements.
J Toxicol Environ Health A 68: 1127-1135. http://dx.doi.org/10.1080/15287390590935987
Sheppard. L; Slaughter. JC: Schildcrout. J: L-JS. L; Lumlev. T. (2005). Exposure and measurement
contributions to estimates of acute air pollution effects. J Expo Sci Environ Epidemiol 15: 366-376.
http://dx.doi.org/10.1038/sj.jea.7500413
Simoens. S: Ayrault M; Wallace. JM. (2007). The flow across a street canyon of variable width- Part 1:
Kinematic description. Atmos Environ 41: 9002-9017.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.013
Simoens. S: Wallace. JM. (2008). The flow across a street canyon of variable widthPart 2: Scalar dispersion
from a street level line source. Atmos Environ 42: 2489-2503.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.12.013
Simon. H; Kimura. Y; Mcgaughev. G: Allen. DT; Brown. SS: Osthoff. HP; Roberts. JM: Byun. D; Lee. D.
(2009). Modeling the impact of C1NO2 on ozone formation in the Houston area. J Geophys Res Atmos
114: DOOF03. http://dx.doi.org/10.1029/2008id010732
Simon. PK: Dasgupta. PK. (1995). Continuous automated measurement of the soluble fraction of
atmospheric paniculate matter. Anal Chem 67: 71-78. http://dx.doi.org/10.1021/ac00097a012
Singer. BC: Hodgson. AT; Hotchi. T; Kim. JJ. (2004). Passive measurement of nitrogen oxides to assess
traffic-related pollutant exposure for the East Bay Children's Respiratory Health Study. Atmos Environ
38: 393-403. http://dx.doi.Org/10.1016/i.atmosenv.2003.10.005
Singh. HB; Salas. L; Herlth. D: Kolyer. R; Czech. E; Avery. M; Crawford. JH; Pierce. RB; Sachse. GW:
Blake. PR; Cohen. RC: Bertram. TH; Perring. A; Wooldridge. PJ: Dibb. J: Huev. G: Hudman. RC:
Turquetv. S: Emmons. LK; Flocke. F; Tang. Y; Carmichael. GR; Horowitz. LW. (2007). Reactive
nitrogen distribution and partitioning in the North American troposphere and lowermost stratosphere. J
Geophys Res 112: D12S04. http://dx.doi.org/10.1029/2006JD007664
Skalska. K; Miller. JS: Ledakowicz. S. (2010). Trends in NO(x) abatement: a review [Review]. Sci Total
Environ 408: 3976-3989. http://dx.doi.Org/10.1016/i.scitotenv.2010.06.001
Smith. LA; Mukerjee. S: Chung. KG: Afghani. J. (2011). Spatial analysis and land use regression of VOCs
and NO2 in Dallas, Texas during two seasons. J Environ Monit 13: 999-1007.
http://dx.doi.org/10.1039/cOem00724b
Spengler. JD; Brauer. M; Samet. JM; Lambert. WE. (1993). Nitrous acid in Albuquerque, New Mexico,
homes. Environ Sci Technol 27: 841-845. http://dx.doi.org/10.1021/es00042a005
Spicer. CW: Coutant. RW: Ward. GF; Joseph. DW: Gaynor. AJ; Billick. IH. (1989). Rates and mechanisms
of NO2 removal from indoor air by residential materials. Environ Int 15: 643-654.
http://dx.doi.org/10.1016/0160-4120(89)90087-l
Stavrakou. T; Miiller. JF; Boersma. KF; De Smedt. I; Van Per a. RJ. (2008). Assessing the distribution and
growth rates of NO x emission sources by inverting a 10-year record of NO 2 satellite columns.
Geophys Res Lett 35: L10801. http://dx.doi.org/10.1029/2008GL033521
Steinbacher. M; Zellweger. C: Schwarzenbach. B; Bugmann. S: Buchmann. B; Ordonez. C: Prevot. ASH;
Hueglin. C. (2007). Nitrogen oxides measurements at rural sites in Switzerland: Bias of conventional
measurement techniques. J Geophys Res Atmos 112: D11307. http://dx.doi.org/10.1029/2006JD007971
STi (Sonoma Technology Inc). (2011). Summary of results from near-road NO2 monitoring pilot study.
(STI-910308-4211-FR). Research Triangle Park, NC: U.S. Environmental Protection Agency.
http://www.epa.gov/ttnamtil/files/nearroad/2011decNearRoadNO2PilotStudv.pdf
Stieb. DM; Burnett. RT; Smith-Doiron. M; Brion. O; Shin. HH; Economou. V. (2008). A new
multipollutant, no-threshold air quality health index based on short-term associations observed in daily
time-series analyses. J Air Waste Manag Assoc 58: 435-450.
November 2013 2-124 DRAFT: Do Not Cite or Quote
-------
Strak. M; Hoek. G: Godri. KJ; Gosens. I; Mudwav. IS: van Oerle. R; Spronk. HM; Cassee. FR; Lebret E;
Kelly. FJ; Harrison. RM; Brunekreef. B; Steenhof. M; Janssen. NA. (2013). Composition of PM affects
acute vascular inflammatory and coagulative markers - the RAPTES project. PLoS ONE 8: e58944.
http://dx.doi.org/10.1371/iournal.pone.0058944
Strickland. MJ; Darrow. LA; Klein. M; Flanders. WD; Sarnat. JA; Waller. LA; Sarnat. SE; Mulholland. JA;
Tolbert. PE. (2010). Short-term associations between ambient air pollutants and pediatric asthma
emergency department visits. Am J Respir Crit Care Med 182: 307-316.
http://dx.doi.org/10.1164/rccm.200908-1201OC
Strickland. MJ: Darrow. LA; Mulholland. JA: Klein. M; Flanders. WD: Winquist. A; Tolbert. PE. (2011).
Implications of different approaches for characterizing ambient air pollutant concentrations within the
urban airshed for time-series studies and health benefits analyses. Environ Health Global Access Sci
Source 10: 36. http://dx.doi.org/10.1186/1476-069X-10-36
Su. JG: Brauer. M; Ainslie. B; Steyn. D; Larson. T; Buzzelli. M. (2008a). An innovative land use regression
model incorporating meteorology for exposure analysis. Sci Total Environ 390: 520-529.
http://dx.doi.0rg/10.1016/i.scitotenv.2007.10.032
Su. JG: Brauer. M; Buzzelli. M. (2008b). Estimating urban morphometry at the neighborhood scale for
improvement in modeling long-term average air pollution concentrations. Atmos Environ 42: 7884-
7893. http://dx.doi.0rg/10.1016/i.atmosenv.2008.07.023
Su. JG: Jerrett M; Beckerman. B. (2009a). A distance-decay variable selection strategy for land use
regression modeling of ambient air pollution exposures. Sci Total Environ 407: 3890-3898.
http://dx.doi.0rg/10.1016/i.scitotenv.2009.01.061
Su. JG: Jerrett. M; Beckerman. B; Wilhelm. M; Ghosh. JK; Ritz. B. (2009b). Predicting traffic-related air
pollution in Los Angeles using a distance decay regression selection strategy. Environ Res 109: 657-
670. http://dx.doi.0rg/10.1016/j.envres.2009.06.001
Suh. HH; Zanobetti. A. (201 Ob). Exposure error masks the relationship between traffic-related air pollution
and heart rate variability. J Occup Environ Med 52: 685-692.
http://dx.doi.org/10.1097/JOM.Ob013e3181e8071f
Szpiro. AA; Paciorek. CJ: Sheppard. L. (2011). Does more accurate exposure prediction necessarily
improve health effect estimates? Epidemiology 22: 680-685.
http://dx.doi.org/10.1097/EDE.Ob013e3182254cc6
Takahashi. N; Yamazaki. K; Sobukawa. H; Shinjoh. H. (2007). The low-temperature performance of NOx
storage and reduction catalyst. Appl Catal B-Environ 70: 198-204.
http://dx.doi.0rg/10.1016/i.apcatb.2005.10.029
Thornton. JA: Braban. CF; Abbatt. JPD. (2003). N2O5 hydrolysis on sub-micron organic aerosol: the effect
of relative humidity, particle phase, and particle size. Phys Chem Chem Phys 5: 4593-4603.
Thornton. JA: Kercher. JP; Riedel. TP; Wagner. NL; Cozic. J: Hollowav. JS: Dube. WP; Wolfe. GM;
Quinn. PK; Middlebrook. AM; Alexander. B; Brown. SS. (2010). A large atomic chlorine source
inferred from mid-continental reactive nitrogen chemistry. Nature 464: 271-274.
http://dx.doi.org/10.1038/nature08905
Timonen. KL; Vanninen. E; De Hartog. J: Ibald-Mulli. A; Brunekreef. B; Gold. PR; Henrich. J: Hoek. G:
Lanki. T; Peters. A; Tarkiainen. T; Tiittanen. P; Kreyling. W: Pekkanen. J. (2006). Effects of ultrafine
and fine paniculate and gaseous air pollution on cardiac autonomic control in subjects with coronary
artery disease: The ULTRA study. J Expo Sci Environ Epidemiol 16: 332-341.
http://dx.doi.org/10.1038/sj.jea.7500460
Tokairin. T; Kitada. T. (2005). Study on the effect of porous fence on air quality and traffic noise level
around a double-decked road structure. Environ Monit Assess 105: 121-143.
http://dx.doi.org/10.1007/sl0661-005-3160-9
November 2013 2-125 DRAFT: Do Not Cite or Quote
-------
Tolbert. PE; Klein. M; Peel. JL; Sarnat. SE; Sarnat. JA. (2007). Multipollutant modeling issues in a study of
ambient air quality and emergency department visits in Atlanta. J Expo Sci Environ Epidemiol 17: S29-
S35. http://dx.doi.org/10.1038/si.ies.7500625
Tonnesen. S: Jeffries. HE. (1994). Inhibition of odd oxygen production in the carbon bond four and generic
reaction set mechanisms. Atmos Environ 28: 1339-1349. http://dx.doi.org/10.1016/1352-
2310(94)90281-X
Tonse. SR; Brown. NJ; Harlev. RA; Jin. L. (2008). A process-analysis based study of the ozone weekend
effect. Atmos Environ 42: 7728-7736.
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park,NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (1996). Air quality criteria for ozone and related
photochemical oxidants [EPA Report]. (EPA/600/P-93/004AF). Research Triangle Park, NC.
U.S. EPA (U.S. Environmental Protection Agency). (1998b). SLAMS/NAMS/PAMS network review
guidance [EPA Report]. (EPA/454/R-98/003). Research Triangle Park, NC.
U.S. EPA (U.S. Environmental Protection Agency). (2006). Air quality criteria for ozone and related
photochemical oxidants [EPA Report]. (EPA/600/R-05/004AF). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=149923
U.S. EPA (U.S. Environmental Protection Agency). (2008a). Integrated science assessment for oxides of
nitrogen - Health criteria: Annexes [EPA Report]. (EPA/600/R-08/072). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for particulate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2010a). Final regulatory impact analysis (RIA) for the
NO2 National Ambient Air Quality Standards (NAAQS) [EPA Report]. Research Triangle Park, NC.
http://www.epa.gov/ttnecasl/regdata/RIAs/FinalNO2RIAfulldocument.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2011). 2008 National Emissions Inventory data and
documentation. Available online at http://www.epa.gov/ttn/chief/net/2008inventory.html (accessed
February 17,2011).
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for ozone and
related photochemical oxidants (final report) [EPA Report]. (EPA/600/R-10/076F). Washington, DC.
http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=247492
Unal. A; Hu. Y; Chang. ME; Talat Odman. M; Russell. AG. (2005). Airport related emissions and impacts
on air quality: Application to the Atlanta International Airport. Atmos Environ 39: 5787-5798.
http://dx.doi.0rg/10.1016/i.atmosenv.2005.05.051
Valin. LC: Russell. AR; Cohen. RC. (2013). Variations of OH radical in an urban plume inferred from NO 2
column measurements. Geophys Res Lett 40: n/a-n/a. http://dx.doi.org/10.1002/grl.50267
van der A. RJ; Eskes. HJ; Boersma. KF; van Noije. TPC: Van Roozendael. M; De Smedt. I; Peters. DHM.
U; Meijer. EW. (2008). Trends, seasonal variability and dominant NOx source derived from a ten year
record of NO2 measured from space. J Geophys Res Atmos 113: D04302.
http://dx.doi.org/10.1029/2007id009021
Van Roosbroeck. S: Li. R; Hoek. G: Lebret. E; Brunekreef. B; Spiegelman. D. (2008). Traffic-related
outdoor air pollution and respiratory symptoms in children: The impact of adjustment for exposure
measurement error. Epidemiology 19: 409-416. http://dx.doi.org/10.1097/EDE.Ob013e3181673bab
November 2013 2-126 DRAFT: Do Not Cite or Quote
-------
Vardoulakis. S: Lumbreras. J: Solazzo. E. (2009). Comparative evaluation of nitrogen oxides and ozone
passive diffusion tubes for exposure studies. Atmos Environ 43: 2509-2517.
http://dx.doi.0rg/10.1016/i.atmosenv.2009.02.048
Venkatram. A; Isakov. V; Thoma. E; Baldauf. R. (2007). Analysis of air quality data near roadways using a
dispersion model. Atmos Environ 41: 9481-9497. http://dx.doi.Org/10.1016/i.atmosenv.2007.08.045
Vichi. F; De Santis. F. (2012). The measurement of the sink properties of triethanolamine (TEA) as a
coating for collecting NO2 by using annular diffusion denuders. Environ Technol 33: 1065-1069.
http://dx.doi.org/10.1080/09593330.2011.606848
Villena. G: Bejan. I; Kurtenbach. R; Wiesen. P; Kleffmann. J. (2011). Development of a new Long Path
Absorption Photometer (LOPAP) instrument for the sensitive detection of NO2 in the atmosphere.
Atmos Meas Tech 4: 1663-1676. http://dx.doi.org/10.5194/amt-4-1663-2011
Villena. G: Bejan. I; Kurtenbach. R; Wiesen. P; Kleffmann. J. (2012). Interferences of commercial NO2
instruments in the urban atmosphere and in a smog chamber. Atmos Meas Tech 5: 149-159.
http://dx.doi.org/10.5194/amt-5-149-2012
Villeneuve. PJ: Chen. L; Rowe. BH; Coates. F. (2007). Outdoor air pollution and emergency department
visits for asthma among children and adults: A case-crossover study in northern Alberta, Canada.
Environ Health Global Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40
von Kuhlmann. R; Lawrence. MG: Poschl. U; Crutzen. PJ. (2004). Sensitivities in global scale modeling of
isoprene. Atmos Chem Phys 4: 1-17.
Vrijheid. M; Martinez. D; Aguilera. I; Bustamante. M; Ballester. F; Estarlich. M; Fernandez-Somoano. A;
Guxens. M; Lertxundi. N; Martinez. MD; Tardon. A; Sunyer. J. (2012). Indoor air pollution from gas
cooking and infant neurodevelopment. Epidemiology 23: 23-32.
http://dx.doi.org/10.1097/EDE.Ob013e31823a4023
Vutukuru. S: Dabdub. D. (2008). Modeling the effects of ship emissions on coastal air quality: A case study
of southern California. Atmos Environ 42: 3751-3764.
http://dx.doi.0rg/10.1016/i.atmosenv.2007.12.073
Walters. LJ; May. MS: Johnson. DE; Macmann. RS: Woodward. WA. (1999). Time-variability of NO x
emissions from Portland cement kilns. Environ Sci Technol 33: 700-704.
http://dx.doi.org/10.1021/es980203b
Wang. M; Beelen. R; Basagana. X; Becker. T; Cesaroni. G: de Hoogh. K; Dedele. A; Declercq. C:
Dimakopoulou. K; Eeftens. M; Forastiere. F; Galassi. C: Grazuleviciene. R: Hoffmann. B; Heinrich. J:
lakovides. M; Kunzli. N; Korek. M; Lindlev. S: Molter. A; Mosler. G: Madsen. C: Nieuwenhuijsen. MJ;
Phuleria. H; Pedeli. X; Raaschou-Nielsen. O; Ranzi. A; Sugiri. D; Stempfelet. M; Tsai. MY; Lanki. T;
Udvardv. O: Varro. MJ: Wolf. K; Weinmayr. G: Yli-Tuomi. T; Hoek. G: Brunekreef. B. (In Press)
Evaluation of land use regression models for NO2 and paniculate matter in 20 European study areas:
the ESCAPE project. Environ Sci Technol. http://dx.doi.org/10.1021/es305129t
Wang. M; Beelen. R; Eeftens. M; Meliefste. K; Hoek. G: Brunekreef. B. (2012). Systematic evaluation of
land use regression models for NO. Environ Sci Technol 46: 4481-4489.
http://dx.doi.org/10.1021/es204183v
Wang. R; Henderson. SB; Sbihi. H; Allen. RW: Brauer. M. (2013). Temporal stability of land use regression
models for traffic-related air pollution. Atmos Environ 64: 312-319.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.09.056
Wang. Y; Wan. Q; Meng. W: Liao. F; Tan. H; Zhang. R. (2011). Long-term impacts of aerosols on
precipitation and lightning over the Pearl River Delta megacity area in China. Atmos Chem Phys 11:
12421-12436. http://dx.doi.org/10.5194/acp-ll-12421-2011
Wang. YJ; Zhang. KM. (2009). Modeling near-road air quality using a computational fluid dynamics
model, CFD-VIT-RIT. Environ Sci Technol 43: 7778-7783. http://dx.doi.org/10.1021/es9014844
November 2013 2-127 DRAFT: Do Not Cite or Quote
-------
Westcott. NE. (1995). Summertime cloud-to-ground lightning activity around major midwestern urban
areas. J Appl Meteorol 34: 1633-1642. http://dx.doi.Org/10.1175/1520-0450-34.7.1633
Wheeler. AJ; Smith-Doiron. M; Xu. X; Gilbert. NL; Brook. JR. (2008). Intra-urban variability of air
pollution in Windsor, Ontario-measurement and modeling for human exposure assessment. Environ
Res 106: 7-16. http://dx.doi.Org/10.1016/i.envres.2007.09.004
Williams. EJ; Lerner. BM; Murphy. PC: Herndon. SC: Zahniser. MS. (2009). Emissions of NOx, SO2, CO,
and HCHO from commercial marine shipping during Texas Air Quality Study (TexAQS) 2006. J
Geophys Res 114: D21306. http://dx.doi.org/10.1029/2009JD012094
Williams. R; Brook. R; Bard. R; Conner. T; Shin. H; Burnett. R. (2012a). Impact of personal and ambient-
level exposures to nitrogen dioxide and paniculate matter on cardiovascular function. Int J Environ
Health Res 22: 71-91. http://dx.doi.org/10.1080/09603123.2011.588437
Williams. R: Jones. P; Croghan. C: Thornburg. J: Rodes. C. (2012b). The influence of human and
environmental exposure factors on personal NO(2) exposures. J Expo Sci Environ Epidemiol 22: 109-
115. http://dx.doi.org/10.1038/ies.2011.20
Wilson. WE; Brauer. M. (2006). Estimation of ambient and non-ambient components of paniculate matter
exposure from a personal monitoring panel study. J Expo Sci Environ Epidemiol 16: 264-274.
http://dx.doi.org/10.1038/sj.jes.7500483
Wilson. WE; Mage. DT; Grant. LD. (2000). Estimating separately personal exposure to ambient and
nonambient paniculate matter for epidemiology and risk assessment: Why and how. J Air Waste Manag
AssocSO: 1167-1183.
Wu. XM; Fan. ZT; Ohman-Strickland. R (2010). Time-location patterns of a population living in an air
pollution hotspot. J Environ Public Health 2010: 625461. http://dx.doi.org/10.1155/2010/625461
Xie. Y; Paulot. F; Carter. WPL; Nolte. CG: Luecken. DJ; Hutzell. WT; Wennberg. PO; Cohen. RC: Finder.
RW (2013). Understanding the impact of recent advances in isoprene photooxidation on simulations of
regional air quality. Atmos Chem Phys 13: 8439-8455. http://dx.doi.org/10.5194/acp-13-8439-2013
Xue. J: McCurdv. T; Spengler. J: Ozkaynak. H. (2004). Understanding variability in time spent in selected
locations for 7-12-year old children. J Expo Anal Environ Epidemiol 14: 222-233.
http://dx.doi.org/10.1038/sj.jea.7500319
Yokelson. RJ; Crounse. JD; Decarlo. PF; Karl. T; Urbanski. S: Atlas. E; Campos. T; Shinozuka. Y;
Kapustin. V; Clarke. AD; Weinheimer. A; Kna. DJ: Montzka. DP; Hollowav. J: Weibring. P; Flocke. F;
Zheng. W: Toohev. D; Wennberg. PO: Wiedinmyer. C: Mauldin. L; Fried. A; Richter. D; Walega. J:
Jimenez. JL; Adachi. K; Buseck. PR; Hall. S. R.: Shetter. R. (2009). Emissions from biomass burning in
the Yucatan. Atmos Chem Phys 9: 5785-5812. http://dx.doi.org/10.5194/acp-9-5785-2009
Yokelson. RJ: Karl. T; Artaxo. P; Blake. PR; Christian. TJ; Griffith. DWT; Guenther. A; Hao. WM. (2007).
The Tropical Forest and Fire Emissions Experiment: overview and airborne fire emission factor
measurements. Atmos Chem Phys 7: 5175-5196. http://dx.doi.org/10.5194/acp-7-5175-2007
Yuan. T; Remer. LA; Bian. H; Ziemke. JR; Albrecht. R: Pickering. KE; Oreopoulos. L; Goodman. SJ: Yu.
H; Allen. DJ. (2012). Aerosol indirect effect on tropospheric ozone via lightning. J Geophys Res Atmos
117: D18213. http://dx.doi.org/10.1029/2012JD017723
Zeger. SL; Thomas. D: Dominici. F; Samet. JM; Schwartz. J: Dockerv. D; Cohen. A. (2000). Exposure
measurement error in time-series studies of air pollution: Concepts and consequences. Environ Health
Perspect 108: 419-426.
Zhou. Y; Lew. JL (2007). Factors influencing the spatial extent of mobile source air pollution impacts: A
meta-analysis [Review]. BMC Public Health 7: 89. http://dx.doi.org/10.1186/1471-2458-7-89
Zhu. DZ; Nussbaum. NJ; Kuhns. HP; Chang. MCO: Sodeman. D; Moosmuller. H; Watson. JG. (2011).
Real-world PM, NO(x), CO, and ultrafme particle emission factors for military non-road heavy duty
diesel vehicles. Atmos Environ 45: 2603-2609. http://dx.doi.Org/10.1016/i.atmosenv.2011.02.032
November 2013 2-128 DRAFT: Do Not Cite or Quote
-------
Zhu. Y; Fung. DC: Kennedy. N; Hinds. WC: Eiguren-Fernandez. A. (2008). Measurements of ultrafine
particles and other vehicular pollutants inside a mobile exposure system on Los Angeles freeways. J Air
Waste Manag Assoc 58: 424-434.
Zidek. JV: Shaddick. G: Meloche. J: Chatfield. C: White. R. (2007). A framework for predicting personal
exposures to environmental hazards. Environ Ecol Stat 14: 411-431. http://dx.doi.org/10.1007/sl0651-
007-0028-x
November 2013 2-129 DRAFT: Do Not Cite or Quote
-------
CHAPTER 3 DOSIMETRY AND MODES OF
ACTION FOR INHALED OXIDES OF NITROGEN
3.1 Introduction
1 This chapter has two main purposes. The first is to describe the principles that underlie
2 the dosimetry of NO2 and NO and to discuss factors that influence it. The second is to
3 describe the modes of action that may lead to health effects that will be presented in
4 Chapter 4 and Chapter 5. This chapter is not intended to be a comprehensive overview,
5 but rather, it updates the basic concepts derived from NO2 and NO literature presented in
6 the 1993 AQCD and 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008a. c, 1993) and
7 introduces the recent relevant literature.
8 In Section 3.2. particular attention is given to chemical properties of inhaled NO2 and NO
9 that affect absorption, distribution, metabolism, and elimination. The net contribution of
10 inhaled NO2 and NO and subsequent reaction products are discussed in relation to those
11 endogenously occurring. Because there have been few NO2 dosimetry studies published
12 since the 1993 AQCD (U.S. EPA. 1993). much of that information has been pulled
13 forward into the current document and is discussed in the context of more recent
14 research. The topics of dosimetry and modes of action are bridged by reactions of NO2
15 with components of the extracellular lining fluid (ELF) and by reactions of NO with
16 heme proteins, processes which play roles in both uptake and biological responses.
17 Section 3J3 highlights findings of studies published since the 2008 ISA (U.S. EPA.
18 2008a. c) that provide insight into the biological pathways affected by exposure to NO2
19 and NO. Since common mechanisms lead to health effects from both short- and long-
20 term exposure to NO2 and NO, these pathways are discussed in this chapter rather than in
21 later chapters. The related sections of health effects chapters are indicated. Earlier studies
22 that represent the current state of the science are also discussed. Studies conducted at
23 more environmentally-relevant concentrations of NO2 and NO are of greater interest,
24 since mechanisms responsible for effects at low concentrations may not be identical to
25 those occurring at high concentrations. Some studies at higher concentrations are
26 included if they were early demonstrations of key mechanisms or if they are recent
27 demonstrations of potentially important new mechanisms.
November 2013 3-1 DRAFT: Do Not Cite or Quote
-------
3.2 Dosimetry of Inhaled Oxides of Nitrogen
3.2.1 Introduction
1 This section provides a brief overview of NO2 and NO dosimetry and updates
2 information provided in the 2008 ISA (U.S. EPA, 2008c). Dosimetry refers to the
3 measurement or estimation of the amount of a compound, or its reaction products,
4 absorbed and/or generated at specific sites in the respiratory tract during an exposure.
5 New to this ISA is the inclusion of basic information regarding the endogenous
6 production of NO2 and NO. It is important to consider the net contribution of inhaled
7 NO2 and NO and subsequent reaction products in relation to those endogenously
8 occurring.
9 Ambient NO2 concentrations are highest in the winter months near major roadways
10 during weekday morning hours and decrease moderately during the afternoon (see
11 Atlanta, GA data in Figure 2-14 and Figure 2-15). One-hour average, near-road
12 (15 meters) NO2 concentrations in Los Angeles, CA range from 3 ppb to 80 ppb with
13 median values of about 40 ppb in the winter and 30 ppb in the summer months of 2009
14 (Polidori and Fine. 2012). Away from major roadways, 1-hour average NO2
15 concentrations may still reach 50 to 70 ppb with median NO2 concentrations between
16 roughly 10 to 30 ppb depending on the season and distance from roadways (Polidori and
17 Fine. 2012). As will be discussed, due to its high reactivity, it is unlikely that these
18 concentrations of inhaled NO2 will diffuse through the ELF to reach the respiratory tract
19 epithelium and less likely that NO2 itself becomes systemically distributed. Therefore
20 endogenous steady state levels of NO2 in distant tissues are unlikely to be affected by
21 inhaled NO2 at ambient concentrations. The balance of reaction products from inhaled
22 NO2 relative to endogenous levels will also be considered.
23 Similar to NO2, ambient NO concentrations are highest in the winter months near major
24 roadways during weekday morning hours, but decrease to very low levels during the
25 afternoon (see Atlanta, GA data in Figure 2-14 and Figure 2-15). One-hour average, near-
26 road (15 meters) NO concentrations in Los Angeles, CA range from 0 ppb to over 400
27 ppb with median values of about 50 ppb in the winter and 20 ppb in the summer months
28 of 2009 (Polidori and Fine. 2012). Away from major roadways, 1-hour average NO
29 concentrations may still reach 250 ppb, but median NO concentrations are 5 ppb or less
30 (Polidori and Fine. 2012). Comparison of NOX data from Zhu et al. (2008) for the same
31 roadway (Interstate 710), though on a different year, with that from Polidori and Fine
32 (2012) suggests that on-road NO concentrations may exceed those near road. As will be
November 2013 3-2 DRAFT: Do Not Cite or Quote
-------
1 discussed these ambient NO concentrations are generally in the range of those occurring
2 endogenously in the respiratory tract.
3.2.2 Dosimetry of NO2
3 NO 2 is a highly reactive gas that occurs as a free radical wherein, although technically a
4 resonance structure, the unpaired electron is more localized to the nitrogen atom than
5 either of the oxygen atoms. Once inhaled, NO2 first encounters the aqueous phase of the
6 ELF, which is a contiguous but biologically complex aqueous fluid layer that covers the
7 entire respiratory tract surfaces (Bastacky et al.. 1995). The ELF constituent composition
8 shows appreciable heterogeneity with respect to anatomic site and species. Furthermore,
9 both the alveolar surfaces and the conducting airway surfaces have a monomolecular
10 layer of surface active lipids (Bernhard et al.. 2004; Hohlfeld. 2002; Mercer et al.. 1994).
11 largely fully saturated, which reduce surface tension and may provide a resistive barrier
12 to the interfacial transfer of NO2 (see below). Upon dissolution into the ELF, NO2 is
13 converted from a gas to a non-electrolyte solute, and thus becomes subject to partitioning
14 and reaction/diffusion characteristics like all small molecules. Thus, the ELF represents
15 the initial barrier between NO2 contained within the intra-respiratory tract gas phase and
16 the underlying epithelia (Postlethwait and Bidani. 1990). NO2 chemically interacts with
17 antioxidants, unsaturated lipids, and other compounds in the ELF. It preferentially reacts
18 with one electron donors (e.g., small molecular weight antioxidants, protein thiols, etc.),
19 undergoes radical-radical addition reactions, may also abstract allylic hydrogen atoms
20 from polyunsaturated fatty acids and, through a complex series of reactions, can add to
21 unsaturated fatty acids to generate nitrolipids (Bonacci etal.. 2012; Rudolph etal.. 2010;
22 O'Donnell et al.. 1999). The compounds thought responsible for pulmonary effects of
23 inhaled NO2 are the reaction products themselves or the metabolites of these products in
24 the ELF. Quantifications of absolute NO2 absorption reported in the 1993 AQCD and the
25 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c. 1993) are briefly discussed below
26 for thoroughness.
3.2.2.1 Mechanisms of Absorption of NO2
27 At the time of the 1993 AQCD (U.S. EPA. 1993). it was thought that inhaled NO2
28 probably reacted with the water molecules in the ELF to form nitrous acid (HNO2) and
29 nitric acid (HNO3). However, some limited data suggested that the absorption of NO2
30 was linked to reactive substrates in the ELF and subsequent nitrite (NO2~) production. By
31 the time of the 2008 ISA (U.S. EPA. 2008c). chemical reactions between NO2 with ELF
November 2013 3-3 DRAFT: Do Not Cite or Quote
-------
1 constituents were more readily recognized as governing NO2 absorption in the respiratory
2 tract.
3.2.2.1.1 Reaction with ELF Water
3 Previous studies have demonstrated that it is not NO2 but instead it is the NO2 dimer,
4 N2O4, that reacts with water to yield nitrite and nitrate (NO3~) (Schwartz and White.
5 1983; England and Corcoran. 1974) with more recent sophisticated analyses introducing
6 greater detail (Finlayson-Pitts et al.. 2003). However, in aqueous solutions, NO2
7 undergoes rapid reaction with many solutes, in particular with solutes that are easily
8 oxidized. Furthermore, at environmentally relevant concentrations of NO2, the direct
9 reactions of NO2 with dissolved substrates also become important because, at
10 equilibrium, there is very little N2O4 compared to NO2. For example, using the delta
11 Gibbs energies of formation of gaseous NO2 and N2O4 (Chase. 1998), one can calculate
12 that at equilibrium, when the concentration of NO2 is 1,000 ppb and 100 ppb, there are
13 1.48 x 105 and 1.48 x 106, respectively, molecules of NO2 for each molecule of N2O4.
14 Thus, at environmental exposure levels there are approximately 1.5 million NO2
15 molecules for each N2O4 molecule. At these concentrations it is far more likely for NO2
16 (compared to N2O4) to penetrate into the aqueous milieu. Ensuing reactions of NO2 with
17 dissolved reactive substrates also become more likely than reaction with a second NO2
18 molecule (to form N2O4). Although during reactive uptake by pure water all reaction
19 occurs via N2O4 regardless of the concentration of NO2, this process becomes unlikely,
20 and instead occurs via direct reactions of NO2, in the presence of dissolved reactive
21 substrates and at low, environmentally relevant concentrations of NO2. The latter
22 conditions resemble reactive uptake of NO2 by the ELF that would entail direct reactions
23 of NO2, for example, with dissolved small molecular weight antioxidants like
24 glutathione, ascorbate, or urate.
25 Enami et al. (2009) revisited the discussions regarding NO2 reaction with water versus
26 ELF solutes. Because the authors postulate that NO2 effects are largely due to nitrate
27 formation and acidification via proton production, this issue warrants some discussion.
28 The claim by Enami et al. (2009) that "antioxidants catalyze the hydrolytic
29 decomposition of NO2 .. .but are not consumed in the process" is disconcerting in view of
30 the vast existing environmental health literature that regards NO2 as an oxidant gas
31 (Prvor et al.. 2006: Augusto et al.. 2002: Ford et al.. 2002: Kirsch et al.. 2002: Wardman.
32 1998; Postlethwait et al.. 1995; Huie. 1994; Netaetal.. 1988; Finlavson-Pitts et al.. 1987;
33 Kikugawa and Kogi. 1987; Priitzetal.. 1985; Pryor and Lightsev. 1981). Enami et al.
34 (2009) only measured nitrate and thereby these data do not strongly support their
35 contention, except to suggest perhaps that some hydrolysis of NO2 may be occurring
November 2013 3-4 DRAFT: Do Not Cite or Quote
-------
1 since nitrate was detected. Moreover, nitrite data are important because any excess nitrite
2 formed (reaction with water generally yield a 1:1 ratio of nitrite and nitrate; thus, a yield
3 of nitrite above 1 would be considered in excess) would be a main product formed in
4 many one-electron oxidations by NO2. Thus, by not measuring nitrite, an important index
5 to assess oxidation by NO2 was missed.
6 It should also be noted that Enami et al. (2009) conducted their experiments in the
7 absence of oxygen which is important regarding the applicability of their model to the
8 lung. At environmentally relevant concentrations and physiologic temperatures
9 intrapulmonary gas phase NO2 will exist in its monomeric form, plus in the presence of
10 aqueous phase reactive substrates, nitrite, but little or no nitrate, is formed during
11 controlled in vitro exposures. Thus, broad reactivity of NO2 with a diversity of reactive
12 substrates (solutes) within the ELF facilitates chemical interactions with antioxidants,
13 lipids, and proteins/peptides/amino acids.
3.2.2.1.2 Governing Determinants of NO2 Absorption within the Respiratory
Tract
14 The absorption of inhaled NO2 into the ELF is governed by a process termed "reactive
15 absorption" that involves dissolution followed by chemical reaction with ELF reactive
16 substrates (Postlethwait and Bidani. 1990). as well as reactions within the interfacial
17 region. Due to the limited aqueous solubility of NO2 and thus the rapid saturation of the
18 aqueous phase interfacial thin film (Bidani and Postlethwait 1998). the net flux of NO2
19 into reactant-free water is constrained by the relatively slow direct reaction of NO2 with
20 water (see above) relative to its free radical reactions with biological substrates (further
21 discussion below). However, since in the presence of aqueous phase reactants NO2
22 absorption is robust, it is the rapid reactions with ELF substrates that maintain the net
23 driving force for NO2 mass transfer from the intrapulmonary gas phase into the ELF
24 (Bidani and Postlethwait. 1998; Postlethwait and Bidani. 1994; Postlethwait et al.. 1991a;
25 Postlethwait and Bidani. 1990). Concentrations of "free" solute NO2 are likely negligible
26 due to its reaction-mediated removal. Empirical evidence suggests that acute NO2 uptake
27 in the lower respiratory tract is rate-governed by chemical reactions of NO2 with ELF
28 constituents rather than solely by gas solubility in the ELF, wherein the reaction between
29 NO2 and water does not significantly contribute to the absorption of inhaled NO2
30 (Postlethwait and Bidani. 1994. 1990). Absorption was also observed to increase with
31 increasing temperature, an indication of chemical reaction rather than aqueous solubility
32 where solubility increases with temperature decrements (Postlethwait and Bidani. 1990).
33 Postlethwait et al. (1991b) proposed that inhaled NO2 (< 10,000 ppb) did not penetrate
34 the ELF to reach underlying sites and suggested that cytotoxicity likely was initiated by
November 2013 3-5 DRAFT: Do Not Cite or Quote
-------
1 products formed during NO2 reactions with ELF constituents. Subsequently, the reactive
2 absorption of NO2 was examined in a number of studies that sought to identify the
3 substrates that predominantly drive NO2 reactive absorption and to quantify the mass
4 transfer kinetics of NO2 in the respiratory tract. Uptake was observed to be first-order
5 with respect to NO2 at concentrations less than 10,000 ppb, was aqueous substrate-
6 dependent, and was saturable meaning that the absolute amount of NO2 uptake would
7 reach a maximum value even if reactive substrate concentrations were in significant
8 excess (Postlethwait et al.. 1991a. b).
9 The absorption of inhaled NO2 is thought to be coupled with either free radical-mediated
10 hydrogen abstraction to form HNO2 (Postlethwait and Bidani. 1994. 1989) or electron
11 transfer from ELF anionic species that directly reduces NO2 to nitrite (Adgent et al..
12 2012). Both mechanisms produce an organic radical from the initial ELF substrate. At
13 physiologic pH, any formed FiNO2 subsequently dissociates to FT and nitrite. The
14 concentration of the resulting nitrite is likely insufficient to alter physiological function
15 since basal nitrite levels may not change appreciably due to an environmental exposure.
16 Consequently, by default, effects are probably attributable to the organic radical,
17 secondary oxidants formed (Adgent et al.. 2012; Velsor etal.. 2003; Velsor and
18 Postlethwait. 1997) and/or the proton load although the ELF buffering capacity is
19 anticipated to compensate for environmentally-relevant exposure-related proton
20 generation. Nitrite will diffuse into the underlying epithelial cells and vascular space
21 wherein, in the presence of red blood cells, nitrite is oxidized to nitrate (Postlethwait and
22 Bidani. 1989; Postlethwait and Mustafa. 1981).
23 Postlethwait et al. (1995) sought to determine the preferential absorption substrates for
24 NO2 in the ELF lavaged from male Sprague-Dawley rats. Because bronchoalveolar
25 lavage (BAL) fluid collected from rats may be diluted up to 100-fold relative to the
26 native ELF (the dilution will be procedure specific), the effect of concentrating the BAL
27 fluid on NO2 absorption was also investigated. A linear association was found between
28 the first-order rate constant for NO2 absorption and the relative concentration of the BAL
29 fluid constituents. This suggested that concentration of the reactive substrates in the ELF
30 determines, in part, the rate of NO2 absorption. The absorption due to specific ELF
31 constituents was also examined in chemically pure solutions. Albumin, and reduced
32 cysteine, glutathione, ascorbate and urate were the hydrophilic moieties found to be the
33 most active substrates for NO2 absorption. Unsaturated fatty acids (such as oleic, linoleic,
34 and linolenic) were also identified as active absorption substrates and thought to account
35 for up to 20% of NO2 absorption. Vitamins A and E exhibited the greatest reactivity of
36 the substrates that were examined. However, the low concentrations of urate (rodent and
37 some primate ELF contains significantly less urate than humans due to differences in
38 nitrogenous waste metabolism) and vitamins A and E were thought to preclude them
November 2013 3-6 DRAFT: Do Not Cite or Quote
-------
1 from being appreciable substrates in vivo. The authors concluded that ascorbate and
2 glutathione were the primary NO2 absorption substrates in rat ELF. Postlethwait et al.
3 (1995) also found that the pulmonary surfactant, dipalmitoyl phosphatidylcholine, was
4 relatively unreactive towards NO2 but subsequent studies documented that compressed
5 monomolecular interfacial films of dipalmitoyl phosphatidylcholine inhibit NO2
6 absorption in vitro (Connor et al.. 2001). Similar to the bell-shaped dose/response related
7 to NO 2 reaction with antioxidants, documenting whether surface active phospholipids
8 (surfactant) inhibit NO2 mass transfer in vivo is extremely challenging due to the fact that
9 any in situ manipulations that disrupt the surface tension lowering actions of surfactant
10 lead to a plethora of pathophysiologic sequelae. However, even though such potentially
11 important influences on NO2 mass transfer have not been verified in vivo, modeling
12 studies could estimate how such effects would influence the intrapulmonary distribution
13 of inhaled NO2, local mass transfer rates, and thus dosimetry.
3.2.2.1 .3 Reaction/Diffusion of ELF NO2, Potential for Penetration to
Underlying Cells
14 Since rapid ELF reactions constrain the diffusion of solute NO2, exposure-related cellular
15 perturbations within the lung are expected to be related to the ELF-derived products
16 generated during reaction with NO2, rather than by solute NO2 per se. In support of this
17 concept, one can estimate the distance (d) that NO2 is able to diffuse before it chemically
18 reacts with ELF constituent molecules in the ELF (e.g., antioxidants, proteins, lipids, etc.)
19 using the Einstein-Smoluchowski equation:
Equation 3-1
20 Where D is the molecular diffusion coefficient of NO2 and tau (T) is the time that NO2 is
21 allowed to diffuse into the ELF medium which is constrained by its rates of reaction and
22 can be set to its half-life assuming pseudo first-order kinetics will apply.
November 2013 3-7 DRAFT: Do Not Cite or Quote
-------
The transit time (T) then has the form:
ln(2)
Equation 3-2
where the term t-'i Ki^i represents the summation of the products of the rate constants
ki) and concentrations (ct) for all the reactive substances that are present in the ELF.
Replacing T in Equation 3-1 yields:
2D ln(2)
Equation 3-3
5 This approach to estimate the penetration distance into the ELF was originally applied by
6 Pryor (1992) to the lung surface penetration of ozone and later by Ford et al. (2002) for
7 the diffusion distance of NO2 in the cytoplasm and blood plasma. A diffusion coefficient
8 D for NO2 in water at 25 °C equal to 1.4 x 10~9 m2/sec has been reported and will be used
9 in calculations. In the lung, the D for NO2 would be increased by temperature and
10 decreased by the higher viscosity of the ELF compared to water.
11 In considering the classes of ELF biomolecules that react with NO2, one may focus on
12 the water-soluble small molecular weight antioxidants (SMWAOs; e.g., ascorbate, urate,
13 and glutathione), which exist in the ELF in high concentrations and are very reactive
14 toward NO2 and consequently have large kjCj terms. Lipids, on the other hand would not
15 be expected to decrease considerably the transit time of NO2 because only those lipids
16 containing fatty acids with two or more double bonds have significant reactivity towards
17 NO2 and the lipids in the ELF are highly saturated.
18 The reaction rate constants for the SMWAOs of 3.5 x 107 M^sec"1, 2 x 107 M^sec"1, and
19 2 x 107 M^sec"1 were assumed for ascorbate, urate and glutathione, respectively (Ford et
20 al.. 2002). These rates were determined in solution using the pulse radiolysis fast kinetics
21 technique; the kinetics of ascorbate and urate were directly monitored, while for the case
November 2013 3-8 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
8
9
10
of glutathione, ABTS [2,2'-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid)] was used
to produce the intense chromophore ABTS*" from its reaction with the glutathiyl radical.
Species and anatomical loci must be considered when selecting appropriate
concentrations of reactive ELF biomolecules. Table 3-1 illustrates the SMWAO
composition differences between human and rat bronchoalveolar ELF, and the
differences between human nasal and bronchoalveolar ELF (Squadrito et al.. 2010; Van
der Vliet et al.. 1999). Because the ELF only reaches dimensions that NO2 is predicted to
penetrate (i.e., 0.2 to 0.6 um) in isolated regions of the alveolar spaces (Bastacky et al..
1995). it is reasonable to assume that NO2 per se does not directly interact with most
apical surfaces of the respiratory tract epithelial (Postlethwait et al.. 1991b).
Table 3-1 Small molecular weight antioxidant concentrations in ELF and
predicted penetration distances for NO2.
Species - site
Human -
nasal
Human -
bronchoalveolar
Rat-
bronchoalveolar
Ascorbate Urate Glutathione
Substrate Concentration, c, (uM)
28 ±19 225 ±105 <0.5
40 ±18 207 ±167 109 ±64
1,004 ±325 81 ±27 43 ±15
Rate constant, kj (M~1sec~1)
3.5 x 107 2xlQ7 2xlQ7
n
]>><
(sec'1)
5.5 x 103
7.7 x 103
3.8 x 104
ELF
penetration (um)
0.6
0.5
0.2
Substrate concentrations from Van der Vliet et al. (1999) for human and from Squadrito et al. (2010) for rat; Reaction
rate constants from Ford et al. (2002).
3.2.2.2 ELF Interactions with NO2
n
12
13
14
3.2.2.2.1 In Vitro Studies
Small molecular weight antioxidants vary appreciably across anatomic sites and species.
For example, due to the lack of urate oxidase, humans, primates, and select other species
have increased levels of urate and conversely, rodent concentrations of urate are small
compared to humans. Such differences need to be recognized when considering
November 2013
3-9
DRAFT: Do Not Cite or Quote
-------
1 preferential reactive absorption substrates and the profile of products formed via reaction
2 with NO 2. Glutathione and ascorbate are the primary NO2 absorption substrates in rat
3 ELF with near 1:1 stoichiometric yields of NO2 uptake:nitrite formation, suggesting one
4 electron reduction of NO2 is a predominant reaction pathway that also yields the
5 corresponding organic radical (Postlethwait et al.. 1995).
6 Beyond cell-specific differential susceptibility and the airway luminal concentration of
7 NO2, site-specific injury was proposed to depend on rate of bioactive reaction product
8 formation relative to the extent of quenching (detoxification) of these products within the
9 ELF. Velsor and Postlethwait (1997) investigated the mechanisms of acute epithelial
10 injury from NO2 exposure. The maximal levels of membrane oxidation were observed at
11 low antioxidant levels versus null (absent antioxidants) or high antioxidant levels.
12 Glutathione- and ascorbate-related membrane oxidation was superoxide- and hydrogen
13 peroxide-dependent, respectively. The authors proposed that increased absorption of NO2
14 occurred at the higher antioxidant concentrations, but little secondary oxidation of the
15 membrane occurred because the reactive species (e.g., superoxide and hydrogen
16 peroxide) generated during absorption were quenched. A lower rate of NO2 absorption
17 occurred at the low antioxidant concentrations, but oxidants were not quenched and so
18 were available to interact with the cell membrane. Further in vitro analyses also
19 suggested that exposure-related responses may not be strictly linear with respect to the
20 inhaled NO2 dose (concentration and/or time) since the dependence of NO2 absorption
21 and biologic target oxidation demonstrated a bell-shaped function with respect to the
22 initial antioxidant concentration (Adgent et al.. 2012; Velsor et al.. 2003). Since the ELF
23 varies throughout the respiratory tract, the heterogeneous distribution of epithelial injury
24 observed from NO2 exposures may be explained, in part, by the ELF-dependent effects
25 on local NO2 uptake and product formation. However, it should be noted that while these
26 dose/response relationships have been documented in vitro, in vivo validation has not yet
27 been accomplished due to the complexities in reproducibly modulating in situ ELF
28 compositions. Importantly, such results are difficult to directly extrapolate to the in vivo
29 situation as precise rates of NO2 uptake, and thus product formation, are a function of gas
30 phase NO2 concentration, aqueous substrate concentrations, surface area, gas flow and
31 related impacts on boundary layer diffusive resistance, pH, temperature, and others
32 (Adgent etal. 2012: Bidani and Postlethwait. 1998).
November 2013 3-10 DRAFT: Do Not Cite or Quote
-------
3.2.2.2.2 Human Studies
In vivo studies
1 In vitro studies have clearly illustrated the role of antioxidants in mediating NO2 uptake
2 and membrane oxidation; however, the temporal dynamics of biological responses to
3 NO 2 that occur in vivo are far more complex. Recognizing the rapid reactions of inhaled
4 NO 2 with various biological substrates, the short half-life of some primary and secondary
5 reaction products as well as the continuous turnover of the ELF, specific chemical species
6 do not likely persist at any given anatomic locale for any appreciable time.
7 Antioxidant levels vary spatially between lung regions and temporally with NO2
8 exposure. Kelly etal. (1996a) examined the effect of a four-hour NO2 (2,000 ppb)
9 exposure on antioxidant levels in bronchial lavage (BL) fluid and BAL fluid of 44
10 healthy nonsmoking adults (19-45 years, median 24 years). The baseline concentrations
11 of urate and ascorbate were strongly correlated between the BL fluid and BAL fluid
12 within individuals (r = 0.88, p <0.001; r = 0.78, p = 0.001; respectively), whereas the
13 concentrations of glutathione in the BL fluid and BAL fluid were not correlated. At
14 1.5 hours after the NO2 exposure, urate and ascorbate were significantly reduced in both
15 lavage fractions while glutathione levels were significantly increased but only in BL
16 fluid. By 6 hours post-exposure, ascorbate levels had returned to baseline in both lavage
17 fractions, but urate had become significantly increased in both lavage fractions and
18 glutathione levels remained elevated in BL fluid. By 24 hours post-exposure, all
19 antioxidant levels had returned to baseline. The levels of glutathione in BAL fluid did not
20 change from baseline at any time point in response to NO2 exposure. The depletion of
21 urate and ascorbate, but not glutathione has also been observed with ex vivo exposure of
22 human BAL fluid to NO2 (Kelly etal.. 1996R
23 Human and animal results stemming from samples obtained after exposure should be
24 viewed with appropriate caution. As detailed below, secondary reactions within the ELF,
25 sample handling and, importantly, the temporal sequence of exposure relative to sample
26 acquisition may all confound data interpretation. Because the ELF is a dynamic
27 compartment, sample obtained after exposure (>30 minutes) may not reflect biochemical
28 conditions that were present during exposure. This is a critical point as while there is
29 some value in quantifying the net short term effects on ELF composition due to exposure,
30 the biological consequences of exposure are largely a function of the ELF conditions
31 during exposure, which initiate the cascades leading to alterations in cell signaling, cell
32 injury, inflammation, etc. Thus, placing ELF measures within the context of ELF
33 turnover time, clearance of "stable" reaction products, and species generated/regenerated
November 2013 3-11 DRAFT: Do Not Cite or Quote
-------
1 as a consequence of secondary redox reactions should all be incorporated during data
2 interpretation.
Ex vivo studies
3 The depletion of urate and ascorbate, but not glutathione has also been observed with
4 ex vivo exposure of human BAL fluid to NO2. Kelly et al. (1996b) collected BAL fluid
5 from male lung cancer patients (n = 16) and exposed the BAL fluid ex vivo at 37 °C to
6 NO2 (50 to 2,000 ppb; 4 hours) or O3 (50 to 1,000 ppb; 4 hours). Kelly and Tetlev (1997)
7 also collected BAL fluid from lung cancer patients (n = 12; 54 ± 16 years) and exposed
8 the BAL fluid ex vivo to NO2 (50 to 1,000 ppb; 4 hours). Both studies found that NO2
9 depletes urate and ascorbate, but not glutathione from BAL fluid. Kelly et al. (1996b)
10 noted a differential consumption of the antioxidants with urate loss being greater than
11 that of ascorbate which was lost at a much greater rate than glutathione. Kelly and Tetlev
12 (1997) found that the rates of urate and ascorbate consumption were correlated with their
13 initial concentrations in the BAL fluid, such that higher initial antioxidant concentrations
14 were associated with a greater rate of antioxidant depletion. Illustrating the complex
15 interaction of antioxidants, these studies also suggest that glutathione oxidized by NO2
16 may be again reduced by urate and/or ascorbate.
17 Nonetheless, such results must be placed in the context of secondary redox reactions as
18 the reported measurements reflect net effects on individual antioxidants but may lend
19 limited insights into the initial reactions of NO2 within the ELF, and by extension, what
20 bioactive products may be formed and how differences in ELF constituent profiles
21 govern biological outcomes. A clear example is evident in the work of Ford et al. (2002)
22 who characterized the reaction of the glutathione (GSH) radical (GS*) with urate (UH2~)
23 at a pH (6.0) slightly below the recognized ELF pH (-6.8 to 7.0). NO2 more readily
24 reacts with glutathione than urate, producing GS* and nitrite (NO2~). However, the
25 subsequent reaction GS* + UH2 ^ GSH + UH*~ has a rate constant of- 3 x 107 M"1 sec"1
26 which could translate to an initial NO2 reaction with glutathione followed by reduction of
27 the thiyl radical by urate, resulting in an apparent, but potentially inaccurate, conclusion
28 of direct loss of urate during subsequent analyses. In addition, some reports have
29 suggested observations that include detecting significant levels of the ascorbate oxidation
30 product dehydroascorbate (DHA). As with the example of secondary urate oxidation,
31 such observations need to be evaluated with caution as the half-life of DHA under
32 biological conditions is very short (minutes; the ascorbyl radical dismutation produces
33 reduced ascorbate and DHA; and DNA spontaneous decomposes to its keto acid) and
34 since high redox couples are maintained in the ELF and it is constantly turning over due
35 to secretion and mucociliary clearance, it is unlikely that any appreciable accumulation of
November 2013 3-12 DRAFT: Do Not Cite or Quote
-------
1 DHA would occur. Therefore, care must be taken to avoid introducing methodological
2 artifacts (e.g., ascorbate oxidation during sample acquisition, handling, and/or storage)
3 that could significantly confound data interpretation. Consequently, understanding of the
4 precise and preferential substrates is needed to discern the genesis of species differences
5 and the products formed that account for NO2 exposure-related cellular perturbations.
3.2.2.3 Regional and Total Respiratory Absorption of NO2
6 There has been very limited work related to the quantification of NO2 uptake since the
7 1993 AQCD for Oxides of Nitrogen (U.S. EPA. 1993) or the subsequent 2008 ISA (U.S.
8 EPA. 2008c). Consequently, only an abbreviated discussion of this topical area is
9 included.
3.2.2.3.1 Experimental Studies of NO2 Uptake
Upper Respiratory Tract Absorption
10 The nasal uptake of NO2 has been experimentally measured in dogs, rabbits, and rats
11 under conditions of unidirectional flow. Yokoyama (1968) reported 42.1 ± 14.9% (mean
12 ± SD) uptake of NO2 in the isolated nasal passages of two dogs (3.5 L/min) and three
13 rabbits (0.75 L/min) exposed to 4,000 and 41,000 ppb NO2. Uptake did not appear to
14 depend on the exposure concentration and was relatively constant over a 10 to 15 minute
15 period. Cavanagh and Morris (1987) measured uptakes of 28% and 25% uptake of NO2
16 (40,400 ppb) in the noses of four naive and four previously exposed rats (0.10 L/min),
17 respectively. Uptake was not affected by a 4-hour prior exposure (naive versus previously
18 exposed rats) to 40,400 ppb NO2 and was constant over the 24-minute period during
19 which uptake was determined.
20 Kleinman and Mautz (1991) measured the penetration of NO2 through the upper airways
21 during inhalation in six tracheostomized dogs exposed to 1,000 or 5,000 ppb NO2.
22 Uptake in the nasal passages was significantly greater at 1,000 ppb than at 5,000 ppb,
23 although the magnitude of this difference was not reported. The mean uptake of NO2
24 (1,000 ppb) in the nasal passages decreased from 80% to 70% as the ventilation rate
25 increased from about 3 to 7 L/min. During oral breathing, uptake was not dependent on
26 concentration. The mean oral uptake of NO2 (1,000 and 5,000 ppb) decreased from 60%
27 to 30% as the ventilation rate increased from 3 to 7 L/min. Although nasal uptake tended
28 to be greater than oral uptake, the difference was not statistically significant. However,
29 the greater nasal than oral uptake on NO2 is consistent with what is observed for O3 as
November 2013 3-13 DRAFT: Do Not Cite or Quote
-------
1 described in Chapter 5 of the 2013 ISA for Ozone and Related Photochemical Oxidants
2 (U.S.EPA.2013b).
3 Overall, NO2 fractional absorption (uptake efficiency) in the upper respiratory tract is
4 greater in the nasal than oral passage and decreases with increasing ventilation rates. This
5 causes a greater proportion of inhaled NO2 to be delivered to the lower respiratory tract at
6 higher ventilation rates associated with exercise. In humans, the breathing pattern shifts
7 from nasal to oronasal during exercise relative to rest. Since the nasal passages scrub gas
8 phase NO2 more efficiently than the mouth and uptake efficiency decreases with
9 increasing flow, exercise delivers a disproportionately greater quantity of the inhaled
10 mass to the lower respiratory tract, where the NO2 is readily absorbed. Additionally,
11 children tend to have a greater oral breathing contribution than adults at rest and during
12 exercise (Bennett et al.. 2008; Becquemin et al.. 1999). Chadhaetal. (1987) found that
13 the majority (11 of 12) of patients with asthma or allergic rhinitis also breathe oronasally
14 at rest. Thus, compared to healthy adults, children and individuals with asthma might be
15 expected to have greater NO2 penetration into the lower respiratory tract. The dose rate to
16 the lower airways of children compared to adults is increased further because children
17 breathe at higher minute ventilations relative to their lung volumes.
Lower Respiratory Tract Absorption
18 Postlethwait and Mustafa (1989) investigated the effect of exposure concentration and
19 breathing frequency on the uptake of NO2 in isolated perfused rat lungs. To evaluate the
20 effect of exposure concentration, the lungs were exposed to NO2 (4,000 to 20,000 ppb)
21 while ventilated at 50 breaths/min with a VT of 2.0 mL. To examine the effect of
22 breathing frequency, the lungs were exposed to NO2 (5,000 ppb) while ventilated at
23 30-90 breaths/min with a VT of 1.5 mL. All exposures were for 90 minutes. The uptake
24 of NO2 ranged from 59 to 72% with an average of 65% and was not affected by exposure
25 concentration or breathing frequency. A combined regression showed a linear
26 relationship between NO2 uptake and total inspired dose (25 to 330 ug NO2). Illustrating
27 variability in NO2 uptake measurements, Postlethwait and Mustafa (1989) observed 59%
28 NO2 uptake in lungs ventilated at 30 breaths/min with a VT of 1.5 mL, whereas,
29 Postlethwait and Mustafa (1981) measured 35% NO2 uptake for the same breathing
30 condition. In another study, 73% uptake of NO2 was reported for rat lungs ventilated 50
31 breaths/min with a VT of 2.3 mL (Postlethwait et al.. 1992). It should be noted that
32 typical breathing frequencies are around 80, 100, and 160 breaths/min for rats during
33 sleep, rest, and light exercise, respectively (de Winter-Sorkina and Cassee. 2002). Hence,
34 the breathing frequencies at which NO2 uptake has been measured are lower than for rats
35 breathing normally. Furthermore, one must consider the potential impacts of how NO2
36 uptake was measured (mass balance; wet chemical versus automated analyzer that may or
November 2013 3-14 DRAFT: Do Not Cite or Quote
-------
1 may not include a dilution component due to the sampling rate) and the lack of perfusion
2 of the bronchial circulation in isolated rat lungs (Postlethwait et al.. 1990). In addition to
3 measuring upper respiratory tract uptakes, Kleinman and Mautz (1991) also measured
4 NO2 uptake in the lower respiratory tract of tracheostomized dogs. In general, there was
5 about 90% NO2 uptake that was independent of ventilation rates from 3 to 16 L/min.
Total Respiratory Tract Absorption
6 Bauer et al. (1986) measured the uptake of NO2 (300 ppb) in 15 adult asthmatics exposed
7 for 30 minutes (20 minutes at rest, then 10 minutes exercising on a bicycle ergometer) via
8 a mouthpiece during rest and exercise. There was a statistically significant increase in
9 uptake from 72% during rest to 87% during exercise. The minute ventilation also
10 increased from 8.1 L/min during rest to 30.4 L/min during exercise. Hence, exercise
11 increased NO2 uptake by 5-fold in these subjects. In an earlier study of seven healthy
12 adults in which subjects were exposed to a NO2/NO mixture containing 290 to 7,200 ppb
13 NO2 for brief (but unspecified) periods, Wagner (1970) reported that NO2 uptake
14 increased from 80% during normal respiration (VT, 0.4 L) to 90% during maximal
15 respiration (VT, 2 to 4 L). Kleinman and Mautz (1991) also measured the total respiratory
16 tract uptake of NO2 (5,000 ppb) in female beagle dogs while standing at rest or
17 exercising on a treadmill. The dogs breathed through a small face mask. Total respiratory
18 tract uptake of NO2 was 78% during rest and increased to 94% during exercise. In large
19 part, this increase in uptake may be due to the increase in VT from 0.18 L during rest to
20 0.27 L during exercise. Coupled with an increase in minute ventilation from 3.8 L/min
21 during rest to 10.5 L/min during exercise, the uptake rate of NO2 was 3-fold greater for
22 the dogs during exercise than rest.
3.2.2.3.2 Dosimetry Models of NO2 Uptake
23 There is a paucity of theoretical studies investigating NO2 dosimetry. The original
24 seminal dosimetry models of Miller etal. (1982) were developed before much of the
25 above information regarding NO2 reaction/diffusion within the ELF had been obtained.
26 In this model, there was a strong distinction between uptake and dose. Uptake referred to
27 the amount of NO2 being removed from gas phase per lung surface area (ug/cm2),
28 whereas, dose referred to the amount of NO2 per lung surface area (ug/cm2) that diffused
29 through the ELF and reached the underlying tissues.
30 Miller et al. (1982) and subsequently Overton (1984) did not attempt to predict the
31 amount of reactants in the ELF or the transport of reactants to the tissues. They assumed
32 reactions of NO2 with constituents in the ELF as protective in that these reactions
November 2013 3-15 DRAFT: Do Not Cite or Quote
-------
1 reduced the flux of NO2 to the tissues. Others have postulated that NO2 reactants formed
2 in the ELF, rather than NO2 itself, could actually cause adverse responses fVelsor and
3 Postlethwait 1997; Postlethwait and Bidani. 1994; Overton. 1984). Two studies
4 examined the influence of age on reactive gas dosimetry in humans (Ginsberg et al..
5 2005; Sarangapani et al.. 2003). Overall, these modeling studies predict that the net NO2
6 uptake (NO2 flux to air-liquid interface) is relatively constant from the trachea to the
7 terminal bronchioles and then rapidly decreases in the pulmonary region. The pattern of
8 net NO2 uptake rate is expected to be similar between species and unaffected by age in
9 humans. However, the NO2 uptake per unit surface area may be several times higher in
10 infants compared to adults, due to the fact that children under age 5 have much a much
11 smaller airway surface area in the extrathoracic (nasal) and alveolar regions (Sarangapani
12 et al.. 2003).
13 The predicted tissue dose and dose rate of NO2 (NO2 flux to liquid-tissue interface) is
14 low in the trachea, increases to a maximum in the terminal bronchioles and the first
15 generation of the pulmonary region, and then decreases rapidly with distal progression.
16 The site of maximal NO2 tissue dose is predicted to be fairly similar between species,
17 ranging from the first generation of respiratory bronchioles in humans to the alveolar
18 ducts in rats. However, estimates of NO2 penetration in Table 3-1 showed that NO2 is not
19 expected to go deeper than 0.2 to 0.6 um into the ELF before reacting with substrates.
20 The production of toxic NO2 reactants in the ELF and the movement of the reactants to
21 the tissues have not been modeled.
22 Contrary to what in vitro studies have shown (Velsor and Postlethwait. 1997). modeling
23 studies have generally considered NO2 reactions in the ELF to be protective. The
24 complex interactions among antioxidants, spatial differences in antioxidants across
25 respiratory tract regions, temporal changes in ELF constituent levels in response to NO2
26 exposure, and species differences in antioxidant defenses need to be considered in the
27 next generation of dosimetric models. Current dosimetry models are inadequate to put
28 response data collected from animals and humans on a comparative footing with each
29 other and with exposure conditions in epidemiologic studies.
November 2013 3-16 DRAFT: Do Not Cite or Quote
-------
3.2.2.4 Endogenous Generation, Metabolism, Distribution, and
Elimination of NO2
1 Along with CO, NO2 is a criteria pollutant believed to be produced endogenously in the
2 lung.1 This endogenous production and function may have important implications for the
3 interpretation of health effects studies. It is also interesting to note that organisms tend to
4 be less sensitive to endogenously produced oxidants. For example, cells are quite
5 resistant to hydrogen peroxide, an endogenous product of oxygen reduction in aerobic
6 cells. NO2 may be produced endogenously by various processes, including the
7 acidification of nitrite (2 FT + 2 NO2~ ^ 2HNO2 ^ H2O + N2O3 •* NO + NO2) (as can
8 transpire in phagolysosomes), the decomposition of peroxynitrite and/or the
9 nitrosoperoxylcarbonate anion (ONOCT + CO2 -> ONOOCO2" -> CO3 *" + NO2), and
10 the action of peroxidases when using nitrite and H2O2 as substrates. Nitrated proteins
11 occur where tyrosine residues are first oxidized to a tyrosyl radical intermediately
12 followed by radical-radical addition of NO2 to produce 3-nitrotyrosine. NO2 is the
13 terminal nitrating agent and the presence of nitrated proteins provides solid evidence for
14 the endogenous production of NO2 per se. Endogenous NO2 is expected to increase with
15 dietary consumption of nitrite and nitrate (which occurs in substantial concentrations in
16 some leafy vegetables, e.g., spinach) as well as during immune responses and
17 inflammation. There is no known antioxidant enzymatic process for the decomposition of
18 NO2 but this is probably due to the spontaneous reactions that NO2 undergoes with small
19 molecular weight antioxidants, such as glutathione and ascorbate, which forms nitrite and
20 the antioxidant radicals. These reactions are so fast they only allow NO2 to diffuse small
21 distances in the submicrometer range before it reacts (see above. Ford et al., 2002). NO2
22 is only slightly hydrophobic (Squadrito and Postlethwait. 2009) and faces no significant
23 physical barriers to readily traverse biological membranes, but in view of its high
24 reactivity, it is unlikely that NO2 becomes systemically distributed, and therefore its
25 endogenous steady state levels in distant tissues are unlikely to be affected, for example,
26 by inhaled NO2.
27 With regard to the lung, understanding the balance between endogenous products and
28 those derived from inhaled ambient NO2 is a complex and challenging issue. Because
29 inhaled NO2 predominantly undergoes univalent reduction to nitrite during reactive
30 absorption, changes in nitrite concentrations can be used as a surrogate for initial
31 considerations of how inhaled NO2 compares with that produced endogenously. As an
32 example, rat lung ELF contains low (iM to nM levels of nitrite with nitrate being
33 substantially more prevalent. Due to salivary and gut microflora nitrate reductase activity,
1 Evidence in support of a claim for endogenously produced ozone (e.g.. Babior et al.. 2003) has received serious
criticism (Pryoretal.. 2006: Kettle et al.. 2004: Sies. 2004: Smith. 2004) and is here considered controversial. A
useful discussion of the issues can be found in Drahl (2009).
November 2013 3-17 DRAFT: Do Not Cite or Quote
-------
1 in combination with reactions of nitrite, especially with heme proteins, that yield nitrate,
2 there is a constant cyclic flux of nitrite <-» nitrate with nitrate being the primary excretion
3 product in urine. In a rat, with numerous simplifying conditions, assuming a gas phase
4 concentration of 200 ppb NO2, a minute ventilation of 150 mL/min, an exposure time of
5 4 hours, quantitative conversion of NO2 to nitrite, 70% uptake efficiency, an ELF volume
6 of 150 (iL, and no ELF clearance [even though nitrite has been shown to diffuse out of
7 the ELF quickly (Postlethwait and Bidani. 1989)1. this would result in the net
8 accumulation of approximately 0.3 (imoles of nitrite. If the NO2-derived nitrite were
9 evenly distributed throughout the ELF pool, this would equate to an additional 2 mM
10 concentration of nitrite. However, in vitro studies using isolated lungs have not reported
11 increases of this magnitude consequent to 10,000-20,000 ppb NO2 exposures,
12 demonstrating that the ELF is a dynamic compartment and that small molecular weight
13 reaction products (even though charged) move readily from the respiratory tract surface
14 to the vascular space, further evidenced by nitrate which does not undergo many of the
15 numerous reaction pathways possible for nitrite. Both nitrite and nitrate levels are very
16 diet dependent and diet represents the primary source for both. Estimates of nitrite/nitrate
17 stemming from NO production via nitric oxide synthase (NOS) suggest that endogenous
18 NO production, even during inflammatory states, is at best modest compared to dietary
19 intake, although under specific conditions plasma levels have been shown to transiently
20 increase due to non-dietary, endogenous biological activities. Thus, in compilation while
21 it is clear that endogenous NO2 is produced, how environmental exposures at current
22 ambient NO2 concentrations impact the overall balance of nitrite and nitrate, and more
23 importantly, compares with endogenous production rates/amounts remains essentially
24 unknown and introduces an appreciable degree of uncertainty when attempting to place
25 low concentration ambient exposures into a biological context.
3.2.3 Dosimetry of NO
26 NO occurs within the respiratory tract gas phase due to: (1) inhalation of ambient NO and
27 (2) off-gassing from its endogenous production within pulmonary tissues, airspace
28 surface inflammatory cells, and blood. The net uptake of NO within the gas exchange
29 regions depends on the balance between the intrapulmonary gas phase concentration
30 (discussed below) relative to the inhaled ambient concentration.
31 While NO exists as a free radical, it is much less reactive than many other radical species
32 except for select chemical interactions that are largely related to radical-radical reactions
33 such as with the superoxide radical anion (O2*~, that produces peroxynitrite; ONOO"),
34 thiyl radicals (e.g., cysteine, Cys*; glutathione, GS*; that produce S-nitrosothiols;
35 RSNO), organic peroxyl radicals (ROO*) (Madei et al.. 2008; Goldstein et al.. 2004). and
November 2013 3-18 DRAFT: Do Not Cite or Quote
-------
1 with heme-containing proteins such as hemoglobin (Pacher et al.. 2007). Although the
2 radical-based reactions generally occur at near diffusion controlled rates, the prevalence
3 of non-NO radical species at any given time is low. Thus, in terms of the overall uptake
4 and tissue diffusion of NO within the lung, interception due to reactions is not expected
5 to consume appreciable amounts of the total NO involved in mass transfer from the
6 alveolar to the vascular space. Inhaled NO uptake occurs against the background of
7 endogenous NO production which is derived primarily from the catalytic activities of the
8 several isoforms of nitric oxide synthase (NOS) (Forstermann and Sessa. 2012).
9 Additional endogenously-generated NO may also occur from the acidification of nitrite in
10 the presence of electron donors, such as within phagolysosomes, by dissociation of S-
11 nitrosothiols, and by complex interactions within red blood cells that likely lead to the
12 release of NO (Weitzberg et al.. 2010). In combination, this results in the appearance of
13 NO within the intrapulmonary gas phase, which can be measured in expired breath and is
14 routinely labeled as either "eNO" or expressed as the fractional amount of expired gas
15 "FeNO".
16 Reported eNO concentrations from the lower respiratory tract span a broad range (~5 to
17 >300 ppb) with nasal/sinus concentrations generally accepted as being greater than what
18 is measured coming from the lower respiratory tract (e.g., See and Christiani. 2013;
19 Alexanderson et al.. 2012; Gelb etal.. 2012; Nodaet al.. 2012; Taylor. 2012; Bautista et
20 al.. 2011; Linhares et al.. 2011; Olinetal.. 1998). eNO has been reported to be affected
21 by a variety of factors including disease state, diet, sex (or height), species, smoking
22 history, environmental exposures, etc. Although eNO from the lower respiratory tract is
23 increased by asthma, this is not the case for nasal NO (ATS/ERS. 2005).
24 For the general U.S. population, results of 2007-2011 NHANES survey show a geometric
25 mean eNO of 9.7 ppb in children (n = 1855; 6-11 years of age; 10% with current asthma)
26 and 13.3 ppb in teenagers and adults (n = 11,420; 12-80 years of age; 8% with current
27 asthma) (See and Christiani. 2013). In healthy, never smokers (558M, 573F; 25-75 years
28 of age), Olin et al. (2007) reported a geometric mean eNO of 16.6 ppb (95% reference
29 interval, 6 to 47 ppb). The eNO levels increased with age and height of the individuals,
30 but did not depend on sex. In healthy children (23M, 28F, 1-5 years of age), a geometric
31 mean eNO of 7 ppb (95% CI: 3, 12) has been reported (van der Heijden et al.. In Press).
32 The eNO levels in these children were unrelated to age, height, weight or sex. These eNO
33 levels correspond to NO output rates of about 40-50 nL/min from the lower respiratory
34 tract of healthy adults and about 20-30 nL/min for healthy children.
35 Kharitonov et al. (2005) reported nasal NO concentrations of 750 ppb (95% CI: 700, 810)
36 in children (n = 20; 10 ± 3 [SD] years) and 900 ppb (95% CI: 870, 930) in adults (n = 29;
37 38 ± 11 years). Another study of healthy adults (n = 10; 18-35 years of age) found a nasal
November 2013 3-19 DRAFT: Do Not Cite or Quote
-------
1 NO concentration of 670 ppb. Higher NO concentrations (9,100 ± 3,800 ppb; n = 5) have
2 been reported for the paranasal sinuses of healthy adults (Lundberg et al.. 1995). Asthma
3 and current rhinitis do not appear to affect nasal NO concentrations (Alexanderson et al..
4 2012: Kharitonov et al.. 2005). Nasal NO is reduced by exercise (ATS/ERS. 2005). The
5 nasal NO concentrations described above correspond to NO output rates of about 300
6 nL/min for the nasal airways of adults with or without asthma and 230 nL/min for
7 children with or without asthma. Nasal NO output rates of healthy primates in the range
8 of 200 to 450 nL/min (ATS/ERS. 2005). With a NO output of 730 nL/min, a large
9 contribution to nasal NO appears to derive from the paranasal sinuses. Based on these
10 NO output rates, the nasal passages may contribute, on average, roughly 15-20 ppb NO to
11 the lower respiratory tract during rest.
12 The other primary approach to non-invasive assessment of the respiratory tract surface is
13 expired breath condensate (EEC) which captures aerosolized materials contained in
14 exhaled air, including those directly related to reactive nitrogen chemistry; e.g., nitrite,
15 nitrate, 3-nitrotyrosine. Unfortunately this relatively new field of analyzing exhaled
16 constituents has encountered numerous situations where concentrations of eNO and EEC
17 constituents are unrelated (Ravaetal.. 2012; Dressel etal.. 2010; Malinovschi et al..
18 2009: Cardinale et al.. 2007: Vints etal. 2005: Chambers and Avres. 2001: Olin et al..
19 2001: Zetterquist et al.. 1999: Olin etal.. 1998: Jilmaetal.. 1996). Given the endogenous
20 production of NO and the lack of a correlation between the two measurements, neither
21 eNO nor EEC can be employed as a metric of exposure history with any significant
22 degree of specificity for inhaled ambient NO.
23 The absorption of inhaled NO proceeds similar to oxygen and carbon monoxide. Because
24 blood acts as a near "infinite" sink for NO, it has been proposed as an alternative to CO
25 for measuring pulmonary diffusing capacity (e.g., Chakraborty et al.. 2004: Heller et al..
26 2004). NO absorption follows Henry's law for dissolution into the aqueous phase
27 followed by diffusion into the vascular space where it interacts with RBC hemoglobin to
28 ultimately form nitrate. Thus, due to its chemical conversion, NO net flux from alveolar
29 gas phase to the blood occurs when the alveolar concentration exceeds that found in
30 tissue/blood. Mass transfer resistances may be encountered (Borland et al.. 2010;
31 Chakrabortv et al.. 2004) but their combined effects are likely small due to the ppb
32 concentrations of NO. The formation of RSNO within the ELF may contribute to the
33 overall uptake (Toroketal.. 2012) but it remains unclear the precise extent of
34 contribution since formation of RSNO requires several steps due to the slow direct
35 reactivity of NO with reduced thiols. Ambient NO levels are likely similar to those
36 endogenously occurring within the lung airspaces, except during morning commutes or
37 near major roadways where they may possibly exceed endogenous levels. Consequently,
38 one can reasonably predict that exposure to ambient NO may not significantly affect the
November 2013 3 -20 DRAFT: Do Not Cite or Quote
-------
1 overall sequelae of NO absorption, metabolism, or downstream impacts on vascular
2 homeostasis, or on lung and systemic biological processes. Importantly, it should be
3 noted that in the clinical setting, therapeutic administration is a very different situation
4 wherein > 10,000 ppb NO may be administered continuously for prolonged periods.
3.2.4 Metabolism, Distribution, and Elimination of Products Derived from
Inhaled Oxides of Nitrogen
5 As stated earlier, NO2 absorption may generate some nitrous acid (HNO2), which
6 subsequently dissociates to fT and nitrite. Nitrite enters the underlying epithelial cells
7 and subsequently the blood. In the presence of red blood cells and/or heme proteins,
8 nitrite is oxidized to nitrate (Postlethwait and Mustafa. 1981). Nitrate is the primary oxide
9 of nitrogen stable product subsequently excreted in the urine. There has been concern that
10 inhaled NO2 may lead to N-nitrosamine production, many of which are carcinogenic,
11 since NO2 can produce nitrite and nitrate (in blood). Nitrate can be converted to nitrite by
12 bacterial reduction in saliva, the gastrointestinal tract, and the urinary bladder. Nitrite has
13 been found to react with secondary amines to form N-nitrosamines. This remains
14 speculative since nitrosamines are not detected in tissues of animals exposed by
15 inhalation to NO2 unless precursors to nitrosamines and/or inhibitors of nitrosamine
16 metabolism are co-administered. Rubenchik et al. (1995) could not detect 7V-
17 nitrosodimethylamine (NDMA) in tissues of mice exposed to 4,000 to 4,500 ppb NO2 for
18 1 hour. However, NDMA was found in tissues if mice were simultaneously given oral
19 doses of amidopyrine and 4-methylpyrazole, an inhibitor of NDMA metabolism.
20 Nevertheless, endogenous NO2 production and the cyclic interconversion of nitrite and
21 nitrate may provide the precursors that drive nitrosamine formation, especially since ELF
22 nitrite is swallowed. However, because ambient NO2 contributes only modest amounts of
23 oxides of nitrogen relative to dietary intake, any substantial contribution to systemic
24 nitrosamine formation is not likely. Thus, the relative importance of inhaled NO2 in
25 endogenous 7V-nitrosamine formation has yet to be demonstrated. Metabolism of inhaled
26 NO2 may also transform other chemicals that may be present in the body, in some cases
27 into mutagens and carcinogens. Van Stee et al. (1983) reported 7V-nitrosomorpholine
28 (NMOR), production in mice gavaged with 1 g of morpholine/kg body weight per day
29 and then exposed (5-6 hours daily for 5 days) to 16,500-20,500 ppb NO2. 7V-
30 nitrosomorpholine is a nitrosamine that is a potent animal carcinogen. The single site
31 containing the greatest amount of NMOR was the gastrointestinal tract, as would be
32 expected due to the pH dependent facilitation of TV- nitrosation chemistry. Later, Van Stee
33 et al. (1995) exposed mice to approximately 20,000 ppb 15NO2 and to 1 g/kg morpholine
34 simultaneously. 7V-nitrosomorpholine was found in the body of the exposed mice. Ninety-
35 eight point four percent was labeled with 15N that was derived from the inhaled 15NO2
November 2013 3-21 DRAFT: Do Not Cite or Quote
-------
1 and 1.6% was derived presumably from endogenous sources. Inhaled NO2 may also be
2 involved in the production of mutagenic (and carcinogenic) nitroderivatives of other co-
3 exposed compounds, such as PAHs, via nitration reactions. Miyanishi et al. (1996) co-
4 exposed rats, mice, guinea pigs and hamsters to 20,000 ppb NO2 and various PAHs
5 (pyrene, fluoranthene, fluorene, anthracene, or chrysene). Nitro derivatives of these
6 PAHs, which were found to be highly mutagenic in the Ames/X typhimurium assay, were
7 excreted in the urine of these animals. Specifically, the nitrated metabolite of pyrene (1-
8 nitro-6/8-hydroxypyrene and l-nitro-3hydroxypyrene) was detected in the urine. Further
9 studies indicated that these metabolites are nitrated by an ionic reaction in vivo after the
10 hydroxylation of pyrene in the liver.
3.2.5 Summary
11 The uptake of inhaled NO2 in the respiratory tract is governed by "reactive absorption"
12 that involves chemical reactions with antioxidants, unsaturated lipids, and other
13 compounds in the ELF. In vitro studies have clearly illustrated the role of antioxidants in
14 mediating NO2 uptake. The rapid reactions of NO2 with ELF substrates maintain a net
15 driving force for NO2 mass transfer from the intrapulmonary gas phase into the ELF.
16 Concentrations of "free" solute NO2 are likely negligible due to its reaction-mediated
17 removal. Thus, it is not NO2 itself, but rather its reaction products that are believed to
18 interact with the apical surfaces of the respiratory tract epithelial. At high substrate
19 concentrations, oxidative/cytotoxic products are at least partially quenched due to
20 secondary antioxidant reactions. At low substrate concentration, ELF-derived
21 oxidants/cytotoxic products have a lower probably of being intercepted by unreacted
22 antioxidants and instead may reach underlying targets.
23 Exercise, relative to rest, increases the dose rate of NO2 to the respiratory tract because of
24 greater NO2 penetration through the extrathoracic airways and a greater intake rate of
25 NO2. The uptake of NO2 by the upper respiratory tract decreases with increasing
26 ventilation rates occurring with activity. This causes a greater proportion of inhaled NO2
27 to be delivered to the lower respiratory tract. In humans, the breathing pattern shifts from
28 nasal to oronasal during exercise relative to rest. Since the nasal passages scrub gas phase
29 NO2 more efficiently than the mouth and uptake efficiency decreases with increasing
30 flow, exercise delivers a disproportionately greater quantity of the inhaled mass to the
31 lower respiratory tract, where the NO2 is readily absorbed. Experimental studies have
32 shown exercise increases the dose rate of NO2 to the respiratory tract by 3- to 5-times
33 compared to resting exposures.
November 2013 3 -22 DRAFT: Do Not Cite or Quote
-------
1 Compared to healthy adults, children and individuals with asthma might be expected to
2 have greater NO2 penetration into the lower respiratory tract. Children tend to have a
3 greater oral breathing contribution than adults at rest and during exercise. Limited data
4 also suggest that patients with asthma or allergic rhinitis breathe oronasally at rest. Since
5 the nasal passages scrub gas phase NO2 more efficiently, a greater quantity of the inhaled
6 NO2 may reach the lower respiratory tract of oronasally breathing individuals. The dose
7 rate to the lower airways of children compared to adults is increased further because
8 children breathe at higher minute ventilations relative to their lung volumes.
9 Current dosimetry models for NO2 do not adequately consider reactive absorption and
10 secondary reactions that affect the probability of oxidants/cytotoxic products reaching
11 target sites. It is unclear to what extent environmental exposures at current ambient NO2
12 concentrations might affect the overall balance of nitrite and nitrate or how ambient NO2
13 uptake compares with endogenous production rates/amounts.
14 The uptake of inhaled NO occurs against the background of endogenous NO production.
15 In terms of the overall uptake and tissue diffusion of NO within the lung, interception due
16 to reactions is not expected to consume appreciable amounts of the total NO involved in
17 mass transfer from the alveolar to the vascular space. The absorption of inhaled NO
18 proceeds similar to oxygen and carbon monoxide. Blood acts as a near "infinite" sink for
19 NO. Absorption of NO follows Henry's law for dissolution into the aqueous phase
20 followed by diffusion into the vascular space where it interacts with RBC hemoglobin to
21 ultimately form nitrate. Ambient NO levels are likely similar to those endogenously
22 occurring within the lung airspaces, except during morning commutes or near major
23 roadways where they may possibly exceed endogenous levels. Given the high
24 endogenous levels of NO in the respiratory tract, exposure to ambient NO may not
25 generally affect the overall sequelae of its absorption, metabolism, or downstream
26 impacts on vascular homeostasis or on lung and systemic biological processes.
3.3 Modes of Action for Inhaled Oxides of Nitrogen
3.3.1 Introduction
27 Mode of action refers to a sequence of key events and processes that result in a given
28 toxic effect (U.S. EPA. 2005). Elucidation of mechanisms provides a more detailed
29 understanding of these key events and processes (U.S. EPA. 2005). The purpose of this
30 section of Chapter 3 is to describe the key events and pathways that may contribute to
31 health effects resulting from short-term and long-term exposures to NO2 and NO. Most
32 of the emphasis will be placed on studies of NO2 and NO, the two most prevalent forms
November 2013 3 -23 DRAFT: Do Not Cite or Quote
-------
1 of NOY. The extensive research carried out over several decades in humans and in
2 laboratory animals has yielded numerous studies on mechanisms by which NO2 and NO
3 exert their effects. This section will discuss some of the representative studies with
4 particular emphasis on studies published since the 2008 ISA for Oxides of Nitrogen (U.S.
5 EPA. 2008a. c) and on studies in humans that inform biological mechanisms underlying
6 responses to NO2 and NO.
7 NO 2 is a free radical and a highly reactive oxidant gas (Table 3-2). It is we 11-appreciated
8 that secondary oxidation products, which are formed as a result of NO2 exposure, initiate
9 numerous responses at the cellular, tissue and whole organ level of the respiratory
10 system. Exposure to NO2 may also have effects outside the respiratory tract. NO is a free
11 radical gas that is more selective in its reactivity than NO2 (Table 3-2). Once inhaled, NO
12 rapidly passes through the alveolar capillary barrier into the circulation where it avidly
13 binds to hemoglobin. Subsequent reactions with hemoglobin lead to the generation of
14 circulating nitrate, nitrite, and methemoglobin.
Table 3-2 Chemical properties of NO2 and NO that inform modes of action.
NO2 NO
Free radical gas Free radical gas
Somewhat hydrophobic Very hydrophobic
Very reactive Selectively reactive
Less diffusible More diffusible
Reactions with unsaturated fatty acids, thiols, and low Radical-radical reactions with
molecular weight antioxidants 1 j superoxide to form peroxynitrite
2) thiyl radicals to form S-nitrosothiols
3) organic peroxyl radicals
Reacts with amino acids, proteins and lipids to form Reacts with heme-containing proteins, transition metals and
nitrated species oxygen
Initiates free radical reactions and lipid peroxidation Quenches free radical reactions
Metabolites include nitrite and nitrate Metabolites include nitrite and nitrate
15 Both NO2 and NO are formed endogenously in cells and tissues (Sections 3.2.2.4 and
16 3.2.3). Formation of endogenous NO is catalyzed by nitric oxide synthases (NOS). In
17 addition, three pathways contribute to the formation of endogenous NO2: (1) acidification
18 of nitrite usually occurring in the phagolysosomes, (2) reaction with carbonate to form
19 nitrosoperoxylcarbonate anion which decomposes to carbonate anion and NO2, and
20 (3) the reaction of peroxidases using nitrite and hydrogen peroxide as substrates. These
November 2013 3 -24 DRAFT: Do Not Cite or Quote
-------
1 enzymatic and non-enzymatic pathways are increased during immune responses and
2 inflammation, leading to higher endogenous levels of NO and NO2. Furthermore, dietary
3 consumption of nitrate leads to enhanced circulating levels of NO2 and NO species due to
4 activity of the enterosalivary cycle ("Weitzberg and Lundberg. 2013; Lundberg et al..
5 2011). Interconversion of reactive nitrogen species (i.e., nitrite, nitrate, and NO) has also
6 been demonstrated in tissue and extracellular compartments. The contribution of
7 environmentally-relevant levels of inhaled NO2 and NO to levels of circulating nitrite
8 and nitrate is thought to be minimal. However inhaled NO2 may act on the same targets
9 as endogenous NO2 produced during inflammation in the respiratory tract (Ckless et al.,
10 2011). Since endogenous NO2 is thought to contribute to the development of lung
11 disease, inhaled NO2 may further this process.
12 The following subsections describe the current understanding of potential pathways and
13 modes of action responsible for the pulmonary and extrapulmonary effects of inhaled
14 NO2 and NO. For NO2, this includes the formation of secondary oxidation products,
15 activation of neural reflexes, initiation of inflammation, alteration of epithelial barrier
16 function, enhancement of bronchial smooth muscle reactivity, modification of
17 innate/adaptive immunity and remodeling of airways and alveoli. Mechanisms underlying
18 the extrapulmonary effects of NO2 are not well-understood however activation of neural
19 reflexes and release of NO2 metabolites or mediators from the lung to the bloodstream
20 are possibilities. Inhaled NO may impact the pulmonary and systemic vasculature
21 through interaction with heme proteins. Other effects of NO may be due to circulating
22 nitrite, nitrate, and methemoglobin; due to interactions with redox active transition
23 metals, and due to reactions with thiyl and superoxide radicals. Since endogenous NO is
24 an important mediator of cell signaling, inhaled NO has the potential to disrupt cell
25 signaling.
3.3.2 NO2
3.3.2.1 Formation of secondary oxidation products
26 The 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993) summarized biochemical
27 effects observed in the respiratory tract after NO2 exposure. These effects have been
28 attributed to the strong oxidizing potential of NO2 resulting in the formation of reactive
29 oxygen species (ROS). Key responses include oxidation of membrane polyunsaturated
30 fatty acids, thiol groups and antioxidants. Chemical alterations of lipids, amino acids,
31 proteins and enzymes can lead to functional changes in membranes, enzymes and
32 oxidant/antioxidant status. For example, lipid peroxidation of unsaturated fatty acids in
November 2013 3 -25 DRAFT: Do Not Cite or Quote
-------
1 membranes may alter membrane fluidity and permeability. As a result, epithelial barrier
2 functions may be impaired and phospholipases may be activated leading to the release of
3 arachidonic acid. In addition, oxidation of protein thiols may result in enzyme
4 dysfunction. Further, consumption of low molecular weight antioxidants by NO2 may
5 result in decreased antioxidant defenses. Effects may occur directly through the action of
6 NO 2 or secondarily due to reaction products, such as nitrogen or oxygen radicals,
7 generated via NO2-mediated chemical reactions. Later effects may occur due to release of
8 ROS/RNS by leukocytes responding to cell damage.
9 As summarized in the 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993),
10 considerable attention has been paid to the effects of NO2 on the antioxidant defense
11 system in the ELF and in respiratory tract tissue. Studies employing in vitro systems
12 point to the ability of antioxidants to both react with NO2 to form reactive intermediates
13 and to quench those reactive intermediates species (Velsor and Postlethwait 1997).
14 Studies in humans and animals exposed to NO2 have demonstrated changes in low
15 molecular weight antioxidants such as glutathione, ascorbate and a-tocopherol, and in the
16 activities of enzymes responsible for glutathione synthesis or maintenance of redox
17 status. For example, a controlled human exposure study found depletion of urate and
18 ascorbate but not glutathione in BAL fluid 1.5 hours following a 4-hour exposure to 2000
19 ppb NO2 (Kelly etal., 1996a). While these results may be interpreted as evidence that
20 NO2 prefers to react with urate or ascorbate over glutathione, an alternative interpretation
21 is that glutathione reacts with NO2 and that the product of the reaction is reduced by
22 other antioxidants. Other studies have found that antioxidant status modulates the effects
23 of NO2 inhalation. For example in a controlled human exposure study, supplementation
24 with ascorbate and a-tocopherol decreased the levels of lipid peroxidation products found
25 in BAL fluid following a 3-hour exposure to 4,000 ppb NO2 (Mohsenin. 1991).
26 Additionally, changes in lung antioxidant enzyme activity have been reported in animals
27 exposed to NO2 (U.S. EPA, 2008c). For example, long term exposure to NO2 resulted in
28 decreased glutathione peroxidase activity in weanling mice which were a-tocopherol
29 deficient while supplementation with a-tocopherol resulted in an increase in glutathione
30 peroxidase activity (Ayaz and Csallany. 1978). Thus, NO2 inhalation is capable of
31 perturbing glutathione-dependent reactions. These changes may reflect altered cell
32 populations since injury induced by NO2 exposure may result in the influx of
33 inflammatory cells or the proliferation of resident epithelial or mesenchymal cells.
34 Changes in cell populations due to proliferative repair may also account for changes
35 phase II enzymes involved in antioxidant defense as well as in phase I and glycolytic
36 enzymes which have been observed following NO2 exposure.
37 Nitrite is a primary product of the chemical reactions of NO2 in the respiratory tract. As
38 discussed in Section 3.2.2.1.2. nitrite formed in the ELF can diffuse into respiratory tract
November 2013 3 -26 DRAFT: Do Not Cite or Quote
-------
1 epithelial cells and subsequently into the vascular space. While the effects of nitrite on
2 the epithelial cell are not well known, it is unlikely that nitrite is responsible for the
3 toxicity of NO2. Interestingly, numerous studies have explored the effects of increased
4 systemic nitrite on various tissues and organs. Nitrite has been found to protect against
5 ischemia-reperfusion injury in the heart and other organs (Weitzberg and Lundberg.
6 2013). In addition, systemic nitrite administration prevented airway and epithelial injury
7 due to exposure to chlorine gas in rats (Yadav et al., 2011). Further, nitrite is known to
8 have a direct relaxing effect on smooth muscle (Folinsbee. 1992) suggesting that it may
9 play a role in bronchodilation.
10 Besides nitrite, nitrated proteins, fatty acids and lipids may be formed in the respiratory
11 tract following NO2 exposure although experimental evidence is currently lacking
12 (Sections 3.2.2 and 3.2.2.4). Nitration of proteins may cause inhibition of protein function
13 and/or induce antigenicity. The presence of nitrated amino acids, such as 3-nitrotyrosine,
14 in cells or tissues is viewed as an indicator of endogenous NO2 and peroxynitrite
15 formation. Nitrated (or nitro) fatty acids have a direct relaxing effect on smooth muscle,
16 perhaps even on airway smooth muscle (Que et al.. 2009; Lima et al.. 2005). Further
17 discussion of the biological effects of these products of NO2 metabolism is found in
18 Section 3.3.4.
19 Toxicity resulting from NO2 exposure is likely due to a product derived from the initial
20 ELF substrate and/or secondary oxidants formed. These reaction products may not be
21 long-lived due to short half-lives and/or continuous turnover of theELF. Studies in vitro
22 have demonstrated quenching of NO2-derived secondary oxidants that is dependent on
23 concentrations of antioxidants. Thus quenching of reaction products by ELF antioxidants
24 may limit damage to respiratory epithelium (Velsor and Postlethwait 1997). The
25 heterogeneous distribution of epithelial injury due to NO2 may reflect ELF-dependent
26 local effects, since the ELF is non-uniform in composition and quantity along the
27 respiratory tract.
3.3.2.2 Activation of neural reflexes
28 NO2 is classified as a pulmonary irritant (Alarie. 1973). Pulmonary irritants stimulate
29 afferent nerve endings in the lung resulting in increased respiratory rate and decreased
30 tidal volume and subsequent rapid shallow breathing. Sometimes pulmonary irritants also
31 stimulate mild bronchoconstriction, bradycardia, and hypotension (Alarie. 1973). All of
32 these pathways involve the vagal nerve.
33 In guinea pigs, concentration (5,200-13,000 ppb) and time (2-4 hours) dependent
34 exposures to NO2 by nose-cone resulted in statistically significant stimulated respiratory
November 2013 3 -27 DRAFT: Do Not Cite or Quote
-------
1 rates and decreased tidal volumes that were reversible when animals were returned to
2 clean air (Murphy et al.. 1964). In contrast, no changes in these respiratory parameters
3 were observed with 4-hour exposures to 16,000 and 50,000 ppb NO. Another study in
4 guinea pigs exposed to 7,000-146,000 ppb NO2 for 1 hour demonstrated a concentration-
5 dependent increase in respiratory rate 10 minutes following exposure and a
6 concentration-dependent decrease in tidal volume 10 minutes, 2 hours, and 19 hours
7 following exposure (Silbaugh et al.. 1981). NO2 exposure-induced increases in
8 respiratory rate have also been reported in rats (Freeman et al.. 1966) and mice (McGrath
9 and Smith. 1984). but not in humans (Bvlin et al.. 1985). In mice, statistically significant
10 increases in respiratory rate and decreases in tidal volume were found in response to an 8
11 minute exposure to 100,000 ppb NO2, but not to 15,000 or 50,000 ppb NO2 (McGrath
12 and Smith. 1984). In rats, continuous exposure to 800 ppb and higher concentrations of
13 NO2 resulted in elevated respiratory rates throughout life (Freeman et al.. 1966). In
14 human subjects, respiratory rates tended to decrease in humans exposed to 0-480 ppb for
15 20 minutes. The authors proposed that NO2 did not act as a pulmonary irritant in humans
16 at this exposure level (Bvlinetal.. 1985). In mice, the increase in respiratory rate
17 observed in response to 100,000 ppb NO2 for 8 minutes (McGrath and Smith. 1984) was
18 lessened by continuous pre-exposure to 5,000 ppb NO2 for 3 days, suggesting the
19 development of a tolerance or attenuated response to NO2 (U.S. EPA. 1993).
20 NO2 has been shown to elicit a small increase in airway resistance consistent with mild
21 bronchoconstriction in humans but not rabbits or guinea pigs [(Alarie. 1973) and below].
22 One study in human subjects at rest found a non-monotonic response to NO2 in terms of
23 airway resistance (Bvlin et al., 1985). In this study, airway resistance was increased after
24 20 minutes of exposure to 250 ppb NO2 and was decreased after 20 minutes exposure to
25 480 ppb NO2. The authors suggested that reflex bronchoconstriction occurred at the
26 lower concentration and that other mechanisms counteracted this effect at the higher
27 concentration. It should be noted that no increase in respiratory rate was observed in this
28 study and that the authors proposed that NO2 did not act as a pulmonary irritant at this
29 exposure level. Other controlled human exposure studies found no change in airway
30 resistance with acute exposures of 530-1,100 ppb NO2, and increases in airway resistance
31 with acute exposures above 1,600-2,500 ppb in healthy human subjects (U.S. EPA.
32 1993). Human subjects with chronic lung disease exposed acutely to 2,100 ppb NO2 also
33 exhibited increased airway resistance (von Nieding and Wagner. 1979). In addition, both
34 FEVi and FVC were decreased in healthy human subjects exposed to 2,000 ppb NO2 for
35 4 hours (Blomberg et al.. 1999). These changes in pulmonary function are consistent with
36 reflex bronchoconstriction. Since the response was lessened with each successive
37 exposure on 4 consecutive days, the authors suggested the development of a tolerance or
38 attenuated response.
November 2013 3 -28 DRAFT: Do Not Cite or Quote
-------
1 Some evidence points to NO2 exposure-induced histamine release from mast cells, rather
2 than reflex bronchoconstriction, as the mechanism underlying changes in airway
3 resistance (von Nieding and Wagner, 1979). This includes a study in rats whereby mast
4 cell degranulation occurred after acute exposure to 500-1,000 ppb NO2 (Thomas et al..
5 1967). In addition, a histamine-suppressive agent, but not atropine which inhibits vagal
6 responses, or |3-agonists blocked NO2-mediated increases in airway resistance in healthy
7 humans and in humans with chronic lung disease exposed to 5,000-8,000 ppb NO2 for
8 5 minutes (von Nieding and Wagner. 1979). This study also demonstrated a decrease in
9 arterial PO2 and increase in the arterial to alveolar PO2 gradient, reflecting impaired gas
10 exchange, in humans with chronic lung disease immediately following 15 minutes of
11 exposure to 4,000 and 5,000 ppb (but not 2,000 ppb) NO2 (von Nieding and Wagner.
12 1979). More recent studies in animals have provided experimental evidence for a
13 relationship between lipid peroxidation/oxidative stress and the release of histamine by
14 allergen-activated mast cells (Beaven. 2009; Gushchin et al., 1990). Taken together, these
15 studies suggest that NO2 exposure-induced lipid peroxidation may promote mast cell-
16 mediated changes in pulmonary function, albeit at high concentrations.
17 There is also experimental support for NO2 exposure-induced cardiovascular reflexes. An
18 acute exposure to NO2 in an occupational setting resulted in tachycardia in one case
19 report (U.S. EPA. 1993; Bates etal.. 1971) while rats exposed acutely to 20,000 ppb or
20 higher concentrations of NO2 exhibited bradycardia (U.S. EPA. 1993; Tsubone et al..
21 1982). This latter response was abolished by injection of atropine, which inhibits vagal
22 responses. Furthermore, a decreased heart rate that was not accompanied by an increase
23 in respiratory rate was observed in mice exposed to 1,200 and 4,000 ppb NO2 for 1
24 month (Suzuki etal.. 1981). These results suggest that the decreased heart rate was due to
25 a different mechanism than rapid stimulation of irritant receptors by NO2. Subsequent
26 studies by this same group found an increase in respiratory rate following a 24-hour
27 exposure to 5,000 ppb NO2, while exposure to 10,000 and 20,000 ppb NO2 for 24 hours
28 resulted in increased respiratory rates, impaired gas exchange, increased lung wet weight
29 and increased lung water content (U.S. EPA. 2008c. 1993; Suzuki etal.. 1982; Suzuki et
30 al.. 1981). These results suggest that some cardiovascular effects observed after exposure
31 to high concentrations of NO2 may be secondary to pulmonary edema which is known to
32 stimulate pulmonary irritant receptors. Recently a controlled human exposure study
33 reported an effect on heart rate variability, which is a measure of autonomic tone, at
34 much lower concentrations of NO2 (Huang et al.. 2012b). While there was no indication
35 of pulmonary edema in this study, a statistically significant increase in levels of the injury
36 marker lactate dehydrogenase (LDH) was found in BAL fluid. Altered heart rate
37 variability found in epidemiologic studies (Section 4.3.3.1) is consistent with a possible
38 effect of NO2 exposure on autonomic tone.
November 2013 3 -29 DRAFT: Do Not Cite or Quote
-------
1 In summary, NO2 is a pulmonary irritant that may activate reflexes through vagal
2 pathways to increase respiratory rate, decrease tidal volume, stimulate reflex
3 bronchoconstriction and induce bradycardia. Responses are rapid, concentration-
4 dependent and variable between species. Evidence that reflex responses occur in humans
5 is weak since no increases in respiratory rate have been reported as a result of NO2
6 exposure. Some findings attributed to reflex bronchoconstriction in humans may be due
7 to alternative pathways such as mast cell degranulation. However, the recent
8 demonstration that NO2 exposure results in altered heart rate variability suggests the
9 possible activation of a neural reflex in humans. Attenuation of NO 2 -mediated responses
10 may occur with continuous or intermittent exposure. Lessening of the breathing pattern
11 response occurred in rodents exposed acutely and continuously to NO2 but not in rodents
12 exposed chronically and continuously to NO2. Attenuation of NO2-mediated changes in
13 pulmonary function occurred in human subjects exposed intermittently over several days.
3.3.2.3 Initiation of inflammation
14 As summarized in the 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993). NO2
15 exposure-induced membrane perturbations resulted in the release of arachidonic acid and
16 the formation of eicosanoid products (Section 4.2.4.2). Animal toxicological studies have
17 found increases in concentrations of eicosanoids in BAL fluid immediately following
18 exposure to NO2 (Schlesinger et al.. 1990). Controlled human exposure studies have also
19 demonstrated increased levels of eicosanoids immediately following NO2 exposure
20 (Torres etal.. 1995). Eicosanoids play an important role in the recruitment of neutrophils.
21 Interestingly higher concentrations and longer durations of exposure to NO2 resulted in
22 inhibited eicosanoid production (Robison and Forman. 1993; Schlesinger et al.. 1990).
23 Recently, acute exposure of mice to 10,000 ppb and higher concentrations of NO2 was
24 shown to activate NFxB in airway epithelium (Ather et al.. 2010; Bevelander et al..
25 2007). NFKB activation resulted in the production of pro-inflammatory cytokines.
26 Inflammation and acute lung injury in this model were found to be dependent on an
27 active NFxB pathway. Increased levels of cytokines have also been documented
28 following NO2 exposure in controlled human studies (Section 4.2.4.1) (U.S. EPA. 2008c;
29 Devlin etal.. 1999). The cell signaling pathways responsible for upregulating cytokines
30 at these lower levels of exposure to NO2 are not clear.
31 Studies in rodents exposed acutely (1 hour to 3 days) to NO2 (500-5,000 ppb) have
32 demonstrated airways inflammation mainly consisting of neutrophils and macrophages,
33 and sometimes of mast cells and lymphocytes, by histological technique or sampling of
November 2013 3-30 DRAFT: Do Not Cite or Quote
-------
1 BAL fluid [as summarized in (Sandstrom et al.. 1990)1 (Poynter et al.. 2006; Pagani et
2 al.. 1994V
3 Numerous studies in healthy human subjects exposed to NO2 have documented airways
4 inflammation in endobronchial biopsy tissue and in sputum, bronchial wash fluid and
5 BAL fluid (Section 4.2.4.1). Many of these studies were conducted while subjects were
6 exercising intermittently and exposed to 1,500-4,000 ppb NO2 for a few hours.
7 Neutrophilia was a prominent feature (U.S. EPA. 2008c: Frampton et al.. 2002; Devlin et
8 al.. 1999; Azadniv et al.. 1998; Blomberg et al.. 1997). In addition, other types of
9 inflammatory cells, including macrophages, lymphocytes and mast cells, have been
10 demonstrated (Frampton et al.. 2002; Sandstrom et al.. 1991; Sandstrom et al.. 1990).
11 Controlled human exposure studies have also evaluated the effects of repeated NO2
12 exposures on airways inflammation. While neutrophilic inflammation was persistent over
13 4 consecutive days of exposure to 2000 ppb NO2, other aspects of the lavage cell
14 response were different compared to single exposure responses (Blomberg et al.. 1999).
15 Repeated exposures also led to the upregulation of cytokines characteristic of the Th2
16 inflammatory response and also to upregulation of ICAM-1 in respiratory epithelium
17 (U.S. EPA. 2008c: Pathmanathan et al.. 2003). Upregulation of ICAM-1 suggests a
18 potential mechanism for the persistent leukocyte influx that was observed (Blomberg et
19 al.. 1999). In a study of repeated exposure to 4000 ppb over 6 consecutive days, numbers
20 of mast cells, macrophages and total lymphocytes were decreased compared with
21 responses to a single exposure (Sandstrom et al.. 1992; Rubinstein et al.. 1991).
22 Furthermore, repeated exposure to 1,500 ppb NO2 resulted in reduction in lymphocyte
23 subpopulations (Sandstrom et al.. 1992). Allergic inflammatory responses to NO2 are
24 discussed in Sections 3.3.2.6.2. 3.3.2.6.3. and 4.2.4.3.
3.3.2.4 Alteration of epithelial barrier function
25 Lipid peroxidation and altered phospholipid composition in the respiratory tract
26 following NO2 exposure may affect membrane fluidity and airway epithelial barrier
27 function. NO2 exposure-induced inflammation may further impair epithelial barrier
28 function. This could potentially lead to the loss of ELF solutes or proteins that could
29 diffuse down their concentration gradient from the lung to the blood. Increases in
30 vascular permeability may also occur, leading to the influx of plasma proteins such as
31 albumin into the airway lumen.
32 As summarized in the 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993). numerous
33 studies have demonstrated increases in the injury biomarkers, protein, albumin, LDH and
34 shed epithelial cells in BAL fluid following exposure to NO2 (Sections 4.2.4.1 and
November 2013 3-31 DRAFT: Do Not Cite or Quote
-------
1 4.2.4.2). Since LDH can be oxidatively inactivated, use of this indicator may
2 underestimate the extent of injury during oxidative stress. Many, but not all, of these
3 effects were observed at concentrations that are higher than ambient-relevant levels.
4 Ascorbate deficiency enhanced protein levels in the BAL fluid of NO 2 -exposed guinea
5 pigs, suggesting a role for BAL fluid ascorbate in preventing the deleterious effects of
6 NO2 (Hatch etal.. 1986). Similarly, a-tocopherol deficiency enhanced lipid peroxidation
7 in NO2-exposed rats (Sevanian et al.. 1982a). Recently, selenium deficiency was found to
8 enhance the injury response in rats exposed to 1,000-50,000 ppb (acute, subacute and
9 chronic exposures) NO2 (de Burbure et al.. 2007). Both BAL fluid total protein and
10 serum Clara Cell secretory protein (CC16) were increased in selenium-deficient rats
11 exposed to NO2. Selenium supplementation diminished this response providing evidence
12 that the selenium-containing enzyme glutathione peroxidase played an important
13 mitigating role.
14 Increases in lung permeability due to high concentrations of NO2 are known to cause
15 death from pulmonary edema (Lehnert et al.. 1994; Gray et al.. 1954). At lower
16 concentrations, more subtle effects have been reported. Exposure of rats to 5,000 ppb and
17 10,000 ppb NO2 for 3 or 25 days resulted in epithelial degeneration and necrosis and
18 proteinaceous edema (Earth etal.. 1995). while exposure to 1,000-10,000 ppb NO2 for 1
19 and 3 days resulted in concentration-dependent increases in BAL fluid protein (Muller et
20 al.. 1994). BAL fluid protein was also elevated in guinea pigs exposed for 1 week to 400
21 ppb NO? (Sherwin and Carlson. 1973).
22 High concentrations of NO2 (70,000 ppb, 30 minutes) were found to enhance
23 translocation of instilled antigen from the lung to the blood stream of guinea pigs
24 (Matsumura. 1970). More subtle increases in lung permeability due to NO2 exposure
25 could enhance the translocation of an antigen to local lymph nodes and circulation (U.S.
26 EPA. 2008c; Gilmour et al.. 1996) and/or to the immunocompetent and inflammatory
27 cells underlying the epithelium which are involved in allergic reactions (Jenkins et al..
28 1999). However, increased lung permeability following exposure to NO2 does not always
29 lead to allergic sensitization (Alberg etal.. 2011). In addition, increased epithelial
30 permeability may contribute to the activation of neural reflexes and the stimulation of
31 smooth muscle receptors (Dimeo etal.. 1981) by allowing greater access of agonist.
32 Susceptibility to NO2 exposure-induced cytotoxicity was investigated in several mice
33 strains with differing genetic backgrounds (Kleeberger et al.. 1997). This study found a
34 strong genetic component of NO2 susceptibility that differed from the genetic component
35 involved in susceptibility to O3. In addition, the genetic component contributed to the
36 attenuation of responses that was seen following repeated exposures.
November 2013 3-32 DRAFT: Do Not Cite or Quote
-------
3.3.2.5 Enhancement of bronchial smooth muscle reactivity
1 Exposure to NO2 enhances the inherent reactivity of airway smooth muscle in healthy
2 and asthmatic human subjects (Folinsbee, 1992) (Section 4.2.2) and in animal models
3 (see below). This "airway responsiveness" is defined as the sensitivity of airways to a
4 variety of natural or pharmacological stimuli (O'Byrne et al.. 2009). Airway
5 hyperresponsiveness (AHR) is a key feature of asthma, which is a chronic inflammatory
6 disease of the airways. As summarized in 2008 ISA (U.S. EPA, 2008c) and in Section
7 4.2.2. numerous studies found that human subjects who were exposed to NO2 were more
8 sensitive to the nonspecific stimuli methacholine than human subjects who were exposed
9 to air. Asthmatics exhibited greater sensitivity than nonasthmatics when similarly
10 exposed. In addition, several studies found that NO2 exposure enhanced airways
11 responsiveness to specific stimuli such as allergens in mild allergic asthmatics.
12 Exercise during exposure to NO2 appeared to modify airway responsiveness in
13 asthmatics (Folinsbee. 1992) (Section 4.2.2.2). Mechanisms by which this occurs are not
14 understood but two hypotheses have been postulated. First, exercise-induced
15 refractoriness, which has been demonstrated in some asthmatics, may alter
16 responsiveness to NO2 (Magnussen et al., 1986). A second hypothesis is that nitrites
17 formed by NO2-mediated reactions in the ELF mediate compensatory relaxation of
18 airway smooth muscle (Folinsbee. 1992). Exercise would increase the total dose of NO2
19 to the respiratory tract, thus increasing nitrite formation. Recent studies have shown that
20 reactive nitrogen species have bronchodilatory effects. For example, endogenous S-
21 nitrosothiols are an important modulator of airway responsiveness in asthmatics and in
22 eosinophilic inflammation (Lee et al., 20lib; Que et al., 2009).
23 Animal toxicological studies have also demonstrated NO2-enhanced responsiveness of
24 airways to nonspecific and specific challenges, as summarized in the 2008 ISA and 1993
25 AQCD (U.S. EPA. 2008a. c, 1993) and in Sections 4.2.2 and 5.2.8. Exposures ranged
26 from acute to subchronic in these studies and results suggest that more than one
27 mechanism may have contributed to the observed AHR. Acute exposure of guinea pigs to
28 NO2 (10 minutes, 7,000 ppb and higher) resulted in concentration-dependent AHR to
29 histamine, which was administered immediately after exposure (Silbaugh et al., 1981).
30 This response was short-lived since no enhanced responsiveness was seen at 2 and 19
31 hours post-exposure to NO2. The rapidity of the response suggests reflex
32 bronchoconstriction (Section 3.3.2.2) as a possible underlying mechanism. A 7-day
33 exposure to 4,000 ppb NO2 also increased airway responsiveness to histamine in guinea
34 pigs (Kobavashi and Shinozaki. 1990). Eicosanoids were proposed to play a role in this
35 transient response. In addition, a recent study in mice sensitized and challenged with
36 ovalbumin found that short-term exposure to NO2 (25,000 ppb but not 5,000 ppb, 3 days)
November 2013 3-33 DRAFT: Do Not Cite or Quote
-------
1 resulted in AHRto methacholine (Poynter et al.. 2006). This enhanced sensitivity
2 correlated with an increase in numbers of eosinophils, suggesting eosinophilic
3 inflammation as a possible underlying mechanism in this model of allergic airways
4 disease. A subchronic study demonstrated dose-dependent increases in airway
5 responsiveness to histamine in NO2-exposed guinea pigs (1,000-4,000 ppb, 6-12 weeks)
6 (U.S. EPA. 2008c: Kobavashi and Miura. 1995). Specific airways resistance in the
7 absence of a challenge agent was also increased, which indicates the development of
8 airways obstruction. This finding suggests airway remodeling as a possible underlying
9 mechanism for AHR. Another subchronic exposure study found a delayed bronchial
10 response, measured as increased respiration rate and suggestive of AHR, in guinea pigs
11 sensitized and challenged with C. albicans and exposed to NO2 (Kitabatake et al.. 1995).
12 Mechanisms underlying the effects of NO2 on airway responsiveness are not well
13 understood. Effects of NO2 exposure on redox status in the respiratory tract should be
14 considered since asthma pathogenesis, including airway inflammation,
15 hyperresponsiveness and remodeling, may be under redox control (Comhair and
16 Erzurum. 2010; Kloek et al.. 2010). In support of this mechanism, supplementation with
17 the antioxidant ascorbate was found to prevent nonspecific AHR in asthmatic subjects
18 exposed to NO2 (Mohsenin. 1987b).
19 Furthermore, different inflammatory pathways may underly NO2-mediated AHR
20 (Krishna and Holgate. 1999). First, there is some evidence that mast cell activation may
21 contribute to NO2 exposure-induced AHR. As discussed in Section 3.3.2.2. acute
22 exposure to NO2 led to mast cell activation in rats and humans. Histamine released by
23 mast cells can directly bind to receptors on smooth muscle cells and cause contraction.
24 This response would have the appearance of reflex bronchoconstriction but would not
25 involve neural pathways. Secondly, neutrophils and other inflammatory cell types can
26 release mediators such as IL-13, IL-17 and TNF-a that can alter the calcium sensitivity of
27 the smooth muscle and enhance a contractile response to a stimulus (Kudo et al.. 2013).
28 Thirdly, chronic inflammation can lead to structural changes in the airway walls that
29 enhance the contractile response of the smooth muscle to a given stimuli (Cockcroft and
30 Davis. 2006c). Whether or not NO2 exposure enhances intrinsic contractility of airway
31 smooth muscle by these mechanisms is unknown. Fourthly, increased peroxynitrite
32 formation occurring during inflammatory states may play a role. Generally, peroxynitrite
33 is produced by reaction of NO and superoxide, subsequently reacts with CO2 to form the
34 nitrosoperoxylcarbonate anion (ONOOCO2~), which decomposes to carbonate radical and
35 NO2 (Section 3.2.2.4). Recent studies have provided evidence that endogenous
36 peroxynitrite contributes to AHR in animal models of allergic airways disease (Section
37 3.3.2.6.2). These studies demonstrate that NO metabolism is dysfunctional in inflamed
38 lungs and enhances peroxynitrite formation. Amelioration of the dysfunction resulted in
November 2013 3-34 DRAFT: Do Not Cite or Quote
-------
1 less nitrative stress and reduced AHR (Ahmad etal. 2011; Mabalirajan et al., 2010b;
2 Maarsingh et al.. 2009; Maarsingh et al.. 2008).These studies highlight the possibility that
3 inhaled NO2 can add to the lung burden of endogenous NO2 which is found in and
4 contributes to AHR and allergic airway disease in animal models (Section 3.3.2.6.2).
5 It has been hypothesized that NO 2 -mediated impairment of epithelial barrier function
6 may play a role in antigen accessibility to immune tissue, resulting in activation of
7 immune responses that trigger AHR (Section 3.3.2.4). In addition, NO2-mediated
8 impairment of epithelial barrier function could allow greater access of mediators to
9 sensory receptors on nerve fibers. This could lead to enhanced activation of neural
10 pathways and airway smooth muscle contraction (Hesterberg et al.. 2009; Cockcroft and
11 Davis. 2006c). Conditions where epithelial barrier function is impaired may increase
12 responses to allergens, mediators and/or NO2. Allergic inflammation, especially
13 eosinophil activation and release of eosinophil cationic protein (ECP), may cause damage
14 to the airway epithelium in allergic airways disease (Ohashi et al.. 1994). This damage
15 may result in epithelial shedding and mucociliary dysfunction. Epithelial shedding could
16 lead to greater exposure of sensory nerve endings and enhanced activation of neural
17 reflexes by mediators. In addition, epithelial shedding and mucociliary dysfunction may
18 allow greater access of allergens to the airway epithelium and submucosa. This may
19 explain the close relationship which has been observed between epithelial shedding and
20 AHR. While NO2 exposure has been shown to enhance the immune response to allergens
21 (Section 3.3.2.6.2). it is not know whether this mechanism is responsible for
22 NO2-mediated AHR in allergic airways disease.
3.3.2.6 Modification of innate/adaptive immunity
23 Host defense depends on effective barrier function and on innate immunity and adaptive
24 immunity (Al-Hegelan et al., 2011). The effects of NO2 on barrier function in the airways
25 were discussed above (Section 3.3.2.4). This section focuses on the mechanisms by
26 which NO2 impacts innate and adaptive immunity. Both tissue damage and foreign
27 pathogens are triggers for the activation of the innate immune system. This results in the
28 influx of inflammatory cells such as neutrophils, mast cells, basophils, eosinophils,
29 monocytes and dendritic cells and the generation of cytokines such as TNF-a, IL-1, IL-6,
30 KC and IL-17. Further, innate immunity encompasses the actions of complement and
31 collections, and the phagocytic functions of macrophages, neutrophils and dendritic cells.
32 In addition to immune cells, airway epithelium contributes to innate immune responses.
33 Innate immunity is highly dependent on cell signaling networks involving TLR4 in
34 airway epithelium and other cell types. Adaptive immunity provides immunologic
November 2013 3-35 DRAFT: Do Not Cite or Quote
-------
1 memory through the actions of B and T-cells. Important links between the two systems
2 are provided by dendritic cells and antigen presentation.
3.3.2.6.1 Impairment of host defenses
3 As summarized in the 2008 ISA (U.S. EPA. 2008c). potential mechanisms by which NO2
4 exposure may impair host defenses include ciliary dyskinesis, damage to ciliated
5 epithelial cells, and altered alveolar macrophage function, all of which may contribute to
6 altered mucociliary transport and/or clearing of the lung of infectious and non-infectious
7 particles. Altered alveolar macrophage function and other potential mechanisms such as
8 increases in pro-inflammatory mediators and cytokines, increased IgE concentrations,
9 interactions with allergens and altered lymphocyte subsets, reflect modification of innate
10 and/or adaptive immunity. These changes may underly susceptibility to infection, which
11 have been observed in animals exposed to NO2.
12 Controlled human studies have demonstrated reduced mucociliary clearance due to
13 depressed ciliary function, depressed alveolar macrophage phagocytic activity and
14 superoxide production and altered humoral- and cell-mediated immunity following
15 exposure to 1,500-4,000 ppb NO2 for a few hours (Frampton et al.. 2002; Devlin et al..
16 1999; Helledav et al.. 1995; Sandstrom et al.. 1992; Sandstrom et al.. 1992; Sandstrom et
17 al.. 1991) (Section 4.2.5). Studies involving repeated daily exposure to 1,500 ppb NO2
18 (but not 600 ppb NO2) found reductions in lymphocyte subpopulations (Sandstrom et al..
19 1992; Rubinstein et al.. 1991; Sandstrom et al.. 1990). Furthermore, repeated daily
20 exposure to 2,000 ppb NO2 resulted in upregulation of ICAM-1 in bronchial biopsy
21 specimens (Pathmanathan et al.. 2003). These findings suggest a potential mechanism
22 underlying susceptibility to viral infection since ICAM-1 is a major receptor for rhino
23 and respiratory syncytial viruses. Finally, enhanced susceptibility of airways epithelium
24 to influenza viral infection was suggested in a study involving exposure to 1,000-3,000
25 ppb NO2 over 3 days, although statistical significance was not achieved (Goings et al..
26 1989). Humans exposed to 600 and 1,500 ppb NO2 for 3 hours exhibited an increased
27 injury response, as measured in bronchial epithelial cells, resulting from influenza and
28 respiratory syncytial virus (Frampton et al.. 2002). Epidemiologic evidence for
29 associations between exposure to NO2 and increased respiratory infections in children is
30 consistent with these results (Section 4.2.5.1).
31 As summarized in the 2008 ISA (U.S. EPA. 2008a. c) and 1993 AQCD (U.S. EPA.
32 1993). studies in NO2-exposed animals (500-10,000 ppb) have demonstrated altered
33 mucociliary clearance and several changes in alveolar macrophages. This includes
34 morphological evidence of damage to alveolar macrophages (membrane bleb formation
35 and mitochondrial damage), decreased viability and decreased function (decreased
November 2013 3-36 DRAFT: Do Not Cite or Quote
-------
1 superoxide production, decreased phagocytic capacity and decreased migration towards a
2 stimulus) (Robison et al.. 1993; Davis etal. 1992; Roseetal.. 1989a: Schlesinger et al..
3 1987; Schlesinger and Gearhart. 1987; Suzuki etal., 1986; Greene and Schneider. 1978;
4 Powell etal.. 1971). Infectivity models have shown increased mortality and decreased
5 bactericidal activity (U.S. EPA. 2008c; Jakab. 1987; Miller et al.. 1987; Ehrlich. 1980;
6 Ehrlich et al.. 1977). as a result of NO2 exposure. Further discussion is found in Section
7 4.2.5 and 5.2.9.
3.3.2.6.2 Exacerbation of allergic airways disease
8 Inhaled allergens activate an acute immune response in allergen-sensitive individuals.
9 This response is characterized by early and late phases. Key players in the early asthmatic
10 response are mast cells and basophils which release mediators following allergen binding
11 to IgE receptors on their cell surfaces. These mediators include histamine and cysteinyl
12 leukotrienes which bind airway smooth muscle receptors and induce contraction.
13 Mediators also activate T lymphocyte subsets (i.e., CD4+ T-cells) resulting in the release
14 of T helper cell (Th2) cytokines that can cause airway smooth muscle contraction and
15 recruit mast cells. They also promote the influx and activation of eosinophils and
16 neutrophils. Airway mucosal eosinophilia is characteristic of asthma and rhinitis.
17 Eosinophils exert their effects via degranulation or cytolysis resulting in release of ECP
18 and other mediators (Erjefalt et al.. 1999). Th2 cytokines also activate B lymphocyte
19 resulting in the production of allergen-specific IgE. These responses initiated by Th2
20 cytokines contribute to the late asthmatic response, which is characterized by airway
21 obstruction generally occurring 3-8 hours following an antigen challenge (Cockcroft and
22 Davis. 2006c) and to other responses occurring greater than 3-8 hours following an
23 antigen challenge.
Exogenous NO2
24 As summarized in the 2008 ISA (U.S. EPA. 2008c) and in Section 4.2.4.3. exposure to
25 NO2 affects several steps in the acute immune response to inhaled allergens. Several
26 controlled human studies found that NO2 exposure enhanced airways responsiveness to
27 specific stimuli such as house dust mite (HDM) allergen (Jenkins et al.. 1999; Tunnicliffe
28 etal.. 1994) in mild allergic asthmatics. Further, repeated exposure to NO2 resulted in an
29 enhanced response to a dose of allergen that was asymptomatic when given alone (Strand
30 et al.. 1998). Airway responses were measured during the first 2 hours after allergen
31 challenge which falls within the timeline of the early phase asthmatic response. These
32 results provide evidence that NO2 exposure exacerbates the early phase asthmatic
November 2013 3-37 DRAFT: Do Not Cite or Quote
-------
1 response to allergen challenge, as measured by enhanced airway smooth muscle cell
2 contraction.
3 Controlled human exposure studies also demonstrated that NO2 exposure exacerbated the
4 late phase asthmatic response to allergen challenge in mild allergic asthmatics. Airway
5 obstruction, measured as a spontaneous fall in FEVi occurring after resolution of the
6 early asthmatic response (generally 3-8 hours after an antigen challenge) was observed in
7 asthmatic subjects exposed to 400 ppb NO2 for 1 hour (Tunnicliffe et al.. 1994) and to
8 250 ppb NO2 for 30 minutes for 4 consecutive days (Strand et al., 1998). Other studies
9 measured cell counts and mediators characteristic of the late phase asthmatic response.
10 Increased numbers of polymorphonuclear leukocytes and increased levels of ECP in BAL
11 fluid, both indicators of inflammatory response to allergen challenge, were reported
12 following exposure to 260 ppb NO2 for 15-30 minutes (Barck et al.. 2005a: Barck et al..
13 2002). Furthermore, increased ECP levels were observed in sputum and blood and an
14 increase in myeloperoxidase (indicator of neutrophil activation) was seen in blood. In
15 subjects with allergic rhinitis, NO2 exposure (400 ppb for 6 hours) increased eosinophil
16 activation, measured by ECP in nasal lavage, following nasal allergen provocation (Wang
17 et al.. 1995a). These studies suggest that exposure to NO2 may prime eosinophils for
18 subsequent activation by allergen in previously sensitized individuals (Davies et al..
19 1997; Wang et al.. 1995b). However, another study found decreased sputum eosinophils
20 6 hours after HDM challenge in HDM-sensitive asthmatics exposed to 400 ppb NO2 for 3
21 hours (Witten et al.. 2005).
22 Late phase responses were also investigated in animal models of allergic airways disease.
23 Increased specific immune response to HDM allergen, including enhanced antigen-
24 specific serum IgE, and increased lung inflammation were demonstrated in Brown
25 Norway rats sensitized to and challenged with HDM allergen followed by 3-hour
26 exposure to 5,000 ppb NO2 (Gilmour et al.. 1996). Similarly, a recent study showed that
27 NO2 exposure (25,000 ppb, 6 hours/day for 3 days) increased the degree and duration of
28 the allergic inflammatory response in mice sensitized and challenged with ovalbumin
29 (Poynter et al.. 2006). Both neutrophilic and eosinophilic airway inflammation were
30 found in these studies; exposure of mice to a lower concentration of NO2 (5,000 ppb)
31 failed to induce this response. Two other studies in ovalbumin sensitized and challenged
32 mice found decreased eosinophilic inflammation in response to 5,000 ppb NO2 (Hubbard
33 et al.. 2002; Proust et al.. 2002). These results in animal models provide some evidence of
34 NO2-mediated enhancement of late phase responses, however results were somewhat
35 inconsistent. It is important to note that eosinophil activation and eosinophil influx reflect
36 different processes and that only the study by Hubbard et al. (2002) measured markers of
37 activation. The ovalbumin sensitized and challenged mouse model may not mimic the
38 eosinophil degranulation or cytolysis that is characteristic of asthma and allergic rhinitis
November 2013 3-38 DRAFT: Do Not Cite or Quote
-------
1 in humans (Malm-Erjefalt et al., 2001). Hence species-related differences may account
2 for the differences in results of animal and controlled human exposure studies.
3 Collectively, these studies demonstrate that inhaled NO2 enhanced both early and late
4 phase responses to inhaled allergens in humans with asthma and allergy. Furthermore,
5 exposure to NO2 augmented allergic inflammation in some rodent models of allergic
6 airways disease. These results provide evidence for NO2-induced exacerbation of allergic
7 airways disease both in the presence and absence of an allergen challenge.
Endogenous NO2
8 Several recent animal toxicological studies have explored the role of endogenous NO and
9 peroxynitrite on allergic airways disease in animal models. In one study, upregulating the
10 enzyme eNOS (and presumably NO production) decreased airway inflammation, AHR
11 and remodeling in a mouse model of asthma (Ahmad et al., 2011). Asthma phenotype-
12 related features such as cell infiltrates, mucus hypersecretion, peribronchial collagen and
13 Th2 cytokines were also diminished. Further, decreased iNOS expression and 3-
14 nitrotyrosine immunostaining in airway epithelium were reported, as were diminished
15 epithelial injury and apoptosis. Since 3-nitrotyrosine is a marker of NO2/peroxynitrite
16 formation, these findings suggest that an increase in NO may have resulted in reduced
17 peroxynitrite. While it is known that NO rapidly reacts with superoxide to form
18 peroxynitrite, and that superoxide levels are increased in inflammation, it is also known
19 that an excess of NO will react with peroxynitrite and quench its reactivity. In fact,
20 Stenger etal. (2010) found that high concentrations of inhaled NO prevented the
21 formation of 3-nitrotyrosine in the lungs of neonatal mice exposed to hyperoxia.
22 In a second set of studies, increased levels of the NOS substrate L-arginine were found to
23 decrease airway inflammation and AHR in a guinea pig model of asthma (Maarsingh et
24 al.. 2009). Similarly, increased L-arginine levels reduced peroxynitrite formation and
25 AHR in a mouse model of asthma (Mabalirajan et al., 2010b). Markers of allergic
26 inflammation such as eosinophilia and Th2 cytokines, markers of oxidative and nitrative
27 stress, and markers of airway remodeling such as goblet cell metaplasia and subepithelial
28 fibrosis, were also decreased. Increased L-arginine levels also reduced mitochondrial
29 dysfunction and airway injury (Mabalirajan et al., 2010a). Limitation of L-arginine is
30 known to uncouple NOS enzyme activity resulting in the production of superoxide in
31 addition to NO. This situation is commonly found in disease models and leads to
32 peroxynitrite formation. Increasing L-arginine availability is a common strategy used to
33 prevent enzyme uncoupling and peroxynitrite formation. Another approach was
34 employed in a study by North et al. (2009) where inhibition of the enzyme arginase 1,
35 (arginase 1 decreases arginine availability), was found to decrease AHR in a mouse
November 2013 3-39 DRAFT: Do Not Cite or Quote
-------
1 model of asthma. Similar findings were reported using arginase inhibition in a guinea pig
2 model of allergic asthma where arginase was upregulated (Maarsingh et al.. 2008).
3 Inhibition of arginase resulted in amelioration of the asthma phenotype. These effects
4 were attributed to decreased enzyme uncoupling, thus promoting the formation of NO,
5 diminishing the generation of superoxide and reducing the formation of peroxynitrite. In
6 contrast, a different study found that arginase inhibition resulted in increased S-
7 nitrosylated and nitrated proteins, increased inflammation, mucous metaplasia, NFxB
8 activation and AHR in a mouse model of asthma (Ckless et al.. 2008). However, antigen
9 specific IgE and IL-4 levels were reduced. Thus, only some features of the asthma
10 phenotype were ameliorated by arginase inhibition. The authors suggested that
11 peroxynitrite, whose presence was indicated by the increase in nitrated proteins in mice
12 treated with arginase, may have contributed to the enhanced AHR in this model.
13 Evidence for similar pathways in humans is provided by a study in which endogenous
14 markers of reactive nitrogen and oxygen chemistry were measured in individuals with
15 and without asthma (Anderson et al.. 2011). Levels of total nitrite and nitrate were higher
16 in the BAL fluid of subjects with asthma compared to healthy subjects. Upregulation of
17 iNOS was observed and it was greater in distal airways compared with more proximal
18 airways of asthmatics. In addition, levels of DHE+ cells capable of producing reactive
19 oxygen species (such as superoxide) were higher in both the bronchial wash and BAL
20 fluid of asthmatics compared with healthy subjects. Levels of arginase were also higher
21 in BAL fluid of asthmatics compared with healthy subjects. These results suggest that
22 uncoupling of NOS and/or NOS dysfunction resulting in enhanced peroxynitrite/ NO2
23 formation may contribute to the asthma phenotype. They also provide biological
24 plausibility for results of another study demonstrating a correlation between increased
25 airway responsiveness and the induction of iNOS, the induction of arginase, and the
26 production of superoxide in subjects with asthma.
27 Collectively, these studies provide evidence that the balance between endogenous NO
28 and peroxynitrite influences features of the asthma phenotype in animal models of asthma
29 and possibly in adults with asthma. Enhanced levels of superoxide, which are
30 characteristic of asthma and other inflammatory states, favor the formation of
31 peroxynitrite at the expense of NO. Evidence from experimental studies indicates that
32 peroxynitrite and other reactive nitrogen species are found in and contribute to allergic
33 airway disease in animal models. Inhaled NO2 may exacerbate allergic airways disease
34 by adding to the lung burden of reactive nitrogen species in inflammatory states.
November 2013 3 -40 DRAFT: Do Not Cite or Quote
-------
3.3.2.6.3 Th2 skewing and allergic sensitization
1 A controlled human exposure study demonstrated that repeated daily exposures to NO2
2 resulted in increased expression of IL-5, IL-10, IL-13 and ICAM-1 in respiratory
3 epithelium following the last exposure (Pathmanathan et al.. 2003) (Section 4.2.4.3).
4 These interleukins are characteristic of a Th2 inflammatory response. IL-5 is known to
5 promote eosinophilia, while IL-13 is known to promote mucus production and AHR
6 (Bevelander et al.. 2007). These findings suggest a potential mechanism whereby
7 repeated exposure to NO2 may exert a pro-allergic influence. Further, upregulation of
8 ICAM-1 suggests a potential mechanism for persistent leukocyte influx. A separate study
9 by these same investigators found that neutrophilic inflammation was persistent over the
10 4 days of repeated exposure (Blomberg et al.. 1999).
11 In addition, two studies in animals examined the effects of longer-term exposures to NO2
12 on the development of allergic responses (Sections 4.2.4.3 and 5.2.6.2). In one study,
13 exposure of guinea pigs to 3000 or 9000 ppb NO2 increased the numbers of eosinophils
14 in nasal epithelium and mucosa after two weeks (Ohashi et al., 1994). In the other,
15 exposure to 4000 ppb NO2 for 12 weeks led to enhanced IgE-mediated release of
16 histamine from mast cells isolated from guinea pigs (Fujimaki andNohara. 1994). This
17 response was not found in mast cells from rats similarly exposed. Both studies provide
18 further evidence for NO2 having a pro-allergic influence.
19 A recent study in mice provides evidence that NO2 may act as an adjuvant promoting the
20 development of allergic airways disease in response to a subsequent inhalation exposure
21 to ovalbumin (Bevelander et al.. 2007). Findings included AHR, mucous cell metaplasia
22 and eosinophilic inflammation, as well as ovalbumin-specific IgE and IgGl and CD4+
23 T-cells biased towards to Th2 and Thl7 phenotypes in the blood. These results are
24 consistent with an allergic asthma phenotype in humans. Furthermore, eosinophilic
25 inflammation, mucus gene upregulation and ovalbumin-specific IgE production were
26 found to be dependent on Toll receptor 2 (TLR2) and MyD88 pathways. TLR2 is known
27 to promote dendritic cell maturation, inflammation and Th2 skewing. A subsequent study
28 in the same model found that NO2 exposure had several effects on pulmonary CD1 lc+
29 dendritic cells, including increased cytokine production, upregulation of maturation
30 markers, increased antigen uptake, migration to the lung-draining lymph node and
31 improved ability to stimulate naive CD4+ T-cells (Hodgkins et al., 2010). Dendritic cells
32 are key players in adaptive immune responses by regulating CD4+ mediated T cell
33 responses through the presentation of antigens in the draining lymph node. Further,
34 dendritic cells can express a distinct pattern of co-stimulatory molecules and produce
35 cytokines which create an environment for T cell polarization, thus skewing the T helper
36 cell response. Changes reported in these two studies are consistent with the promotion of
November 2013 3-41 DRAFT: Do Not Cite or Quote
-------
1 allergic sensitization and suggest a role for TLR2 in mediating this effect. A third study
2 by these same investigators found that NO2 exposure resulted in antigen specific IL-17A
3 generation from ThlV cells, which is characteristic of the severe asthma phenotype which
4 is unresponsive to glucocorticoid treatment in humans (Martin etal.. 2013). Although all
5 studies involved 1-hour exposures to high concentrations of NO2 (10,000-15,000 ppb),
6 they are included here because they describe potentially new mechanisms by which NO2
7 exerts its effects. It should additionally be noted that airway inflammation is seen in mice
8 exposed to 15,000 ppb, but not 10,000 ppb, NO2 for 1 hour and that pulmonary damage
9 is minimal in this model (Martin et al., 2013).
10 It should be noted that another study failed to find that NO2 acted as an adjuvant in a
11 mouse model of allergic airway disease (Alberg et al., 2011). In this study the exposure
12 consisted of 5,000 or 25,000 ppb NO2 for 4 hours and followed exposure to ovalbumin
13 which was administered intranasally. Adjuvant activity was measured as the production
14 of allergen-specific IgE antibodies. Methodological differences in study design involving
15 the timing between ovalbumin and NO2 exposures and the route of ovalbumin exposure
16 may account for differences in findings between this study and others. In fact, Bevelander
17 et al. (2007) found that NO2 promoted allergic sensitization when exposure occurred
18 prior (but not subsequent) to ovalbumin.
19 It has been hypothesized that both endogenous and exogenous reactive nitrogen and
20 oxygen species can alter the balance between tolerance and allergic sensitization due to
21 an inhaled agent (Ckless etal.. 2011). Some activities of dendritic and T-cells, such as
22 maturation of antigen presenting capacity of dendritic cells, dendritic cell stimulation of
23 CD4+ T-cells and polarization of T-cells are redox-sensitive. Endogenous reactive
24 nitrogen and oxygen species are produced by a variety of respiratory tract cells including
25 epithelial, dendritic, T lymphocytes, macrophages, neutrophils and eosinophils,
26 especially during inflammation. Peroxynitrite formation, myeloperoxidase activity and/or
27 nitrite acidification may also be enhanced during inflammation and contribute to
28 endogenous NO2 levels. Reactive oxygen and nitrogen species are thought to promote the
29 allergic phenotype. Air pollution-derived exogenous reactive nitrogen and oxygen species
30 can potentially contribute to oxidative/nitrative stress in the respiratory tract and
31 influence the adaptive immune response that occurs once dendritic cells are activated.
32 Thus, recent studies suggest the possibility of an interaction between inhaled NO2 and
33 NO2 endogenously formed in the respiratory tract.
34 Collectively these studies in humans and animals provide evidence that NO2 exposure
35 may lead to the development of allergic responses via Th2 skewing and allergic
36 sensitization.
November 2013 3 -42 DRAFT: Do Not Cite or Quote
-------
3.3.2.7 Remodeling of airways and alveoli
1 As summarized in the 2008 ISA (U.S. EPA. 2008a. c) and 1993 AQCD (U.S. EPA.
2 1993). numerous studies have examined morphological changes in the respiratory tract
3 resulting from chronic NO2 exposure. The sites and types of morphological lesions
4 produced by exposure to NO2 were similar in all species when effective concentrations
5 were used (U.S. EPA. 1993). The centriacinar region exhibited the greatest sensitivity to
6 NO2 exposure, while the nasal cavity was not much affected. Cells most injured in the
7 centriacinar region were the ciliated cells of the bronchiolar epithelium and type 1 cells of
8 the alveolar epithelium. These were replaced with nonciliated bronchiolar and type II
9 cells, respectively, which were relatively resistant to continued NO2 exposure. Some
10 lesions rapidly resolved post-exposure. One study found that collagen synthesis rates
11 were increased in NO2-exposed rats. Since collagen is an important structural protein in
12 the lung and since increased total lung collagen is characteristic of pulmonary fibrosis, it
13 was proposed that NO2 exposure may cause fibrotic-like diseases.
14 Exposure to NO2 was also found to enhance pre-existing emphysema in animal models
15 (U.S. EPA. 2008c). Other studies demonstrated that NO2 exposure induced air space
16 enlargements in the alveolar region and suggested that chronic exposures could result in
17 permanent alterations resembling emphysema-like diseases (U.S. EPA. 1993). A recent
18 study confirmed and extended these findings. NO2 exposure in rats (10,000 ppb for 21
19 days) caused increased apoptosis of alveolar epithelial cells and enlargement of air spaces
20 (Fehrenbach et al.. 2007). Further, alveolar septal cell turnover was increased and
21 changes in extracellular matrix were noted. However, there was no loss of alveolar walls
22 (i.e., total alveolar wall volume or total alveolar surface area) indicating that the lesions
23 induced did not meet the 1985 National Heart Lung and Blood Institute definition of
24 human emphysema (U.S. EPA. 1993).
25 A chronic study in rats exposed to 9,500 ppb NO2 for 7 hours/day, 5 days/week for 24
26 months found an additional response (Mauderly et al.. 1990). Bronchiolar epithelium was
27 observed in centriacinar alveoli, and this response progressed with increasing length of
28 exposure. This has been termed "alveolar bronchiolization" (Nettesheim et al.. 1970).
29 reflecting the replacement of one type of epithelium by another. Long-term consequences
30 of alveolar bronchiolization are not known.
31 The relationship between NO2 exposure-induced morphologic changes in animal models
32 and impaired lung function growth seen in epidemiological studies is not clear. Effects of
33 NO2 exposure on lung morphology in rats has been shown to be age-dependent (U.S.
34 EPA. 2008a. c, 1993). Six-week old rats exposed to NO2 for 6 weeks were more sensitive
35 to the effects of NO2 exposure than one day-old rats exposed for six weeks (Chang et al..
36 1986). In humans, the respiratory and immune systems are immature in newborns and the
November 2013 3 -43 DRAFT: Do Not Cite or Quote
-------
1 respiratory system continues to develop until about 20 years of age. This suggests the
2 potential for NO2 exposure-induced permanent morphological changes in humans if
3 exposure should occur during critical windows of development. However, experimental
4 evidence to substantiate this claim is currently lacking.
3.3.2.8 Transduction of extrapulmonary responses
5 While the respiratory tract has been viewed as the primary target of the effects of inhaled
6 NO 2, effects outside the respiratory tract have been demonstrated in controlled human
7 exposure and toxicological studies (U.S. EPA. 2008a. 1993). These include
8 hematological effects and effects on the heart, central nervous system, liver, kidneys and
9 on reproduction and development. Some studies have explored the potential
10 carcinogenicity of NO2. Many, but not all, of these extrapulmonary effects have been
11 observed in animal models at concentrations that are higher than ambient-relevant levels.
12 Epidemiologic evidence of associations between NO2 exposure and extrapulmonary
13 effects has also been described (Section 4.3).
14 Given the reactivity of NO2, extrapulmonary effects would likely be due to NO2 reaction
15 products rather than to NO2 itself. One mechanism by which a NO2-derived reaction
16 product could mediate extrapulmonary effects would involve the activation of pulmonary
17 irritant receptors (Section 3.3.2.2). As summarized in the 2008 ISA and 1993 AQCD
18 (U.S. EPA. 2008a. c, 1993). effects of chronic NO2 exposure in animal models include a
19 reduction in PaO2 and a reduction in heart rate. This reduction in heart rate was not
20 accompanied by an increase in respiratory rate, suggesting that pulmonary irritant
21 receptors were not involved (Section 3.3.2.2 and 5.3.3). However, altered vagal responses
22 were seen in animals exposed acutely to a high concentration of NO2, but not to a lower
23 concentration over several weeks. Much weaker evidence exists for activation of
24 pulmonary irritant receptors in humans (Section 3.3.2.2). Controlled human exposure
25 studies have examined the effects of NO2 on heart rate and heart rate variability (Section
26 4.3.3). Older studies and one newer study failed to find statistically significant changes in
27 heart rate at ambient-relevant concentrations of NO2. However, changes in heart rate
28 variability found in one recent study suggest the possibility that NO2 exposure may lead
29 to effects on autonomic nervous system that are mediated via pulmonary irritant receptors
30 (Huang etal.. 2012a).
31 Alternatively, NO2-derived reaction products in the lung may "spillover" into the
32 circulation or lead to the "spillover" of other mediators into the circulation. One reaction
33 product of inhaled NO2, nitrite, is known to gain access to the circulation. In the presence
34 of red blood cell hemoglobin, nitrite is oxidized to nitrate (Postlethwait and Mustafa.
November 2013 3 -44 DRAFT: Do Not Cite or Quote
-------
1 1981) and nitrosylhemoglobin and methemoglobin are formed. Rapid appearance of
2 nitrite and nitrate in the blood was demonstrated in rats exposed for 1-2 hours to
3 5,000-40,000 ppb NO2 (Odaetal.. 1981). Elevated levels of blood nitrite and nitrate were
4 maintained as long as the exposure to NO2 continued. A small increase in levels of
5 nitrosylhemoglobin, but not methemoglobin, was detected in blood. The lack of
6 accumulation of methemoglobin was likely due to reduction of methemoglobin to
7 hemoglobin catalyzed by methemoglobin reductase. Two other studies measured
8 methemoglobin in the blood of mice exposed to NO2, with conflicting results (U.S. EPA.
9 1993).
10 Nitrite has known effects on blood cells, vascular cells and other tissues. Much recent
11 attention has been paid to nitrite's systemic vasodilatory effects that occur under hypoxic
12 conditions. As discussed in the 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993).
13 one controlled human exposure study demonstrated that NO2 exposure for a few hours
14 resulted in a reduction in blood pressure (Linn et al.. 1985b). which is consistent with the
15 systemic vasodilatory properties of nitrite under conditions of low oxygen. However
16 studies from other laboratories did not see this effect. Furthermore, dosimetric
17 considerations suggest that contributions of nitrite derived from ambient NO2 to plasma
18 levels of nitrite are small compared to nitrite derived from dietary sources.
19 Although unknown, NO 2-derived reaction products or mediators may transduce an
20 oxidative or other stress signal from the lung to the circulation. As summarized in the
21 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993). two controlled human exposure
22 studies involving NO2 inhalation over several hours found effects on circulating red
23 blood cells including reduced hemoglobin and hematocrit levels; one of these also found
24 reduced acetylcholinesterase activity (Frampton et al.. 2002; Posin et al.. 1978) (Section
25 4.3.6.2). Studies in animals have demonstrated decreases in red blood cell number as well
26 as increases in diphosphoglycerate, sialic acid, and methemoglobin following several
27 days of NO2 exposure (Section 4.3.6.3). However, changes in hematocrit and hemoglobin
28 did not occur following longer-term exposure to NO2. Additionally, blood lipids were
29 altered by exposure to NO2 for several weeks in obese rats. Studies in animal models
30 have demonstrated increases in blood glutathione levels resulting from NO2 exposure
31 (U.S. EPA. 2008c). Recent controlled human exposure studies (Section 4.3.6.2) also
32 found NO2 exposure-induced changes in blood lipids and increased levels of plasma
33 soluble lectin-like receptor for oxidized low-density lipoprotein (Channell et al.. 2012;
34 Huang etal. 2012a). Changes in peripheral blood inflammatory cells and tissue markers
35 of inflammation have also been observed following exposure to NO2. As summarized in
36 the 2008 ISA (U.S. EPA. 2008c). controlled human exposure studies demonstrated
37 changes in lymphocyte numbers and subsets in the peripheral blood following exposure
38 to NO2 (Frampton et al.. 2002; Sandstrom et al.. 1992). More recently, markers of
November 2013 3 -45 DRAFT: Do Not Cite or Quote
-------
1 inflammation were observed in myocardial tissue of NO 2 -exposed rats (Li etal.. 201 la)
2 (Section 4.3.6.3).
3 As summarized in the 2008 ISA and 1993 AQCD (U.S. EPA. 2008a. c, 1993). NO2
4 exposure in animals results in effects on the brain, the liver and xenobiotic metabolism. A
5 recent study in rats also demonstrated neurological effects of exposure to 2,500-10,000
6 ppb NO2 for seven days (Li etal.. 2012a) (Section 5.4.4.1). In addition, animal studies
7 demonstrated reproductive and developmental effects resulting from exposure to NO2
8 during gestation (U.S. EPA. 2008c). This included decreased litter size and neonatal
9 weight, lipid peroxidation of maternal lungs and placenta, retarded intrauterine
10 development, and disturbances in neuromotor development (Section 5.4). Epidemiologic
11 evidence of associations between exposure to NO2 and reproductive and developmental
12 effects has also been described (Section 5.4). The involvement in these effects of
13 NO2-mediated activation of pulmonary irritant receptors or spillover of NO2 metabolites
14 or inflammatory mediators from the lung to the circulation is not known.
15 There is no clear evidence that NO2 acts as a carcinogen (U.S. EPA. 2008a. c, 1993)
16 (Section 5.6). However, NO2 may act as a tumor promoter at the site of contact, possibly
17 due to its ability to produce cellular damage and promote regenerative cell proliferation.
18 In addition, It has been shown to be genotoxic and mutagenic in some systems, including
19 human nasal epithelial mucosa cells ex vivo where urban level concentrations were used
20 (Koehler et al.. 2011. 2010). Some studies demonstrated that inhaled NO2 at high
21 concentrations can contribute to the formation of mutagens and carcinogens if other
22 precursor chemicals are found in body; e.g., N-nitrosomorpholine from morpholine and
23 nitro-pyrene from pyrene (U.S. EPA. 2008c) (Section 3.2.4). However inhaled ambient
24 NO2 may not contribute significantly to the body burden of nitrite that can be derived
25 from other NO2 sources.
3.3.3 NO
26 As summarized in the 2008 ISA, 1993 AQCD (U.S. EPA. 2008a. c, 1993). and a recent
27 review (Hill etal.. 2010). the synthesis of endogenous NO in cells is catalyzed by three
28 different isoforms of NO synthases (eNOS, iNOS, nNOS). NO is involved in intracellular
29 signaling in virtually every cell and tissue. In general, low levels of endogenous NO play
30 important roles in cellular homeostasis, while higher levels are important in cellular
31 adaptation and still higher levels are cytotoxic. Further, signaling functions of NO may be
32 altered in the presence of acute inflammation (Hill etal.. 2010).
33 Like NO2, NO is a free radical. However, it is more selectively reactive than NO2 (Hill et
34 al.. 2010). In addition, it is more hydrophobic and can more easily cross cell membranes
November 2013 3 -46 DRAFT: Do Not Cite or Quote
-------
1 and diffuse much greater distances compared with NO2. As a result it there may be
2 overlap between endogenous and exogenous NO in terms of biological targets and
3 pathways. The following discussion focuses on mechanisms underlying the effects of
4 both endogenous and exogenous inhaled NO.
5 Since NO has a high affinity for heme-bound iron, many of its actions are related to its
6 interactions with heme proteins (Hill et al., 2010). For example, activation of the heme
7 protein guanylate cyclase is responsible for the smooth muscle relaxation and
8 vasodilation of pulmonary and systemic vessels, and possibly for bronchodilator effects.
9 Inhaled NO rapidly diffuses across the alveolar capillary barrier and reacts with soluble
10 guanylate cyclase in the pulmonary arterial smooth muscle. At the same time, inhaled NO
11 rapidly diffuses into the circulation and reacts with red blood cell hemoglobin to form
12 nitrosylhemoglobin, which is subsequently oxidized to methemoglobin and nitrate.
13 Increased blood concentrations of nitrosylhemoglobin and methemoglobin have been
14 reported in mice exposed for 1 hour to 20,000-40,000 ppb NO, as well as in mice
15 exposed chronically to 2,400 and 10,000 ppb NO (U.S. EPA. 1993). Some
16 S-nitrosohemoglobin may be formed in partially deoxygenated blood ("Wennmalm et al..
17 1993). NO can also disrupt iron-sulfur centers in proteins (Hill et al.. 2010). Furthermore,
18 redox reactions of NO and transition metals such as iron and copper facilitate S-
19 nitrosylation of protein and non-protein thiols. Binding of NO to iron- and copper-
20 containing proteins in the mitochondria may play an important role in mitochondrial
21 respiration. NO also rapidly reacts with superoxide, an oxygen-derived free radical, to
22 produce the potent oxidant peroxynitrite (Hill et al.. 2010). Peroxynitrite subsequently
23 reacts with CO2 to form the nitrosoperoxylcarbonate anion (ONOOCO2), followed by
24 decomposition to carbonate radical and NO2 (Section 3.2.2.4).
25 Endogenous NO is formed in the respiratory tract at high levels (Section 3.2.3) and it has
26 physiologic functions. The paranasal sinuses are a major source of NO in air derived
27 from the nasal airways, with average levels of 9,100 ppb NO (n = 5) measured in the
28 sinuses (Lundberg et al., 1995). Expression of iNOS was found to be higher in epithelial
29 cells of the paranasal sinuses than in epithelial cells of the nasal cavity. This NO derived
30 from nasal airways is thought to play a role in sinus host defense through bacteriostatic
31 activity. In addition, NO derived from nasal airways was found to modulate pulmonary
32 function in humans through effects on pulmonary vascular tone and blood flow
33 (Lundberg et al.. 1996). In healthy subjects, a comparison of nasal and oral breathing
34 demonstrated that nasal airway NO enhanced transcutaneous oxygen tension. In intubated
35 patients, nasal airway NO increased arterial oxygenation and decreased pulmonary
36 vascular resistance. Additionally, endogenous NO has been shown to act as a
37 bronchodilator (Belvisi etal.. 1992). Endogenous NO produced at high concentrations by
38 phagocytic cells is also known to participate in the killing of bacteria and parasites; this
November 2013 3 -47 DRAFT: Do Not Cite or Quote
-------
1 contributes to host defense (U.S. EPA. 2008c). Another effect of endogenous NO on host
2 defense is modulation of ciliary beat frequency (Jain et al.. 1993). Specifically, NO
3 derived from more distal airways was found to increase ciliary beat frequency.
4 Furthermore, endogenous NO production can be upregulated during inflammation
5 (Anderson et al., 2011). In fact, induction of iNOS in proximal or distal airways of
6 asthmatics results in levels of NO in exhaled breath as high as 20-50 ppb (Alving et al..
7 1993; Hamidetal.. 1993).
8 Endogenous NO has known pro- and anti-inflammatory effects and thus its role in
9 inflammatory lung disease is not clear. While it is known that both eNOS and iNOS
10 contribute to NO production in the lung, the relatively low levels of NO produced by
11 eNOS are thought to be more important in metabolic homeostasis (Ahmad et al., 2011).
12 Some evidence points to a role of iNOS-derived NO in the pathogenesis of asthma since
13 it has been correlated with inflammation, epithelial injury and clinical exacerbations of
14 asthma (Anderson et al.. 2011) (Section 3.3.2.6.2). Furthermore, preferential iNOS
15 upregulation was found in the distal airways compared with more proximal airways in
16 asthmatics. This is of interest since asthma is a disease of the small airways. As
17 mentioned above, signaling functions of NO may be altered in the presence of acute
18 inflammation (Hill et al.. 2010) which is characterized by enhanced levels of superoxide.
19 Superoxide reacts with NO to form peroxynitrite, which has been shown in animal
20 models to play a role in the pathogenesis of allergic airways disease (Section 3.3.2.6.2).
21 NO exposure has been shown to alter pulmonary function, morphology and vascular
22 function (U.S. EPA. 2008a. c, 1993). Studies in animals have demonstrated that inhaled
23 NO reversed acute methacholine-induced bronchoconstriction (Hogman et al.. 1993;
24 Dupuy et al.. 1992). This was observed with exposures of 5000 ppb NO in guinea pigs
25 and 80,000 ppb in rabbits. Chronic inhalation exposures have been found to alter the
26 morphology of the alveolar septal units in rats (Mercer etal.. 1995). This effect was not
27 seen with chronic inhalation exposures to NO2 at similar concentrations (500 ppb with
28 twice daily spikes of 1,500 ppb). In addition, inhaled NO has been shown to alter
29 transferrin and red blood cells in mice. Further, acute inhalation exposure of NO
30 decreased pulmonary vascular resistance in pigs and reduced pulmonary arterial pressure
31 in a rodent model of chronic pulmonary hypertension. A recent study also found that
32 inhaled NO (1,000, 5,000, 20,000 and 80,000 ppb) selectively dilated pulmonary blood
33 vessels, improved ventilation-perfusion mismatch, and reduced hypoxemia-induced
34 pulmonary vascular resistance in a pig model (Lovich et al., 2011).
35 Inhaled NO is used clinically at concentrations higher than those which are
36 environmentally relevant. Although it can cause both pulmonary and systemic
37 vasodilation, effects on pulmonary vasculature occur at relatively lower concentrations
November 2013 3 -48 DRAFT: Do Not Cite or Quote
-------
1 than required for vasodilation of systemic vessels. This selectivity for pulmonary
2 vasculature is likely due to the rapid scavenging of NO by hemoglobin in the blood.
3 Hence, inhaled NO has been used to mitigate pulmonary hypertension in newborns and
4 adults. High concentrations of inhaled NO are also known to alter ciliary beating and
5 mucus secretion in the airways, to increase renal output, to alter distribution of systemic
6 blood flow, to alter coagulation, fibrinolysis, and platelet functions and to modulate the
7 inflammatory response (U.S. EPA. 2008c).
8 Endogenous NO is an important mediator of cardiovascular homeostasis. It has anti-
9 inflammatory and anti-thrombotic effects, is cytoprotective and induces antioxidant
10 defenses (Wang and Widlansky. 2009). Two recent studies in animal models demonstrate
11 that high concentrations of inhaled NO may result in vascular toxicity. One of these
12 studies found rapid formation of plasma nitrites/nitrates and aortic S-nitrosothiols in rats
13 exposed acutely to NO (Knuckles et al., 2011). Plasma nitrites/nitrates doubled after an
14 hour of exposure to 3,000 ppb NO and tripled after an hour of exposure to 10,000 ppb
15 NO. These changes were accompanied by an enhanced constriction response to
16 endothelin-1 in coronary arterioles, which reflected altered vasomotor tone. Although this
17 latter effect appears to run counter to the vasodilator role of NO, it should be noted that
18 high concentrations of NO, as were used in this study, are known to inhibit eNOS activity
19 in other models (Griscavage et al., 1995). The increase in aortic eNOS content reported is
20 consistent with enzyme inactivation and turnover. Another recent animal toxicological
21 study conducted in ApoE"7" mice, a model of atherosclerosis, found the exposure to very
22 high concentrations of inhaled NO over the course of a week (17,000 ppb NO for 6
23 hours/day for 7 days) led to increases in mRNA for aortic endothelin-1 and MMP-9, as
24 well as to enhanced vascular gelatinase activity (Campen et al. 2010). These effects,
25 which are biomarkers of vascular remodeling and plaque vulnerability, were not seen
26 with 2,000 ppb NO2. The authors suggested that the activity of eNOS was uncoupled,
27 resulting in oxidative stress due to the production of superoxide instead of or in addition
28 to NO. Both of these studies suggest that inhaled NO has the potential to disrupt normal
29 signaling processes mediated by endogenous NO.
30 As mentioned above, endogenous NO plays key signaling roles in virtually every cell and
31 tissue (Hill et al.. 2010). and, as such, is an important mediator of homeostasis. Inhaled
32 NO at high enough concentrations has the potential to have beneficial or deleterious
33 effects on multiple organ systems. An important consideration is whether effects are
34 mediated by an NO metabolite, by the release of NO from a metabolite that serves as a
35 storage pool of NO or through methemogobin formation in the blood. Further discussion
36 of the biological functions of NO metabolites is found below.
November 2013 3 -49 DRAFT: Do Not Cite or Quote
-------
3.3.4 Metabolites of NO and NO2
3.3.4.1 Nitrites/Nitrates
1 Recently it has been proposed that nitrite is a storage form of NO since it can be reduced
2 back to NO under conditions of low oxygen tension in a reaction catalyzed by
3 deoxyhemoglobin (Gladwin et al.. 2005). In addition, nitrite is a signaling molecule in its
4 own right and does not require conversion to NO for this activity (Bryan. 2006). Nitrite
5 can increase cGMP levels and HSP20 expression, decrease CYP450 activity and alter
6 HO-1 expression (Bryan et al.. 2005). Nitrite is also bactericidal (Major et al.. 2010).
7 Furthermore, under acidic conditions, nitrite can react with thiols to form S-nitrosothiols.
8 Nitrite also reacts with hemoglobin to form iron-nitrosyl-hemoglobin and with
9 oxyhemoglobin to form nitrate. Nitrite acts as a vasodilator under hypoxic conditions,
10 through a reaction catalyzed by deoxyhemoglobin (Cosby et al.. 2003). The venous
11 circulation may be more sensitive to nitrite than the arterial circulation (Maher et al..
12 2008).
13 A recent study found that inhaled nitrite decreased pulmonary blood pressure in newborn
14 lambs with hemolysis-induced pulmonary vasoconstriction (Blood etal.. 2011). Nitrite
15 was converted to NO in lung tissue by a mechanism that did not require reaction with
16 deoxyhemoglobin in the circulation. This mechanism resulted in increased exhaled NO
17 gas as well as the relaxation of vascular smooth muscle which led to pulmonary
18 vasodilation. Although concentrations of inhaled nitrite employed were high (0.87 mol/L
19 sodium nitrite), this study is discussed here because it illustrates a novel biological
20 activity of lung nitrite that is normally formed by reactions of NO2 and NO in the ELF
21 and/or the blood.
3.3.4.2 S-Nitrosothiols
22 Exogenous and endogenous NO can increase S-nitrosothiols, protein S-glutathionylation
23 and thiol oxidation by cysteinyl thiol-dependent pathways (Hill etal.. 2010). All of these
24 post-translational protein modifications can act as redox switches to initiate cell signaling
25 events or alter enzyme activity. While NO does not react directly with thiol groups, it can
26 form S-nitrosothiols via reactions with thiyl groups and through intermediate formation
27 of N2O3 or metal nitrosyls. S-nitrosothiols are thought to serve as a storage or delivery
28 form of NO and play a role in cell signaling.
29 High concentrations of S-nitrosothiols are found in the lung where they act as
30 endogenous bronchodilators (Que et al.. 2009). In addition, S-nitrosothiols suppress
November 2013 3 -50 DRAFT: Do Not Cite or Quote
-------
1 inflammation by decreasing activation of the transcription factor NFxB (Marshall and
2 Stamler. 2001). Furthermore, augmentation of airway S-nitrosothiols by ethyl nitrite
3 inhalation protected against LPS-induced lung injury in an animal model (Marshall et al.,
4 2009). Several findings suggest an inverse relationship between endogenous airway S-
5 nitrosothiol levels and AHR. First, levels of airway S-nitrosoglutathione levels were
6 decreased in children with asthmatic respiratory failure and in adults with asthma (Que et
7 al., 2009; Gaston et al., 1998). Second, the enzyme nitrosoglutathione reductase
8 (GSNOR), which regulates airway S-nitrosoglutathione content, was expressed at higher
9 levels in BAL cell lysates in human asthmatics than in nonasthmatics (Que et al., 2009).
10 GSNOR expression was inversely correlated with S-nitrosoglutathione content. In
11 addition, GSNOR activity in BAL fluid was increased and inversely correlated with AHR
12 in human asthma (Que et al., 2009). Third, levels of airway S-nitrosothiols were inversely
13 correlated with AHR in human subjects with eosinophilic inflammation (Lee et al..
14 201 Ib).
3.3.4.3 Nitrated fatty acids and lipids
15 Nitration of fatty acids and lipids can occur in vivo under conditions of inflammation,
16 infection or ischemia/reperfusion; following exposure to exogenous NO2 and NO; and
17 possibly by reaction with nitrite (Higdon etal.. 2012; Khoo et al.. 2010). Nitrated fatty
18 acids (also known as nitro-fatty acids) can release NO which stimulates vascular smooth
19 muscle relaxation through cGMP-dependent pathways in vitro (Lima et al., 2005).
20 However, most of the cell signaling effects of nitrated fatty acids in vivo is likely due to
21 posttranslational modification of proteins (Khoo etal.. 2010). These electrophilic species
22 react with susceptible thiol groups in transcription factors (Higdon et al.. 2012; Bonacci
23 etal.. 2011).
24 Nitro-fatty acids such as nitro-oleic acid and nitro-linoleic acid are anti-inflammatory
25 (Bonacci etal.. 2011) and vasculoprotective (Khoo etal.. 2010). These effects are
26 mediated via activation of PPARy and the ARE pathway and suppression of NFxB and
27 STAT-1 pathways (Bonacci et al.. 2011). In a mouse model, nitro-oleic acid upregulated
28 vascular eNOS and HO-1 and inhibited angiotensisn Il-induced hypertension (Khoo et
29 al.. 2010; Zhang etal.. 2010). Nitro-oleic acid protected against ischemia/reperfusion
30 injury in a mouse model (Rudolph et al., 2010). Nitro-oleic acid also activated matrix
31 metalloproteinases (a pro-inflammatory effect) through thiol alkylation in vitro and
32 inhibited matrix metalloproteinase expression in macrophages through activation of
33 PPARy (Bonacci etal.. 2011). Matrix metalloproteinase was also suppressed in a mouse
34 model of atherosclerosis.
November 2013 3-51 DRAFT: Do Not Cite or Quote
-------
3.3.4.4 Nitrated amino acids and proteins
1 Peroxynitrite and NO2 can react with amino acids to produce nitrated amino acids and
2 proteins (Hill et al., 2010). These products can also be formed from nitrite and peroxide
3 in a reaction catalyzed by myeloperoxidase. Nitration of proteins may cause inhibition of
4 protein function and/or induce antigenicity. The presence of nitrated amino acids, such as
5 3-nitrotyrosine, in cells or tissues is an indicator of NO2 and/or peroxynitrite formation.
3.3.5 Summary
6 This section summarizes the key events and pathways that contribute to health effects
7 resulting from short-term and long-term exposures to NO2 and NO (Table 3-3). Both
8 older studies and studies published since the 2008 ISA (U.S. EPA. 2008c) and both
9 studies conducted in humans and studies conducted in animals provide insight into the
10 biological mechanisms which are affected by exposure to NO2 and NO. While studies
11 conducted at more environmentally-relevant concentrations are of greater interest, some
12 studies at higher concentrations are recent demonstrations of potentially important new
13 mechanisms.
NO2: Formation of secondary oxidation products
14 Studies in in vitro systems have shown that antioxidants react with NO2 to form reactive
15 intermediates and subsequently quench those reactive intermediates species. Consistent
16 with these findings, studies in humans and animals exposed to NO2 have demonstrated
17 changes in low molecular weight antioxidants such as glutathione, ascorbate and
18 a-tocopherol and that modulating antioxidant status alters levels of injury biomarkers.
19 Health effects resulting from NO2 exposure are likely due to reactive intermediates or
20 secondary oxidation products formed following initial reaction with ELF substrates.
21 These reaction products likely activate the following pathways.
NO2: Activation of neural reflexes
22 NO2 is classified as a pulmonary irritant. Irritant responses such as altered breathing
23 patterns and bradycardia have been demonstrated in animal models using high
24 concentrations of NO2. Increased airway resistance has been observed in humans but not
25 in animals exposed to high concentrations of NO2. This has been attributed to reflex
26 bronchoconstriction, however non-neural mechanisms may underly this response since
27 the changes in airway resistance were not accompanied by altered breathing patterns. A
November 2013 3 -52 DRAFT: Do Not Cite or Quote
-------
1 recent controlled human exposure study reported an effect of NO2 exposure on heart rate
2 variability, which is a measure of autonomic tone, at much lower concentrations of NO2.
3 This suggests the activation of a neural reflex in humans.
NO2: Initiation of inflammation
4 Increased levels of respiratory tract eicosanoids have been found in NO2 -exposed humans
5 and animals. These products may arise from lipid peroxidation and/or membrane
6 perturbation and play a role in neutrophil recruitment. A recent study demonstrated NFxB
7 activation in airway epithelium of animals exposed to high concentrations of NO2. NFxB
8 activation generally results in the synthesis and/or release of proinflammatory cytokines.
9 The influx of inflammatory cells in NO2-exposed animals and humans consists of
10 neutrophils and other cells types. Repeated exposure of humans to NO2 led to
11 upregulation of the adhesion cell molecule ICAM-1, persistent neutrophil inflammation,
12 and the upregulation of Th2 cytokines, the latter of which is characteristic of a pro-
13 allergic response. Studies involving repeated exposure of humans to NO2 also found a
14 reduction in lymphocyte subpopulations and other changes which may be characteristic
15 of impaired host defense mechanisms. NO2 exposure-induced inflammation generally
16 occurs at higher concentrations in humans however some effects of NO2 on allergic
17 inflammation may occur at lower concentrations (see below).
NO2: Alteration of epithelial barrier function
18 Epithelial barrier function is an important component of host defense. Increases in injury
19 biomarkers characteristic of impaired barrier function have been measured in the
20 respiratory tract of NO2-exposed humans and animals. This impairment may result from
21 lipid peroxidation and/or membrane perturbation and generally occurs only at higher
22 concentrations in humans. Under extreme conditions, increases in lung permeability may
23 cause death due to pulmonary edema. More subtle increases in lung permeability may
24 enhance the translocation of an antigen to the immune cells underlying the epithelium
25 which are involved in allergic responses or may contribute to activation of neural reflexes
26 and/or the stimulation of smooth muscle receptors by allowing greater access of agonist.
NO2: Enhancement of bronchial smooth muscle reactivity
27 Exposure to NO2 enhances the inherent reactivity of airway smooth muscle in healthy
28 and asthmatic human subjects and in animals. Several mechanisms may play a role in this
29 response, including impaired epithelial barrier function (as mentioned above),
30 enhancement of neural pathways leading to airway smooth muscle contraction,
November 2013 3-53 DRAFT: Do Not Cite or Quote
-------
1 oxidative/nitrative stress, mast cell activation and enhanced smooth muscle cell
2 contractility. Inflammation, especially allergic inflammation (see below), and airway
3 remodeling may contribute to these processes. Experiments in animals demonstrated that
4 NO2 exposure resulted in mast cell degranulation, production of eicosanoids (which may
5 sensitize receptors on nerve fibers) and eosinophilic inflammation. Studies in humans
6 also suggested a role for mast cell degranulation in NO 2 -exposure induced AHR.
NO2: Modification of innate/adaptive immunity
7 Innate and adaptive immunity encompass many different cell types and processes which
8 provide defense against foreign pathogens, repair tissue damage, and/or promote an
9 allergic phenotype. Studies in animals and humans exposed to NO2 demonstrated
10 impairment of mucociliary clearance and alveolar macrophage phagocytic activity. In
11 addition, NO2 exposure resulted in increased mortality in animal infectivity models and
12 altered lymphocyte subpopulations in humans (repeated exposures). These changes are
13 characteristic of impaired host defense. In addition, studies in humans and animals found
14 that NO2 exposure resulted in exacerbation of asthma and allergic airways disease. In
15 humans, both early and late phase asthmatic responses (e.g., AHR and eosinophil
16 activation following a specific allergen challenge) were enhanced by NO2 exposure. In
17 allergic animals, late phase responses (IgE, eosinophilia) were enhanced by NO2
18 exposure. Furthermore, recent studies in animal models of allergic airways disease and in
19 human subjects with asthma provide evidence that increased levels of endogenous
20 peroxynitrite/NO2 are linked to AHR and other features of allergic airways disease.
21 These findings raise the possibility that exogenous and endogenous NO2 act through
22 similar pathways to exacerbate allergic airways disease. In addition, NO2 exposure
23 resulted in Th2 skewing/allergic sensitization in animals and human subjects. This
24 included increased expression of Th2 cytokines in human respiratory epithelium and the
25 development of nasal eosinophilia and enhanced mast cell responses in animals.
26 Furthermore recent studies provide evidence that NO2 may act as an adjuvant promoting
27 the development of allergic airways disease in response to an inhaled allergen. These
28 findings suggest that exposure to NO2 may lead to the development of allergic airways
29 disease.
NO2: Remodeling of airways and alveoli
30 Morphologic changes in response to NO2 exposure generally consist of epithelial damage
31 in the centriacinar region, followed by hyperplastic repair. Emphysematous-like lesions
32 have been reported at high concentrations of NO2. The relationship between morphologic
33 changes observed in animal models and decrements in pulmonary function is unclear.
November 2013 3 -54 DRAFT: Do Not Cite or Quote
-------
NO2: Transduction of extrapulmonary responses
1 NO2 exposure results in effects outside the respiratory tract in both humans and animals,
2 especially at higher concentrations. Given the reactivity of NO2, these extrapulmonary
3 effects are likely secondary to the formation of an NO 2 reaction product. Two possible
4 mechanisms underlying extrapulmonary responses are activation of neural reflexes via
5 pulmonary irritant receptors and spillover of NO2 reaction products or inflammatory
6 mediators from the respiratory tract into the circulation. Experimental evidence in
7 animals demonstrated that exposure to high concentrations of NO2 activates vagal
8 pathways and results in bradycardia. Evidence for pulmonary irritant responses in
9 humans is weak although a recent study demonstrating NO2 exposure-induced changes in
10 heart rate variability is suggestive of this possibility. Another recent study found that
11 NO2 exposure resulted in the presence of a vasoactive substance in the plasma of human
12 volunteers, providing evidence that spillover of a reaction product or inflammatory
13 mediator into the circulation may transduce the signal from the respiratory tract.
NO: Initiating events
14 Due to its hydrophobic nature, selective reactivity and high affinity for heme proteins,
15 inhaled NO rapidly diffuses across the alveolar capillary barrier and reacts with soluble
16 guanylate cyclase in the pulmonary arterial smooth muscle and with hemoglobin in red
17 blood cells. Inhaled NO at high concentrations is used clinically and results in selective
18 pulmonary vasodilation.
Metabolites of NO2 and NO
19 Major metabolites of NO2 and NO include nitrites and nitrates, S-nitrosothiols, and
20 nitrated fatty acids, lipids, amino acids and proteins. With the exception of nitrate, all
21 have been shown to be biologically active. However, there is little available evidence to
22 support a role for these metabolites in mediating the effects of NO2 and NO considered in
23 this ISA.
November 2013 3-55 DRAFT: Do Not Cite or Quote
-------
Table 3-3 Biological pathways, key events and endpoints.
Biological Pathways
Key Events
Downstream Endpoints
NO2: Initiating Event-formation of
secondary oxidation products
Lipid Peroxidation
Antioxidant depletion
TROS/RNS
Thiol oxidation
NO2: Activation of neural reflexes
Altered breathing patterns
Possible reflex bronchoconstriction
Bradycardia
Airway resistance
NO2: Initiation of inflammation
^Eicosanoid production
NFKB Activation
ICAM-1 upregulation
Neutrophil influx
^Pro-inflammatory cytokines
Persistent neutrophil influx
NO2: Alteration of epithelial barrier
function
t BAL fluid protein or albumin
t Access of antigen to immune cells
t Access of agonist to airway smooth
muscle
t Access of mediators to irritant
receptors
Impaired host defense
Allergic inflammation
AHR
AHR
NO2: Enhancement of bronchial
smooth muscle reactivity
TROS/RNS
Mast cell degranulation
Allergic inflammation
Enhanced airway smooth muscle cell
contractility: inflammatory mediators
or airway remodeling
Sensitization of irritant receptors
AHR
NO2: Modification of innate/adaptive
immunity
Altered mucociliary clearance
Altered macrophage phagocytosis
Altered lymphocyte subsets
^Early phase and late phase
responses to allergen (AHR, allergic
inflammation)
^Late phase responses in allergic
model (allergic inflammation)
^Endogenous NO2 and peroxynitrite
Upregulation of Th2 cytokines
Nasal eosinophilia
Enhanced mast cell responses
Adjuvant activity: dendritic cell
activation
Impaired host defense: ^Animal
infectivity
Exacerbation of asthma and allergic
airways disease
Th2 skewing/Allergic sensitization
NO2: Remodeling of airways and
alveoli
Morphologic changes to centriacinar Unknown
region
November 2013
3-56
DRAFT: Do Not Cite or Quote
-------
Table 3-3 (Continued): Biological pathways, key events and endpoints.
Biological Pathways
Key Events
Downstream Endpoints
NC>2: Transduction of
extrapulmonary responses
Activation of neural reflex
Possibly altered HRV
Spillover into circulation
Presence of vasoactive substance in
plasma
NO: Initiating events
Rapid diffusion across alveolar
capillary barrier
Reaction with heme proteins
Lack of respiratory tract effects
Selective pulmonary vasodilation
Metabolites of NO2 and NO
Various biological activities
Unknown
3.4 Summary
i
2
3
4
5
6
7
This chapter provides a foundation for understanding how exposure to the gaseous air
pollutants NO2 and NO may lead to health effects. This encompasses the many steps
between uptake into the airways and biological responses that result from activation of
intra- and inter-cellular signaling pathways in a variety of tissues. Key to this process is
the reaction of NO2 with components of the ELF of the respiratory tract and the reaction
of NO with heme proteins in the circulation. These chemical interactions are responsible
for targeting of oxides of nitrogen species to different tissues. Inhaled NO2 may
contribute to the endogenous body burden of NO2 species, especially in the respiratory
tract.
November 2013
3-57
DRAFT: Do Not Cite or Quote
-------
References for Chapter 3
Adgent. MA; Squadrito. GL; Ballinger. CA; Krzywanski. DM; Lancaster. JR. Jr; Postlethwait. EM. (2012).
Desferrioxamine inhibits protein tyrosine nitration: Mechanisms and implications. Free Radic Biol Med
53: 951-961. http://dx.doi.Org/10.1016/i.freeradbiomed.2012.06.003
Ahmad. T; Mabaliraian. U; Sharma. A; Aich. J; Makhija. L; Ghosh. B; Agrawal. A. (2011). Simvastatin
improves epithelial dysfunction and airway hyperresponsiveness: From asymmetric dimethyl-arginine
to asthma. Am J Respir Cell Mol Biol 44: 531-539. http://dx.doi.org/10.1165/rcmb.2010-0041OC
Al-Hegelan. M; Tighe. RM; Castillo. C: Hollingsworth. JW. (2011). Ambient ozone and pulmonary innate
immunity [Review]. Immunol Res 49: 173-191. http://dx.doi.org/10.1007/sl2026-010-8180-z
Alarie. Y. (1973). Sensory irritation by airborne chemicals [Review]. CRC Crit Rev Toxicol 2: 299-363.
http://dx.doi.org/10.3109/10408447309082020
Alberg. T; Nilsen. A; Hansen. JS: Nygaard. UC: L0vik. M. (2011). Nitrogen dioxide: No influence on
allergic sensitization in an intranasal mouse model with ovalbumin and diesel exhaust particles. Inhal
Toxicol 23: 268-276. http://dx.doi.org/10.3109/08958378.2011.566898
Alexanderson. C: Olin. AC: Dahlman-Hoglund. A; Finizia. C: Toren. K. (2012). Nasal nitric oxide in a
random sample of adults and its relationship to sensitization, cat allergen, rhinitis, and ambient nitric
oxide. American Journal of Rhinology & Allergy 26: e99-e!03.
http://dx.doi.org/10.2500/aira.2012.26.3777
Alving. K; Weitzberg. E; Lundberg. JM. (1993). Increased amount of nitric oxide in exhaled air of
asthmatics. Eur Respir J 6: 1368-1370.
Anderson. JT; Zeng. M; Li. Q; Staplev. R; Moore. DR. II; Chenna. B; Fineberg. N; Zmijewski. J: Eltoum.
IE; Siegal. GP; Gaggar. A; Barnes. S: Velu. SE; Thannickal. VJ: Abraham. E; Patel. RP; Lancaster. JR.
Jr; Chaplin. DP; Dransfield. MT; Deshane. JS. (2011). Elevated levels of NO are localized to distal
airways in asthma. Free Radic Biol Med 50: 1679-1688.
http://dx.doi.0rg/10.1016/i.freeradbiomed.2011.03.015
Ather. JL; Alcorn. JF; Brown. AL; Guala. AS: Suratt. BT; Janssen-Heininger. YM; Poynter. ME. (2010).
Distinct functions of airway epithelial nuclear factor-kappaB activity regulate nitrogen dioxide-induced
acute lung injury. Am J Respir Cell Mol Biol 43: 443-451. http://dx.doi.org/10.1165/rcmb.2008-
0416OC
ATS/ERS (American Thoracic Society and European Respiratory Society). (2005). ATS/ERS
recommendations for standardized procedures for the online and offline measurement of exhaled lower
respiratory nitric oxide and nasal nitric oxide, 2005. Am J Respir Crit Care Med 171: 912-930.
http://dx.doi.org/10.1164/rccm.200406-710ST
Augusto. O: Bonini. MG: Amanso. AM; Linares. E; Santos. CC: De Menezes. SL. (2002). Nitrogen dioxide
and carbonate radical anion: two emerging radicals in biology [Review]. Free Radic Biol Med 32: 841-
859.
Ayaz. KL; Csallany. AS. (1978). Long-term NO2 exposure of mice in the presence and absence of vitamin
E. II. Effect of glutathione peroxidase. Arch Environ Occup Health 33: 292-296.
Azadniv. M; Utell. MJ; Morrow. PE; Gibb. FR; Nichols. J: Roberts. NJ. Jr; Speers. DM; Torres. A; Tsai. Y;
Abraham. MK; Voter. KZ; Frampton. MW. (1998). Effects of nitrogen dioxide exposure on human host
defense. Inhal Toxicol 10: 585-601. http://dx.doi.org/10.1080/089583798197556
Babior. BM; Takeuchi. C: Ruedi. J: Gutierrez. A; Wentworth. R (2003). Investigating antibody-catalyzed
ozone generation by human neutrophils. PNAS 100: 3031-3034.
http://dx.doi.org/10.1073/pnas.0530251100
November 2013 3-58 DRAFT: Do Not Cite or Quote
-------
Barck. C: Lundahl. J; Hallden. G: Bylin. G. (2005 a). Brief exposures to NO2 augment the allergic
inflammation in asthmatics. Environ Res 97: 58-66. http://dx.doi.Org/10.1016/i.envres.2004.02.009
Barck. C: Sandstrom. T; Lundahl. J; Hallden. G: Svartengren. M; Strand. V: Rak. S: Bylin. G. (2002).
Ambient level of NO2 augments the inflammatory response to inhaled allergen in asthmatics. Respir
Med 96: 907-917.
Barth. PJ: Muller. B; Wagner. U; Bittinger. A. (1995). Quantitative analysis of parenchyma! and vascular
alterations in NO2-induced lung injury in rats. Eur Respir J 8: 1115-1121.
http://dx.doi.org/10.1183/09031936.95.08071115
Bastacky. J: Lee. CY; Goerke. J: Koushafar. H; Yager. D; Kenaga. L; Speed. TP; Chen. Y; Clements. JA.
(1995). Alveolar lining layer is thin and continuous: Low-temperature scanning electron microscopy of
rat lung. J Appl Physiol 79: 1615-1628.
Bates. DV: Macklem. PT; Christie. RV. (1971). Respiratory function in disease: An introduction to the
integrated study of the lung (2nd ed.). Philadelphia, PA: W.B. Saunders Company.
Bauer. MA; Utell. MJ; Morrow. PE; Speers. DM; Gibb. FR. (1986). Inhalation of 0.30 ppm nitrogen
dioxide potentiates exercise-induced bronchospasm in asthmatics. Am Rev Respir Dis 134: 1203-1208.
Bautista. AP; Eisenlohr. CP; Lanz. MJ. (2011). Nasal nitric oxide and nasal eosinophils decrease with
levocetirizine in subjects with perennial allergic rhinitis. American Journal of Rhinology & Allergy 25:
383-387. http://dx.doi.org/10.2500/aira.2011.25.3668
Beaven. MA. (2009). Our perception of the mast cell from Paul Ehrlich to now. Eur J Immunol 39: 11-25.
http://dx.doi.org/10.1002/eii.200838899
Becquemin. MM; Bertholon. JF; Bouchikhi. A; Malarbet. JL; Roy. M. (1999). Oronasal ventilation
partitioning in adults and children: Effect on aerosol deposition in airways. Radiat Prot Dosimetry 81:
221-228.
Belvisi. MG: Stretton. CD; Miura. M; Verleden. GM; Tadjkarimi. S: Yacoub. MH; Barnes. PJ. (1992).
Inhibitory NANC nerves in human trachea! smooth muscle: A quest for the neurotransmitter. J Appl
Physiol 73: 2505-2510.
Bennett. WD; Zeman. KL; Jarabek. AM. (2008). Nasal contribution to breathing and fine particle
deposition in children versus adults. J Toxicol Environ Health A 71: 227-237.
http://dx.doi.org/10.1080/15287390701598200
Bernhard. W: Haslam. PL; Floros. J. (2004). From birds to humans: new concepts on airways relative to
alveolar surfactant [Review]. Am J Respir Cell Mol Biol 30: 6-11. http://dx.doi.org/10.1165/rcmb.2003-
0158TR
Bevelander. M; Mavette. J: Whittaker. LA; Paveglio. SA; Jones. CC: Robbins. J: Hemenwav. D; Akira. S:
Uematsu. S: Poynter. ME. (2007). Nitrogen dioxide promotes allergic sensitization to inhaled antigen. J
Immunol 179: 3680-3688.
Bidani. A; Postlethwait EM. (1998). Kinetic determinants of inhaled reactive gas absorption. In MP
Hlastala; HT Robertson (Eds.), Complexity in structure and function of the lung (pp. 243-296). New
York, NY: Marcell Dekker Inc.
Blomberg. A; Krishna. MT; Bocchino. V; Biscione. GL; Shute. JK; Kelly. FJ; Frew. AJ; Holgate. ST;
Sandstrom. T. (1997). The inflammatory effects of 2 ppm NO2 on the airways of healthy subjects. Am J
Respir Crit Care Med 156: 418-424.
Blomberg. A; Krishna. MT; Helledav. R; Soderberg. M; Ledin. MC: Kelly. FJ: Frew. AJ: Holgate. ST;
Sandstrom. T. (1999). Persistent airway inflammation but accomodated antioxidant and lung function
responses after repeated daily exposure to nitrogen dioxide. Am J Respir Crit Care Med 159: 536-543.
Blood. AB; Schroeder. HJ; Terry. MH; Merrill-Henry. J: Bragg. SL; Vrancken. K; Liu. T; Herring. JL;
Sowers. LC: Wilson. SM; Power. GG. (2011). Inhaled nitrite reverses hemolysis-induced pulmonary
vasoconstriction in newborn lambs without blood participation. Circulation 123: 605-612.
http://dx.doi.org/10.1161/CIRCULATIONAHA.110.001073
November 2013 3-59 DRAFT: Do Not Cite or Quote
-------
Bonacci. G: Baker. PR; Salvatore. S. R.: Shores. D; Khoo. NK; Koenitzer. J. R.; Vitturi. DA; Woodcock. S.
R.; Golin-Bisello. F; Cole. MP; Watkins. S: St Croix. C: Batthyany. CI; Freeman. BA; Schopfer. FJ.
(2012). Conjugated linoleic acid is a preferential substrate for fatty acid nitration. J Biol Chem 287:
44071-44082. http://dx.doi.org/10.1074/ibc.M112.401356
Bonacci. G: Schopfer. FJ: Batthyany. CI: Rudolph. TK; Rudolph. V; Khoo. NK; Kellev. EE; Freeman. BA.
(2011). Electrophilic fatty acids regulate matrix metalloproteinase activity and expression. J Biol Chem
286: 16074-16081. http://dx.doi.org/10.1074/ibc.Mlll.225029
Borland. CDR; Dunningham. H; Bottrill. F; Vuylsteke. A; Yilmaz. C: Dane. DM; Hsia. CCW. (2010).
Significant blood resistance to nitric oxide transfer in the lung. J Appl Physiol 108: 1052-1060.
http://dx.doi.org/10.1152/iapplphvsiol.00904.2009
Bryan. NS. (2006). Nitrite in nitric oxide biology: Cause or consequence? A systems-based review
[Review]. Free Radic Biol Med 41: 691-701. http://dx.doi.Org/10.1016/i.freeradbiomed.2006.05.019
Bryan. NS: Fernandez. BO; Bauer. SM; Garcia-Saura. MF; Milsom. AB; Rassaf. T; Malonev. RE; Bharti.
A; Rodriguez. J: Feelisch. M. (2005). Nitrite is a signaling molecule and regulator of gene expression in
mammalian tissues. Nat Chem Biol 1: 290-297. http://dx.doi.org/10.1038/nchembio734
Bylin. G: Lindvall. T; Rehn. T; Sundin. B. (1985). Effects of short-term exposure to ambient nitrogen
dioxide concentrations on human bronchial reactivity and lung function. Eur Respir J 66: 205-217.
Campen. MJ; Lund. AK; Doyle-Eisele. ML; McDonald. JD; Knuckles. TL; Rohr. AC: Knipping. EM;
Mauderly. JL. (2010). A comparison of vascular effects from complex and individual air pollutants
indicates a role for monoxide gases and volatile hydrocarbons. Environ Health Perspect 118: 921-927.
http://dx.doi.org/10.1289/ehp.0901207
Cardinale. F; Tesse. R: Fucilli. C: Loffredo. MS: lacoviello. G: Chinellato. I; Armenio. L. (2007).
Correlation between exhaled nitric oxide and dietary consumption of fats and antioxidants in children
with asthma [Letter]. J Allergy Clin Immunol 119: 1268-1270.
http://dx.doi.0rg/10.1016/i.iaci.2007.01.028
Cavanagh. DG: Morris. JB. (1987). Mucus protection and airway peroxidation following nitrogen dioxide
exposure in the rat. J Toxicol Environ Health 22: 313-328.
http://dx.doi.org/10.1080/15287398709531074
Chadha. TS: Birch. S: Sackner. MA. (1987). Oronasal distribution of ventilation during exercise in normal
subjects and patients with asthma and rhinitis. Chest 92: 1037-1041.
http://dx.doi.0rg/10.1378/chest.92.6.1037
Chakrabortv. S: Balakotaiah. V; Bidani. A. (2004). Diffusing capacity reexamined: Relative roles of
diffusion and chemical reaction in red cell uptake of O2, CO, CO2, and NO. J Appl Physiol 97: 2284-
2302. http://dx.doi.org/10.1152/iapplphvsiol.00469.2004
Chambers. DC: Ayres. JG. (2001). Effects of nitrogen dioxide exposure and ascorbic acid supplementation
on exhaled nitric oxide in healthy human subjects. Thorax 56: 774-778.
http://dx.doi.org/10.1136/tfaorax.56.10.774
Chang. LY; Graham. JA; Miller. FJ: Ospital. JJ: Crapo. JD. (1986). Effects of subchronic inhalation of low
concentrations of nitrogen dioxide: I. The proximal alveolar region of juvenile and adult rats. Toxicol
Appl Pharmacol 83: 46-61. http://dx.doi.org/10.1016/0041-008X(86)90321-2
Channell. MM; Paffett. ML; Devlin. RB; Madden. MC: Campen. MJ. (2012). Circulating factors induce
coronary endothelial cell activation following exposure to inhaled diesel exhaust and nitrogen dioxide in
humans: Evidence from a novel translational in vitro model. Toxicol Sci 127: 179-186.
http://dx.doi.org/10.1093/toxsci/kfs084
Chase. MW. Jr. (1998). NIST-JANAF thermochemical tables, monograph 9 (Fourth Edition ed.).
Washington DC: American Chemical Society, http://www.nist.gov/srd/reprints.cfm
November 2013 3 -60 DRAFT: Do Not Cite or Quote
-------
Ckless. K; Hodgkins. SR; Ather. JL; Martin. R; Poynter. ME. (2011). Epithelial, dendritic, and CD4(+) T
cell regulation of and by reactive oxygen and nitrogen species in allergic sensitization [Review].
Biochim Biophys Acta 1810: 10251034. http://dx.doi.Org/10.1016/i.bbagen.2011.03.005
Ckless. K; Lampert. A; Reiss. J: Kasahara. D; Poynter. ME; Irvin. CG: Lundblad. LKA; Norton. R; van der
Vliet. A; Janssen-Heininger. YMW. (2008). Inhibition of arginase activity enhances inflammation in
mice with allergic airway disease, in association with increases in protein S-nitrosylation and tyrosine
nitration. JImmunol 181: 4255-4264.
Cockcroft. DW: Davis. BE. (2006c). Mechanisms of airway hyperresponsiveness [Review]. J Allergy Clin
Immunol 118: 551-559. http://dx.doi.Org/10.1016/i.jaci.2006.07.012
Comhair. SAA; Erzurum. SC. (2010). Redox control of asthma: Molecular mechanisms and therapeutic
opportunities [Review]. Antioxid Redox Signal 12: 93-124. http://dx.doi.org/10.1089/ARS.2008.2425
Connor. LM; Bidani. A; Goerke. J: Clements. JA; Postlethwait. EM. (2001). NO2 interfacial transfer is
reduced by phospholipid monolayers. J Appl Physiol 91: 2024-2034.
Cosby. K; Partovi. KS: Crawford. JH; Patel. RP; Reiter. CD; Martyr. S: Yang. BK; Waclawiw. MA; Zalos.
G: Xu. X; Huang. KT; Shields. H; Kim-Shapiro. DB; Schechter. AN; Cannon. RO; Gladwin. MT.
(2003). Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates the human circulation. Nat
Med9: 1498-1505. http://dx.doi.org/10.1038/nm954
Davies. RJ: Rusznak. C: Calderon. MA; Wang. JH; Abdelaziz. MM; Devalia. JL. (1997). Allergen-irritant
interaction and the role of corticosteroids [Review]. Allergy 52: 59-65. http://dx.doi.org/10.1111/i. 1398-
9995.1997.tb04873.x
Davis. JK; Davidson. MK; Schoeb. TR; Lindsev. JR. (1992). Decreased intrapulmonary killing of
Mycoplasma pulmonis after short-term exposure to NO2 is associated with damaged alveolar
macrophages. Am Rev Respir Dis 145: 406-411. http://dx.doi.Org/10.1164/ajrccm/145.2 Pt 1.406
de Burbure. CY; Heilier. JF; Neve. J: Becker. A; Albrecht. C: Borm. PJA; Gromadzinska. J: Wasowicz. W:
Rydzynski. K; Bernard. AM. (2007). Lung permeability, antioxidant status, and NO2 inhalation: A
selenium supplementation study in rats. J Toxicol Environ Health A 70: 284-294.
http://dx.doi.org/10.1080/15287390600884875
de Winter-Sorkina. R: Cassee. FR. (2002). From concentration to dose: Factors influencing airborne
paniculate matter deposition in humans and rats. (RIVM Report 650010031). Bilthoven, Netherlands:
National Institute for Public Health and the Environment.
http://www.rivm.nl/en/Documents and_publications/Scientific/Reports/2003/februari/From concentrati
on to dose factors influencing airborne_particulate matter deposition in humans and rats?sp=cml2
bXE9ZmFsc2U7c2VhcmNoYmFzZTO 1 MDI5MDtvaXZtcT 1m YWxzZTs=&pagenr=5 03 0
Devlin. RB; Horstman. DP; Gerritv. TR; Becker. S: Madden. MC. (1999). Inflammatory response in
humans exposed to 2.0 PPM nitrogen dioxide. Inhal Toxicol 11: 89-109.
http://dx.doi.org/10.1080/089583799197195
Dimeo. MJ; Glenn. MG: Holtzman. MJ; Sheller. JR; Nadel. JA; Boushev. HA. (1981). Threshold
concentration of ozone causing an increase in bronchial reactivity in humans and adaptation with
repeated exposures. Am Rev Respir Dis 124: 245-248.
Powell. AR; Kilburn. KH; Pratt. PC. (1971). Short-term exposure to nitrogen dioxide: Effects on
pulmonary ultrastructure, compliance, and the surfactant system. Arch Intern Med 128: 74-80.
http://dx.doi.org/10.1001/archinte.1971.00310190078009
Drahl. C. (2009). Probing for in-body ozone. Chem Eng News 87: 40-42. http://dx.doi.org/10.1021/cen-
v087n025.pQ40
Dressel. H; Miiller. F; Fischer. R; Rommelt H; Hohlfeld. JM; Behr. J: Huber. RM; Nowak. D; Jorres. RA.
(2010). Independent information of nonspecific biomarkers in exhaled breath condensate. Respiration
80: 401-409. http://dx.doi.org/10.1159/000319945
November 2013 3-61 DRAFT: Do Not Cite or Quote
-------
Dupuv. PM; Shore. SA; Drazen. JM; Frostell. C: Hill. WA; Zapol WM. (1992). Bronchodilator action of
inhaled nitric oxide in guinea pigs. J Clin Invest 90: 421-428. http://dx.doi.org/10.1172/JCI115877
Ehrlich. R. (1980). Interaction between environmental pollutants and respiratory infections. Environ Health
Perspect35: 89-99.
Ehrlich. R; Findlav. JC: Fenters. JD; Gardner. DE. (1977). Health effects of short-term inhalation of
nitrogen dioxide and ozone mixtures. Environ Res 14: 223-231. http://dx.doi.org/10.1016/0013-
9351(77)90034-2
Enami. S: Hoffmann. MR; Colussi. AJ. (2009). Absorption of inhaled NO(2) [Letter]. J Phys Chem B 113:
7977-7981. http://dx.doi.org/10.1021/ip902667x
England. C: Corcoran. WH. (1974). Kinetics and mechanisms of the gas-phase reaction of water vapor and
nitrogen dioxide. Ind Eng Chem Fundam 13: 373-384. http://dx.doi.org/10.1021/il60052a014
Erjefalt JS: Greiff. L; Andersson. M; Matsson. E; Petersen. H; Linden. M; Ansari. T; Jefferv. PK; Persson.
CG. (1999). Allergen-induced eosinophil cytolysis is a primary mechanism for granule protein release in
human upper airways. Am J Respir Crit Care Med 160: 304-312.
http://dx.doi.0rg/10.1164/ajrccm.160.l.9809048
Fehrenbach. H; Zimmermann. G: Starke. E; Bratu. VA; Conrad. D; Yildirim. AO; Fehrenbach. A. (2007).
Nitrogen dioxide induces apoptosis and proliferation but not emphysema in rat lungs. Thorax 62: 438-
446. http://dx.doi.org/10.1136/thx.2006.062364
Finlavson-Pitts. BJ; Sweetman. LL; Weissbart. B. (1987). A Fourier transform infrared spectrometry study
of the reactions of phosphatidylcholines with gaseous N2O5 and NO2. Toxicol Appl Pharmacol 89:
438-448.
Finlavson-Pitts. BJ: Wingen. LM; Sumner. AL; Svomin. D; Ramazan. KA. (2003). The heterogeneous
hydrolysis of NO2 in laboratory systems and in outdoor and indoor atmospheres: An integrated
mechanism. Phys Chem Chem Phys 5: 223-242. http://dx.doi.org/10.1039/b208564j
Folinsbee. LJ. (1992). Does nitrogen dioxide exposure increase airways responsiveness? Toxicol Ind Health
8: 273-283.
Ford. E; Hughes. MN; Wardman. P. (2002). Kinetics of the reactions of nitrogen dioxide with glutathione,
cysteine, and uric acid at physiological pH. Free Radic Biol Med 32: 1314-1323.
http://dx.doi.org/10.1016/S0891-5849(02)00850-X
Forstermann. U; Sessa. WC. (2012). Nitric oxide synthases: Regulation and function [Review]. European
Heart Journal (Online) 33: 829-837. http://dx.doi.org/10.1093/eurheartj/ehr304
Frampton. MW: Boscia. J: Roberts. NJ. Jr; Azadniv. M; Torres. A; Cox. C: Morrow. PE; Nichols. J:
Chalupa. D: Frasier. LM; Gibb. FR; Speers. DM; Tsai. Y; Utell. MJ. (2002). Nitrogen dioxide exposure:
Effects on airway and blood cells. Am J Physiol 282: L155-L165.
Freeman. G: Furiosi. NJ: Havdon. GB. (1966). Effects of continuous exposure of 0.8 ppm NO2 on
respiration of rats. Arch Environ Health 13: 454-456.
http://dx.doi.org/10.1080/00039896.1966.10664595
Fujimaki. H; Nohara. O. (1994). Changes in the response of lung mast cells isolated from rats and guinea
pigs exposed to nitrogen dioxide. Inhal Toxicol 6: 515-520.
http://dx.doi.org/10.3109/08958379409040508
Gaston. B; Sears. S: Woods. J: Hunt. J: Ponaman. M; Mcmahon. T; Stamler. J. (1998). Bronchodilator S-
nitrosothiol deficiency in asthmatic respiratory failure. Lancet 351: 1317-1319.
http://dx.doi.org/10.1016/S0140-6736(97)07485-0
Gelb. AF; Barnes. PJ: George. SC: Ricciardolo. FL; Dimaria. G: Zamel. N. (2012). Review of exhaled
nitric oxide in chronic obstructive pulmonary disease. J Breath Res 6: 047101.
http://dx.doi.Org/10.1088/1752-7155/6/4/047101
November 2013 3 -62 DRAFT: Do Not Cite or Quote
-------
Gilmour. MI; Park. P; Selgrade. MK. (1996). Increased immune and inflammatory responses to dust mite
antigen in rats exposed to 5 ppm NO2. Fundam Appl Toxicol 31: 65-70.
Ginsberg. GL; Foos. BP; Firestone. MR (2005). Review and analysis of inhalation dosimetry methods for
application to children's risk assessment [Review]. J Toxicol Environ Health A 68: 573-615.
http://dx.doi.org/10.1080/15287390590921793
Gladwin. MT; Schechter. AN; Kim-Shapiro. DB; Patel. RP; Hogg. N; Shiva. S: Cannon. RO. Ill; Kelm. M;
Wink. DA; Espev. MG: Oldfield. EH; Pluta. RM; Freeman. BA; Lancaster. J. R.; Feelisch. M;
Lundberg. JO. (2005). The emerging biology of the nitrite anion. Nat Chem Biol 1: 308-314.
http://dx.doi.org/10.1038/nchembioll05-308
Goings. SAJ; Kulle. TJ; Bascom. R; Sauder. LR; Green. DJ; Hebel. JR; Clements. ML. (1989). Effect of
nitrogen dioxide exposure on susceptibility to influenza A virus infection in healthy adults. Am Rev
RespirDis 139: 1075-1081. http://dx.doi.Org/10.1164/ajrccm/139.5.1075
Goldstein. S: Lind. J: Merenyi. G. (2004). Reaction of Organic Peroxyl Radicals with NO 2 and NO in
Aqueous Solution: Intermediacy of Organic Peroxynitrate and Peroxynitrite Species. J Phys Chem A
108: 1719-1725. http://dx.doi.org/10.1021/ip037431z
Gray. EL; Patton. FM; Goldberg. SB; Kaplan. E. (1954). Toxicity of the oxides of nitrogen. II. Acute
inhalation toxicity of nitrogen dioxide, red fuming nitric acid, and white fuming nitric acid. AMAArch
Ind Health 10: 418-422.
Greene. ND; Schneider. SL. (1978). Effects of NO2 on the response of baboon alveolar macrophages to
migration inhibitory factor. J Toxicol Environ Health 4: 869-880.
http://dx.doi.org/10.1080/15287397809529708
Griscavage. JM; Hobbs. AJ; Ignarro. LJ. (1995). Negative modulation of nitric oxide synthase by nitric
oxide and nitroso compounds. In LJ Ignarro; F Murad (Eds.), Nitric oxide: Biochemistry, molecular
biology, and therapeutic implications (pp. 215-234). San Diego, CA: Academic Press.
http://dx.doi.org/10.1016/S1054-3589(08)61088-l
Gushchin. IS: Petvaev. IM; Tsinkalovskv. OR. (1990). Kinetics of oxygen metabolism indices in the course
of histamine secretion from rat mast cells. Agents and Actions 30: 85-88.
http://dx.doi.org/10.1007/BF01969005
Hamid. 0: Springall. PR; Riveros-Moreno. V; Chanez. P; Howarth. P; Redington. A; Bousquet. J: Godard.
P; Holgate. S: Polak. JM. (1993). Induction of nitric oxide synthase in asthma. Lancet 342: 1510-1513.
http://dx.doi.org/10.1016/S0140-6736(05)80083-2
Hatch. GE; Slade. R: Selgrade. MK; Stead. AG. (1986). Nitrogen dioxide exposure and lung antioxidants in
ascorbic acid-deficient guinea pigs. Toxicol Appl Pharmacol 82: 351-359.
http://dx.doi.org/10.1016/0041-008X(86)90212-7
Helledav. R: Huberman. D; Blomberg. A; Stjernberg. N; Sandstrom. T. (1995). Nitrogen dioxide exposure
impairs the frequency of the mucociliary activity in healthy subjects. Eur Respir J 8: 1664-1668.
http://dx.doi.org/10.1183/09031936.95.08101664
Heller. H; Korbmacher. N; Gabler. R; Brandt. S: Breitbach. T; Juergens. U: Grohe. C: Schuster. KD.
(2004). Pulmonary 15NO uptake in interstitial lung disease. Nitric Oxide 10: 229-232.
http://dx.doi.0rg/10.1016/i.niox.2004.06.001
Hesterberg. TW: Bunn. WB; McClellan. RO; Hamade. AK; Long. CM; Valberg. PA. (2009). Critical review
of the human data on short-term nitrogen dioxide (NO2) exposures: Evidence for NO2 no-effect levels
[Review]. Crit Rev Toxicol 39: 743-781. http://dx.doi.org/10.3109/10408440903294945
Higdon. A; Piers. AR; Oh. J: Landar. A; Darlev-Usmar. VM. (2012). Cell signalling by reactive lipid
species: new concepts and molecular mechanisms [Review]. Biochem J 442: 453-464.
http://dx.doi.org/10.1042/BJ20111752
November 2013 3 -63 DRAFT: Do Not Cite or Quote
-------
Hill. BG: Dranka. BP; Bailey. SM; Lancaster. JR; Darlev-Usmar. VM. (2010). What part of NO don't you
understand? Some answers to the cardinal questions in nitric oxide biology [Review]. J Biol Chem 285:
19699-19704. http://dx.doi.org/10.1074/ibc.R110.101618
Hodgkins. SR; Ather. JL; Paveglio. SA; Allard. JL; Leclair. LA; Suratt. BT; Boyson. JE; Poynter. ME.
(2010). NO2 inhalation induces maturation of pulmonary CDllc+ cells that promote antigenspecific
CD4+T cell polarization. RespirRes 11: 102. http://dx.doi.org/10.1186/1465-9921-ll-102
Hogman. M; Frostell. C: Arnberg. H; Hedenstierna. G. (1993). Inhalation of nitric oxide modulates
methacholine-induced bronchoconstriction in the rabbit. Eur Respir J 6: 177-180.
Hohlfeld. JM. (2002). The role of surfactant in asthma. Respir Res 3: 4. http://dx.doi.org/10.1186/rrl76
Huang. W: Zhu. T; Pan. X; Hu. M; Lu. SB; Lin. Y; Wang. T; Zhang. Y; Tang. X. (2012a). Air pollution and
autonomic and vascular dysfunction in patients with cardiovascular disease: Interactions of systemic
inflammation, overweight, and gender. Am J Epidemiol 176: 117-126.
http://dx.doi.org/10.1093/aie/kwr511
Huang. YC: Rappold. AG: Graff. DW: Ohio. AJ; Devlin. RB. (2012b). Synergistic effects of exposure to
concentrated ambient fine pollution particles and nitrogen dioxide in humans. Inhal Toxicol 24: 790-
797. http://dx.doi.org/10.3109/08958378.2012.718809
Hubbard. AK; Symanowicz. PT; Thibodeau. M; Thrall. RS: Schramm. CM; Cloutier. MM; Morris. JB.
(2002). Effect of nitrogen dioxide on ovalbumin-induced allergic airway disease in a murine model. J
Toxicol Environ Health A 65: 1999-2005. http://dx.doi.org/10.1080/00984100290071397
Huie. RE. (1994). The reaction kinetics of NO2 [Review]. Toxicology 89: 193-216.
http://dx.doi.org/10.1016/0300-483X(94)90098-l
Jain. B; Rubinstein. I; Robbins. RA; Leise. KL; Sisson. JH. (1993). Modulation of airway epithelial cell
ciliary beat frequency by nitric oxide. Biochem Biophys Res Commun 191: 83-88.
http://dx.doi.org/10.1006/bbrc.1993.1187
Jakab. GJ. (1987). Modulation of pulmonary defense mechanisms by acute exposures to nitrogen dioxide.
Experientia, Suppl 51: 235-242.
Jenkins. HS: Devalia. JL: Mister. RL; Sevan. AM; Rusznak. C: Davies. RJ. (1999). The effect of exposure
to ozone and nitrogen dioxide on the airway response of atopic asthmatics to inhaled allergen: dose- and
time-dependent effects. Am J Respir Crit Care Med 160: 33-39.
http://dx.doi.0rg/10.1164/ajrccm.160.l.9808119
Jilma. B; Kastner. J: Mensik. C: Vondrovec. B; Hildebrandt. J: Krejcv. k; Wagner. OF; Eichler. H. g.
(1996). Sex differences in concentrations of exhaled nitric oxide and plasma nitrate. Life Sci 58: 469-
476. http://dx.doi.org/10.1016/0024-3205(95)02311-9
Jorres. R; Nowak. D; Grimminger. F; Seeger. W: Oldigs. M; Magnussen. H. (1995). The effect of 1 ppm
nitrogen dioxide on bronchoalveolar lavage cells and inflammatory mediators in normal and asthmatic
subjects. Eur Respir J 8: 416-424. http://dx.doi.org/10.1183/09031936.95.08030416
Kelly. FJ; Blomberg. A; Frew. A; Holgate. ST; Sandstrom. T. (1996a). Antioxidant kinetics in lung lavage
fluid following exposure of humans to nitrogen dioxide. Am J Respir Crit Care Med 154: 1700-1705.
http://dx.doi.0rg/10.1164/ajrccm.154.6.8970358
Kelly. FJ; Cotgrove. M; Mudwav. IS. (1996b). Respiratory tract lining fluid antioxidants: the first line of
defence against gaseous pollutants. Cent Eur J Public Health 4: 11-14.
Kelly. FJ: Tetlev. TD. (1997). Nitrogen dioxide depletes uric acid and ascorbic acid but not glutathione
from lung lining fluid. Biochemistry 325: 95-99.
Kettle. AJ; Clark. BM; Winterbourn. CC. (2004). Superoxide converts indigo carmine to isatin sulfonic
acid: Implications for the hypothesis that neutrophils produce ozone. J Biol Chem 279: 18521-18525.
http://dx.doi.org/10.1074/ibc.M400334200
November 2013 3 -64 DRAFT: Do Not Cite or Quote
-------
Kharitonov. SA; Walker. L; Barnes. PJ. (2005). Repeatability of standardised nasal nitric oxide
measurements in healthy and asthmatic adults and children. Respir Med 99: 1105-1114.
http://dx.doi.0rg/10.1016/i.rmed.2005.02.019
Khoo. NK; Rudolph. V; Cole. MP; Golin-Bisello. F; Schopfer. FJ; Woodcock. SR; Batthyany. C: Freeman.
BA. (2010). Activation of vascular endothelial nitric oxide synthase and heme oxygenase-1 expression
by electrophilic nitro-fatty acids. Free Radic Biol Med 48: 230-239.
http://dx.doi.0rg/10.1016/i.freeradbiomed.2009.10.046
Kikugawa. K; Kogi. K. (1987). Oxidation products of linoleate and linolenate exposed to nitrogen dioxide.
Chem Pharm Bull (Tokyo) 35: 344-349. http://dx.doi.org/10.1248/cpb.35.344
Kirsch. M; Korth. HG: Sustmann. R; De Groot. H. (2002). The pathobiochemistry of nitrogen dioxide
[Review]. Biol Chem 383: 389-399. http://dx.doi.org/10.1515/BC.2002.043
Kitabatake. M; Yamamoto. H; Yuan. PF; Manjurul. H; Murase. S: Yamauchi. T. (1995). Effects of exposure
to NO2 or SO2 on bronchopulmonary reaction induced by Candida albicans in guinea pigs. J Toxicol
Environ Health 45: 75-82. http://dx.doi.org/10.1080/15287399509531981
Kleeberger. SR: Zhang. LY; Jakab. GJ. (1997). Differential susceptibility to oxidant exposure in inbred
strains of mice: Nitrogen dioxide versus ozone. Inhal Toxicol 9: 601-621.
http://dx.doi.org/10.1080/089583797198006
Kleinman. MT; Mautz. WJ. (1991). The effects of exercise on dose and dose distribution of inhaled
automotive pollutants. (Research Report no. 45). Cambridge, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=45
Kloek. J; Mortaz. E; van Ark. I; Lilly. CM; Nijkamp. FP; Folkerts. G. (2010). Glutathione prevents the
early asthmatic reaction and airway hyperresponsiveness in guinea pigs. J Physiol Pharmacol 61: 67-72.
Knuckles. TL; Buntz. JG: Paffett. M; Channell. M; Harmon. M; Cherng. T; Lucas. SN; Mcdonald. JD;
Kanagv. NL; Campen. MJ. (2011). Formation of vascular S-nitrosothiols and plasma nitrates/nitrites
following inhalation of diesel emissions. J Toxicol Environ Health A 74: 828-837.
http://dx.doi.org/10.1080/15287394.2011.570225
Kobavashi. T; Miura. T. (1995). Concentration- and time-dependent increase in specific airway resistance
after induction of airway hyperresponsiveness by subchronic exposure of guinea pigs to nitrogen
dioxide. Fundam Appl Toxicol 25: 154-158. http://dx.doi.org/10.1006/faat. 1995.1049
Kobavashi. T; Shinozaki. Y. (1990). Effect of subacute exposure to nitrogen dioxide on the airway
responsiveness of guinea pig. Agents Actions Suppl 31: 71-74.
Koehler. C: Ginzkev. C: Friehs. G: Hackenberg. S: Froelich. K; Scherzed. A; Burghartz. M; Kessler. M;
Kleinsasser. N. (2010). Aspects of nitrogen dioxide toxicity in environmental urban concentrations in
human nasal epithelium. Toxicol Appl Pharmacol 245: 219-225.
http://dx.doi.0rg/10.1016/i.taap.2010.03.003
Koehler. C: Ginzkev. C: Friehs. G: Hackenberg. S: Froelich. K; Scherzed. A; Burghartz. M; Kessler. M;
Kleinsasser. N. (2011). Ex vivo toxicity of nitrogen dioxide in human nasal epithelium at the WHO
defined 1-h limit value. Toxicol Lett 207: 89-95. http://dx.doi.Org/10.1016/i.toxlet.2011.08.004
Krishna. MT; Holgate. ST. (1999). Inflammatory mechanisms underlying potentiation of effects of inhaled
aeroallergens in response to nitrogen dioxide in allergic airways disease [Review]. Clin Exp Allergy 29:
150-154. http://dx.doi.org/10.1046/i. 1365-2222.1999.00512.x
Kudo. M: Khalifeh Soltani. SM: Sakuma. SA: McKlerov. W: Lee. TH: Woodruff. PG: Lee. JW: Huang. K:
Chen. C: Arjomandi. M; Huang. X; Atabai. K. (2013). Mfge8 suppresses airway hyperresponsiveness in
asthma by regulating smooth muscle contraction. PNAS 110: 660-665.
http://dx.doi.org/10.1073/pnas.1216673110
Lee. BJ; Jeung. YJ; Lee. JY; Choi. DC. (20lib). Increased S-nitrosothiol levels in nonasthmatic
eosinophilic bronchitis compared with cough variant asthma. Int Arch Allergy Immunol 156: 99-103.
http://dx.doi.org/10.1159/000321919
November 2013 3 -65 DRAFT: Do Not Cite or Quote
-------
Lehnert. BE; Archuleta. DC: Ellis. T; Session. WS: Lehnert. NM; Gurlev. LR; Stavert. DM. (1994). Lung
injury following exposure of rats to relatively high mass concentrations of nitrogen dioxide. Toxicology
89: 239-277. http://dx.doi.org/10.1016/0300-483X(94)90100-7
Li. H; Chen. L; Guo. Z; Sang. N; Li. G. (2012a). In vivo screening to determine neurological hazards of
nitrogen dioxide (NO2) using Wistar rats. J Hazard Mater 225-226: 46-53.
http://dx.doi.0rg/10.1016/i.jhazmat.2012.04.063
Li. H; Han. M; Guo. L; Li. G: Sang. N. (2011 a). Oxidative stress, endothelial dysfunction and
inflammatory response in rat heart to NO inhalation exposure. Chemosphere 82: 1589-1596.
http://dx.doi.0rg/10.1016/i.chemosphere.2010.ll.055
Lima. ES: Bonini. MG: Augusto. O: Barbeiro. HV; Souza. HP; Abdalla. PS. (2005). Nitrated lipids
decompose to nitric oxide and lipid radicals and cause vasorelaxation. Free Radic Biol Med 39: 532-
539. http://dx.doi.0rg/10.1016/j.freeradbiomed.2005.04.005
Linhares. D; Jacinto. T; Pereira. AM; Fonseca. JA. (2011). Effects of atopy and rhinitis on exhaled nitric
oxide values - a systematic review. Clin Transl Allergy 1:8. http://dx.doi.org/10.1186/2045-7022-l-8
Linn. WS: Solomon. JC: Trim. SC: Spier. CE; Shamoo. DA; Venet. TG: Avol. EL; Hackney. JD. (1985b).
Effects of exposure to 4 ppm nitrogen dioxide in healthy and asthmatic volunteers. Arch Environ Occup
Health 40: 234-239.
Lovich. MA: Bruno. NK: Plant. CP: Wei. AE: Vasauez. GB: Johnson. BJ: Fine. DH: Gilbert. RJ. (2011).
Use of ultra pure nitric oxide generated by the reduction of nitrogen dioxide to reverse pulmonary
hypertension in hypoxemic swine. Nitric Oxide 24: 204-212.
http://dx.doi.0rg/10.1016/i.niox.2011.04.006
Lundberg. JO; Carlstrom. M; Larsen. FJ; Weitzberg. E. (2011). Roles of dietary inorganic nitrate in
cardiovascular health and disease [Review]. Cardiovasc Res 89: 525-532.
http://dx.doi.org/10.1093/cvr/cvq325
Lundberg. JON; Farkas-Szallasi. T; Weitzberg. E; Rinder. J; Lidholm. J; Anggaard. A; Hokfelt. T;
Lundberg. JM; Alving. K. (1995). High nitric oxide production in human paranasal sinuses. Nat Med 1:
370-373. http://dx.doi.org/10.1038/nm0495-370
Lundberg. JON; Settergren. G: Gelinder. S: Lundberg. JM; Alving. K; Weitzberg. E. (1996). Inhalation of
nasally derived nitric oxide modulates pulmonary function in humans. Acta Physiol Scand 158: 343-
347. http://dx.doi.Org/10.1046/i.1365-201X.1996.557321000.x
Maarsingh. H; Bossenga. BE; Bos. IS: Volders. HH; Zaagsma. J: Meurs. H. (2009). L-arginine deficiency
causes airway hyperresponsiveness after the late asthmatic reaction. Eur Respir J 34: 191-199.
http://dx.doi.org/10.1183/09031936.00105408
Maarsingh. H; Zuidhof. AB; Bos. IS: van Duin. M; Boucher. JL; Zaagsma. J: Meurs. H. (2008). Arginase
inhibition protects against allergen-induced airway obstruction, hyperresponsiveness, and inflammation.
Am J Respir Crit Care Med 178: 565-573. http://dx.doi.org/10.1164/rccm.200710-1588OC
Mabalirajan. U; Ahmad. T; Leishangthem. GD; Dinda. AK; Agrawal. A; Ghosh. B. (2010a). L-arginine
reduces mitochondria! dysfunction and airway injury in murine allergic airway inflammation. Int
Immunopharmacol 10: 1514-1519. http://dx.doi.Org/10.1016/i.intimp.2010.08.025
Mabalirajan. U; Ahmad. T; Leishangthem. GD: Joseph. DA; Dinda. AK; Agrawal. A; Ghosh. B. (201 Ob).
Beneficial effects of high dose of L-arginine on airway hyperresponsiveness and airway inflammation in
a murine model of asthma. J Allergy Clin Immunol 125: 626-635.
http://dx.doi.0rg/10.1016/i.jaci.2009.10.065
Madej. E; Folkes. LK; Wardman. P; Czapski. G: Goldstein. S. (2008). Thiyl radicals react with nitric oxide
to form S-nitrosothiols with rate constants near the diffusion-controlled limit. Free Radic Biol Med 44:
2013-2018. http://dx.doi.0rg/10.1016/i.freeradbiomed.2008.02.015
Magnussen. H; Reuss. G: Jorres. R. (1986). Airway response to methacholine during exercise induced
refractoriness in asthma. Thorax 41: 667-670.
November 2013 3 -66 DRAFT: Do Not Cite or Quote
-------
Maher. AR; Milsom. AB; Gunaruwan. P; Abozguia. K; Ahmed. I; Weaver. RA; Thomas. P; Ashrafian. H;
Born. GVR; James. PE; Frenneaux. MR (2008). Hypoxic modulation of exogenous nitrite-induced
vasodilation in humans. Circulation 117: 670-677.
http://dx.doi.org/10.1161/CIRCULATIONAHA.107.719591
Major. TA; Panmanee. W: Mortensen. JE; Gray. LD; Hoglen. N; Hassett. DJ. (2010). Sodium nitrite-
mediated killing of the major cystic fibrosis pathogens Pseudomonas aeruginosa, Staphylococcus
aureus, and Burkholderia cepacia under anaerobic planktonic and biofilm conditions. Antimicrob
Agents Chemother 54: 4671-4677. http://dx.doi.org/10.1128/AAC.00379-10
Malinovschi. A; Janson. C: Holm. L; Nordvall. L; Alving. K. (2009). Basal and induced NO formation in
the pharyngo-oral tract influences estimates of alveolar NO levels. J Appl Physiol 106: 513-519.
http://dx.doi.org/10.1152/iapplphvsiol.91148.2008
Malm-Erjefalt. M; Persson. CG: Erjefalt JS. (2001). Degranulation status of airway tissue eosinophils in
mouse models of allergic airway inflammation. Am J Respir Cell Mol Biol 24: 352-359.
http://dx.doi.0rg/10.1165/ajrcmb.24.3.4357
Marshall. HE; Potts. EN; Kelleher. ZT; Stamler. JS: Foster. WM; Auten. RL. (2009). Protection from
lipopolysaccharide-induced lung injury by augmentation of airway S-nitrosothiols. Am J Respir Crit
Care Med 180: 11-18. http://dx.doi.org/10.1164/rccm.200807-1186OC
Marshall. HE; Stamler. JS. (2001). Inhibition of NF-kappa B by S-nitrosylation. Biochemistry 40: 1688-
1693. http://dx.doi.org/10.1021M002239y
Martin. RA; Ather. JL; Lundblad. LKA; Suratt. BT; Boyson. JE; Budd. RC: Alcorn. JF; Flavell. RA;
Eisenbarth. SC: Poynter. ME. (2013). Interleukin-1 receptor and caspase-1 are required for the Thl7
response in NO2-promoted allergic airway disease. Am J Respir Cell Mol Biol 48: 655-664.
http://dx.doi.org/10.1165/rcmb.2012-0423OC
Matsumura. Y. (1970). The effects of ozone, nitrogen dioxide, and sulfur dioxide on the experimentally
induced allergic respiratory disorder in guinea pigs: II. The effects of ozone on the absorption and the
retention of antigen in the lung. Am Rev Respir Dis 102: 438-443.
Mauderly. JL; Cheng. YS: Gillett. NA; Griffith. WC: Henderson. RF; Pickrell. JA; Ulblff. RK. (1990).
Influence of preexisting pulmonary emphysema on susceptibility of rats to chronic inhalation exposure
to nitrogen dioxide. Inhal Toxicol 2: 129-150. http://dx.doi.org/10.3109/08958379009145250
McGrath. JJ; Smith. PL. (1984). Respiratory responses to nitrogen dioxide inhalation: Subtitle: Nitrogen
dioxide. J Environ Sci Health, Part A: Environ Sci Eng 19: 417-431.
http://dx.doi.org/10.1080/10934528409375168
Mercer. RR; Costa. PL; Crapo. JD. (1995). Effects of prolonged exposure to low doses of nitric oxide or
nitrogen dioxide on the alveolar septa of the adult rat lung. Lab Invest 73: 20-28.
Mercer. RR; Russell. ML; Roggli. VL; Crapo. JD. (1994). Cell number and distribution in human and rat
airways. Am J Respir Cell Mol Biol 10: 613-624.
Miller. FJ; Graham. JA; Raub. JA; Illing. JW: Menache. MG: House. DE; Gardner. DE. (1987). Evaluating
the toxicity of urban patterns of oxidant gases. II. Effects in mice from chronic exposure to nitrogen
dioxide. J Toxicol Environ Health 21: 99-112. http://dx.doi.org/10.1080/15287398709531005
Miller. FJ: Overton. JH; Myers. ET; Graham. JA. (1982). Pulmonary dosimetry of nitrogen dioxide in
animals and man. In L ed RandGrant (Ed.), Air pollution by nitrogen oxides (pp. 377-386). Amsterdam:
Elsevier Scientific Publishing Company.
Miyanishi. K; Kinouchi. T; Kataoka. K; Kanoh. T; Ohnishi. Y. (1996). In vivo formation of mutagens by
intraperitoneal administration of polycyclic aromatic hydrocarbons in animals during exposure to
nitrogen dioxide. Carcinogenesis 17: 1483-1490.
Mohsenin. V. (1987b). Effect of vitamin C on NO2-induced airway hyperresponsiveness in normal
subjects: Arandomized double-blind experiment. Am Rev Respir Dis 136: 1408-1411.
http://dx.doi.0rg/10.1164/airccm/136.6.1408
November 2013 3 -67 DRAFT: Do Not Cite or Quote
-------
Mohsenin. V. (1991). Lipid peroxidation and antielastase activity in the lung under oxidant stress: Role of
antioxidant defenses. J Appl Physiol 70: 1456-1462.
Miiller. B; Schafer. H; Barth. P; von Wichert. P. (1994). Lung surfactant components in bronchoalveolar
lavage after inhalation of NO2 as markers of altered surfactant metabolism. Lung 172: 61-72.
http://dx.doi.org/10.1007/BF00185077
Murphy. SD; Ulrich. CE; Frankowitz. SH; Xintaras. C. (1964). Altered function in animals inhaling low
concentrations of ozone and nitrogen dioxide. Am Ind Hyg Assoc J 25: 246-253.
http://dx.doi.org/10.1080/00028896409342583
Neta. P; Huie. RE; Ross. AB. (1988). Rate constants for reactions of inorganic radicals in aqueous solution.
J Phys Chem Ref Data 17: 1027-1284. http://dx.doi.Org/10.1063/l.555808
Nettesheim. P; Hanna. MG. Jr; Dohertv. DG: Newell. RF; Hellman. A. (1970). Effect of chronic exposure
to air pollutants on the respiratory tracts of mice: Histopathological findings. In P Nettesheim; MG
Hanna, Jr.; JW Deatherage, Jr. (Eds.), Morphology of experimental respiratory carcinogenesis:
proceedings of a biology division, Oak Ridge National Laboratory conference; May; Gatlinburg, TN
(pp. 19701970). Oak Ridge, TN: U.S. Atomic Energy C.
Noda. N; Takeno. S; Fukuiri. T; Hirakawa. K. (2012). Monitoring of oral and nasal exhaled nitric oxide in
eosinophilic chronic rhinosinusitis: A prospective study. American Journal of Rhinology & Allergy 26:
255-259. http://dx.doi.org/10.2500/ajra.2012.26.3772
North. ML; Khanna. N; Marsden. PA; Grasemann. H; Scott. JA. (2009). Functionally important role for
arginase 1 in the airway hyperresponsiveness of asthma. Am J Physiol Lung Cell Mol Physiol 296:
L911-L920. http://dx.doi.org/10.1152/aiplung.00025.2009
O'Byrne. PM; Gauvreau. GM; Brannan. JD. (2009). Provoked models of asthma: What have we learnt?
[Review]. Clin Exp Allergy 39: 181-192. http://dx.doi.org/10.1111/i. 1365-2222.2008.03172.x
O'Donnell. VB; Eiserich. JP; Chumlev. PH; Jablonsky. MJ; Krishna. NR; Kirk. M; Barnes. S; Darlev-
Usmar. VM; Freeman. BA. (1999). Nitration of unsaturated fatty acids by nitric oxide-derived reactive
nitrogen species peroxynitrite, nitrous acid, nitrogen dioxide, and nitronium ion. Chem Res Toxicol 12:
83-92. http://dx.doi.org/10.1021/tx980207u
Oda. H; Tsubone. H; Suzuki. A; Ichinose. T; Kubota. K. (1981). Alterations of nitrite and nitrate
concentrations in the blood of mice exposed to nitrogen dioxide. Environ Res 25: 294-301.
Ohashi. Y; Nakai. Y; Sugiura. Y; Ohno. Y; Okamoto. H; Tanaka. A; Kakinoki. Y; Havashi. M. (1994).
Nitrogen dioxide-induced eosinophilia and mucosal injury in the nose of the guinea pig. Acta
Otolaryngol 114: 547-551.
Olin. AC; Aldenbratt. A; Ekman. A; Ljungkvist. G; Jungersten. L; Alving. K; Toren. K. (2001). Increased
nitric oxide in exhaled air after intake of a nitrate-rich meal. Respir Med 95: 153-158.
http://dx.doi.org/10.1053/rmed.2000.1010
Olin. AC; Bake. B; Toren. K. (2007). Fraction of exhaled nitric oxide at 50 mL/s: Reference values for
adult lifelong never-smokers. Chest 131: 1852-1856. http://dx.doi.org/10.1378/chest.06-2928
Olin. AC; Hellgren. J; Karlsson. G; Ljungkvist. G; Nolkrantz. K; Toren. K. (1998). Nasal nitric oxide and
its relationship to nasal symptoms, smoking and nasal nitrate. Rhinology 36: 117-121.
Overton. JH. Jr. (1984). Physicochemical processes and the formulation of dosimetry models. In FJ Miller;
DB Menzel (Eds.), Fundamentals of extrapolation modeling of inhaled toxicants: Ozone and nitrogen
dioxide (pp. 93-114). Washington, DC: Hemisphere Publishing Corporation.
Pacher. P; Beckman. JS; Liaudet. L. (2007). Nitric oxide and peroxynitrite in health and disease [Review].
Physiol Rev 87: 315-424. http://dx.doi.org/10.1152/phvsrev.00029.2006
Pagani. P; Romano. M; Erroi. A; Ferro. M; Salmona. M. (1994). Biochemical effects of acute and subacute
nitrogen dioxide exposure in rat lung and bronchoalveolar lavage fluid. Arch Environ Contam Toxicol
27: 426-430. http://dx.doi.org/10.1007/BF00213182
November 2013 3 -68 DRAFT: Do Not Cite or Quote
-------
Patfamanathan. S: Krishna. MT; Blomberg. A; Helledav. R; Kelly. FJ; Sandstrom. T; Holgate. ST; Wilson.
SJ: Frew. AJ. (2003). Repeated daily exposure to 2 ppm nitrogen dioxide upregulates the expression of
IL-5, IL-10, IL-13, and ICAM-1 in the bronchial epithelium of healthy human airways. Occup Environ
Med 60: 892-896. http://dx.doi.org/10.1136/oem.60.ll.892
Polidori. A; Fine. PM. (2012). Ambient concentrations of criteria and air toxic pollutants in close proximity
to a freeway with heavy-duty diesel traffic. Final report. South Coast Air Quality Management District.
http ://www. aqmd.gov/tao/AQ-Reports/I71 OFwv Studv.pdf
Posin. C: Clark. K; Jones. MP; Patterson. JV: Buckley. RD; Hackney. JD. (1978). Nitrogen dioxide
inhalation and human blood biochemistry. Arch Environ Occup Health 33: 318-324.
Postlethwait. EM; Bidani. A. (1989). Pulmonary disposition of inhaled NO2-nitrogen in isolated rat lungs.
Toxicol Appl Pharmacol 98: 303-312.
Postlethwait. EM; Bidani. A. (1990). Reactive uptake governs the pulmonary air space removal of inhaled
nitrogen dioxide. J Appl Physiol 68: 594-603.
Postlethwait. EM; Bidani. A. (1994). Mechanisms of pulmonary NO2 absorption [Review]. Toxicology 89:
217-237.
Postlethwait. EM; Bioani. A; Evans. MJ. (1990). The effect of NO2 exposure on perfusate distribution in
isolated rat lungs: pulmonary versus bronchial circulation. Toxicol Appl Pharmacol 106: 456-461.
http://dx.doi.org/10.1016/0041-008X(90)90340-Z
Postlethwait. EM; Langford. SD: Bidani. A. (1991a). Interfacial transfer kinetics of NO2 into pulmonary
epithelial lining fluid. J Appl Physiol 71: 1502-1510.
Postlethwait. EM; Langford. SD: Bidani. A. (1991b). Transfer of NO2 through pulmonary epithelial lining
fluid. Toxicol Appl Pharmacol 109: 464-471.
Postlethwait. EM; Langford. SD: Bidani. A. (1992). Kinetics of NO2 air space absorption in isolated rat
lungs. JAppl Physiol 73: 1939-1945.
Postlethwait. EM; Langford. SD: Jacobson. LM; Bidani. A. (1995). NO2 reactive absorption substrates in
rat pulmonary surface lining fluids. Free Radic Biol Med 19: 553-563. http://dx.doi.org/10.1016/0891-
5849(95)00058-6
Postlethwait. EM; Mustafa. MG. (1981). Fate of inhaled nitrogen dioxide in isolated perfused rat lung. J
Toxicol Environ Health 7: 861-872. http://dx.doi.org/10.1080/15287398109530029
Postlethwait. EM; Mustafa. MG. (1989). Effect of altered dose rate on NO2 uptake and transformation in
isolated lungs. J Toxicol Environ Health 26: 497-507. http://dx.doi.org/10.1080/15287398909531272
Povnter. ME; Persinger. RL; Irvin. CG: Butnor. KJ; van Hirtum. H; Slav. W: Heintz. NH; Robbins. J:
Hemenwav. D; Taatjes. DJ: Janssen-Heininger. Y. (2006). Nitrogen dioxide enhances allergic airway
inflammation and hyperresponsiveness in the mouse. Am J Physiol Lung Cell Mol Physiol 290: LI 44-
L152. http://dx.doi.org/10.1152/ajplung.00131.2005
Proust. B; Lacroix. G: Robidel. F; Marliere. M; Lecomte. A; Vargaftig. BB. (2002). Interference of a short-
term exposure to nitrogen dioxide with allergic airways responses to allergenic challenges in BALB/c
mice. Mediators Inflamm 11: 251-260. http://dx.doi.org/10.1080/096293502900000113
Priitz. WA; Monig. H; Butler. J: Land. EJ. (1985). Reactions of nitrogen dioxide in aqueous model systems:
Oxidation of tyrosine units in peptides and proteins. Arch Biochem Biophys 243: 125-134.
http://dx.doi.org/10.1016/0003-9861(85)90780-5
Pryor. WA. (1992). How far does ozone penetrate into the pulmonary air/tissue boundary before it reacts?
Free Radic Biol Med 12: 83-88. http://dx.doi.org/10.1016/0891-5849(92)90060-T
Prvor. WA: Houk. KN; Foote. CS: Fukuto. JM; Ignarro. LJ; Squadrito. GL; Davies. KJA. (2006). Free
radical biology and medicine: It's a gas, man! [Review]. Am J Physiol Regul Integr Comp Physiol 291:
R491-R511. http://dx.doi.org/10.1152/aipregu.00614.2005
November 2013 3 -69 DRAFT: Do Not Cite or Quote
-------
Pryor. WA; Lightsev. JW. (1981). Mechanisms of nitrogen dioxide reactions: Initiation of lipid peroxidation
and the production of nitrous acid. Science 214: 435-437.
http://dx.doi.org/10.1126/science.214.4519.435
Que. LG: Yang. Z; Stamler. JS: Lugogo. NL; Kraft. M. (2009). S-nitrosoglutathione reductase: An
important regulator in human asthma. Am J Respir Crit Care Med 180: 226-231.
http://dx.doi.org/10.1164/rccm.200901-0158OC
Rava. M; Varraso. R; Decoster. B; Huyvaert. H; Le Moual. N; Jacquemin. B; Kiinzli. N; Kauffmann. F;
Zerimech. F; Matran. R; Nadif. R. (2012). Plasma and exhaled breath condensate nitrite-nitrate level in
relation to environmental exposures in adults in the EGEA study. Nitric Oxide 27: 169-175.
httrj://dx.doi.org/10.1016/i.niox.2012.06.007
Robison. TW: Forman. HJ. (1993). Dual effect of nitrogen dioxide on rat alveolar macrophage arachidonate
metabolism. Exp Lung Res 19: 21-36.
Robison. TW: Murphy. JK; Beyer. LL; Richters. A; Forman. HJ. (1993). Depression of stimulated
arachidonate metabolism and superoxide production in rat alveolar macrophages following in vivo
exposure to 0.5 ppm NO2. J Toxicol Environ Health 38: 273-292.
http://dx.doi.org/10.1080/15287399309531718
Rose. RM; Pinkston. P; Skornik. WA. (1989a). Altered susceptibility to viral respiratory infection during
short-term exposure to nitrogen dioxide. Res Rep Health Eff Instl-24.
Rubenchik. BL; Glavin. AA; Galenko. PM; Kilkichko. AA; Olenick. IP; Artemov. KV. (1995). Gaseous
nitrogen dioxide increases the endogenous synthesis of carcinogenic N-nitrosodimethylamine in
animals. J Environ Pathol Toxicol Oncol 14: 111-115.
Rubinstein. I; Reiss. TF; Bigbv. BG: Stites. DP; Boushev. HA. Jr. (1991). Effects of 0.60 PPM nitrogen
dioxide on circulating and bronchoalveolar lavage lymphocyte phenotypes in healthy subjects. Environ
Res 55: 18-30.
Rudolph. V; Rudolph. TK; Schopfer. FJ; Bonacci. G: Woodcock. SR; Cole. MP; Baker. PR; Ramani. R;
Freeman. BA. (2010). Endogenous generation and protective effects of nitro-fatty acids in amurine
model of focal cardiac ischaemia and reperfusion. Cardiovasc Res 85: 155-166.
http://dx.doi.org/10.1093/cvr/cvp275
Sandstrom. T; Andersson. MC: Kolmodin-Hedman. B; Stjernberg. N; Angstrom. T. (1990).
Bronchoalveolar mastocytosis and lymphocytosis after nitrogen dioxide exposure in man: a time-kinetic
study. Eur Respir J 3: 138-143.
Sandstrom. T; Helledav. R: Bjermer. L; Stjernberg. N. (1992). Effects of repeated exposure to 4 ppm
nitrogen dioxide on bronchoalveolar lymphocyte subsets and macrophages in healthy men. Eur Respir J
5: 1092-1096.
Sandstrom. T; Ledin. MC: Thomasson. L; Helledav. R: Stjernberg. N. (1992). Reductions in lymphocyte
subpopulations after repeated exposure to 1.5 ppm nitrogen dioxide. Br J Ind Med 49: 850-854.
Sandstrom. T; Stjernberg. N; Eklund. A; Ledin. MC: Bjermer. L; Kolmodin-Hedman. B; Lindstrom. K;
Rosenhall. L; Angstrom. T. (1991). Inflammatory cell response in bronchoalveolar lavage fluid after
nitrogen dioxide exposure of healthy subjects: A dose-response study. Eur Respir J 4: 332-339.
Sarangapani. R; Gentry. PR; Covington. TR; Teeguarden. JG: Clewell. HJ. III. (2003). Evaluation of the
potential impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of
vapors. Inhal Toxicol 15: 987-1016. http://dx.doi.org/10.1080/08958370390226350
Schlesinger. RB; Driscoll. KE; Gunnison. AF; Zelikoff. JT. (1990). Pulmonary arachidonic acid metabolism
following acute exposures to ozone and nitrogen dioxide. J Toxicol Environ Health 31: 275-290.
http://dx.doi.org/10.1080/15287399009531456
Schlesinger. RB; Driscoll. KE; Vollmuth. TA. (1987). Effect of repeated exposures to nitrogen dioxide and
sulfuric acid mist alone or in combination on mucociliary clearance from the lungs of rabbits. Environ
Res 44: 294-301. http://dx.doi.org/10.1016/80013-9351(87)80238-4
November 2013 3 -70 DRAFT: Do Not Cite or Quote
-------
Schlesinger. RB; Gearhart. JM. (1987). Intermittent exposures to mixed atmospheres of nitrogen dioxide
and sulfuric acid: effect on particle clearance from the respiratory region of rabbit lungs. Toxicology 44:
309-319. http://dx.doi.org/10.1016/0300-483X(87)90032-l
Schwartz. SE: White. WH. (1983). Kinetics of reactive dissolution of nitrogen oxides into aqueous solution.
In SE Schwartz (Ed.), Trace atmospheric constituents: Properties, transformations, and fates (pp. 1-116).
New York, NY: Wiley.
See. KG: Christiani. DC. (2013). Normal values and thresholds for the clinical interpretation of exhaled
nitric oxide levels in the US general population: Results from the National Health and Nutrition
Examination Survey 2007-2010. Chest 143: 107-116. http://dx.doi.org/10.1378/chest.12-0416
Sevanian. A; Elsaved. N; Hacker. AD. (1982a). Effects of vitamin E deficiency and nitrogen dioxide
exposure on lung lipid peroxidation: Use of lipid epoxides and malonaldehyde as measures of
peroxidation. J Toxicol Environ Health 10: 743-756. http://dx.doi.org/10.1080/15287398209530292
Sherwin. RP; Carlson. DA. (1973). Protein content of lung lavage fluid of guinea pigs exposed to 0.4 ppm
nitrogen dioxide. Arch Environ Health 27: 90-93.
Sies. H. (2004). Ozone in arteriosclerotic plaques: searching for the "smoking gun". Angew
Chem Int Ed Engl 43: 3514-3515. http://dx.doi.org/10.1002/anie.200460118
Silbaugh. SA; Mauderly. JL; Macken. CA. (1981). Effects of sulfuric acid and nitrogen dioxide on airway
responsiveness of the guinea pig. J Toxicol Environ Health 8: 31-45.
http://dx.doi.org/10.1080/15287398109530048
Smith. LL. (2004). Oxygen, oxysterols, ouabain, and ozone: A cautionary tale [Review]. Free Radic Biol
Med 37: 318-324. http://dx.doi.Org/10.1016/i.freeradbiomed.2004.04.024
Squadrito. GL; Postlethwait. EM. (2009). On the hydrophobicity of nitrogen dioxide: could there be a
"lens" effect for NO(2) reaction kinetics. Nitric Oxide 21: 104-109.
http://dx.doi.0rg/10.1016/i.niox.2009.05.008
Squadrito. GL; Postlethwait. EM; Matalon. S. (2010). Elucidating mechanisms of chlorine toxicity:
Reaction kinetics, thermodynamics, and physiological implications [Review]. Am J Physiol Lung Cell
Mol Physiol 299: L289-L300. http://dx.doi.org/10.1152/aiplung.00077.2010
Stenger. MR; Rose. MJ; Joshi. MS: Rogers. LK; Chicoine. LG: Bauer. JA; Nelin. LD. (2010). Inhaled nitric
oxide prevents 3-nitrotyrosine formation in the lungs of neonatal mice exposed to >95% oxygen.
Lung 188: 217-227. http://dx.doi.org/10.1007/s00408-010-9235-6
Strand. V: Svartengren. M; Rak. S: Barck. C: Bylin. G. (1998). Repeated exposure to an ambient level of
NO2 enhances asthmatic response to a nonsymptomatic allergen dose. Eur Respir J 12: 6-12.
http://dx.doi.org/10.1183/09031936.98.12010006
Suzuki. AK; Tsubone. H; Ichinose. T; Oda. H; Kubota. K. (1981). [Effects of subchronic nitrogen dioxide
exposure on arterial blood pHa, PaCO2 and PaO2 in rats]. Nippon Eiseigaku Zasshi 36: 816-823.
Suzuki. AK; Tsubone. H; Kubota. K. (1982). Changes of gaseous exchange in the lung of mice acutely
exposed to nitrogen dioxide. Toxicol Lett 10: 327-335. http://dx.doi.org/10.1016/0378-4274(82)90226-0
Suzuki. T; Ikeda. S: Kanoh. T; Mizoguchi. I. (1986). Decreased phagocytosis and superoxide anion
production in alveolar macrophages of rats exposed to nitrogen dioxide. Arch Environ Contam Toxicol
15: 733-739. http://dx.doi.org/10.1007/BF01054920
Taylor. DR. (2012). Advances in the clinical applications of exhaled nitric oxide measurements. J Breath
Res 6: 047102. http://dx.doi.Org/10.1088/1752-7155/6/4/047102
Thomas. HV; Mueller. PK; Wright. R. (1967). Response of rat lung mast cells to nitrogen dioxide
inhalation. J Air Pollut Control Assoc 17: 33-35. http://dx.doi.org/10.1080/00022470.1967.10468939
Torok. JA; Brahmaiothi. MV; Zhu. H; Tinch. BT; Auten. RL; Mcmahon. TJ. (2012). Transpulmonary flux
of S-nitrosothiols and pulmonary vasodilation during nitric oxide inhalation: role of transport. Am J
Respir Cell Mol Biol 47: 37-43. http://dx.doi.org/10.1165/rcmb.2011-0439OC
November 2013 3-71 DRAFT: Do Not Cite or Quote
-------
Tsubone. H; Oda. H; Suzuki. AK; Kubota. K. (1982). Electrocardiographic abnormalities in rats by acute
exposure to nitrogen dioxide. Toxicol Lett 12: 125-129. http://dx.doi.org/10.1016/0378-4274(82)90174-
6
Tunnicliffe. WS: Surge. PS: Ayres. JG. (1994). Effect of domestic concentrations of nitrogen dioxide on
airway responses to inhaled allergen in asthmatic patients. Lancet 344: 1733-1736.
http://dx.doi.org/10.1016/s0140-6736(94)92886-x
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park,NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment [EPA
Report]. (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum, http://www.epa.gov/cancerguidelines/
U.S. EPA (U.S. Environmental Protection Agency). (2008a). Integrated science assessment for oxides of
nitrogen - Health criteria: Annexes [EPA Report]. (EPA/600/R-08/072). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for ozone and
related photochemical oxidants (final report) [EPA Report]. (EPA/600/R-10/076F). Washington, DC.
http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=247492
van der Heijden. HHA. CM; Brouwer. ML; Hoekstra. F; van der Pol. P; Merkus. PJF. M. (In Press)
Reference values of exhaled nitric oxide in healthy children 1-5 years using off-line tidal breathing.
Pediatr Pulmonol. http://dx.doi.org/10.1002/ppul.22796\
Van der Vliet A: O'Neill. C: Cross. C: Koostra. J: Volz. W: Halliwell B: Louie. S. (1999). Determination of
low-molecular-mass antioxidant concentrations in human respiratory tract lining fluids. American Journal of
Physiology 276: L289-L296.
Van Stee. EW: Sloane. RA; Simmons. JE; Brunnemann. KD. (1983). In vivo formation of N-
nitrosomorpholine in CD-I mice exposed by inhalation to nitrogen dioxide and by gavage to
morpholine. JNatl Cancer Inst 70: 375-379.
Van Stee. EW: Sloane. RA; Simmons. JE; Moorman. MP; Brunnemann. KD. (1995). Endogenous
formation of N-nitrosomorpholine in mice from 15NO2 by inhalation and morpholine by gavage.
Carcinogenesis 16: 89-92.
Velsor. LW: Ballinger. CA; Patel. J: Postlethwait. EM. (2003). Influence of epithelial lining fluid lipids on
NO2-induced membrane oxidation and nitration. Free Radic Biol Med 34: 720-733.
http://dx.doi.org/10.1016/S0891-5849(02)01370-9
Velsor. LW: Postlethwait. EM. (1997). NO2-induced generation of extracellular reactive oxygen is
mediated by epithelial lining layer antioxidants. Am J Physiol 17: L1265-L1275.
Vints. AM; Oostveen. E; Eeckhaut. G: Smolders. M; De Backer. WA. (2005). Time-dependent effect of
nitrate-rich meals on exhaled nitric oxide in healthy subjects. Chest 128: 2465-2470.
http://dx.doi.org/10.1378/chest. 128.4.2465
von Nieding. G: Wagner. HM. (1979). Effects of NO2 on chronic bronchitics. Environ Health Perspect 29:
137-142.
Wagner. HM. (1970). Absorption of NO and NO2 in MIK- and MAK-concentrations during inhalation.
[Absorption von NO und NO2 in MIK- und MAK-konzentrationen bei der inhalation]. Staub,
Reinhaltung der Luft 30: 380-381.
Wang. J; Widlansky. ME. (2009). Lifestyle choices and endothelial function: Risk and relevance [Review].
Curr Vase Pharmacol 7: 209-224. http://dx.doi.org/10.2174/157016109787455581
November 2013 3 -72 DRAFT: Do Not Cite or Quote
-------
Wang. JH; Devalia. JL; Puddle. JM; Hamilton. SA; Davies. RJ. (1995 a). Effect of six-hour exposure to
nitrogen dioxide on early-phase nasal response to allergen challenge in patients with a history of
seasonal allergic rhinitis. J Allergy Clin Immunol 96: 669-676.
Wang. JH; Puddle. J: Devalia. JL; Davies. RJ. (1995b). Nitrogen dioxide increases eosinophil activation in
the early-phase response to nasal allergen provocation. Int Arch Allergy Immunol 107: 103-105.
Wardman. P. (1998). Nitrogen dioxide in biology: Correlating chemical kinetics with biological effects. In
ZB Alfassi (Ed.), N-centered radicals (pp. 155-179). Chichester, UK: Wiley.
Weitzberg. E; Hezel. M; Lundberg. JO. (2010). Nitrate-nitrite-nitric oxide pathway: implications for
anesthesiology and intensive care [Review]. Anesthesiology 113: 1460-1475.
http://dx.doi.org/10.1097/ALN.Ob013e3181fcf3cc
Weitzberg. E; Lundberg. JO. (2013). Novel aspects of dietary nitrate and human health. Annu Rev Nutr 33:
129-159. http://dx.doi.org/10.1146/annurev-nutr-071812-161159
Wennmalm. A; Benthin. G: Edlund. A; Jungersten. L; Kieler-Jensen. N; Lundin. S: Westfelt. UN;
Petersson. AS: Waagstein. F. (1993). Metabolism and excretion of nitric oxide in humans. An
experimental and clinical study. Circ Res 73: 1121-1127. http://dx.doi.Org/10.1161/01.RES.73.6.1121
Witten. A; Solomon. C: Abbritti. E; Arjomandi. M; Zhai. W: Kleinman. M; Balmes. J. (2005). Effects of
nitrogen dioxide on allergic airway responses in subjects with asthma. J Occup Environ Med 47: 1250-
1259.
Yadav. AK; Doran. SF; Samal. AA; Sharma. R; Vedagiri. K; Postlethwait EM; Squadrito. GL; Fanucchi.
MV; Roberts. LJ; Patel. RP: Matalon. S. (2011). Mitigation of chlorine gas lung injury in rats by post-
exposure administration of sodium nitrite. Am J Physiol Lung Cell Mol Physiol 300: L362-L369.
http://dx.doi.org/10.1152/ajplung.00278.2010
Yokoyama. E. (1968). Uptake of SO2 and NO2 by the isolated upper airways. Koshu Eisei Kenkyu 17:
302-306.
Zetterquist. W: Pedroletti. C: Lundberg. JO; Alving. K. (1999). Salivary contribution to exhaled nitric
oxide. Eur Respir J 13: 327-333. http://dx.doi.Org/10.1034/i.1399-3003.1999.13bl8.x
Zhang. J; Villacorta. L; Chang. L; Fan. Z; Hamblin. M; Zhu. T; Chen. CS: Cole. MP; Schopfer. FJ; Deng.
CX; Garcia-Barrio. MT; Feng. YH; Freeman. BA; Chen. YE. (2010). Nitro-oleic acid inhibits
angiotensin II-induced hypertension. Circ Res 107: 540-548.
http://dx.doi.org/10.1161/CIRCRESAHA.110.218404
Zhu. Y; Fung. DC: Kennedy. N; Hinds. WC: Eiguren-Fernandez. A. (2008). Measurements of ultrafme
particles and other vehicular pollutants inside a mobile exposure system on Los Angeles freeways. J Air
Waste Manag Assoc 58: 424-434.
November 2013 3 -73 DRAFT: Do Not Cite or Quote
-------
CHAPTER 4 INTEGRATED HEALTH EFFECTS
OF SHORT-TERM EXPOSURE TO OXIDES OF
NITROGEN
4.1 Introduction
1 This chapter summarizes, integrates, and evaluates the evidence for various health effects
2 associated with short-term (i.e., 1 month or less, see Section 1.4) exposure to oxides of
3 nitrogen. The chapter sections comprise evaluations of the epidemiologic, controlled
4 human exposure, and animal toxicological evidence for the effects of short-term exposure
5 to oxides of nitrogen on health outcomes related to respiratory effects (Section 4.2).
6 cardiovascular effects (Section 4.3). as well as total mortality (Section 4.4). Reproductive
7 and developmental effects have also been examined in relation to short-term exposure to
8 oxides of nitrogen. However, this evidence is evaluated with long-term exposure studies
9 in Chapter 5. because associations are often compared among various short- and long-
10 term exposure periods that are difficult to distinguish.
11 Individual sections for major outcome categories (i.e., respiratory, cardiovascular,
12 mortality) begin with a summary of conclusions from the 2008 ISA for Oxides of
13 Nitrogen followed by an evaluation of recent (i.e., published since the completion of the
14 2008 ISA for Oxides of Nitrogen) studies that builds upon evidence from previous
15 reviews. Within each of these sections, results are organized into smaller groups of
16 related endpoints (e.g., airway hyperresponsiveness, pulmonary inflammation, lung
17 function) and then specific scientific discipline (i.e., epidemiology, toxicology).
18 Sections for each of the major outcome categories (i.e., respiratory, cardiovascular, total
19 mortality) conclude with an integrated summary of the assessment of evidence and
20 conclusions regarding causality. A determination of causality was made for each major
21 outcome category by evaluating the evidence for each category independently with the
22 causal framework (described in the Preamble to the ISA). Findings for cause-specific
23 mortality (i.e., respiratory, cardiovascular) are used to assess the continuum of effects and
24 inform the causality determinations for respiratory and cardiovascular effects. The
25 causality determination for total mortality (Section 4.4) is based primarily on the
26 evidence for non-accidental causes of mortality combined, but is also informed by the
27 extent to which evidence for the spectrum of cardiovascular and respiratory effects
28 provides biological plausibility for NO2-related total mortality.
29 Judgments regarding causality were made by evaluating the evidence over the full range
30 of exposure or ambient concentrations in animal toxicological, controlled human
November 2013 4-1 DRAFT: Do Not Cite or Quote
-------
1 exposure, and epidemiologic studies considered relevant to this ISA (i.e., up to 5,000 ppb
2 NO 2 or NO) as described in Section 1.1. Studies that examined higher concentrations of
3 oxides of nitrogen were evaluated particularly to inform mode of action. Causality
4 determinations were made by evaluating evidence for the consistency of findings across
5 multiple studies, the coherence of findings across related endpoints and across
6 disciplines, and the extent to which chance, confounding (i.e., bias due to a correlation
7 with exposures or ambient concentrations of oxides of nitrogen and relationship with the
8 outcome), and other biases could be ruled out with reasonable confidence. This
9 evaluation involved consideration of the strength of study design and analytical methods
10 as well as the potential for selection bias, publication bias, and confounding. Aspects
11 used in the evaluation and integration of evidence to form a causal classification are
12 described in more detail in the Preamble.
13 Epidemiologic studies of short-term exposure to oxides of nitrogen rely primarily on
14 temporal variation in exposure (e.g., day-to-day variations in ambient NO2
15 concentrations) and health effects. For the assessment of potential confounding,
16 epidemiologic studies of short-term exposure were evaluated for the extent to which they
17 considered other factors associated with health outcomes and exhibited similar temporal
18 variation as exposures to oxides of nitrogen. These potential confounding factors can
19 include meteorological factors, season, long-term time trends, medication use, and
20 copollutant exposures. Epidemiologic studies varied in the extent to which they
21 considered potential confounding. Because no single study considered all potential
22 confounding factors, and not all potential confounding factors were examined in the
23 collective body of evidence, residual confounding by unmeasured factors is possible.
24 Residual confounding also may result from poorly measured factors. However, potential
25 confounding was assessed as the extent to which the collective literature base examined
26 factors well documented in the literature to be associated with exposure to oxides of
27 nitrogen and health outcomes (e.g., meteorological factors, others specified above).
28 Epidemiologic studies examine various averaging times of NO2, NO and NOX
29 concentrations (e.g., 24-h avg, daily 1-h max) and present effect estimates for
30 associations with health outcomes scaled to various changes in concentrations, e.g.,
31 interquartile range for the study period or an arbitrary unit such as 10 ppb. To increase
32 comparability among studies, the ISA presents effect estimates for a given averaging time
33 scaled to the same increment. For short-term exposure, effect estimates are scaled to a
34 20-ppb increase for 24-h avg NO2 or NO, a 40-ppb increase for 24-h avg NOX, a 30-ppb
35 increase for 1-h max NO2 or NO, and a 60-ppb increase for 1-h max NOX. These
36 increments were derived by calculating the U.S. nationwide percentile distributions for
37 various averaging times, and they represent the estimated difference between the median
38 (a typical pollution day) and the 95th percentile (a more polluted day) for a given
39 averaging time.
November 2013 4-2 DRAFT: Do Not Cite or Quote
-------
1 Controlled human exposure and animal toxicological studies can provide direct evidence
2 for health effects related to pollutant exposures. They also can be used to address
3 uncertainties in the epidemiologic evidence, for example, potential confounding.
4 Experimental studies additionally can provide biological plausibility for observed effects
5 by describing key events to inform modes of action. Thus, the integration of evidence
6 across a spectrum of related endpoints, including cause-specific mortality, and across
7 disciplines was used to inform uncertainties for any particular endpoint or discipline due
8 to factors such as publication bias, selection bias, and confounding by copollutant
9 exposures.
4.2 Respiratory Effects
4.2.1 Introduction
10 The 2008 ISA for Oxides of Nitrogen concluded that there was sufficient evidence to
11 infer that a causal relationship is likely to exist between short-term exposure to NO2 and
12 respiratory effects (U.S. EPA. 2008c). This conclusion was based primarily on a large
13 body of epidemiologic evidence consistently indicating associations between increases in
14 ambient NO2 concentrations and increases in respiratory-related hospital admissions and
15 emergency department (ED) visits with supporting evidence for increases in respiratory
16 symptoms in children with asthma. Copollutant modeling results generally indicated
17 robust associations with adjustment for copollutants, such as ozone (O3), carbon
18 monoxide (CO), or particulate matter (i.e., PM25, PMio), supporting an independent
19 effect of ambient NO2 exposure. Biological plausibility was provided, in particular, by
20 observations of NO2-induced airway hyperresponsiveness (AHR) in adults with asthma
21 following <1 to 6-hour exposures to NO2 at concentrations in the range of 100 to 300
22 ppb, on the order of peak 1-h maximum ambient concentrations examined in
23 epidemiologic studies. The 2008 ISA for Oxides of Nitrogen also noted some support for
24 pulmonary inflammation and impaired host defenses in controlled human exposure and
25 animal toxicological studies, albeit at higher concentrations of 1,500 to 5,000 ppb NO2.
26 There was less consistent evidence for NO2-associated lung function decrements, as
27 examined in controlled human exposure and epidemiologic studies. However,
28 epidemiologic studies in adults and children found associations with lung function
29 measured by supervised spirometry (U.S. EPA. 2008c).
30 Substantial evidence in support of a relationship between short-term NO2 exposure and
31 respiratory effects was provided by the coherence of findings across disciplines for
32 related outcomes. Specifically, evidence for increases in AHR and pulmonary
November 2013 4-3 DRAFT: Do Not Cite or Quote
-------
1 inflammation in controlled human exposure, animal toxicological, and to a limited extent,
2 epidemiologic studies provided biological plausibility for respiratory symptoms in
3 children with asthma which in turn, provided biological plausibility for increases in
4 asthma hospital admissions and ED visits. Although there was coherence of evidence
5 across related outcomes and disciplines, a major uncertainty that remained regarding the
6 respiratory effects of short-term ambient NO2 exposure was the high correlations of NO2
7 with other traffic-related pollutants and the potential for NO2 to serve primarily as an
8 indicator for another or mixture of combustion-related pollutants.
9 As will be described in the following sections, consistent with the body of evidence
10 presented in the 2008 ISA for Oxides of Nitrogen, recent studies continue to demonstrate
11 respiratory effects related to short-term NO2 exposure. The majority of the recent
12 evidence is from epidemiologic studies, which expand on the evidence for ambient
13 NO2-associated increases in respiratory hospital admissions and ED visits (including
14 those for asthma), increases in pulmonary inflammation, oxidative stress, and respiratory
15 symptoms in children with asthma, as well as increases in respiratory mortality. The
16 discussion of the evidence is organized by outcome with results from recent studies,
17 where available, evaluated in the context of those from previous studies.
4.2.2 Airway Hyperresponsiveness
18 Inhaled pollutants such as NO2 may have direct effects on lung function or they may
19 enhance the inherent responsiveness of the airways to challenge by bronchoconstricting
20 agents. Challenge agents can be classified as nonspecific (e.g., histamine, SO2, cold air)
21 or specific (i.e., allergen). Nonspecific agents can be differentiated between "direct"
22 stimuli (e.g., histamine, carbachol, and methacholine) which act on airway smooth
23 muscle receptors and "indirect" stimuli (e.g., exercise, cold air) which act on smooth
24 muscle through intermediate pathways, especially via inflammatory mediators (Cockcroft
25 and Davis. 2006c). Specific allergen challenges (e.g., house dust mite, cat allergen) also
26 act "indirectly" via inflammatory mediators to initiate smooth muscle contraction and
27 bronchoconstriction. This section primarily focuses on changes in airway responsiveness
28 to bronchial challenge attributable to NO2 in individuals with asthma. Discussed in
29 Section 3.3.2.5. toxicological studies have demonstrated increased airway responsiveness
30 to nonspecific challenges following short-term exposures to 4,000 ppb NO2 (Kobayashi
31 and Shinozaki. 1990). Described in Section 4.2.4.3 (Allergic Inflammation), altered
32 responses to specific allergens following NO2 exposure have also been demonstrated in
33 human and animal studies. There is a wide range of airways responsiveness that is
34 influenced by many factors, including medications, cigarette smoke, air pollutants,
35 respiratory infections, occupational exposures, disease status, and respiratory irritants.
November 2013 4-4 DRAFT: Do Not Cite or Quote
-------
1 There are several notable changes and additions to the discussion of airway
2 responsiveness in this ISA as compared to Section 3.1.3 of the 2008 ISA for Oxides of
3 Nitrogen (U.S. EPA. 2008c). Only one new experimental study (Riedl et al.. 2012) and a
4 new meta-analysis (Goodman et al. 2009) of NO2 associated increases in airway
5 responsiveness have been published since the 2008 ISA. Accordingly, this ISA focuses
6 primarily on meta-analyses of airway responsiveness data rather than specifics of
7 individual studies. Relative to Table 3.1-2 in the 2008 ISA, Table 4-1 and Table 4-2
8 include new or previously not included data (namely, specific allergen challenges had
9 been intentionally excluded) for 155 subject exposures from nine studies (Riedl et al..
10 2012; Wittenetal.. 2005; Barck et al.. 2002: Jenkins et al.. 1999: Strand et al.. 1998:
11 Strand et al.. 1997: Tunnicliffe et al.. 1994: Morrow and Utell. 1989a: Orehek et al..
12 1976). Based on data in Table 4-1 and Table 4-2. an updated meta-analysis (Table 4-3)
13 and detailed methodology are provided. Consistent with conclusions reached in the 2008
14 ISA, the updated meta-analysis (Table 4-3) shows that in individuals exposed to NO2 at
15 rest, increases in nonspecific airway responsiveness occur in the range of 200 and 300
16 ppb NO2 for 30 minute exposures and at 100 ppb NO2 for 60 minute exposures in
17 individuals with asthma. Finally, a section has been added evaluating the potential of
18 various factors to affect airways hyperresponsiveness independently or in conjunction
19 with NO2 exposure (Section 4.2.2.3).
20 Responses to bronchial challenge are typically quantified in terms of the provocative dose
21 (PD) or concentration (PC) of an agent required to produce a 20% reduction in forced
22 expiratory volume in 1 second (FEVi) (PD20, PC20, respectively) or a 100% increase in
23 specific airway resistance (PDioo, PCioo, respectively). In the general population, airway
24 responsiveness is log-normally distributed with individuals having AHR tending to be
25 those with asthma (Postma and Boezen. 2004): although, the airway responsiveness of
26 individuals with asthma extends into the normal range (Cockcroft. 2010). Along with
27 symptoms, variable airway obstruction, and airway inflammation, AHR is a primary
28 feature in the clinical definition and characterization of asthma severity (Reddel et al..
29 2009). In asthma, there is a strong relationship between the degree of nonspecific airway
30 responsiveness and the intensity of the early airway response to specific allergens to
31 which individuals have become sensitized (Cockcroft and Davis. 2006a).
32 Due to their predisposition for AHR, individuals with asthma generally require a lower
33 PD of a bronchial challenge agent than healthy individuals to produce a given reduction
34 in lung function. In Morrow and Utell (1989a). the average PD of carbachol producing a
35 given change in lung function in individuals with mild-to-moderate asthma was 16 times
36 lower than in age-matched healthy controls. Similarly, Hazucha et al. (1983) reported a
37 10-12 times lower average baseline PDioo to methacholine in individuals with mild
38 asthma than healthy age-matched controls. The PDs for asthma in Morrow and Utell
November 2013 4-5 DRAFT: Do Not Cite or Quote
-------
1 (1989a) did not overlap with those of the healthy controls, whereas Hazuchaetal. (1983)
2 observed an overlap with two of fifteen subjects with asthma being relatively
3 unresponsive to bronchial challenge. The bronchoconstrictive response to indirect acting
4 agents (especially specific allergens) can be more difficult to predict and control than the
5 bronchoconstrictive response to nonspecific agents that act directly on airway smooth
6 muscle receptors (O'Byrne et al.. 2009). Consequently, most of the available literature
7 relevant to the evaluation of the effects of NO2 on AHR has focused primarily on the
8 responses of individuals with asthma to bronchial challenge with "nonspecific"
9 bronchoconstricting agents (e.g., methacholine, SO2, cold air).
10 In healthy adults without asthma or AHR, there is likely little or no clinical significance
11 of transient, small increases in airway responsiveness following low-level NO2 inhalation
12 exposures. In individuals with asthma, however, transient changes in AHR in response to
13 inhaled pollutants may have clinical consequences. Increased airway responsiveness is
14 linked with airway inflammation and airway remodeling (Chetta et al.. 1996). increased
15 risk for exacerbations (Van Schayck et al., 1991). reduced lung function (Xuan et al..
16 2000). and increased symptoms (Murray et al.. 1981). A variety of environmental
17 challenges can transiently increase AHR and worsen asthma control, including allergen
18 exposures (Strand et al.. 1997; Brusasco etal.. 1990). viral infections (Cheung et al..
19 1995; Fraenkel et al.. 1995). cigarette smoke (Tashkin et al.. 1993). O3 (Kehrl et al..
20 1999). and other respiratory irritants (Kinsellaet al.. 1991). An exposure that worsens
21 AHR to one agent in subjects with asthma may also enhance airway responsiveness to
22 other challenge agents. Transient increases in AHR following NO2 or other pollutant
23 exposures have the potential to increase symptoms and worsen asthma control, even if the
24 pollutant exposure does not cause acute decrements in lung function.
4.2.2.1 Healthy Individuals
25 The 2008 ISA for Oxides of Nitrogen reported that increases in nonspecific airway
26 responsiveness were observed in the range of 1,500 to 2,000 ppb NO2 for 3-hour (3-h)
27 exposures in healthy adults (U.S. EPA. 2008c). Studies of airway responsiveness in
28 healthy individuals were generally conducted using volunteers of 18 to 35 + years of age.
29 Mohsenin (1988) found that a 1-h resting exposure to 2,000 ppb NO2 increased
30 responsiveness to methacholine. A mild increase in responsiveness to carbachol was
31 observed following a 3-h exposure to 1,500 ppb NO2 with moderate intermittent exercise
32 (VE = 40 L/min; 10 of 30 minutes) (Frampton et al.. 1991). Kulle and Clements (1988)
33 also showed a tendency for greater FEVi decrements from methacholine challenge
34 following 2-h resting exposures to 2,000 and 3,000 ppb NO2. Resting exposures to 100
35 ppb NO2 for 1 hour have not affected carbachol or methacholine responsiveness of
November 2013 4-6 DRAFT: Do Not Cite or Quote
-------
1 healthy subjects (Ahmed et al., 1983b; HazuchaetaL 1983). Two meta-analyses of the
2 available literature confirm significant effects of NO2 exposures above 1,000 ppb, but not
3 below, on airway responsiveness in healthy individuals (Kjaergaard and Rasmussen.
4 1996; Folinsbee. 1992). More recent studies of airways responsiveness in healthy
5 individuals following NO2 exposure are not available.
4.2.2.2 Individuals with Asthma
6 The 2008 ISA for Oxides of Nitrogen reported that increases in nonspecific airway
7 responsiveness were observed in the range of 200 and 300 ppb NO2 for 30-minute
8 exposures and at 100 ppb NO2 for 60-minute exposures in individuals with asthma (U.S.
9 EPA. 2008c). Enhanced airway responsiveness to allergens was found in individuals with
10 asthma at exposures as low as 260 ppb for 30 minutes (U.S. EPA. 2008c). Detailed
11 descriptions of individual studies are provided in the 1993 AQCD for Oxides of Nitrogen
12 (U.S. EPA. 1993) and 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
13 As an update to Table 3.1-2 in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
14 Table 4-1 and Table 4-2 present studies for which individual data were available to
15 evaluate the fraction of subjects whose airway responsiveness increased or decreased
16 following exposure to NO2. In general, the subjects recruited for these studies ranged in
17 age from 18 to 50 years with the exception of Avol etal. (1989) who studied children
18 aged 8-16 years. The disease status of subjects was mild asthma in most studies, but
19 ranged from inactive asthma up to severe asthma in a few studies. For studies that
20 assessed AHR at multiple time points post-exposure or over repeated days of exposure,
21 the data from the first time point and first day of exposure were selected for inclusion in
22 Table 4-1 and Table 4-2 in an attempt to reduce the heterogeneity between studies.
23 Selection of the earliest time point assessing AHR was, in part, due to late phase
24 responses (3-8 hours post-allergen challenge) being mechanistically different from early
25 phase responses (<30 minutes post-allergen challenge) (O'Byrne et al.. 2009; Cockcroft
26 and Davis. 2006c). It should be noted that Table 4-1 and Table 4-2 are sorted by NO2
27 exposure concentration and, as such, studies that evaluated multiple NO2 exposure
28 concentrations appear in multiple rows. The statistical significance reported in studies for
29 changes in AHR following NO2 exposure compared to filtered air is also provided in
30 these tables. Based on all listed studies, the general tendency of most studies is toward
31 increased AHR following NO2 exposure with some studies reaching statistical
32 significance. Fewer studies showed no effect or a tendency for decreased AHR following
33 NO2. Published since the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). the one
34 recent study reported a statistically significant decrease in AHR following NO2, but the
35 authors attributed the protective effect of NO2 to chance (Riedl etal.. 2012).
November 2013 4-7 DRAFT: Do Not Cite or Quote
-------
Table 4-1 Resting exposures to NO2 and airway responsiveness in subjects with asthma.
Reference
Ahmed et al. (1983b)
Ahmed et al. (1983a)
Hazuchaetal. (1983)
Oreheketal. (1976)
Tunnicliffe et al. (1994)
Bvlin etal. (1988)
Oreheketal. (1976)
Jorres and Maqnussen
(1990)
Barck et al. (2002)
Strand etal. (1997)
Strand etal. (1998)
Bvlin etal. (1988)
Tunnicliffe etal. (1994)
N
20
20
15
20
8
20
4
14
13
18
16
20
8
NO2
(PPb)
100
100
100
100
100
140
200
250
260
260
260
270
400
Exp.
(min)
60
60
60
60
60
30
60
30
30
30
30
30
60
Challenge
Type
CARB
RAG
METH
CARB
HDM
HIST
CARB
SO2
SIR, TIM
SIR, TIM
SIR
HIST
HDM
End
Point
sGaw
sGaw
sRaw
sRaw
FEV-,
sRaw
sRaw
sRaw
FEV1
sRaw
FEV1
sRaw
FEV1
Time Post-
exp (min)
NA
IM
20
IM
IM
25
IM
27
240
240
240
25
IM
Change in
AHRa
+
13
10
6
14
3
14
3
11
5
9
11
14
8
-
7
8
7
3
5
6
0
2
7
9
4
6
0
Average PD ± SEb
Air
6.0 ±2.4
9.0 ±5.7
1.9±0.4
0.56 ±0.08
-14.62
AFEV-,
0.39 ±0.07
0.60 ±0.10
46.5 ±5.1
-5 ±2
AFEV-,
860 ± 450
-0.1 ±0.8
AFEV-,
0.39 ±0.07
-14.62
AFEV-,
NO2
2.7 ±0.8
11.7±7.6
2.0 ± 1.0
0.36 ±0.05
-14.41
AFEV-,
0.28 ±0.05
0.32 ± 0.02
37.7 ±3.5
-4 ±2
AFEV-,
970 ± 450
-2.5 ± 1.0
AFEV-,
0.24 ± 0.04
-18.64
AFEV-,
p-valuec
NA
n.s.
n.s.
<0.01d
n.s.
n.s.
n.s.
p<0.01
n.s.
n.s.
0.03
<0.01
0.009
November 2013
4-8
DRAFT: Do Not Cite or Quote
-------
Table 4-1 (Continued): Resting exposures to NO2 and airway responsiveness in subjects with asthma.
Reference
Bvlin etal. (1985)
Mohsenin (1987a)
Bvlin etal. (1988)
N
8
10
20
NO2
(PPb)
480
500
530
Exp.
(min)
20
60
30
Challenge
Type
HIST
METH
HIST
End
Point
sRaw
pEF
sRaw
Time Post-
exp (min)
20
IM
25
Change in
AHRa
+
5
7
12
-
0
2
7
Average PD ± SEb
Air
>30
9.2 ±4.7
0.39 ±0.07
NO2
>20
4.6 ±2.6
0.34 ±0.08
p-valuec
0.04
0.042
n.s.
Abbreviations: BIR, birch; CARS, carbachol; COLD, cold-dry air; FEN/!, forced expiratory volume in 1 s; HIST, histamine; IM, immediately after exposure; METH, methacholine; NA, not
available; NO2, nitrogen dioxide; n.s., less than marginal statistical significance, p >0.10; pEF, partial expiratory flow at 40% vital capacity; RAG, ragweed; SO2, sulfur dioxide; sGaw,
specific airway conductance; sRaw, specific airway resistance; TIM, timothy.
aChange in AHR: number of individuals showing increased (+) or decreased (-) AHR after NO2 compared to air.
bPD ± SE, Arithmetic or geometric mean provocative dose (PD) ± standard error (SE). See individual papers for PD calculation and dosage units. AFEV! indicates the change in
response at a constant challenge dose.
°Statistically significance of increase in AHR to bronchial challenge following NO2 exposure compared to filtered air. Statistical tests varied between studies, e.g., sign test, t-test,
analysis of variance.
Statistically significance for all asthmatics from analysis by Dawson and Schenker (1979). Oreheket al. (1976) only tested for differences in subsets of individuals classified as
"responders" and "nonresponders."
November 2013
4-9
DRAFT: Do Not Cite or Quote
-------
Table 4-2 Exercising exposures to NO2 and airway responsiveness in subjects with asthma.
Reference
Roqeretal. (1990)
Kleinman et al. (1983)
Jenkins et al. (1999)
Jorres and Maqnussen
(1991)
Strand et al. (1996)
Avoletal. (1988)
Avoletal. (1989)
Bauer et al. (1986)
Morrow and Utell
(1989a)
Roqeretal. (1990)
Rubinstein et al.
(1990)
Riedletal. (2012)
Riedletal. (2012)
Jenkins et al. (1999)
Witten et al. (2005)
N
19
31
11
11
19
37
34
15
20
19
9
15
15
10
15
NO2
(PPb)
150
200
200
250
260
300
300
300
300
300
300
350
350
400
400
Exp.
(min)
80
120
360
30
30
120
180
30
240
80
30
120
120
180
180
Challenge
Type
METH
METH
HDM
METH
HIST
COLD
COLD
COLD
CARB
METH
SO2
METH
CAT
HDM
HDM
End
Point
sRaw
FEV-,
FEV-,
sRaw
sRaw
FEV1
FEV-,
FEV-,
FEV-,
sRaw
sRaw
FEV-,
FEV-,
FEV-,
FEV-,
Time Post-
exp (min)
120
IM
IM
60
30
60
60
60
30
120
60
90
90
IM
IM
Change in
AHRa
+
10d
20
6
6
13
11d
12d
9
T
8d
4
6
4
7
8
-
7d
7
5
5
5
16d
21d
3
2e
9d
5
7
11
3
7
Average PD ± SEb
Air
3.3 ±0.7
8.6 ±2.9
2.94
0.41 ± 1.6
296 ± 76
-8.4 ± 1.8
AFEV-,
-5 ±2
AFEV-,
0.83 ±0.12
3.31 ± 8.64e
AFEV1
3.3 ±0.7
1.25 ±0.23
7.5 ±2.6
-6.9 ± 1.7
AFEV-,
3.0
550 ± 240
NO2
3.1 ±0.7
3.0± 1.1
2.77
0.41 ± 1.6
229 ± 56
-10.7 ±2.0
AFEV-,
-4 ±2
AFEV1
0.54 ±0.10
-6.98 ± 3.35 E
AFEV-,
3.3 ±0.8
1.31 ±0.25
7.0 ±3.8
-0.5 ± 1.7
AFEV-,
2.78
160 ±60
p-valuec
n.s.
<0.05
n.s.
n.s.
0.08
n.s.
n.s.
<0.05
n.s.
n.s.
n.s.
n.s.
<0.05f
0.018
n.s.
November 2013
4-10
DRAFT: Do Not Cite or Quote
-------
Table 4-2 (Continued: Exercising exposures to NO2 and airway responsiveness in subjects with asthma.
Reference
Avoletal. (1988)
Roger et al. (1990)
N
37
19
NO2
(PPb)
600
600
Exp.
(min)
120
80
Challenge
Type
COLD
METH
End
Point
FEV-,
sRaw
Time Post-
exp (min)
60
120
Change in
AHRa
+
13°
11d
-
16°
8D
Average PD ± SEb
Air
-8.4 ± 1.8
AFEV-,
3.3 ±0.7
NO2
-10.4 ±2.2
AFEV-,
3.7± 1.1
p-valuec
n.s.
n.s.
Abbreviations: AHR, airway hyperresponsiveness; CARS, carbachol; CAT, cat allergen; COLD, cold-dry air; FEN/!, forced expiratory volume in 1 s; HDM, house dust mite allergen;
HIST, histamine; IM, immediately after exposure; METH, methacholine; NO2, nitrogen dioxide; n.s., less than marginal statistical significance, p >0.10; SO2, sulfur dioxide; sRaw,
specific airway resistance.
aChange in AHR: number of individuals showing increased (+) or decreased (-) AHR after NO2 compared to air.
bPD ± SE, Arithmetic or geometric mean provocative dose (PD) ± standard error (SE). See individual papers for PD calculation and dosage units. AFEV! indicates the change in
response at a constant challenge dose.
°Statistically significance of increase in AHR to bronchial challenge following NO2 exposure compared to filtered air. Statistical tests varied between studies, e.g., sign test, t-test,
analyses of variance.
dNumber of individuals having an increase or decrease in AHR is from Folinsbee (1992).
eData for 0.25% carbachol challenge from Appendix H of Morrow and Utell (1989b).
'Significantly greater AFEV! in response to a constant challenge dose following exposure to filter air than NO2, i.e., a protective effect of NO2 exposure.
November 2013
4-11
DRAFT: Do Not Cite or Quote
-------
1 Three meta-analyses in the peer-reviewed literature have assessed the effects of NO2
2 exposure on airway responsiveness in individuals with asthma (Goodman et al.. 2009;
3 Kjaergaard and Rasmussen. 1996; Folinsbee. 1992). Kjaergaard and Rasmussen (1996)
4 reported statistically significant effects of NO2 exposure on the airway responsiveness of
5 subjects with asthma exposed to less than or equal to 300 ppb NO2, but not for exposures
6 in excess of 300 ppb NO2. With consideration given to activity level during exposure,
7 Folinsbee (1992) found statistically significant increases in airway responsiveness of
8 subjects with asthma exposed to NO2 at rest across all concentration ranges (namely,
9 <200 ppb, 200 to 300 ppb, and >300 ppb). However, there was no significant effect of
10 NO2 exposures on responsiveness during exercise. For instance, following exposures
11 between 200 and 300 ppb NO2, 76% of subjects exposed at rest had statistically increased
12 responsiveness, whereas only 52% of subjects exposed with exercise tended to have
13 increased responsiveness. The analyses of Folinsbee (1992) and Kjaergaard and
14 Rasmussen (1996) effectively assessed nonspecific responsiveness since few studies of
15 allergen responsiveness were available.
16 The analyses conducted by Folinsbee (1992) were detailed in the 1993 AQCD for Oxides
17 of Nitrogen (U.S. EPA. 1993). Results of these analyses appeared in Table 15-10 of that
18 AQCD and supported the conclusion that NO2 exposure increases airway responsiveness
19 in individuals with asthma. The results of a slightly modified analysis focusing
20 exclusively on nonspecific responsiveness appeared in Table 3.1-3 on the 2008 ISA for
21 Oxides of Nitrogen (U.S. EPA. 2008c). The overall conclusion of that modified analysis
22 was that NO2 exposures conducted during rest, but not exercise, in the range of 200 and
23 300 ppb NO2 for 30-minute exposures and at 100 ppb NO2 for 60-minute exposures
24 increased nonspecific responsiveness in individuals with asthma. Due to differences in
25 study protocols (e.g., rest versus exercise) in the NO2-AHR literature, the original
26 (Folinsbee. 1992) and updated meta-analyses in the 2008 ISA for Oxides of Nitrogen
27 (U.S. EPA. 2008c) assessed only the fraction of individuals experiencing increased or
28 decreased airway responsiveness following NO2 exposure.
29 A recent study by Goodman et al. (2009) provided meta-analyses and meta-regressions
30 evaluating the effects of NO2 exposure on airway responsiveness in subjects with asthma.
31 By considering studies of specific allergen and nonspecific responsiveness following
32 NO2 exposure, Goodman et al. (2009) evaluated a larger number of studies than the
33 analysis in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). which was limited to
34 nonspecific responsiveness in subjects with asthma in an attempt to reduce the
35 heterogeneity between studies. Goodman et al. (2009) evaluated changes in three
36 endpoints following NO2 exposure relative to a control air exposure: (1) the fraction of
37 subjects with asthma experiencing increases in responsiveness, (2) the PD of the
38 bronchial challenge agent, and (3) the FEVi response to the challenge agent. Overall,
November 2013 4-12 DRAFT: Do Not Cite or Quote
-------
1 statistically significant effects of NO2 exposure on each of these three endpoints were
2 observed. Consistent with the meta-analysis provided in the 2008 ISA for Oxides of
3 Nitrogen (U.S. EPA. 2008c). Goodman et al. (2009) found 64% (95% CI: 58%, 71%) of
4 subjects with asthma exposed at rest to NO2 experience an increase in airway
5 responsiveness, whereas there was no effect of NO2 exposure during exercise with 52%
6 (95% CI: 43%, 60%) having an increase in responsiveness. Additionally, NO2 exposure
7 resulted in a reduction in PD and increased the FEVi decrement following bronchial
8 challenge.
9 Goodman et al. (2009) concluded that, "NO2 is not associated with clinically relevant
10 effects on AHR at exposures up to 600 ppb based primarily on the small magnitude of
11 effects and the overall lack of exposure-response associations." Relative to therapeutic
12 agents used to treat airway responsiveness, which may be considered effective if they
13 more than double the PD for methacholine, the authors further concluded that the effects
14 of NO2 exposure on airway responsiveness were sufficiently small so as not to be
15 considered adverse. By this assessment, the authors concluded that a -50% change in the
16 PD would be considered adverse, whereas the effect of NO2 exposure was a -27% (95%
17 CI: -37%, -18%) reduction in the PD. Stratifying by rest and exercise exposure, the
18 NO2-induced changes in PD were -30% (95% CI: -38%, -22%) and -24% (95% CI:
19 -40%, -7%), respectively. The appropriateness of weighing the deleterious effects of a
20 generally unavoidable ambient exposure using the criteria for judging the efficacy of
21 beneficial therapeutic agents is not clear. Based on the lack of a monotonic increase in
22 responsiveness with exposure, the authors also suggested that NO2 is not a causal factor.
23 The nature of the relationship between NO2 exposure and airway responsiveness, as well
24 as factors potentially affecting within- and between-study variability in observed
25 responses, is discussed later in this section.
26 Based on the summary data in Table 4-1 and Table 4-2. the fraction of individuals
27 experiencing a NO 2-induced increase in airway responsiveness can be assessed in a
28 manner consistent with the analysis conducted by Folinsbee (1992). The magnitude of
29 NO 2 -induced changes in PD were not evaluated due to considerable variability in
30 exposure protocols and the potential for this variability in protocols to affect estimates of
31 PD (see Section 4.2.2.3). Specifically, a two-tailed sign test was used to assess the
32 statistical significance of directional changes in AHR between the NO2 and filter air
33 exposure days. The nonparametric sign test, which assumes only that the responses of
34 each subject are independent and makes no assumptions about the distribution of the
35 response data, is appropriate to test the null hypothesis that observed values have the
36 same probability of being positive or negative. This test allows estimation of whether a
37 significant fraction of individuals experience an increase or decrease in airway
38 responsiveness, but does not provide information on the magnitude of the change in that
November 2013 4-13 DRAFT: Do Not Cite or Quote
-------
1 endpoint. The significance of a two-tailed sign test may be calculated in Microsoft®
2 Office Excel® 2007 as:
3 For AHR+ > (AHR+ + AHR )/2
4 p-value = 2 * (1-BINOMDIST(AHR+ -1, (AHR+ + AHR), 0.5,TRUE))
5 For AHR+ < (AHR+ + AHR )/2
6 p-value = 2 * BINOMDIST(AHR+, (AHR+ + AHR), 0.5,TRUE)
7 For AHR+ = AHR
8 p-value = 1.00
9 where: AHR+ and AHR" are the number of individuals in Table 4-1 and Table 4-2 having
10 an increase or decrease in airway responsiveness, respectively, under a specified set of
11 conditions (i.e., NO2 concentration, exercise versus rest during exposure, nonspecific
12 versus allergen challenge); the BINOMDIST function in Excel® returns the binomial
13 distribution probability given the total number of increases (AHR+), number experiencing
14 a change (AHR+ + AHR"), probability of a change (0.5), for the cumulative distribution
15 (indicated by the logical, TRUE); and the multiplication by two provides the probability
16 for a two-tailed test. Table 4-3. Table 4-4. and Table 4-5 present the fraction of
17 individuals experiencing aNO2-induced increase in airway responsiveness to nonspecific
18 agents, specific allergens, and all challenge types, respectively.
19 Table 4-3 updates Table 3.1-3 of the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c)
20 and is consistent with the prior conclusion that increases in nonspecific airway
21 responsiveness (following resting NO2 exposures) occur in the range of 200 and 300 ppb
22 NO2 for 30-minute exposures and at 100 ppb NO2 for 60-minute exposures in individuals
23 with asthma. Increases in airways responsiveness were not observed following the
24 exercising exposures to NO2. As discussed in Section 4.2.2.3. the literature on airway
25 responsiveness supports the development of a refractory period following bouts of
26 exercise. An effect of exercise refractoriness is consistent with NO2-induced increases in
27 airway responsiveness following resting but not excercising exposures.
November 2013 4-14 DRAFT: Do Not Cite or Quote
-------
Table 4-3 Fraction of subjects with asthma having NO2-induced increase in
airway hyperresponsiveness to a nonspecific challenge.
NO2 Concentration, ppb
[NO2]= 100
100<[NO2]<200
200<[NO2]<300
[NO2]>300
All [NO2]
All Exposures3'13
0.66 (50; p = 0.033)
0.66 (87; p = 0.005)
0.59(199; p = 0.011)
0.57(94; n.s.)
0.60(380; p<0.001)
Exposure with
Exercisea'b
—
0.59(17; n.s.)
0.55(163; n.s.)
0.49(61; n.s.)
0.54(241; n.s.)
Exposure at
0.66 (50; p =
0.67 (70; p =
0.78 (36; p =
0.73 (33; p =
Resta'b
0.033)
0.006)
0.001)
0.014)
0.71 (139; p<0.001)
Abbreviations: n.s., less than marginal statistical significance (p >0.10)
"Data are the fraction of subjects with asthma having an increase in airway responsiveness following NO2 versus air exposure.
Values in parentheses are number of individuals with asthma having a change in responsiveness and the p-value for a two-tailed
sign test.
bAnalysis is for the 380 subjects with asthma in Table 4-1 and Table 4-2 having a change (+/-) in nonspecific AHR.
Table 4-4 Fraction of subjects with asthma having NO2-induced increase in
specific airway hyperresponsiveness to an allergen challenge.
NO2 Concentration, ppb
[NO2]= 100
100<[NO2]<200
200<[NO2]<300
[NO2]>300
All [NO2]
All Exposures3'"
0.50(26; n.s.)
0.50(26; n.s.)
0.55(56; n.s.)
0.56(48; n.s.)
0.55(130; n.s.)
Exposure with
Exercise3'"
—
—
0.55(11; n.s.)
0.48(40; n.s.)
0.49(51; n.s.)
Exposure at Resta'b
0.50(26; n.s.)
0.50(26; n.s.)
0.56(45; n.s.)
1.00(8; p = 0.008)
0.58(79; n.s.)
Abbreviations: n.s., less than marginal statistical significance (p >0.10)
aSee Footnote "a" of Table 4-3.
bAnalysis is for the 130 subjects with asthma in Table 4-1 and Table 4-2 having a change (+/-) in specific allergen AHR.
November 2013 4-15 DRAFT: Do Not Cite or Quote
-------
Table 4-5 Fraction of subjects with asthma having NO2-induced increase in
airway hyperresponsiveness regardless of challenge types.
NO2 Concentration
[NO2]= 100
100<[NO2]<200
200<[NO2]<300
[NO2]>300
All [NO2]
, ppb All Exposures3'13
0.61 (76; p = 0.08)
0.62(113; p = 0.014)
0.58 (255; p = 0.008)
0.57(142; n.s.)
0.59(510; p<0.001)
Exposure during
Exercisea'b
—
0.59(17; n.s.)
0.55(174; n.s.)
0.49(101; n.s.)
0.53 (292; n.s.)
Exposure at Resta'b
0.61 (76; p = 0.08)
0.63(96; p = 0.018)
0.65(81; p = 0.007)
0.78(41; p<0.001)
0.67(218; p<0.001)
Abbreviations: n.s., less than marginal statistical significance (p >0.10)
aSee Footnote "a" of Table 4-3.
bAnalysis is for the 51 0 subjects with asthma in Table 4-1 and Table 4-2 having a change (+/-) in AHR.
In general. Table 4-4 does not show significant effects of NO? exposure on airway
1
2 responsiveness to allergen challenge. This may be, in part, due to the small number of
3 individuals in the analysis. The lack of statistical significance in Table 4-4 does not
4 necessarily diminish the potential importance of allergen exposures. Eighty percent of
5 children with asthma are thought to be sensitized to common household allergens
6 (O'Byrne et al.. 2009). Individuals with asthma may experience an early phase response
7 to allergen challenge with declines in lung function within 30 minutes; and,
8 approximately half of those having an early phase response also have a late phase
9 response with a decline in lung function 3-8 hours after the challenge (O'Bvrne et al..
10 2009; Cockcroft and Davis. 2006c). The early response, which may be reversed with
11 bronchodilators, primarily reflects release of histamine and other mediators by airway
12 mast cells; whereas, the late response reflects enhanced airways inflammation and
13 mucous production and requires steroidal treatment. Studies have reported NO2-induced
14 effects on allergen responsiveness for both the early phase (Jenkins et al.. 1999; Strand et
15 al.. 1998; Tunnicliffe et al.. 1994) and late phase (Strand et al.. 1998; Tunnicliffe et al..
16 1994). These effects were observed following 30-minute resting exposures to
17 concentrations as low as 260 ppb NO2. The degree of airway responsiveness is not only a
18 function of the concentration of inhaled allergen, but also the degree of sensitization as
19 measured by the level of allergen-specific IgE and responsiveness to nonspecific agents
20 (Cockcroft and Davis. 2006a). These factors make it difficult to predict the level of
21 responsiveness to an allergen, and although rare, severe bronchoconstriction can occur
22 with inhalation of very low allergen concentrations (O'Bvrne et al.. 2009). Given the
23 ubiquity of allergens and potential severity of effects, the responsiveness to allergens in
24 animals and humans is also addressed in Sections 3.3.2.6 (Modification of
25 Innate/Adaptive Immunity) and 4.2.4.3 (Allergen Inflammation).
November 2013 4-16 DRAFT: Do Not Cite or Quote
-------
1 With all challenge types considered, Table 4-5 shows statistically significant increases in
2 airway responsiveness across all exposure concentration in subjects with asthma exposed
3 at rest. However, given differing mechanisms of effect (see discussion of Bronchial
4 Challenge Agent in Section 4.2.2.3). preference should be given to the analysis of
5 nonspecific responsiveness (Table 4-3) over the combined analysis of specific and
6 nonspecific agents (Table 4-5).
4.2.2.3 Factors Affecting Airway Hyperresponsiveness and Dose-
response
Exercise
7 In considering why increases in airway responsiveness occurred only after resting
8 exposure to NO2, Folinsbee (1992) and Bylin (1993) suggested that exercise itself may
9 affect the mechanisms responsible for increased responsiveness. Based on the literature at
10 that time, both of these authors noted that exercise may cause a refractory period during
11 which airway responsiveness to challenge is diminished. Specifically, airway
12 responsiveness to methacholine had been observed to be reduced following exercise
13 (Inman et al.. 1990). A more rapid reversal of methacholine-induced bronchoconstriction
14 had been found following periods of exercise than rest (Freedman et al.. 1988). The
15 refractory period from exercise had also been found to correlate with the responsiveness
16 to methacholine, i.e., individuals who experienced a smaller bronchoconstrictive response
17 following repeated bouts of exercise subsequently also had a smaller response to
18 methacholine challenge (Magnussen et al.. 1986). Recent literature continues to support
19 the possibility that exercise may lead to a period of reduced airway responsiveness. The
20 review by O'Byrne et al. (2009) noted with repeated bouts of exercise, the
21 bronchoconstrictive response to exercise can be abolished in many individuals with
22 asthma. The most probable mechanism explaining this exercise refractory period is the
23 release of inhibitory prostaglandins that partially protect the airways. There may also be
24 changes in eicosanoids associated with NO2 exposure itself (Sections 3.3.2.3 and
25 4.2.4.1). Refractory periods following exercise of 40 minutes to 3 hours has been
26 reported (Dryden et al.. 2010).
27 Controlled NO2 exposure studies may also provide some insight into the effect of
28 exercise on airways responsiveness. Torres and Magnussen (1991) and Strand et al.
29 (1996) provide individual subject PDioo for methacholine and histamine, respectively, on
30 both a control day (no exposure, no exercise) and following a filtered air exposure with
31 exercise. There was a slight tendency for the PD100 to be lower following the filtered air
32 exposures relative to control (no exposure, no exercise) with roughly 53% of the
November 2013 4-17 DRAFT: Do Not Cite or Quote
-------
1 individuals having a lower PD100 following filtered air (with exercise). Thus, these two
2 studies do not support an effect of exercise on AHR in studies evaluating effects of NO2
3 exposure. However, a comparison of two studies that utilized the same challenge agent
4 following the same duration of NO2 exposure and nearly the same exposure
5 concentration does support the conclusion that exercise diminishes the subsequent
6 responsiveness to bronchial challenge. Torres and Magnussen (1990) found a statistically
7 significant increase in airway responsiveness to SO2 in subjects with asthma following
8 exposure to 250 ppb NO2 for 30 minutes at rest; whereas, Rubinstein et al. (1990) found
9 no change in responsiveness to SO2 inhalation following exposure of subjects with
10 asthma to 300 ppb NO2 for 30 minutes with 20 minutes of exercise.
11 Overall, the literature on airway responsiveness supports the development of a refractory
12 period following bouts of exercise. An effect of exercise refractoriness is consistent with
13 greater increases in airway responsiveness following resting than exercising exposures to
14 NO2 as was shown in Table 4-3.
Bronchial Challenge Delivery and Assessment
15 Variations in methods for administering the bronchoconstricting agents may substantially
16 affect the results (Cockcroft and Davis. 2006b: Cockcroft et al.. 2005). A repeated
17 measures study of 55 subjects with asthma evaluating two ATS recommended methods of
18 methacholine delivery found a highly significant (p <0.00001), two-fold difference in
19 PC20 which was attributable to the delivery method (Cockcroft and Davis, 2006b). Even
20 in the same subjects exposed by the same investigators in the same facility to the same
21 bronchial challenge agent, there can be a doubling dose difference due to the delivery
22 method. The difference observed by Cockcroft and Davis (2006b) may, in part, be due to
23 the use of full vital capacity inspirations with breath-hold as part of the delivery
24 technique that yielded the higher PC20. The maximal lung inflations are recognized to
25 induce bronchodilation. The full vital capacity inspiration required for FEVi
26 measurements when assessing airway response to challenge may also cause a partial
27 reversal of bronchospasm versus the use of other measures such as specific airway
28 resistance (sRaw). Variations in the delivery of bronchial challenge agents and methods
29 of assessing airway response may affect comparisons of provocative doses between NO2
30 studies.
Bronchial Challenge Agent
31 Bronchial challenge agents differ in the mechanisms by which they cause
32 bronchoconstriction, acting either "directly" or "indirectly" on bronchial smooth muscle
33 receptors. The asthmatic response to specific allergens may include an early response
November 2013 4-18 DRAFT: Do Not Cite or Quote
-------
1 (within 30 minutes of challenge), which primarily reflects release of histamine and other
2 mediators by airways mast cells as well as a response (typically 3-8 hours post
3 challenge), which reflects enhanced airways inflammation and mucous production
4 (O'Byrne et al.. 2009; Cockcroft and Davis. 2006c). The degree of early airway
5 responsiveness to allergen challenge is not only a function of the concentration of inhaled
6 allergen, but also the degree of sensitization as measured by the level of allergen-specific
7 IgE and responsiveness to nonspecific agents (Cockcroft and Davis. 2006a). Even
8 similarly delivered nonspecific, direct acting agents may differently affect the lung. In a
9 comparison of responses to methacholine and histamine in healthy volunteers not having
10 AHR, Verbanck et al. (2001) reported that histamine caused an overall narrowing of the
11 airways (i.e., similar between parallel lung regions), whereas methacholine caused a
12 differential narrowing of parallel airways which altered ventilation distribution. The
13 differential effects of these two direct acting agents may, in part, be due to their differing
14 target receptors and the distribution of these receptors in the airways (O'Byrne et al.,
15 2009). Comparison of the airway responsiveness between bronchial challenge agents is
16 complicated by the differing mechanisms by which they initiate bronchoconstriction.
Subject Selection
17 Exercise is a major trigger of asthma symptoms in between 60 and 90 percent of people
18 with asthma (Dryden et al.. 2010). In their study of NO2 effects on airway
19 responsiveness, Roger etal. (1990) reported that all their volunteers with asthma
20 experienced either cold air or exercise-induced bronchoconstriction. Morrow and Utell
21 (1989a) reported that, "Many of the asthmatic subjects were unable to undertake the
22 carbachol challenge after either NO2 or air exposures, presumably because of
23 pre-existing exercise-induced bronchoconstriction." Consequently, in their study, data on
24 changes in airway responsiveness were only available for 9 of 20 subjects (see Table
25 4-2). Thus, the existence of exercise-induced bronchospasm and symptoms may have
26 caused an underlying difference in the health status of subjects for which airway
27 responsiveness was evaluated between studies utilizing resting versus exercising
28 exposures.
Medication Usage
29 It is recommended that short-acting bronchodilators be stopped 8 hours before and long-
30 acting bronchodilators 36 hours before the bronchial challenge (Reddel et al.. 2009).
31 Even after withholding salmeterol (a long-acting bronchodilator) for 24 hours, there is
32 still a greater than two-fold reduction in airway responsiveness relative to an unmedicated
33 baseline (Reddel et al.. 2009). There was a wide range in restrictions on asthma
November 2013 4-19 DRAFT: Do Not Cite or Quote
-------
1 medication usage between NO2 studies. For example, Hazuchaetal. (1983) required that
2 subjects not receive steroid therapy or daily bronchodilator therapy for a month prior to
3 bronchial challenge testing. Other studies recorded asthma medication usage and asked
4 subjects to refrain from usage for defined periods of time depending on the medication,
5 such as 8 hours for short-acting bronchodilators (e.g., Witten etal. 2005; Avol et al.
6 1988). Restrictions were far less in some studies, for example, Kleinman et al. (1983)
7 asked subjects to withhold bronchodilators for at least 4 hours prior to exposure, but
8 subjects were not excluded from analysis since usage was generally balanced between
9 filtered air and NO2 exposure days. Still other studies provided no indication of asthma
10 medications or prohibitions for study inclusion (e.g.. Bylin et al., 1988). Pretreatment
11 (500 mg, 4 times per day for 3 days) with ascorbic acid was shown to prevent
12 NO2-induced increases in airway responsiveness of healthy individuals (Mohsenin.
13 1987b). The use of asthma medications or dietary supplements may have affected the
14 ability of studies to identify effects of NO2 on airway responsiveness and may have
15 affected observed provocative doses.
Airway Caliber
16 Bvlin (1993) suggested that NO2 may have a direct effect on airway smooth muscle,
17 possibly relaxing and inducing mild bronchodilation at higher NO2 doses. Consistent
18 with this supposition, statistically significant increases in sRaw following a 20-minute
19 resting exposure to 240 ppb NO2 and significant decreases in sRaw following exposure
20 to 480 ppb NO2 has been reported in healthy individuals (Bvlin etal. 1985). and
21 individuals with asthma exhibited similar trends in sRaw responses to NO2 exposure.
22 Bronchoconstriction shifts the deposition site of challenge agents proximally, whereas
23 bronchodilation shifts the deposition site more distally. Decreasing the surface dose in the
24 bronchi may in turn decrease the airway responsiveness to the challenge.
25 The importance of particle dosimetry (which is affected by factors such as inhaled
26 particle size, airway dimensions, and breathing rates) on airways responsiveness has been
27 investigated by numerous investigators, some of the more conclusive findings are
28 described here. Moss and Oldham (2006) reported a PC of methacholine producing a
29 200% increase in airway resistance in Balb/c mice of 12x lower than in B6C3F1 mice.
30 However, the B6C3F1 mice airways were 1.6x larger and their ventilation was smaller by
31 0.9x than the Balb/c mice. Given these differences in airway size and breathing rate, the
32 estimated dose of methacholine delivered to the airways was equivalent between the
33 species. Wanner etal. (1985) found a strong correlation between the decrease in FEVi
34 following histamine challenge and the estimated dose to the airways of 10 smokers (r =
35 -0.82, p <0.005) and 10 nonsmokers (r = -0.83, p <0.005). In a study of 19 individuals
November 2013 4-20 DRAFT: Do Not Cite or Quote
-------
1 with asthma, Casset et al. (2007) found that the PD20 of house dust mite (RDM) allergen
2 increased with decreasing inhaled particle size from 10 urn to 1 um (mass median
3 aerodynamic diameter). These studies demonstrate lower airway responsiveness for distal
4 versus proximal deposition of challenge agents. Thus, these studies are supportive of the
5 supposition proposed by Bylin (1993).
6 The recent review by Brannan and Lougheed (2012) has also specifically identified
7 reduced airway caliber as a predictor of airway responsiveness. However, other studies
8 not described here have concluded that airway caliber was not a predictor of airway
9 responsiveness. Although this may seem counter to the dosimetric discussion above,
10 simply considering airway caliber may not adequately capture the complexity and
11 anatomical heterogeneity of lung disease from asthma. In a comparison of individuals
12 with asthma and healthy controls, Laube etal. (1992) reported that increasing
13 heterogeneity in particle deposition was significantly associated with decreasing PD20 to
14 methacholine. Heterogeneity in deposition is, in part, due to heterogeneity in ventilation
15 distribution. In another study of individuals with asthma, Downie et al. (2007) found
16 heterogeneity in ventilation distribution to be a predictor of airway responsiveness
17 independent of airway inflammation and airway caliber.
18 The literature more strongly suggests an effect of the surface dose of challenge agents to
19 the conducting airways on airways responsiveness than as a function of airway caliber,
20 per se. The dose of bronchial challenge agents to the conducting airways may have been
21 affected by numerous factors within and between studies evaluating the effect of NO2 on
22 airway responsiveness. Although it is clear that such factors could contribute to
23 variability within and between studies, the available information is insufficient to support
24 an effect such as decreased airway responsiveness at higher NO2 concentrations due to
25 bronchodilation.
Effect of Time of Challenge Post-exposure
26 With respect to the data in Table 4-1 and Table 4-2. bronchial challenges were delivered
27 an average of 60 minutes post-exposure. For nonspecific agents, on average, challenges
28 were delivered 16 minutes following resting exposures and 67 minutes following exercise
29 exposures (p <0.01). Although challenges may take upwards of 40 minutes to complete
30 (Mohsenin. 1987a). the difference in the time when challenge agents were delivered
31 could plausibly affect differences in airway responsiveness among studies.
32 Strand etal. (1996) exposed exercising adults with asthma to 260 ppb NO2 for 30
33 minutes. Responsiveness to histamine was assessed at 30-minutes, 5-hours, 27-hours, and
34 7-days post-exposure. The PDi00 tended (p = 0.08) to decrease after 30 minutes, became
November 2013 4-21 DRAFT: Do Not Cite or Quote
-------
1 significantly decreased by 5 hours (p = 0.03), and returned to baseline by 27-hours post
2 NO2 exposure compared to filtered air. Although the PDioo following NO2 exposure was
3 fairly constant between 30 minutes and 5 hours, the PD10o following filtered air was
4 increased at the 5-hour time point which may have contributed to the significant
5 difference between NO2 and filtered air after 5 hours. This 5-hour time point is just
6 beyond reported refractory periods following exercise of 40 minutes to 3 hours (Dryden
7 et al., 2010). A comparison across other NO2 studies of human subjects for an effect of
8 challenge delivery timing is not possible due to differences in NO2 concentration and
9 exposure duration. Silbaugh et al. (1981) found a rapid return to baseline responsiveness
10 in guinea pigs by two hours post exposure.
11 Although there is strong evidence for a refractory period following exercise, the existing
12 data on airway responsiveness following NO2 exposure are insufficient to assess the
13 influence of challenge delivery timing on airway responsiveness in those studies.
Effect of Repeated NO2 Exposures
14 To mimic a daily commute, Strand et al. (1998) exposed adults with asthma on four
15 sequential days to either filtered air or 260 ppb NO2 for 30 minutes during rest. The early
16 phase response to allergen challenge was significantly increased by NO2 exposure. The
17 allergen-induced fall in FEVi for the 4 days was, on average, -2.5 versus -0.4% after air
18 (p = 0.018). The late phase response to allergen challenge was also significantly greater
19 after NO2 with an average decrement in FEVi of-4.4 versus -1.9% after air (p = 0.009)
20 for the 4 days. This study suggests that the effect of NO2 exposure on airway
21 responsiveness to allergen challenge is relatively constant over several contiguous days
22 of repeated NO2 exposure.
Extraneous Factors
23 Although some early studies progressively increased NO2 exposure concentrations for
24 safety purposes, the majority of controlled human exposure studies investigating the
25 effects of NO2 are of a randomized, controlled, crossover design in which subjects were
26 exposed, without knowledge of the exposure condition and in random order to clean
27 filtered air (the control) and, depending on the study, to one or more NO2 concentrations.
28 The filtered air control exposure provides an unbiased estimate of the effects of the
29 experimental procedures on the outcome(s) of interest. Comparison of responses
30 following this filtered exposure to those following NO2 exposure allows for estimation of
31 the effects of NO2 itself on an outcome measurement while controlling for independent
32 effects of the experimental procedures. Furthermore, the studies by Hazucha et al. (1983)
33 and Strand et al. (1997) provided AHR data at the time of enrollment in their study and
November 2013 4-22 DRAFT: Do Not Cite or Quote
-------
1 AHR data following resting exposures to filtered air. Little to no discernible change was
2 observed between AHR at inclusion and following the resting exposure which suggests
3 that experimental procedures (other than exposure to NO2) did not affect AHR. In the
4 study by Jenkins et al. (1999). although the average PD20 to HDM were similar following
5 a 3- and 6-hour filter air exposure with exercise, 5 of 10 subjects had greater than a 2x
6 difference in their PD2o between the two air exposures. Unfortunately, this study does not
7 allow the contribution of daily variability in airway responsiveness versus an effect of
8 exposure duration to be discerned.
Dose-response
9 Folinsbee (1992) noted that greater NO2 doses occur with exercise due to both the
10 increased ventilation rates and a tendency for increased exposure duration. However, in
11 his meta-analyses, the effects of NO2 exposure on airway responsiveness were found
12 following resting, but not exercising exposures to NO2. The lack of a clear dose-response
13 relationship may suggest that some factors cause a diminution of responses at higher
14 versus lower intake doses.
15 The dose-response of NO2 on airway responsiveness may be modulated by a number of
16 factors that have been described in this section. The finding of greater airway
17 responsiveness following exposures at rest than exercise, despite a lower intake dose of
18 NO2 during the resting exposures, is consistent with an effect of exercise refractoriness.
19 Issues related to subject selection and medication may have reduced observed effects of
20 NO2 on airway responsiveness and contributed to variability within and among studies.
21 The choice of bronchial challenge agent and method of delivery each also would have
22 likely contributed to variability between studies. Limited evidence also suggests airway
23 dilation at higher intake doses could reduce airway responsiveness. Overall, the effects of
24 exercise refractoriness and potential for some individuals with asthma with exercise-
25 induced bronchoconstriction to be excluded from the evaluation of airway responsiveness
26 appear to be the most likely contributors to not readily finding effects of NO2 on airway
27 responsiveness at higher intake doses occurring with exercise. Other methodological
28 differences, if randomly occurring, between studies such as the choice of challenge
29 agents, challenge delivery method, physiological endpoint used to quantify airway
30 responses, severity of disease, and asthma medication usage would likely add variability
31 to assessment of airway responsiveness and thereby bias data toward the null of no
32 discernible dose-response.
33 A few studies have investigated the effects of NO2 exposure on airway responsiveness at
34 more than one concentration. Intra-study evaluation of a potential dose-response reduces
35 the inherent variability and uncertainty occurring with inter-study comparisons. Bylin et
November 2013 4-23 DRAFT: Do Not Cite or Quote
-------
1 al. (1988) found statistically significant effects of NO2 on airway responsiveness at 270
2 ppb, but not 140 ppb. Orehek et al. (1976) provided responsiveness data for four
3 individuals following exposure to both 100 and 200 ppb NO2. Of these four individuals,
4 three had similar PDi00 between the two exposures, one individual had a doubling
5 difference in the PD100 (0.42 mg at 200 ppb versus 0.94 mg at 100 ppb). Tunnicliffe et al.
6 (1994) found a significant and larger increase in airway responsiveness at 400 ppb as
7 compared to tendency for increased responsiveness at 100 ppb. These three studies
8 (Tunnicliffe et al.. 1994; Bylinetal.. 1988; Orehek etal.. 1976). for resting exposure to
9 NO2 are supportive of increasing airways responsiveness with increasing NO2
10 concentration in individuals with asthma. The dose-response evidence from studies that
11 used exercising protocols is less compelling. Roger etal. (1990) did not find a change in
12 airway responsiveness at either 150 or 300 ppb NO2. Jenkins et al. (1999) found
13 significant increases in airway responsiveness to allergens following a 3-hour exposure to
14 400 ppb NO2, but not following a 6-hour exposure to 200 ppb NO2 despite equivalence
15 in terms of the total intake dose (concentration x exposure duration).
16 Several inter-study differences likely contribute to variability and uncertainty in cross
17 study comparisons of provocative dose and lung function response to bronchial challenge
18 agents. Evaluation of the proportional change in these outcomes following NO2 and
19 filtered air exposure as performed by Goodman et al. (2009) should allow for a valid
20 comparison across studies since the air control would, theoretically, adjust for many
21 methodological differences between studies. However, even after this adjustment, clear
22 differences between resting and exercising exposures exist, presumably because exercise
23 itself causes real effects on airway responsiveness. It may not be possible to adequately
24 remove the influence of some methodological factors such as exercise that so
25 substantially affect the airways or the determination of airway responsiveness in
26 individuals with asthma. Thus, it is not clear to what extent inter-study assessments of the
27 dose-response relationship between NO2 exposure and airway responsiveness are
28 affected by methodological biases of studies. The few studies having evaluated effects at
29 multiple NO2 concentrations, especially those using resting exposure, are supportive of a
30 dose-response relationship showing increasing airway responsiveness with increasing
31 NO2 exposure concentration.
4.2.2.4 Summary of Airway Hyperresponsiveness
32 There is a wide range of airway responsiveness influenced by many factors, including
33 exercise, medications, cigarette smoke, air pollutants, respiratory infections, disease
34 status, and respiratory irritants. The airway responsiveness of individuals with asthma is
35 generally greater than in healthy age-matched controls; although, the airway
November 2013 4-24 DRAFT: Do Not Cite or Quote
-------
1 responsiveness of those with asthma extends into the normal range. Nonspecific
2 bronchial challenge agents causing bronchoconstriction may act directly (i.e., histamine,
3 carbachol, and methacholine) on airway smooth muscle receptors or act indirectly (i.e.,
4 exercise, cold air) though intermediate pathways, especially via inflammatory mediators.
5 Specific allergens also act indirectly on smooth muscle to initiate bronchoconstriction.
6 Likely affecting the observed changes in airway responsiveness due to NO2 exposure,
7 there are methodological differences between NO2 studies including subject activity level
8 (rest versus exercise) during NO2 exposure, asthma medication usage, choice of airway
9 challenge agent (e.g., direct and indirect nonspecific stimuli), method of administering
10 the bronchoconstricting agents, and physiological endpoint used to assess airway
11 responsiveness. These intra-study differences likely contribute to considerable variability
12 and uncertainty in comparison of factors such as the provocative dose and lung function
13 response to bronchial challenge agents. Studies that evaluated effects of NO2 in subjects
14 with asthma at multiple NO2 concentrations under resting conditions generally show
15 increasing airway responsiveness with an increase in exposure concentration.
16 Controlled human exposure studies have shown significant effects of NO2 exposure on
17 airway responsiveness in both healthy individuals and those with asthma. In healthy
18 individuals, increases in nonspecific airway responsiveness were observed in the range of
19 1,500 to 2,000 ppb NO2 for 3-hour exposures. In those with asthma, statistically
20 significant effects on responsiveness to nonspecific challenge were reported following
21 exposures as low as 100 ppb NO2, although most studies showing significant effects were
22 in the range of 300 ppb NO2 or greater. Enhanced airway responsiveness to allergens in
23 asthmatics was found at exposures of as low as 260 ppb for 30 minutes. Evidence to
24 describe key events to inform modes of action for NO2-induced AHR include the effects
25 of NO2 on bronchial smooth muscle reactivity (Section 3.3.2.5) and innate/adaptive
26 immunity (Section 3.3.2.6). Given the methodological differences between studies,
27 several meta-analyses including those in Table 4-3. Table 4-4. and Table 4-5 have
28 assessed the fraction of individuals experiencing a change in airway responsiveness. In
29 individuals exposed to NO2 at rest, increases in nonspecific airway responsiveness (Table
30 4-3) occur in the range of 200 and 300 ppb NO2 for 30-minute exposures and at 100 ppb
31 NO2 for 60-minute exposures in individuals with asthma.
4.2.3 Lung Function
32 Compared with evidence for AHR, the 2008 ISA for Oxides of Nitrogen reported weak
33 evidence for the direct effects of NO2 exposure on changes in lung function, particularly
34 in controlled human exposure studies and epidemiologic studies of adults (U.S. EPA.
November 2013 4-25 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
2008c). The evidence was weak in healthy adults and those with asthma or chronic
obstructive pulmonary disease (COPD) alike. In previous epidemiologic studies, the most
robust evidence comprised associations in children in the general population between
increases in ambient NO2 concentration and decrements in lung function as measured by
supervised spirometry. Evidence in children with asthma was based on unsupervised lung
function measurements and was inconsistent. Most recent studies were epidemiologic and
supported associations between ambient NO2 concentrations and lung function
decrements in children with asthma and children in the general population.
9
10
11
12
13
14
15
16
17
4.2.3.1 Epidemiologic Studies
Collectively, previous and recent studies found associations between increases in ambient
NO2 concentrations and decrements in supervised spirometry measures (primarily FEVi)
in children with asthma and children in the general population. Across the various
populations examined, results are less consistent for lung function measured under
unsupervised conditions, primarily peak expiratory flow (PEF) at home. Most results
indicate lung function decrements in association with NO2; associations were
inconsistent for NO and NOX. Ambient concentrations of oxides of nitrogen, locations,
and time periods for epidemiologic studies of lung function are presented in Table 4-6.
Table 4-6 Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Study3
Greenwald et al.
(2013)
Holquinetal. (2007)
Martins et al. (2012)
Spira-Cohen et al.
(2011)
Location
El Paso, TX
Ciudad Juarez,
Mexico
Viseu,
Portugal
Bronx, NY
Study Period
Mar-June 2010
2001-2002
Jan and June,
2006 and 2007
Spring 2002,
Spring/Fall
2004,
Spring 2005
Exposure
Metric
Analyzed
96-h avg NO2
1-weekavg NO2
1 -week avg
N02b
6-h avg NO2
(9 a.m.-3 p.m.)
Mean/Median
Concentration
(PPb)
School A: 6.5
School B: 17.5
18.2
Across
4 periods:
4.5, 3.5, 9.8,
8.2C
NR
Upper Percentile
Concentrations
(PPb)
NR
-
Max across
4 periods:
4.6,4.0, 10.9, 9.4C
NR
November 2013
4-26
DRAFT: Do Not Cite or Quote
-------
Table 4-6 (Continued): Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Study3
Liu et al. (2009b)
Dales et al. (2009a)
Barraza-Villarreal et
al. (2008)
Hernandez-Cadena
et al. (2009)
Delfino et al.
(2008a)
Mortimer et al.
(2002)
O'Connor et al.
(2008)
Odaiima et al.
(2008)
Gillespie-Bennett et
al. (2011)
Wiwatanadate and
Trakultivakorn
(2010)
Just et al. (2002)
Yamazaki et al.
(2011)
McCreanor et al.
(2007)
Location
Windsor, ON,
Canada
Mexico City,
Mexico
Mexico City,
Mexico
Riverside, CA
Whittier, CA
Bronx and East
Harlem, NY;
Chicago, IL;
Cleveland, OH;
Detroit, Ml;
St. Louis, MO;
Washington, DC
Boston, MA;
Bronx, NY;
Chicago, IL;
Dallas, TX;
New York, NY;
Seattle, WA;
Tucson, AZ
Fukuoka,
Japan
Bluff, Dunedin,
Christchurch, Porirua,
Hutt Valley,
New Zealand
Chiang Mai,
Thailand
Paris, France
Yotsukaido,
Japan
London,
U.K.
Study Period
Oct-Dec 2005
June 2003-
June2005
May-Sept 2005
July-Dec 2003
July-Dec 2004
June-Aug 1993
Aug 1998-
July2001
Apr-Sept 2002
Oct 2002 -
Mar 2003
Sept 2006
Aug 2005-June
2006
Apr-June 1996
Oct-Dec 2000
Nov-March
2003-2005
Exposure
Metric
Analyzed
24-h avg NO2
8-h max NO2
1-h max NO2
24-h avg
personal NO2
24-h avg central
site NO2
4-h avg NO2
(6 a.m.-10 a.m.)
24-h avg NO2
3-h avg
' (7 p.m. -10 p.m.)
NO2
4-week avg NO2
24-h avg NO2
24-h avg NO2
1-h avg
(6 p.m. -7 p.m.)
NO2
2-h avg
(10:30 a.m. -
12:30 p.m.)
NO2
Mean/Median
Concentration
(PPb)
19.8
37.4
57
28.6
25.0
NR
NR
20.0
11.0
3.9
17.2
28.6C
32.6
Oxford St:
75.5C
Hyde Park:
11. 5C
Upper Percentile
Concentrations
(PPb)
95th: 29.5
Max: 77.6
75th: 69
Max: 116
Max: 105.7
Max: 29.2
NR
NR
Max: 51.3
Max: 49.0
NR
90th: 26.5
Max: 37.4
Max: 59.0C
Max: 154C
Max: 77.7C
November 2013
4-27
DRAFT: Do Not Cite or Quote
-------
Table 4-6 (Continued): Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Study3
Qian et al. (2009a)
Silkoffetal. (2005)
Harreetal. (1997)
Peacock et al.
(2011)
Canova et al. (2010)
Wiwatanadate and
Liwsrisakun (2011)
Laqorio et al. (2006)
Maestrelli et al.
(2011)
Hiltermann et al.
(1998)
Hiqqins et al.
(2000): Hiqqins et
al. (1995)
Park et al. (2005)
Steerenberq et al.
(2001)
Linnetal. (1996)
Moshammer et al.
(2006)
Oftedal et al. (2008)
Chanq et al. (2012)
Location
Boston, MA
Denver, CO
Madison, Wl
New York City, NY
Philadelphia, PA
San Francisco, CA
Denver, CO
Christchurch,
New Zealand
London,
U.K.
Padua,
Italy
Chiang Mai,
Thailand
Rome,
Italy
Padua,
Italy
Bilthoven,
the Netherlands
Widnes, Runcorn,
U.K.
Incheon,
Korea
Utrecht,
the Netherlands
Bilthoven,
the Netherlands
Upland, Rubidoux,
Torrance, CA
Linz,
Austria
Oslo,
Norway
Taipei,
Taiwan
Study Period
Feb1997-Jan
1999
Winters
1999-2000
2000-2001
June-Aug 1994
Oct1995-
Oct1997
Summer/Fall
2004,
Winter/Summer
/Fall 2005
Aug 2005-June
2006
May-June,
Nov-Dec 1999
1999-2003
July-Oct 1995
Aug, year NR
Mar-June 2002
Feb-Mar1998
School yr,
1992-1994
School yr,
2000-2001
Nov2001-Dec
2002
Dec 1996-May
1997
Exposure
Metric
Analyzed
24-h avg NO2
24-h avg NO2
24-h avg NO2
1-h max NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO
24-h avg NO2
24-h avg NO
24-h avg NO2
8-h avg NO2
(12-8 a.m.)
24-h avg NO2
6-day avg NO2
Mean/Median
Concentration
(PPb)
20.8
16
29
NR
51.4
27.2C
17.2
37.6C
Across
seasons and
years:
20.9-37.0C
11. 2C
NR
Control days:
31.6
Dust days: 20.7
28.2C
30.2C
25.5C
7.4C
33
9.3C
14.4C
31.8
Upper Percentile
Concentrations
(PPb)
75th: 25.5
Max: 60.7
75th: 30, Max: 54
75th: 36, Max: 54
NR
75th: 56
48. 1C
90th: 26.5
Max: 37.4
Max: 54. 3C
Range of 75th:
23.0-42.5C
22. 5C
Max: 44. T
Max: 44. T
Max: 168C
Max: 49.5C
Max: 85.6C
Max: 96
75th: 11. 4C
Max: 59.2C
75th: 41. 7
November 2013
4-28
DRAFT: Do Not Cite or Quote
-------
Table 4-6 (Continued): Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Study3
Castro et al. (2009)
Baqheri Lankarani
etal. (2010)
Eenhuizen et al.
(2013)
Peacock et al.
(2003)
Scarlett et al. (1996)
Timonen and
Pekkanen (1997)
Ranzi et al. (2004)
Ward et al. (2000)
van der Zee et al.
(2000): van der Zee
etal. (1999)
Roemer et al.
(1998)
Straketal. (2012)
Weichenthal et al.
(2011)
Thaller etal. (2008)
Schindler et al.
(2001)
Cakmak et al.
(2011 a)
Location
Rio de Janeiro,
Brazil
Tehran,
Iran
3 study areas,
the Netherlands
Rochester upon
Medway, U.K.
Surrey, U.K.
Kuopio,
Finland
Emiglia-Romagna,
Italy
West Midlands,
U.K.
Rotterdam,
Bodegraven/Reeuwij,
Amsterdam, Meppel,
Nunspeet,
the Netherlands
Multiple locations:
Sweden, Finland,
Norway,
the Netherlands,
Germany,
Czech Republic,
Hungary, Italy, Greece
Bilthoven,
the Netherlands
Ottawa, ON,
Canada
Galveston, TX
Aarau, Basel, Davos,
Geneva, Lugano,
Montana, Payerne,
Wald, Switzerland
14 Canadian cities
Study Period
May, June,
Sept, Oct 2004
NR
Oct 2000-
Nov 2001
Nov1996-
Feb 1997
June-July 1994
Feb-Apr1994
Feb-May 1999
Jan-Mar 1997
May-July 1997
Winter
1992-1993
Winter
1993-1994
Winter
1994-1995
Winter
1993-1994
Mar-Oct 2009
NR
Summers 2002,
2003, 2004
NR
Mar 2006-
Mar2007
Exposure
Metric
Analyzed
24-h avg NO2
24-h avg NO2
24-h NO
24-avg NOX
24-h avg NO2
24-h avg NO2
1-h max NO2
1-h max NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
5-h avg NOx
5-h avg NO2
1-h avg NO2
24-h avg NO2
1-h max NO2
24-h avg NOX
1-h max NOx
24-h avg NO2
24-h avg NO2
Mean/Median
Concentration
(PPb)
49.2C
75.5, 17.6C
51.6, 40.4C
72.9, 38.8C
16.0C
17.4, 17.1, 19.2
28.5,28.1, 31.8
34.9
Urban: 14.9C
Suburban: 7.4C
Urban: 37.0C
Rural: 18.51C
NR
27.1, 17.6C
25.5, 13.3C
25.0, 11. T
Across
locations:
6.7-39.8c
36
20
High traffic: 4. 8
Low traffic: 4.6
1.2
3.2
1.3
3.6
19.5C
12.6
Upper Percentile
Concentrations
(PPb)
Max: 115C
Max: 119, 25.5C
Max: 85.1, 11 Oc
Max: 122, 94.7C
75th: 23.2C
Max: 47.9C
Max: 39, 39, 43
Max: 67, 71, 98
Max: 82
Max:41.5c
Max:27.1c
NR
NR
NR
Max: 50, 44.2C
Max: 40.4, 28.7C
Max: 43.6, 30.3C
Max: 96
Max: 34
Max: 1 1
Max: 10
Max: 7.1
Max: 27.7
Max: 7.4
Max: 36.6
69.3C
95th: 29.4
November 2013
4-29
DRAFT: Do Not Cite or Quote
-------
Table 4-6 (Continued): Mean and upper percentile oxides of nitrogen concentrations in
epidemiologic studies of lung function.
Study3
Steinvil et al. (2009)
Sonetal. (2010)
Location
Tel Aviv,
Israel
Ulsan,
Korea
Exposure Mean/Median
Metric Concentration
Study Period Analyzed (ppb)
Sept 2002- 24-havgNO2 19.3
Nov 2007
2003-2007 24-havgNO2 21.4
Upper Percentile
Concentrations
(PPb)
75th: 25.3
Max: 59.9
75th: 26.1
Max: 44.8
"Studies presented in order of first appearance in the text of this section.
bSubject-level exposure estimates calculated from outdoor NO2 at schools and other locations plus time activity patterns.
""Concentrations converted from ug/m3 to ppb using the conversion factor of 0.532 assuming standard temperature (25 °C) and
pressure (1 atm).
NR = not reported.
Children with Asthma
1 In contrast with studies reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
2 2008c). several recent studies of children with asthma conducted spirometry under
3 supervised conditions, and most indicated a relationship with short-term NO2 exposure
4 (Figure 4-1 and Table 4-7). Studies of supervised spirometry measured lung function
5 weekly, biweekly, or seasonally. Evidence for lung function measured daily by subjects
6 at home was less consistent. Among these studies, some reported an association with
7 NO2 (Gillespie-Bennett et al.. 2011; Delfino et al.. 2008a; O'Connor et al.. 2008).
8 whereas others did not (Wiwatanadate and Liwsrisakun. 2011; Odajima et al.. 2008; Just
9 et al.. 2002; Mortimer et al.. 2002). Results were inconsistent between U.S. multicity
10 studies (NCICAS, ICAS) (O'Connor et al.. 2008; Mortimer etal.. 2002). However,
11 several studies that reported no association with home lung function measurements did
12 not provide quantitative results, including NCICAS (Odajima et al.. 2008; Just et al..
13 2002; Mortimer et al.. 2002). Thus, assessing the relative magnitude and precision of
14 their results was not possible. A relationship between ambient NO2 and PEF was
15 indicated in children diagnosed with asthma in a recent meta-analysis ("Weinmayr et al..
16 2010) that included mostly European studies as well as some studies reviewed in the
17 2008 ISA for Oxides of Nitrogen.
18 With respect to the populations examined, most studies assessed asthma as parental
19 report of physician-diagnosed asthma. Children were recruited mostly from schools,
20 supporting the likelihood that study populations were representative of the general
21 population of children with asthma. Study populations represented a range of asthma
22 severity, as ascertained by Global Initiative for Asthma guidelines or medication use, ED
23 visit, or hospital admission for asthma in the previous year. Based on a priori hypotheses,
24 results did not demonstrate larger NO2-associated decrements in lung function in children
November 2013 4-30 DRAFT: Do Not Cite or Quote
-------
1 with asthma than children without asthma (Barraza-Villarreal et al.. 2008; Holguin et al..
2 2007). Post-hoc analyses pointed to stronger associations among children with asthma not
3 taking ICS (Hernandez-Cadena et al.. 2009; Liu et al.. 2009b) or not taking controller
4 bronchodilators (Delfino et al.. 2008a). The limited results for larger associations in ICS
5 nonusers together with observations for NO2-associated lung function decrements in
6 populations with high prevalence of atopy (53-100%) (Martins et al.. 2012; O'Connor et
7 al., 2008; Holguin et al., 2007) are supported by findings for NO2-induced increases in
8 allergic inflammation (Section 4.2.4.3) and findings for mast cell degranulation (which
9 leads to histamine release) in mediating NO 2 -induced lung function decrements (Section
10 3.3.2.2). Bronchodilator use has been shown to reduce AHR in response to a challenge
11 agent (Section 4.2.2.3).
12 For children with asthma, key evidence for NO2-associated lung function decrements was
13 provided by studies with strong exposures assessment characterized by personal
14 monitoring (Delfino et al.. 2008a). modeling outdoor measurements at school and other
15 locations with time-activity data (Martins et al., 2012) or outdoor school monitoring
16 (Greenwald et al.. 2013; Spira-Cohen et al.. 2011; Holguin et al.. 2007). These studies
17 examined limited lags of NO2 exposure and were similar in finding associations with
18 multiday (i.e., lag 0-1 avg, 0-4 avg) averages of 24-h avg NO2. Studies that measured or
19 modeled personal exposures provided evidence of an effect of outdoor NO2 on lung
20 function. Among children with asthma in the Los Angeles, CA area, slightly larger
21 decrements in % predicted FEVi were found with personal NO2 (-1.5 [95% CI: -2.3,
22 -0.57] per 20-ppb increase in lag 0 day NO2) than central site NO2 (-1.3 [95% CI: -2.4,
23 -0-15]) (Delfino et al., 2008a). But, a Spearman correlation of 0.43 between personal and
24 central site NO2 indicated that ambient NO2 had some influence on personal exposures.
25 Among children with wheeze in Portugal, indoor school and home NO2 concentrations
26 were below the limit of detection (Martins et al.. 2012). and time-weighted average of
27 microenvironmental NO2 concentrations have shown agreement with personal NO2
28 (Section 2.6.5.2).
29 Among studies of outdoor school NO2, associations with FEVi were found with lag 0-6
30 day avg NO2 in populations in El Paso, TX, and Ciudad Juarez, Mexico, which are
31 located along the U.S./Mexico border (Greenwald et al.. 2013; Holguin et al.. 2007)
32 (Figure 4-1 and Table 4-7). Between two El Paso schools, associations were limited to
33 the school characterized by a larger percentage of Mexican-American children, higher
34 BMI, and higher outdoor pollutant concentrations (Greenwald et al.. 2013). No
35 association with FEVi was found in children with asthma in Bronx, NY with school NO2
36 averaged over the 6-h school day (Spira-Cohen et al.. 2011). An effect of outdoor NO2
37 was indicated by similar FEVi decrements for outdoor and indoor NO2 in an El Paso
38 school (Greenwald et al.. 2013) and larger lung function decrements for home outdoor
November 2013 4-31 DRAFT: Do Not Cite or Quote
-------
than indoor NO2 among children in five New Zealand towns (Gillespie-Bennett et al.,
2011). The latter results have weaker implications as multiple daily lung function
measures were related to a single 4-week average of NO2.
Study
Greenwald et al. (In
press)
NO2 Exposure
Metrics
Subgroup
lag 0-6 avg, 24-h avg School A
School B
Holguin et al. (2007) lag 0-6 avg, 24-h avg
Martins et al. (2012) lag 0-6 avg, 24-h avg
Spira-Cohen et al.
(2011)
Liuetal. (2009)
Barraza-Villarreal et
al.(2008)
lag 0, 6-h avg
lag 0-3 avg, 24-h avg
lag 1-4 avg, 8-h max Asthma
No Asthma
Delflno et al. (2008) lag 0, 24-h avg
-30.0-25.0-20.0-15.0-10.0-5.0 0.0 5.0 10.0
Percent change in FE\A, per 20 or 25 ppb increase in NO2 (95% Cl)a
All subjects
No Bronchodilator
Bronchodilator
Dales etai. (2009)
lagO, 12-h avg
-30,0-25.0-20.0-15.0-10.0 -5.0 0.0 5.0 10.0
Change in % predicted FEV1 per 20 or 25 ppb increase in NO2 (95% Cl)a
Note: All results are from recent studies and are organized by population examined and then generally in order of decreasing study
strength (e.g., exposure assessment method, potential confounding considered).
"Effect estimates are standardized to a 20-ppb increase for 24-h avg NO2, and a 25-ppb increase for 6-h to 12-h avg or 8-h max
NO2. Study details and quantitative results are reported in Table 4-7.
Figure 4-1 Associations between ambient or personal NO2 concentrations
and FEV1 in children with asthma.
November 2013
4-32
DRAFT: Do Not Cite or Quote
-------
Table 4-7
Study
Children with
Greenwald et
al. (201 3)t
Holquin et al.
(2007)1
Epidemiologic studies of lung function in children and adults with respiratory disease.
Study Population and
Methodological Details
Asthma
El Paso, TX
N = 38, mean age 10 yr, 76% Mexican-
American
Repeated measures. Supervised
spirometry. Examined weekly for 13
weeks. 413-441 observations.
Recruitment from schools in low and high
traffic area. No information on participation
rate. School record of physician-
diagnosed asthma. GLM with subject as
random effect and adjusted for potential
confounding by school, temperature,
relative humidity, indoor NO.
Ciudad Juarez, Mexico
N = 194, ages 6-12 yr, 78% mild,
intermittent asthma, 58% with atopy.
Repeated measures. Supervised
spirometry. Examined biweekly for 4 mo.
87% participation. Self-report of physician-
diagnosed asthma. Linear and nonlinear
mixed effects model with random effect for
subject and school adjusted for sex, BMI,
day of week, season, maternal and
paternal education, passive smoking
exposure.
Subgroup Effect Estimate
Exposure Metrics Lag day Analyzed (95% Cl)
Analyzed Analyzed (if applicable) Single-Pollutant Model3
FEV-i:
NO2-School 0-4 avg School A 2.3% (-15, 24%)
outdoor SchoolB -17% (-32, 0.12%)
NO2-School indoor School A 38% (-12, 1 16%)
SchoolB -14% (-32, 7.2%)
All 24-h avg
NO2-School 0-6 avg FEV-i:
outdoor Asthma, n = 31 -1.2% (-2.5, 0.06%)
24-h avg
Homes 397 meters
from schools
Copollutant
Examination
No copollutant model
BC, SO2 (central site)
associated with FEV-i.
- Moderate correlation
with NO2 (Pearson r =
0.62, -0.22).
School BTEX
associated with FEV-i ,
highly correlated (r =
0.77).
No copollutant model.
No association with
PM2.5, EC.
Weak to moderate
correlations with NO2.
Spearman r = 0.30 for
PM2.5, 0.49 for EC.
Road density at home
not school associated
with lung function.
November 2013
4-33
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study Population and
Study Methodological Details
Martins et al. Viseu, Portugal
(2012)}: N = 51, mean age 7.3 (SD: 1.1)
Subgroup Effect Estimate
Exposure Metrics Lag day Analyzed (95% Cl) Copollutant
Analyzed Analyzed (if applicable) Single-Pollutant Model3 Examination
NO2-Subject-level 0-6 avg8
yr, 53% 24-h avg
FEV-i:
-24% (-45, -2.8%)
ForFEV-i:
w/PM-io: -31%
(-96,
with atopy.
Repeated measures. Supervised
spirometry. 4 measurements over 2
different seasons. Recruitment from urban
and suburban schools. -66% participation
rate. Parental report of wheeze in previous
12 mo. GEE adjusted for age, sex,
parental smoking, parental education,
atopy, time of visit, average temperature,
relative humidity. Also included height,
weight, older siblings, mold or dampness
in home, fireplace in home, pets in home
because changed at least 1 pollutant
effect estimate >10%.
Estimated from
school outdoor
NO2, 20 city
monitors,
MM5/CHIMERE
modeling, and daily
activity patterns
FEV-i/FVC:
-11% (-21, 0.49%)
FEF 25-75%^
-39% (-71,-6.0%)
FEV-i after bronchodilator:
18% (3.4, 32%)
spirometry. Examined daily for 1 mo. 454
observations. Recruitment from schools
by referrals from school nurses. Parental
report of physician-diagnosed asthma.
Mixed effects model with subject as
random effect adjusted for height, sex,
temperature. Adjustment for school
(indicator of season) did not alter results.
89% time indoors.
Most children walk
to school.
35%)
w/benzene: -3.7%
(-33, 25%)
w/ethylbenzene: -18%
(-50, 14%)
Benzene robust to
NC>2 adjustment,
PM-io and
ethylbenzene
attenuated.
Correlations negative
or weakly positive.
Spearman r = -0.72 to
-0.55 for PM-io, -0.43
to0.14forVOCs.
Spira-Cohen et Bronx, NY
al- (2011)4: N = 40, ages 10-12 yr, 100% nonwhite,
44% with asthma ED visit or hospital
admission in previous 12 mo.
NO2-school 0
6-h avg
(9 a.m.-3 p.m.)
FEV-,:
0.23% (-1.6, 2.1%)
PEF:
0.92% (-1.0, 2.8%)
Personal EC
associated with lung
function and was
robust to NO2
adjustment.
November 2013
4-34
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Liu et al.
(2009btt
Dales et al.
(2009a)t
Study Population and
Methodological Details
Windsor, ON, Canada
N = 182, ages 9-14 yr
Repeated measures. Supervised
spirometry. Examined weekly for 4 weeks,
same day of week. 672 observations.
Recruitment from schools. No information
on participation rate. Parental report of
physician-diagnosed asthma. Mixed effect
model with random effect for subject and
adjusted for testing period, temperature,
relative humidity, daily medication use.
Windsor, ON, Canada
N = 182, ages 9-14 yr
Repeated measures. Same cohort as
above. Unsupervised peak flow.
Examined daily for 4 weeks, same day of
week. 672 observations. Recruitment from
schools. No information on participation
rate. Parental report of physician-
diagnosed asthma. Mixed effect model
with random effect for subject and
adjusted for sex, testing period, day of
week, daily mean temperature, relative
humidity, time spent outdoors.
Subgroup
Exposure Metrics Lag day Analyzed
Analyzed Analyzed (if applicable)
NO2-Central site 0
24-h avg
Average of 2 sites.
Most subjects live
within 10 km of 0_3
sites.
NO2-Central site 0
12-h avg
(8 a.m.-8 p.m.)
Average of 2 sites.
Most subjects live
within 10 km of sites
Mean 1.6 and 2.2
h/day spent
outdoors.
Effect Estimate
(95% Cl)
Single-Pollutant Model3
FEV-,:
-1.2% (-3.2, 0.84%)
FEF 25-75%:
-4.8% (-8.6, -0.94%)
FEV-,:
-2.3% (-5.5, 0.92%)
FEF 25-75%:
-8.0 (-14, -1.6%)
Evening % predicted
FEV-,:
-0.23 (-1.1, 0.59)
Diurnal change FEV-i:
-0.69% (-1.3, 0.07%)
Copollutant
Examination
FEV-i w/PM25: 1.2%
(-3.8, 6.4%)
FEV-i w/SO2: -2.0%
(-6.9, 3.1%)
PM2.5 association
robust to NO2
adjustment, SO2
attenuated. Spearman
r = 0.71 forPM25,
0.18forSO2.
Evening FEV-,: NO2
becomes positive with
PM2.5 adjustment.
Diurnal change FEV-i:
NO2 robust to
adjustment for PM2.5
orSO2. SO2 and
PM2.5 attenuated
slightly.
Moderate correlation
between NO2 and
PM2s. Pearson r =
0.68.
November 2013
4-35
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Barraza-
Villarreal et al.
(2008)1
Hernandez-
Cadena et al.
(2009)1
Study Population and
Methodological Details
Mexico City, Mexico
N = 163-179, ages 6-14 yr, 54% persistent
asthma, 89% with atopy.
Repeated measures. Supervised
spirometry. Examined every 15 days for
mean 22 weeks. 1,503 observations.
Children with asthma recruited from
pediatric clinic. Children without asthma
were friends or schoolmates. Asthma
severity assessed by pediatric allergist.
Linear mixed effects model with random
effect for subject and adjusted for sex,
BMI, lag 1 minimum temperature, ICS
use, time. Adjustment for outdoor
activities, smoking exposure, anti-allergy
medication use, and season did not alter
results.
Mexico City, Mexico
N - 85, ages 7-12 yr, 62% mild,
intermittent asthma, 90% with atopy.
Exposure Metrics
Analyzed
NO2-Central site
8-h max
Monitors within 5
km of school or
home.
Spearman
correlation
coeffiecient for
school vs. central
site: r= 0.21
NO2-Central site
1-h max
Site within 5 km of
Subgroup
Lag day Analyzed
Analyzed (if applicable)
1 -4 avg
Asthma,
n = 126
No asthma,
n = 50
Asthma,
n = 129
No asthma,
n = 45
0
Effect Estimate
(95% Cl)
Single-Pollutant Model3
FEV-i
0% (-0.87, 0.87%)
-0.64% (-2.1, 0.82%)
FVC
-0.11% (-1.2, 0.97%)
-0.91% (-2.6, 0.76%)
% change FEV-i after
bronchodilator use
-39% (-64, 5.4%)
Copollutant
Examination
No copollutant model.
PM2 5 associated with
FEV-i and FVC.
Moderate correlation
with NO2. Pearson r =
61.
No copollutant model.
O3, not PM2.5
associated with FEV-i.
Cross-sectional. Supervised spirometry.
Recruitment from asthma and allergy
clinic. Atopy and asthma severity
assessed at clinic. Linear regression
adjusted for sex, pet ownership in
previous 12 mo, visible mold in home, lag
1 max temperature. Adjustment for age
and passive smoking exposure did not
alter results. Did not examine potential
confounding by SES.
home or school
24-h avg and
8-h max similar
results but less
precise
November 2013
4-36
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Yamazaki et al.
(2011 )t
Delfino et al.
(2008a)t
Study Population and
Methodological Details
Yotsukaido, Japan
N = 17, ages 8-15 yr.
Repeated measures. Supervised peak
flow collected before medication use.
Examined daily during long-term stay in
hospital. No air conditioning in hospital.
Permitted to go outside if asthma stable.
Lack of generalizability. 1,198 morning
and evening observations. GEE adjusted
for sex, age, height, temperature, day of
week, temporal trends.
Riverside, Whittier, CA
N = 53, ages 9-18 yr, persistent asthma
and exacerbation in previous 12 mo.
Repeated measures. Home spirometry.
Examined daily for 1 to 16 10-day periods.
416 observations. Recruitment by referral
from school nurses. Parent report of
physician-diagnosed asthma. Non-
smokers from nonsmoking homes. No
information on participation rate. Mixed
effects model with random effect for
subject with pollutant concentrations
centered on subject mean and adjusted
for personal relative humidity, personal
temperature, and follow-up period.
Adjustment for city, beta agonist use,
weekend, gas stove use did not alter
results.
Exposure Metrics
Analyzed
NO2-Central site
1-h avg
(6 p.m. -7 p.m.)
Monitor adjacent to
hospital.
NO2-Personal
24-h avg
Monitoring checked
daily.
NO2-Central site
24-h avg
Central site with 5
or 10 km of homes
Central site and
personal r = 0.43
Subgroup
Lag day Analyzed
Analyzed (if applicable)
0
0-1 avg All subjects
0 All subjects
No
bronchodilator,
n = 37
Bronchodilator
use, n = 16
0 All subjects
Effect Estimate
(95% Cl)
Single-Pollutant Model3
No quantitative data. PEF
decreases with increasing
NO2 0 to 23 hours before
measurement.
Stronger associations at 0
h and 12 h.
% predicted FEV1
-1.7 (-3.2, -0.19)
-1.5 (-2.3, -0.57)
-1.7 (-2.7, -0.75)
-0.70 (-2. 9, 1.5)
-1.3 (-2.4, -0.15)
Copollutant
Examination
Only 3-pollutant model
analyzed.
PM2.s also associated
with evening PEF.
Moderate correlation
with PM2.5. r = 0.62,
w/ 1-h max PM25: -1.3
(-2.8, 0.22)
Weak correlation with
M(~)o ^nparmpn r ~
• 1 *t\^2 . O|JG0IIII0II 1 —
0.38 for personal
PM2.5, 0.36 for central
site.
PM2.5 robust to NO2
adjustment.
• Lack of associations
with EC, OC.
Central site NO2
w/personal PM2.5:
-0.86 (-2.6, 0.89)
November 2013
4-37
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Lag day
Analyzed
Subgroup
Analyzed
(if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Mortimer et al. Bronx and East Harlem, NY Chicago, IL
(2002) Cleveland, OH Detroit, Ml St. Louis, MO
Washington, DC (NCICAS)
N = 846, ages 4-9 yr.
Repeated measures. Home peak flow.
Examined daily for four 2-week periods.
Recruitment from ED visits and clinics.
Parent report of physician-diagnosed
asthma and symptoms in previous 12 mo,
or asthma symptoms for >6 weeks and
symptoms with exercise or cold exposure
or family history of asthma.
Representative of full cohort except for
greater asthma medication use. Mixed
effects model adjusted for city, follow-up
period, day of study, 24-h rainfall,
12-h avg temperature.
NO2-Central site
4-h avg
(6 a.m.-10 a.m.)
Average of all city
monitors.
Single-
day lags 1
to 6
1-5 avg
1 -4 avg
0-4 avg
0-3 avg
No quantitative data. Only
reported no association
with PEF.
O3 associated with
PEF.
O'Connor et al. Boston, MA Bronx, NY Chicago, IL Dallas,
(2008)1 TX New York, NY Seattle, WA Tucson, AZ
(ICAS).
N = 861, ages 5-12 yr, persistent asthma
and atopy, 82% black or Hispanic.
Repeated measures. Home spirometry.
Examined for 2 weeks every 6 mo for 2 yr.
Recruitment from intervention of physician
feedback. Mixed effects model adjusted
for site, month, site*month interaction,
temperature, intervention group.
NO2-Central site
24-h avg
All monitors close to
home and not near
industrial source.
Median distance to
site = 2.3 km
1 -5 avg
% predicted FEV-i:
-1.3 (-1.9,-0.78)
% predicted PEF:
-1.6 (-2.2, 1.1)
Only 3-pollutant model
analyzed.
PM2.s, SO2, CO, O3
also associated.
Moderate correlations
with NO2. r = 0.59 for
PM25and SO2, 0.54
for CO. Weak
correlation with O^. r
= -0.31
Just et al. Paris, France
(2002) N = 82, ages 7-1 Syr, asthma attack in
previous 12 mo and daily asthma
medication use, 90% atopy
Repeated measures. Home peak flow.
Examined daily for 3 mo. Recruitment
from hospital outpatients. GEE adjusted
for time trend, day of week, pollen,
temperature, humidity.
NO2-Central site
24-h avg
Average of 11 sites
NR
No quantitative data. Only
reported no relationship
with PEF.
No copollutant model.
Os associated with
PEF. No correlation
with NO2. Pearson r =
0.09.
November 2013
4-38
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Lag day
Analyzed
Subgroup
Analyzed
(if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Wiwatanadate
and
Trakultivakorn
(2010)t
Chang Mai, Thailand
N = 31, ages 4-11 yr, asthma plus
symptoms in previous 12 mo, 52% mild
intermittent
Repeated measures. Home peak flow.
Examined daily for 1 yr. Recruitment from
allergy clinic. GLM with random effect for
subject and adjusted for time trend, day of
week, height, weight, atmospheric
pressure, temperature, sunshine duration.
NO2-Central site
24-h avg
1 site within 25 km
of homes.
forage, sex, height, growth of child,
temperature.
PEF in L/min
-1.8 (-5.4, 1.8)
2.6 (-1.2, 6.4)
Odaiima et al.
(2008)1
Fukuoka, Japan
N = 70, ages 4-11 yr, 66% with asthma
exacerbation
Repeated measures. Home peak flow.
Examined daily for 1 yr. >15,000
observations. Recruitment from hospital
where received treatment. GEE adjusted
NO2-Central site
3-h avg
24-h avg
1 site
No quantitative data. Only
reported no association
with PEF.
Only 3-pollutant model
analyzed.
SPM associated with
PEF in warm season.
Weak correlation with
NO2. r = 0.30 for
24-h avg.
Gillespie-
Bennett et al.
(2011)t
Bluff, Dunedin, Christchurch, Porirua, Hutt
Valley, New Zealand
N = 358, ages 6-13 yr
Cross-sectional. Home spirometry.
Multiple measures of lung function, 1 NO2
measurement. Recruitment from a home
heating intervention. 77% participation.
Mixed effects model with log-transformed
NO2 and random effect for subject.
Adjustment for age, sex, region, ethnicity,
intervention, income, temperature did not
alter results.
NO2-outdoor home 4-week
avg
NO2-indoor home
Per log increase NO2:
Evening FEV-i (ml_)
-88 (-191, 15)
Evening PEF (L/min)
-10 (-21, 0.98)
Evening FEV-i (ml_)
-13 (-26,-0.38)
Evening PEF (L/min)
-0.97 (-2.3, 0.36)
No copollutant model.
No other pollutants
examined.
November 2013
4-39
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Adults with
Study Population and
Methodological Details
Respiratory Disease
McCreanor et London, U.K.
al- (2007) N = 31 mild asthma, 32 moderate asthma,
ages 19-55 yr, all with airway
hyperresponsiveness, 84% with atopy
Randomized cross-over natural
experiment. Supervised spirometry.
Exposure on busy road and park. 55
observations. Recruitment from
advertisements and volunteer databases.
Mixed effects model with random effect for
subject and adjusted for temperature,
relative humidity.
Qian et al.
(2009btt
Boston, MA; New York, NY; Denver, CO;
Philadelphia, PA; San Francisco, CA;
Madison, Wl.
N = 119, ages 12-65 yr, persistent
asthma, nonsmokers
Repeated measures. Home PEF. No
information on participation rate. Study
population representative of full cohort.
Examined daily for 16 weeks. >14,000
observations. Trial of asthma medication,
a priori comparison of medication
regimens. Linear mixed effects model
adjusted for age, sex, race/ethnicity,
center, season, week, daily average
temperature, daily average humidity.
Adjustment for viral infections did not alter
results.
Exposure Metrics
Analyzed
NO2-on site of
outdoor exposure
2-h avg
NO2-Central site
24-h avg
Average of all
monitors within 20
miles of subject
ZIP code centroid
Subgroup Effect Estimate
Lag day Analyzed (95% Cl)
Analyzed (if applicable) Single-Pollutant Model3
2-h FEV-,:
-1.2%(-2.2,-0.22%)
FEF 25-75%:
-4. 3% (-7. 9, -0.65%)
22-h FEV^-0.73%
Post- (-2.0,-0.51%)
exposure FEF2s-75%: -4. 2% (-8. 7,
-0.61%)
PEF
0 All subjects -0.69% (-1.3, -0.06%)
Placebo -0.28% (-1 .4, 0.85%)
Beta-agonist -1 . 1 % (-2. 1 , -0.05%)
ICS -0.62% (-1.6, 0.38%)
0-2 avg All subjects -0.62% (-1.6, 0.38%)
Copollutant
Examination
FEF25-75% w/UFP:
-0.45%
(-0.73,-0.17%)
w/EC:-1.0%(-2.1,
0.06%)
w/PM25: -0.84% (-2.0,
0.34%)
Moderate correlation
with NO2. Spearman r
= 0.58 for UFP, EC,
0.60 for PM2.5
w/SO2: -0.52% (-3.9,
2.9%)
w/PM10:-3.6%(-7.7,
0.46%)
w/O3:-3.1%(-5.9,
-0.24%)
SO2 slightly
attenuated with NO2
adjustment.
PM-io, O3 not
associated with PEF.
Correlations NR.
November 2013
4-40
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Silkoffetal.
(2005)
Harre et al.
(1997)
Peacock et al.
(201 1)t
Study Population and
Methodological Details
Denver, CO
N = 34 with COPD, mean age 66, 67 yr,
panel 1 and 2, 75% with severe COPD
Repeated measures. Home PEF.
Recruitment from outpatient clinics,
research registries, advertisements. Mixed
effects model with random effect for
subjects and adjusted for temperature,
relative humidity, barometric pressure.
Christchurch, New Zealand
N = 40 with COPD, ages 55-83 yr,
nonsmokers
Repeated measures. Home PEF.
Recruitment from doctors' offices, COPD
support group, advertisements. 66%
participation. Log linear model adjusted for
temperature, wind speed, day of study,
CO, PM-io, SO2.
London, U.K.
N = 28-94 with COPD, 40-83 yr
Repeated measures. Home PEF.
Examined daily for 21-709 days.
Recruitment from outpatient clinic. GEE
adjusted for temperature, season. Lung
function measures adjusted for indoor
temperature and time spent outdoors.
Exposure Metrics
Analyzed
NO2-Central site
24-h avg
1 city site
NO2-central site
24-h avg
# sites NR
NO2-central site
1-h max
1 city site
Subgroup Effect Estimate
Lag day Analyzed (95% Cl)
Analyzed (if applicable) Single-Pollutant Model3
0 No quantitative data.
1 Negative, positive, and
2 null associations across
NO2 lags.
1 PEF
-0.72% (-1.5, 0.07%)
1 PEF:
0.1 7% (0.03, 0.32%)
PEF >20% below
predicted
OR: 1.0(0.86, 1.2)
Symptomatic fall in PEF
OR: 1.1 (0.97, 1.3)
Copollutant
Examination
No copollutant model.
Mixed positive,
negative, null
associations for
PM2.5, PM-io, O3
Only 4-pollutant model
analyzed.
PM-io, CO, SO2 not
associated with PEF.
Symptomatic fall in
PEF
ORw/PM-io: 0.97
(0.81, 1.2)
ORw/BC: 1.1 (0.84,
1.3)
November 2013
4-41
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Subgroup
Lag day Analyzed
Analyzed (if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Canova et al. Padua, Italy
(2010)* N = 19, ages 15-44 yr, 81 %
moderate/severe asthma
Repeated measures. Home PEF/FEV-i.
Examined for five 30-day periods for 2 yr.
Recruitment from prescription database of
subjects with mean >6 prescription/yr for 3
yr. 50% subjects provided fewer than 33%
maximum observations. GEE adjusted for
temperature, humidity, atmospheric
pressure, ICS use, smoking status.
NO2-central site
24-h avg
2 city sites
0, 1,2, 3
(single-
day)
0-1 avg
0-3 avg
No quantitative data.
NO2 shows null
associations with PEF and
FEV-i.
CO associated with
PEF. Moderate
correlation with NO2.
Spearman r = 0.48.
Wiwatanadate
and
Liwsrisakun
(201 Pi
Chiang Mai, Thailand
N = 121 with asthma and symptoms in
previous 12 mo, ages 13-78 yr, 48%
moderate/severe persistent asthma.
Repeated measures. Home PEF.
Examined daily for 10 mo. Recruited from
allergy clinic patients. GLM with random
effect for subject and adjusted for sex,
age, asthma severity, day of week, weight,
pressure, temperature, sunshine duration,
rain.
NO2-central site
24-h avg
1 city site
Evening PEF:
1.8(0.60, 3.0)
Average PEF:
-0.40 (-0.80, 0)
Units of PEF not reported.
Only multipollutant
models analyzed. No
associations with
PM-io, SO2, O3.
Hiltermann et
al. (1998)
Bilthoven, the Netherlands
N = 60 with asthma, ages 18-55 yr, all
with airway hyperresponsiveness, 87%
with atopy.
Repeated measures. Home PEF.
Examined daily for 4 mo. Recruitment
from outpatient clinic. Model adjusted for
allergen concentrations, smoking
exposure, day of week, temperature,
linear and quadratic term for study day
NO2-central site
24-h avg
1 city site
0
0-6 avg
Diurnal change PEF
-0.75 (-8.1, 6.6) L/min
-3.0 (-16, 10) L/min
No association found
with O3, PM-io, or BS.
November 2013
4-42
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Hiqqins et al.
(1995)
Hiqqins et al.
(2000)
Park et al.
(2005)
Study Population and
Methodological Details
Widnes, Runcorn, U.K.
N = 47 with asthma, 10 with COPD, 10
with asthma and COPD, 8 with wheeze,
ages 18-81 yr, 70% with atopy.
N = 31 with asthma, 3 with COPD, 63%
with atopy.
Repeated measures. Home PEF.
Examined daily for 28 days. Recruitment
from doctors' offices. Model specifics
including covariates NR.
Incheon, Korea
N = 64 with asthma, ages 16-75 yr, 31%
with severe asthma.
Exposure Metrics
Analyzed
NO2-central site
24-h avg
1 site per city
NO2-central site
24-h avg
1 0 city sites
Subgroup
Lag day Analyzed
Analyzed (if applicable)
0, 1, 0-1
avg
0
0
Effect Estimate
(95% Cl)
Single-Pollutant Model3
No quantitative data. NO2
reported to have some
effect on PEF.
<33th percentile: ref
33-67th percentile
-3.3 (-7.9, 1.4)
>67th percentile
-4.7 (-9.6, 0.21)
PEF in L/min
0.45 (-1.0, 1.9)
Copollutant
Examination
O3 and SO2
associated with PEF.
NO2 effect estimate
attenuated with
adjustment for SO2.
Os and spore count
associated with PEF.
PM-io and CO
associated with PEF.
Repeated measures. Home PEF.
Examined daily for 3-4 mo. Recruited from
medical center. GEE model, covariates
NR.
Maestrelli et al. Padua, Italy
I2011)t N = 32, mean age 39.6 (SD: 7.5) yr, 81%
persistent asthma.
Repeated measures. Supervised
spirometry. 6 measures over 2 yr. 166
observations. Selected from database as
beta-agonist users (>6/yr for 3 yr),
diagnosis clinically confirmed. Drop outs
did not differ from participants. GEE
adjusted for daily average temperature,
humidity, atmospheric pressure, asthma
medication use, current smoking status.
NO2-central site
24-h avg
2 city sites
% predicted FEV-,:
1.1 (-6.6, 8.7)
No copollutant model.
Os, SO2, personal or
central site PM2.5,
PM-io not associated
with FEV-i.
Correlations NR.
November 2013
4-43
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Laqorio et al.
(2006)
Children in the
Steerenberq et
al. (2001)
(Linn et al.,
1996)
Study Population and
Methodological Details
Rome, Italy
N = 11, ages 18-64 yr, 100% mild,
intermittent asthma, N = 11, ages 40-64
yr, COPD
Repeated measures. Supervised
spirometry. Examined 2/weekfortwo 1-
mo periods. Mean 9, 15
observations/subject. Recruitment of non-
smokers from outpatient clinic. GEE
adjusted for season, temperature,
humidity, beta-agonist use
General Population
Utrecht, Bilthoven, the Netherlands
N = 126, ages 8-13 yr, 28% respiratory
disease, 20% allergy.
Repeated measures. Supervised PEF.
FvsminpH "1/\A/PPk fnr 7 ft \A/PPkQ
C.AC1I 1 III 1 CU 1 / VVCCI\ 1 Ul / O VVCCI\O.
Recruitment from urban and suburban
schools. 65% participation. Mixed effects
model adjusted for sex, age, #cigarettes
smoked in home, presence of a cold,
history of respiratory symptoms and
allergy. No consideration for potential
confounding by meteorological factors.
Upland, Rubidoux, Torrance, CA
N - 269, 4th-5th grades
Repeated measures. Supervised
spirometry. Examined 1 week/season for
2 yr. Recruitment from schools. Repeated
measures ANOVA adjusted for year, day,
temperature, rain. Time spent outdoors =
101-136 min across seasons and
communities.
Exposure Metrics
Analyzed
NO2-central site
24-h avg
Average of 5 city
sites
NO2 - central site
15-h
(8 a.m.-11 p.m.)
24-h avg
NO - central site
15-h
(8 a.m. -11 p.m.)
i~) A U. _, ,_
24-h avg
Site within 2 km of
schools
NO2 - central site
24-h avg
# sites NR, no site
in Torrance
r = 0.61 correlation
with personal NO2
Subgroup
Lag day Analyzed
Analyzed (if applicable)
0 Asthma
0-1 avg COPD
15-h avg Urban
Suburban
0-2 avg Urban
Suburban
15-h avg Urban
Suburban
0-2 avg Urban
Suburban
0
Effect Estimate
(95% Cl)
Single-Pollutant Model3
% predicted FEV-i :
-2.0 (-3.2, -0.75)
-2.3 (-3.6, -1.0)
PEFmL/min-17(-35,0)
7, p>0.05
0, p >0.05
6, p >0.05
1, p>0.05
0, p >0.05
-6 (-12, 0)
6, p >0.05
p.m. FEV-i mL
-5.2 (-13, 2.3)
p.m. FVC
-3.6 (-12, 4.6)
Diurnal change FEV-i
-7.8 (-14, -1.5)
Diurnal change FVC
-2.2 (-9.6, 4.9)
Copollutant
Examination
No copollutant model.
Lung function
associated with PM2.s,
PM-io in adults with
COPD not asthma.
Moderate correlations
with NO2. Spearman r
= 0.43 for PM2.5, 0.45
for PM-io.
No copollutant model.
PM-io and BS also
associated with PEF.
Correlations NR.
Associations found
with PM2.5, weak for
3'
NO2 association
reported to lose
statistical significance
with PM2.5 adjustment.
Weak correlation with
PM2.5. r=0.25
November 2013
4-44
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Moshammer et
al. (2006)
Oftedal et al.
(2008)1
Ghana et al.
(2012)t
Study Population and
Methodological Details
Linz, Austria
N = 163, ages 7-10 yr,
Repeated measures. Supervised
spirometry. Examined every 2 weeks for
school yr. Recruitment from schools. GEE
model, covariates not specified.
Oslo, Norway
N =2,170, ages 9-1 0 yr, 5. 5% with
asthma
Cross-sectional. Supervised spirometry.
Recruitment from a birth cohort. 67%
participation, 60% follow-up. Examined
subjects had more "Westernized" parents.
Linear regression adjusted for age, sex,
height, BMI, current asthma, early life
maternal smoking, parental ethnicity,
education, smoking, and atopy, lag 1-3
temperature, neighborhood variables (%
married, % with income 2.
Taipei, Taiwan
N= 2,919, ages 12-16 yr
Cross-sectional. Supervised spirometry.
Recruitment from schools. Regression
model adjusted for residence in district,
age, sex, height, weight, temperature,
rainfall.
Exposure Metrics
Analyzed
NO2 - central site
8-h avg
(12 a.m.-8 a.m.)
Site adjacent to
school
NC>2-dispersion
model
NO2 - central site
24-h avg
1 city site
NC>2 - central site
4-h avg
(8 a.m.-12 p.m.)
10-h avg
(8 a.m. -6 p.m.)
Average of 5 city
sites within 2 km of
schools
Subgroup Effect Estimate
Lag day Analyzed (95% Cl)
Analyzed (if applicable) Single-Pollutant Model3
0 FEV-,:
-4.1% (-6.4, -1.7%)
FVC:
-2. 7% (-5.1, -0.33%)
1-3 avg Results in figure
1-7 avg Lag 1-3 avg similar to 1-7,
1-30 avg slightly smaller than 1-30
avg
Central site no
association.
FEV-, in mL:
0 -25 (-57, 7.5)
1 -41 (-70, -11)
2 -2.5 (-50, 45)
Copollutant
Examination
w/PM2.5:
-4.7% (-7.3, -2.0).
PM2.5 results
attenuated or become
positive. Associations
also found for PM-i,
PM-io.
Moderate correlations
with NO2. r = 0.53 for
PM-i, 0.54forPM25,
0.62 for PM-io.
No copollutant model.
No association
reported for PM2.5.
Correlations among
pollutants = 0.83-0.95
Short-term association
attenuated with
adjustment for early or
lifetime NO2. r =
0.46-0.77
No copollutant model.
Associations also
found with SO2, CO,
O3, PM-io
November 2013
4-45
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Castro et al.
(2009)1
Baqheri
Lankarani et al.
(201 0)t
Study Population and
Methodological Details
Rio de Janeiro, Brazil
N = 118, ages 6-1 5 yr, 18.4% with asthma
Repeated measures. Supervised PEF.
Recruitment from school. Examined daily
for 6 weeks. 9-122 observations/subject.
Mixed effects model with random effect for
subject and adjusted for weight, height,
sex, age, asthma, smoking exposure, time
trend, temperature, relative humidity.
Tehran, Iran
N = 562, elementary school.
Repeated measures. Examined daily for 6
weeks. 158 case-days. Case crossover
with control dates as two weeks before
and after case date. Conditional logistic
regression adjusted for daily temperature,
lag 0-6 avg PM-io.
Subgroup
Exposure Metrics Lag day Analyzed
Analyzed Analyzed (if applicable)
NO2 -school
24-h avg 1
School was within 2 1-2 avg
km of homes. 1-3 avg
NO - central site
24-h avg 0-6 avg
2 city sites
Effect Estimate
(95% Cl)
Single-Pollutant Model3
PEF, Liters/minute
0.04 (-0.58, 0.65)
-0.60 (-1.3, 0.14)
-1.7(0.02)
PEF <50% predicted
OR: 18(1, 326)
Copollutant
Examination
No copollutant model.
Associations also
found with PM-io,
weaker associations
with CO, SO2.
PM-io associated with
decreased odds of
large PEF decrement.
Eenhuizen et 3 study areas, the Netherlands
N = 880, age 8 yr,
Cross-sectional. Recruitment from
intervention study of mattress allergy
covers. Valid data on 49% subjects, who
had higher parental education, less likely
to have pets. Linear regression adjusted
for sex, age, height, weight, prenatal
smoke exposure, smoking in home, gas
stove, parental allergy, dampness in
home, parental education, season,
temperature, humidity.
- central site
1 site
Interrupter resistance
kPA*s/L (+ = worse)
0 (-0.04, 0.04)
-0.02 (-0.06, 0.03)
No associations with
PM-io or BS.
Moderate correlations
with NO2. Pearson r =
0.47 for PM-io, 0.60 for
BS.
November 2013
4-46
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Peacock et al.
(2003)
Scarlett et al.
(1996)
Timonen and
Pekkanen
(1997)
Ranzi et al.
(2004)
Study Population and
Methodological Details
Rochester upon Medway, U.K.
N = 177, ages 7-13 yr, 25% with wheeze
Repeated measures. Home PEF.
Examined daily for 13 weeks. 14-63
observations/subject. Recruitment from
rural and urban schools. Individual subject
regressions adjusted for day of week,
date, and temperature. Pooled estimates
obtained using weighting method.
Surrey, U.K.
N = 154, ages 7-1 1 yr, 9% with wheeze
Repeated measures. Supervised
spirometry. Examined daily for 6 weeks.
Recruitment from school. Lung function
adjusted for machine, operator, day of
week. Individual subject regressions
adjusted for temperature, humidity, pollen.
Pooled estimates obtained using
weighting method.
Kuopio, Finland
— "IRQ sriPQ 7 "1*7 \/r f^hilH rpn \A/ith f^ni in h
— IDC7, dU Co 1 \ £. y 1 , L/1 1 IIU 1 d 1 Will 1 L/UUU 1 1
Repeated measures. Home PEF.
Examined daily for 3 mo. Recruitment
from schools. 86% participation. Linear
mixed model adjusted for time trend,
weekend, minimum temperature, relative
humidity,
Emiglia-Romagna, Italy
N = 1 18, ages 6-1 1 yr, 77% with asthma,
67% with atopy.
Repeated measures. Home PEF.
Examined daily for 12 weeks. Recruitment
from schools. GLM adjusted for sex,
medication use, symptom status,
temperature, humidity
Exposure Metrics
Analyzed
NO2 -schools
24-h avg
1-h max
NO2 -school
1-h max
NO2 - central site
94 h a\/n
z.^ 1 1 d V u
# sites NR
26% missing data
were modeled, r =
0.58
NC>2 - central site
24-h avg
# sites NR
Subgroup Effect Estimate
Lag day Analyzed (95% Cl)
Analyzed (if applicable) Single-Pollutant Model3
0-4 avg PEF: -0.20 (-3.0, 2.6)
PEF>20%:2.3 (1.0,5.4)
PEF: 1.2 (-1.5, 3.9)
PEF>20%:1. 3 (0.5,3.4)
1 FEVo.75
0.30 (-0.29, 0.89%)
FVC
5.5% (-5.1, 17%)
0 FEV-,:
I Irhan 1 1 M4 "3,c-i\
\J\ \Jo\\ I I 1 I t, OO 1
Suburban -6.5 (-40, 27)
1-4 avg PEF:
Urban 13 (-24, 50)
Suburban -22 (-87, 43)
0 No quantitative data.
Figure shows null
association in group with
and without atopy.
Copollutant
Examination
No copollutant model.
PM2.5also associated
with PEF decrement
>20%.
Correlation NR.
Association found with
PM-io
Weak to moderate
correlations with NO2.
r — r\ r\~7
r — u.u/ .
Associations found for
SO2 in urban group.
Weak correlations
with NO2. r = 0.22.
PM2.s associated with
PEF in urban group.
November 2013
4-47
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Subgroup
Lag day Analyzed
Analyzed (if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Ward et al. West Midlands, U.K.
£000) N = 147, age 9 yr, 24% with symptoms,
31% with atopy.
Repeated measures. Home PEF.
Examined daily for 2 8-week periods.
Recruitment from schools. Individual
subject regressions adjusted for time
trend, day of week, meteorological
variables, pollen count. Individual
regressions pooled with weighting
method.
NO2 - central site
24-h avg
2 sites
0, 1,2, 3,
0-4 avg
No quantitative data.
Figure shows no
association across lags,
except 0 in symptomatic
group
No spool
Associations with
PM2.5 equally
inconsistent.
van der Zee et
al. (1999)
Rotterdam, Bodegraven/Reeuwijk,
Amsterdam, Meppel, Nunspeet, the
Netherlands
N = 633, ages 7-1 1 yr, 63% with
symptoms, 26 and 38% with asthma
NO2 - central site
24-h avg
1 site per
community
0
0-4 avg
Urban
Suburban
Urban
Suburban
OR: 0.96(0.79, 1.2)
OR: 0.77(0.54, 1.1)
OR: 1.1 (0.93, 1.3)
OR: 0.99(0.72, 1.4)
Associations found for
PM-io, BS, SO4, SO2
Correlations NR.
Repeated measures. Home PEF.
Examined daily for 3 mo. Recruitment
from general population. Logistic
regression adjusted for minimum
temperature, day of week, time trend,
influenza,
Roemer et al. Germany, Finland, the Netherlands,
(1998) Czech Republic, Norway, Italy, Greece,
Hungary, Sweden - 26 locations
N = 2,010, ages 6-12 yr, atopy
prevalence: 7-81%
Repeated measures. Home PEF.
Examined daily for 2 mo. Regression
model adjusted for minimum temperature,
school-day, time trend. Individual panel
results combined in a meta-analysis.
NO2 - central site
24-h avg
0
0-6 avg
PEF, Liters/minute
0.15 (-0.19, 0.49)
0.23 (-1.2, 1.6)
Association found with
PM-io and BS, but not
consistently across
lags
November 2013
4-48
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Lag day
Analyzed
Subgroup
Analyzed
(if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Adults in the General Population
Strak et al.
(2012)t
Utrecht area, the Netherlands
N = 31, adults ages 19-26 yr, all healthy,
nonsmoking
Repeated measures. Supervised
spirometry. Examined 3-7 times. 107
observations. Recruitment from university.
Well defined outdoor exposures at various
traffic/non-traffic sites. Outcomes
measured before and after outdoor
exposures. Heart rate maintained during
intermittent exercise. Higher probability of
associations found by chance alone.
Mixed effects model adjusted for
temperature, relative humidity, season,
high/low pollen, respiratory infection.
NO2 - on site of
outdoor activity
5-h avg
NOX - on site of
outdoor activity
5-h avg
0-h
2-h
18-h
0-h
2-h
18-h
Post-
exposure
FVC:
-4. 3% (-7.4, -1.0)
-3.5% (-6.5, -0.43%)
-4.5% (-7.4, -1.4)
-1.6% (-2.6, -0.51%)
-2.0% (-4.9, -0.16%)
-2.5% (-5.4, -0.69%)
FVC w/PNC:
NO2:
-3.0% (-7.2, 1.4%)
NOX:
-0.11% (-2.6, 2.5%)
Moderate to high
correlation with NO2.
Spearman r = 0.56,
0.75
PNC attenuated with
NO2 and NOX
Weichenthal et Ottawa, Canada
N = 42, adults ages 19-58 yr, from
nonsmoking homes, 95% white, 62% with
allergies, 33% with asthma
Repeated measures. Supervised
spirometry. Most examined 3 times. 118
observations. 1-h outdoor exposures
during cycling in low and high traffic areas.
Recruitment from public advertisements.
Differential exposure measurement error
for personal PM and VOCs and central
site NO2. Mixed effects models with
random subject effect adjusted for
temperature during cycling, average heart
rate. Adjustment for relative humidity, day
of week did not affect results.
NO2 - central site
1-h avg 1-h
1 site 4-h
Post-
exposure
FEV-i in liters
0.54 (-0.15, 1.2) L
0.40 (-0.12, 0.92) L
No copollutant model.
Lung function not
associated with O3 or
VOCs, UFP, BC,
PM2.5
November 2013
4-49
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Study Population and
Methodological Details
Exposure Metrics
Analyzed
Subgroup
Lag day Analyzed
Analyzed (if applicable)
Effect Estimate
(95% Cl)
Single-Pollutant Model3
Copollutant
Examination
Thaller et al. Galveston, TX
(2008)* N = 142, lifeguards at work, ages 16-27
yr, 13% with asthma, 22% with allergies.
Repeated measures. Supervised
spirometry. Recruitment from worksite.
1,140 observations. Self-report of
physician-diagnosed asthma. GLM,
covariates not specified.
NO2 and NOX-
central site
24-h avg, 1-h max
1 site 2.5-7.6 miles
from beaches
0
No quantitative data. NO2
and NOx reported not to
be significantly associated
with lung function.
Schindler et al. Aarau, Basel, Davos, Geneva, Lugano,
(2001) Montana, Payerne, Wald, Switzerland
N = 3,912, ages 18-60 yr, nonsmokers
Cross-sectional. Supervised spirometry.
Recruitment from registry and SALPADIA
cohort. Sample representative of full
cohort. Regression model adjusted for
sex, age, height, weight, day of week,
temperature, relative humidity. Adjustment
for asthma medication or wheeze did not
alter results.
NC>2 - central site
24-h avg 0
1 site per city 0-3 avg
FEV-,:
-2.5% (-4.5,-0.48%)
-2.9% (-5.9, 0.21%)
w/TSP:-1.2%(-3.!
1.6%)
Van Per Zee et Rotterdam, Bodegraven/Reeuwijk,
Amsterdam, Meppel, Nunspeet, the
Netherlands
N = 274, ages 50-70 yr, no symptoms in
previous 12 mo
Repeated measures. Home PEF.
Examined daily for 3 mo. Recruitment
from mailings. Logistic regression
adjusted for minimum temperature, day of
week, time trend, influenza,
NC>2 - central site
24-h avg
1 site per
community
PEF decrement >10%
Urban OR: 0.85 (0.59, 1.2)
Suburban OR: 0.72 (0.50, 1.05)
0-4 avg Urban
Suburban
OR: 0.46(0.20, 1.08)
OR: 0.56(0.27, 1.16)
No copollutant model.
PEF associated with
PM-io and SO4 in
urban group.
' Wide range of
correlations with NO2.
Spearman r =
0.16-0.72 for PM10,
0.25-0.65 for BS
November 2013
4-50
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Cakmak et al.
(2011 att
Steinvil et al.
(2009)1
Study Population and
Methodological Details
15 cities, Canada
N = 5,01 1 , ages 6-79 yr, mean age 39 yr
Cross-sectional. Supervised spirometry.
Recruitment by random sampling of
households. GLMM adjusted for age, sex,
income, education, smoking, random
effect for site. Adjustment for temperature
and relative humidity did not alter results.
Tel Aviv, Israel
N = 2,380, mean age 43 (SD: 1 1 ) yr,
healthy nonsmokers
Cross-sectional. Supervised spirometry.
Recruitment from ongoing survey of
individuals attending health center. Linear
regression adjusted for sex, age, height,
BMI, exercise intensity, education,
temperature, relative humidity, season,
year.
Subgroup
Exposure Metrics Lag day Analyzed
Analyzed Analyzed (if applicable)
NO2 - central site
24-h avg 0
# sites NR
NC>2 - central site
24-h avg 0
3 sites within 11 km 5
of homes 0-6 avg
Effect Estimate
(95% Cl)
Single-Pollutant Model3
% predicted FEV1
-1.6 (-2.9, -0.35)
FEV-, ml_:
-16 (-64, 33)
-55 (-103, -6.3)
-97 (-181, -13)
Copollutant
Examination
No copollutant model.
O3 and PM2.5also
associated with lung
function. Correlations
NR.
For lag 5
w/SO2: -7.8 (-72, 56)
w/CO:-19(-88, 50)
SO2 and CO results
robust with adjustment
for NO2.
High correlations with
NO2. Pearson r = 0.70
for SO2, 0.75 for CO.
November 2013
4-51
DRAFT: Do Not Cite or Quote
-------
Table 4-7 (Continued): Epidemiologic studies of lung function in children and adults with respiratory disease.
Study
Son et al.
(201 0)t
Study Population and
Methodological Details
Ulsan, Korea
N = 2,102, mean age 45 (SD: 17) yr,
mean % predicted FEV-i: 83%.
Cross-sectional. Supervised spirometry.
Recruitment during meeting of residents.
Regression model adjusted for age, sex,
BMI. Did not consider potential
confounding by weather, season, or time
trend. High correlation among exposure
assessment methods, r = 0.84-0.96.
Subgroup
Exposure Metrics Lag day Analyzed
Analyzed Analyzed (if applicable)
NO2-avg 13 central
site 0-2 avg
NO2-nearest site
Inverse distance
weighting
Kriging
All 24-h avg
Effect Estimate
(95% Cl)
Single-Pollutant Model3
% predicted FVC:
-7.9 (-10, -5.6)
-6.9 (-8.8, -5.0)
-6. 9 (-9.1, -4. 7)
-7.4 (-9.8, -5.1)
Copollutant
Examination
Associations found
with PM-io, O3, SO2,
CO NOT pffprt
estimate slightly
reduced with
adjustment for O3. No
copollutant model with
PM-io orSO2.
Note: Studies are organized by population examined and then generally in order of study strength (e.g., exposure assessment method, potential confounding considered). GLM =
Generalized linear model, BC = black carbon, SO2 = sulfur dioxide, BTEX = benzene, toluene, Ethylbenzene, xylene, BMI = body mass index, PM25 = particulate matter less than 2.5
urn in aerodynamic diameter, EC = elemental carbon, VOCs = volatile organic compounds, ICS = inhaled corticosteroid, SES = socioeconomic status, O3 = ozone, GEE = generalized
estimating equations, OC = organic carbon, NCICAS = National Cooperative Inner-city Asthma Study, ICAS = Inner City Asthma Study, CO = carbon monoxide, SPM = suspended
particulate matter, UFP = ultrafine particles, PMio = particulate matter less than 10 urn in aerodynamic diameter, NR = not reported, PNC = Particle number concentration, TSP = Total
suspended particles.
"Effect estimates were standardized to a 20-ppb increase in 24-h avg NO2, a 25-ppb increase in 5-h to 12-h avg or 8-h max NO2, a 30-ppb increase 1-h or 2-h avg NO2, and a 50-ppb
increase in 5-h avg NOX.
JRecent study published since the 2008 ISA for Oxides of Nitrogen.
November 2013
4-52
DRAFT: Do Not Cite or Quote
-------
1 Compared with more spatially resolved estimates of NO2 exposure, evidence for
2 associations with lung function decrements was less robust for NO2 measured at central
3 sites. Among studies that measured ambient NO2 at central sites, some found associations
4 with lung function decrements (Yamazaki et al.. 2011; Dales et al.. 2009a; Hernandez-
5 Cadena et al.. 2009; Liu et al.. 2009b; O'Connor etal.. 2008). In the U.S. multicity ICAS
6 cohort, a 20-ppb increase in 24-h avg NO2 at lag 1-5 day avg was associated with a
7 -1.3-point (95% CI: -1.9, -0.78) change in percent predicted FEVi (O'Connor et al..
8 2008). Many studies reported lack of association (Wiwatanadate and Trakultivakorn.
9 2010; Barraza-Villarreal et al.. 2008; Odaiima et al.. 2008; Just et al.. 2002; Mortimer et
10 al.. 2002). Several studies did not report quantitative results, but among children (in
11 Mexico City, Mexico, and Thailand), various lung function parameters showed no or
12 imprecise associations with NO2 (Wiwatanadate and Trakultivakorn. 2010; Barraza-
13 Villarreal et al.. 2008) (Table 4-7). NO2 exposures were assigned as ambient
14 measurements from a site located within 5 or 10 km of subjects' homes or schools,
15 measurements averaged among city monitors, or measurements from 1 site. The central
16 site NO2 assessment method did not appear to influence results.
17 Adjustment for potential confounding varied among studies but in most cases included
18 temperature. Several studies adjusted for (or considered in preliminary analyses) relative
19 humidity; a few studies examined day of the week, smoking exposure, or asthma
20 medication use. Few studies analyzed copollutant models, and while Holguin et al.
21 (2007) found that neither PM2 5 nor elemental carbon (EC) was associated with FEVi
22 among children with asthma in Ciudad Juarez, Mexico, most studies found associations
23 with PM25 as well as with PMi0, black carbon (BC), EC, sulfur dioxide (SO2), ozone
24 (O3) or volatile organic compounds (VOCs). A wide range of correlations with NO2 were
25 reported for PM2 5 (r = 0.30-0.71). Negative or weakly positive correlations were reported
26 for other pollutants (e.g., -0.72 for PMi0 to 0.18 for SO2). In copollutant models, NO2
27 effect estimates were attenuated in some cases and robust in others. Copollutant effect
28 estimates adjusted forNO2 generally were robust. Among children with wheeze in
29 Portugal, the association of modeled outdoor NO2 with FEVi was attenuated (-3.7%
30 [95% CI: -33, 25%] per 20-ppb increase in 1-week avg NO2) with adjustment for
31 benzene (Spearman r = -0.42 to 0.14). Among children with asthma in Windsor, Ontario,
32 Canada, associations of 12-h avg and 24-h avg NO2 with FEVi became positive with
33 adjustment for highly correlated (r = 0.71) PM25 (Dales et al.. 2009a; Liu et al.. 2009b)
34 (Table 4-7). NO2 associations with FEVi diurnal change were robust to PM2 5 and SO2
35 adjustment (Dales et al.. 2009a). In a more detailed copollutant analysis of personal and
36 central site measures, Delfino et al. (2008a) found the association of personal NO2 with
37 FEVi to be robust (-1.3-point [95% CI: -2.8, 0.22] change in percent predicted FEVi per
38 20-ppb increase in NO2) to adjustment for personal PM2 5, which was weakly correlated
November 2013 4-53 DRAFT: Do Not Cite or Quote
-------
1 with personal NO2 (Spearman r = 0.38). Adjustment for personal PM2 5 (r = 0.32)
2 reduced but did not eliminate the association of central site NO2 with FEVi (-0.86-point
3 [95% CI: -2.6, 0.89] change per 20-ppb increase in NO2). Results from copollutant
4 analyses with personal and central site NO2 indicate that ambient NO2 may partly serve
5 as an indicator of personal PM2 5 but also provide evidence for independent effects on
6 FEVi of personal and ambient NO2.
Adults with Respiratory Disease
7 Most previous and recent studies of lung function in adults with asthma or COPD were
8 based on PEF measured at home and produced inconsistent associations with ambient
9 NO2 concentrations (U.S. EPA. 2008c). Ambient NO2-associated decreases in PEF were
10 found in a recent multicity U.S. study of adults with asthma (Oian et al.. 2009b). The few
11 previous studies with supervised spirometry found associations with ambient NO2
12 concentrations (McCreanor et al.. 2007; Lagorio et al.. 2006) whereas the recent study
13 did not (Maestrelli et al.. 2011) (Table 4-7). The majority of studies found no
14 NO 2 -associated lung function decrements in adults with asthma or COPD (Maestrelli et
15 al.. 2011; Canovaet al.. 2010; Hiltermann et al.. 1998) or mixed associations (Peacock et
16 al..2011; Wiwatanadate and Liwsrisakun. 2011; Park etal.. 2005; Silkoff et al.. 2005;
17 Higgins et al.. 1995) among the various lung function parameters or NO2 exposure lags
18 examined. Most studies recruited subjects from outpatient clinics or doctors' offices, and
19 the nonrandom selection of the general population may produce study populations less
20 representative of the asthma population. There were more studies of adults with asthma
21 than adults with COPD, but evidence was equally inconsistent for the two conditions.
22 Similar associations between increases in ambient NO2 concentrations and FEVi
23 decrements were found in adults with COPD and asthma in Rome, Italy (Lagorio et al..
24 2006).
25 With respect to exposure assessment, most studies examined 24-h NO2. In a study of
26 adults with COPD in London, U.K., 1-h max NO2 showed mixed associations among the
27 various lung function measures examined (Peacock et al.. 2011). Among all studies, no
28 clear pattern of association was found for a particular lags of exposure (0, 1, 2, or 2- to
29 7-day avg). NO2 exposures were assessed primary from central site measurements, and
30 results were equally inconsistent for NO2 exposures assigned from 1 site or averaged
31 from multiple city sites. However, in a study in London, U.K., with stronger exposure
32 assessment, NO2 measured on site of outdoor exposures on a high-traffic road (allowing
33 only diesel buses and taxis) and in a park was associated with decrements FEVi and
34 FEF25_75o/0 in adults with mild to moderate asthma (McCreanor et al.. 2007). Results
35 indicated associations 2- to 22-hours after exposure. A 30-ppb increase in 2-h avg NO2
November 2013 4-54 DRAFT: Do Not Cite or Quote
-------
1 with a lag of 2 hours was associated with a -1.24 L (95% CI: -2.26, -0.23) change in
2 FEVj and -4.4 L/min (95% CI: -8.1, -0.68) change in FEF25.75%
3 In addition to the inconsistent evidence in adults with asthma or COPD, there is
4 uncertainty regarding an independent association of NO2 from that of copollutants.
5 Lagorio et al. (2006) found lung function decrements in association with NO2 but not
6 PM2 5 or PM10. But, in most cases, associations were also found for copollutants
7 (Peacock etal.. 2011: Oian et al.. 2009b: McCreanor et al.. 2007: Higgins et al.. 2000V
8 Studies reporting copollutant correlations found moderate correlations (Spearman r =
9 0.43-0.60 for UFP, EC, PM2 5, PM10) (McCreanor et al.. 2007: Lagorio et al.. 2006V In
10 the few studies with copollutant modeling, NO2-PEF effect estimates were attenuated
11 with adjustment for SO2 in the U.S. multicity study of adults with asthma (Oian et al..
12 2009b') and for PM10 and black smoke (BS) in the study of adults with COPD in London,
13 U.K. (Peacock et al.. 2011). Copollutant effect estimates were robust or less attenuated
14 with adjustment for NO2. The London walking study, with pollutants measured on site of
15 outdoor exposures, provided some evidence for an independent association for NO2.
16 NO2-associated decrements in FEVi were attenuated to near null with adjustment for
17 UFP, EC, or PM2 5 (McCreanor et al.. 2007). Associations with FEF25_75o/0 decreased in
18 magnitude and precision with copollutant adjustment but remained negative (e.g., -0.45%
19 [95% CI: -0.73, 0.17%] per 30-ppb increase in 2-h avg NO2 with adjustment for UFP,
20 Spearman r = 0.58). These results indicate that the decrements in some lung function
21 parameters associated with near-road exposures of relatively short duration (2 hours)
22 were attributable to NO2.
Children in the General Population
23 As in other populations, the 2008 ISA for Oxides of Nitrogen indicated associations
24 between ambient NO2 concentrations and lung function decrements in children in the
25 general population as measured using supervised spirometry but not home PEF (U.S.
26 EPA. 2008c). All recent studies conducted supervised spirometry, and most found
27 associations with ambient NO2 concentrations. Studies recruited children from schools,
28 and reflecting the general population, examined groups of children with prevalence of
29 respiratory conditions such as asthma and allergy of 5 to 72%. Several recent studies
30 were cross-sectional. Most found associations with adjustment for time-vary ing factors
31 such as weather as well as between-subject factors such as height, weight, smoking
32 exposure, and SES (Chang etal.. 2012: Oftedal et al.. 2008). In children in the
33 Netherlands, no association was found between 24-h avg NO2 and interrupter resistance
34 (Eenhuizen et al.. 2013). a measure of airway resistance. In controlled human exposure
35 studies, examination of airway resistance has been limited to adults. But, some evidence
36 shows NO2-induced increases in airway resistance (Section 3.3.2.2).
November 2013 4-55 DRAFT: Do Not Cite or Quote
-------
1 A relatively large body of previous studies, which were conducted in various European
2 countries, did not provide evidence of NO2-associated decrements in PEF. These studies
3 were similar to studies of supervised lung function in examining study populations that
4 included children with symptoms, asthma, or atopy, 24-h avg NO2 measured at central
5 sites, and exposures lagged 0 to 3 days or averaged over 5 to 7 days. Collectively, results
6 indicated null associations (Ranzi et al.. 2004; Ward et al.. 2000; van der Zee et al.. 1999)
7 or NO2-associated increases in PEF (Peacock et al., 2003; Roemeretal.. 1998; Timonen
8 and Pekkanen. 1997). including a population with 77% asthma and 70% atopy prevalence
9 (Ranzi et al.. 2004).
10 With respect to exposure assessment, a majority of evidence was for NO2. NO was
11 associated with lung function among children in Iran (Bagheri Lankarani et al.. 2010) but
12 not consistently among children in the Netherlands (Steerenberg et al.. 2001). With the
13 exception of Scarlett et al. (1996). studies found associations with NO2 measured at or
14 next to schools (Castro et al.. 2009; Moshammer et al.. 2006). Lung function decrements
15 were also associated with home outdoor NO2 estimated with dispersion modeling but not
16 measured at central sites (Oftedal et al.. 2008). These model estimates corresponded well
17 with measured outdoor concentrations. Among children in three southern California
18 communities, Linn et al. (1996) indicated that central site NO2 measurements may
19 represent temporal variation in personal exposures by finding a correlation of 0.61
20 between the two metrics. Most studies examined 24-h avg NO2. Associations also were
21 found with NO2 averaged over 4 to 10 hours (Chang etal. 2012; Moshammer et al..
22 2006) and 8-h max NO2 (Barraza-Villarreal et al.. 2008) but not 1-h max NO2 (Chang et
23 al.. 2012; Scarlett et al.. 1996). Lung function decrements were found with NO2 lagged 0
24 or 1 day and averaged over 3, 7, or 30 days, without a clear difference among lags in the
25 magnitude of association. Among children in Oslo, Norway, associations with lung
26 function were larger for lag 1-30 avg NO2 than lag 1-2 avg or 1-7 avg (Oftedal et al..
27 2008). With adjustment for age 1-year or lifetime NO2, the association for short-term
28 NO2 was attenuated, but exposure metrics were highly correlated (r = 0.70-0.77).
29 In most studies, an association of ambient NO2 with lung function in children
30 independent of copollutants was not clearly distinguished. Among children in Oslo,
31 Norway, decrements in lung function were associated with increases in ambient NO2 not
32 PM10 or PM2 5 (no quantitative results reported) (Oftedal et al.. 2008). But, other studies
33 found associations with PMi0, PM25, and BS, which showed a range of correlations with
34 NO2 (r = 0.25-0.62). Few studies conducted copollutant modeling. In a study of children
35 in three southern California communities, Linn etal. (1996) did not provide quantitative
36 results for copollutant analyses and only indicated that NO2 effect estimates lost
37 statistical significance with adjustment for PM2 5. Robust PM2 5 -adjusted effects were
38 estimated among children in Austria, whose exposures were assessed from a monitoring
November 2013 4-56 DRAFT: Do Not Cite or Quote
-------
1 site adjacent to the school (Moshammer et al., 2006). A 25-ppb increase in lag 1 of
2 8-h avg NO2 (12 a.m.-8 a.m.) was associated with a -4.1% change (95% CI: -6.4, -1.7%)
3 in FEVj in the single-pollutant model and a -4.7% change (95% CI: -7.3, -2.0%) with
4 adjustment for PM2 5 (r = 0.54). PM2 5 effect estimates were attenuated or became
5 positive with adjustment for NO2. While these results provide evidence for an
6 independent association with NO2, other model covariates were not specified, and
7 potential confounding by other factors such as weather cannot be assessed.
Adults in the General Population
8 In studies reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). increases
9 in ambient NO2 concentration were associated with decrements in lung function in adults
10 in the general population as measured by supervised spirometry (Schindler et al., 2001)
11 but not by peak flow meter (Van Per Zee et al.. 2000). Recent studies examined lung
12 function with supervised spirometry but also provided inconsistent evidence. Studies
13 examined a wide range of ages (i.e., 18-79 years) and a mix of healthy populations and
14 those including adults with asthma or allergies, but these factors did not appear to
15 influence results. Van Per Zee et al. (2000) found no association in adults with or without
16 respiratory symptoms.
17 Studies noteworthy for examining lung function before and after repeated outdoor
18 exposures produced mixed evidence for associations with NOX averaged over 1 to 5
19 hours (Straketal. 2012; Weichenthal etal.. 2011; Thaller et al.. 2008). Associations
20 were not found in adults cycling in various traffic and non-traffic locations or lifeguards
21 working outdoors, whose NO2 exposures were assessed from a central site (Weichenthal
22 et al., 2011; Thaller et al., 2008). However, decreases in FVC and FEVi were found in
23 healthy adults after 5-hour NO2 and NOX exposures measured on site of outdoor activity
24 (Straketal.. 2012). A 25-ppb increase inNO2 and 50-ppb increase inNOX was
25 associated with a 4.3% (95% CI: 1.0, 7.4%) and 1.6% (0.51, 2.6%) decrease in FVC,
26 respectively, immediately after exposures. Decrements persisted 18 hours after exposure.
27 Other studies of adults in the general population also were mixed in finding ambient
28 NO2-associated lung function decrements. Contributing to the supporting evidence were
29 recent cross-sectional studies, including a study of 15 Canadian communities that
30 adjusted for age, sex, income, and smoking and considered potential confounding by
31 temperature and humidity (Cakmak etal.. 201 la). Studies similarly assessed 24-h avg
32 NO2 from central sites, primarily 1 per city. Son etal. (2010) was unique in comparing
33 various methods of exposure assessment. NO2-associated decreases in FVC in subjects
34 ages 7-97 years in Ulsan, Korea were similar among NO2 averaged across 13 city
35 monitors, measured at the nearest monitor, and estimated by spatial interpolation methods
November 2013 4-57 DRAFT: Do Not Cite or Quote
-------
1 (i.e., inverse distance weighting, kriging). Overall, results from this study may have
2 weaker implications because potential confounding by meteorological factors or other
3 time-varying factors was not considered. With respect to lags of exposure, lung function
4 decrements in adults were found with lag 0 day NO2 and lags 0-2 day and 0-3 day avg
5 NO2 (Cakmak et al.. 201 la: Son etal. 2010; Schindler etal.. 2001). Among healthy
6 adults in Tel Aviv, Israel, Steinvil et al. (2009) found decreases in FEVi in association
7 with increases in lag 0-6 day avg of 24-h avg NO2, but results generally were mixed
8 among the various lags of exposure examined.
9 In adults, an NO2-associated decrement in lung function independent of copollutants was
10 not clearly demonstrated. Studies also found associations with various PM components,
11 SO2, CO, and O3; and copollutant correlations were often high (r = 0.56 to 0.75). Some
12 studies found NO2 effect estimates to be attenuated with adjustment for TSP (Schindler
13 etal.. 2001) or SO2 or CO (Steinvil et al.. 2009) but found copollutant effect estimates to
14 remain robust to NO2 adjustment. Son etal. (2010) found robust NO2 associations with
15 FVC with adjustment for O3 but did not analyze copollutant models with PMi0, SO2, or
16 CO because they covaried with NO2. In contrast, Straketal. (2012). who measured
17 pollutants on the locations of outdoor exposures, found that FVC remained associated
18 with NO2 with adjustment for particle number concentration (PNC), O3, PM2 5, PMi0,
19 PM2.5-io, PM25 absorbance, EC, and metal components of PM25 (e.g., -3.0% [95% CI:
20 -7.2, 1.4%] per 25-ppb increase in 5-h avg NO2 with adjustment for PNC). The
21 association between NOX and FVC was attenuated with adjustment for PNC.
22 Copollutants were weakly to moderately correlated with NO2 and NOX, except for EC
23 and absorbance (r = 0.67-0.87). Effect estimates for EC, absorbance, and PNC were
24 attenuated with adjustment for NO2 or NOX, indicating that NO2 or NOX may have
25 confounded the associations of copollutants.
4.2.3.2 Controlled Human Exposure Studies
26 Most controlled human exposure studies examined adults, and consistent with
27 epidemiologic findings, numerous controlled human exposure studies, most of which
28 were reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). generally
29 reported minimal effects on lung mechanics in healthy adults or those with asthma or
30 COPD. Study details are presented in Table 4-8. but overall, exposures ranged from 200
31 to 4,000 ppb NO2 for 75 minutes to 6 hours, and most studies incorporated exercise in the
32 exposure period to assess lung function during various physiological conditions.
33 Among healthy adults, Huang et al. (2012b) conducted a study to examine the health
34 effects of NO2 exposure alone and in combination with exposure to concentrated ambient
November 2013 4-58 DRAFT: Do Not Cite or Quote
-------
1 particles (CAPs). Healthy adults did not experience any pulmonary function changes
2 during, immediately after, or 18 hours after exposure to 500 ppb NO2 for 2 hours with
3 intermittent exercise. These results are consistent with previously published studies in
4 healthy adults. For example, Hackney et al. (1978) demonstrated that exposure to 1,000
5 ppb for 2 hours/day for 2 consecutive days did not induce pulmonary function changes
6 with the exception of a 1.5% drop in forced vital capacity (FVC) after exposure on the
7 second day. Similarly, Frampton et al. (1989) reported no differences in lung function
8 before, during, or after exercise or after exposure to 600 or 1,500 ppb NO2 for 3 hours or
9 a 3 hour base of 50 ppb NO2 with intermittent peaks of 2,000 ppb. These results were
10 replicated in studies in healthy adults at similar concentrations (Frampton et al.. 2002;
11 Devlin etal.. 1999). Rasmussen et al. (1992) reported that healthy subjects exposed to
12 2,300 ppb NO2 for 5 hours even had slight improvements, though not statistically
13 significant, in FVC and FEVi during and after NO 2 exposure compared to air.
14 Since ambient NO2 exposures occur with copollutants, other studies have examined
15 co-exposures to NO2 and O3. Hazucha et al. (1994) found no effect on pulmonary
16 function after exposure to 600 ppb NO2 for 2 hours. However, significantly greater
17 reductions in FEVi and forced expiratory flow were observed after a subsequent O3
18 exposure (Hazucha et al.. 1994). Exposure of aerobically trained young men and women
19 to 600 ppb NO2 or 600 ppb NO2 + 300 ppb O3 for 1 hour resulted in an increase in
20 airway resistance with co-exposure, though the increase in resistance with co-exposure
21 was significantly less than O3 alone and NO2 exposure along did affect lung function
22 (Adams etal.. 1987).
23 Controlled human exposure studies examining potentially at-risk lifestages or
24 populations, including older adults and those diagnosed with COPD and asthma, also
25 have not consistently found decrements in lung function with NO2 exposure. Healthy,
26 older adults exposed to 300-400 ppb NO2 for 2-3 hours did not experience decrements in
27 lung function compared to air controls, though there were slight differences between
28 smokers and nonsmokers in FEVi (Gong etal.. 2005; Morrow et al.. 1992). Exposure of
29 older adults diagnosed with COPD to 300 ppb NO2 for 3 hours showed consistent
30 reductions in FVC that reached significance at the end of exposure (Morrow et al.. 1992).
31 while Vagaggini et al. (1996) reported decreases in FEVi in subjects with COPD exposed
32 to 300 ppb NO2 for 1 h. In contrast, Linn etal. (1985a) and Gong et al. (2005) reported
33 that exposure to 400-2,000 ppb for 1-2 hours has no effect on lung function in adults with
34 COPD.
35 Whereas NO2 consistently induced increases in AHR in adults with asthma (Section
36 4.2.2.2). direct changes in lung function or airway resistance were not consistently found.
37 Linn etal. (1985b) exposed adults with asthma and healthy adults to 4,000 ppb NO2 for
November 2013 4-59 DRAFT: Do Not Cite or Quote
-------
1 75 minutes and reported no changes in airway resistance after NO2 exposure in either
2 group. Kleinman et al. (1983) found no significant changes in forced expiratory flows or
3 airway resistance after exposure to 200 ppb NO2 for 2 hours with light exercise; however,
4 Bauer etal. (1986) reported significant decrements in forced expiratory flow rates in
5 adults with asthma after exposure to 300 ppb NO2 for 30 minutes. Torres and Magnussen
6 (1991) found no changes in lung function in adults with asthma exposed to 250 ppb NO2
7 for 30 minutes; however, exposure to 1,000 ppb NO2 for 3 hours with intermittent
8 exercise (adjusted to individual maximum workload) resulted in small reductions in
9 FEVi in adults with asthma (Torres etal.. 1995).
10 Subjects with asthma have also been screened for pulmonary function changes in
11 response to pollutant co-exposures. Koenig et al. (1987) exposed adolescents with asthma
12 to 300 ppb NO2 in combination with 120 ppb O3, with or without 70 (ig/m3 H2SO4 or
13 50 ppb HNO3, and reported no changes in pulmonary function. Jenkins etal. (1999)
14 investigated the effects of exposure to 200 ppb NO2 for 6 hours (with or without 200 ppb
15 O3) or 400 ppb NO2 for 3 hours (with or without 400 ppb O3) and found no change in
16 lung function in adults with asthma following NO2 exposures. Significant decreases in
17 FEVi were found following the 3-hour exposure to O3 and O3 + NO2.
18 Overall, ambient concentrations of NO2 do not consistently contribute to decrements in
19 lung function in controlled human exposure studies in adults with chronic respiratory
20 disease, particularly COPD. Studies more frequently report NO 2-induced decrements in
21 lung function, but these also vary across studies. This evidence suggests that NO2
22 exposure has minimal effects on lung function in the range of 200-4,000 ppb for 30
23 minutes to 6 hours.
November 2013 4-60 DRAFT: Do Not Cite or Quote
-------
Table 4-8 Controlled human exposure studies of NO2 and lung function.
Study
Adams et al. (1987)
Bauer et al. (1986)
Blomberq et al.
(1999)
Devlin etal. (1999)
Frampton et al.
(1989)
Frampton et al.
(1991)
Frampton et al.
(2002)
Gong et al. (2005)
Hackney et al.
(1978)
Disease status3;
n; Sex; Age
(mean ± SD)
(1-3)n = 20M,
20 F; F=21.4±
1.5yr
M= 22.7 ±3.3
Asthma
n = 15; 33 ±7.8
yr
n = 8 M, 4 F; 26
yr (Range: 21-32
yr)
n = 11 M; Range:
18-35yr
(1)n = 7M, 2F;
30 yr (Range:
24-37 yr)
(2)n = 11 M, 4F;
25 yr (Range:
1 9-37 yr)
(1)n = 7M, 2F;
29.9 ± 4.2 yr
(2)n = 12 M, 3F;
25.3 ± 4.6 yr
(3)n = 11 M, 4F;
23.5 ± 2.7 yr
(1,2)n = 12M, 9
F; F=27.1 ±4.1
yr
M= 26.9 ± 4.5 yr
Healthy: n = 2 M,
4F; 68 ± 11 yr
COPD: n = 9 M,
9 F; 72 ± 7 yr
n = 16 M; 26.9 ±
5.0 yr
Exposure Details (Concentration; Duration)
(1)600ppbNO2for1 h,
(2) 300 ppb O3 for 1 h,
(3) 600 ppb NO2 and 300 ppb O3 for 1 h;
(1-3) Exercise during entire exposure at VE= 75
L/min (M) and VE= 50 L/min (F)
300 ppb for 30 min (20 min at rest, 10 min of
exercise at VE>3 times resting
2,000 ppb, 4 h/day for 4 days; Exercise 15 min
on/1 5 min off at workload of 75 W
2,000 ppb for 4 h;
Exercise for 15 min on/15 min off at VE= 50
L/min
(1) 600 ppb for 3h,
(2) 1,500 ppb for 3h;
(1,2) Exercise 10 min on/20 min off at VE= ~4
times resting
(1 ) 600 ppb for 3 h,
(2) 1,500 ppb for 3 h,
(3) 50 ppb forSh + 2,000 ppb peak for 15
min/h;
(1-3) Exercise 10 min on/20 min off at VE= ~4
times resting
(1 ) 600 ppb for 3 h,
(2) 1,500 ppb for 3 h;
(1 ,2) Exercise 1 0 min on/20 min off at VE= 40
L/min
(1)400ppbNO2for2h
(2) 200 ug/m3 CAPs for 2 h
(3) 400 ppb NO2 + 200 ug/m3 CAPs for 2 h
(1-3) Exercise 15 min on/15 min off at VE= ~2
times resting
1,000 ppb, 2 h/day for 2 days; Exercise 15 min
on/15 min off at VE= 2 times resting
Endpoints Examined
Before and after
exposure
Before, during, and
after exposure
Before and after
exposure
Aerosol bolus
dispersion (deposition,
FEV-i and SRaw)
Pulmonary function
tests before, during,
and after exposure
Pulmonary function
tests before, during,
and after exposure,
airway reactivity 30 min
post-exposure
Pulmonary function
tests before and after
exposure
Pulmonary function
tests before and
immediately after
exposure and 4 h and
22-h post-exposure
Pulmonary function
tests before and after
each exposure
November 2013
4-61
DRAFT: Do Not Cite or Quote
-------
Table 4-8 (Continued): Controlled human exposure studies of NO2 and lung function.
Study
Hazucha et al.
(1994)
Huanq et al.
(2012b)
Jenkins et al.
(1999)
Kleinman et al.
Koeniq et al. (1987)
Linnetal. (1985b)
Linn et al. (1985a)
Disease status3;
n; Sex; Age
(mean ± SD)
n = 21 F;22.9±
3.6 yr
(1)n = 11 M, 3F
(2) n = 6 M, 7 F
(3) n = 7 M, 6 F;
24.6 ± 4.3 yr
Asthma
n = 9M, 2F;
31.2 ± 6.6 yr
Asthma
n = 12 M, 19 F;
31 ± 11 yr
Healthy
(1)n = 3M, 7F
(2)n = 4M, 6F
Asthma
(1)n=4M,6F
(2) n = 7 M, 3 F
14.4 yr
(Range: 12-19 yr)
Healthy: n =
16 M, 9F;
Range: 20-36 yr
Asthma:
n = 12 M, 11 F;
Range: 18-34 yr
COPD
n = 13 M, 9F
(1 never smoker,
13 former
smokers, and
8 current
smokers);
60.8 ± 6.9 yr
Exposure Details (Concentration; Duration)
(1 ) 600 ppb NO2 for 2 h, air for 3 h, 300 ppb O3
for2h,
(2) air for 5 h, 300 ppb O3 for 2 h;
(1 ,2) Exercise for 1 5 min on/1 5 min off at VE=
35 L/min
(1)500ppbNO2for2 h,
(2) 500 ppb NO2 + 73.4 ± 9.9 ug/m3 CAPs for 2
h,
(3) 89.5 ± 10.7ug/m3for2h;
(1-3) Exercise 15 min on/15 min off at VE= 25
L/min
(1)200ppbNO2for6 h
(2) 200 ppb NO2 + 100 ppb O3 for 6 h
(3)400ppbNO2for3h
(4) 400 ppb NO2 + 200 ppb O3 for 3 h
(1-4) Exercise 10 min on/40 min off atVE=
32L/min)
200 ppb for 2 h;
Exercise 15 min on/15 min off at VE= ~2 times
resting
(1) 120ppbNO2,
(2)180ppbNO2;
(1-2) Exposures were 30 min at rest with 10
min of moderate exercise
4,000 ppb for 75 min;
Two 15-min periods of exercise at VE= 25 L/min
and 50 L/min
500, 1,000, and 2,000 ppb for 1 h;
Exercise 15 min on/15 min off VE= 16 L/min
Endpoints Examined
Pulmonary function
tests before, during,
and after exposure,
airway reactivity after
exposure
Pulmonary function
tests before,
immediately after and
18 h after exposure
Pulmonary function
tests before and after
exposure
Pulmonary function
testing before and after
exposure
Pulmonary function
tests before, during,
and after exposure
Airway resistance
before, during, and
after exposure
Pulmonary function
tests before, during,
and after exposure
November 2013
4-62
DRAFT: Do Not Cite or Quote
-------
Table 4-8 (Continued): Controlled human exposure studies of NO2 and lung function.
Study
Disease status3;
n; Sex; Age
(mean ± SD) Exposure Details (Concentration; Duration)
Endpoints Examined
Morrow et al. (1992)
Healthy: n = 10
M, 10 F (13 never
smokers, 4
former smokers,
3 current
smokers)
COPD: n = 13M,
7 F (14 current
smokers, 6
former smokers);
59.9 ± 7.0 yr
300 ppbfor4 h;
Three 7-min periods of exercise at VE= ~4
times resting
Pulmonary function
tests before, during,
and after exposure and
24-h post-exposure
Rasmussen et al.
(1992).
Vaqaqqini et al.
(1996)
n = 10M, 4F; 2,300 ppb for 5 h
34.4 yr (Range:
22-66 yr)
Healthy: n = 7 M; 300 ppb for 1 h;
34 ± 5 V Exercise at VE= 25 L/min
Asthma: n = 4 M,
4F; 29 ± 14 yr
COPD: n = 7M;
58 ± 12 yr
Pulmonary function
tests before, 2 times
during, and 3 times
after exposure
Pulmonary function
tests before and 2 h
after exposure
"Subjects were healthy individuals unless described otherwise.
1
2
o
3
4
5
6
7
8
9
10
11
12
13
14
15
4.2.3.3 Summary of Studies of Lung Function
Recent epidemiologic evidence consistently indicates associations between short-term
increase in ambient NO2 concentration and decrements in lung function in children with
asthma (ages 6-18 years), as measured by supervised spirometry. The evidence in
children with asthma that was reviewed in the 2008 ISA for Oxides of Nitrogen was
based primarily on unsupervised PEF measurement and was inconsistent (U.S. EPA.
2008c). The results for FEVi are not supported by findings from a controlled human
exposure study of adolescents with asthma, which showed no effect. Previous and recent
epidemiologic evidence indicates ambient NO2-associated decrements in lung function in
children in the general population. Most studies recruited children from schools,
increasing the likelihood that study populations were representative of the general
population. Consistent with overall results, studies did not find more marked effects of
ambient NO2 exposure on lung function in children with asthma than in children without
asthma (Barraza-Villarreal et al.. 2008; Holguin et al.. 2007). Bronchodilator use was
associated with smaller ambient NO2-associated lung function in children with asthma
(Delfino et al.. 2008a) but not adults with asthma (Qian et al.. 2009b). Recent
November 2013
4-63
DRAFT: Do Not Cite or Quote
-------
1 epidemiologic studies of adults with asthma or COPD or adults in the general population
2 also measured lung function with spirometry, but collectively studies were mixed in
3 finding associations with NO2. Similarly, most controlled human exposure studies did
4 not find ambient-relevant NO2 exposures (200-4,000 ppb) of 30 minutes to 6 hours to
5 consistently induce decrements in lung function healthy adults, adults with asthma, or
6 adults with COPD.
7 In addition to the lack of experimental evidence to directly support epidemiologic
8 observations in children, the mechanisms underlying NO2-related decrements in lung
9 function are not well delineated. There is not strong evidence in humans for the direct
10 effects of inhaled NO2 on activating neural reflexes; however, there is some evidence for
11 mast cell degranulation mediating changes in lung function (Section 3.3.2.2). Mast cell
12 degranulation leads to histamine release, describing a role for allergic inflammation in
13 mediating NO2-induced lung function decrements. These findings provide biological
14 plausibility for the ambient NO2-associated decreases in lung function found in
15 populations of children with asthma that have high prevalence of atopy (53-100%) and
16 groups of children with asthma not using anti-inflammatory ICS (Hernandez-Cadena et
17 al., 2009; Liu et al., 2009b). Controlled human exposure studies of adults with atopic
18 asthma and the single study in adolescents with asthma did not find lung function
19 decrements with NO2 exposures of 120-400 ppb (for 30 minutes to 6 hours) but did find a
20 decrease with 1,000 ppb NO2 (3 hours).
21 Key epidemiologic evidence was provided by studies with relatively strong exposure
22 assessment characterized by measuring or modeling personal exposures (Martins et al.,
23 2012: Delfino et al.. 2008a: Oftedal et al.. 2008) and measuring NO2 at schools
24 (Greenwald et al., 2013; Holguin et al., 2007) or on site of outdoor exposures (Strak et
25 al.. 2012; McCreanor et al.. 2007). Lung function was also associated with NO2
26 measured at one available city monitoring site, the closest site, or averaged among city-
27 wide monitors. Comparisons among central site exposure assessment methods did not
28 clearly indicate differences in association, but stronger associations with lung function
29 were found for individual-level NO2 than central site NO2 (Delfino et al.. 2008a: Oftedal
30 et al., 2008) and for school outdoor NO2 than indoor NO2 (Greenwald et al.. 2013).
31 A majority of the supporting evidence was for 24-h avg NO2, with more variable results
32 for NO or NOX. For shorter averaging times (1-h max, 3- to 10-h avg), a few studies
33 found associations with lung function (Chang et al., 2012; Moshammer et al., 2006);
34 most did not (Peacock et al.. 2011; Spira-Cohen et al.. 2011; Barraza-Villarreal et al..
35 2008; Odajima et al., 2008; Mortimer et al., 2002; Scarlett et al.. 1996). In adults with
36 outdoor exposures in traffic and non-traffic locations, lung function decrements were
37 associated with NO2 averaged over 2 to 5 hours and measured on site of outdoor
November 2013 4-64 DRAFT: Do Not Cite or Quote
-------
1 exposures (Strak et al., 2012; McCreanor et al., 2007). Decrements in lung function were
2 found with lags of 2-22 hours after outdoor exposures (Strak et al.. 2012; McCreanor et
3 al., 2007), ambient NO2 concentrations lagged 0 or 1 day, and multiday averages of 2 to
4 30 days. Evidence was not more robust for a particular lag of NO2 exposure. Some
5 studies found stronger associations for multiday average than single-day concentrations
6 (Castro et al.. 2009: Liu et al.. 2009b: Delfino et al.. 2008a): others found similar
7 associations (Qian et al., 2009b; Lagorio et al., 2006; Schindler et al., 2001). The range of
8 mean ambient NO2 concentrations was 4.5-49.2 ppb for 24-h avg NO2 and 11.5-75 ppb
9 for l-to5-havgNO2.
10 In most studies, an association of ambient NO2 with lung function independent of
11 copollutants was not clearly demonstrated. However, some studies provided supporting
12 evidence. Some studies found associations with NO2 but not copollutants such as PM2 5,
13 PMin. EC. CO. or SO? (Oftedal et al.. 2008; Holguin et al.. 2007; Lagorio et al.. 2006). A
14 wide range of correlations were reported between NO2 and copollutants (r = 0.18-0.75).
15 In studies with copollutant modeling, NO 2-lung function associations were attenuated
16 (Liu et al.. 2009b: Qian et al.. 2009b: Schindler etal. 2001) or reported to lose statistical
17 significance (Linn et al.. 1996) with adjustment for copollutants such as PM2 5, PMi0,
18 EC, BC, SO2, and UFP where pollutants were measured at central sites. Copollutant
19 effect estimates were robust to adjustment for NO2. Among studies that measured or
20 modeled personal NO2 or measured NO2 at schools or on site of outdoor activity, most
21 found NO2 associations with adjustment for pollutants such as PM2 5, PMi0, EC, PM25
22 metal components, PNC, and O3 (Strak etal.. 2012; Delfino et al.. 2008a: McCreanor et
23 al., 2007; Moshammer et al.. 2006). The attenuation of copollutant effect estimates with
24 adjustment for NO2 in some of these studies indicated that NO2 may have confounded
25 copollutant associations (Strak etal.. 2012; Delfino et al., 2008a; Moshammer et al.,
26 2006).
4.2.4 Pulmonary Inflammation, Injury, and Oxidative Stress
27 The evidence for NO2-related increases in AHR (Section 4.2.2) shows coherence with
28 evidence indicating effects on pulmonary inflammation, as pulmonary inflammation can
29 mediate AHR (Section 3.3.2.5). Both are characteristic features of respiratory conditions
30 such as asthma and COPD, and both lines of evidence describe key events informing the
31 modes of action by which ambient NO2 exposure may lead to increases in respiratory
32 symptoms (Section 4.2.6) as well as respiratory hospital admissions and ED visits
33 (Section 4.2.7). The initiation of inflammation by NO2 exposure is supported by
34 observations of NO2-induced increases in eicosanoids, which mediate recruitment of
35 neutrophils (Section 3.3.2.3). Further, NO2-induced increases in ROS and RNS may
November 2013 4-65 DRAFT: Do Not Cite or Quote
-------
1 initiate inflammation (Section 3.3.2.1). as many transcription factors regulating
2 expression of pro-inflammatory cytokines are redox sensitive. Both pulmonary
3 inflammation and oxidative stress can cause tissue injury. However, there is not strong
4 evidence for the effects of NO2 on pulmonary oxidative stress or injury.
5 The 2008 ISA for Oxides of Nitrogen described evidence for NO2-induced increases in
6 pulmonary inflammation in some controlled human exposure studies and animal
7 toxicological studies (U.S. EPA. 2008c). There was coherence with findings from the few
8 available epidemiologic studies in children with asthma and children in the general
9 population, which found associations between short-term increases in ambient NO2
10 concentrations and increases in exhaled nitric oxide (eNO). In particular, coherence is
11 found among disciplines for NO2-associated increases in allergic inflammation. Recent
12 studies, most of which were epidemiologic, continued to find NO 2 -associated increases
13 in pulmonary inflammation, pulmonary injury, and oxidative stress. Biological indicators
14 of pulmonary inflammation, injury, and oxidative stress included those measured in
15 exhaled breath or bronchoalveolar or nasal lavage fluid. Indicators of systemic
16 inflammation in blood are evaluated in the context of cardiovascular effects in Section
17 4.3.
4.2.4.1 Controlled Human Exposure Studies
18 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) cited several studies addressing
19 the effects of NO2 exposure on markers of airway inflammation (i.e., differential cell
20 counts, cytokines, prostaglandins), injury (i.e., lactate dehydrogenase (LDH) and protein
21 concentrations), and oxidative stress (i.e., antioxidant molecules and enzymes). Study
22 details are presented in Table 4-9. but overall, the study protocol typically used in these
23 studies includes a single- or multi-day exposure to NO2 (50-5,000 ppb) followed 1 to 24
24 hours later by collection of bronchial wash or bronchoalveolar lavage fluid (BALF). The
25 coherence and biological significance of effects across studies is difficult to evaluate
26 given the variety of exposures and timing of when effects were measured, but there is
27 evidence for pulmonary inflammation that is most consistently demonstrated by increases
28 in polymorphonuclear cells (PMNs).
29 Several studies reported increases in PMNs and other inflammatory markers following
30 NO2 exposure. In a study by Frampton et al. (2002). adults exposed to 1,500 ppb NO2
31 had increased PMNs in BALF though PMNs were not statistically significantly increased
32 after exposure to 600 ppb, consistent with results from an earlier study (Frampton et al..
33 1989). No change in BALF protein concentration was reported, but lymphocytes were
34 decreased in peripheral blood and increased in BALF after 600 ppb NO2. Consistent with
November 2013 4-66 DRAFT: Do Not Cite or Quote
-------
1 Frampton et al. (2002). several studies reported an increase in PMNs in BALF or
2 bronchial wash from adults exposed to 2,000 ppb NO2 under varying exposure durations
3 and patterns (Solomon et al., 2000; Blomberg et al., 1999; Devlin et al., 1999; Azadniv et
4 al.. 1998). Other cell populations, LDH, and protein concentration were not altered
5 following NO2 exposure in these studies. In an additional study, Helleday et al. (1994)
6 found that bronchial PMNs were increased in nonsmoking adults while alveolar PMNs
7 were increased in smoking adults 24 hours after a brief exposure to 3,500 ppb NO2. With
8 respect to cytokine profiles, Devlin etal. (1999) reported increased IL-6 and IL-8 in
9 BALF from adults exposed to 2,000 ppb, whereas 2,000 ppb exposure repeated over 4
10 days did not increase expression of IL-6 and IL-8 in biopsies of the bronchial epithelium
11 (Pathmanathan et al.. 2003); however this study did find significant increases in IL-5 and
12 IL-13, both of which contribute to allergic inflammation (Section 4.2.4.3). Based on this
13 group of studies, NO2 exposure can induce pulmonary inflammation in healthy human
14 adults, though evidence does not demonstrate NO2-induced pulmonary injury.
15 The few available studies did not consistently demonstrate NO2-induced pulmonary
16 inflammation in adults with asthma. Torres etal. (1995) exposed healthy adults and those
17 with asthma to 1,000 ppb NO2 and performed bronchoscopy 1 hour later. The
18 macroscopic appearance of the bronchial epithelium was altered after exposure in adults
19 with asthma compared to healthy controls; however, this was not accompanied by any
20 changes in cell counts in the BALF. Eicosanoid levels were also measured; thromboxane
21 B2 was increased in healthy adults and those with asthma following NO2 exposure while
22 prostaglandin D2 was increased and 6-keto prostaglandin Fla was decreased after
23 exposure only in adults with asthma. Vagaggini et al. (1996) observed a decrease in
24 eosinophils in sputum collected from adults with asthma following a 1-hour exposure to
25 300 ppb NO2, though this decrease was not statistically significant.
26 Studies have also measured effects of NO2 exposure on antioxidant capacity, but results
27 across studies are mixed. Blomberg et al. (1999) found no changes in glutathione,
28 ascorbic acid, or uric acid levels following exposure to 2,000 ppb NO2. Kelly et al.
29 (1996a) examined the kinetics of antioxidant response in the respiratory tract after
30 exposure to 2,000 ppb NO2 and found reduced levels of uric acid in bronchial wash and
31 BALF 1.5 hours post-exposure, elevated levels at 6 hours, and control levels by 24 hours.
32 Ascorbic acid decreased in bronchial wash and BALF at 1.5 hours but returned to
33 baseline levels by 6 hours. Glutathione was increased at 1.5 and 6 hours in the bronchial
34 wash, but no changes in glutathione were found in the BALF or for reduced glutathione
35 and malondialdehyde at any time after exposure. These observations suggest that
36 pulmonary antioxidants are modulated by NO2 exposure.
November 2013 4-67 DRAFT: Do Not Cite or Quote
-------
Table 4-9 Controlled human exposure studies of NO2 and pulmonary
inflammation, injury, and oxidative stress.
Study
Disease status3; n,
Sex; Age
(mean ± SD)
Exposure Details
(Concentration; Duration)
Endpoints Examined
Azadniv et al.
(1998)
Blomberq et
al. (1999)
Devlin et al.
(1999)
Frampton et
al. (1989)
Frampton et
al. (2002)
Helledav et al.
(1994)
Huanq et al.
(2012b)
n = 11 M, 4F;
Early Phase:
28.1 ±3.5yr
Late Phase:
27.4 ± 4.2 yr
n = 8M, 4F;
26 yr (Range:
21-32 yr)
n = 10 M;
Range: 18-35 yr
(1)n = 7M, 2F;
30 yr (Range:
24-37 yr)
(2)n = 11 M, 4F;
25 yr (Range:
1 9-37 yr)
(1,2)n = 12M, 9F;
F=27.1 ±4.1 yr
M= 26.9 ± 4.5 yr
n = 8 nonsmokers;
Median: 26 yr
(Range: 24-35 yr),
8 smokers,
Median: 29 yr
(Range: 28-32 yr)
(1)n = 11 M, 3F
(2) n = 6 M, 7 F
(3) n = 7 M, 6 F;
24.6 ± 4.3 yr
2,000 ppbfor6 h;
Exercise for approximately 10 of
every 30 min at VE= 40 L/min
2,000 ppb, 4 h/day for 4 days;
Exercise 15 min on/ 15 min off at
workload of 75 W
2,000 ppb for 4 h;
Exercise for 15 min on/15 min off
atVE= 50 L/min
(1 ) 600 ppb for 3 h,
(2) 50 ppb for 3 h +
2,000 ppb peak for 15 min/h;
(1,2) Exercise 10 min on/20 min
off at VE= ~4 times resting
(1 ) 600 ppb for 3 h,
(2) 1,500 ppb for 3 h;
(1,2) Exercise 10 min on/20 min
off at VE= 40 L/min
3,500 ppb for 20 min;
Exercise last 15 min at 75 Watts
(1)500ppbNO2for2h,
(2) 500 ppb NO2 +
73.4±9.9|jg/m3CAPsfor2h,
(3) 89.5 ± 10.7 |jg/m3 CAPS for
2h;
(1-3) Exercise 15 min on/15 min
off at VE= 25 L/min
BALF analysis 1 h and 18 h after
exposure. Protein concentration,
differential cell counts.
Cell counts from bronchial biopsies,
BW, and BALF 1.5-h post-exposure;
protein concentration, IL-8, MPO,
hyaluronic acid, glutathione, ascorbic
acid, and uric acid in BALF and BW
1.5-h post-exposure, blood
parameters
Bronchial and alveolar lavage fluid
contents 16h post-exposure. LDH
activity, t-PA activity, IL-6 activity, IL-8
activity, PGE2 levels, total protein,
ascorbate, urate, and glutathione.
BALF cell viability and differential
counts 3.5-h post-exposure, IL-1
activity in BALF cells
Bronchial and alveolar lavage fluid
cell viability and differential counts
3.5-h post-exposure, peripheral blood
characterization
Bronchial wash and BALF analysis.
Protein concentration, differential cell
counts,
Cell counts and concentrations of
IL-6, IL-8, a1-antitrypsin, and LDH in
BALF 18-h post-exposure
November 2013
4-68
DRAFT: Do Not Cite or Quote
-------
Table 4-9 (Continued): Controlled human exposure studies of NO2 and pulmonary
inflammation, injury, and oxidative stress.
Study
Jorres et al.
(1995)
Kelly et al.
(1996a)
Pathmanathan
et al. (2003)
Riedl et al.
(2012)
Solomon et al.
(2000)
Vaqaqqini et
al. (1996)
Disease status3; n,
Sex; Age
(mean ± SD)
Healthy: n = 5 M, 3 F
27 yr (Range:
21-33 yr)
Asthma: n = 8 M, 4 F
27 ± 5 yr
n = 44;
Median: 24 yr
(Range: 19-45yr)
n = 12;
26 yr (Range:
21-32yr)
n = 31; Range:
18-50yr
n = 11 M, 4F;
29.3 ± 4.8 yr
Healthy: n = 7 M;
34 ± 5 yr
Exposure Details
(Concentration; Duration)
; 1,000 ppb for 3 h;
Exercise 10 min on/10 min off at
individual's maximum workload
2,000 ppbfor4 h;
Exercise 15 min on/15 min off at
75 W
2,000 ppb for 4 h/day for 4 days;
Exercise 15 min on/15 min off at
75 W
350 ppb for 2 h;
Exercise 15 min on/15 min off at
VE= 15-20L/min
2,000 ppb for 4 h/day for 3 days;
Exercise 30 min on/30 min off at
VE= 25 L/min
300 ppb for 1 h;
Exercise at VF= 25 L/min
Endpoints Examined
BALF analysis 1 h after exposure
(cell counts, histamine,
prostaglandins
Antioxidant concentrations and
malondialdehyde in BALF and
bronchial wash at 1.5, 6, or24-h post-
exposure
Quantification of cytokines in airway
biopsies by immunohistochemistry
Sputum sample cell count (alveolar
macrophages, lymphocytes, PMNs,
and eosinophils).
Bronchial wash and BALF analysis
immediately after exposure.
Differential cell counts, LDH,
peripheral blood parameters
Cell counts in sputum 2-h post-
exposure
Asthma: n = 4 M, 4 F;
29 ± 14 yr
COPD: n = 7 M;
58 ± 12 yr
aSubjects were healthy individuals unless described otherwise.
1
2
o
J
4
5
4.2.4.2 Toxicological Studies
Animal lexicological studies reported limited evidence of pulmonary injury with
ambient-relevant short-term exposures to NO2 but more consistently indicate effects with
long-term exposure (Section 5.2.7.2). Few studies have been published since the 2008
ISA for Oxides of Nitrogen (U.S. EPA. 2008c) and the results reported are consistent
with those from past studies. Study details are presented in Table 4-10.
6
7
Pulmonary Inflammation
Animal studies have examined similar endpoints to those in controlled human exposure
studies (Section 4.2.4.1) to assess pulmonary inflammation after NO2 exposure, but
effects of NO2 are inconsistent across disciplines. While studies in humans demonstrated
November 2013
4-69
DRAFT: Do Not Cite or Quote
-------
1 increases in BALF PMNs after NO2 exposure, several studies in animals found no
2 significant changes in BALF inflammatory cells and mediators in rodents exposed to
3 5.000 ppb NO7 for up to 7 days (Povnter et al., 2006; Muller et al., 1994; Pagani et al.,
4 1994; Mustafa et al.. 1984). Schlesinger (1987a). however, did report an increase in
5 PMNs in BALF from rabbits exposed to 1,000 ppb NO2 for 3, 7, and 14 days, though all
6 exposures included H2SO4.
7 A series of studies also investigated changes in arachidonic acid metabolism in response
8 to NO2 exposure. Robison and Forman (1993) exposed rats or rat alveolar macrophages
9 (AMs) ex vivo to NO2 at concentrations as low as 100 ppb and found that in vivo
10 exposure led to significant decreases in eicosanoid levels in as little as 4 hours, while
11 ex vivo exposure of AMs led to significant increases in cyclooxygenase and lipoxygenase
12 activity and slight, but not significant, increases in eicosanoids. Schlesinger etal. (1990)
13 studied similar endpoints in rabbits exposed to NO2 for 2 hours and found an increase in
14 thromboxane B2 in BALF at 1,000 ppb, but not at 3,000 ppb. This study also investigated
15 effects of O3 and co-exposures of NO2 and O3 and suggested that eicosanoid response is
16 more sensitive to O3 exposure than NO2.
Pulmonary Injury
17 In addition to NO2-induced changes in inflammatory cells and mediators, studies have
18 also assessed pulmonary injury at the morphologic and molecular level. For example,
19 Muller etal. (1994) did not find evidence of changes in surfactant or lipid content in
20 BALF at concentrations below 10,000 ppb; however histopathological assessments in
21 lung tissues from this study suggested morphologic changes in the respiratory airways
22 including thickened interstitium and inflammatory cell accumulation. A study published
23 by Earth etal. (1995) expanded upon these structural observations and reported
24 pulmonary injury at 10,000 ppb that includes diffuse alveolar damage, epithelial
25 degeneration and necrosis, proteinaceous oedema, inflammatory cell influx, and
26 compensatory proliferation and differentiation. Few morphologic studies have
27 incorporated ambient-relevant NO2 exposures; however, Earth etal. (1995) reported that
28 slight interstitial edema was present following a 5,000 ppb exposure for 3 days, though
29 this edema was not present after a 25-day exposure. In another study, Earth and Muller
30 (1999) also found slight modifications to the bronchiolar epithelium after 3 days of
31 exposure, though the bronchi appeared normal. The proliferative index of Clara cells
32 increased in the bronchioles and bronchi relative to air controls following a 3-day
33 exposure to 5,000 ppb, but the number of Clara cells was only increased in the
34 bronchioles; no changes were observed following a 25-day exposure. Additionally, Last
35 and Warren (1987) found increased collagen synthesis, a feature of fibrosis, in lung
36 homogenates obtained from rats exposed to 5,000 ppb NO2, which was enhanced with
November 2013 4-70 DRAFT: Do Not Cite or Quote
-------
1 concurrent exposure to H2SO4 or NaCl. Overall, short term exposure to NO2 appears to
2 induce minor morphologic changes in the respiratory tract, though long-term studies
3 (Section 5.2.10) report more profound impacts of exposure.
4 In addition to pulmonary injury observed at the morphologic level, molecular markers of
5 injury have also been described in some studies. Continuous exposure to 400 ppb NO2
6 for one week resulted in increased BALF protein in guinea pigs on a Vitamin C-deficient
7 diet (Sherwin and Carlson. 1973) while a 2,000 ppb exposure for 1-3 weeks increased
8 LDH levels in alveolar lung sections (Sherwin et al., 1972). Hatch et al. (1986) also
9 reported increased BALF protein levels in NO2 exposed Vitamin C-deficient guinea pigs.
10 Gregory etal. (1983) exposed rats to 1,000 and 5,000 ppb NO2 for up to 15 weeks and
11 found early increases in LDH in BALF. Rose etal. (1989b) did not find any changes in
12 LDH in BALF following a 6-day exposure to 5,000 ppb, though slight increases in
13 albumin were reported, suggesting mild pulmonary injury. In contrast to these studies, a
14 number of studies have shown that NO2 exposure below 5,000 ppb does not result in an
15 increase in BALF protein and LDH levels in a variety of models (Robison et al., 1993;
16 Robison and Forman. 1993; Schlesinger et al.. 1990; Last and Warren. 1987).
Oxidative Stress and Antioxidant Status
17 Oxidant gases are known to impair antioxidant defenses and contribute to oxidant stress
18 in the upper and lower airways. The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c)
19 did not discuss the toxicological evidence relating to effects of NO2 on antioxidants or
20 oxidative stress, but a limited number of previously published studies have evaluated
21 oxidative stress at ambient-relevant concentrations. For example, Ichinose et al. (1988)
22 exposed rats and guinea pigs to 400 ppb NO2 for 2 weeks and found that levels of lipid
23 peroxides and antioxidants (non-protein sulfhydryls, Vitamin C, and Vitamin E) were not
24 affected in lung homogenates. Furthermore, there was no change in activity levels of
25 antioxidant enzymes including glucose-6-phosphate dehydrogenase, 6-phosphogluconate
26 dehydrogenase, glutathione S-transferase (GSH), glutathione peroxidase (GPx),
27 glutathione reductase, and superoxide dismutase (SOD) after NO2 exposure; however,
28 combined exposure with O3 did demonstrate synergistic effects on antioxidant systems.
29 Studies have also investigated the effects of NO2 on glutathione and oxidized glutathione
30 levels in the BALF and peripheral blood. Pagani et al. (1994) found that rats exposed to
31 5,000 ppb NO2 for 24 hours had increased total and oxidized glutathione in peripheral
32 blood, though the increase in oxidized glutathione alone was not significant. Conversely,
33 significant increases in oxidized glutathione were reported in the BALF, whereas total
34 glutathione was slightly diminished, de Burbure et al. (2007) reported decreased GPx in
35 plasma immediately and 48 hours after exposure to 1,000 ppb NO2 for 28 days, whereas
November 2013 4-71 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
GPx increased in the BALF. GSH and SOD also increased in BALF after exposure,
though SOD returned to control levels by 48 hours post-exposure. Rats exposed to 5,000
ppb NO2 for 5 days also had reduced levels of GPx in plasma and increased levels of
GPx and GST in BALF. SOD also increased in BALF, but only 48 hours post-exposure.
Oxidized lipids were transiently increased immediately after exposure in BALF and were
not affected in the subacute exposure. Other studies have reported effects of NO2 on
antioxidant levels or enzyme activity, but those exposures were above ambient-relevant
concentrations of NO2.
Other studies have reported that Vitamin C or E deficiency enhances the effects of NO2
in the lung, which is plausible given that both vitamins have antioxidant activity in the
airways and neutralize reactive oxygen species. Guinea pigs with a Vitamin C-deficient
diet had increased BALF protein and lipids following exposure to 1,000 ppb NO2 for 72
hours or 4,800 ppb for 3 hours relative to air controls or guinea pigs with a normal diet
(Hatch etal.. 1986: Selgrade et al.. 1981). Additionally, exposure to 5,000 ppb for 72
hours resulted in 50% mortality in Vitamin C-deficient guinea pigs. Similarly, rats with
diets deficient in Vitamin E had increases in lipid peroxidation and protein content in
lung homogenates following a 7-day exposure to 3,000 ppb NO2 (Elsayed and Mustafa.
1982; Sevanian et al.. 1982b). Additional support for an influence of Vitamin E is
provided by observations that NO2-induced increases in BALF protein or decreases in
glutathione peroxidase activity were attenuated in animals fed Vitamin E-supplemented
diets, relative to animals not supplemented with Vitamin E (Guth and Mavis. 1986; Ayaz
and Csallany. 1978). These studies demonstrate that antioxidants, particularly Vitamin C
and E can modify the effects of NO2 on pulmonary injury in animals.
Table 4-10 Animal toxicological studies of NO2 and pulmonary inflammation,
injury, and oxidative stress.
Species (Strain);
Study Lifestage; Sex; n
Exposure Details
(Concentration; Duration)
Endpoints Examined
Barth et al.
(1995)
Rat (Sprague Dawley);
Male, n = 7/group
5,000, 10,000, and 20,000 ppb NO2
for 3 or 25 days
Histological evaluation,
morphometry, parenchymal and
vascular damage, pulmonary
arterial thickness, average
medial thickness
Barth and Rat (Sprague Dawley); 5,000, 10,000, and 20,000 ppb NO2
Muller(1999) Ma|6i n = 5/group for 3 or 25 days
Clara cell morphology, cellular
proliferation, epithelial
proliferation
November 2013
4-72
DRAFT: Do Not Cite or Quote
-------
Table 4-10 (Continued): Animal toxicological studies of NO2 and pulmonary inflammation,
injury, and oxidative stress.
Species (Strain);
Study Lifestage; Sex; n
Exposure Details
(Concentration; Duration)
Endpoints Examined
de Burbure
et al. (2007)
Greqorv et
al. (1983)
Hatch et al.
(1986)
Rat (Wistar);
8 weeks; Male,
n = 8/group
Rat (Fischer 344);
14-16 weeks;
n = 4-6/group
Guinea Pig (Hartley);
Young adult;
n =5 -16/group
(1)1,OOOppbNO2for6h/day,
5 days/week for 4 weeks;
(2) 10,000 ppb NO2 for6 h/day,
5 days/week for 4 weeks;
(3) 5,000 ppb NO2 for 6h/day for
5 days;
(1-3) Animals had selenium-deficient
or selenium-supplemented diets.
(1) 1,000 and 5,000 ppb NO2 for
7 h/day for 5 days/week for up to
15 weeks;
(2) 1,000 ppb NO2 for 0.5 h, 5,000
ppb NO2 for 1.5 h;
(3) 1,000 ppb NO2 for 3 h, 5,000 ppb
NO2for1.5h;
(4) 1,000 ppb NO2 for 0.5 h for
5 days/week for up to 15 weeks
4,800 ppb NO2 for 3 h in deficient
and normal animals; 4,500 ppb NO2
for 16 h;
BALF lipid peroxidation,
antioxidative enzyme levels,
protein concentration, cell
counts, oxidant production, and
selenium levels, peripheral blood
parameters
Histopathological evaluation,
BALF and lung homogenate
biochemical analysis (protein
concentration, LDH, glucose-6-
phosphate dehydrogenase,
alkaline phosphatase,
glutathione reductase, and
glutathione peroxidase)
BALF protein and antioxidant
concentrations
normal diets
Ichinose et
Mice (ICR), Hamster
(Golden), Rat (Wistar),
Guinea Pig (Hartley);
10 weeks; Male
400 ppb NO2, 400 ppb O3, and
400 ppb NO2 + 400 ppb O3 for
24 h/day for 2 weeks
Lipid peroxidation, antioxidative
protective enzymes, total
proteins, TEA reactants, non-
protein sulfhydryls in lung
homogenates
Last and
Warren
Rat (Sprague Dawley);
Male
5,000 ppb NO2, 1.0 mg/nr NaCI or
H2SO4, 5,000 ppb NO2 + 1.0 mg/m3
NaCI, 5,000 ppb NO2 + 1.0 mg/m3
H2SO4 for 23.5 h/day for 1, 3, or 7
days
Collagen synthesis, BALF protein
content and lavagable enzyme
activities
Muller et al.
(1994)
Rat (Sprague Dawley);
Male, n = 4
800, 5,000, and 10,000 ppb NO2 for 1
and 3 days
BALF cell counts and protein
concentration, phospholipid
component, SP-A, morphological
changes,
Mustafa et
al. (1984)
Mice (Swiss Webster);
8 weeks;
Male, n = 6/group
(1) 4,800 ppb NO2;
(2) 4,500 ppb O3;
(2) 4,800 ppb NO2 + 4,500 ppb O3;
(1-3) for 8 h/day for 7 days
Physical and biochemical lung
parameters (lung weight, DMA,
protein contents, oxygen
consumption, sulfhydryl
metabolism, NADPH generating
enzyme activities)
Ohashi et al.
(1994)
Guinea Pig (Hartley);
Female, n = 10/group
3,000 and 9,000 ppb NO2 for 6 h/day,
6 times/week for 2 weeks
Ciliary activity and morphological
observations
Pagan! et al. Rat (CD Cobs); 5,000 and 10,000 ppb NO2 for 24 h
(1994) Ma|e and 7 days
Analysis of BALF and superoxide
anion production by AMs
Povnter et al.
(2006)
Mice (C57BL/6);
n = 5/group
5,000 and 25,000 ppb NO2 for
6 h/day for 1, 3, or 5 days
Analysis of BALF and
histopathological evaluation
November 2013
4-73
DRAFT: Do Not Cite or Quote
-------
Table 4-10 (Continued): Animal toxicological studies of NO2 and pulmonary inflammation,
injury, and oxidative stress.
Study
Robison and
Forman
(1993)
Robison et
al. (1993)
Rose et al.
(1989b)
Sherwin et
al. (1972)
Sherwin and
Carlson
(1973)
Schlesinqer
(1987a)
Schlesinqer
etal. (1990)
Species (Strain);
Lifestage; Sex; n
Rat (Sprague Dawley);
Male
Rat (Sprague Dawley);
n >4/group
Mice(CD-l);
4-6 weeks;
n >4/group
Guinea Pig;
Male, n = 4/group
Guinea Pig;
Rabbit (New Zealand
White);
Male, n = 5/group
Rabbit (New Zealand
White);
Male, n = 3/group
Exposure Details
(Concentration; Duration)
100, 1,000, 5,000, and 20,000 ppb
NO2for1, 2, and 4 h
500 ppb NO2 for 8h/day for 0.5, 1 , 5,
or 10 days
(1) 1,000, 2,500, and 5,000 ppb NO2
for 6 h/day for 2 days; intratracheal
inoculation with murine
Cytomegalovirus; 4 additional days (6
h/day) of exposure
(2) re-inoculation 30 days (air) post-
primary inoculation
2,000 ppb NO2 continuously for 7, 14,
or 21 days
400 ppb NO2 continuously for 1 week
0.5 mg/m3 H2SO4 + 300 ppb NO2,
0.5 mg/m3 H2SO4 + 1,000 ppb NO2
for 2h/day for 2, 6, or 1 3 days
(1) 1,000, 3,000, or 10,000 ppb NO2
for2h;
(2) 3,000 ppb NO2 + 300 ppb O3 for
2h;
(3)100, 300, or 1,000 ppb O3 for 2 h
Endpoints Examined
Enzymatic production of
arachidonate metabolites in AMs,
cyclooxygenase products
Bal fluid cell counts and
arachidonate metabolite levels,
AM arachidonate metabolism,
respiratory burst activity, and
glutathione content
Infection 5 and 10 days post-
inoculation, histopathological
evaluation, and analysis of BALF
(LDH, albumin, macrophages)
Histopathological evaluation,
cellular damage by LDH staining
Protein concentration in BALF
Cell counts in BALF, AM function
Eicosanoids in BALF
1
2
o
5
4
5
6
7
8
9
10
4.2.4.3 Allergic Inflammation
As described in Section 4.2.2.2. controlled human exposure studies in adults with asthma
and allergy demonstrated increases in AHR in response to NO2 exposure or to NO2
followed by an allergen challenge. These observations are supported by several findings
in controlled human exposure and animal toxicological studies that NO2 exposure or
NO 2 exposure followed by an allergen challenge resulted in increased indicators of
allergic inflammation. This includes increases in Th2 cytokines and IgE and the influx
and/or activation of eosinophils and neutrophils. Results provide evidence thatNO2
exposure may lead to exacerbation of allergic airways disease (discussed below and in
Section 3.3.2.6) and the development of allergic airways disease (discussed below and in
Section 3.3.2.6). These results provide support for epidemiologic evidence of
November 2013
4-74
DRAFT: Do Not Cite or Quote
-------
1 NO2-associated increases in inflammation in children with asthma and allergy (Section
2 4.2.4.4V
Exacerbation of allergic airways disease
3 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) described several studies that
4 examined inflammatory responses in adults with mild allergic asthma who were exposed
5 to NO2 followed by a specific allergen challenge (Table 4-11). In a series of studies from
6 the Karolinska Institute in Sweden, adults at rest were exposed to air or 260 ppb NO2 for
7 15-30 minutes followed by an antigen (birch or timothy pollen) challenge 4 hours later.
8 BALF and bronchial wash fluid were collected 19 hours after allergen challenge. NO2
9 exposure for 30 minutes increased PMN in BALF and bronchial wash fluid and increased
10 eosinophil cationic protein (ECP) in bronchial wash fluid compared with air exposure
11 (Barck et al.. 2002). Reduced cell viability of BALF cells and reduced volume of BALF
12 were also reported. ECP is released by activated eosinophils, is toxic to respiratory
13 epithelial cells, and is thought to play a role in the pathogenesis of airway injury in
14 asthma. In a subsequent study, Barck et al. (2005a) exposed adults with mild allergic
15 asthma to air or NO2 for 15 minutes on day 1, and twice on day 2, for 15 minutes with
16 allergen challenges following all of the exposures. NO2 exposure induced an increased
17 level of ECP in both sputum and blood and increased myeloperoxidase levels in blood.
18 These results suggest that NO2 may prime circulating eosinophils and enhance activation
19 of airway eosinophils and neutrophils in response to an inhaled allergen. Nasal responses
20 to nasal allergen challenge were also examined following a 30-minute exposure to NO2
21 (Barck et al.. 2005b). No enhancement of nasal allergen responses was observed in adult
22 subjects. As noted in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). these
23 studies were we 11-designed and indicate that brief exposures to 260 ppb NO2 can enhance
24 allergen responsiveness in individuals with asthma.
25 Additional studies have been performed using longer NO2 exposures (Table 4-11). Wang
26 etal. (1999); Wang etal. (1995a): Wang etal. (1995b) found that exposure of adults to
27 400 ppb NO2 for 6 hours enhanced allergen responsiveness in the nasal mucosa in
28 subjects with allergic rhinitis. Mixed grass pollen was used as the challenge agent and
29 was administered immediately after the NO2 exposure. Responses included increased
30 numbers of eosinophils and increased levels of myeloperoxidase and ECP in nasal lavage
31 fluid collected 30 minutes after the allergen challenge. Witten et al. (2005) did not
32 observe enhanced airway inflammation with allergen challenge in adults with asthma and
33 allergy to house dust mite (HDM) allergen who were exposed to 400 ppb NO2 for 3
34 hours with intermittent exercise. HDM allergen was administered immediately after the
35 NO2 exposure and a decrease in sputum eosinophils was found 6 hours later (Witten et
36 al.. 2005). Sputum ECP levels were increased although this change did not reach
November 2013 4-75 DRAFT: Do Not Cite or Quote
-------
1 statistical significance. The authors suggested that their findings may be explained by a
2 decreased transit of eosinophils across the bronchial mucosa occurring concomitantly
3 with NO2-induced eosinophilic activation. Other investigators have noted that numbers
4 of eosinophils do not always correlate with allergic disease activity (Eriefalt et al.. 1999).
5 Airway mucosal eosinophilia is a characteristic feature of asthma and rhinitis; eosinophils
6 exert their effects via degranulation or cytolysis resulting in release of ECP and other
7 mediators. However under conditions favoring eosinophil cytolysis, ECP concentrations
8 may be high and numbers of eosinophils may be low.
9 A recent study of adults with mild allergic asthma also did not provide evidence of
10 NO2-induced increases in allergic asthma (Table 4-11) (Riedl et al.. 2012). Exposure to
11 350 ppb NO2 for 2 hours with intermittent exercise followed by methacholine challenge
12 1.5 hours later resulted in increased levels of blood IgM, and decreased levels of sputum
13 IgG4, IL-4, eotaxin, RANTES and fibrinogen measured 22 hours after exposure. Subjects
14 that were exposed to NO2 followed by cat allergen 1.5 hours later did not exhibit changes
15 in sputum cell counts measured 22 hours after exposure. While these results are not
16 consistent with NO2 enhancing allergen-induced airway inflammatory responses, it
17 should be noted that markers of eosinophil activation were not measured.
18 As noted in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). differing findings
19 between the studies in allergic individuals could be due to differences in timing of the
20 allergen challenge, the use of multiple or single allergen challenges, the use of BALF
21 versus sputum versus nasal lavage fluid, exercise versus rest during exposure, and
22 differences in subjects. Furthermore, study protocols varied in the timing of biological
23 sample collection post-exposure to NO2 or allergen.
24 Allergic inflammatory responses were also investigated in animal models of allergic
25 airways disease (Table 4-12). These studies involved sensitization and challenge with an
26 antigen followed by exposure to NO2. In one study in rats, which were sensitized and
27 challenged with RDM allergen, exposure to NO2 (5,000 ppb, 3 hours) enhanced specific
28 immune responses and increased the numbers of lymphocytes, neutrophils, and
29 eosinophils in the airways (Gilmour et al., 1996). In this study, the most pronounced
30 responses occurred when rats were exposed to NO2 immediately after sensitization and
31 immediately after challenge with RDM antigen. Rats exposed to NO2 twice had
32 increased levels of antigen-specific IgG and IgA and increased levels of IgE in BALF 7
33 days post-exposure to NO2. In addition, an increase in the ratio of inflammatory cells
34 (lymphocytes, neutrophils and eosinophils) to alveolar macrophages was observed 7 days
35 post-exposure to NO2 although the total number of lavagable cells was not changed.
36 In several studies in mice, which were sensitized and challenged with ovalbumin, NO2
37 exposure over several hours or days failed to increase allergic inflammatory responses.
November 2013 4-76 DRAFT: Do Not Cite or Quote
-------
1 Exposures to 700 or 5,000 ppb NO2 for 3 hours on a single day, for 2 hours on 3
2 consecutive days or for 6 hours on 3 consecutive days either reduced or had no effect on
3 indicators of eosinophil inflammation such as eosinophil counts, eosinophil peroxidase
4 activity, and total cellularity (Poynter et al.. 2006; Hubbard et al.. 2002; Proust et al..
5 2002). Other findings included decreases in IL-5 levels in the BALF at both 24 and 72
6 hours after exposure to 5,000 ppb NO2 and reductions in perivascular and peribronchial
7 cellular infiltrates after exposure to 700 ppb NO2. Others have noted that the ovalbumin-
8 induced airway inflammation in mice does not involve significant eosinophil
9 degranulation or cytolysis, which are characteristic features of asthma and allergic rhinitis
10 in humans (Malm-Erjefalt et al., 2001). This suggests that species-related differences may
11 account for NO2-induced decreases in eosinophilic inflammation seen in mouse models.
12 Mechanisms underlying the NO2-induced decrease in airways eosinophilia are unknown.
13 In summary, several high quality controlled human exposure studies of adults with
14 asthma and allergy found that exposures to 260 ppb NO2 for 15-30 minutes or 400 ppb
15 NO2 for 6 hours increased inflammatory responses to an allergen challenge. These
16 responses included increases in number and activation of eosinophils and neutrophils.
17 Allergic inflammation was also enhanced by a 3-hour exposure to 5,000 ppb NO2 in a rat
18 model of allergic airways disease, as demonstrated by increases in IgE levels and
19 numbers of eosinophils and neutrophils. These results provide evidence for NO2-induced
20 exacerbation of allergic airways disease both in the presence and absence of an allergen
21 challenge (Section 3.3.2.6).
Development of allergic airways disease
22 While the majority of studies of allergic inflammation were conducted in individuals with
23 asthma and allergy or in animal models of allergic airways disease, two studies were
24 conducted in naive individuals or animals. As reviewed in the 2008 ISA for Oxides of
25 Nitrogen (U.S. EPA. 2008c). one study examined the effects of repeated NO2 exposures
26 in bronchial biopsy tissue obtained from healthy human subjects who were exercising at a
27 light rate (Table 4-11). Exposure to 2,000 ppb NO2 for 6 hours on 4 consecutive days
28 increased expression of the cytokines IL-5, IL-10, and IL-13 and the intercellular
29 adhesion molecule ICAM-1 in bronchial epithelium (Tathmanathan et al.. 2003). IL-5 and
30 IL-13 are characteristic of a Th2 inflammatory response, with IL-5 known to promote
31 eosinophilia and IL-13 known to promote AHR and mucus production. A study in guinea
32 pigs also provides evidence for the development of pro-allergic responses since increases
33 in eosinophils were found in the nasal epithelium and submucosa following a two-week
34 exposure to 3,000 ppb NO2 (Ohashi et al.. 1994) (Table 4-12). The observed increase in
35 numbers of airway eosinophils and expression of Th2 cytokines in these two studies
November 2013 4-77 DRAFT: Do Not Cite or Quote
-------
suggest that inhaled NO2 may promote Th2 skewing and allergic sensitization (Section
3.3.2.6V
Table 4-11 Controlled human exposure studies of NO2 and allergic
inflammation.
Study
Disease status3;
Age; N; Sex
Exposure Details
Endpoints Examined
Barck et al.
(2002)
Adults with mild
asthma and
allergy to birch or
timothy pollen
Mean age: 29 yr
N = 6M, 7F
Histamine inhalation test to confirm
airway hyperresponsiveness
266 ppb NO2 for 30 min
Inhaled allergen challenge 4 h after
pollutant exposure
Albumin in serum samples
BW and BAL cell parameters-
volume recovered, cell viability, total
cell counts, macrophage
concentrations, percentage of
neutrophils, # eosinophils, # mast
cells (performed 19 h after allergen
challenge)
ECP, MPO, IL-5, IL-8, eotaxin,
ICAM-1
Barck et al.
(2005a)
Adults with mild
asthma and
allergy to birch or
timothy pollen
Mean age: 32 yr
N = 10M, 8F
260 ppb NO2
Day 1: one 15 min exposure with
bronchial challenge 4 h after exposure
Day 2: two 15 min exposures with
bronchial challenge 3 h after 2nd
exposure
Total and differential cells counts of
induced sputum and venous blood
(samples taken on morning of days
1-3)
ECP, MPO in sputum
Barck et al.
Adults with rhinitis
and mild asthma
Mean age = 31 yr
N = 9M, 7F
Seasonal allergy confirmed by positive
nasal challenge of allergen
AHR confirmed by histamine test
260 ppb NO2
Nasal allergen challenge 4h after
exposure
Total and differential cell counts and
cell viability in NAL (performed
before exposure, before allergen
challenge, and 1 h, 4 h and 18 h
after challenge)
ECP and MPO in NAL fluid and
blood
Wang et al.
(1995a):
Wang et al.
(1995b)
Adults with
seasonal rhinitis
Mean age: 26 yr
N=6M, 10F
Nasal provocation with grass pollen
allergen to confirm increase in nasal
airways resistance
(1)400ppbNO2for6h
(2) 400 ppb NO2 for 6 h + allergen
challenge
Nasal lavage for inflammatory
mediators fluid- ECP, MCT, MPO,
IL-8 (30 min after allergen
challenge)
Wang et al.
(1999)
Adults with grass
allergy
Mean age: 32 yr
N = 8M, 8F
NAR tests at rest, after saline, and
after allergen challenge to confirm
reactivity for inclusion in study
(1) 200 ug Fluticasone propionate (FP)
+ 400 ppb NO2 for 6 h
(2) Matched placebo + 400 ppb NO2
for6h
NAL-total and differential cell counts
(30 min after allergen challenge)
Immunoassay of NAL fluid- ECP,
RANTES
November 2013
4-78
DRAFT: Do Not Cite or Quote
-------
Table 4-11 (Continued): Controlled human exposure studies of NO2 and allergic
inflammation.
Study
Witten et al.
(2005)
Riedl et al.
(2012)
Pathmanathan
et al. (2003)
Disease status3;
Age; N; Sex
Adults with
asthma and HDM
allergy
Mean age: 32 yr
N = 6M, 9F
Phase 1: Adults
with mild asthma
Mean age: 37 yr
N = 10 M, 5F
Phase 2: Adults
with mild asthma
and cat allergy
Mean age: 36 yr
N=6M, 9F
Healthy adults
Mean age: 26 yr
N = 8M, 4F
Exposure Details
Inhaled allergen challenge to
determine Predicted Allergen PC20
400 ppb NO2 for 3 h w/ intermittent
exercise
2nd inhaled allergen challenge,
starting at four doubling doses less
than APC20 and doubling until 20%
decrease in FEV-i
Inhalation challenge to detect
bronchoconstrictive response (phase
1: methacholine; phase 2: cat allergen)
(1) 100ug/m3DEPfor2hwith
intermittent exercise
(2) 350 ppb NO2 control for 2 h with
intermittent exercise
2,000 ppb NO2 for 4 h/day for 4 days
Endpoints Examined
Total and differential cell counts in
induced sputum- macrophages,
lymphocytes, neutrophils and
eosinophils (samples taken at 6 and
26 h after allergen challenge)
Total counts and differential cell
counts (alveolar macrophages,
lymphocytes, PMNs, eosinophils) in
induced sputum (taken 22 h after
exposure)
Induced sputum fluid assay-
RANTES, eotaxin, ECP, IgG, lgG4,
IgA, IgM, IgE
Cat-specific IL-4, IL-5, IL-8, IL-12,
GM-CSF, IFN-Y, TNF-a, tryptase
Biomarkers in bronchial epithelium-
exotoxin, GM-CSF, Gro-a, I-CAM 1,
IL-5, IL-6, IL-8, IL-10, IL-13, total
and active NF-K(3, and TNF-a
(fiberoptic bronchoscopy after end of
last exposure)
"Subjects were healthy individuals unless described otherwise.
November 2013
4-79
DRAFT: Do Not Cite or Quote
-------
Table 4-12 Animal toxicological studies of NO2 and allergic inflammation.
Study
Species (Strain);
Lifestage; Sex; N
Exposure Details
Endpoints Examined
Gilmour et al. Rats (Brown Immunization with 100 ug antigen (D.
(1996) Norway); six farina and D. pteronyssinuss) + killed
weeks; F; Bordetella pertussis in 0.3 ml_ saline
n = 5/group Challenge with 50 ug allergen (2
weeks after immunization), followed
by:
5,OOOppbNO2for3h
Endpoints examined 7 days after
exposure:
Total and differential cell counts
from lung lavage
Antigen-specific IgG, IgA, IgE
antibodies in serum and lavage fluid
Lymphocyte proliferation
responsiveness
Proust et al. Mice (BALB/c); Immunization with injection of 10 ug
(2002) 6-7 weeks; M; OVA (day 0 and day 7)
n = 5/group Challenge with either 10 ug OVA or
saline control (day 14)
Exposure following OVA/saline
challenge:
(1) 5,000 ppbNO2
(2) 20,000 ppbNO2
Challenge to 0.1 M aerosol of
methacholine for 20 sec
Endpoints examined 24 h after
exposure:
BALF total and differential cell
counts
EPO activity
Immunoassay of IL-4, IL-5
Anti-OVA IgE and lgG1 in serum
Lung histology
Hubbard et al.
(2002)
Mice (CB57BI/6);
adult; M/F
Sensitization by weekly injections of 25
ug OVA for 3 weeks
Challenge with 20 mg/m3 OVA aerosol
for 1 h for 3 days or 10 days
Exposure following OVA aerosol
challenge:
(1)700ppbNO2for2h
(2)5,000ppb NO2for2h
Total and differential cell counts
from lung lavage (24 h after
exposure)
Histology analysis (24 h after
exposure)
Povnter et al. Mice (C57BL/6) Sensitization by 20 ug of OVA via i.p.
(2006) injections on day 0 and 7
Challenge with OVA aerosol (1% in
PBS) for 30 min on days 14-16
Exposures subsequent to OVA
challenge:
(1)5,000ppb NO2 for 6 h/day for 1, 3,
5 days
(2) 25,000 ppb NO2 for 6 h/day for 1,
3, 5 days
*Select groups given 20-day recovery
period
Methacholine challenge (0, 3.125,
12.5, 50 mg/mL in aerosol)
Endpoints examined after last day of
exposure or after 20 day recovery:
BALF-total and differential cell
counts; LDH
Histopathology analysis
mRNA levels of Gob5, MucSAC,
Th2, dendritic cell chemokine
CCL20 and eotaxin-1
November 2013
4-80
DRAFT: Do Not Cite or Quote
-------
Table 4-12 (Continued): Animal toxicological studies of NO2 and allergic inflammation.
Species (Strain);
Study Lifestage; Sex; N Exposure Details Endpoints Examined
Ohashi et al. Guinea pigs (1) 3,000 ppb NO2 for 6 h/day, Pathology of mucosal samples:
(1994) (Hartley); F; 6 day/week, for 2 weeks accumulation of eosinophils,
n = 10/group (2) g 0QO ppb NO2 for 6 h/day epithelial injury, mucociliary
6 day/week, for 2 weeks ' dysfunction (taken 24 h after end of
exposure period)
4.2.4.4 Epidemiologic Studies
1 The observations described in the preceding sections for NO2-induced increases in
2 inflammation, particularly increases in allergic inflammation, provide support for the
3 epidemiologic evidence for associations of ambient or personal NO2 with increases in
4 inflammation in children with asthma and allergy. Evidence also supports associations in
5 children in the general population but is inconsistent in adult populations. The number of
6 these epidemiologic studies has increased dramatically since the 2008 ISA for Oxides of
7 Nitrogen, and recent studies expand on previous studies with additional examination of
8 potential copollutant confounding and potential at-risk populations. Ambient NO2
9 concentrations, locations, and time periods for epidemiologic studies of pulmonary
10 inflammation and oxidative stress are presented in Table 4-13.
11 As in previous studies, the majority of evidence is for eNO. Across studies, eNO was
12 collected with a similar protocol, following the guidelines established by the American
13 Thoracic Society (ATS. 2000a). eNO assessment methods also accounted for NO in the
14 collection room, although eNO has not been shown to be a reliable indicator of exposure
15 (Section 3.2.3). Although not examined in controlled human exposure or animal
16 toxicological studies of NO2 exposure, several observations support epidemiologic
17 findings. NO2 exposure has been shown to increase some pro-inflammatory cytokines
18 and increase neutrophils and eosinophils (Sections 3.3.2.6. 4.2.4.1. 4.2.4.2). which can
19 activate inducible nitric oxide synthase or produce NO in the lung during an
20 inflammatory response (Barnes and Liew. 1995). Further, eNO commonly is higher in
21 children and adults with asthma and increases during acute exacerbations (Carraro et al..
22 2007: Jones etal.. 2001: Kharitonov and Barnes. 2000).
November 2013 4-81 DRAFT: Do Not Cite or Quote
-------
Table 4-1 3
Study3
Liu et al.
(2009b)
Barraza-
Villarreal et al.
(2008)
Delfino et al.
(2006)
Sarnat et al.
(2012)
Greenwald et
al. (2013)
Holquin et al.
(2007)
Martins et al.
(2012)
Flamant-Hulin
etal. (2010)
Linetal. (2011)
Berhane et al.
(2011)
Romieu et al.
(2008)
Patel et al.
(2013)
Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of pulmonary inflammation and oxidative
stress.
Location
Windsor, ON,
Canada
Mexico City,
Mexico
Riverside, CA
Whittier, CA
Riverside, CA
Whittier, CA
El Paso, TX and
Ciudad Suarez,
Mexico
El Paso, TX
Ciudad Juarez,
Mexico
Viseu, Portugal
Clermont-
Ferrand, France
Beijing, China
13 Southern CA
communities
Mexico City,
Mexico
New York City,
NY
Study Period
Oct-Dec 2005
June2003-June
2005
Aug-Dec 2003
July-Nov2004
Jan-Mar 2008
Mar-June 2010
2001-2002
Jan and June,
2006 and 2007
NR
June 2007
Sept 2007
Dec 2007
June 2008
Sept 2008
Sept-June
2004-2005
Jan-Oct 2004
May-June 2005
Oxide of
Nitrogen Metric
Analyzed
24-h avg NO2
8-h max NO2
24-h avg NO2
8-h max NO2
96-h avg NO2
96-h avg NO2
1-week avg NO2
1 -week avg
N02b
5-day avg NO2
24-h avg NO2
24-h avg NO2
8-h max NO2
24-h avg NO2
Mean/Median
Concentration
(PPb)
19.8
37.4
Personal: 24.3
Personal: 30.9
Central Site: 39.3
Central Site: 35.1
El Paso schools:
4.5, 14.2, Central
sites: 14.0, 18.5,
20.5
Ciudad Juarez
schools: 18.7,27.2,
Central site: None
School A: 6.5
School B: 17.5
18.2
Across 4 periods:
4.5, 3.5, 9.8, 8.2C
Schools <14.0: 10.1
Schools >14.0: 17.4
24.3
30.4
45.3
26.6
25.9
NR
35.3
Median: 23.3
Upper Percentile
Concentrations (ppb)
95th: 29.5
Max: 77.6
Max: 47.6
Max: 106
Max: 72.4
Max: 96
NR
NR
NR
Max across 4
4.6,4.0, 10.9,
periods:
9.4C
Across schools:
75th: 14.0C
Max: 19.7C
NR
NR
NR
NR
NR
NR
Max: 73.5
NR
November 2013
4-82
DRAFT: Do Not Cite or Quote
-------
Table 4-13 (Continued): Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of pulmonary inflammation and oxidative
stress.
Study3
Jalaludin et al.
(2004)
Qian et al.
(2009a)
Maestrelli et al.
(2011)
Steerenberq et
al. (2001)
Chen et al.
(2012a)
Salam et al.
(2012)
Steerenberq et
al. (2003)
Steenhof et al.
(2013)
Strak et al.
(2012)
Adamkiewicz et
al. (2004)
Weichenthal et
al. (2011)
Chimenti et al.
(2009)
Madsen et al.
(2008)
Location
Western and
Southwestern
Sydney, Australia
Boston, MA;
Denver, CO;
Madison, Wl;
New York City,
NY; Philadelphia,
PA; San
Francisco, CA
Padua, Italy
Utrecht
Bilthoven
the Netherlands
New Taipei City,
Taiwan
the Netherlands,
city NR
the Netherlands,
city NR
Steubenville, OH
Ottawa, ON,
Canada
Palermo, Sicily,
Italy
Oslo, Norway
Study Period
Feb-Dec 1994
Feb1997-Jan
1999
1999-2003
Feb-Mar1998
Oct-June 2007;
June-Nov 2009
2004-2007,
school year
May-June,
year NR
Mar-Oct 2009
Sept-Dec 2000
NR
Nov
Feb
July, year NR
Jan-June 2000
Oxide of
Nitrogen Metric
Analyzed
1 5-h avg
(6 a.m. -9 p.m.)
NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO
24-h avg NO2
24-h avg NO
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO
5-h avg NO2
5-h avg NOX
1-h avg NO2
24-h avg NO2
1-h avg NO
24-h avg NO
1-h avg NO2
7-day avg NO2
24-h avg NO2
7-day avg NO2
Mean/Median
Concentration
(PPb)
15.0
23.6
Range of Means
across seasons and
years: 20.9-37.0C
28.2C
30.2C
25.5C
7.4C
21.7
19.0
17.3C
6.3C
36
20
9.2
10.9
15
11.2
High traffic: 4.8
Low traffic: 4.6
31. T
27.1C
33.9C
NR
NR
Upper Percentile
Concentrations (ppb)
Max: 47.0
75th: 28.8
Max: 48.1
Range of 75th:
23.0-42.5C
Max: 44.7C
Max: 168C
Max: 49.5C
Max: 85.6C
NR
Max: 39.4
Max: 28.3C
Max: 34.5C
Max: 96
Max: 34
75th: 12.8, Max:
75th: 14.6, Max:
75th: 16.1, Max:
75th: 14.2, Max:
Max: 1 1
Max: 10
NR
NR
NR
NR
NR
32.9
23.8
215
70.7
November 2013
4-83
DRAFT: Do Not Cite or Quote
-------
Table 4-13 (Continued): Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of pulmonary inflammation and oxidative
stress.
Oxide of Mean/Median
Nitrogen Metric Concentration Upper Percentile
Study3 Location Study Period Analyzed (ppb) Concentrations (ppb)
Timonen et al. Amsterdam, Nov1998- 24-h avg NO2 22.T 75th: 28.7, Max:49.7c
(2004) the Netherlands June 1999
Erfurt, Germany Oct1998- 15.4C 75th: 19.6, Max:43.5c
Apr 1999
Helsinki, Finland Nov1998- 16.5C 75th: 18.9, Max:35.9c
Apr 1999
"Studies presented in order of first appearance in the text of this section.
bSubject-level exposure estimates calculated from outdoor NO2 at schools and other locations plus time activity patterns.
""Concentrations converted from ug/m3to ppb using the conversion factor of 0.532 for NO2 and 0.815 for NO assuming standard
temperature (25 °C) and pressure (1 atm).
NR = not reported.
Children with Asthma
1 Several recent and previous studies found associations between short-term increases in
2 ambient NO2 concentration and increases in pulmonary inflammation in children with
3 asthma. Children were recruited mostly from schools, supporting the likelihood that study
4 populations were representative of the general population of children with asthma.
5 Asthma was assessed as self or parental report of physician-diagnosed asthma, but studies
6 varied in whether they assessed asthma severity or required current symptoms in subjects.
7 Across studies, associations varied in strength and precision; however, most results
8 indicated a pattern of increasing eNO with increasing short-term NO2 exposure (Figure
9 4-2 and Table 4-14). Most studies analyzed multiple endpoints, pollutants, lags of
10 exposure, or subgroups; however, with a few exceptions (Liu et al.. 2009b; Barraza-
11 Villarreal et al.. 2008). there was a pattern of association found across the multiple
12 comparisons reducing the likelihood of associations found by chance alone or publication
13 bias.
14 Key evidence was provided from studies with strong NO2 exposure assessment,
15 comparison of various exposure metrics, and examination of copollutant confounding.
16 These studies examined a limited number of exposure lags but specified them a priori.
17 Across studies, associations were found with multiday averages of NO2 (i.e., 0-1 avg to
18 0-6 avg) (Figure 4-2 and Table 4-14). with Delfino et al. (2006) finding a stronger
19 association of eNO with lag 0-1 avg than lag 0 or 1 day NO2. Strong exposure
20 assessment was characterized as personal monitoring (Delfino et al.. 2006). estimation of
21 subject-level outdoor exposures based on monitoring, modeling, and daily activity
22 patterns (Martins et al.. 2012). or monitoring at schools (Greenwald et al.. 2013; Sarnat et
November 2013 4-84 DRAFT: Do Not Cite or Quote
-------
1 al.. 2012; Holguin et al.. 2007). In comparisons with central site NO2, associations with
2 eNO were similar to personal NO2 among children with asthma in Riverside and
3 Whittier, CA, although personal NO2 was examined as a 24-h avg, and central site NO2
4 (1 per community) was analyzed as an 8-h max (Delfino et al.. 2006). Based on
5 interquartile range, a 17-ppb increase in 24-h avg personal NO2 was associated with a 1.6
6 (95% CI: 0.43, 2.8)-ppb increase in eNO, and a 12-ppb increase in 8-h max central site
7 NO2 was associated with a 1.4 (95% CI: 0.39, 2.3)-ppb increase in eNO. Personal and
8 central site NO2 were moderately correlated (Spearman r = 0.46), indicating that despite
9 the potential for greater exposure measurement error due to spatial variability in ambient
10 NO2 concentrations and variation in time-activity patterns (Sections 2.5.1. 2.5.2, 2.5.3.
11 2.6.5.2). daily variation in ambient NO2 represents some daily variation in personal NO2
12 exposures that is associated with eNO. Among children with wheeze in Portugal who
13 spent more than 22 hours per day indoors, a 20-ppb increase in 1-week avg subject-level
14 NO2 was associated with a 13.9 ppb (95% CI: -12.4, 40.2) increase in eNO and a -2.6
15 unit (95% CI: -4.8, -1.4) decrease in EEC pH (Martins etal. 2012). School and home
16 indoor NO2 concentrations were nondetectable, providing support for an association with
17 ambient NO2. Further, time-weighted averages of microenvironmental NO2 have shown
18 good agreement with personal NO2 (Section 2.6.5.2).
November 2013 4-85 DRAFT: Do Not Cite or Quote
-------
Study Exposure Metric
Children with Asthma
Delfinoetal.(2006) NO2Personal
NO2Central site
Sarnatetal. (2012) NO2School
NO2Central site
Greenwaldetal. (In press) NO2School
Martinsetal. (2012) NO2Modeled
Liu etal. (2009) NO2Centralsite
Lin etal. (2011) NO2Centralsite
Barraza-Villarrealetal. (2008) NO2Centralsite
Berhaneetal. (2011) NO2Centralsite
Adults with Asthma
Qianetal. (2009)
Children in the General Population
Lin etal. (2011) NO2Centralsite
Barraza-Villarrealetal. (2008) NO2Centralsite
Berhaneetal. (2011) NO2Centralsite
Streeren berg etal. (2001)
NO2 Central Site
NO Central Site
Adults in the General Population
Straketal. (In press)
Weichenthal etal. (2011)
Adamkiewiczetal. (2004)
NO2on site
NOxon site
NO2 Central Site
NO Central Site
Exposure Subgroup
Lag
0-1 avg All subjects
No anti-inflamm med use
Anti-inflamm med use
ICSuse
Anti-LTandlCS use
All subjects
0-3 avg All subjects
NolCS use
ICSuse
All subjects
0-3 avg
0-6 avg
0
0-2 avg
0
0, 8-h max
1 -6 avg
SchoolA
School B
NO2Centralsite 0
All subjects
Placebo
Beta-agonistuse
ICSuse
Maestrellietal. (2011) NO2Centralsite 0
0 No asthma
0, 8-h max No asthma
0-6 avg
No asthma
Respiratory allergy
No Resp allergy
Urban
Suburban
Urban
Suburban
5-h
NO2 Central Site 1-h
0
0
1-h
-10 -5 0 5 10 15 20 25 30 35 40 45 50
Percent change in eNO per 20,25, or 60 ppb increase in NO2, NO, orNOx (95% Cl)a
Note: Studies are organized by study population and then generally in order of decreasing study strength (e.g., exposure
assessment method, potential confounding considered). Red=recent studies, Black=previous studies, Circles=NO2, Squares=NO,
Triangles=NOx.
"Effect estimates are standardized to a 20 ppb, 25 ppb, 30 ppb, and 60-ppb increase for 24-h avg NO2 or NO, 8-h max NO2 or NO,
1-h avg NO2 or NO, and 5-h avg NOX, respectively. Study details and quantitative results reported in Table 4-14.
Figure 4-2 Associations of personal or ambient NO2, NO, or NOX with
exhaled nitric oxide (eNO) in various populations.
November 2013
4-86
DRAFT: Do Not Cite or Quote
-------
Table 4-14 Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of
Oxide of Nitrogen Nitrogen
Metrics Analyzed Lag Day
Subgroup
Analyzed
(if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Children with Asthma
Delfino et al. Riverside, Whittier, CA
{2Q06) N _ 45] ages g_18 yr] persistent asthma
and exacerbation in previous 12 mo.
Repeated measures. Examined for 4-8 10-
day follow-up periods, 372 observations.
Recruitment in schools of nonsmokers from
nonsmoking homes. No information on
participation rate. Self report of physician-
diagnosed asthma. Mixed effects model
with random effect for subject with pollutant
concentrations centered on subject mean
and adjusted for personal relative humidity,
personal temperature, and follow-up
period. Adjustment for city, daily beta
agonist use, weekend did not alter results.
NO2-Personal 0
24-h avg 0_1 ayg
Compliance assessed
with motion detectors.
Monitoring checked
daily.
NC>2-central site 0
8-h max 0-1 avg
All subjects
All subjects
No anti-
inflammatory
medication
use, n = 14
Anti-
inflammatory
medication
use, n = 31
ICS use,
n = 19
Anti-LT + ICS
use, n = 12
1.1% (-2.0, 4.3%)
7.5% (2.0, 13%)
2.6% (-9.9, 15%)
9. 3% (3.1, 16%)
7.0% (0.23, 14%)
9.1% (-3.7, 22%)
0.98% (-5.4, 7.3%)
13% (3.8, 23%)
No quantitative results for
copollutant model.
-w/PM2s EC orOC' NO2
results robust. Decrease in
precision. Copollutant
results robust to NO2
adjustment.
Weak correlations for
personal exposures.
Spearman r= 0.20-0.31.
Stronger correlations for
. central site pollutants, r =
0.25-0.70).
November 2013
4-87
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Effect Estimate3
Oxide of Subgroup (95% Cl)
Nitrogen Analyzed Single-Pollutant
Lag Day (if applicable) Modelb
Copollutant Examination
Sarnat et al. El Paso, TX and Ciudad Suarez, Mexico
(2012)$ N = 29 per city, ages 6-12 yr, asthma and
current symptoms.
Repeated measures. Examined weekly for
16 weeks, 697 observations. Recruitment
from schools representing a gradient of
traffic, subjects from nonsmoking homes.
No information on participation rate. Self
report of physician-diagnosed asthma.
GLM with subject as random effect and
adjustment for school, temperature, relative
humidity, indoor NO. Adjustment for
medication use, cold symptoms did not
alter results.
NC>2-School outdoor 0-4 avg All subjects
No ICS use,
n = 10
ICS use,
n = 19
6.3% (2.5, 10%)
6.6% (2.6, 11%)
1.1% (-8.9, 12%)
NO2-School indoor
NO2-Central site
1 site in El Paso, TX
All 24-h avg
0.53% (0.11, 1.0%)
1.7% (-1.0, 4.5%)
w/O3: 8.8% (4.6, 13%)
No copollutant model with
PM2.5 or PM-io-2.5.
PM2.5and PMi0.2.5
associated with eNO,
weakly correlated with
'NO2.
• Spearman r = -0.39 to 0.32
for PM25;-0.24 to 0.04 for
PM 10-2.5.
Greenwald et
al. (2013)t
El Paso, TX
N = 38, mean age 10 yr, 76% Mexican-
American
Repeated measures. Examined weekly for
13 weeks, 436 observations. Recruitment
from schools in low and high traffic area.
No information on participation rate. School
record of physician-diagnosed asthma.
GLM with subject as random effect and
adjusted for school, temperature, relative
humidity, indoor NO.
NO2-School outdoor 0-4 avg
NO2-School indoor
All 24-h avg
School A
School B
-0.86% (-38, 58)
30% (-3.1, 73%)
School A
School B
-16% (-53, 47%)
14% (-19, 60%)
No copollutant model.
BC, VOCs (central site)
1 associated with eNO.
Moderate correlations with
NO2. Pearson r =
0.47-0.62.
BTEX associated with
eNO. Highly correlated with
NO2(r = 0.77).
Holquin et al. Ciudad Juarez, Mexico
(2007)* N = 194, ages 6-12 yr, 78% mild,
intermittent asthma, 58% with atopy.
Repeated measures. Examined biweekly
for 4 mo. 87% participation. Self-report of
physician-diagnosed asthma. Linear and
nonlinear mixed effects model with random
effect for subject and school adjusted for
sex, BMI, day of week, season, maternal
and paternal education, passive smoking
exposure
NO2-School outdoor
24-h avg
Homes 397 meters
from schools
0-6 avg Asthma, n = 31 No quantitative No copollutant model.
No asthma, n = results reported for Road density but not PM2
41 eNO- No or EC associated with
association was e|\|Q
reported.
November 2013
4-8
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study
Martins et al.
(2012)t
Study Population and
Methodological Details
Viseu, Portugal
N = 51, mean age 7.3 (SD: 1.1)yr, 53%
Oxide of Nitrogen
Metrics Analyzed
NO2-Subject-level
24-h avg
Oxide of
Nitrogen
Lag Day
0-6 avg
Subgroup
Analyzed
(if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
14% (-12, 40%)
EEC pH:
Copollutant Examination
For EEC pH only:
w/PM-io: 0.08 (-3.0,
3.6)
with atopy.
Repeated measures. 4 measurements over
2 different seasons. Recruitment from
urban and suburban schools. -66%
participation rate. Parental report of
wheeze in previous 12 mo. GEE adjusted
for age, sex, parental smoking, parental
education, atopy, time of visit, average
temperature, relative humidity. Also
included height, weight, older siblings,
mold/dampness in home, fireplace in
home, pets in home because changed at
least 1 pollutant effect estimate >10%.
Estimated from school
outdoor NO2, 20 city
monitors,
MM5/CHIMERE
modeling, and daily
activity patterns
-2.6% (-3.8, -1.4%) w/benzene: -1.7 (-3.6,
0.26)
w/ ethylbenzene:
-1.6 (-3.7, 0.49)
PM-io robust to NO2
adjustment. VOCs
attenuated to null.
Correlations negative or
weakly positive. Spearman
r = -0.72 to -0.55 for PM10,
-0.43 to 0.14 for various
VOCs.
Flamant-Hulin
etal. (201(m
Clermont-Ferrand, France NC>2-School outdoor 0-4 avg
N ~~ 104 mean age' 107 (SD' 0 7) yr
Cross-sectional. Recruitment from schools.
69% participation rate. Self or parental
report of lifetime asthma. For some
subjects, eNO measured up to 1 week
mother's birth region, parental education, NO2-School indoor
family history of allergy, prenatal and
childhood smoking exposure. Did not A|| 24_n avg
consider potential confounding by weather.
Asthma
No asthma
Asthma
No asthma
> 14.3 vs. <14.3
ppb NO2
log eNO
0(-0.14, 0.14)
-0.09 (-0.22, -0.04)
0(-0.13, 0.14)
-0.1 6 (-0.11, -0.20)
No copollutant model.
PM2.s, acetylaldehyde
associated with eNO.
November 2013
4-89
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Liu et al. Windsor, ON, Canada
N = 182, ages 9-14 yr
Repeated measures. Examined weekly for
4 weeks, same day of week. 672
observations. Recruitment from schools.
No information on participation rate.
Parental report of physician-diagnosed
asthma. Mixed effect model with random
effect for subject and adjusted for testing
period, temperature, relative humidity, daily
medication use.
NO2-Central site
24-h avg
Average of 2 sites.
Most subjects live
within 10 km of sites.
0
1
0-2
0
1
0-2
17% (-5.8, 47%) For TEARS only
7.7% (-12, 32%) w/PM2.5: 31% (-30, 145%)
1.5% (-32, 50%) w/SO2: 43% (-10, 126%)
PM2.5 estimate less
attenuated with NO2
111%) adjustment. High
67%) correlation with PM2.5,
334%) weak with SO2. Spearman
r =0.71 for PM25, 0.18 for
SO2.
TEARS:
48% (3.9,
22% (-11,
131% (23,
Lin et al. Beijing, China
I2011)t N = 36, ages 9-12 yr, 22% with asthma.
Repeated measures before and after
Olympics. Examined daily for five 2-week
periods. 1,581 observations. Recruitment
from school. Subjects selected from 437
initial respondents. GEE adjusted for
temperature, relative humidity, BMI,
asthma.
NO2-Central site
24-h avg
Site 650 meters from
schools.
All subjects
Asthma
No asthma
22% (18, 26%)
23% (16, 31%)
22% (18, 26%)
Asthma
No Asthma
12% (4.0, 20%)
9.5% (5.8, 13%)
w/BC: 5.6% (0.38, 11%)
w/PM2.5: 14% (9.5, 19%)
BC robust to NO2
• adjustment, PM25 reduced
but positive.
Weak to moderate
correlations with NO2.
Spearman r = 0.30 for
PM2.5, 0.68 for BC.
November 2013
4-90
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study
Barraza-
Villarreal et al.
(2008)1
Romieu et al.
(2008)t
Study Population and
Methodological Details
Mexico City, Mexico
N - 163-179, ages 6-14 yr, 54% persistent
asthma, 89% with atopy.
Repeated measures. Examined every 15
days for mean 22 weeks. 1,004
observations. Children with asthma
recruited from pediatric clinic. Children
without asthma were friends or
schoolmates, 72% atopy. Asthma severity
assessed by pediatric allergist. Linear
mixed effects model with random effect for
subject and adjusted for sex, BMI, lag 1
minimum temperature, ICS use, time.
Adjustment for outdoor activities, smoking
exposure, anti-allergy medication use, and
season did not alter results.
Mexico City, Mexico
N = 107, mean age 9.5 yr. 48% persistent
Oxide of
Oxide of Nitrogen Nitrogen
Metrics Analyzed Lag Day
NO2-Central site 0
8-h max
Monitors within 5 km
of school or home.
Spearman correlation
coeffiecient for school
vs. central site:
r = 0.21
NO2-Central site 0
8-h max
Subgroup
Analyzed
(if applicable)
Asthma,
n ~ 126
No asthma,
n = 50
Asthma,
n = 129
No asthma,
n = 45
Asthma,
n = 119
No Asthma,
n = 44
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
8.4% (7.9, 9.0%)
18% (16, 20%)
IL-8:
1.2% (1.1, 1.3%)
1.1 (0.93, 1.2%)
pH:
-0.5% (-1.5, 0.5%)
0.25% (-1.7, 2.2%)
Log MDA
0.13 (-0.10, 0.35)
Copollutant Examination
No copollutant model.
PM2.5 and O^ also
associated with eNO and
IL-8.
Weak to moderate
correlations with NO2.
Pearson r = 0.61 for PM25,
0.21 forO3.
No copollutant model.
O3, PM2s, distant to
asthma, 90% with atopy.
Repeated measures. EEC collected every
2 weeks for 2-16 weeks. 480 observations.
Recruitment from allergy clinic. No
information on participation rate. 25% EEC
samples below detection limit, assigned
random value 0-4.1 nmol. MDA associated
with wheeze and asthma medication use.
GEE model adjusted for sex, school shift,
temperature, chronological time.
Adjustment for outdoor activities, parental
smoking did not alter results.
Similar results for
1-h max and
24-h avg.
Monitors within 5 km
of school or home.
closest avenue, and 4.5-h
traffic count also
associated with MDA.
Moderate correlation with
NO2. Pearson r = 0.44 for
O3 and 0.54 for PM2.5.
November 2013
4-91
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Berhane et al. 13 Southern CA towns
(2011)* N = 2,240, ages 6-9 yr
Cross-sectional. Recruitment from schools.
Parental report of physician-diagnosed
asthma and history of respiratory allergy.
Linear regression adjusted for community,
age, sex, race/ethnicity, asthma, asthma
medication use, history of respiratory
allergy, eNO collection time, BMI
percentile, smoking exposure, parental
education, questionnaire language,
season, multiple temperature metrics, eNO
collected outdoors.
NO2-Central site
24-h avg
Sites in each
community.
Asthma,
n = 169
No asthma,
n = 2,071
-6.9% (-33, 20%)
11% (-3.2, 25%)
Respiratory
allergy,
n = 1,167
No respiratory
allergy,
n = 1,073
15% (2.6, 27%)
8.9% (-3.6, 21%)
No copollutant model.
PM2.s, PM-io, O3
associated with eNO in all
groups.
Moderate correlations with
NO2. Pearson r = 0.47 for
PM2.5, 0.49 for PM-io, 0.15
for O3.
Adults with Asthma
Qian et al.
(2009att
Boston, MA; New York, NY; Denver, CO;
Philadelphia, PA; San Francisco, CA;
Madison, Wl.
N = 119, ages 12-65 yr, 100% persistent
asthma
Repeated measures. Examined every 2-4
NO2-Central site 0
24-h avg
Average of all
monitors within 20
miles of subject
ZIP code centroid
All subjects
Placebo
Beta-agonist
use
ICS use
1.1% (0.52, 1.7)
0.79% (-0.08, 1.7)
0.86% (0.08, 1.6)
1.8% (0.62, 2.9)
w/PM10: 0.69% (-0.09, 1.8)
w/O3: 0.94% (0.43, 1.5)
w/SO2: 1.2% (0.52, 1.9)
Copollutant effect
estimates attenuated.
_ Correlations NR.
information on participation rate. Study
population representative of full cohort.
Trial of asthma medication, a priori
comparison of medication regimens. Linear
mixed effects model adjusted for age, sex,
race/ethnicity, center, season, week, daily
average temperature, daily average
humidity. Adjustment for viral infections did
not alter results.
0-3
All subjects 0.94% (0.09, 1.8%)
November 2013
4-92
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study
Maestrelli et
al. (2011 n
Study Population and
Methodological Details
Padua, Italy
N = 32, mean age 39.6 (SD: 7.5) yr, 81%
persistent asthma.
Repeated measures. Examined 6 times
Oxide of Nitrogen
Metrics Analyzed
NO2-Central site
24-h avg
2 sites in city
Oxide of
Nitrogen
Lag Day
0
Subgroup
Analyzed
(if applicable)
All subjects,
n = 32
Nonsmokers,
n = 22
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
3.1 (-89, 95)
2.9 (-120, 126)
Copollutant Examination
No copollutant model.
O3 and SO2 but not
personal or central site
PM2.5 or PM-io associated
with eNO. Correlations NR.
agonist users (>6/yr for 3 yr), diagnosis
clinically confirmed. Drop outs did not differ
from participants. GEE adjusted for daily
average temperature, humidity,
atmospheric pressure, asthma medication
use, current smoking status.
EBCpH
All subjects, (-1.1,1.1)
n = 32
Nonsmokers, 0.01 (-1.4, 1.4)
n = 22
Children in the General Population
Patel et al. New York City, NY
(2P_13):t: N = 36, ages 14-19 yr, 94% nonwhite, 50%
with asthma.
Repeated measures. EEC collected 2/week
for 4 weeks. 217 observations. Recruitment
from schools. 89-90% participation rate. A
priori recruitment of children with and
without asthma or atopy. Self-report of
physician-diagnosed asthma and
symptoms in previous 12 mo. Mixed effects
model with random effects for subject and
adjusted for school, daily average
temperature, 8-h max O^. Adjustment for
day of week and humidity did not alter
results.
NO2-Central site
24-h avg
Site 14 km from
schools.
0
0-3 avg
0
0-3 avg
EEC 8-isoprostane
1.7(0.63, 2.7) log
3.1 (1.3, 4.9) log
EBCpH
-0.05 (-0.79, 0.68)
-0.11 (-1.2, 1.0)
No copollutant model.
BC also associated with
EEC pH and 8-isoprostane.
School BC moderately to
highly correlated with NO2.
Pearson r = 0.62, 0.80.
November 2013
4-93
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Chen et al. New Taipei City, Taiwan
N = 100, mean age 10.6 (SD: 2.5 yr), 33%
asthma, 33% allergic rhinitis
Repeated measures. Examined 3-4
times/mo for 10 mo. 824 observations.
Recruitment from schools. A priori
recruitment of children with and without
asthma or atopy. Participants similar to
nonparticipants. Atopy confirmed by study
physician. Mixed effects model adjusted for
age, BMI, upper respiratory infection,
asthma/allergic rhinitis attack, asthma
medication use, temperature, humidity, day
of week, sampling time, sex, school,
parental education, parental atopy,
smoking exposure at home.
NO2-Central site 0
24-h avg 1
1 site 2.5 km from 2
schools, most homes 3
1 km of schools
No quantitative
No copollutant model.
data. N02 reported Associations found for
not to affect
eosinophils, PMNs,
monocytes, IL-8
PM2.5, O3 but not CO.
Weak to moderate to
correlations with NO2.
Pearson r = 0.61 for PM2.5,
-0.01 forO3.
Steerenberq
etal. (2001)
Utrecht and Bilthoven, the Netherlands
N — 1 9fi anpQ ft 1 ^ \/r 9ft% rpQniratnrv
disease, 20% allergy.
Repeated measures. Examined 1/weekfor
7-8 weeks. Recruitment from urban and
suburban schools. 65% participation.
Nonstandardized eNO collection. Mixed
effects model adjusted for sex, age,
#cigarettes smoked in home, presence of a
cold, history of respiratory symptoms and
allergy. No consideration for potential
confounding by meteorological factors.
NO2-Central site
NO - central site
All 24-h avg
Site within 2 km of
schools
0-6 avg Urban
Suburban
Urban
Suburban
0-6 avg Urban
Suburban
Urban
Suburban
44% (0, 88%)c
10%, p>0.05
IL-8 (units NR)
OR: 1.1, p>0.05
OR: 1.0, p>0.05
6.6% (0, 13%)c
7.3% (0, 15%)c
IL-8 (units NR)
OR: 1.1, p>0.05
OR: 0.95, p>0.05
No copollutant model.
PM-io and BS also
associated with eNO, IL-8,
uric acid, urea.
November 2013
4-94
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Steerenberq the Netherlands
et al. (2003) N = 115, ages 7-12 yr, 75% with pollen
and/or HDM atopy, 59% with asthma.
Cross-sectional. Recruitment from schools.
72% participation rate. Allergic sensitization
confirmed by skin prick test.
Nonstandardized eNO collection. Linear
regression model adjusted for age, sex,
gas cooking, unvented water heater,
smoking exposure, presence of a cold. No
consideration for potential confounding by
meteorological factors.
NO2 and NO-central
site
24-h avg
Site within 2 km of
schools
1
0-6 avg
No quantitative
data.
Lag 1 NO and NO2
associated with
eNO in group with
HDM and pollen
atopy.
No consistent
association for Lag
0-6 avg NO or
NO2.
No copollutant model. Lag
1 CO, PM2.5, pollen
associated with eNO.
Patel et al.
(2013)t
New York City, NY
N = 36, ages 14-19 yr, 94% nonwhite, 50%
with asthma.
Repeated measures. EEC collected 2/week
for 4 weeks. 217 observations. Recruitment
from schools. 89-90% participation rate. A
priori recruitment of children with and
without asthma or atopy. Self-report of
physician-diagnosed asthma and
symptoms in previous 12 mo. Mixed effects
model with random effects for subject and
adjusted for school, daily average
temperature, 8-h max O^. Adjustment for
day of week and humidity did not alter
results.
NO2-Central site
24-h avg
Site 14 km from
schools.
0
0-3 avg
0
0-3 avg
EEC 8-isoprostane
1.7(0.63, 2.7) log
3.1 (1.3, 4.9) log
EBCpH
-0.05 (-0.79, 0.68)
-0.11 (-1.2, 1.0)
No copollutant model.
BC also associated with
EEC pH and 8-isoprostane.
School BC moderately to
highly correlated with NO2.
Pearson r = 0.62, 0.80.
November 2013
4-95
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study
Study Population and
Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of
Nitrogen
Lag Day
Subgroup
Analyzed
(if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Salam et al. Anaheim, Glendora, Long Beach, Mira
(2012)t Loma, Riverside, San Dimas, Santa
Barbara, Upland, CA, Children's Health
Study
N = 940, ages 6-11 yr, 14% with asthma,
56% with respiratory allergy.
Cross-sectional. Recruitment from schools.
Subjects representative of full cohort. Two
different methods used for eNO
measurement. Linear regression model
adjusted for age, sex, ethnicity, asthma,
respiratory allergy, parental education,
smoking exposure, community, month of
eNO collection. No consideration for
potential confounding by meteorological
factors.
NO2-Central site
24-h avg
1 site per community
1-7 avg
iNOS promoter
methylation
0.40% (-1.0, 1.8%)
iNOS methylation
not strong predictor
of eNO.
No copollutant model.
PM2.5 associated with
higher iNOS promoter
methylation.
Weak correlation with NO2.
Spearman r = 0.36.
Adults in the General Population
Strak et al. Utrecht area, the Netherlands
I2Q12)t N = 31, adults ages 19-26 yr, all healthy,
nonsmoking
NO2 and NOx-on site 0-h
of outdoor activity post-
5-h avg exposure
NO2
NOX
20% (-5.4, 45%)
10% (-3.8, 24%)
w/PNC:-21%(-53, 11%)
for NO2,-6.2% (-15, 2.6%)
for NOX.
November 2013
4-96
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Steenhof et Repeated measures. Examined 3-7 times.
107 observations. Recruitment from
university. Well-defined outdoor exposures
at various sites: underground train station,
two traffic sites, farm, and urban
background site. Outcomes measured
before and after outdoor exposures. Heart
rate maintained during intermittent
exercise. Multiple comparisons could
results in higher probability of associations
found by chance alone. Mixed effects
model adjusted for temperature, relative
humidity, season, high/low pollen,
respiratory infection.
2-h
NO2
IL-6
66% (-10, 142)
NAL protein
60% (0, 121 )c
w/EC: 12% (-17, 41%) for
NO2, 1.6% (-8.0, 11%) for
NOX.
PNC results robust to
NO2/NOX adjustment. EC
and Abs attenuated.
For IL-6:
w/PNC: 95% (0, 190%)
w/OC:67%(-10, 144)
Copollutant results robust.
Robust NO2 results found
for NAL protein. Moderate
to high correlations with
NO2/NOx. Spearman r =
0.56, 0.75 for PNC, 0.74,
0.87 for Abs, 0.67, 0.87 for
EC.
Weichenthal Ottawa, Canada
etal. (2011)j N = 42] adu|ts ages 19-58 yr, from
nonsmoking homes, 95% white, 62% with
allergies, 33% with asthma
Repeated measures. Most examined 3
times. 118 observations. 1-h outdoor
exposures during cycling in low and high
traffic areas. Recruitment from public
advertisements. Mixed effects models with
random subject effect adjusted for
temperature during cycling, average heart
rate. Adjustment for relative humidity, day
of week did not affect results.
NO2-Central site
1 -h avg
1 site
Potential differential
exposure
measurement error
for personal PM
species and VOCs
and central site NO2.
1-h
4-h
Post-
exposure
-5.3% (-57, 46%)
-32% (-70, 6.7%)
No Copollutant model.
PM2 5 associated with
eNO.
Moderate correlation with
NO2. Spearman r= 0.31
for low traffic site, 0.45 for
high traffic site.
November 2013
4-97
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study Population and
Study Methodological Details
Oxide of Nitrogen
Metrics Analyzed
Oxide of Subgroup
Nitrogen Analyzed
Lag Day (if applicable)
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
Copollutant Examination
Chimenti et al. Palermo, Sicily, Italy
i2009)t N = 9] ma|e adu|ts mean age 40 (SD: 3.8)
yr, all healthy, nonsmoking.
Repeated measures. Examined during 3
outdoor races. Statistical analyses limited
to correlation analyses. No consideration
for potential confounding factors or
repeated measures.
NO2-Central site
Averaging Time NR
10 sites
No correlations
with plasma PMN
or eosinophils.
No results reported
forCC16.
No copollutant model.
Associations found with
and PM2.5.
Madsen et al. Oslo, Norway
(ZOOSJi N = 1,004, male adults ages 67-77 yr, 10%
with respiratory disease.
Cross-sectional. Recruitment from a larger
cohort to represent a range of home
outdoor NC>2. GLM adjusted for age,
alcohol consumption, smoking status, hour
of examination, respiratory disease, BMI, #
cigarettes/day, smoking exposure,
education, temperature.
NO2-Central site
NC>2-residential
estimated with
dispersion model
No information on
model validation.
0-7 avg
CC16
30% (7.8, 57%)
3.8% (-7.3, 16%)
No copollutant model.
Associations found for
PM2.5 (central site and
home).
Moderate correlation with
NC>2. Spearman r for home
= 0.59.
Adamkiewicz Steubenville, OH
et al. (2004) N = 29, adults median age 71 yr,
nonsmoking, 28% with asthma, 24% with
COPD.
Repeated measures. Examined weekly for
12 weeks. 138-244 total observations. GLM
with subject-specific intercept and adjusted
for time of day, day of week, study week,
temperature, pressure, relative humidity.
Several NO2 measurements missing.
NO2-central site
24-h avg
NO-central site
1-h avg
24-h avg
1 site
15% (-9.7, 39%)
8.4% (0.21, 17%)
14% (4.4, 24%)
NOw/PM25: 9.2% (-1.7,
20%)
• PM2.5 result robust.
Correlations NR.
Ambient NO robust to
adjustment for indoor NO.
November 2013
4-98
DRAFT: Do Not Cite or Quote
-------
Table 4-14 (Continued): Epidemiologic studies of pulmonary inflammation, injury, and oxidative stress.
Study
Timonen et al.
(2004)
Study Population and
Methodological Details
Amsterdam, the Netherlands; Erfurt,
Germany; Helsinki, Finland
N = 121, adults mean ages 71, 65, 68 yr,
100% with coronary heart disease, 18%
with asthma, 19% with COPD.
Repeated measures. Examined every 2
weeks for 6 mo. 1 ,249 total observations.
GLM adjusted for different covariates
depending on city but included time trend,
temperature, relative humidity, barometric
pressure, weekday.
Oxide of Nitrogen
Metrics Analyzed
NC>2-central site
24-h avg
1 site per city
Oxide of
Nitrogen
Lag Day
0
0-4 avg
Subgroup
Analyzed
(if applicable)
Amsterdam
Erfurt
Helsinki
Amsterdam
Erfurt
Helsinki
Effect Estimate3
(95% Cl)
Single-Pollutant
Modelb
CC16
14% (-8.4, 41%)
15% (-13, 52%)
13% (-18, 54%)
26% (-12, 80%)
0.38% (-34, 52%)
59% (-15, 195%)
Copollutant Examination
No quantitative results for
copollutant model.
PNC and PM25 associated
withCC16.
adjustment.
Weak to high correlation
with NO2. Spearman r =
0.82 Erfurt, 0.35 and 0.49
other cities.
Note: Studies are organized by population examined and then generally in order of study strength (e.g., exposure assessment method, potential confounding considered). ICS =
inhaled corticosteroid, LT = leukotrienes, GLM = generalized linear mixed effects model, GEE = generalized estimating equation, EEC = exhaled breath condensate, eNO = exhaled
nitric oxide, TEARS = thiobarbituric acid reactive substances, BMI = body mass index, IL-8 = interleukin-8, MDA = malondialdehyde, NR = not reported, PMNs = polymorphonuclear
leukocytes, HDM = house dust mite, iNOS = inducible nitric oxide synthase, NAL = nasal lavage, CC16 = clara cell protein.
"Results are presented for exhaled nitric oxide unless otherwise specified.
bEffect estimates are standardized to a 20 ppb for 24-h avg NO2 or NO, 25 ppb for 8-h max NO2 or NO, 30-ppb increase for 1-h avg or 5-h avg NO2 or NO, and a 60-ppb increase for
5-h avg NOX.
°95% Cl estimated for p = 0.05 based on reported p-value <0.05.
JRecent studies published since the 2008 ISA for Oxides of Nitrogen.
November 2013
4-99
DRAFT: Do Not Cite or Quote
-------
1 Evidence also pointed to associations of eNO in children with asthma with NO2
2 concentrations measured outside schools. Of the studies conducted in communities along
3 the Texas/Mexico border, most found NO2-associated increases in eNO. In comparisons
4 of NO2 exposure metrics, eNO was more strongly associated with outdoor school NO2
5 than central site NO2 (Sarnatetal.. 2012) and school indoor NO2 (Greenwald et al.
6 2013; Sarnatetal.. 2012) (Figure 4-2 and Table 4-14). In the Texas/Mexico study, a
7 20-ppb increase in 96-h avg NO2 concentration was associated with increases in eNO of
8 6.3% (95% CI: 2.5, 10.2%) for outdoor school, 0.5% (95% CI: 0.1,1.0%) for indoor
9 school, and 1.7% (95% CI: -1.0, 4.5%) for central site. There was evidence of association
10 with central site NO2, which was moderately to strongly correlated (Spearman r =
11 0.63-0.91) with school NO2 (Sarnat et al.. 2012). The results suggest that the central site
12 measures captured temporal variation in school based measures. However, the variability
13 in NO2 found across schools (coefficient of variation = 59%) indicates that the stronger
14 associations of eNO with school NO2 may be attributable to school measurements better
15 representing spatial variability in NO2. Spatial variability has been characterized to
16 influence exposure measurement error (Section 2.6.5.2). Holguin et al. (2007) did not
17 find an association with eNO in children with asthma in Ciudad Juarez schools. No
18 association was found in a study of children in France that had weaker methodology
19 characterized by cross-sectional design and comparison of eNO between low and high
20 NO2 (means 10.1 and 17.4, respectively for lag 1-4 day avg) (Flamant-Hulin et al.. 2010).
21 Several studies found associations of pulmonary inflammation or oxidative stress with
22 ambient NO2 measured at central sites. Included among these studies were those using
23 central sites located within 5 km of subjects' homes or schools (Lin et al.. 2011: Barraza-
24 Villarreal et al.. 2008: Romieu et al.. 2008). In particular, among children in Beijing,
25 China examined before and after the 2008 summer Olympics, NO2 measured within
26 650 meters of subjects' schools (lag 0 day of 24-h avg) was associated with eNO. NO2
27 was not associated with eNO in a cross-sectional study of children with asthma in 13
28 southern California communities (Berhane et al.. 2011). Central site NO2 was associated
29 with indicators of inflammation such as IL-8 and exhaled breath condensate pH and
30 indicators of oxidative stress. Among children in Windsor, ON, Canada, an increase in
31 24-h avg NO2 (measured at one of two sites within 10 km of subjects' homes) was
32 associated with increases in exhaled breath condensate thiobarbituric acid reactive
33 substances (TEARS), an indicator of lipid peroxidation, weakly associated with eNO, and
34 not associated with another lipid peroxidation indicator, 8-isoprostane (Liu et al.. 2009b).
35 With regard to confounding, most studies adjusted for temperature and humidity, with a
36 few additionally evaluating asthma medication use (Sarnatetal.. 2012: Liu et al.. 2009b:
37 Delfino et al.. 2006). Most studies found associations with copollutants such as PM25,
November 2013 4-100 DRAFT: Do Not Cite or Quote
-------
1 PM10, PM10_2.5, EC, organic carbon (OC), VOCs, O3, and SO2. These copollutants
2 showed a wide range of correlations withNO2 (Pearson or Spearman r = 0.18 to 0.80).
3 Among studies with personal or school NO2 monitoring, some provided evidence for the
4 effects of NO2 independent from other pollutants. Among children with asthma in
5 southern California, robust eNO-NO2 associations were found with adjustment for
6 personal PM25, EC, and OC, which were weakly correlated with personal NO2
7 (Spearman r = 0.20-0.33) (Delfino et al.. 2006). Effect estimates for copollutants were
8 robust to adjustment for NO2. Reporting copollutant-adjusted results only for EEC pH,
9 Martins et al. (2012) found that associations for subject-level estimates of outdoor NO2
10 exposure were relatively robust to adjustment for VOCs, which showed no or negative
11 correlations with NO2 (range of Spearman correlation coeffiecient across four visits:
12 r = -0.42 to 0.03). VOC estimates were attenuated to the null with adjustment for NO2.
13 NO2 effect estimates were attenuated to null with PMi0, but PMi0 and NO2 were
14 negatively correlated (r = -0.55 to -0.82). The studies conducted in El Paso, TX and
15 Ciudad Juarez, Mexico did not analyze copollutant models with the copollutants
16 associated with eNO (Greenwald et al.. 2013; Sarnatetal.. 2012). In the El Paso schools,
17 increases in eNO were found with increases in both school NO2 and BTEX (benzene,
18 toluene, ethylbenzene, xylene, traffic-related VOCs) (Greenwald et al.. 2013). Because of
19 the high correlation (Pearson r = 0.77) between NO2 and BTEX, an independent
20 association for NO2 is not discernible. However, analyses of the combined El
21 Paso/Ciudad Juarez population indicated an independent association of ambient school
22 NO2 with eNO. BTEX was not analyzed, but PM10, PM 10-2.5, and PM2 5 were weakly
23 correlated with NO2 in most schools (r = -0.28 to 0.34) (Sarnatetal., 2012). Further,
24 NO2 associations were less variable across schools than were copollutant associations,
25 and in a school in Ciudad Juarez, NO2 but not copollutants, was associated with eNO. An
26 independent association for NO2 was not clearly demonstrated for children in Windsor,
27 ON, Canada because of a high NO2-PM2 5 correlation (r = 0.71) (Liu et al.. 2009b). In a
28 copollutant model, the effect estimates for NO2 were attenuated somewhat with
29 adjustment for SO2 and largely with adjustment for PM25 (Figure 4-2 and Table 4-14).
30 Smaller changes were found in the SO2 and PM25 estimates with adjustment for NO2.
31 Studies of children with asthma did not clearly identify potential factors that may modify
32 ambient NO2-associated increases in pulmonary inflammation, primarily based on post-
33 hoc analyses. Associations were not found to differ by sex (Sarnatetal.. 2012; Liu et al..
34 2009b; Delfino et al., 2006). Results did not clearly indicate whether use of anti-
35 inflammatory ICS influences ambient NO2-associated increases in pulmonary
36 inflammation. Larger associations were found in children with daily ICS use (Delfino et
37 al.. 2006). but in children not using ICS in other studies (Sarnatetal.. 2012; Liu et al..
38 2009b). The latter studies did not report frequency of ICS use. Because of the
November 2013 4-101 DRAFT: Do Not Cite or Quote
-------
1 heterogeneity in the definition of ICS use and lack of assessment of ICS compliance, ICS
2 use could represent well-controlled or more severe asthma across populations, which
3 could contribute to the inconsistent evidence for effect modification by ICS use. Several
4 studies specified comparisons between children with and without asthma a priori. While
5 children with asthma had higher eNO, results indicated no difference in associations with
6 NO2 between groups (Patel et al.. 2013; Lin etal.. 2011; Flamant-Hulin et al.. 2010;
7 Holguin et al., 2007) or larger associations in children without asthma (Berhane et al.,
8 2011; Barraza-Villarreal et al.. 2008V
Adults with Asthma
9 In the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). there were no epidemiologic
10 studies of pulmonary inflammation specifically in adults with asthma. Recent studies,
11 both of which examined predominately adults with persistent asthma and central site NO2
12 and adjusted for temperature and humidity, had contrasting results. A U.S. multicity
13 (Boston, MA; New York, NY; Philadelphia, PA; San Francisco, CA; Madison, WI) study
14 nested within an asthma medication trial found an association with eNO that was robust
15 to copollutant adjustment (Qian et al.. 2009a). Among all subjects, a 20-ppb increase in
16 lag 0 day of 24-h avg NO2 (averaged from monitors located within 20 miles of subjects'
17 homes) was associated with a 0.26 ppb (95% CI: 0.12, 0.40) increase in eNO. A similar
18 increase in eNO was found for lag 0-3 day avg NO2 but not lags 1, 2 or 3. A larger effect
19 was estimated in the daily ICS group than the placebo or beta-agonist groups only for lag
20 0 day NO2. Among children and adults with asthma in Padua, Italy, a large percentage of
21 whom reported ICS use, lag 0 day of 24-h avg ambient NO2 was not associated with eNO
22 or exhaled breath condensate pH (Maestrelli et al.. 2011).
23 The U.S. multicity study provided evidence for the independent effects of ambient
24 NO2-assoiciated exposure. eNO was associated with 24-h avg SO2, weakly associated
25 with 24-h avg PMi0, but not with 24-h avg O3. The NO2-eNO association was slightly
26 attenuated (0.16 ppb [95% CI: -0.02, 0.34] increase per 20-ppb increase in lag 0 day
27 NO2) with adjustment for 24-h avg PMi0 and increased with adjustment for O3 and SO2
28 (Qian et al., 2009a). In turn, adjustment for NO2 attenuated effect estimates for the
29 examined copollutants, indicating that the copollutant associations were confounded by
30 NO2.
Children in the General Population
31 Together with the few studies described in the 2008 ISA for Oxides of Nitrogen (U.S.
32 EPA. 2008c). most recent studies found associations between increases in ambient NO or
33 NO2 and increases in pulmonary inflammation or oxidative stress in populations of
November 2013 4-102 DRAFT: Do Not Cite or Quote
-------
1 children not restricted to those with asthma. Many of these studies also examined
2 children with asthma and similarly analyzed multiple endpoints, pollutants, lags of
3 exposure, or subgroups. With a few exceptions (Patel et al.. 2013; Steerenberg et al..
4 2001). there was a pattern of association found across the multiple comparisons reducing
5 the likelihood of associations found due to chance alone or publication bias. Results were
6 consistent for eNO (Figure 4-2 and Table 4-14). In the Southern California Children's
7 Health Study, cross-sectional analyses indicated that NO2 was associated with eNO
8 (Berhane et al.. 2011) but not methylation of inducible nitric oxide synthase (Salam et al..
9 2012). which is the enzyme thought to induce increases in eNO. However, these results
10 do not appear to be discordant since methylation was not associated with eNO. Based on
11 more limited examination, associations were found with IL-8 (Barraza-Villarreal et al..
12 2008; Steerenberg et al.. 2001) and exhaled breath condensate 8-isoprostane (Patel et al..
13 2013). Associations were not found with exhaled breath condensate pH or inflammatory
14 cells such as PMNs or eosinophils (Patel et al.. 2013; Chen et al.. 2012a; Barraza-
15 Villarreal et al.. 2008; Steerenberg et al.. 2001). Ambient NO2 was associated with
16 pulmonary inflammation and oxidative stress in populations of children with prevalence
17 of asthma ranging from 7.5 to 59% and allergy ranging from 20 to 89% (Patel et al..
18 2013; Berhane et al.. 2011; Lin etal.. 2011; Steerenberg etal.. 2003; Steerenberg et al..
19 2001). With the exception of Flamant-Hulin et al. (2010). studies demonstrated
20 associations in groups without asthma or allergy (Berhane etal.. 2011; Lin etal.. 2011;
21 Steerenberg et al.. 2003). suggesting that increases in ambient NO2 exposure may
22 increase pulmonary inflammation in healthy children.
23 NO- and NO2-(24-h avg) associated increases in pulmonary inflammation were reported
24 in some (Lin etal.. 2011; Steerenberg et al.. 2003; Steerenberg et al.. 2001) but not all
25 (Chen etal.. 2012a; Flamant-Hulin et al.. 2010; Holguin et al.. 2007) studies with
26 exposure ascertained from schools or central sites located 0.65 to 2 km from subjects'
27 schools. However, Chen etal. (2012a) did not provide quantitative results, and results
28 from Flamant-Hulin et al. (2010) were based on cross-sectional comparisons of high and
29 low school-based NO2.
30 With the exception of Steerenberg et al. (2001). repeated measures studies adjusted for
31 temperature (Patel etal.. 2013; Lin etal.. 2011; Barraza-Villarreal et al.. 2008). Lin et al.
32 (2011) additionally adjusted for relative humidity. Many cross-sectional studies adjusted
33 for age, sex, parental education, and smoking exposure (Salam etal.. 2012; Berhane et
34 al., 2011; Flamant-Hulin et al.. 2010; Steerenberg et al.. 2003). A limitation of the
35 evidence in children overall was the uncertainty in discerning an independent association
36 of NO2 exposure among multiple pollutants examined. Studies found increases in
37 pulmonary inflammation and oxidative stress in association with copollutants such as
38 PM25, PM10, BC, O3, SO2, carbon monoxide (CO), and pollen. In the few studies that
November 2013 4-103 DRAFT: Do Not Cite or Quote
-------
1 reported copollutant correlations, weak correlations were reported for O3 (r = 0.15, 0.21)
2 (Berhane et al.. 2011; Barraza-Villarreal et al.. 2008). and moderate to strong correlations
3 were reported for PM and BC (r = 0.47-0.80) (Patel et al.. 2013; Berhane et al.. 2011;
4 Barraza-Villarreal et al.. 2008). Evidence for an independent association with NO2 was
5 found among children in Beijing, China examined before and after the 2008 Olympics in
6 a study that was noteworthy also for the large number of measurements per child and
7 measurement of pollutants at a site 650 meters from schools (Lin et al.. 2011). NO2 effect
8 estimates adjusted for BC or PM2 5 were attenuated 2 to 4 fold but remained positive
9 (e.g., 22% [95% CI: 18, 26%] increase in eNO per 20-ppb increase in lag 0 day of
10 24-h avg NO2 to 5.6% [95% CI: 0.38, 11%] with adjustment for BC). Adjustment for
11 NO2 attenuated the association with PM2 5 but not BC. These results indicated that some
12 of the NO2 association was confounded, by BC in particular, but NO2 was also
13 independently associated with eNO in this population of children in Beijing.
Adults in the General Population
14 Among a few studies reviewed in the 2008 ISA for Oxides of Nitrogen that were
15 conducted in older adults (U.S. EPA. 2008c) and recent studies conducted in older adults
16 and adults performing outdoor exercise, results point to increases in pulmonary
17 inflammation more clearly in association with increases in 24-h avg ambient NO or NO2
18 than NOX or NO2 averaged up to 5 hours. Copollutant-adjusted associations were found
19 with 24-h avg NO (Adamkiewicz et al.. 2004) and with 5-h avg NOX or NO2 for some
20 outcomes (Steenhof et al.. 2013; Strak et al.. 2012). Pulmonary inflammation was
21 indicated as increases in eNO, nasal lavage IL-6, and indicators of pulmonary injury and
22 lung permeability such as Clara cell protein (CC16) and nasal lavage protein levels. The
23 findings for pulmonary injury and lung permeability have weak support from controlled
24 human exposure and animal toxicological studies as that evidence was inconsistent
25 (Sections 4.2.4.1. 4.2.4.2). With respect to species of oxides of nitrogen, associations
26 were found with ambient NO, NO2, and NOX.
27 Ambient NO2 was not consistently associated with increases in pulmonary inflammation
28 in populations of mostly healthy adults performing outdoor exercise for <1 to 5 hours. A
29 strength of these studies is the examination of effects in subjects while outdoors, whose
30 exposures are likely to be more correlated with ambient NO2 measured at central sites
31 than they are for individuals in indoor locations. In these studies, subjects had 3-5
32 separate outdoor exposure periods, in some cases, in locations representing a gradient of
33 traffic volume. Among adults running or cycling outdoors for 35-90 minutes, eNO and
34 inflammatory cell counts were not associated with NO2 measured at central sites
35 fWeichenthal et al.. 2011; Chimenti et al.. 2009) (Figure 4-2 and Table 4-14). However,
36 increases in eNO and nasal lavage IL-6 and protein were found in healthy adults in
November 2013 4-104 DRAFT: Do Not Cite or Quote
-------
1 association with 5-h avg NOX and NO2 measured on the site of outdoor exposures
2 (Steenhof et al.. 2013; Strak et al.. 2012). which account for spatial variability better than
3 central site measurements. Increases in eNO and nasal lavage IL-6 and protein were
4 found immediately after and 2 hours after exposures ended but not the morning after,
5 indicating a transient increase in pulmonary inflammation. Further, the multiple analyses
6 conducted across pollutants, including several PM2 5 components, increases the potential
7 for associations to be found by chance alone (Strak et al.. 2012). eNO also was associated
8 with EC, Absorbance coefficient (Abs), and PNC (Strak et al.. 2012). whereas IL-6 also
9 was associated with PM2 5 and OC (Steenhof et al.. 2013). Independent associations for
10 NOX and NO2 were not clearly demonstrated because of high copollutant correlations
11 (e.g., r = 0.75 for NOX and PNC and 0.71 for NO2 and EC). In copollutant models,
12 associations of eNO with NOX and NO2 were attenuated with adjustment for EC or Abs
13 and became negative with adjustment for PNC (Strak et al.. 2012). The PNC effect
14 estimate was robust to adjustment for NOX or NO2. However, the associations of nasal
15 lavage IL-6 and protein with NO2 remained after adjustment for PNC or other
16 copollutants (e.g., 66% [95% CI: -10, 144%] increase in IL-6 per 30-ppb increase in
17 5-h avg NO2 and 95% [95% CI: 0, 190%] with adjustment for PNC). Thus, in this study
18 of well-defined outdoor exposures, there was evidence of confounding of NO2-eNO
19 associations by PNC but independent associations of NO2 with IL-6 and nasal lavage
20 protein.
21 Increases in pulmonary inflammation were associated with 24-h avg NO or NO2
22 measured at central monitoring sites among older adults (mean ages: 65-71 years)
23 (Madsen et al.. 2008; Adamkiewicz et al.. 2004). including those with coronary heart
24 disease (Timonen et al.. 2004). Multiday averages of NO2 (e.g., lag 0-4 day avg, 0-7 day
25 avg) were associated with CC16 (Madsen et al.. 2008; Timonen et al.. 2004). However,
26 there was uncertainty regarding independent associations of NO2 as Madsen et al. (2008)
27 found an association with central site not home NO2, and each study found associations
28 with other pollutants. There was evidence of an independent association between lag 0 of
29 24-h avg NO and eNO among older adults in Steubenville, OH (Adamkiewicz et al..
30 2004). In a copollutant model, the NO effect estimate decreased, and the 24-h avg PM2 5
31 effect estimate increased. However, the NO effect estimate remained positive.
November 2013 4-105 DRAFT: Do Not Cite or Quote
-------
4.2.4.5 Summary of Studies of Pulmonary Inflammation, Injury,
and Oxidative Stress
1 Overall, results from recent epidemiologic studies augmented the body of evidence
2 reviewed in the 2008 ISA for Oxides of Nitrogen indicating associations between short-
3 term increases in ambient NO2 exposure and increases in pulmonary inflammation in
4 children with asthma. There was heterogeneity in the strength and precision of
5 associations, but there was a pattern of association found across the various lags of
6 exposure and endpoints examined. A majority of the evidence was for NO2-associated
7 increases in eNO (Figure 4-2 and Table 4-14). The studies of adults with asthma
8 produced contrasting results. The epidemiologic observations are supported by
9 demonstrations of allergic inflammation in several (but not all) controlled human
10 exposure studies of adults with asthma and allergy following exposures to allergen plus
11 260 ppb NO2 for 15-30 minutes or 400 ppb NO2 for 6 hours. Although these findings are
12 in adults with asthma and allergy, the allergic inflammation was characterized by an
13 increase in eosinophil number and activation of eosinophils and/or neutrophils, both of
14 which have been linked with NO production during an inflammatory response. Allergic
15 inflammation was also enhanced by a 3-hour exposure to 5,000 ppb NO2 in a rat model
16 of allergic airways disease. These results provide evidence for NO2-induced exacerbation
17 of allergic airways disease both in the presence and absence of an allergen challenge
18 (Section 3.3.2.6). Observations of increased numbers of airway eosinophils and
19 expression of Th2 cytokines in healthy individuals or guinea pigs following exposure to
20 2,000-3,000 ppb NO2 provide evidence that repeated or prolonged exposure to NO2 may
21 lead to the development of a pro-allergic inflammatory response and promote Th2
22 skewing and allergic sensitization (Section 3.3.2.6). Epidemiologic studies generally did
23 not find NO 2-associated changes in inflammatory cell counts in children with asthma.
24 Results from controlled human exposure studies demonstrated NO 2-induced pulmonary
25 inflammation in healthy adults, most consistently as increases in PMNs in healthy adults.
26 Whereas epidemiologic studies did not consistently find NO 2-associated increases in
27 pulmonary inflammation in adults (examined during outdoor exercise), results were
28 consistent in children in the general population and older adults. In contrast, pulmonary
29 inflammation was not consistently affected in experimental animals with a wide range of
30 NO2 exposures (800-5,000 ppb for 6 hours to 2 weeks).
31 Rather than increases in oxidative stress, in the limited available controlled human
32 exposure studies, results indicated NO2-induced changes in antioxidant concentrations in
33 BALF. Such observations also were made in experimental animals, but changes in
34 antioxidant capacity appeared to be transient. Further, there was heterogeneity in animal
35 toxicological studies that made it difficult to draw conclusions across studies. A few
November 2013 4-106 DRAFT: Do Not Cite or Quote
-------
1 epidemiologic studies reported associations between ambient NO2 concentrations and
2 indicators of lipid peroxidation in exhaled breath condensate of children with asthma and
3 children in the general population. Across disciplines, there were a few observations of
4 NO2-induced pulmonary injury as demonstrated by alterations in epithelial barrier
5 function. However, in experimental studies, results were inconsistent at ambient-relevant
6 NO2 exposures. Effects are concentration dependent and observed at concentrations of
7 NO2 above 5,000 ppb as discussed in Section 3.3.2.4. However, this evidence combined
8 with that from morphologic studies suggest the slight injury to the airway can result from
9 short-term NO2 exposure.
10 With respect to potential at-risk populations, several epidemiologic studies did not find
11 larger NO 2 -associated increases in pulmonary inflammation or oxidative stress among
12 children with asthma than children without asthma (Patel et al.. 2013; Berhane et al..
13 2011; Lin etal.. 2011; Flamant-Hulin et al.. 2010: Barraza-Villarreal et al.. 2008).
14 Evidence did not consistently indicate that NO 2 -associated pulmonary inflammation
15 differed by sex (Sarnat etal.. 2012: Liu et al.. 2009b: Delfino et al.. 2006). ICS use
16 (Sarnat etal.. 2012: Liu et al.. 2009b: Oian et al.. 2009a: Delfino et al.. 2006). or
17 respiratory allergy (Sarnat et al.. 2012: Berhane etal.. 2011: Steerenberg et al.. 2003).
18 A majority of the epidemiologic evidence for pulmonary inflammation and oxidative
19 stress was for NO2; however, associations also were found with NO and NOX. With
20 respect to averaging times, most evidence was for 24-h avg, with a few results indicating
21 associations with 8-h max NO2. Associations with shorter averaging times (i.e., <1 to 5
22 hours), examined primarily in adults performing outdoor exercise, were inconsistent. The
23 ranges of mean concentrations were 6.5-45 ppb for outdoor 24-h avg NO2, 4.5-39 ppb for
24 24-h avg personal NO2, and 6.3-30 ppb for 24-h avg NO. Increases in pulmonary
25 inflammation and oxidative stress were found with single-day NO2 lags of 0 or 1 day and
26 multiday averages of 2- to 7-days. Increases in pulmonary inflammation were found 0 to
27 2 hours after outdoor exposures (Steenhof et al.. 2013: Strak etal.. 2012). Among studies
28 that compared various lags, several found larger associations for multiday average (e.g.,
29 0-1 avg to 0-7 avg) than single-day NO2 concentrations (e.g., 0 or 1) (Patel etal.. 2013:
30 Liu et al.. 2009b: Madsen et al.. 2008: Delfino et al.. 2006: Timonen et al.. 2004:
31 Steerenberg et al.. 2001).
32 The evidence in children with asthma is substantiated by several studies with strong
33 exposure assessment characterized by personal monitoring, modeling individual outdoor
34 exposures, or school-based monitoring. Among studies that compared various exposure
35 assessment methods, Delfino et al. (2006) found similar associations with 24-h avg
36 personal and 8-h max central site NO2, and Sarnat et al. (2012) found stronger
37 associations for school than central site NO?. Observations that indoor school NO? was
November 2013 4-107 DRAFT: Do Not Cite or Quote
-------
1 not associated with pulmonary inflammation (Greenwald et al.. 2013; Sarnat etal.. 2012)
2 or that home or school indoor NO2 concentrations were negligible (Martins et al.. 2012)
3 provide additional support for associations of pulmonary inflammation with ambient
4 NO 2. Increases in pulmonary inflammation in adults also were associated with higher
5 NOX and NO2 measured at the locations of outdoor exposure (Strak etal.. 2012) and in
6 children with NO2 or NO measured 0.65 to 2 km of schools (Lin et al.. 2011; Steerenberg
7 etal.. 2003; Steerenberg etal.. 2001).
8 Most epidemiologic studies found NO2-associated increases in pulmonary inflammation
9 and oxidative stress with adjustment for temperature and relative humidity. Because
10 pulmonary inflammation and oxidative stress were associated with other pollutants (e.g.,
11 PM2 5, EC, other PM constituents) that were correlated with NO2 in many studies, an
12 association of pulmonary inflammation specifically with NO2 exposure was not
13 identified in all studies. However, among the small group of studies that examined
14 copollutant confounding, most demonstrated an independent association with NO2.
15 Particularly among children with asthma, associations of pulmonary inflammation with
16 personal or modeled NO2 exposure were relatively robust in copollutant models with
17 PM25, PMio, EC or VOCs (Martins etal.. 2012; Delfino et al.. 2006). With few
18 exceptions (Strak et al.. 2012; Liu et al.. 2009b). studies found increases in pulmonary
19 inflammation in association with NO2 measured at central sites or at the locations of
20 outdoor exposures, after adjusting for BC, PMio, PM2 5, PNC, SO2, or O3 (Steenhof et
21 al..2Q13; Lin etal.. 2011; Qian et al.. 2009a; Adamkiewicz et al.. 2004). Copollutant
22 associations adjusted for NO2 were robust in some cases (Lin etal.. 2011; Delfino et al..
23 2006; Adamkiewicz et al.. 2004) and attenuated in other cases (Steenhof et al.. 2013;
24 Martins etal.. 2012; Qian et al.. 2009a). Thus, in some studies, NO2 appeared to
25 confound copollutant associations. Overall, results from copollutant modeling provided
26 evidence for the effects of NO2 on pulmonary inflammation independent of the effects of
27 other ambient pollutants. Sarnat et al. (2012) found increases in eNO in association with
28 outdoor and indoor school NO2. The correlations between NO2 and copollutants differed
29 between the indoor and outdoor environments for BC, PM, and SO2, suggesting that NO2
30 may exist as part of a different pollutant mixture in the indoor and outdoor environments.
31 Thus, the coherence of evidence for eNO related to indoor and outdoor NO2 exposure
32 further supports the independent effects of NO2 exposure on pulmonary inflammation.
4.2.5 Host Defense
33 The respiratory tract is protected from exogenous pathogens and particles through a
34 variety of lung host defense mechanisms that include mucociliary clearance, particle
35 transport and detoxification by alveolar macrophages, and innate and adaptive immunity.
November 2013 4-108 DRAFT: Do Not Cite or Quote
-------
1 Animal toxicological studies provide clear evidence for NO2-induced susceptibility to
2 bacterial or viral infection with some coherence with results from controlled human
3 exposure and epidemiologic studies. Providing mechanistic support for these
4 observations, some toxicological studies show NO2-induced impairments in alveolar
5 macrophage function as characterized by decreased superoxide anion release and
6 diminished phagocytic activity. Effects on mucociliary clearance are not in a consistent
7 direction, but the exact mechanism by which it may impair host defense is not well
8 characterized. Study details for animal toxicological and controlled human exposure
9 studies are presented in Table 4-15 and Table 4-16. respectively. Study details for
10 epidemiologic studies are presented in Table 4-18, Table 4-21. and Table 4-22.
4.2.5.1 Susceptibility to Bacterial or Viral Infection
Toxicological Studies
11 A large body of evidence, provided by studies reviewed in the 2008 ISA for Oxides of
12 Nitrogen (U.S. EPA. 2008c). demonstrates increased susceptibility of rodents to viral or
13 bacterial infection following short-term NO2 exposure. These studies used a variety of
14 experimental approaches that generally includes exposing animals to NO2 or filtered air
15 and then combining treatment groups for a brief exposure to an aerosol of a viable agent,
16 such as Streptococcus zooepidemicus, Streptococcus pyogenesi, Staphylococcus aureus,
17 and Klebsiellapneumoniae. The majority of studies measured mortality over a specified
18 number of days following the challenge, but several studies also examined endpoints
19 such as bacterial counts and clearance. While there are differences in sensitivity across
20 species to various infectious organisms, host defense mechanisms are shared and the
21 infectivity model is well accepted as an indicator of impaired or weakened pulmonary
22 defense.
23 In a series of studies, Goldstein et al. (1974); 1973) examined bactericidal activity and
24 clearance in mice challenged with radiolabeled Staphylococcus aureus either before or
25 after NO2 exposure. The number of bacteria deposited in the lung was not different in
26 NO2-exposed animals compared to controls; however, dose-dependent decreases in
27 bactericidal activity were observed in animals exposed to NO2 for 4 hours after challenge
28 as well as those exposed to NO2 for 17 hours before challenge. While the 4-hour
29 exposure did not yield significant differences compared to air controls at NO2
30 concentrations less than 7,000 ppb, the 17-hour exposure preceding challenge was
31 significant for concentrations greater than 2,300 ppb. Parker etal. (1989) also used
32 radiolabeled bacteria to determine effects of NO2 on susceptibility to infection; this study
33 demonstrated that a 4-hour exposure to 5,000 ppb NO2 was sufficient to reduce
November 2013 4-109 DRAFT: Do Not Cite or Quote
-------
1 bactericidal activity and increase the number of bacteria in the lungs of C3H/HeN and
2 C57BL/6N mice 3 days after challenge compared to control mice, but did not result in an
3 increase in incidence or severity of lung lesions. These results were corroborated in a
4 similar study published by Davis etal. (1991).
5 It is also important to consider differences in response to NO2 that are specific to the
6 infectious organism as Jakab (1988) has demonstrated. A 4-hour exposure to 5,000 ppb
7 NO2 resulted in a decrease in bactericidal activity after challenge with Staphylococcus
8 aureus; however, bactericidal activity against Proteus mirabilis and Pasteurella
9 pneumotropica was not impaired with exposure to NO2 at concentrations less than
10 20,000 ppb. Additionally, Sherwood et al. (1981) observed an increase in the propensity
11 of virulent group C Streptococci, but not Staphylococcus aureus following exposure to
12 1,000 ppb NO2 for 1 hour. In this study, Streptococci infection did not increase the total
13 mortality compared to controls, but NO2-exposed mice died significantly earlier.
14 Several other studies reported that NO2 exposure increases mortality from bacterial
15 infection. Illing etal. (1980) exposed mice to 2,000 ppb NO2 with continuous exercise
16 for 3 hours while Ehrlich et al. (1977) exposed mice to 3,000 ppb NO2 for 3 hours; both
17 studies subsequently exposed mice to an aerosol of Streptococcus pyogenes and
18 measured increased mortality rates compared to control animals exposed to clean air.
19 Increases in mortality from Streptococcus pyogenes infection following NO2 exposure
20 were also reported by Ehrlich et al. (1979). In this study, the relationship between
21 concentration and time was examined, and these factors yielded very different results as
22 the concentration was more important than time in determining mortality. Results were
23 consistent with other studies, though mortality increased post-challenge following a
24 7-day exposure to 3,500 ppb NO2.
25 Ehrlich (1980) conducted similar studies to investigate the effects of NO2 on Klebsiella
26 pneumoniae -induced mortality. Challenge following exposure to 1,500 ppb NO2 for
27 more than 8 hours resulted in increased mortality; however, exposure to 500 ppb for more
28 than 3 months was required to result in NO2-related increases in infection mortality. This
29 study also demonstrated species differences as increases in K. pneumoniae infection
30 mortality in mice were observed after a 2-hour exposure to 3,500 ppb NO2 while
31 hamsters and squirrel monkeys did not experience increases in mortality at NO2
32 concentrations less than 35,000 ppb and 50,000 ppb, respectively (Ehrlich. 1980).
33 Conversely, Purvis and Ehrlich (1963) did not observe increases in K. pneumonia
34 infection mortality in mice following a 2-hour exposure to NO2 at concentrations less
35 than 5,000 ppb.
36 One study examined effects of NO2 peak exposures superimposed on a lower continuous
37 background level of NO2 on susceptibility Streptococcus zooepidemicus infection
November 2013 4-110 DRAFT: Do Not Cite or Quote
-------
1 (Graham et al.. 1987). Mice were exposed to 4,500 ppb NO2 for 1, 3.5 and 7 hours or
2 exposed to these spikes with a continuous background exposure to 1,500 ppb NO2,
3 followed either immediately or 18 hours later with a Streptococcus zooepidemicus
4 challenge. Compared to control animals, the 4,500 ppb spikes alone or the spikes
5 superimposed on a background exposure did not result in differences in mortality from
6 infection; however, combined mortality rates (following the 1-hour exposure to 4,500
7 ppb and the 1-hour exposure to 4,500 ppb with 1,500 ppb background) were significantly
8 increased from immediate challenge after 4,500 ppb NO2 and were proportional to
9 duration of the 4,500 ppb exposure. In animals challenged 18 hours after NO2 exposure,
10 increases in mortality were only significant with 3.5- and 7-hour exposures to 4,500 ppb
11 NO2.
Controlled Human Exposure Studies
12 Compared with animal toxicological studies, controlled human exposure studies provided
13 less consistent evidence for NO2-induced infectivity. Although Pathmanathan et al.
14 (2003) found increased expression of ICAM-1, an extracellular receptor for viruses, in
15 airway biopsies following exposure to 2,000 ppb NO2 for 4 hours per day for 4 days,
16 Frampton et al. (2002) did not find evidence of increased susceptibility to ex vivo viral
17 challenge in bronchial epithelial cells collected from subjects exposed to 600 ppb or
18 1,500 ppb NO2 for 3 hours; however, there was an increase in virus-induced cytotoxicity
19 as measured by LDH release. Consistent with Frampton et al. (2002): Goings et al.
20 (1989) reported no increase infectivity of administered live, attenuated influenza virus in
21 subjects exposed to 1,000, 2,000 or 3,000 ppb NO2 for 2 hours/day for 3 consecutive
22 days. This study, however, lacked a sham control. Although not significant, another study
23 (Frampton et al.. 1989) reported a trend of decreased inactivation of influenza virus in
24 AMs collected from subjects after a 3-hour exposure to 600 ppb NO2.
Epidemiologic Studies
25 Epidemiologic studies examined respiratory infections as hospital admissions (Section
26 4.2.7.3). ED visits (Section 4.2.7.4). and parental reports of incidence and duration of
27 respiratory infections. Several studies found associations with ambient NO2
28 concentrations in children (Mehtaet al.. 2013; Stern et al.. 2013; HEI Collaborative
29 Working Group. 2012; Zemeket al.. 2010; Segala et al.. 2008; Just et al.. 2002). Studies
30 varied in the specific respiratory infection examined (e.g., bronchiolitis, ear infection, any
31 respiratory infection), and some associations were estimated imprecisely, with wide 95%
32 CIs (Figure 4-4 and Figure 4-5). Studies indicated associations in groups with respiratory
33 disease, i.e., children with asthma (Just et al.. 2002) and adults with COPD (Faustini et
November 2013 4-111 DRAFT: Do Not Cite or Quote
-------
1 al.. 2013). A multeity study in Canada did not find an association between ambient NO2
2 concentrations and ED visits for respiratory infections among subjects of all ages (Stieb
3 et al.. 2009).
4 Ambient NO2-associated increases in respiratory infection were found across a wide
5 range of ages in children, from infants ages 0-1 years (Stern etal. 2013) to
6 schoolchildren ages 7-15 years (Just et al.. 2002). Among infants ages 0-1 years, higher
7 1-week average NO2 was associated with longer duration of respiratory infections but not
8 incidence of infections (Stern etal., 2013). Several studies found respiratory infections in
9 children ages 0-5 years in associations with multiday (5- to 7-day) averages of NO2
10 concentration (Mehta et al.. 2013; Stern etal.. 2013; HEI Collaborative Working Group.
11 2012; Zemek et al., 2010). In addition to NO2, respiratory infections were associated with
12 ambient concentrations of copollutants such as BS, CO, and PMi0. Among adults with
13 COPD in Italy, the association between NO2 and LTRI was robust to adjustment for
14 PMio (Faustini et al.. 2013). Other studies did not examine copollutant models, and
15 studies conducted in Paris, France reported high correlations forNO2 with BS, PMi0, and
16 SO2 (r = 0.74-0.92) (Segala et al.. 2008: Just et al.. 2002). adding some uncertainty
17 regarding an independent association with NO2.
4.2.5.2 Mucociliary and Alveolar Clearance
18 Airborne substances that are small enough to be respired may be trapped in the epithelial
19 lining fluid (ELF) in the conducting airways and physically removed or cleared from the
20 airway by ciliated epithelial cells. Recent and previous animal toxicological studies have
21 demonstrated that exposure to high concentrations of NO2, generally above 5,000 ppb,
22 functionally impair pulmonary clearance and damage the ciliated epithelium of the
23 airway; however, exposure to NO2 at concentrations below 5,000 ppb have varying
24 effects on pulmonary clearance in animal toxicological and controlled human exposure
25 studies. The examination of the effect of NO2 on pulmonary clearance has been limited to
26 previous studies, which were reviewed in the 2008 ISA for Oxides of Nitrogen (U.S.
27 EPA. 2008c). No recent studies have been conducted that investigate the effects of NO2
28 on mucociliary clearance.
29 Studies have been conducted in various animal models and provide evidence that NO2
30 exposure can potentially affect mucociliary clearance. Schlesinger (1987b) employed two
31 methods to measure ciliary clearance in rabbits exposed to 310 or 1,030 ppb NO2 for 2
32 hours per day for up to 14 days. Mean residence time of radioactive tracer microspheres
33 was not altered 24 hours following 2, 7, or 14-day exposures; however, patterns in
34 clearance, measured as the fraction of retained radioactive tracer microspheres, were
November 2013 4-112 DRAFT: Do Not Cite or Quote
-------
1 reported be significantly different than controls at both 310 or 1,030 ppb NO2 over 14
2 days of exposure. Similarly, Vollmuth et al. (1986) studied mucociliary clearance in
3 rabbits exposed to 300 or 1,000 ppb NO2 for 2 hours while Ferin and Leach (1975)
4 exposed rats to 1,000 ppb NO2 in conjunction with 900 ppb NO for 7 hours per day, 5
5 days per week for 11 or 22 exposures; both studies reported accelerated clearance of
6 particles. A study published by Ohashi et al. (1994) found different results as guinea pigs
7 exposed to 3,000 or 9,000 ppb NO2 for 6 hours per day, 6 days per week for 2 weeks had
8 concentration-dependent reductions in ciliary activity. This study, however, used excised
9 nasal tissues from exposed animals and reported ciliary beat measured by light refraction
10 which is less representative of ambient human exposure.
11 In a controlled human exposure study of healthy adults, Helleday et al. (1995) used
12 fiberoptic bronchoscopy to measure ciliary activity and found decreased ciliary activity
13 after a brief exposure to 1,500 or 3,500 ppb NO2. In contrast, increases in ciliary activity
14 were reported 24 hours after a 4-hour exposure to 3,500 ppb NO2. It is important to note
15 that baseline measurements for each subject in this study were used as control values, and
16 therefore, the study lacked proper controls and subject blinding.
4.2.5.3 Alveolar Macrophages
17 Resident AMs have an integral role in detoxifying and/or clearing the lung of infectious
18 and noninfectious particles. The ability of AMs to perform this duty is dependent upon
19 several factors including the number and type of AMs, viability, mobility, phagocytic
20 activity, efficient enzyme activity, and secretion of inflammatory mediators. In animal
21 toxicological studies and controlled human exposure studies, examination of the effect of
22 NO2 on AM function has been limited to those reviewed in the 2008 ISA for Oxides of
23 Nitrogen (U.S. EPA. 2008c) as there are no recent studies.
Toxicological Studies
24 Previous studies reported NO2 exposure to induce slight morphological differences and
25 increases in AM numbers in BALF (Hooftman et al.. 1988; Mochitate et al.. 1986;
26 Rombout et al.. 1986: Goldstein et al.. 1977: Powell etal. 1971) and diminished
27 superoxide radical production (indicating reduced respiratory burst) at exposures as low
28 as 500 ppb (see Table 4-15 for study details). Robisonetal. (1990) and Robison et al.
29 (1993) exposed rat AMs to 100-20,000 ppb for 1 hour in vitro and found a dose-
30 dependent decrease in superoxide production, ranging from 81% -55% of control levels
31 after PMA stimulation. Similarly, Sprague Dawley male rats exposed to 500 ppb NO2 for
32 8 hours/day for 0.5, 1, 5, or 10 days had superoxide levels 63-75% of those in air exposed
November 2013 4-113 DRAFT: Do Not Cite or Quote
-------
1 animals after PMA stimulation (Robison et al., 1993). Suzuki etal. (1986) reported
2 comparable observations in AMs isolated from Fisher 344 rats exposed to 4,000 ppb NO2
3 for 3-10 days. Conversely, PMA-stimulated AMs isolated from Sprague Dawley female
4 rats exposed to NO2 below 6,100 ppb showed no change in superoxide production
5 compared to controls (Amoruso et al. 1981). Overall, NO2 exposure appears to decrease
6 the ability of AMs to produce superoxide anion, though inconsistencies are present across
7 studies that could be the results of strain or sex differences in response to NO2.
8 Studies also found variable effects of ambient relevant NO2 exposures on phagocytic
9 capacity of AMs. Rose etal. (1989b) exposed CD-I mice to 1,000 and 5,000 ppb NO2 for
10 6 hours/day for 2 days and reported diminished phagocytosis of colloidal gold particles at
11 both concentrations of NO2. In contrast, NO2 exposure increased uptake of murine
12 cytomegalovirus. Studies report both no change and decreased phagocytosis of latex
13 microspheres. Hooftmanetal. (1988) exposed rats to 4,000, 10,000, or 25,000 ppb NO2
14 for 6 hour/day, 5 days/week and found no changes in phagocytosis of latex microspheres
15 below 10,000 ppb at 1, 2, or 3 weeks. Schlesinger (1987b). however, found decreased
16 phagocytosis of latex microsphere by AMs isolated from rabbits 24 hours after a 2 or
17 6-day exposure at 300 or 1,000 ppb (2 hours/day; all animals were co-exposed to
18 0.5 mg/m3 H2SO4). Suzuki etal. (1986) also reported decreased phagocytic capacity of
19 AMs isolated from rats exposed to 4,000 ppb NO2 for 7 days.
Controlled Human Exposure Studies
20 Similar to animal toxicological studies, controlled human exposure studies did not
21 consistently demonstrate that NO2 concentrations relevant to ambient concentrations can
22 alter AM characteristics (see Table 4-16 for study details). Devlin etal. (1999) exposed
23 healthy subjects to 2,000 ppb NO2 for 4 hours with intermittent exercise and found that
24 AMs isolated from the BALF had decreased phagocytic activity and superoxide
25 production in ex vivo experiments. Conversely, no change in ex vivo macrophage
26 morphology or function was reported after subjects were exposed to 2,000 ppb NO2 for
27 6 hours with intermittent exercise (Azadniv et al.. 1998). In vitro exposure of human
28 AMs for 3 hours at 5,000 ppb NO2 did not result in statistically significant changes in
29 cell viability or neutrophil chemotactic factor (IL-8) or IL-1 release, markers of
30 macrophage activity (Pinkston et al., 1988).
4.2.5.4 Summary of Studies of Host Defense
31 Animal toxicological studies provide clear evidence for short-term NO2 exposure
32 impairing host defense by demonstrating increased mortality from bacterial or viral
November 2013 4-114 DRAFT: Do Not Cite or Quote
-------
1 infection following exposures of experimental animals to 1,500 to 4,500 ppb NO2 for 1 to
2 8 hours. Several studies also demonstrated decreased bactericidal activity following
3 exposures of 1,000 to 5,000 ppb for 1 to 17 hours. Compared with animal toxicological
4 studies, controlled human exposure studies provided less consistent evidence for
5 NO2-induced infectivity assessed as viral titers or inactivation of influenza virus. In
6 humans, NO2 exposures spanned 600-3,000 ppb for 3 hours for a single or 3-day
7 exposure. The evidence from animal toxicological studies provides biological plausibility
8 for the associations observed in epidemiologic studies between increases in ambient NO2
9 concentrations (5- to 7-day averages) and increases in respiratory infections as
10 ascertained by hospital admissions, ED visits, and parental reports. Studies varied in the
11 specific respiratory infection examined (e.g., bronchiolitis, ear infection, any respiratory
12 infection), and some associations were estimated imprecisely, with wide 95% CIs (Figure
13 4-4 and Figure 4-5). Epidemiologic evidence indicated associations in children and in
14 study populations with respiratory disease (i.e., children with asthma, adults with COPD).
15 Whereas an association between NO2 and LTRI in adults with COPD was robust to
16 adjustment for PMi0 (Faustini et al. 2013). most epidemiologic studies did not examine
17 copollutant models, and associations were found with highly correlated copollutants such
18 as BS, PMio, and SO2 (r = 0.74 to 0.92).
19 Also providing biological plausibility for NO 2-induced impaired host defense, studies
20 have characterized potential mechanisms underlying susceptibility to infection. Although
21 results vary across studies, some animal toxicological and controlled human exposure
22 studies found NO2 exposure to decrease the ability of AMs to produce superoxide anion
23 and decrease phagocytic activity. Such observations were made with NO2 exposures of
24 300 to 5,000 ppb. There was heterogeneity across studies in animal species, strain, and
25 sex that could have contributed to inconsistencies observed in response to NO2. Results
26 for the effects of NO2 exposure on pulmonary clearance were more variable with a
27 majority of studies reporting increased pulmonary clearance after relevant NO2 exposure.
November 2013 4-115 DRAFT: Do Not Cite or Quote
-------
Table 4-1 5
Study
Amoruso et al.
(1981)
Davis et al.
(1991)
Dowell et al.
(1971)
Ehrlich et al.
(1977)
Ehrlich (1980)
Gardner et al.
(1979)
Goldstein et al.
(1973)
Goldstein et al.
(1974)
Animal toxicological studies of NO2 and lung host
Species (Strain);
Lifestage;
Sex; n
Rat (Sprague-
Dawley); Female, n
= 4/group
Mice (C57BL/6N);
8-10 weeks;
n = 6/group
Dog (beagle);
n = 11
Mice(CF-l);
5-8 weeks; Female;
n = 5-88/group
(1,2) Mice; 6-8
weeks;
n > 88/group
(3) Mice, hamsters,
and squirrel
monkeys
Mice (Swiss albino);
Female;
n = 20/group
Mice (Swiss albino);
Male; n = 30/group
Mice (Swiss albino);
Male; n = 30/group
Exposure Details (Concentration; Duration)
1,300, 1,900, and3,OOOppbNO2for3h
5,000 ppb NO2 for 4 h;
Mycoplasma pulmonis challenge immediately
after exposure
3,000 ppb NO2 for 1 h
0, 1,500, 2,000, 3,500, and 5,000 ppb NO2 for
3h;
Streptococcus pyogenes challenge
immediately after exposure
(1 ) 500 ppb NO2 continuously for 1 week - 1 yr
(2) 1 ,500 ppb NO2 continuously for 2 h - 3 mo
(3) 1,500-50,000 ppb NO2 for 2 h
(1-3) Klebsiella pneumoniae challenge
immediately after exposure
(1 ) 500 ppb NO2 continuously for 7 days - 1 yr
(2) 1 ,500 ppb NO2 continuously for 2 h - 21
days
(3) 1 ,500 ppb NO2 7 h/day for 7 h - 1 1 days
(3) 3,500 ppb NO2 continuously for 30 min - 16
days
(4) 3,500 ppb NO2 7 h/day for 7 h - 13 days
(1-4) Streptococcus pyogenes challenge
immediately after exposure
(1) Staphylococcus aureus challenge
immediately before exposure;
0, 1,900, and 3,800 ppb NO2 for 4 h
(2) 0, 1,000, and 2,300 ppb NO2 for 17 h;
Staphylococcus aureus challenge immediately
after exposure
(1) 1,740 ppb NO2 + 110 ppb O3
(2) 1,490 ppb NO2 + 200 ppb O3
(3) 2,300 ppb NO2 + 200 ppb O3
defense.
Endpoints Examined
Analysis of BALF and
superoxide production by
AMs (PMA stimulation)
Bacterial clearance,
bactericidal activity
Histopathological
evaluation and lung
surfactant properties
Mortality
(1-3) Mortality
Mortality
(1) Bacterial counts and
bactericidal activity 5 h
after challenge (i.e., 1 h
after exposure)
(2) Bacterial counts and
bactericidal activity 0 h
and 4 h after challenge
Bacterial counts,
bactericidal activity, and
bacterial clearance 0 h
and 4 h after challenge
(4) 1,780 ppb NO2 + 270 ppb O3
(5) 4,180 ppb NO2 + 210 ppb O3
(1-5) 17 h; Staphylococcus aureus challenge
immediately after exposure
November 2013
4-116
DRAFT: Do Not Cite or Quote
-------
Table 4-15 (Continued): Animal toxicological studies of NO2 and lung host defense.
Study
Goldstein et al.
(1977)
Graham et al.
(1987)
Hooftman et al.
(1988)
Illina et al.
(1980)
Mochitate et al.
(1986)
Parker et al.
(1989)
Purvis and
Ehrlich(1963)
Robison et al.
(1990)
Robison and
Forman (1993)
Rombout et al.
(1986)
Rose et al.
(1988)
Rose et al.
(1989b)
Schlesinqer
(1987b)
Species (Strain);
Lifestage;
Sex; n
Rat (Sprague-
Dawley); Female
Mice (CD-1); 4-6
weeks; n =
5-12/group
Rats (Wistar); Male;
n = 10/group
Mice (CD-1);
5-6 weeks; Female;
n = 16/group
Rats (Wistar);
Male; 19-23 weeks;
n = 6/group
Mice (C57BL/6N
and C3H/HeN);
6-10 weeks
Mice (Swiss
Webster and
albino);
n >25/group
Rats (Sprague-
Dawley)
Rats (Sprague-
Dawley)
Rats (Wistar); Male,
6 weeks; n =
3-6/group
Mice (CD-1); 4-6
weeks; n >4/group
Rabbits (New
Zealand white);
Male, n = 5/group
Exposure Details (Concentration; Duration)
500, 1,000, and 2,400 ppb NO2 for 1 and 2 h
(1) 4,500 ppb NO2 for 1, 3.5, and 7 h
(2) 1,500 ppb NO2 continuously with a daily
spike of 4,500 ppb for 1, 3.5, and 7 h;
(1-2) Streptococcus zooepidemicus challenge
immediately and 18 h after exposure
3,000 ppb NO2 for 6 h/day, 5 days/week up to
21 days
1,000 ppb, 3,000 ppb NO2, and air for 3 h;
With or without continuous exercise;
Streptococcus pyogenes challenge
immediately after exposure
4,000 ppb NO2 continuously up to 10 days
0 and 5,000 ppb NO2 for 4 h;
Mycoplasma pulmonis challenge immediately
after exposure
1,500, 2,500, 3,500, and 5,000 ppb NO2 for2h;
Klebsiella pneumoniae challenge 0-27-h post-
exposure
100, 500, and 1,000 ppb NO2 for 1 h; AMs
exposed ex vivo
100, 2,000, and 5,000 ppb NO2 for 1 -4 h;
AMs exposed ex vivo
500, 1,390, and 2,800 ppb NO2 for
1,2,4, 8, 16, and 28 days
(1) 1,000, 2,500, and 5,000 ppb NO2 for 6
h/day for 2 days; intratracheal inoculation with
murine Cytomegalovirus', 4 additional days (6
h/day) of exposure
(2) re-inoculation 30 days (air) post-primary
inoculation
300 or 1 ,000 ppb NO2 for 2 h/day for 2, 6, and
13 days
Endpoints Examined
Agglutination of AMs
Mortality
Histopathological
evaluation, analysis of
BALF, and AM function
and morphology
Mortality after
Streptococcus pyogenes
challenge
BALF cell counts and MA
function and morphology
Histopathological
evaluation, bacterial
infection and clearance 4
h up to 7 days post-
challenge, BALF cell
counts
Mortality
Viability, LTB4
production, neutrophil
chemotaxis, superoxide
production
Arachidonate metabolite
production induced by
treatment with a calcium
ionophore)
Histopathological
evaluation
Infection 5 and 10 days
post-inoculation,
histopathological
evaluation, and analysis
ofBALF(LDH, albumin,
macrophages)
Viability and AM activity
(mobility, attachment,
and phagocytosis)
November 2013
4-117
DRAFT: Do Not Cite or Quote
-------
Table 4-15 (Continued): Animal toxicological studies of NO2 and lung host defense.
Study
Species (Strain);
Lifestage;
Sex; n
Exposure Details (Concentration; Duration) Endpoints Examined
Sherwood et al.
(1981)
Mice (Swiss albino);
Male; n =
8-24/group
1,000 ppb NO2 for 24 and 48 h;
Streptococcus (group C) challenge immediately
after exposure
Bacterial counts 0 - 48-h
post-challenge, bacterial
clearance,
histopathological
evaluation, mortality
Suzuki et al.
(1986)
Rats (Fischer 344);
Male, 7 weeks;
n = 8/group
4,000 NO2 ppb for 1, 3, 5, 7, and 10 days
AM activity (phagocytosis
and superoxide
production), SOD and
G-6-PD activity
Table 4-16 Controlled human exposure studies of lung host defense.
Study
n, Sex; Age
(mean ± SD)
Exposure Details
(Concentration; Duration)
Endpoints Examined
Azadniv et al.
(1998)
n = 11 M, 4F;
Early Phase: 28.1 ±
3.5 yr
Late Phase: 27.4 ±
4.2 yr
2,000 ppb for 6 h;
Exercise for approximately 10 of every 30
min at VE= 40 L/min
Alveolar macrophage
function 1 h (early phase)
and 18 h (late phase) after
exposure
Devlin et al.
(1999)
n = 11 M; Range:
18-35yr
2,000 ppb for 4h;
Exercise for 15 min on/15 min off at VE= 50
L/min
BALF macrophage
superoxide production and
phagocytosis
Frampton et al.
(1989)
(1)n = 7M, 2F; 30
yr (Range: 24-37)
(2)n = 11 M, 4F;25
yr (Range: 19-37)
(1) 600 ppb for 3h,
(2) 1,500 ppb for 3h;
(1,2) Exercise 10 min on/20 min off atVE=
~4 times resting
BALF cell viability and
differential counts 3.5h
post-exposure, inactivation
of influenza virus by BALF
cells, IL-1 activity in BALF
cells
(Frampton et al.
2002)
(1,2)n = 12M, 9F;
F=27.1 ±4.1 yr
M= 26.9 ± 4.5 yr
(1) 600 ppb for 3h,
(2) 1,500 ppb for 3h;
(1,2) Exercise 10 min on/20 min off atVE=
40 L/min
Bronchial and alveolar
lavage fluid cell viability
and differential counts
3.5h post-exposure,
influenza and RSV
challenge in BALF cells,
peripheral blood
characterization
Goings et al.
(1989)
(1)n = 44
(2) n = 43
(3) n = 65; Range:
18-35yr
(1)2,000 ppb for 2h
(2) 3,000 ppb for 2h
(3) 1,000 or 2,000 ppb for 2h
Nasal wash virus isolation
and count 4d after virus
administration. Serum and
nasal wash antibody
response 4w after virus
administration.
November 2013
4-118
DRAFT: Do Not Cite or Quote
-------
Table 4-16 (Continued): Controlled human exposure studies of lung host defense.
Study
Helledav et al.
(1995)
(Pinkston et al..
1988)
n, Sex; Age
(mean ± SD)
n = 8 nonsmokers;
Median: 26 yr
(Range: 24-35), 8
smokers, Median: 29
yr (Range: 28-32)
Human alveolar
macrophages
isolated from 14 M
and 1 F; 29 ± 3.9 yr
Exposure Details
(Concentration; Duration)
3,500 ppbfor20 min;
Exercise last 15 min at 75 W
5,000 ppbforS h (ex vivo)
Endpoints Examined
Bronchial wash and BALF
analysis (protein
concentration, differential
cell counts, AM function)
Cell viability and release of
neutrophil chemotactic
factor and IL-1
4.2.6 Respiratory Symptoms and Asthma Medication Use
1 The evidence described in preceding sections for NO2-induced AHR (Section 4.2.2) and
2 increases in pulmonary inflammation (Section 4.2.4) characterizes key events to inform
3 the mode of action by which NO2 exposure may increase respiratory symptoms.
4 Epidemiologic studies reviewed in the 2008 ISA for Oxides of Nitrogen consistently
5 found increased respiratory symptoms in children with asthma and children in the general
6 population in association with higher indoor NO2, personal NO2, and ambient NO2
7 concentrations (U.S. EPA. 2008c). There was weak support from a controlled human
8 exposure study of adolescents. NO2-associated increases in respiratory symptoms in
9 adults with asthma were found in previous epidemiologic studies but inconsistently found
10 in controlled human exposure studies. Controlled human exposure studies in healthy
11 adults generally did not observe respiratory symptoms with NO2 exposure. Recent
12 studies, most of which were epidemiologic, continued to demonstrate associations
13 between short-term increases in ambient NO2 concentration and increases in respiratory
14 symptoms in children with asthma and children in the general population.
15
16
17
18
19
20
4.2.6.1 Epidemiologic Studies
The most robust evidence for associations between ambient oxides of nitrogen and
respiratory symptoms is demonstrated for NO2 for children with asthma and children in
the general population. Across the various populations examined, symptom data were
collected by having subjects or their parents complete daily diaries for periods of two
weeks to several months. Heterogeneity in the number of consecutive days of follow-up
and the frequency of diary collection from study subjects did not appear to influence
November 2013
4-119
DRAFT: Do Not Cite or Quote
-------
results. Ambient NO2 and NOX concentrations, locations, and time periods for
epidemiologic studies of respiratory symptoms are presented in Table 4-17.
Table 4-17 Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of respiratory symptoms.
Study3
Mortimer et al.
(2002)
O'Connor et al.
(2008)
Schildcrout et al.
(2006)
Sarnatetal. (2012)
Zoraetal. (2013)
Mannetal. (2010)
Gent et al. (2003)
Holquinetal. (2007)
Location
Bronx and East
Harlem, NY;
Chicago, IL;
Cleveland, OH;
Detroit, Ml;
St. Louis, MO;
Washington, DC
Boston, MA
Bronx, NY
Chicago, IL
Dallas, TX
New York, NY
Seattle, WA
Tucson, AZ
Albuquerque, NM;
Baltimore, MD;
Boston, MA;
Denver, CO;
San Diego, CA;
St. Louis, MO;
Toronto, ON,
Canada
El Paso, TX and
Ciudad Suarez,
Mexico
El Paso, TX
Fresno/Clovis, CA
New Haven county,
CT
Ciudad Juarez,
Mexico
Study Period
June-Aug 1993
Aug 1998-
July2001
Nov1993-
Sept 1995
Jan-Mar 2008
Mar-June 2010
Winter-Summer,
2000-2005
Aug 200-
Feb 2004
2001-2002
Exposure
Metric
Analyzed
4-h avg NO2
(6 a.m.-
10 a.m.)
24-h avg NO2
24-h avg NO2
96-h avg NO2
96-h avg NO2
24-h avg NO2
NO2 - avg
time NR
1-week avg
NO2
Mean/Median
Concentration
(PPb)
NR
NR
Across cities:
17.8-26.0
El Paso
schools: 4.5,
14.2, Central
sites: 14.0,
18.5,20.5
Ciudad Juarez
schools: 18.7,
27.2, Central
site: None
School 1: 9.3
School 2: 3.4
Median: 18.6
NR
18.2
Upper Percentile
Concentrations (ppb)
NR
NR
90th: Across cities
26.7-36.9 ppb
NR
Max: 16.2
Max: 7.5
75th: 24.7
Max: 52.4
NR
NR
November 2013
4-120
DRAFT: Do Not Cite or Quote
-------
Table 4-17 (Continued): Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of respiratory symptoms.
Study3
Spira-Cohen et al.
(2011)
Pateletal. (2010)
Barraza-Villarreal et
al. (2008),
Escamilla-Nufiez et
al. (2008)
Romieu et al.
(2006)
Gillespie-Bennett et
al. (2011)
Just et al. (2002)
Schwartz et al.
(1994)
Moon et al. (2009)
Andersen et al.
(2008a)
Ward et al. (2002)
Rodriguez et al.
(2007)
Stern et al. (2013)
Peeletal. (2011)
Wiwatanadate and
Liwsrisakun (2011)
Maestrelli et al.
(2011)
Hiltermann et al.
(1998)
Location
Bronx, NY
New York City and
nearby suburb, NY
Mexico City, Mexico
Mexico City, Mexico
Bluff, Dunedin,
Christchurch,
Porirua, Hutt Valley,
New Zealand
Paris, France
Watertown, MA;
Kingston-Harriman,
TN; St. Louis, MO;
Steubenville, OH;
Portage, Wl;
Topeka, KS
Seoul, Incheon,
Busan, Jeju, Korea
Copenhagen,
Denmark
Birmingham,
Sandwell, U.K.
Perth, Australia
Bern, Basel,
Switzerland
Atlanta, GA
Chiang Mai,
Thailand
Padua, Italy
Bilthoven, the
Netherlands
Study Period
Spring 2002,
Spring/Fall
2004, Spring
2005
2003-2005,
months NR
June 2003-
June2005
Oct1998-
Apr2000
Sept 2006
Apr-June 1996
Apr-Aug,
1984-1988
Apr-May 2003
Dec 1998-
Dec 2004
Jan-Mar 1997
May-July 1997
June 1996-
July 1998
Apr 1999-
Feb2011
Aug 1998-
Dec 2002
Aug 2005-
June2006
1999-2003
July-Oct1995
Exposure
Metric
Analyzed
6-h avg NO2
(9 a.m. -3 p.m.)
24-h avg NO2
8-h max NO2
1-h max NO2
4-week avg
NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NOX
24-h avg NO2
1-h max NO2
24-h avg NO2
24-h avg NO2
1-h max NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
Mean/Median
Concentration
(PPb)
NR
NR
37.4
66
3.9
28.6b
13.3
NR
11.8
15.2
18
13.3
18
7
Rural: 8.1b
Urban: 25.6b
41.7
17.2
Across
seasons and
years:
20.9-37.0b
11. 2b
Upper Percentile
Concentrations (ppb)
NR
NR
Max: 77.6
Max: 298
NR
Max: 59.0b
75th: 24.1
Max: 44.2
NR
75th: 14.6
75th: 18.4
Max: 35
Max: 29
Max: 48
Max: 24
NR
NR
90th: 65.6
Max: 109.2
90th: 26.5
Max: 37.4
75th: 23.0-42.5b
22.5b
November 2013
4-121
DRAFT: Do Not Cite or Quote
-------
Table 4-17 (Continued): Mean and upper percentile concentrations of oxides of nitrogen in
epidemiologic studies of respiratory symptoms.
Study3
Boezen et al. (1998)
Forsberq et al.
(1998)
Feo Brito et al.
(2007)
Annesi-Maesano et
al. (2012b)
Karakatsani et al.
(2012)
von Klot et al.
(2002)
Laurent et al. (2009)
Carlsen et al.
(2012)
Silkoffetal. (2005)
Harreetal. (1997)
Peacock et al.
(2011)
Hiqqins et al. (1995)
Desqueyroux et al.
(2002)
Location
Amsterdam
Meppel,
the Netherlands
Landskrona,
Sweden
Ciudad Real
Puertollano,
Spain
Multiple metropolitan
locations, France
Amsterdam,
the Netherlands
Athens, Greece
Birmingham, U.K.
Helsinki, Finland
Erfurt, Germany
Strasbourg, France
Reykjavik, Iceland
Denver, CO
Christchurch,
New Zealand
London, U.K.
Widnes, Runcorn,
U.K.
Paris, France
Study Period
Winter
1993-1994
Jan-Mar,
yrNR
May-June
2000-2001
May-Aug 2004
Oct 2002-
Mar2004
Sept 1996-
Nov1997
2004, all yr
Mar 2006-
Dec 2009
Winters
1999-2000
2000-2001
June-Aug 1994
Oct 1995-
Oct1997
Aug, year NR
Oct 1995-
Mar1996
Apr-Sept 1996
Exposure
Metric
Analyzed
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
24-h avg NO2
Dispersion
model
24-h avg NO2
1-h max
24-h avg NO2
24-h avg NO2
1-h max NO2
24-h avg NO2
1-h max NO2
Mean/Median
Concentration
(PPb)
24.5b
14.2b
16.2b
17.4b
29.5b
9.9b
20.4b
21. 2b
18.3b
12.1b
24.5b
18.6b
11. 7b
27.4b
16
29
NR
51.4
NR
31. 4b
26. 1b
Upper Percentile
Concentrations (ppb)
Max: 40.4b
Max: 28.9b
38. 1b
Max: 35.6b
Max: 100.5b
Max: 38.9b
51. 8b
59.0b
44.2b
41. 4b
Max: 63.3b
NR
52.9b
Max: 64.4b
75th: 30, Max:
75th: 36, Max:
NR
75th: 56
Max: 44.7b
Max:68.1b
Max: 56.4b
54
54
aStudies presented in order of first appearance in the text of this section.
""Concentrations converted from |jg/m3 to ppb using the conversion factor of 0.532 assuming standard temperature (25 °C) and
pressure (1 atm).
NR = not reported.
November 2013
4-122
DRAFT: Do Not Cite or Quote
-------
Children with Asthma
1 Several recent studies add to the evidence for increases in respiratory symptoms in
2 children with asthma associated with increases in ambient NO2. Across previous and
3 recent studies, there was heterogeneity in the strength and precision of association;
4 however, most results indicated a pattern of elevated risk of respiratory symptoms across
5 the various symptoms and lags of NO2 exposure examined (Figure 4-3 and Table 4-18).
6 The robustness of association was demonstrated in a meta-analysis of 24 mostly
7 European studies and some U.S. studies, including several reviewed in the 2008 ISA for
8 Oxides of Nitrogen. There was some evidence of publication bias with exclusion of the
9 multicounty European PEACE studies, but with adjustment for publication bias, an
10 increase in 24-h avg NO2 was associated with increased risk of asthma symptoms
11 (Weinmayr et al.. 2010). Across all studies, the most consistent results were for total
12 respiratory or asthma symptoms, wheeze, and cough. Increases in ambient NO2
13 concentrations were not consistently associated with increases in rescue inhaler or beta-
14 agonist use in children with asthma (Patel et al.. 2010; Romieu et al., 2006; Schildcrout et
15 al.. 2006: Seealaetal.. 1998).
November 2013 4-123 DRAFT: Do Not Cite or Quote
-------
Study
Symptom Composite
Schildcroutetal. (2006)
Just etal. (2002)
Segalaetal. (1998)
Mortimer etal. (2002)
Spira-Cohen etal. (2011)
O'Connor etal. (2008)
Wheeze
Mann etal. (2010)
Gent etal. (2003)
Pateletal. (2010)
Jalaludin etal. (2004)
Barraza-Villarreal et al. (2008)
Escamilla-Nunezetal. (2008)
Asthma Medication Use
Schildcroutetal. (2006)
Romieuetal. (2006)
Jalaludin etal. (2004)
Segalaetal. (1998)
NO2Averaging Time
24-h avg
24-h avg
24-h avg
4-h avg
6-h avg
24-h avg
24-h avg
NR
24-h avg
1 5-h avg
1 -h max
1 -h max
24-h avg
1 -h max
1 5-h avg
24-h avg
NO2Lag Subgroup
0
^^^
0 Mildasthma
1-6 avg
0
1-1 9 avg
2
0
0
0 —
0
1
0
1-6 avg GSTM1 null -•
GSTM1 positive
0
0 Mild 3sth tri3 ~~
*•
•—
-•—
-^
•*-
•
• —
+•
»
•
-•-
•-
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Odds ratio per 20,25, or 30 ppb increase in NO2 (95% Cl)a
Note: Studies are presented in order of decreasing study strength (e.g., exposure assessment method, potential confounding
considered). Red=recent studies, Black=previous studies. Study details and quantitative results are reported in Table 4-18.
aEffect estimates are standardized to a 20-ppb increase for 24-h avg or 15-h avg NO2, 25 ppb for 4-h avg, 6-h avg or 8-h max NO2,
and 30 ppb for 1-h max NO2.
Figure 4-3 Associations of ambient NO2 concentrations with respiratory
symptoms and asthma medication use in children with asthma.
November 2013
4-124
DRAFT: Do Not Cite or Quote
-------
Table 4-18 Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Study Population and Methodological Details
Oxide of
Nitrogen
Metrics
Analyzed
Oxide of
Nitrogen
Lag Day
Subgroup
Analyzed
(if
applicable)
Odds Ratio
(95% Cl)
Single-Pollutant Copollutant
Model3
Examination
Children with Asthma
Mortimer et Bronx and East Harlem, NY; Chicago, IL; Cleveland,
al. (2002) OH; Detroit, Ml; St. Louis, MO; Washington, DC
(NCICAS cohort)
N = 846, ages 4-9 yr.
Repeated measures. Daily symptom data collected
for 2-week periods every 3 mo. Recruitment from ED
visits and clinics. Parent report of physician-
diagnosed asthma and symptoms in previous 12 mo,
or asthma symptoms for >6 weeks and symptoms
with exercise or cold exposure or family history of
asthma. Representative of full cohort except for
greater asthma medication use. Mixed effects model
adjusted for city, follow-up period, day of study, 24-h
rainfall, 12-h avg temperature
NO2-Central site Lag 1-6
4-h avg av9
(6 a.m.-10 a.m.) Largest
Average of all OR
city monitors.
Morning symptoms:
1.80(1.03, 3.15)
w/O3 (summer): 1.66
(0.90, 3.06)
Weak correlation with
NO2. r = 0.27.
O3 effect estimate also
slightly attenuated.
SO2 and PM-|0 also
associated with
symptoms. Correlations
NR.
O'Connor et
al. (2008)
Boston, MA; Bronx, NY; Chicago, IL; Dallas, TX; New
York, NY; Seattle, WA; Tucson, AZ (ICAS cohort)
N = 861, ages 5-12 yr, persistent asthma and atopy,
82% black or Hispanic.
Repeated measures. Symptom data collected for 2
week period every 2 mo for 2 yr. Recruitment from
intervention of physician feedback. Mixed effects
model adjusted for site, month, sitexmonth
interaction, temperature, intervention group.
NO2-Central site 1-19 avg
24-h avg
All monitors
close to home
and not near
industrial source.
Median distance
to site = 2.3 km.
Wheeze-cough
1.17(0.99, 1.37)
Slow Play
1.25(1.04, 1.51
Missed school in 2
week period
1.65(1.18,2.32)
Only 3-pollutant model
analyzed.
PM2.5, SO2, CO and O3
also associated
Moderate correlations
with NO2. r = 0.59 for
PM2.5and SO2, 0.54 for
CO. Negative
correlation for O3. r =
-0.31
November 2013
4-125
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Schildcrout
et al. (2006)
Zora et al.
(2013)
Sarnat et al.
(2012)
Study Population and Methodological Details
Albuquerque, NM; Baltimore, MD; Boston, MA;
Denver, CO; San Diego, CA; St. Louis, MO; Toronto,
ON, Canada (CAMP cohort)
N = 990, ages, 5-12 yr, mild to moderate asthma
Repeated measures. Daily symptom diaries for
21-201 days. GEE for individual cities combined for
study-wide estimates. City-specific models adjusted
for day of week, ethnicity, annual family income,
response to methacholine, maximum temperature,
humidity, temperaturexhumidity, calendar date.
Pollutant analyzed as daily deviation from subject
mean.
El Paso, TX
N = 36, mean age 9.3 (SD: 1.5) yr, 47% with atopy.
Repeated measures. Daily asthma control
questionnaire given by parents for 13 weeks, weekly.
Questionnaire ascertains symptoms, activity
limitations, asthma medication use. Parent report of
physician-diagnosed asthma. Linear mixed effects
model adjusted for random subject effect and
humidity, temperature, school.
El Paso, TX and Ciudad Juarez, Mexico
N = 29 per city, ages, 6-12 yr, asthma and current
symptoms.
Repeated measures. Daily symptom diaries.
Recruitment from schools representing a gradient of
traffic, subjects from nonsmoking homes. No
information on participation rate. Self report of
physician-diagnosed asthma. GLM with subject as
random effect and adjustment for school,
temperature, relative humidity. Excluded potential
confounding by medication use, cold symptoms.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-Central site 0
Average of
multiple sites
within 50 miles of
ZIP code
0-2 sum
NO2-school 0-4 avg
outdoor
24-h avg
NO2-school 0-4 avg
outdoor
24-h avg
Subgroup Odds Ratio
Analyzed (95% Cl)
(if Single-Pollutant
applicable) Model3
Asthma symptoms:
1.06(1.00, 1.12)
Rescue Inhaler use:
1.04(1.00, 1.08)
Asthma symptoms:
1.05(1.01, 1.09)
Asthma control
score
No allergy, -0.29 (-1.07, 0.49)
n = 19
Allergy, 0.56 (-0.17, 1.28)
n = 17
No quantitative
results reported.
Associations
reported to be
consistent with the
null.
Copollutant
Examination
Joint effect models
NO2 + CO:
1.07(1.0, 1.14)
NO2 + SO2:
1.06(0.98, 1.15)
NO2 + PMi0:
1.06(0.99, 1.13)
Moderate to high
pnirplatinn^ with
\j\J\ 1 CI0LIUI lo Will 1
NO2. r =
0.23 to 0.68 for SO2,
0.26 to 0.64 for PM10,
0.63 to 0.92 for CO.
No copollutant models
analyzed for subgroups.
BC, benzene, toluene,
also associated with
- poorer asthma controls.
Correlations with NO2
weak to high.
Spearman r =
0.29 to 0.56 for BC,
0.37 to 0.71 for
benzene,
0.16 to 0.71 for toluene
November 2013
4-126
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Holquin et
al. (2007)
Spira-Cohen
etal. (2011)
Study Population and Methodological Details
Ciudad Juarez, Mexico
N = 194, ages 6-12 yr, 78% mild, intermittent asthma,
58% with atopy.
Repeated measures. Daily symptom diaries given by
parents for 4 mo, checked biweekly. 87%
participation. Parent-report of physician-diagnosed
asthma. Linear and nonlinear mixed effects model
with random effect for subject and school adjusted for
sex, BMI, day of week, season, maternal and
paternal education, passive smoking exposure
Bronx, NY
N ~~ 40 ages 10-1 2 yr 100%nonwhite 44% with
asthma ED visit or hospital admission in previous 12
mo.
Repeated measures. Daily symptom diaries for 1 mo,
checked daily. 454 observations. Recruitment from
schools by referrals from school nurses. Parental
report of physician-diagnosed asthma. Mixed effects
model with subject as random effect adjusted for
height, sex, temperature. Adjustment for school
(indicator of season) did not alter results. 89% time
indoors.
Oxide of Subgroup
Nitrogen Oxide of Analyzed
Metrics Nitrogen (if
Analyzed Lag Day applicable)
NO2-School 0-6 avg Asthma,
outdoor n = 31
24-h av9 No asthma,
Homes n = 41
397 meters from
schools.
NO2-School 0,1,0-1
outdoor avg
6-h avg
(9 a.m.-3 p.m.)
walk to school.
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
No quantitative
results reported. Air
pollutant exposures
reported not to be
associated with
respiratory
outcomes.
Total symptoms:
1.05(0.92, 1.20)
Wheeze:
1.10(0.87, 1.39)
Copollutant
Examination
Road density at home
and school reported not
to be associated with
respiratory symptoms.
Personal EC associated
with symptoms with
NO2 adjustment.
No quantitative data
reported.
November 2013
4-127
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Patel et al.
(2010)
Barraza-
Villarreal et
al. (2008)
Study Population and Methodological Details
New York City and nearby suburb, NY
N = 249, ages 14-20 yr, 57 with asthma, 192 without
asthma.
Repeated measures. Daily symptom diaries for 4-6
weeks, collected weekly. Recruitment from schools.
Self-report of physician-diagnosed asthma. GLMM
with random effect for subject and school and
adjusted for weekend, daily maximum 8-h avg Os,
urban location. Adjustment for season, pollen counts
did not alter results.
Mexico City, Mexico
N ~~ 163-179 ages 6-14 yr 54% persistent asthma
89% with atopy.
Repeated measures. Symptom data collected every
15 days for mean 22 weeks. Children with asthma
recruited from pediatric clinic. Children without
asthma were friends/schoolmates. Asthma severity
assessed by pediatric allergist. Linear mixed effects
model with random effect for subject and adjusted for
sex, BMI, lag 1 minimum temperature, ICS use, time.
Adjustment for outdoor activities, smoking exposure,
anti-allergy medication use, and season did not alter
results.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-Central site 0
24-h avg
1 site 2. 2-9.0 km
from schools, 1
site 40 km from
schools
NO2-Central site 0
1-h max
Site within 5 km
of school or
Spearman °'1 sum
correlation
coeffiecient for
school vs. central
site: r= 0.21
Subgroup
Analyzed
(if
applicable)
All subjects
Asthma,
n = 57
No asthma,
n = 192
All subjects
Asthma,
n = 57
No asthma,
n = 192
Asthma,
n = 126
No asthma,
n = 45
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
Wheeze
1.04(0.92,
1.16(0.93,
0.88(0.75,
1.17)
1.45)
1.03)
Chest tightness
1.10(0.96,1.25)
1.25(1.00,
0.96(0.75,
Wheeze
1.09(1.03,
Cough
1.09(1.04,
Cough
1.28(1.04,
1.55)
1.23)
1.15)
1.14)
1.57)
Copollutant
Examination
No copollutant model
with BC.
BC also associated with
symptoms.
Across locations
moderately to highly
correlated with NO2.
Speaiman i - 0.56-0.90
for BC.
No copollutant model.
PM2.5and O3 also
associated with
symptoms.
Moderate correlations
with NO2. Pearson r =
0.61 for PM25, 0.21 for
03.
November 2013
4-128
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Escamilla-
Nufiez et al.
(2008)
Romieu et
al. (2006)
Study Population and Methodological Details
Mexico City, Mexico
N ~~ 197 ages 6-14 yr 147 with asthma 43% with
persistent asthma, 89% atopy; 50 without asthma,
79% with atopy. Part of same cohort as above.
Linear mixed effects model with random effect for
subject and adjusted for asthma severity, atopy, lag 1
minimum temperature, time, sex. Adjustment for
outdoor activities, smoking exposure, season did not
alter results.
Mexico City, Mexico
N = 151, mean age 9 yr, mild or moderate asthma.
Repeated measures. Daily symptom diaries 61-92
days per subject, collected weekly. Recruitment from
allergy clinic as part of a Vitamin C/E
supplementation trial. Diagnosis by clinical
examination. GEE adjusted for supplementation
group, minimum temperature, smoking exposure,
asthma severity, time.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-Central site 1
1 -h max
Site within 5 km
of school or
home.
NO2-Central site 1-6 avg
1-h max
Site within 5 km
of home
Subgroup
Analyzed
(if
applicable)
Asthma
n = 147'
No Asthma,
n = 50
Asthma,
n = 147
Genotype
GSTM1 null
GSTM1
positive
GSTP1 lie/lie
or Ile/Val
GSTP1
Val/Val
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
Cough
1.07 (1.02,
1.23(1.03,
Wheeze:
1.08(1.02,
Cough
1.09(1.00,
1.19(1.11,
1.19(1.11,
1.08(0.99,
1 12)
1 • t£-i
1.47)
1.14)
1.19)
1.27)
1.27)
1.18)
Copollutant
Examination
Only multipollutant
model with O3 and
PM2.5. NO2
associations persist.
No copollutant model.
Associations with O^
found with different
variants.
Moderate correlation
with NO2. Pearson r =
0.57 for O3 and PM-io
BD use
GSTM1 null 0.94(0.87,1.02)
GSTM1 1.09(1.02,1.17)
positive
GSTP1 lie/lie 1.08(1.02, 1.14)
or Ile/Val
GSTP1
Val/Val
0.94(0.85, 1.04)
November 2013
4-129
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Just et al.
(2002)
Gillespie-
Bennett et
al. (2011)
Gent et al.
(2003)
Study Population and Methodological Details
Paris, France
N = 82, ages 7-15 yr, asthma attack in previous 12
mo and daily asthma medication use, 90% atopy
Repeated measures. Daily symptom diaries for 3 mo,
collected weekly. Recruitment from hospital
outpatients. GEE adjusted for time trend, day of
week, pollen, temperature, humidity.
Bluff, Dunedin, Christchurch, Porirua, Hutt Valley,
New Zealand
N = 358, ages 6-13 yr
Cross-sectional. Daily symptom diaries for 1 12 days.
Recruitment from a home heating intervention. 77%
participation. Mixed effects model with log-
transformed NO2 and random effect for subject.
Adjustment for age, sex, region, ethnicity,
intervention, income, temperature did not alter
results.
New Haven county, CT
N = 149, ages 4-1 2 yr
Repeated measures. Daily symptom diaries reported
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-Central site 0
24-h avg
Average of 1 1
sites
NO2-outdoor 4-week
home avg
24-h avg
1 measure per
subject
NC>2-indoor
home
24-h avg
Multiple
measures per
subject
NO2-central site 0
Avg time NR
# sites NR
Subgroup Odds Ratio
Analyzed (95% Cl)
(if Single-Pollutant
applicable) Model3
Asthma
1.75(0.82, 3.70)
Night cough
2.11 (1.20, 3.71)
Respiratory infection
7*1 n /o co or\ j\\
.19 (2.53, 20.4)
Per log increase
NO2
Lower respiratory
symptoms: 1.09
(0.78, 1.51)
Reliever Inhaler:
1.47(0.96,2.26)
Lower respiratory
symptoms: 1.14
(1.12, 1.16)
Reliever Inhaler:
1.14(1.11, 1.17)
Not reported
Copollutant
Examination
No copollutant model.
BS associated with
cough and infection.
High correlation with
NO2. Pearson r = 0.92.
No copollutant model.
No other pollutants
examined.
w/source apportionment
factor of EC, Zn, Pb,
Cu, Se:
monthly. Recruitment from larger cohort, pediatric
asthma clinic, and school. Parent report of physician
diagnosed asthma. GEE adjusted for season, day of
week, date, and motor vehicle factor obtained by
source apportionment.
1.08(0.99, 1.18).
Factor results robust to
NO2 adjustment.
Moderate correlation
with NO2. Pearson r =
0.49.
November 2013
4-130
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Mann et al.
(2010)
Jalaludin et
al. (2004)
Study Population and Methodological Details
Fresno, Clovis, CA
N = 315, ages 6-11 yr, 47% mild persistent asthma,
25% moderate to severe asthma, 63% with atopy.
Repeated measures. Daily symptom diaries for 14
days every 3 mo. Recruitment from schools,
advertisements, physician's offices, local media.
ARIMA with imputed wheeze values for 7.6% days
and GEE adjusted for fitted daily mean wheeze,
home ownership, race, sex, asthma severity, panel
group, 6-month cohort, 1-h minimum temperature.
Adjustment for medication use did not alter results.
Sydney, Australia
N = 125, mean age 9.6 yr, 45 with wheeze, asthma,
and AHR, 60 with wheeze and asthma, 20 with
wheeze
Repeated measures. Daily symptom diary mailed in
monthly for 11 mo. Recruitment from schools. Parent-
report of physician-diagnosed asthma. GEE adjusted
for time trend, temperature, humidity, number of
hours spent outdoors, total pollen and alternaria,
season.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-central site 2
24-h avg
1 site within 20
km of homes
NO2-central site 0
1 5-h avg
(6 a.m.-9 p.m.)
Site within 2 km
of schools
Subgroup
Analyzed
(if
applicable)
All subjects
Fungi
allergic, n =
85
Cat allergic,
n = 49
Boys,
intermittent
asthma, n =
47
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
Wheeze
1.24(1.05, 1.48)
1.61 (1.24, 2.08)
1.73(1.14,2.62)
2.58 (1.61, 4.13)
Wheeze:
1.02(0.86, 1.22)
Wet cough:
1.13(1.00, 1.27)
Beta agonist use:
1.02(0.93, 1.13)
Copollutant
Examination
w/PM 10-2.5 all subjects
1.14(0.95, 1.37).
PM 10-2.5 association
robust to NC>2
adjustment.
— Weak correlation with
NO2. r = 0.12.
NO2 associations found
in children with asthma
and AHR but examined
only in multipollutant
model with Qs and
PMio.
Negative or weak
correlation with NC>2. r=
-0.31 forO3, 0.26 for
PMio.
November 2013
4-131
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Seaala et al
(1998)
Children in
Schwartz et
al. (1994)
Moon et al.
(2009)
Study Population and Methodological Details
. Greater Paris area, France
N = 43 mild asthma, 41 moderate asthma, 89%
atopy, 69% ICS users, ages 7-15 yr.
Repeated measures. Daily symptom diary for
25 weeks, collected weekly. Recruitment from
outpatients of children's hospital. GEE adjusted for
day of week, time trend, temperature, humidity, age,
sex.
the General Population
Watertown, MA; Kingston-Harriman, TN; St. Louis,
MO; Steubenville, OH; Portage, Wl; Topeka, KS
N = 1,844, grades 2-5.
Repeated measures. Daily symptom diaries for 5 mo,
collected every 2 weeks. Recruitment from schools.
Logistic regression adjusted for lag 1 temperature,
day of week, city.
Seoul, Incheon, Busan, Jeju, Korea
N = 696, ages NR.
Repeated measures. Daily symptom diaries for 2 mo.
Recruitment from schools. 69% participation rate.
GEE adjusted for temperature, relative humidity,
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NO2-central site
24-h avg 0
# sites NR
3
3
0
NO2-central site
24-h avg 0
1 site per 0-3 avg
community
1
NO2-central site 0
24-h avg
# sites NR
Subgroup
Analyzed
(if
applicable)
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
Incident asthma
Mild asthma,
n=43
Moderate
asthma
n = 41
Mild asthma,
n = 43
Moderate
asthma,
n = 41
All subjects
Seoul
Incheon
Busan
Jeju
1.89(1.13, 3
1.31 (0.84,2
Beta agonist
1.27(0.82, 1
1.56(0.51,4
Cough
1.21 (0.92, 1
1.61 (1.08,2
LRS:
1.44(0.96,2
LRS
1.02(1.00, 1
1.08(0.99, 1
1.08(0.99, 1
1.04(0.96,1.
0.97(0.89, 1
.15)
.02)
use
.97)
.7)
.59)
.40)
.16)
.05)
.18)
.18)
12)
.06)
Copollutant
Examination
No copollutant model.
Associations also found
with PM-,3 and SO2.
Moderate correlations
with NO2. Pearson r =
0.55 for PM13, 0.61 for
BS 0 54 for SO2
Cough w/PMio:
1.46(0.98,2.19)
w/O3: 1.61 (1.08,2.41)
w/SO2: 1.42(0.90,
2.22)
PM-io less attenuated
with adjustment for
NO2. O3 robust, SO2
reduced.
Moderate correlations
with NO2. r = 0.35 for
PM-io, 0.35 for PM2.5,
0.28 for sulfate
No copollutant model.
Association also found
with CO. Correlation
NR.
November 2013
4-132
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Andersen et
al. (2008a)
Study Population and Methodological Details
Copenhagen, Denmark
Repeated measures. Daily symptom diaries from
birth to 3 yrs, collected every 6 mo. Recruitment from
birth cohort. 95% follow up participation. Mean 805
observations/subject. GEE adjusted for age, sex,
smoking exposure, paternal asthma, temperature,
calendar season.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NC>2-central site 24-h avg
24-h avg
NOx-central site
24-h avg
1 site with 1 5 km
Subgroup
Analyzed
(if
applicable)
Age 0-1 yr
Age 2-3 yr
Age 0-1 yr
Age 2-3 yr
Odds Ratio
(95% Cl)
Single-Pollutant
Model3
Wheeze
3.13(1.27, 7.77)
1.71 (0.94, 3.10
3.26(1.14, 9.26)
1.80(0.87, 3.72)
Copollutant
Examination
For age 0-1 yr
NO2 w/PM-io:
2.46 (0.72, 8.4)
NOxwith PMi0:
2.36(0.59, 9.37)
NO2 and NOX results
also attenuated with
UFP adjustment. PM-io
associations also
attenuated.
Moderate correlations
for PM-io with NO2 and
NOx. Spearman r =
0.40, 0.43. High
correlations for CO with
NO2andNOx. r=0.74,
0.75
Ward et al.
(2002)
Rodriguez
et al. (2007)
Birmingham, Sandwell, U.K.
N = 162, age 9 yr, 27% with asthma, 31% with atopy
Repeated measures. Daily symptom diaries for 2 8-
week periods, collected weekly. Recruitment from
schools. 61% participation rate. Logistic regression
adjusted for time trend, temperature, school day.
Perth. Australia
atopic disease.
Repeated measures. Daily symptom diary from birth
to age 5 yr. Recruitment from birth cohort. >80%
follow-up participation until yrs 4and5. GEE adjusted
for temperature, humidity.
NO2-central site 0
24-h avg
Multiple sites
NO2-central site 0
1-h max
24-h avg
Average of 10
sites
Cough
Winter 0.78(0.57,1.09)
Summer 1.14(1.01, 1.27)
Wheeze (unit NR)
1.00(0.99, 1.01)
Cough
1.01 (1.00, 1.02)
Wheeze 1.01 (0.98,
1.04)
Cough 1.03(1.00,
1.06)
No copollutant model.
PM2s associated with
cough.
No copollutant model.
Associations also found
forPM2.5, BSatlagO.
November 2013
4-133
DRAFT: Do Not Cite or Quote
-------
Table 4-18 (Continued): Epidemiologic studies of respiratory symptoms and asthma medication use in children.
Study
Stern et al.
(2013)
Peel et al.
(2011)
Study Population and Methodological Details
Bern, Base, Switzerland
N = 366, ages 0-1 yr
Repeated measures. Symptoms reported weekly by
telephone for 1 yr. Recruitment from birth cohort.
High follow-up participation. GAM adjusted for study
week, sex, siblings, nursery care, prenatal maternal
smoking, postnatal maternal smoking, birth weight,
maternal atopy, parental education
Atlanta, GA area
N = 4,277, mean age 46 days, 84% premature births
Repeated measures. Followed for mean 42 days.
111,000 person-days. Recruitment from referral
center for home cardiorespiratory monitoring of
infants. Limited generalizability. Apnea events
collected electronically. GEE adjusted for long-term
trends, age.
Oxide of
Nitrogen Oxide of
Metrics Nitrogen
Analyzed Lag Day
NC>2-central site 5
1 -week avg
2 site, urban and
rural
NC>2-central site 0-1 avg
1-h max
1 site
Subgroup Odds Ratio
Analyzed (95% Cl)
(if Single-Pollutant
applicable) Model3
Daytime respiratory
symptom composite
1.20(1.04, 1.39)
Respiratory tract
infection duration
NO2<26 ppb: ref
NO2>26ppb: 1.18
(1.00, 1.39)
Apnea
1.02(0.96, 1.08)
Copollutant
Examination
No copollutant model.
PM-io lag 7 associated
with respiratory
symptoms.
Correlation NR.
w/O3: 1.00(0.96, 1.05)
O3 association robust to
NC>2 adjustment.
Moderate correlation
with NO2. Spearman r =
0.45. No association
with PM-io, coarse PM
Note: Studies are organized by population examined and then generally in order of study strength (e.g., exposure assessment method, potential confounding considered). NCICAS =
National Cooperative Inner-city Asthma Study, ICAS = Inner City Asthma Study, CAMP = Childhood Asthma Management Program, GEE = generalized estimating equations, BMI =
body mass index, GLMM = Generalized linear mixed model, ICS = inhaled corticosteroid, LRS = lower respiratory symptoms, GLM = generalized linear model, NR = not reported,
GAM = generalized additive model, BD = Bronchodilator
"Effect estimates are standardized to a 20 ppb for 24-h avg NO2, 25 ppb for 8-h max, a 30-ppb increase for 1-h max NO2, and 40-ppb increase in 24-h avg NOX
November 2013
4-134
DRAFT: Do Not Cite or Quote
-------
1 Study populations were recruited from schools, asthma or allergy clinics, or doctors'
2 offices. Asthma assessment included parental report of physician-diagnosed asthma or
3 clinical examination. Neither of these methodological issues appeared to affect whether
4 an association was found. In a-priori-determined comparisons of children with and
5 without asthma, one found stronger associations in children with asthma (Patel et al.,
6 2010); another found stronger associations in children without asthma but with 72%
7 prevalence of atopy (Barraza-Villarreal et al., 2008; Escamilla-Nunez et al., 2008).
8 Many asthma study populations had high prevalence of atopy (47-100%), and larger
9 NO2-associated increases in symptoms were found in children with asthma who also had
10 allergies (Zoraet al.. 2013; Mann et al.. 2010). These results were based on 16% to 47%
11 of the study populations. However, they are supported by findings from experimental
12 studies demonstrating increases in NO2-induced allergic responses in adults with asthma
13 and animal models of allergic disease (Section 4.2.4.3). Study populations also varied in
14 asthma severity; some studies examined mostly children with mild, intermittent asthma
15 and others examined children with persistent asthma. Comparisons by asthma severity
16 indicated larger NO2-related increases in respiratory symptoms among children with
17 mild, intermittent asthma than severe or moderate asthma (Mann et al.. 2010; Segala et
18 al.. 1998). but these results also were based on small numbers. Jalaludin et al. (2004)
19 found elevated risk in children with asthma and AHR, not children without AHR, but in a
20 3-pollutant model that is difficult to interpret because of potential multicollinearity.
21 The evidence in children with asthma is substantiated by results from several U.S.
22 multicity studies indicating associations between increases in ambient NO2 concentration
23 and increases in a composite index of respiratory symptoms. These studies each
24 examined 6-8 cities across the U.S., and only four cities were common to multiple studies
25 (Table 4-18). In each study, NO2 exposures were assigned as the average of
26 measurements from multiple city monitors, and concentrations lagged up to 6 days were
27 associated with symptoms. In NCI CAS and CAMP, respectively, ORs for asthma
28 symptoms were 1.48 (95% CI: 1.03, 3.15) for a 30-ppb increase in lag 1-6 day avg of
29 4-h avg (6 a.m.-lO a.m.) NO? (Mortimer et al.. 2002) and 1.05 (95% CI: 1.01, 1.09) for a
30 20-ppb increase in lag 0-2 day sum of 24-h avg NO2 (Schildcrout et al., 2006). The recent
31 multicity ICAS found increases in symptoms, slow play, and missed school in association
32 with a 19-day avg of 24-h avg NO2 (O'Connor et al.. 2008).
33 While a strength of the study was the proximity of most subjects to a monitor (median 2.3
34 km), a limitation is the examination of 19-day avg concentrations of NO2. Most other
35 evidence, whether from multi- or single-city, indicates associations of respiratory
36 symptoms with shorter lags of NO2 up to a few days. There is lack of biological
37 plausibility for symptoms occurring with NO2 exposure on the order of weeks, and there
November 2013 4-135 DRAFT: Do Not Cite or Quote
-------
1 is greater potential for residual temporal confounding. ICAS could not examine shorter
2 lag periods because symptom data were collected with a time resolution of two weeks.
3 Among the studies covering smaller geographic regions, primarily one or two cities,
4 many examined outdoor NO2 exposures at home or schools. School-based NO2
5 exposures were imprecisely associated with symptoms among children in Bronx, NY
6 (6-h avg school-day (9 a.m.-3 p.m.) (Spira-Cohen et al., 2011). and not consistently
7 associated with respiratory symptoms among children with asthma in El Paso, TX and
8 Ciudad Juarez, Mexico (4-day avg)(Zora et al., 2013; Sarnat et al., 2012; Holguin et al.,
9 2007). Zoraetal. (2013) found a larger association between higher outdoor school 4-day
10 avg NO2 and poorer asthma control (composite of symptoms, activity limitation and
11 asthma medication use) among children with asthma who also had allergies. Outdoor
12 home NO2 was weakly associated with respiratory symptoms among children with
13 asthma in multiple New Zealand towns (Gillespie-Bennett et al., 2011). However, daily
14 symptoms were analyzed with a single 4-week sample of NO2, which cannot represent
15 temporal variability in exposure. Home indoor NO2, which was represented as up to four
16 measurements per subject, showed stronger associations with respiratory symptoms and
17 reliever inhaler use.
18 Among studies examining NO2 exposures assigned from central sites, most found
19 associations with respiratory symptoms in children with asthma. In many cases, including
20 multiple Mexico City studies, the monitor was within 2-5 km of children's homes or
21 schools (Barraza-Villarreal et al.. 2008; Escamilla-Nunez et al.. 2008; Romieu et al..
22 2006). Results were equally robust for NO2 measured at 1 central site per location or
23 averaged over multiple sites (Mann et al.. 2010; Patel et al.. 2010; Gent et al.. 2003; Just
24 et al.. 2002).
25 Collectively, the multicity and single-city studies demonstrated increases in respiratory
26 symptoms among children with asthma in association with 24-h avg NO2. Studies
27 conducted in Mexico City found associations with 1-h max NO2 (Barraza-Villarreal et
28 al.. 2008; Escamilla-Nunez et al.. 2008; Romieu et al.. 2006). NO2 averaged over 4 hours
29 in the morning was associated with asthma symptoms in the multicity NCI CAS cohort
30 (Mortimer et al.. 2002). but school-day 6-h avg NO2 showed weak associations with
31 symptoms in children in South Bronx, NY (Spira-Cohen et al.. 2011). Collectively,
32 respiratory symptoms were associated with NO2 lagged 0 to 2 days and averaged over 2
33 to 7 days. In comparison of various lags, several found stronger associations with
34 multiday averages of NO2 than single-day lags (Mannetal. 2010; Patel etal.. 2010;
35 Escamilla-Nunez et al., 2008; Romieu et al., 2006; Mortimer et al.. 2002). But some
36 found no difference (Schildcrout et al.. 2006; Just et al.. 2002) between single- and multi-
37 day lags of NO2 exposure.
November 2013 4-136 DRAFT: Do Not Cite or Quote
-------
1 In addition to NO2, most studies found associations with copollutants such as PM10,
2 PM25, PMio-2.5, EC, BC, BS, SO2, and O3. These copollutants showed a wide range of
3 correlations with NO2 (r = 0.23-0.64), with higher correlations reported for CO and
4 BS/BC (r = 0.54-0.92). Multicity studies analyzed multipollutant models (3 pollutants)
5 which have limited implications because of potential multicollinearity (O'Connor et al..
6 2008: Mortimer et al. 2002) or joint effect models (Schildcrout et al.. 2006). In CAMP,
7 joint effects of NO2 with CO, SO2, or PM10 were similar to NO2 single-pollutant effects
8 (Schildcrout et al.. 2006). In studies covering smaller regions, NO2-wheeze associations
9 were robust to adjustment for PM10.2 5 among children in California (OR: 1.14 [95% CI:
10 0.95, 1.37] per 20-ppb increase in lag day 2 of 24-h avg NO2) (Mann etal.. 2010) and to
11 adjustment for source apportionment factor comprising EC, zinc, lead, copper, and
12 selenium in New Haven County, CT (OR: 1.08 [95% CI: 0.99, 1.18] per unit increase in
13 lag 0 NO2, unit not reported) (Gent et al.. 2003). Thus, the few copollutant-adjusted
14 results provide evidence for an independent association of NO2 with respiratory
15 symptoms in children. In the El Paso, TX study, outdoor school NO2, BC, and VOCs
16 were associated with poorer asthma control in children who also had allergies that were
17 not in the whole study population (Zoraet al.. 2013). Copollutant models only were
18 analyzed for the whole study population, and it is not clear whether NO2 associations
19 confounded these copollutants that showed a wide range of correlations with NO2
20 (Spearman r = 0.19-0.71). An independent effect of NO2 exposure is supported by
21 numerous studies that show increases in respiratory symptoms in association with
22 increases in indoor NO2 averaged over 3 to 7 days or a 4-week average (Belanger et al..
23 2013: Luetal.. 2013: Gillespie-Bennett et al.. 2011: Hansel et al.. 2008). Previous
24 findings indicated reductions in respiratory symptoms after an intervention to switch to
25 flued gas heaters led to a reduction in indoor classroom NO2 concentrations (Pilotto et
26 al.. 2004). Although potential differences in pollutant mixtures between the indoor and
27 outdoor environments have not been well characterized, a recent study found that
28 correlations between NO2 and copollutants differed between the indoor and outdoor
29 environments for BC, PM, and SO2 (Sarnat et al.. 2012). suggesting that NO2 may exist
30 as part of a different pollutant mixture in the indoor and outdoor environments.
Adults with Respiratory Disease
31 Previous and recent evidence indicates associations of ambient NO2 concentrations with
32 respiratory symptoms and asthma medication use among adults with asthma or bronchial
33 hyperresponsiveness. However, among adults with COPD, examined primarily in studies
34 reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). associations of
35 ambient oxides of nitrogen with respiratory symptoms were inconsistent. Most studies
36 were conducted in Europe.
November 2013 4-137 DRAFT: Do Not Cite or Quote
-------
Asthma
1 Adults with asthma or bronchial hyperresponsiveness were recruited primarily from
2 clinics, doctors' offices, and administrative databases and represented a mix of asthma
3 severity and prevalence of ICS use and atopy. NO2-associated increases in respiratory
4 symptoms (e.g., total respiratory symptoms, cough, wheeze) were found in most studies
5 (Maestrelli et al.. 2011; Wiwatanadate and Liwsrisakun. 2011; von Klot et al.. 2002;
6 Boezen etal. 1998; Forsberg et al., 1998). However, a study of adults in four European
7 countries found no association (Karakatsani et al.. 2012). Although NO2 induced allergic
8 inflammatory responses in subjects with asthma and animal models of allergic disease
9 (Section 4.2.4.3). ambient NO2 was not associated with respiratory symptoms with adults
10 with asthma and allergy (Feo Brito et al., 2007) or with severe allergic rhinitis in affected
11 adults (Annesi-Maesano et al.. 2012b). In addition to respiratory symptoms, evidence
12 pointed to ambient NO 2 -associated increases in asthma medication use in adults,
13 primarily bronchodilators but also ICS. These associations were found in studies of
14 individual subjects (von Klot et al., 2002; Forsberg et al.. 1998; Hiltermann et al., 1998)
15 and time-series or case-crossover studies of asthma medication sales (Carlsen et al.. 2012;
16 Laurent et al.. 2009).
17 Most studies assigned NO2 exposure from a single central site located in the community.
18 Beta-agonist sales were associated with NO2 estimated at the census block level using a
19 dispersion model which showed high correlation with ambient concentrations (r = 0.87)
20 (Laurent et al.. 2009). Symptoms and asthma medication sales were associated with
21 increases in 24-h avg ambient NO2, with Carlsen et al. (2012) finding stronger
22 associations of beta-agonist sales with 1-h max than 24-h avg NO2. Across studies,
23 respiratory symptoms were associated with lag 0 NO2. Increases in medication use or
24 sales, in particular, were associated more strongly with increases in multiday averages of
25 NO2 (i.e., lag 3-5 avg, 0-5 avg, 0-6 avg) than with single-day lags (Carlsen et al., 2012;
26 von Klot et al.. 2002; Hiltermann et al.. 1998). Medication use or sales also were
27 associated with 2-week average NO2, for which it is more difficult to control for
28 confounding by weather (Carlsen et al.. 2012; von Klot et al.. 2002).
29 For both respiratory symptoms and medication, most studies found associations with
30 copollutants such as SO2, CO, BS, PM10, PM25, and UFP. Copollutants were not
31 associated with most respiratory symptoms among adults with asthma in the Netherlands
32 (Boezen et al.. 1998). Few studies conducted copollutant analyses. Wiwatanadate and
33 Liwsrisakun (2011) did not provide quantitative data but only reported that lag 5 NO2
34 was not associated with nighttime symptoms with adjustment for lag 5 SO2. Among
35 adults with asthma in Germany, the association between lag 0-4 day avg NO2 and
36 medication use was robust to adjustment for UFP or PM2 5 (e.g., OR: 1.31 [95% CI: 1.06,
37 1.61] per 20-pbb increase in lag 0-14 day avg of 24-h avg NO2, with adjustment for UFP,
November 2013 4-138 DRAFT: Do Not Cite or Quote
-------
1 Pearson r = 0.66), but the NO2-wheeze association was attenuated with adjustment for
2 UFP (OR: 1.03 [95% CI: 0.82, 1.29]) (von Klot et al. 2002). Copollutant effect estimates
3 were attenuated with NO2 adjustment. Thus, an independent association was found for
4 medication use, but confounding by UFP was indicated for wheeze.
COPD
5 Studies of adults with COPD were conducted mostly in Europe and found no association
6 (Desqueyroux et al.. 2002; Higgins et al.. 1995) between ambient NO2 and respiratory
7 symptoms or inconsistent associations across the lags of exposure or range of outcomes
8 examined (Peacock et al.. 2011: Silkoff etal.. 2005: Harreetal.. 1997). These studies
9 recruited subjects from clinics and advertisements. Results were equally inconsistent for
10 symptoms such as cough, wheeze, dyspnea, total symptoms and medication use. There
11 was no pattern of association found for either 24-h avg or 1 -h max NO 2 or for a particular
12 lag day of exposure examined (0, 1, or longer). Most of these studies assigned exposures
13 from a single central site, but associations with symptoms and medication were
14 inconsistent for NO2 assigned from the closest site (Desqueyroux et al.. 2002) or site
15 within 5 km (Harre et al.. 1997). In the studies that found associations with specific
16 symptoms or lags of NO2, associations also were found with PM2 5, PMi0, BS and CO
17 (Peacock etal.. 2011: Silkoff etal.. 2005: Harreetal.. 1997). Among adults in New
18 Zealand, an increase in 24-h avg NO2 was associated with an increase in inhaler use in a
19 multipollutant model with CO, PM10, and SO2 (Harre etal.. 1997). which has limited
20 implications because of multicollinearity. A recent study of adults in London, U.K. found
21 that associations between lag day 1 of 1-h max NO2 and dyspnea were null with
22 adjustment for PMi0 or BS (Peacock et al.. 2011). Thus, in the few associations found
23 between increases in ambient NO2 concentration and increases in symptoms or
24 medication among adults with COPD, there was uncertainty regarding independent
25 associations with NO2.
Children in the General Population
26 Together, most previous and recent studies found associations between short-term
27 increases in ambient NO2 and respiratory symptoms in children in the general population,
28 with the strongest evidence provided by the U.S. multicity Six Cities study (Schwartz et
29 al.. 1994). Overall, evidence was more robust for cough than symptoms such as wheeze
30 and shortness of breath, which are associated more with asthma. Associations were
31 demonstrated in school-aged children recruited primarily from schools and also from an
32 allergy clinic or birth cohort, suggesting study populations were representative of the
33 general populations. NO2-associated increases in respiratory symptoms also were found
34 in infants (Stern etal.. 2013: Andersen et al.. 2008a: Peel et al.. 2007). These results may
November 2013 4-139 DRAFT: Do Not Cite or Quote
-------
1 have weaker implications because symptoms such as wheeze are common in infancy and
2 may not necessarily be correlated with respiratory morbidity later in life. Further, Peel et
3 al. (2007) examined apnea in infants on home cardiorespiratory monitors, a group that
4 may not be representative of the general population. The health status of study
5 populations was not always specified. Some studies demonstrated NO2-associated
6 increases in respiratory symptoms in children with high (72, 79%) prevalence of atopy
7 (Barraza-Villarreal et al.. 2008; Escamilla-Nunez et al., 2008) or parental history of
8 asthma (Rodriguez et al.. 2007). Other studies found no association among children
9 without asthma (Patel et al.. 2010) or children with 27% asthma prevalence (Ward et al..
10 2002).
11 A majority of the supporting evidence was for 24-h avg NO2, which was assigned from
12 central sites, one site per city or average of multiple sites per city. Studies in Mexico City
13 demonstrated associations with 1-h max NO2 measured at sites within 5 km of children's
14 schools or homes (Barraza-Villarreal et al.. 2008; Escamilla-Nunez et al.. 2008). Among
15 children in Australia followed from birth to age 5 years, Rodriguez et al. (2007) found
16 slightly larger increases in cough for increases in 24-h avg than 1-h max NO2. Most
17 studies examined multiple lags of NO2, and associations were found with lag day 0 and
18 2- to 5-day averages of NO2. The U.S. multicity study found larger associations with
19 multiday average NO2 than lag day 0 NO2 (Schwartz et al.. 1994). whereas an Australian
20 study found larger associations with lag 0 than lag 0-4 day avg NO2 (Rodriguez et al..
21 2007). In the U.S. Six Cities study, a 20-ppb increase in lag 0-3 day avg of 24-h avg NO2
22 was associated with increased cough with an OR of 1.61 (95% CI: 1.08, 2.43) (Schwartz
23 et al.. 1994). A nonlinear association was found, in which cough was found to increase
24 with increasing NO2 up to the median (among all study cities) of 13 ppb but not with
25 higher NO2 concentrations.
26 Copollutant models were not analyzed in most studies, and respiratory symptoms also
27 were associated with copollutants such as PM10, PM2 5, UFP, BS, CO, SO2, or O3.
28 Studies that reported data indicated a wide range of correlations between NO2 and
29 copollutants, with higher correlations found for UFP, BS, and CO (r = 0.61-0.75).
30 Copollutant models were not analyzed with CO (Moon et al.. 2009; Andersen et al..
31 2008a). In the Copenhagen, Denmark study of infants, NO2 and NOX associations were
32 attenuated and became imprecise with adjustment for PM10 or UFP (Andersen et al..
33 2008a). Odds ratios for PMi0 and UFP also were attenuated; thus, an independent effect
34 was not demonstrated for either NO2 or copollutants. In the U.S. Six Cities study,
35 although NO2 effect estimates were reduced with adjustment for PMi0 or SO2 (r = 0.36
36 and 0.51, respectively), they remained positive (Schwartz et al.. 1994). The ORs for a
37 20-ppb increase in lag 0-3 day avg of NO2 were 1.37 (95% CI: 0.88, 2.13) with PM10
38 adjustment and 1.42 (95% CI: 0.90, 2.28) with SO2 adjustment. The PM10 odds ratio was
November 2013 4-140 DRAFT: Do Not Cite or Quote
-------
1 robust to NO2 adjustment, whereas the SO2 odds ratio was reduced. These results from
2 the Six Cities study indicate some confounding of NO2 associations with symptoms but
3 also support an independent association with NO2.
4.2.6.2 Controlled Human Exposure Studies
4 Similar to epidemiologic studies, controlled human exposure studies did not provide
5 strong evidence for NO 2-induced increases in respiratory symptoms in adults with
6 COPD. However, unlike epidemiologic studies, they also did not provide strong evidence
7 in adults with asthma. In fact, the majority of controlled human exposure studies that
8 assessed respiratory symptoms before, during, or after exposure to NO2 did not find
9 changes, regardless of the subjects' age or disease status. Most of these studies were
10 reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). and the recent study
11 published since the last review does not materially change the previous conclusion. Study
12 details are presented in Table 4-19. but overall, studies involved NO2 exposures of
13 200-2,300 ppb for 2-5 hours and assessment of symptoms 24 hours later.
14 The majority of studies reported no change in symptoms, as measured by symptom score,
15 in healthy subjects or in adults with asthma or COPD (Gong et al.. 2005; Witten et al..
16 2005; Frampton et al., 2002; Torres etal.. 1995; Morrow et al.. 1992; Rasmussen et al..
17 1992; Linn etal.. 1985b; Kleimnan et al.. 1983). though a few studies reported
18 statistically nonsignificant increases in symptom score following NO2 exposure
19 (Frampton et al.. 2002; Hackney et al.. 1978).
20 Other studies in adults with asthma or COPD reported small, statistically significant
21 increases in symptom scores during NO2 exposures of 300-2,000 ppb for 1 hour with
22 exercise (Vagaggini etal.. 1996; LinnetaL. 1985a). though Koenig etal. (1987) found a
23 statistically nonsignificant increase in symptom score in adolescents with COPD
24 following NO2 exposure. Riedl et al. (2012) recently reported an increase in symptom
25 score in adults with asthma during, but not after, exposure to 350 ppb NO2 for 2 hours
26 with alternating periods of exercise. The increase in symptom score corresponded to a
27 subject experiencing a mild increase in any two symptoms or moderate elevation of any
28 one symptom. Symptom scores were not different between air and NO2-exposed subjects
29 when categorically grouped as respiratory, cardiovascular, or miscellaneous; nor were
30 they different when subjects were exposed to allergen after NO2 exposure.
31 Other studies investigated the effects of co-exposure to NO2 and O3 on respiratory
32 symptoms, although Adams etal. (1987) and Hazucha et al. (1994) did not find evidence
33 of effects of NO2 (600 ppb for 1-2 hours) on symptoms related to O3 exposure (200-300
34 ppb for 1-2 hours) when exposed simultaneously or sequentially.
November 2013 4-141 DRAFT: Do Not Cite or Quote
-------
Table 4-1 9
Study
Adams et al.
(1987)
Frampton et
al. (1991)
Gona et al.
(2005)
Hackney et al.
(1978)
Hazucha et
al. (1994)
Jorres et al.
(1995)
Kleinman et
al. (1983)
Koeniq et al.
(1987)
Controlled human exposure studies of respiratory symptoms.
Disease status3; n, Sex;
Age (mean ± SD)
(1-3)n = 20M, 20 F
F=21.4± 1.5yr
M= 22.7 ± 3.3 yr
(1)n = 7M, 2F;
29.9 ± 4.2 yr
(2)n = 12 M, 3F;
25.3 ± 4.6 yr
(3)n = 11 M, 4F;
23.5 ± 2.7 yr
Healthy: n = 2 M, 4 F;
68 ± 11 yr
COPD: n = 9M, 9 F;
72 ± 7 yr
n = 16 M;
26.9 ± 5.0 yr
(1,2)n=21F;
22.9 ± 3.6 yr
Healthy: n = 5 M, 3 F;
27 yr (Range: 21-33)
Asthma: n = 8 M, 4 F;
27 ± 5 yr
Asthma
n = 12 M, 19 F;
31 ± 11 yr
Healthy
(1)n = 3M, 7F
(2) n = 4 M, 6 F
Asthma
(1)n = 4M,6F
(2) n = 7 M, 3 F
14.4 yr (Range: 12-19)
Exposure Details (Concentration; Duration)
(1)600ppbNO2for1 h,
(2) 300 ppb O3 for 1 h,
(3) 600 ppb NO2 and 300 ppb O3 for 1 h;
(1-3) Exercise during entire exposure at VE= 75
L/min (M) and VE= 50 L/min (F)
(1 ) 600 ppb for 3 h,
(2) 1,500 ppb for 3 h,
(3) 50 ppb for 3h + 2,000 ppb peak for 15 min/h;
(1-3) Exercise 10 min on/20 min off at VE= ~4
times resting
(1)400ppbNO2for2h
(2) 200 ug/m3 CAPs for 2 h
(3) 400 ppb NO2 + 200 ug/m3 CAPs for 2 h
(1-3) Exercise 15 min on/15 min off atVE=
~2 times resting
1,000 ppb, 2 h/day for 2 days; Exercise 15 min
on/1 5 min off at VE= 2 times resting
(1 ) 600 ppb NO2 for 2h, air for 3h, 300 ppb O3 for
2h,
(2) air for 5h, 300 ppb O3 for 2 h;
(1,2) Exercise for 1 5 min on/1 5 min off at VE= 35
L/min
1,000 ppb for 3 h;
Exercise 10 min on/10 min off at individual's
maximum workload
200 ppb for 2 h;
Exercise 1 5 min on/1 5 min off at VE= ~2 times
resting
(1)120 ppb NO2,
(2)180ppbNO2;
(1-2) Exposures were 30 min at rest with 10 min of
moderate exercise
Time of
Symptom
Assessment
Following
exposure
Following
exposure
Before, during,
and after
exposure
After each
exposure
Not reported
Immediately
and 6 and 24 h
after exposure
Before,
immediately
after, and day
after exposure
Immediately
after, a day
after, and a
week after
exposure
November 2013
4-142
DRAFT: Do Not Cite or Quote
-------
Table 4-19 (Continued): Controlled human exposure studies of respiratory symptoms.
Study
Linn et al.
(1985b)
Linn et al.
(1985a)
Morrow et al.
(1992)
Rasmussen et
al. (1992).
Riedl et al.
(2012)
Vaqaqqini et
al. (1996)
Witten et al.
(2005)
Disease status3; n, Sex;
Age (mean ± SD)
Healthy: n = 16 M, 9 F;
Range: 20-36 yr
Asthma: n = 12 M, 11 F;
Range: 18-34 yr
COPD
n = 13 M, 9F
(1 never smoker,
13 former smokers, and
8 current smokers);
60.8 ± 6.9 yr
Healthy: n = 10 M, 10 F
(13 never smokers,
4 former smokers,
3 current smokers)
COPD: n = 13M, 7F
(14 current smokers,
6 former smokers);
59.9 ± 7.0 yr
n = 10 M, 4F;
34.4 yr (Range: 22-66)
Asthma
Phase 1 (methacholine
challenge)
n = 10 M, 5F;
37.3 ± 10.9 yr
Phase 2 (cat allergen
challenge)
n = 6M, 9F;
36.1 ± 12.5 yr
Healthy: n = 7 M;
34 ± 5 yr
Asthma: n = 4 M, 4 F;
29 ± 14 yr
COPD: n = 7 M;
58 ± 12 yr
n = 15; 32±8.6yr
Exposure Details (Concentration; Duration)
4,000 ppbfor75 min;
Two 15 min periods of exercise at VE= 25 L/min
and 50 L/min
500, 1,000, and 2,000 ppb for 1 h;
Exercise 15 min on/15 min off VE= 16 L/min
300 ppb for 4 h;
Three 7 min periods of exercise at VE= ~4 times
resting
2,300 ppb for 5 h
350 ppb for2h;
Exercise 15 min on/15 min off atVE= 15-20 L/min
300 ppb for 1 h;
Exercise at VE= 25 L/min
400 ppb for 3 h;
Exercise 30 min on/15 min off VE= 25 L/min;
Inhalation challenge with house dust mite antigen
after NO2 exposure
Time of
Symptom
Assessment
Before, during,
immediately
after, 1 day
after and 1
week after
exposure
Before, during,
immediately
after, 1 day
after, and
1 week after
exposure;
Before, during,
and after
exposure and
24-h post-
exposure
Before, during,
and after
exposure
Before, during,
1-22 h after
exposure,
Before and 2 h
after exposure
Before and
after exposure
and 6 h after
allergen
challenge
aSubjects were healthy individuals unless described otherwise.
November 2013
4-143
DRAFT: Do Not Cite or Quote
-------
4.2.6.3 Summary of Studies of Respiratory Symptoms
1 Consistent with studies reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
2 2008c). recent epidemiologic studies generally found associations between increases in
3 ambient NO2 concentrations and increases in respiratory symptoms among children with
4 asthma and children in the general population. The evidence in children with asthma is
5 weakly supported by findings in a controlled human exposure study of adolescents with
6 asthma. The robustness of epidemiologic evidence was demonstrated by results from U.S.
7 multicity studies (Schildcrout et al.. 2006; Mortimer et al.. 2002; Schwartz et al.. 1994)
8 and a meta-analysis of children with asthma ("Weinmayr et al. 2010). Results did not
9 clearly indicate larger effects in children with asthma than children without asthma (Patel
10 et al.. 2010; Barraza-Villarreal et al.. 2008).
11 Epidemiologic results also indicated associations of ambient NO2 concentrations with
12 respiratory symptoms in adults with asthma. Associations with asthma medication use or
13 sales were found more consistently in adults with asthma than children with asthma.
14 These findings are only weakly supported by results from controlled human exposure
15 studies, as only a few found increases in respiratory symptoms in adults with asthma
16 following exposure to 120-350 ppb NO2 for 30 minutes to 3 hours. Among adults with
17 COPD, both controlled human exposure and epidemiologic studies were inconsistent in
18 showing NO2-related increases in respiratory symptoms. Respiratory symptoms were not
19 examined in healthy adults in epidemiologic studies, but most controlled human exposure
20 studies did not find increases in respiratory symptoms in healthy adults following
21 exposures of 300-2,300 ppb NO2 for 1 to 5 hours.
22 Respiratory symptoms were associated more consistently with NO2 measured at central
23 monitoring sites than schools (Zoraet al.. 2013; Sarnat et al.. 2012; Spira-Cohen et al..
24 2011; Holguin et al.. 2007). Several of the school studies were conducted in the same or
25 neighboring communities. A majority of evidence was for 24-h avg NO2; however,
26 associations also were found with shorter averaging times, 4-h avg (6 a.m.-10 a.m.) in a
27 U.S. multicity study (Mortimer et al.. 2002) and 1-h max (Carlsen et al.. 2012; Barraza-
28 Villarreal et al.. 2008; Rodriguez et al.. 2007). Comparisons of averaging times did not
29 clearly indicate larger ORs for 24-h avg NO2 or 1-h max NO2 (Carlsen et al.. 2012;
30 Rodriguez et al.. 2007). Increases in respiratory symptoms were found with single-day
31 lags of ambient NO2 concentrations of 0 to 7 days and multiday averages of 2 to 7 days.
32 In some studies, associations were inconsistent among the lags examined. Several studies
33 found stronger associations for multiday averages of NO2, including multicity U.S.
34 studies (Mortimer et al.. 2002; Schwartz et al.. 1994) and most studies of asthma
35 medication use or sales (Carlsen et al.. 2012; von Klot et al.. 2002; Hiltermann et al..
November 2013 4-144 DRAFT: Do Not Cite or Quote
-------
1 1998). The range of mean ambient NO2 concentrations in studies was 7-40.4 ppb for
2 24-h avg NO2 and 18-37.4 ppb for shorter averaging times. With respect to the
3 concentration-response relationship, the Six Cities Study of children found a nonlinear
4 relationship with cough that was evident in the lower range of the 24-h avg NO2
5 distribution, i.e., days with <13 ppb NO2 (Schwartz et al.. 1994).
6 Several studies provide evidence for associations between indoor home NO2 exposures
7 with respiratory symptoms among children with asthma (Belanger et al.. 2013; Lu et al..
8 2013; Gillespie-Bennett et al., 2011; Hansel et al., 2008) and between reductions in
9 indoor NO2 and reduction in symptoms (Pilotto et al.. 2004). Evidence is inconsistent in
10 adults with COPD (Hansel et al.. 2013). Indoor NO2 exposures at ice arenas also were
11 associated with respiratory symptoms in hockey players (Salonen et al., 2008). These
12 findings for indoor NO2 support an independent effect of NO2 exposure. Among studies
13 of outdoor NO2, a small percentage examined copollutant models. While some results
14 indicated confounding by copollutants, several observations supported an independent
15 association for ambient NO2 exposure. In studies of children with asthma, associations of
16 NO2 with respiratory symptoms were robust to adjustment for PMi0.2 5, O3, or a source
17 factor comprising EC and PM2 5 metal components (Mann et al.. 2010; Gent et al., 2003;
18 Mortimer et al.. 2002). In the Six Cities study, the NO2-cough association was reduced in
19 magnitude and precision with adjustment for PMi0 or SO2 but remained positive
20 (Schwartz et al.. 1994). indicating only partial confounding of the NO2 association. In
21 adults, adjustment for UFP, which was highly correlated with NO2 (r = 0.63-0.92), did
22 not affect the NO2 association with medication use but attenuated the association with
23 symptoms (Andersen et al., 2008a; von Klot et al.. 2002). Copollutant associations were
24 reduced with adjustment for NO2, indicating that NO2 may have confounded copollutant
25 associations.
26 Biological plausibility for effects in children and adults with asthma is provided by
27 evidence for NO2-induced AHR in adults in asthma (Section 4.2.2) and evidence for
28 NO2-induced increases in allergic inflammation in adults with asthma and animal models
29 of allergic disease (Section 4.2.4.3). Consistent with the latter, ambient NO2-associated
30 increases in respiratory symptoms were found in children with atopy as determined by
31 skin prick test (Barraza-Villarreal et al.. 2008; Escamilla-Nunez et al.. 2008). And, larger
32 NO2-associated increases in respiratory symptoms were found in children with asthma
33 who also had allergy, although based on analyses of small groups (Zoraet al.. 2013;
34 Mann etal.. 2010). In contrast, such associations were not demonstrated in adults with
35 asthma and pollen allergy (Feo Brito et al.. 2007) or adults with severe allergic rhinitis
36 (Annesi-Maesano et al., 2012b). Similarly, a recent controlled human exposure study of
37 adults with allergic asthma did not find NO2-induced increases in allergic inflammation,
38 and increases in symptoms were found only during, not after, exposure. Nonetheless,
November 2013 4-145 DRAFT: Do Not Cite or Quote
-------
1 there is robust evidence for NO2-associated increases in respiratory symptoms in children
2 with asthma and evidence describing underlying mechanisms. Because increases in
3 respiratory symptoms can lead to the seeking of medical treatment, this evidence
4 provides biological plausibility for epidemiologic associations observed between short-
5 term increases in ambient NO2 concentration and increases in respiratory-related hospital
6 admissions, ED visits, and physician visits as described in Section 4.2.7.
4.2.7 Respiratory Hospital Admissions and Emergency Department Visits
4.2.7.1 Summary of Findings from the 2008 ISA for Oxides of
Nitrogen
7 Epidemiologic studies examining the association between short-term NO2 exposures and
8 respiratory-related hospital admissions or ED visits were not available until after the
9 completion of the 1993 AQCD for Oxides of Nitrogen. As a result, the 2008 ISA for
10 Oxides of Nitrogen (U.S. EPA. 2008c). consisted of the first thorough evaluation of
11 respiratory morbidity in the form of respiratory-related hospital admissions and ED visits.
12 Of the studies evaluated, the majority consisted of single-city, time-series studies that
13 examined all respiratory hospital admissions or ED visits with additional cause-specific
14 studies of asthma and COPD. Studies of all respiratory and asthma hospital admissions
15 and ED visits consistently reported positive associations with short-term NO2 exposures.
16 These associations were generally found to be robust and independent of the effects of
17 ambient particles or gaseous copollutants (U.S. EPA. 2008c). The strongest evidence was
18 from respiratory studies that focused on children and older adults (65+ years of age) and
19 asthma studies that focused on all ages and children. The 2008 ISA for Oxides of
20 Nitrogen found limited evidence for associations between short-term NO2 exposure and
21 other respiratory hospital admission and ED visit outcomes, such as COPD. This
22 evidence supporting NO2-associated increases in respiratory-related hospital admission
23 and ED visits contributed heavily to the 2008 ISA for Oxides of Nitrogen conclusion that
24 "there is a likely causal relationship between short-term exposure to NO2 and effects on
25 the respiratory system" (U.S. EPA. 2008c).
4.2.7.2 Evaluation of Studies Published Since 2008 ISA for
Oxides of Nitrogen
26 Since the completion of the 2008 ISA for Oxides of Nitrogen, relatively fewer studies
27 have examined the association between short-term exposure to ambient NOX or NO2 and
November 2013 4-146 DRAFT: Do Not Cite or Quote
-------
1 respiratory-related hospital admissions and ED visits. The following sections characterize
2 recent studies in the context of the collective body of evidence evaluated in the 2008 ISA
3 for Oxides of Nitrogen. Where possible, the emphasis within this section is on multicity
4 studies, which allow for the examination of the relationship of short-term exposures to
5 NOX or NO2 with respiratory-related hospital admissions or ED visits over a large
6 geographic area using a common statistical methodology. The remaining evaluated
7 studies consist of single-city studies conducted over long durations or in areas with large
8 populations with a specific emphasis on studies that provide insight into the relationship
9 between short-term exposure to NOX or NO2 and respiratory-related hospital admissions
10 and ED visits with respect to potential copollutant confounding, seasonal differences in
11 risk estimates, or factors that may modify risk of NO2-related hospital admissions and
12 ED visits.
13 In this ISA, respiratory-related hospital admissions and ED visits are evaluated separately
14 because it is likely that a small percentage of respiratory ED visits will be admitted to the
15 hospital. Therefore, ED visits may represent potentially less serious, but more common,
16 outcomes. Additionally, results are presented as either a collection of respiratory
17 diagnoses or as an individual diagnosis (e.g., asthma). Table 4-20 presents characteristics
18 of studies discussed within this section along with the air quality characteristics of the
19 city, or across all cities, evaluated in each study.
20
November 2013 4-147 DRAFT: Do Not Cite or Quote
-------
Table 4-20 Mean and upper percentile concentrations of respiratory-related hospital admission and emergency
department visit studies published since the 2008 ISA for Oxides of Nitrogen.
Study
Mean Upper Percentile of
Location Type of Visit (ICD9/10) Years Metric Concentration (ppb) Concentrations (ppb)
Hospital Admissions
Cakmak et al. (2006)
Wong et al. (2009)
Dales et al. (2006)
Faustini et al. (2013)
Sonetal. (2013)
Samolietal. (2011)
Ko et al. (2007b)
10 Canadian Hospital Admissions: 1993-2000 24-h avg 21.4 Max: 44- 134
cities All Respiratory (466, 480-486, 490-494,
496)
Hong Kong Hospital Admissions: 1996-2002 24-h avg 31.2 75th: 37.0
All Respiratory (460-519) Max: 89.4
COPD (490-496)
Acute Respiratory Disease
(460-466, 480-487)
11 Canadian Hospital Admissions: 1986-2000 24-h avg 21.8 95th: 21 -43
cities AN Respiratory (799.0, 799.1, 786.0, 769,
768.9, 770.8, 486)
6 Italian cities Hospital Admissions: 2001-2005 24-h avg 24.1-34.6 NR
All Respiratory (460-51 9)
COPD (490-492,494,496)
LRTI (466, 480-487)
8 South Hospital Admissions: 2003-2008 24-h avg 11.5-36.9 NR
Korean cities Respiratory diseases (JOS, J18, J20, J21,
J40-42, J44-46, J67)
Asthma (J45, J46)
Allergic disease (J30, J45, L20)
Athens, Hospital Admissions: 2001-2004 1-h max 44.4 75th: 53.1
Greece Asthma (493, 493.9)
Hong Kong Hospital Admissions: 2000-2005 24-h avg 28.3 75th: 33.8
Asthma (493) Max: 79 5
November 2013
4-148
DRAFT: Do Not Cite or Quote
-------
Table 4-20 (Continued): Mean and upper percentile concentrations of respiratory-related hospital admission and emergency
department visit studies published since the 2008 ISA for Oxides of Nitrogen.
Study
Ko et al. (2007a)
HEI Collaborative
Working Group (2012)
Mehtaetal. (2013)
Seqala et al. (2008)
Grineski et al. (2010)
Location Type of Visit (ICD9/10)
Hong Kong Hospital Admissions:
COPD(491,492, 496)
Ho Chi Minh Hospital Admissions:
City, Vietnam Acute Lower Respiratory Infection (J13-16,
18,21)
Paris, France Hospital Admissions:
Bronchiolitis
Phoenix, AZ Hospital Admissions:
Asthma (493)
Years Metric
2000-2004 24-h avg
2003-2005 24-h avg
1997-2001 24-h avg
2001-2003 1-h max
(Evening hours:
4 p.m. -11 pm)
Mean
Concentration (ppb)
27.2
11.7
27.0
46.0
Upper Percentile of
Concentrations (ppb)
75th: 34.0
Max: 83.8
Max: 29.2
Max: 90.4
75th: 57.0
Max: 79.0
ED Visits
Darrowet al. (2011 a)
Stieb et al. (2009)
Strickland et al. (2010)
Atlanta, GA ED Visits:
All Respiratory (460-466, 477, 480-486,
491-493, 496, 786.09)
7 Canadian ED Visits:
cities Respiratory Infection (464, 466, 480-487)
Asthma (493)
COPD (490-492, 494-496)
Atlanta, GA ED Visits:
Asthma (493.0-493.9, 786.09)
1993-2004
1-h max
24-h avg
Commute
(7 a.m. -10 a.m.,
4 p.m.-7 p.m.)
Day-time
(8 a.m. -7 p.m.)
Night-time
(12 a.m.-6 a.m.)
1992-2003 24-h avg
1993-2004 1-h max
1-h max: 43
24-h avg: 22
Commute: 21
Day-time: 17
Night-time: 25
9.3 22.7
23.3
75th:
1-h max: 53
24-h avg: 28
Commute: 27
Day-time: 22
Night-time: 35
Max:
1-h max: 181
24-h avg: 74
Commute: 97
Day-time: 82
Night-time: 97
75th: 12.3-27.6
NR
November 2013
4-149
DRAFT: Do Not Cite or Quote
-------
Table 4-20 (Continued): Mean and upper percentile concentrations of respiratory-related hospital admission and emergency
department visit studies published since the 2008 ISA for Oxides of Nitrogen.
Study
Strickland et al. (2011)
Lietal. (2011 b)
Villeneuve et al. (2007)
Jalaludin et al. (2008)
Arbex et al. (2009)
Seqala et al. (2008)
Zemeketal. (2010)
Orazzo et al. (2009)
Location
Atlanta, GA
Detroit, Ml
Edmonton,
Canada
Sydney,
Australia
Sao Paulo,
Brazil
Paris, France
Edmonton,
Canada
6 Italian cities
Type of Visit (ICD9/10)
ED Visits:
Asthma (493.0-493.9, 786.09)
ED Visits:
Asthma (493)
ED Visits:
Asthma (493)
Influenza (487)
ED Visits:
Asthma (493)
ED Visits:
Bronchitis (J40, J41, J42)
Emphysema (J43)
COPD (J44)
ED Visits:
Bronchiolitis
ED Visits:
Otitis Media (382.9)
ED Visits:
Wheeze
Mean
Years Metric Concentration (ppb)
1993-2004 1-h max Central monitor: 42.0
Unweighted average:
27.7
Population-weighted
average: 22.0
2004-2006 24-havg 15.7
1992-2002 24-havg 50th: 17.5 Summer
50th: 28.5 Winter
1997-2001 1-h max 23.2
2001-2003 1-h max 63.0
1997-2001 24-havg 27.0
1992-2002 24-havg 21.9
1QQR 9fin9 94 h awn 91 A 41 9
Upper Percentile of
Concentrations (ppb)
NR
75th: 21. 2
Max: 55.2
75th: 22.0 Summer
75th: 35.5 Winter
Max: 59.4
75th: 78.6
Max: 204.6
Max: 90.4
75th: 27.6
NR
November 2013
4-150
DRAFT: Do Not Cite or Quote
-------
Table 4-20 (Continued): Mean and upper percentile concentrations of respiratory-related hospital admission and emergency
department visit studies published since the 2008 ISA for Oxides of Nitrogen.
Study
Location Type of Visit (ICD9/10)
Mean Upper Percentile of
Years Metric Concentration (ppb) Concentrations (ppb)
Physician Visits
Burra et al. (2009)
Toronto,
Canada
Physician Visits:
Asthma (493)
1992-2001 1-hmax
39.2
95th: 60
Max: 105
Sinclair et al. (2010)
Atlanta, GA Physician Visits:
Asthma
Upper Respiratory Infection
Lower Respiratory Infection
1998-2002 1-hmax
1998-2000:49.8
2000-2002: 35.5
1998-2002:41.7
NR
Villeneuve et al.
(2006b)
Toronto,
Canada
Physician Visits:
Allergic rhinitis
1995-2000 24-havg 25.4
Max: 71.7
November 2013
4-151
DRAFT: Do Not Cite or Quote
-------
4.2.7.3 Hospital Admissions Studies
1 Of the recent studies evaluated that examined the association between short-term
2 exposure to concentrations of oxides of nitrogen and respiratory-related hospital
3 admissions, the majority focus on short-term exposures to NO2 in single-city studies.
4 Most of these studies are limited in that they encompass short durations (<5 years) or a
5 small number of daily hospital admissions (i.e., <2 admissions per day). A
6 comprehensive list of these studies can be found in Supplemental Table S4-1. (U.S. EPA.
7 2013d). The following section evaluates the multi- and single-city studies detailed in
8 Table 4-20 that examined the association between short-term NO2 concentrations and
9 respiratory-related hospital admissions.
Respiratory Disease
10 Multicity studies conducted in Canada (Cakmak et al.. 2006; Dales et al.. 2006). Italy
11 (Taustini et al.. 2013) and Korea (Son et al.. 2013). as well as a single-city study
12 conducted in Hong Kong (Wong et al.. 2009) examined the association between short-
13 term NO2 concentrations and hospital admissions for all respiratory diseases, each
14 focusing on a different age range. The results from these studies are consistent with those
15 studies evaluated in the 2008 ISA for Oxides of Nitrogen (Figure 4-4). Several of these
16 studies also examined potential effect modification by SES, influenza, age, and sex; the
17 results of which are discussed later in the section.
18 Cakmak et al. (2006) focused on all ages in 10 Canadian cities with the primary objective
19 of the study being to examine the potential modification of the effect of ambient air
20 pollution on daily respiratory hospital admissions by education and income using a time-
21 series analysis conducted at the city-level (the effect modification analysis is discussed in
22 detail later in the section). The authors calculated a pooled estimate across cities for each
23 pollutant using a random effects model by first selecting the lag day with the strongest
24 association from the city-specific models. For NO2, the mean lag day across cities that
25 provided the strongest association and for which the pooled effect estimate was
26 calculated was 1.4 days. At this lag, Cakmak et al. (2006) reported a 2.3% increase
27 (95% CI: 0.2, 4.5%) in respiratory hospital admissions for a 20-ppb increase in 24-h avg
28 NO2 concentrations. This result is consistent with Wong et al. (2009) in a study
29 conducted in Hong Kong aimed to examine whether influenza modifies the relationship
30 between air pollution exposure and hospital admissions. Wong et al. (2009) observed a
31 3.2% (95% CI: 1.9, 4.5) increase in all respiratory hospital admissions for all ages at lag
32 0-1 days for a 20-ppb increase in 24-h avg NO2 concentrations.
November 2013 4-152 DRAFT: Do Not Cite or Quote
-------
1 Cakmak et al. (2006) also examined the potential confounding by other pollutants but
2 only through the use of a multipollutant model (i.e., two or more additional pollutants
3 included in the model). These models are difficult to interpret due to the potential
4 multicollinearity between pollutants and are not evaluated in this ISA.
5 In an additional multicity study conducted in 11 Canadian cities, Dales et al. (2006)
6 focused on NO2-associated respiratory hospital admissions in neonatal infants (ages
7 0-27 days). The investigators used a statistical analysis approach similar to Cakmak et al.
8 (2006) (i.e., time-series analysis to examine city-specific associations, and then a random
9 effects model to pool estimates across cities). Dales et al. (2006) observed that the mean
10 lag day across cities that provided the strongest association for NO2 was 1 day, which
11 corresponded to 6.5% (95% CI: 3.5, 9.6%) increase in neonatal respiratory hospital
12 admissions for a 20-ppb increase in 24-h avg NO2 concentrations. Similar to Cakmak et
13 al. (2006). Dales et al. (2006) only examined the potential confounding effects of other
14 pollutants on the NO 2 -respiratory hospital admissions association through the use of
15 multipollutant models, which are not informative due to potential multicollinearity issues.
16 The results of Cakmak et al. (2006) and Wong et al. (2009). are further supported by Son
17 et al. (2013) in a study that examined the association between short-term exposures to air
18 pollution and respiratory-related hospital admissions in 8 South Korean cities. It is
19 important to note that South Korea has unique demographic characteristics with some
20 indicators more in line with other more developed countries (e.g., life expectancy, percent
21 of population living in urban areas), but because it represents a rapidly developing Asian
22 country, it is likely to have different air pollution, social, and health patterns than less
23 industrialized Asian nations or Western nations that developed earlier (Son et al.. 2013).
24 In a time-series analysis using a two-stage Bayesian hierarchical model, Son et al. (2013)
25 examined both single-day lags and cumulative lags up to 3 days (i.e., lag 0-3). The
26 authors only presented NO2 results for the strongest lag and observed a 3.6% increase
27 (95% CI: 1.0, 6.1) in respiratory disease hospital admissions at lag 0 for a 20-ppb
28 increase in 24-h avg NO2 concentrations. The authors did not conduct copollutants
29 analyses; however, similar patterns of associations were observed across pollutants that
30 were moderately (PM10 [r = 0.5]; SO2 [r = 0.6]) to highly correlated (CO [r = 0.7]) with
31 NO2.
32 Faustini et al. (2013) focused on examining the relationship between short-term air
33 pollution exposures and respiratory hospital admissions, specifically on the adult
34 population (i.e., individuals 35 years of age and older) in 6 Italian cities. In a time-series
35 analysis the authors examined the lag structure of associations through single-day lags as
36 well as cumulative lags, using cubic polynomial distributed lags, in an attempt to identify
37 whether the NO2 effect on respiratory-related hospital admissions was immediate (lag 0,
November 2013 4-153 DRAFT: Do Not Cite or Quote
-------
1 lag 0-1 days), delayed (lag 2-5 days), or prolonged (lag 0-3, 0-5 days). The authors
2 reported that NO2 was most strongly associated with respiratory hospital admissions at
3 lag 0-5 days (4.6% [95% CI: 0.87, 8.3] for a 20-ppb increase in 24-h avg NO2
4 concentrations), which differs from Cakmak et al. (2006) and Dales et al. (2006) where
5 the strongest effects were observed at lags less than 2 days. However, Faustini et al.
6 (2013) did observe positive associations, although smaller in magnitude (ranging from
7 2.5-2.9%) at the shorter lags (i.e., lag 0 and 0-1 days). Faustini et al. (2013) also indicated
8 that the NO2 association was independent of that with PMi0. In a copollutant model with
9 PM10, the NO2 association with respiratory hospital admissions at lag 0-5 days was
10 attenuated slightly, but remained positive (3.3% [95% CI: -1.1, 7.8]).
Cause-specific Respiratory Outcomes
Asthma
11 The 2008 ISA for Oxides of Nitrogen generally found consistent evidence of a positive
12 association between short-term NO2 exposures and asthma hospital admissions.
13 Generally, studies that examined the effect of short-term NO2 exposures on asthma
14 hospital admissions have been limited to single-cities. It is important to note the results of
15 these studies should be viewed with caution because they tended to include ages <5 years
16 in the study population, which is problematic considering the difficulty in reliably
17 diagnosing asthma within this age range [National Asthma Education and Prevention
18 Program Expert Panel (2007)1. but it is unlikely a systematic positive bias would be
19 introduced. In contrast, to account for this difficulty, the majority of studies on asthma
20 ED visits (Section 4.2.7.4) have excluded ages <2 years in analyses.
21 Samoli et al. (2011). in a time-series study conducted in Athens, Greece, evaluated the
22 association of multiple ambient air pollutants and pediatric asthma hospital admissions
23 for ages 0-14 years. In an all-year analysis, the authors reported a positive association
24 with NO2 (6.4 % [95% CI: -3.8, 17.6]; lag 0 increase for a 30-ppb increase in 1-h max
25 NO2 concentrations); however, the magnitude of the association was small compared to
26 that observed for SO2 and PMi0. An examination of additional lags (quantitative results
27 not presented) revealed similar associations at lag 2 and a 0-2 day distributed lag. In
28 copollutant analyses, NO2 risk estimates were robust when O3 (7.6% [95% CI: -2.7,
29 19.0]) was included in the model, and were attenuated but remained positive with wide
30 confidence intervals when including PM10 in the model (3.1% [95% CI: -7.3, 14.6]).
31 However, there was evidence of confounding of the NO2 association when SO2 was
32 included in the model, which was most highly correlated with NO2 (r = 0.66). In the
33 copollutant model with SO2 the NO2 effect estimate was no longer positive (-4.3% [95%
34 CI: -16.9, 10.2]).
November 2013 4-154 DRAFT: Do Not Cite or Quote
-------
1 Ko et al. (2007b) conducted a more extensive analysis than Samoli etal. (2011) to
2 examine associations between short-term air pollution exposures and asthma hospital
3 admissions at both single- and multi-day lags. In a time-series analysis conducted in
4 Hong Kong, which included all ages, the authors reported positive associations at single-
5 day lags that were smaller in magnitude than those observed in Samoli etal. (2011) (e.g.,
6 3.4% [95% CI: 1.9, 5.4]; lag 0 for a 20-ppb increase in 24-h avg NO2 concentrations).
7 These results are consistent with those of Son et al. (2013) in 8 South Korean cities,
8 which found the strongest association between short-term NO2 exposures and asthma as
9 well as allergic disease hospital admissions, which encompasses asthma, at lag 0 (3.6%
10 [95% CI: 0.5, 6.8] and 3.8% [95% CI: 1.0, 6.6], respectively) for a 20-ppb increase in
11 24-h avg NO 2 concentrations. However, unlike Samoli etal. (2011) and Son etal. (2013).
12 Ko et al. (2007b) found the strongest evidence of an association between short-term NO2
13 exposures and asthma hospital admissions at multi-day lags of 0-3 (10.9% [95% CI: 8.1,
14 13.8]) and 0-4 (10.9% [95% CI: 8.1, 13.4]) days. In a copollutant analysis with O3, the
15 authors reported evidence of a reduction in NO2 risk estimates although they remained
16 positive (2.3% [95% CI: -0.8, 5.8]; lag 0-4 days), which is not consistent with the results
17 of the copollutants analysis in Samoli et al. (2011). This attenuation occurred even
18 though NO2 and O3 were not well correlated (r = 0.34) in Hong Kong.
COPD
19 Of the studies evaluated in the 2008 ISA for Oxides of Nitrogen, relatively few examined
20 the association between short-term NO2 exposure and COPD hospital admissions;
21 however, these studies provided initial evidence of a positive association. Consistent with
22 the 2008 ISA for Oxides of Nitrogen, a few recent studies have focused on the outcome
23 of COPD hospital admissions, and these studies further support the initial evidence
24 observed in the 2008 ISA for Oxides of Nitrogen. Faustini et al. (2013). in the 6 Italian
25 city analysis discussed above, also examined the association between short-term NO2
26 concentrations and COPD hospital admissions. Unlike the pattern of associations
27 observed for total respiratory hospital admissions, the authors observed stronger evidence
28 for immediate (lag 0: 4.6% [95% CI: 0.64, 8.6] for a 20-ppb increase in 24-h avg NO2
29 concentrations) NO2 effects on COPD hospital admissions with positive, albiet smaller
30 associations when examining prolonged effects, (3.3% for lag 0-3 days and 3.1% for lag
31 0-5 days). There was no evidence for delayed effects (lag 2-5 days). In a copollutant
32 model with PMi0, the association between NO2 and COPD hospital admissions remained
33 robust (3.9% [95% CI: -1.7, 9.8]; lag 0).
34 In a study conducted in Hong Kong, Ko et al. (2007a) also examined the lag structure of
35 associations between short-term air pollution exposures and COPD hospital admissions.
36 In analyses of both single-day lags and multiday averages, Ko et al. (2007a) observed the
November 2013 4-155 DRAFT: Do Not Cite or Quote
-------
1 largest magnitude of an association at lags ranging from 0-3 to 0-5 days (10.1% [95% CI:
2 8.5, 12.2)] for a 20-ppb increase in 24-h avg NO2 concentrations at both 0-3 and 0-5 day
3 lags). Although Ko et al. (2007a) reported associations larger in magnitude for multiday
4 averages, the authors also observed a positive association across single day lags, with lag
5 0 having one of the stronger associations (3.4% [95% CI: 1.9, 5.0]), which is of similar
6 magnitude to the lag 0 effect observed in Faustini et al. (2013). Ko et al. (2007a) only
7 examined the potential confounding effects of copollutants through the use of 3 and 4
8 pollutant models, which are difficult to interpret. However, when comparing single-
9 pollutant results for NO2 with the other pollutants examined (O3, PM25, and PM10),
10 similar patterns of associations were observed across pollutants.
Respiratory Infection
11 To date, very few studies have examined the association between short-term NO2
12 exposures and respiratory infection hospital admissions. Overall, these studies have
13 generally not provided consistent evidence of an association (U.S. EPA. 2008c). Few
14 recent studies have examined the association between short-term NO2 exposures and
15 respiratory infection hospital admissions. A time-series study conducted in Ho Chi Minh
16 City, Vietnam (Mehta et al.. 2013; HEI Collaborative Working Group. 2012) examined
17 the association between short-term air pollution exposures and pediatric (ages 28 days - 5
18 years) hospital admissions for acute lower respiratory infections (ALRI, including
19 bronchiolitis and pneumonia). In a time-stratified case-crossover analysis focused only on
20 the average of a 1-6 day lag, there was no evidence of an association between NO2 and
21 ALRI hospital admissions in the all-year analysis (-4.0% [95% CI: -18.0, 12.5] for a
22 20-ppb increase in 24-h avg NO2 concentrations).
23 In an additional study that also examined respiratory infections (i.e., bronchiolitis) in
24 children, Segala et al. (2008) focused on associations with winter (October-January) air
25 pollution because it is the time of year when respiratory syncytial virus (RSV) activity
26 peaks. It has been hypothesized that air pollution exposures may increase the risk of
27 respiratory infections, including bronchiolitis due to RSV (Segala et al.. 2008). Focusing
28 on children <3 years of age in Paris, France, the authors conducted a bidirectional case-
29 crossover analysis along with a time-series analysis to examine air pollution (i.e., PMi0,
30 BS, NO2, SO2) associations with bronchiolitis ED visits (see Section 4.2.7.4) and
31 hospital admissions. Although the authors specify the bidirectional case-crossover
32 approach was used to "avoid time-trend bias", it must be noted that the bidirectional
33 approach has been shown to bias results (Segala etal.. 2008; LevyetaL 2001). In the
34 case-crossover analysis NO2 was associated with bronchiolitis hospital admissions
35 (15.9% [95% CI: 7.7, 29.0], lag 0-4 days for a 20-ppb increase in 24-h avg NO2
36 concentrations); NO2 was not examined in the time-series analysis. Although a positive
November 2013 4-156 DRAFT: Do Not Cite or Quote
-------
1 association was observed, the authors did not conduct copollutants analyses. The lack of
2 copollutant analyses complicates the interpretation of these results because the pollutants
3 were highly correlated, ranging from r = 0.74-0.83.
4 Faustini et al. (2013). in the analysis of air pollution in 6 Italian cities, also examined
5 associations with lower respiratory tract infection (LRTI) hospital admissions. However,
6 the authors only focused on LRTIs in individuals with COPD over the age of 35. Unlike
7 the analyses focusing on only COPD hospital admissions where the strongest associations
8 were for immediate effects, i.e., lag 0 and 0-1 days, for the population of individuals with
9 COPD that had a hospital admission for a LRTI there was no evidence of an effect at
10 these shorter durations; the largest effects were observed at lag 2-5 days (10.0% [95% CI:
11 -2.7, 24.3]). The authors examined the NO2 association with LRTI hospital admissions in
12 copollutant models with PMi0 at lag 0-5 days and, consistent with the other endpoints
13 examined, reported that results remained robust (7.8% [95% CI: -6.5, 24.2]).
Sensitivity Analyses and Effect Modification of Relationships between NO2
and Respiratory-Related Hospital Admissions
Model Specification
14 A question that often arises in the examination of associations between air pollution and a
15 health effect is whether the statistical model employed adequately controls for the
16 potential confounding effects of temporal trends and meteorological conditions. Son et al.
17 (2013). in the study of 8 South Korean cities, conducted a sensitivity analyses to identify
18 if risk estimates changed depending on the degrees of freedom (df) used to control for
19 temporal trends and meteorology covariates (i.e., temperature, humidity, and barometric
20 pressure). The authors reported that this association between short-term NO2 exposures
21 and all of the respiratory hospital admission outcomes examined was sensitive to using
22 less than 6 degrees of freedom (df) per year to control for temporal trends, but was stable
23 when using 6-10 df per year. Additionally, when varying the number of df used for the
24 meteorology covariates from 3 to 6 df as well as the lag structure (i.e., lag 0 and lag 0-3
25 days), the NO2 association remained robust for all respiratory outcomes.
Examination of Seasonal Differences
26 In addition to examining the association between short-term NO2 concentrations and
27 respiratory-related hospital admissions in all-year analyses, some studies also conducted
28 seasonal analyses. When evaluating these studies it is important to note that the
29 differences in the locations examined across studies complicate the ability to draw overall
30 conclusions regarding the seasonal patterns of associations.
November 2013 4-157 DRAFT: Do Not Cite or Quote
-------
1 Son et al. (2013). in the study of 8 South Korean cities, examined potential seasonal
2 differences across respiratory hospital admission outcomes. For all outcomes examined,
3 the association with NO2 was largest in magnitude during the summer (Respiratory
4 Diseases: 8.3% [95% CI: 2.8, 14.3], lag 0; Asthma: 16.2% [95% CI: 5.1, 28.6], lag 0;
5 Allergic Disease: 15.9 [95% CI: 4.6, 28.5], lag 0 for a 20-ppb increase in 24-h avg NO2
6 concentrations). Across the 8 cities, NO2 concentrations were lowest during the summer
7 season (<20 ppb compared to >24 ppb in the other seasons).
8 The asthma hospital admission results of Son et al. (2013) are supported by Samoli et al.
9 (2011) in a study conducted in Athens, Greece. Although risk estimates for asthma
10 hospital admissions were relatively consistent across winter, spring, and autumn, ranging
11 from a 13.1 to a 13.8% increase per 20-ppb increase in 24-h avg NO2, the largest percent
12 increase was observed for the summer (28.7% [95% CI: -3.4, 71.3]).
13 Additional studies conducted in Hong Kong and Ho Chi Minh, Vietnam highlight the
14 vast differences observed in seasonal analyses based on geographic location. In studies
15 conducted in Hong Kong that examined asthma hospital admissions (Ko et al.. 2007b)
16 and COPD hospital admissions (Ko et al.. 2007a). larger associations were observed in
17 the cold season (i.e., December to March) than the warm season.
18 Mehtaetal. (2013) in the examination of acute lower respiratory infection (ALRI)
19 hospital admissions in Vietnam examined potential seasonal differences in associations
20 by dividing the year into the dry (November-April) and rainy seasons (May-October).
21 Unlike the other pollutants examined in the study for which concentrations differed
22 drastically between these seasons, mean NO2, concentrations were similar across
23 seasons: 23.1 ppb in the dry season and 21.2 ppb in the rainy season. However, NO2 was
24 strongly correlated with PMi0 (r = 0.78) only during the dry season; the correlation
25 dropped to r = 0.18 in the rainy season. In seasonal analyses, Mehtaetal. (2013) reported
26 that NO2 was consistently associated with ALRI hospital admissions in the dry season
27 (35.9% [95% CI: 3.0, 79.3] per 20-ppb increase in 24-h avg NO2, lag 1-6 day avg), with
28 no evidence of an association in the rainy season. No pollutants were associated with
29 ALRI hospital admissions during the rainy season. In copollutant analyses for the dry
30 season NO2 was robust to the inclusion of other pollutants (SO2, O3, or PMi0) with the
31 magnitude of the effect remaining constant or increasing slightly. In the dry season
32 analyses, the PMi0 results were robust to all pollutants in copollutant models except NO2,
33 where PM10 risk estimates were dramatically attenuated. These results, along with the
34 high correlation between PMi0 and NO2, complicate the interpretation of the NO2 results.
November 2013 4-158 DRAFT: Do Not Cite or Quote
-------
Lifestage
1 The 2008 ISA for Oxides of Nitrogen found evidence of consistent associations between
2 NO2 and respiratory-related hospital admissions for both children and older adults (65+
3 years of age). Recent studies that conducted age stratified analyses add to this body of
4 evidence.
5 In the study of asthma hospital admissions in Hong Kong, Ko et al. (2007b) found
6 evidence of positive associations across each age range (i.e., 0-14, 14-65, and 65+) with
7 the strongest evidence for ages 0-14 (15.5% [95% CI: 10.9, 20.6]) and 65+ (8.9% [95%
8 CI: 5.4, 13.0]) at lag 0-4 days for a 20-ppb increase in 24-h avg NO2 concentrations. Son
9 et al. (2013) also reported children 0-14 years of age to be at increased risk of respiratory-
10 related hospital admissions (i.e., allergic disease, asthma, and respiratory) compared to
11 ages 15-64, 65-74, and > 75 years. However, the authors observed limited evidence for
12 increased risk in older adults (> 75 years), with older adults exhibiting a greater risk
13 compared to the other age ranges >15 years, but only for allergic diseases (Son et al..
14 2013). Samoli et al. (2011) in the study of pediatric asthma hospital admissions in
15 Athens, Greece reported evidence of increased risk of NO2-associated pediatric asthma
16 ED visits in children 0-4 years of age compared to those 5-14 years of age. However, the
17 interpretation of these results requires caution due to the examination of a separate age
18 category of children less than age 5 years in whom asthma diagnoses are less reliable.
19 Wong et al. (2009) examined potential differences by lifestage for respiratory-related
20 hospital admissions in Hong Kong without taking into consideration influenza intensity.
21 In models that examined the baseline effect of lag 0-1 days NO2 exposure on all ages and
22 those 65 years of age and older, the authors found evidence for positive associations
23 across age ranges for all respiratory, acute respiratory disease, and COPD hospital
24 admissions. Only for all respiratory hospital admissions was the association larger in ages
25 65 years and older (4.0 [95% CI: 2.4, 5.7] for a 20-ppb increase in 24-h avg NO2
26 concentrations) compared to all ages (3.2 [95% CI: 1.9, 4.5]).
Sex
27 Of the studies evaluated a limited number examined whether there were differences in the
28 risk of NO 2 -associated respiratory-related hospital admissions by sex. Cakmak et al.
29 (2006). in the 10 Canadian city study, found evidence of increased risk of NO2-associated
30 respiratory hospital admissions for males (2.6% [95% CI: -0.1, 5.3]) compared to females
31 (0.7% [95% CI: -2.1, 3.3]) at lag 1.4 days for a 20-ppb increase in 24-h avg NO2
32 concentrations. Additionally, in a study of pediatric asthma hospital admissions, Samoli
33 etal. (2011) observed evidence of larger effects in males (13.6% [95% CI: 0.7, 28.2]; lag
34 0 for a 30-ppb increase in 1-h max NO2 concentrations) compared to females (-3.4%
35 [95% CI: -12.4, 6.6]; lag 0) for NO2-associated pediatric asthma ED visits.
November 2013 4-159 DRAFT: Do Not Cite or Quote
-------
Race/Ethnicity
I Of the studies evaluated, only Grineski et al. (2010). in a study conducted in Phoenix,
2 Arizona, examined whether race/ethnicity modified the association between NO2
3 exposure and asthma hospital admissions focusing on children <14 years of age. It is
4 important to note that in this study the authors used evening (4-11 pm) NO2
5 concentrations to represent the time of day when children would not be in school and
6 more likely to be outside (Grineski et al.. 2010). In these analyses, differences in
7 race/ethnicity were examined by using a specific race or ethnicity as a referent category.
8 In this analysis, black children were found to be at increased risk of NO2-related asthma
9 ED visits compared to Hispanic children, but no difference was observed between black
10 and white children. However, among children with the same health insurance status
11 (i.e., private insurance), black children were found to be at increased risk of asthma
12 hospital admissions compared to Hispanic and white children.
Pre-existing disease
13 The potential for pre-existing diseases to modify risk of respiratory-related hospital
14 admissions was examined by Faustini et al. (2013) in 6 Italian cities and Wong et al.
15 (2009) in Hong Kong. Faustini et al. (2013) examined whether individuals with COPD
16 were at increased risk of hospital admissions compared to individuals that had both a
17 LRTI and COPD. The authors found evidence of increased risk of hospital admissions for
18 people with LRTI and COPD (6.9% increase [95% CI: -4.3, 19.4] at lag 0-5 days for a
19 20-ppb increase in 24-h avg NO2 concentrations) compared to individuals with only
20 COPD (3.1% increase [95% CI: -2.6, 9.2]; lag 0-5 days for a 20-ppb increase in 24-h avg
21 NO2 concentrations).
22 Wong et al. (2009) examined the potential modification of the relationship between
23 ambient air pollution, including NO2, and respiratory hospital admissions by influenza
24 intensity in Hong Kong. Influenza intensity was defined as a continuous variable using
25 the proportion of weekly specimens positive for influenza A or B instead of defining
26 influenza epidemics. This approach was used to avoid any potential bias associated with
27 the unpredictable seasonality of influenza in Hong Kong where there are traditionally two
28 seasonal peaks, which is in contrast to the single peaking influenza season in the U.S.
29 (Wong et al., 2009). Across respiratory-related hospital admission endpoints, when
30 examining influenza intensity, Wong et al. (2009) only observed increased risk with
31 higher levels of influenza intensity for COPD hospital admissions in the population 65
32 years of age and older with an additional increase in NO2-associated hospital admissions
33 above baseline of 1.6% (95% CI: 0.2, 3.1).
November 2013 4-160 DRAFT: Do Not Cite or Quote
-------
Socioeconomic Status (SES)
I Potential differences in associations betwen NO2and respiratory-related hospital
2 admissions by SES were examined using measures of insurance status, education level,
3 income, and socioeconomic position. In a study examining insurance status, Grineski et
4 al. (2010) reported evidence that children ages <14 years who had no insurance were at
5 greater risk of NO 2 -associated asthma hospital admissions compared to children who had
6 private insurance or Medicaid.
7 Cakmak et al. (2006) conducted a more extensive analysis of SES indicators in 10
8 Canadian cities. Focusing on education level, the authors reported consistent risk of
9 respiratory hospital admissions across individuals with different levels of educational
10 attainment (-2% increase for
-------
1 term NO2 exposures and respiratory-related hospital admissions remained relatively
2 robust in copollutant models (i.e., similar in magnitude or attenuated slightly, but
3 remaining positive). Overall, in the majority of studies NO2 was not found to be highly
4 correlated with other combustion-related pollutants (i.e., r<0.60 for CO and PM2 5).
5 An examination of model specification indicates the NO2-respiratory hospital admissions
6 relationship is senstive to using less than 6 df per year to account for temporal trends, but
7 robust to alternative lags and df for weather covariates (Son et al.. 2013). Studies that
8 examined potential seasonal differences in associations across outcomes provide some
9 evidence of seasonal differences with NO2 effects being greater in the summer in studies
10 conducted in the U.S., Canada, and Europe, but the season with the largest effect was
11 found to vary depending on the study location (i.e., cold season in some Asian cities). A
12 number of studies examined potential effect modifiers of the association between NO2
13 exposure and respiratory-related hospital admissions. These studies continue to provide
14 evidence of larger effects for children and older adults with preliminary evidence that
15 black children, individuals with pre-existing diseases, and individuals of low SES (i.e., no
16 health insurance, low educational attainment, or low income) may be at increased risk of
17 an NO2-related respiratory hospital admission. Additionally, there was inconsistent
18 evidence for differences in the risk of NO2-related respiratory hospital admissions by sex.
November 2013 4-162 DRAFT: Do Not Cite or Quote
-------
Study
FungetaL (2006)
Cakmaketal.(2006)
Sonetal. (2013)a
Dales etal. (2006)
BumettetaL (2001)
Yang etal. (2005)
FaustinietaL (2013)
Wong etal. (2009)
Yang etal. (2005)
Wong etal. (2009)
BumettetaL (1999)
Sonetal. (2013)
ATSDR (2006)
ATSDR (2006)
Ko etal. (2007)
SamolietaL (2011)
Linn etal. (2000)
Sonetal. (2013)
Ko etal. (2007)
Wong etal. (2009)
Yang etal. (2005)
Moolgavkar(2003)
FaustinietaL (2013)
Mehtaetal. (2013)
Segaketal. (2008)
FaustinietaL (2013)
SSouthKor
Bro
x, NY
Manhattan, NY
HongKong
Athens, Greece
LosAngeles, CA
SSouthKoreanciti
Vancouver, CAN
Cook County, CA
LA County, CA
6 Italian cities
HoChiMin
, Vie
Age
All
Lag
All
All
All
28 days-5
28 days-5
Acute Respiratory Dist
Respiratory Infectioi
15 20 25 30 35 40 45
Note: Effect estimates were standardized to a 20-ppb increase in 24-h avg or 30-ppb increase in 1 -h max NO2 concentrations.
All effect estimates presented [except Segala etal. (2008). which focused on winter time air pollution], are for all-year analyses.
Black symbols = U.S. and Canadian studies from the 2008 ISA for Oxides of Nitrogen,
Red symbols = recent studies.
DL = distributed lag.
a = Respiratory diseases for this study was defined as croup, pneumonia, bronchiolitis, respiratory infection inlcuding bronchitis,
chronic obstructive pulmonary disease, asthma, and pneumonitis;
b = dry season results; and
c = lower respiratory tract infections in individuals with COPD.
Figure 4-4 Percent increase in respiratory-related hospital admissions for a
20 ppb increse in 24-h avg or 30-ppb increase in 1-h max NO2
concentrations from U.S. and Canadian studies evaluated in the
2008 ISA for Oxides of Nitrogen and recent studies.
November 2013
4-163
DRAFT: Do Not Cite or Quote
-------
Table 4-21 Corresponding percent increase (95% Cl) for studies presented in
Figure 4-4.
Study
Location
Age
Lag
Avg Time
% Increase
(95% Cl)
Respiratory Diseases
Funq et al. (2006)
Cakmak et al. (2006)
Sonetal. (201 3)b
Dales et al. (2006)
Burnett et al. (2001 f
Yang et al. (2003f
Faustini et al. (2013)
Wonq et al. (2009)
Yang et al. (2003f
Vancouver, CAN
10 Canadian cities
8 South Korean
cities
11 Canadian cities
Toronto, CAN
Vancouver, CAN
6 Italian cities
Hong Kong
Vancouver, CAN
All
All
All
0-27 days
<2
<3
35+
All
65+
65+
0-2
1.4
0
1
0-1
1
0-5
0-1
0-1
1
24-h avg
24-h avg
24-h avg
24-h avg
1-h max
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
9.1 (1.5,
2.3(0.2,
3.6(1.0,
6.5(3.5,
13.3(5.3,
19.1 (7.4,
4.6(0.9,
3.2(1.9,
4.0(2.4,
19.1 (11.2
17.2)
4.5)
6.1)
9.6)
22.0)
36.3)
8.3)
4.5)
5.7)
, 27.5)
Asthma
Burnett et al. (1999f
Sonetal. (2013)
ATSDR (2006)
Ko et al. (2007b)
Samolietal. (2011)
Linn et al. (2000)
Toronto, CAN
8 South Korean
cities
Bronx, NY
Manhattan, NY
Hong Kong
Athens, Greece
Los Angeles, CA
All
All
All
All
All
0-14
<30
0
0
0-4
0-4
0-4
0
0
24-h avg
24-h avg
24-h avg
24-h avg
24-h avg
1-h max
24-h avg
2.6(0.5,
3.6(0.5,
6.0(1.0,
-3.0 (-18.0
10.9(8.1,
28.7 (-3.4,
2.8(0.8,
4.9)
6.8)
10.0)
, 14.0)
13.8)
71.3)
4.9)
Allergic Disease
Sonetal. (2013)
8 South Korean
cities
All
0
24-h avg
3.8(1.0,
6.6)
COPD
Ko et al. (2007a)
Wonq et al. (2009)
Hong Kong
Hong Kong
All
All
65+
0-3
0-1
0-1
24-h avg
24-h avg
24-h avg
10.1 (8.5,
7.1 (5.1,
4.6(2.4,
12.2)
9.1)
6.8)
November 2013
4-164
DRAFT: Do Not Cite or Quote
-------
Table 4-21 (Continued): Corresponding percent increase (95% CI) for studies presented in
Figure 4-4.
Study
Yang et al. (2005)
Moolqavkar (2003)
Faustini et al. (2013)
Location
Vancouver, CAN
Cook County, IL
LA County, CA
6 Italian cities
Age
65+
65+
65+
35+
Lag
1
0
0
0
Avg Time
24-h avg
24-h avg
24-h avg
24-h avg
% Increase
(95% CI)
19.0(4.0, 37.0)
4.9(1.6, 8.2)
3.6(2.8,4.5)
4.6(0.6, 8.6)
Acute Respiratory Disease
Wong et al. (2009)
Hong Kong
All
65+
0-1
0-1
24-h avg
24-h avg
2.1 (-0.1,4.3)
1.7 (-0.6, 4.0)
Respiratory Infection
Mehta et al. (2013)
Seqala et al. (2008)
Faustini et al. (2013)
Ho Chi Minh,
Vietnam
Paris, France
6 Italian cities
28 days-5
yr
28 days-5
yr
<3
35+
1-6
1-6C
0-4
0-5d
24-h avg
24-h avg
24-h avg
24-h avg
-4.0 (-18.0, 12.5)
35.9 (3.0, 79.3)
15.9(7.7,29.0)
6.9 (-4.3, 19.4)
Note: Studies correspond to studies presented in Figure 4-4.
aU.S. and Canadian studies from the 2008 ISA for Oxides of Nitrogen.
""Respiratory diseases for this study was defined as croup, pneumonia, bronchiolitis, respiratory infection inlcuding bronchitis,
chronic obstructive pulmonary disease, asthma, and pneumonitis.
°Dry season results.
dLower respiratory tract infections in individuals with COPD.
1
2
3
4
5
6
7
8
9
10
11
4.2.7.4 Emergency Department Visits
As mentioned previously, ED visit studies are evaluated separately because they often
represent less serious and more common respiratory-related outcomes. Similar to the
hospital admission studies evaluated above, the majority of ED visit studies that
evaluated respiratory-related outcomes have been conducted in individual cities.
However, compared to the hospital admission studies, a larger number of these studies
have been conducted over longer study durations (i.e., >5 years). The following section
evaluates the multi- and single-city studies detailed in Table 4-20 that examined the
association between short-term NO2 concentrations and respiratory-related ED visits.
Several studies of ED visit are not included in this section because they do not fit the
criteria outlined in Section 4.2.7.2. A comprehensive list of these studies can be found in
Supplemental Table S4-1 (U.S. EPA. 2013d)
November 2013
4-165
DRAFT: Do Not Cite or Quote
-------
Respiratory Disease
1 ED visit studies evaluated in the 2008 ISA for Oxides of Nitrogen that focused on all
2 respiratory outcomes were limited in number and focused almost exclusively on study
3 populations consisting of all ages. These studies reported evidence of consistent positive
4 associations between short-term NO2 exposures and all respiratory ED visits. Building on
5 the studies conducted in Atlanta, GA, and evaluated in the 2008 ISA for Oxides of
6 Nitrogen, Peel et al. (2005). Tolbert et al. (2007). and Darrowetal. (2011 a) conducted an
7 analysis to examine whether the association between short-term air pollution exposures
8 and respiratory ED visits differed depending on the exposure metric used (i.e., 1-h max,
9 24-h avg, commuting period [7:00 a.m. to 10:00 a.m.; 4:00 p.m. to 7:00 p.m.], daytime
10 avg [8:00 a.m. to 7:00 p.m.] and night-time avg [12:00 a.m. to 6:00 a.m.]). To examine
11 the association between the various NO2 exposure metrics and respiratory ED visits, the
12 authors conceptually used a time-stratified case-crossover framework in which control
13 days were selected as those days within the same calendar month and maximum
14 temperature as the case day. However, instead of conducting a traditional case-crossover
15 analysis, the authors used a Poisson model with indicator variables for each of the strata
16 (i.e., parameters of the control days). Darrow et al. (201 la) reported relatively consistent
17 results (using an a priori lag of 1 day) across exposure metrics with the largest estimate
18 found for the night-time average and the smallest for the daytime metrics (Figure 4-5).
19 The correlation between NO2 metrics was not as high compared to that for other
20 pollutants examined in the study (i.e., r <0.80 between 1-h max and all other metrics), but
21 was relatively high for the 24-h avg metric, which is the other metric for NO2 often used
22 in epidemiologic studies.
November 2013 4-166 DRAFT: Do Not Cite or Quote
-------
0
"J 1.02 -
•§ "5
ti €>
5 S* i-°1 -
•go
* 1 10° '
i
S 0.99 -
Partial
r -
n - 1 -
J_ JL, T
J- o
1 0,79 0,59 0,55
| | } I
» ** o
^~ |%| Q
---
0.44
ff
TJ
NO,
Note: Partial Spearman correlation coeffiecient between a priori metrics (shaded in gray) and other pollutant metrics shown above
the x-axis.
Source: Reprinted with permission of Nature Publishing Group Darrow et al. (2011 a)
Figure 4-5 Risk ratio and 95% CIs for associations between various lag 1
NO2 metrics and respiratory ED visits.
i
2
3
4
Cause-specific Respiratory Outcomes
Asthma
In the 2008 ISA for Oxides of Nitrogen there was consistent evidence of a positive
association between short-term NO2 exposures and asthma ED visits with some evidence
of larger associations during warmer months. Strickland et al. (2010) examined the
association between NO2 exposure and pediatric asthma ED visits (ages 5-17 years) in
November 2013
4-167
DRAFT: Do Not Cite or Quote
-------
1 Atlanta, GA, using air quality data over the same years as Darrow et al. (201 la) and
2 Tolbert et al. (2007). However, unlike Darrow et al. (2011 a) and Tolbert et al. (2007).
3 which used a single-site centrally located monitor, Strickland et al. (2010) used
4 population-weighting to combine daily pollutant concentrations across monitors. In this
5 study, the authors developed a statistical model using hospital-specific time-series data
6 that is essentially equivalent to a time-stratified case-crossover analysis (i.e., using
7 interaction terms between year, month, and day-of-week to mimic the approach of
8 selecting referent days within the same month and year as the case day). Strickland et al.
9 (2010) observed a 8.6% (95% CI: 4.2, 13.3) increase in ED visits for a 30-ppb increase in
10 1-h max NO2 concentrations at lag 0-2 days in an all-year analysis. The potential
11 confounding effects of other pollutants on the NO 2 -asthma ED visit relationship was only
12 examined in a copollutant model with O3 and correlations between pollutants were not
13 presented. In the copollutant model, NO2 risk estimates were found to be robust to the
14 inclusion of O3 (quantitative results not presented).
15 Additional evidence for an association between short-term increases in NO2
16 concentrations and asthma ED visits comes from studies conducted in Canada
17 (Villeneuve et al.. 2007) and Australia (Jalaludin et al.. 2008). Villeneuve et al. (2007) in
18 a study conducted in Edmonton, Alberta, Canada, in the population aged 2 years and
19 older, reported evidence of positive associations between short-term NO2 concentrations
20 and asthma ED visits for multiple lag structures (lag 1, lag 0-2, and lag 0-4 days). The
21 authors observed the strongest association for lag 0-4 days (4.5% [95% CI: 0, 7.5] for a
22 20-ppb increase in 24-h avg NO2 concentrations). There was no evidence of an
23 association at lag 0. In this study NO2 and CO were strongly correlated (r = 0.74), but in
24 copollutant models with CO, NO2 associations with asthma ED visits remained robust
25 (quantitative results not provided).
26 In a study conducted in Sydney, Australia focusing on children 1-14 years old, Jalaludin
27 et al. (2008) examined air pollution associations with asthma ED visits for single day lags
28 up to 3 days as well as the average of 0-1 day lags. In addition to conducting the analysis
29 focusing on ages 1-14, the authors also examined whether risks varied among age ranges
30 within this study population. In the 1-14 years of age analysis, Jalaludin et al. (2008)
31 observed a similar magnitude of an association for both lag 0 (7.5% [95% CI: 4.5, 10.5])
32 and lag 0-1 days (7.8% [95% CI: 4.5, 11.1]) for a 30-ppb increase in 1-h max NO2
33 concentrations. An examination of the potential confounding effects of other pollutants
34 was assessed in copollutant models with PM10, PM2 5, O3, CO, or SO2. The NO2-asthma
35 ED visit association was found to remain robust in copollutants models with the
36 magnitude of the effect ranging from 4.2-6.1% increase in asthma ED visits.
37 Additionally, NO2 was moderately to weakly correlated with the other pollutants
38 examined.
November 2013 4-168 DRAFT: Do Not Cite or Quote
-------
1 In contrast with the majority of the evidence, short-term increases in NO2 concentrations
2 were not associated with asthma ED visits in a multicity study conducted in 7 Canadian
3 cities (Stieb et al., 2009). Compared to the other asthma ED visit studies evaluated, mean
4 NO2 concentrations across the cities included in this study were the lowest with all cities
5 having mean 24-h avg concentrations <23 ppb. Stieb et al. (2009) examined the
6 association between short-term NO2 exposure and a number of respiratory-related ED
7 visits for all ages. There was no evidence that NO2 was associated with asthma ED visits
8 at single-day lags of 0 to 2 days (0.0% [95% CI: -2.6, 2.7]; lag 2 for a 20-ppb increase in
9 24-h avg NO2 concentrations). Additionally, there was no evidence of associations
10 between respiratory-related ED visits, including asthma, and air pollution averaged over
11 sub-daily time scales (i.e., 3-h avg of ED visits versus 3-h avg pollutant concentrations).
Wheeze
12 Additional evidence for an association between NO2 and respiratory-related ED visits
13 comes from a study focusing on children, (ages 0-2 years) and conducted in 6 Italian
14 cities (Orazzo et al.. 2009). In this study, Orazzo et al. (2009) used data on wheeze
15 extracted from medical records as an indicator of lower respiratory disease. This study
16 examined daily counts of wheeze in relation to air pollution using a time-stratified case-
17 crossover approach in which control days were matched on day of week in the same
18 month and year as the case day. PMi0, SO2, CO, and O3 were also evaluated, but no
19 correlations with NO2 were reported or copollutants analyses conducted. The authors
20 reported positive associations between short-term 24-h avg NO2 exposures and wheeze
21 ED visits when examining various lag averages (0-1 through 0-6 days) with risk
22 estimates ranging from 1.1% (95% CI: -1.2, 3.4) for lag 0-1 days to 2.5% (95% CI: -0.9,
23 6.0) for lag 0-6 days.
COPD
24 To date, the majority of studies that have examined the association between short-term
25 NO2 exposures and COPD have focused on hospital admissions, with very limited
26 evidence for ED visits. In the 7 Canadian cities discussed previously, consistent with the
27 asthma ED visits results, Stieb et al. (2009) did not find evidence of associations between
28 24-h avg NO2 and COPD ED visits at individual lags ranging from 0 (0.1% [95% CI:
29 -6.1, 6.8] for a 20-ppb increase in 24-h avg NO2) to 2 (-5.2% [95% CI: -12.4, 2.7]) days.
30 Additionally, there was no evidence of consistent associations between any pollutant and
31 COPD ED visits at sub-daily time scales (i.e., 3-h avg of ED visits versus 3-h avg
32 pollutant concentrations).
33 Arbex et al. (2009) also examined the association between COPD and several ambient air
34 pollutants, including NO2, in a single-city study conducted in Sao Paulo, Brazil, for
November 2013 4-169 DRAFT: Do Not Cite or Quote
-------
1 individuals over the age of 40 years. Associations between NO2 exposure and COPD ED
2 visits were examined in both single-day lags (0 to 6 days) and a polynomial distributed
3 lag model (0-6 days). However, for NO2, only those results that were statistically
4 significant were presented, i.e., for individuals 65 years of age and older for lag 5 (4.3%
5 [95% CI: 0.5, 8.3] for a 20-ppb increase in 24-h avg NO2 concentrations) and a
6 distributed lag of 0-5 days (9.6% [95% CI: 0.2, 19.9]). The authors did not conduct
7 copollutant analyses, but NO2 was moderately correlated with PM10 (r = 0.60), SO2
8 (r = 0.63), and CO (r= 0.56).
Respiratory Infection
9 Recently some studies have examined the effect of air pollution on ED visits attributed to
10 respiratory-related infections. Stieb et al. (2009). in their study of 7 Canadian cities, also
11 examined the association between short-term NO2 concentrations and respiratory
12 infection ED visits. The authors reported postive associations at lags of 1 and 2 days, but
13 the confidence intervals were wide, providing little evidence of an association. However,
14 Segala et al. (2008). discussed in Section 4.2.7.3. in a study of winter (October-January)
15 air pollution in Paris, France, reported evidence of an association between short-term
16 NO2 concentrations and bronchiolitis ED visits (11.8% [95% CI: 7.7, 20.1]; lag 0-4 for a
17 20-ppb increase in 24-h avg NO2 concentrations). As mentioned previously the
18 interpretation of these results is complicated by the lack of copollutant analyses and the
19 high correlation between pollutants examined (r = 0.74 to 0.83).
20 In an additional study conducted in Edmonton, Alberta, Canada, Zemek et al. (2010)
21 examined a new outcome for NO2, otitis media (i.e., ear infections) ED visits, for ages
22 1-3 years. Associations were examined for single-day lags of 0 to 4 days in all-year as
23 well as seasonal analyses. The authors observed that NO2 was associated with increases
24 in ED visits for otitis media in the all-year analysis at lag 2 (7.9% [95% CI: 1.6, 12.8] for
25 a 20-ppb increase in 24-h avg NO2 concentrations). Interestingly the pattern of
26 associations for CO was similar to that observed for NO2, but the authors did not report
27 correlations between pollutants or conduct copollutant analyses.
Sensitivity Analyses and Effect Modification of NO2-Respiratory-Related ED
Visits Relationship
Exposure Assignment
28 Questions often arise in air pollution epidemiologic studies with regard to the method
29 used to assign exposure. Darrow et al. (201 la) and Strickland et al. (2011) in studies
30 using ED visit data from Atlanta, GA assessed the effect of spatial variability of air
31 pollutants and various exposure assignment approaches, respectively, on the relationship
November 2013 4-170 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
between short-term NO2 exposures and respiratory-related ED visits. Darrow et al.
(2011 a) in their analysis of multiple exposure metrics and respiratory ED visits also
conducted an additional analysis to examine the spatial variability of each exposure
metric (Figure 4-6).
Unlike O3 and PM25, which were found to be spatially homogenous, there was evidence
that correlations for NO 2 metrics decreased dramatically as distance from the central
monitor increased, especially for the 1-h max and night-time metrics (r<0.20) at 60 km.
The 24-h avg metric was also reduced (r=~0.40), but not as dramatically as the 1-h max.
Although reduced at greater distances, moderate correlations (r -0.50) were reported with
the central monitor for the daytime and commute time metrics. Overall, these results
suggest evidence of potential exposure misclassification for NO2 with increasing distance
from the central monitor across exposure metrics.
1.0
c ;
CD
;s O.B -
m
o
o
.1 a6 *
**-*
J9
§ 0.4 i
c
| :
-------
1 Using data from the warm season from a previous analysis, Strickland et al. (2010)
2 and Strickland et al. (2011) examined the relative influence of different exposure
3 assignment approaches (i.e., central monitor, un-weighted average across available
4 monitors, and population-weighted average) on the magnitude and direction of
5 associations betwen NO2 and pediatric asthma hospital admission. Strickland et al.
6 (2011) reported that effect estimates per interquartile range (IQR) increase in NO2 were
7 similar across the metrics; however, based on a standardized increment, the magnitude of
8 the association between NO2 and pediatric asthma ED visits varied (central monitor
9 [7.9% (95% CI: 4.2, 11.8)] 181 ppb, relative to the first
21 quintile (i.e., NO2 concentrations <28 ppb). Additionally, the LOESS C-R relationship
22 analysis provides evidence indicating elevated NO2 associations along the distribution of
23 concentrations from the 5th to 95th pecentile (Figure 4-7). Collectively, these analyses do
24 not provide evidence of a threshold.
November 2013 4-172 DRAFT: Do Not Cite or Quote
-------
Nitrogen Dioxide Warm Season
in
ro
o:
Q)
+••
ro in
15 20 25 30 35
Concentration (ppb)
Source: Reprinted with permission of the American Thoracic Society, (Strickland et al.. 2010).
Figure 4-7 LOESS Concentration-Response estimates (solid line) and twice-
standard error estimates (dashed lines) from generalized additive
models for associations between 3-day avg (lag 0-2) NO2
concentrations and ED visits for pediatric asthma at the 5th to
95th percentile of NO2 concentrations in the Atlanta, GA area.
i
2
o
3
4
5
6
7
8
9
10
11
12
In a study conducted in Detroit, MI, Li et al. (20lib) examined the C-Rrelationship by
examining if there is evidence of a deviation from linearity. Associations were examined
in both a time-series and time-stratified case-crossover study design assuming: (1) no
deviation from linearity and (2) a change in linearity at 23 ppb (i.e., the maximum
likelihood estimate within the 10th to 95th percentile concentration where a change in
linearity may occur [~80th percentile]). It is important to note that the analysis that
assumed a deviation in linearity did not assume zero risk below the inflection point. The
focus of the analysis was on identifying whether risk increased above that observed in the
linear models at NO2 concentrations above 23 ppb. In the analyses assuming linearity,
effect estimates varied across models for a 0-4 day lag (time series: 2.9% [95% CI: -7.9,
15.1]; case-crossover: 9.1% [95%CI: -0.83, 20.2] fora20-ppb increase in24-havgNO2
concentrations). In the models that assumed a deviation from linearity the authors did not
November 2013
4-173
DRAFT: Do Not Cite or Quote
-------
1 observe evidence of higher risk in either the time-series or case-crossover analyses at
2 NO 2 concentrations greater than 23 ppb.
Seasonal Differences
3 A number of the respiratory-related ED visit studies also conducted seasonal analyses.
4 These studies found seasonal patterns similar to those observed in the hospital admission
5 studies indicating larger effects during the warm or summer months.
6 In the study of 6 Italian cities, Orazzo et al. (2009) reported slightly larger respiratory
7 disease ED visits associations in the summer compared to the winter, but the confidence
8 intervals were wide and overlapping (quantitative results not provided). Evidence of
9 larger effects during warm or summer months was also found in studies of asthma ED
10 visits. Villeneuve et al. (2007) reported associations to be generally stronger in the warm
11 season (e.g., 21.4% [95% CI: 13.6, 31.0] at lag 0-4 days for a 20-ppb increase in 24-h avg
12 NO2 concentrations) than in the cold season in Edmonton, Canada. Additionally,
13 Jalaludin et al. (2008) in a study conducted in Sydney, Australia found evidence of
14 greater effects during the warm months (November-April) compared to the cold months
15 (May-October). These results are consistent with Strickland et al. (2010). which reported
16 stronger associations during the warm season (i.e., May-October) (16.0% [95% CI: 9.1,
17 23.5]; lag 0-2 days) than the cold season (3.8% [95% CI: -1.9, 9.6]; lag 0-2 days) in a
18 study of pediatric asthma ED visits in Atlanta, GA. Additionally, Zemeketal. (2010) in a
19 study of otitis media ED visits in Alberta, Canada reported that the magnitude of the
20 association was larger in the warm months (April-September), 16.1%(95%CI: 3.1,
21 31.2), compared to the cold months, (October-March), 4.7% (95% CI: 0, 11.2) at lag 2 for
22 a 20-ppb increase in 24-h avg NO2 concentrations.
23 In the study of 7 Canadian cities, Stieb et al. (2009) also conducted seasonal analyses, but
24 did not present detailed results. However, the authors did state that no consistent
25 associations were observed between any pollutants and the respiratory outcomes
26 examined during the winter months (October-March).
Lifestage
27 A study that examined respiratory-related ED visits continues to provide evidence that
28 children and older adults are at increased risk of NO2-induced ED visits. Villeneuve et al.
29 (2007) conducted an extensive analysis of the risk of NO 2-associated asthma ED visits
30 across age ranges (i.e., 2-4, 5-14, 15-44, 45-64, 65-74, and 75+ years). In the warm
31 season (April-September), where the greatest effects were estimated, across age ranges
32 the risk of asthma ED visits was greatest for the ages 2-4 years and 75+ years, with
33 elevated risks also observed for 5-14 years of age. There was no evidence of increased
34 risk of NO 2 -associated asthma ED visits in the population 45-64 years of age. Age-
November 2013 4-174 DRAFT: Do Not Cite or Quote
-------
1 specific risk estimates were also examined in copollutant models with CO and overall
2 were relatively robust, except for the age range of 15-44 years (Figure 4-8).
N02
NO,
"3 1.60
CO
CO
o CO
g 1.40
u •
•o
? CO NO?
I 1.20 . •
1.00 . . * . , •
•o
O 0.80
°-60 '2-4 5-14 15-44 45-64 65-74 75+
Agc (in years)
0.40
Note: NO2 odds ratios represent copollutant models with CO, while CO odds ratios represent copollutant models with NO2.
Source: Reprinted with permission of Biomed Central Ltd (Villeneuve et al.. 2007).
Figure 4-8 Age-specific NO2 asthma ED visit effect estimates from
copollutant models with CO in Edmonton, Canada.
3 Arbex et al. (2009) in a study in Sao Paulo, Brazil, examined respiratory ED visits in ages
4 40-64 and > 65 years. Of the outcomes examined, NO2 results were reported only for
5 COPD and the population 65 years of age and older because larger effects were estimated
6 for this age range compared to ages 40-64 years and ages >40 years.
Sex
1 Of the respiratory-related ED visit studies evaluated, only Zemek et al. (2010) examined
8 whether there were differences in risk by sex. Focusing on the lag that showed the
9 strongest association in the combined analysis (lag 2), Zemek et al. (2010) reported
November 2013 4-175 DRAFT: Do Not Cite or Quote
-------
1 evidence of greater risk for NO2-associated otitis media ED visits in females (9.5% [95%
2 CI: 1.6, 17.1] for a 20-ppb increase in 24-h avg NO2 concentrations) compared to males
3 (4.7% [95% CI: -1.6, 12.80]). There especially was a dramatic difference in the warm
4 months (32.9% for females compared to 6.3% for males).
Summary of ED Visit Studies
5 Recent respiratory-related ED visit studies build on the body of evidence from the 2008
6 ISA for Oxides of Nitrogen (Figure 4-9 and Table 4-22). These studies generally provide
7 evidence of consistent positive associations, particularly for respiratory disease and
8 asthma ED visits, with associations primarily observed at lags ranging from 0-2 days. In
9 these studies copollutant confounding was not consistently examined. Of the studies that
10 conducted copollutant modeling, NO2 associations were consistently found to remain
11 robust (i.e., similar in magnitude or attenuated slightly, but remaining positive). Overall,
12 in the majority of studies NO2 was not found to be highly correlated with other
13 combustion-related pollutants, i.e., r <0.60 for CO and PM2 5.
14 In studies that examined various exposure-related issues and respiratory-related ED
15 visits, it was identified that the 1-h max and 24-h avg exposure metrics for NO2 result in
16 similar associations (Darrow et al.. 201 la). It was also found that the exposure
17 assignment approach used can influence the magnitude, but not direction, of the NO2-
18 asthma ED visit risk estimate (Strickland et al.. 2011). Additionally, it was demonstrated
19 that the correlation between the NO2 concentration for various exposure metrics and the
20 NO2 concentration at the monitor drops off as distance from monitor increases, which
21 could lead to exposure misclassification (Darrow et al.. 201 la). Although informative, it
22 should be noted that all of these studies were conducted in one location (i.e., Atlanta,
23 GA) using a similar dataset.
24 An examination of the C-R relationship for NO2-respiratory-related ED visits provides
25 evidence of a linear, no-threshold relationship. An examination of seasonal analyses
26 indicates that associations between NO2 and respiratory ED visits are greater in the warm
27 or summer months, specifically in locations where hospital admissions studies provided
28 similar evidence. Although limited in number, studies that examined potential effect
29 modifiers of the NO2-respiratory-related ED visits relationship continue to support larger
30 effects in children and older adults.
November 2013 4-176 DRAFT: Do Not Cite or Quote
-------
Study
Peeletal. (2005)
Tolbertetal. (2007)
Da rrow etal. (2009)
Tolbertetal. (2000)
Peeletal. (2005)
Itoetal. (2007)
Stieb et al (2009)
Jalaludinetal. (2008)
Peeletal. (2005)
Lietal. (2011)
Location
Atlanta,GA
Atlanta,GA
Atlanta,GA
Atlanta,GA
Detroit, MI
Vnieneuve etal. (2007) Edmonton,Canada
Stricklandetal. (2010) Atlanta.GA
Jaffe etal. (2003)
Orazzoetal. (2009)
Peeletal. (2005)
Stieb et al. (2009)
Arbex et al. (2009)
Segala et al. (2008)
Stieb et al. (2009)
Zemeketal. (2010)
Edmonton, Canad
Age
ATI
An
An
Atlanta.GA An
Atlanta.GA An
New York, NY An
7 Canadian cities AU
Sydney, Australia 1-14
2-18
2-18
5-17
2 Ohio cities 5-34
6 Itattan cities 0-2
Atlanta.GA An
7 Canadian cities AU
Sao Paulo, Brazn 65+
Paris, France < 3
7 Canadian cities AU
1-3
0-1
0
0
0-2
0-4"
0-4b
0-4
0-1
0-6
0-4
2
Respiratory Diseases
Asthma
Note: Black = U.S. and Canadian studies from the 2008 ISA for Oxides of Nitrogen, Red = recent studies. Circles = all-year,
Diamonds = summer/warm, and Squares = winter/cold, a = time-series analysis results; and b = case-crossover analysis results.
Figure 4-9 Percent increase in respiratory-related ED visits for a 20-ppb
increase in 24-h avg or 30-ppb increase in 1-h max NO2
concentrations from U.S. and Canadian studies evaluated in the
2008 ISA for Oxides of Nitrogen and recent studies in all-year and
seasonal analyses.
Table 4-22 Corresponding percent increase (95% Cl) for studies presented in
Figure 4-9.
Study
Location
Age
Avg Time
Season
Lag
% Increase
(95% Cl)
Respiratory Diseases
Peel et al.
Tolbert et
D arrow et
(2005)a
al. (2007f
al. (201 1a)
Atlanta,
Atlanta,
Atlanta,
GA
GA
GA
All
All
All
1-h max
1-h max
1-h max
All
All
All
0-2
0-2
1
0-6
2
2
1
2.
.4
.0
.4
5
(0.9,
(0.5,
(0.8,
(-0.9,
4.1)
3.3)
2.1)
6.0)
Asthma
Tolbert et
Peel et al.
al. (2000f
(2005)a
Ito et al. (2007f
Atlanta,
Atlanta,
GA
GA
New York, NY
All
All
All
1-h max
1-h max
24-h avg
All
All
All
1
0-2
0-1
0.
2.
12
7
1
.0
(-0.8,
(-0.4,
(7.0,
2.3)
4.5)
15.0)
November 2013
4-177
DRAFT: Do Not Cite or Quote
-------
Table 4-22 (Continued): Corresponding percent increase (95% CI) for studies presented in
Figure 4-9.
Study
Stieb et al. (2009)
Jalaludin et al. (2008)
Jalaludin et al. (2008)
Peel et al. (2005)a
Lietal. (2011 b)
Villeneuve et al. (2007)
Strickland et al. (2010)
Jaffe et al. (2003)a
Location
7 Canadian
cities
Sydney,
Australia
Sydney,
Australia
Atlanta, GA
Detroit, Ml
Edmonton, CAN
Atlanta, GA
2 Ohio cities
Age
All
1-14
1-14
2-18
2-18
2+
5-17
5-34
Avg Time
24-h avg
1-h max
1-h max
1-h max
24-h avg
24-h avg
1-h max
24-h avg
Season
All
All
Warm
Cold
All
All
All
All
Warm
All
Warm
Cold
Summer
Lag
2
0-1
0
0
0-2
0-4b
0-4C
0-4
0-2
1
% Increase
(95% CI)
0.0 (-2.6, 2.7)
7.8(4.5, 11.1)
8.4(4.2, 12.5)
4.4 (-1.7, 10.4)
4.1 (0.8, 7.6)
2.9 (-7.9, 15.1)
9.1 (-0.8, 20.2)
4.5(0.0, 7.5)
21.4(13.6, 31.0)
8.6(4.2, 13.3)
16.0(9.1,23.5)
3.8 (-1.9, 9.6)
6.1 (-2.0, 14.0)
Wheeze
Orazzo et al. (2009)
6 Italian cities
0-2
24-h avg
All
0-1
1.1 (-1.2, 3.4)
COPD
Peel et al. (2005)a
Stieb et al. (2009)
Arbex et al. (2009)
Atlanta, GA
7 Canadian
cities
Sao Paulo,
Brazil
All
All
65+
1-h max
24-h avg
24-h avg
All
All
All
0-2
0
0-5 DL
5.3(0.9, 9.9)
0.1 (-6.1,6.8)
9.6(0.2, 19.9)
Respiratory Infection
Seqala et al. (2008)
Stieb et al. (2009)
Paris, France
7 Canadian
cities
<3
All
24-h avg
24-h avg
Winter
All
0-4
2
11.8(7.7,20.1)
0.7 (-1.5, 2.8)
Otitis media
Zemeketal. (2010)
Note:Studies correspond to
Edmonton, CAN
1-3
24-h avg
All
Warm
Cold
2
7.9(1.6, 12.8)
16.1 (3.1, 31.2)
4.7(0.0, 11.2)
studies presented in Figure 4-9.
"Studies evaluated in the 2008 ISA for Oxides of Nitrogen.
bTime-series analysis results.
""Case-crossover analysis results.
November 2013
4-178
DRAFT: Do Not Cite or Quote
-------
4.2.7.5 Outpatient and Physician Visit Studies
1 Several recent studies examined the association between ambient NO2 concentrations and
2 physician or outpatient (non-hospital, non-ED) visits for acute conditions in various
3 geographic locations. Burra et al. (2009) examined asthma physician visits among
4 patients aged 1-17 and 18-64 years in Toronto, Canada in a study focusing on differences
5 by sex and income within age categories. The authors reported evidence of consistently
6 positive associations between short-term increases in NO2 concentrations and asthma
7 physician visits across the single- and multi-day lags examined (i.e., 0, 0-1, 0-2, 0-3, and
8 0-4 days). The magnitude of the effect estimates was found to be similar between sexes,
9 income quintiles, both within and between ages. In a study conducted in Atlanta, GA,
10 Sinclair et al. (2010) examined the association of acute asthma and respiratory infection
11 (e-g-, upper respiratory infections, lower respiratory infections) outpatient visits from a
12 managed care organization. The authors separated the analysis into two time periods (the
13 first 25 months of the study period and the second 28 months of the study period), in
14 order to compare the air pollutant concentrations and relationships between air pollutants
15 and acute respiratory visits for the 25-month time-period examined in Sinclair and
16 Tolsma (2004) (i.e., August 1998-August 2000), and an additional 28-month time-period
17 of available data from the Atlanta Aerosol Research Inhalation Epidemiology Study
18 (AIRES) (i.e., September 2000-December 2002). A comparison of the two time periods
19 indicated that risk estimates across outcomes tended to be larger in the earlier 25-month
20 period compared to the later 28-month period, with evidence of consistently positive
21 associations at lags of 0-2 and 3-5 days for asthma (both child and adult), as well as upper
22 and lower respiratory infections. However, the confidence intervals for outcomes with
23 smaller counts (i.e., approximately 12 per day for adult asthma and LRI, and 18 per day
24 for child asthma compared to 263 per day for URI) were relatively large. An examination
25 of potential seasonal differences in the association between air pollution exposures and
26 child asthma visits produced evidence of larger risk estimates in the warm season at all
27 lags, only in the 25-month period (e.g., warm: 9.6% [95% CI: -7.4, 30.0]; cold: 1.2%
28 [95% CI: -12.4, 16.8] at lag 0-2 days for a 30-ppb increase in 1-h max NO2
29 concentrations), with less consistent evidence for seasonal differences in the 28-month
30 period.
31 (Villeneuve et al.. 2006b) examined the effect of short-term NO2 exposure on allergic
32 rhinitis physician visits among individuals aged 65 or older in Toronto, Canada.
33 The authors reported a strong association between allergic rhinitis physician visits and
34 ambient NO2 concentrations at lag 0 in all-year and cold season (November-April)
35 analyses (results not presented quantitatively), but not any other lag day. A similar
November 2013 4-179 DRAFT: Do Not Cite or Quote
-------
1 pattern of associations was observed with SO2, but the authors did not examine
2 correlations between pollutants or conduct copollutant analyses. Overall, the
3 interpretation of these results is complicated because of the lack of a consistent
4 association across the lags examined.
4.2.8 Respiratory Mortality
5 Studies evaluated in the 2008 ISA for Oxides of Nitrogen that examined the association
6 between short-term NO2 exposure and cause-specific mortality found consistent positive
7 associations with respiratory mortality with some evidence indicating that the magnitude
8 of the effect was larger compared to total and cardiovascular mortality. Recent multi-city
9 studies conducted in Asia (Wong et al.. 2008b). China (Chenetal.. 2012b). and Italy
10 (Chiusolo et al.. 2011) add to the initial body of evidence indicating larger respiratory
11 mortality effects (Section 4.4. Figure 4-17). However, an additional multi-city study
12 conducted in Italy (Bellini et al., 2007). which is an extension of Biggeri et al. (2005).
13 observed relatively consistent risk estimates across mortality outcomes, inconsistent with
14 the results of the original analysis and complicating interpretation of whether there is
15 differential risk among mortality outcomes.
16 The initial observation of potentially larger respiratory mortality risk estimates was
17 further examined in a few studies by examining the potential confounding effects of
18 copollutants. Chenetal. (2012b) in the 17 Chinese cities study (CAPES) found that NO2
19 risk estimates for respiratory mortality were slightly attenuated, but remained positive in
20 copollutant models with PM10 and SO2 (10.1% [95% CI: 5.7, 14.5]; with PM10: 6.9%
21 [95% CI: 3.0, 11.0]; with SO2: 7.2% [95% CI: 3.2, 11.3]; for a 20-ppb increase in
22 24-h avg NO2 concentrations at lag 0-1 days). Chiusolo etal. (2011) also found evidence
23 that associations between short-term NO2 exposure and respiratory mortality remained
24 robust in copollutant models in a study of 11 Italian cities. In both an all year analysis of
25 NO2 with PM10 (NO2: 14.1% [95% CI: 2.9, 26.4]; NO2 + PM10: 13.7% [95% CI: 3.0,
26 25.5]; for a 20-ppb increase in NO2 concentrations at lag 1-5 days), and a warm season
27 (April-September) analysis of NO2 with O3 (NO2: 42.4% [95% CI: 16.6, 73.9];
28 NO2 + O3: 44.6% [95% CI: 15.0, 81.9]; for a 20-ppb increase in NO2 concentrations at
29 lag 1-5 days) NO2 associations with respiratory mortality were relatively unchanged.
30 Overall, the limited number of studies that have examined the potential confounding
31 effects of copollutants on the NO2-respiratory mortality relationship indicate that
32 associations remain robust.
33 Of the studies evaluated, only the studies conducted in Italy examined potential seasonal
34 differences in the NO2-respiratory mortality relationship (Chiusolo et al.. 2011; Bellini et
November 2013 4-180 DRAFT: Do Not Cite or Quote
-------
1 al.. 2007). In a study of 15 Italian cities, Bellini et al. (2007) found that risk estimates for
2 respiratory mortality were dramatically increased in the summer from 1.5% to 9.4% for a
3 20-ppb increase in 24-h avg NO2 concentrations at lag 0-1 days, respectively, with no
4 evidence of an association in the winter. These results were further confirmed in a study
5 of 10 Italian cities (Chiusolo et al.. 2011). which also observed an increase in risk
6 estimates for respiratory mortality in the warm season (i.e., April - September) compared
7 to all-year analyses. Chiusolo et al. (2011) did not conduct winter season analyses.
8 Although the respiratory mortality results are consistent with those observed in the total
9 mortality analyses conducted by Bellini et al. (2007) and Chiusolo et al. (2011). as
10 discussed in Section 4.4. studies conducted in Asian cities observed much different
11 seasonal patterns and it remains unclear if the seasonal patterns observed for total
12 mortality would be similar to those observed for respiratory mortality in these cities.
4.2.9 Summary and Causal Determination
13 Evidence indicates that a causal relationship exists between short-term NO2 exposure and
14 respiratory effects based primarily on the coherence among multiple lines of evidence
15 that indicate increases in asthma morbidity. There also is some support for NO2-related
16 effects on impaired host defense, COPD, and respiratory mortality, but coherence among
17 various lines of evidence is limited. The determination of a causal relationship represents
18 a change from the "sufficient to infer a likely causal relationship" determined in the 2008
19 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
20 Consistent with previous findings, recent epidemiologic results continue to demonstrate
21 associations between increases in ambient NO2 concentrations and increases in hospital
22 admissions and ED visits for asthma and respiratory symptoms and pulmonary
23 inflammation in children with asthma. As in the 2008 ISA for Oxides of Nitrogen,
24 biological plausibility is demonstrated by NO2-induced AHR, pulmonary inflammation,
25 and impaired host defenses in controlled human exposure and animal toxicological
26 studies. However, recent studies address the key uncertainty identified in the 2008 ISA
27 for Oxides of Nitrogen regarding the potential for NO2 to serve as an indicator for
28 another combustion-related pollutant or mixture. Recent results from copollutant models
29 additionally demonstrate ambient NO2-associated increases in asthma and respiratory
30 effects in diverse geographic locations with adjustment for copollutants such as PM, PM
31 components, O3, SO2, and CO. The evidence for a causal relationship between short-term
32 exposure to NO2 and respiratory effects is detailed below using the framework described
33 in Table II of the Preamble to this ISA. The key evidence as it relates to the causal
34 framework is presented in Table 4-23.
November 2013 4-181 DRAFT: Do Not Cite or Quote
-------
1 A causal relationship between short-term NO2 exposure and respiratory effects is
2 strongly supported by the coherence of evidence across related outcomes and disciplines
3 indicating increases in asthma morbidity. Epidemiologic studies consistently demonstrate
4 associations between increases in ambient NO2 concentration and increases in asthma
5 hospital admissions and ED visits (Sonetal.. 2013; Li etal.. 20lib; Strickland et al.,
6 2010; Ito et al.. 2007; Ko et al.. 2007b; Villeneuve et al. 2007) among subjects of all
7 ages and in children. Risk estimates ranged from a 2.1% to 12% increase per 20-ppb
8 increase in 24-h avg NO2 or 30-ppb increase in 1-h max NO2. These observations are
9 supported by evidence in children and adults with asthma for increases in respiratory
10 symptoms (Mann etal., 2010; Schildcrout et al., 2006; Gent et al., 2003; Mortimer et al..
11 2002). the major reason for seeking medical treatment. Recent epidemiologic evidence
12 substantiates the robustness of NO2-associated asthma morbidity with additional results
13 from studies conducted in diverse locations in the U.S., Canada, and Asia, including
14 several multicity studies. Individual epidemiologic studies examined multiple outcomes
15 and lags of exposure; however, the pattern of association observed with NO2 does not
16 indicate that a higher probability of findings due to chance explains the evidence. The
17 epidemiologic findings specifically for respiratory symptoms are only weakly supported
18 by findings from controlled human exposure studies, as NO2-induced (120-4,000 ppb)
19 increases in respiratory symptoms were found in some but not all studies of adults with
20 asthma and one study of adolescents with asthma (Section 4.2.6.2).
21 Key biological plausibility for NO 2-associated asthma morbidity is provided by findings
22 of NO2-induced increases in AHR in controlled human exposure studies of adults with
23 asthma. AHR can be a key contributor to increases in respiratory symptoms such as
24 wheeze. Controlled human exposure studies demonstrated increases in AHR in adults
25 with asthma at rest with the lowest NO2 exposures examined in experimental studies, i.e.,
26 200 to 300 ppb NO2 for 30 minutes and 100 ppb for 1 hour (Section 4.2.2.2). Exposures
27 in this range and up to 4,000 ppb were not found consistently to have a direct effect on
28 changes in lung function in controlled human exposure of adults with asthma, and
29 epidemiologic evidence in adults also is inconsistent. However, epidemiologic studies of
30 children with asthma found NO2-associated decrements in lung function measured under
31 supervised conditions. Several of these study populations had high prevalence of atopy
32 (e.g., 53-90%). Allergen-induced airway obstruction is a mechanism that can lead to lung
33 function decrements and respiratory symptoms. Thus, the epidemiologic evidence for
34 NO2-associated lung function decrements in children with asthma supports the evidence
35 for increases in respiratory symptoms in children with asthma.
36 A causal relationship between short-term NO2 exposure and respiratory effects also is
37 supported by evidence characterizing potential mechanisms for NO2-induced AHR and
38 respiratory symptoms. Whereas changes in lung permeability were not consistently found
November 2013 4-182 DRAFT: Do Not Cite or Quote
-------
1 in experimental animals with ambient-relevant NO2 exposures (Section 4.2.4.2).
2 controlled human exposure studies and animal toxicological studies indicate the effects of
3 NO2 on pulmonary oxidative stress (Sections 4.2.4.1 and 4.2.4.2) and modification of
4 adaptive and innate immunity (Sections 3.3.2.6 and 4.2.4.3). NO2 exposures of 260-400
5 ppb enhanced allergic inflammation in humans with allergic asthma or animal models of
6 allergic disease, as characterized by increases in Th2 cytokines, IgE, eosinophil
7 activation, myeloperoxidase levels, and PMNs (Section 4.2.4.3). As allergic
8 inflammation promotes bronchoconstriction and airway obstruction, the evidence
9 describes key events to inform the mode of action for NO2-associated increases in
10 respiratory symptoms found in populations of children with asthma with atopy
11 prevalence ranging from 53 to 100%. NO 2 -related pulmonary inflammation also was
12 demonstrated as increases in PMNs with 3- to 6-hour exposures to 1,500-3,500 ppb NO2
13 in healthy adults (Section 4.2.4.1) and increases in eicosanoids (Section 4.2.4.1). which
14 are involved in PMN recruitment. Several epidemiologic studies demonstrated
15 associations between increases in ambient NO2 concentrations and increases in
16 pulmonary inflammation in individuals with asthma (Section 4.2.4.4). Recent
17 epidemiologic studies in children added to the robustness of evidence by demonstrating
18 associations with measures of NO2 that account for spatial heterogeneity in ambient
19 concentrations, including measures of personal and school outdoor exposures. Further,
20 the recruitment of children from schools supports the likelihood that study populations
21 were representative of the general population of children with asthma.
22 Additional evidence indicates that the respiratory effects of short-term NO2 exposure
23 extend beyond those specifically related to asthma morbidity. Epidemiologic studies
24 demonstrate associations between ambient NO2 concentrations and hospital admissions
25 and ED visits for all respiratory causes combined (Faustini etal.. 2013; Darrow et al.
26 201 la: Cakmak et al.. 2006; Dales et al.. 2006). Ambient NO2-associated increases in
27 pulmonary inflammation and respiratory symptoms (Sections 4.2.4.4 and 4.2.6.1) also
28 were found in children in the general population. Ambient NO2 concentrations were
29 associated with respiratory infections, particularly in children; however, some effects
30 were estimated with imprecision (Figure 4-4 and Figure 4-5. and Sections 4.2.5.1 and
31 4.2.7.3). There is clear evidence for impaired host defense demonstrated as mortality
32 from bacterial or viral infection in animal models exposed to 1,500-5,000 ppb NO2
33 (Section 4.2.5.1). Evidence also describes key events to inform the mode of action for
34 impaired host defense, including NO2-induced increases in pulmonary inflammation
35 found in controlled human exposure and animal toxicological studies and diminished
36 alveolar macrophage function found in some but not all studies (Section 4.2.5.3). Effects
37 on pulmonary clearance were more variable (Section 4.2.5.2). Evidence supports
38 associations of ambient NO2 exposure with COPD exacerbations, primarily as increases
39 in hospital admissions and ED visits for COPD (Sections 4.2.6.2 and 4.2.6.1). However,
November 2013 4-183 DRAFT: Do Not Cite or Quote
-------
1 in adults with COPD, NO2 was not consistently related with increases in respiratory
2 symptoms lung function decrements in epidemiologic or controlled human exposure
3 studies. Epidemiologic studies also demonstrated NO2-associated increases in respiratory
4 mortality (Section 4.2.8). The spectrum of respiratory effects that can explain NO2-
5 related increases in respiratory mortality is not entirely clear. There is consistent evidence
6 for the effects of NO2 exposure on asthma, but limited coherence among outcomes and
7 disciplines for effects on COPD and respiratory infections. Among the leading causes of
8 mortality, COPD and respiratory infections are the ones related to respiratory causes
9 (HovertandXu.2012).
10 Previous uncertainty regarding the independent effects of NO2 exposure on respiratory
11 outcomes is reduced by recent epidemiologic observations that NO2 remains associated
12 with respiratory effects with statistical adjustment for temperature, relative humidity,
13 season, long-term time trends, and particularly copollutant concentrations. In a few
14 studies, NO2 associations were largely explained by copollutant exposure; however,
15 among the studies that conducted copollutant modeling, most found that NO2
16 associations persisted with adjustment for copollutants such as PMi0, PM25, PMi0-2.5
17 (Figure 4-10). BC, EC, UFP, PNC (Figure 4-11). SO2, O3, and CO (Figure S4-1U.S.
18 EPA. 2013d). Many NO2 associations remained robust to copollutant adjustment,
19 although confounding by any particular copollutant was examined to a limited extent and
20 not all potentially correlated pollutants were examined. Further, the interpretation of
21 copollutant models can be limited and methods to adjust for multiple copollutants
22 simultaneously are not reliable. Thus, the potential for residual confounding is recognized
23 (Section 1.5). In some cases, the loss of precision in NO2 effect estimates was magnified
24 because the increment used to standardize effect estimates is much larger than the
25 variability in NO2 concentrations reported in the study (Martins etal.. 2012; Strak et al..
26 2012; Mann etal.. 2010; Schwartz et al.. 1994). There also was some evidence that NO2
27 exposure confounded respiratory effects associated with copollutants. Among
28 copollutants, NO2 typically is more highly correlated with CO, BC, and UFP (Figure
29 2-20). Time series provided evidence for NO2 associations independent of CO, and panel
30 studies indicated associations independent of BC and UFP. There was previous evidence
31 of NO2-associated respiratory effects with adjustment for copollutants, but recent
32 findings strengthen the evidence with additional studies of respiratory hospital
33 admissions and ED visits conducted in diverse locations and additional panel studies with
34 strong exposure assessment by modeling individual subjects' outdoor exposures or
35 monitoring exposures during outdoor activity (Martins et al.. 2012; Strak etal.. 2012).
36 Other lines of evidence indicate that NO2-associated respiratory effects are independent
37 of copollutants. Larger ambient NO2-associated increases in respiratory hospital
38 admissions and ED visits were found in the warm season, during which the potential for
November 2013 4-184 DRAFT: Do Not Cite or Quote
-------
1 confounding by PM2 5 may be lower because of lower NO2-PM2 5 correlations in the
2 warm than cold season (Section 2.6.4.1). Correlations between NO2 and O3 remain low
3 in the warm season. Recent studies continue to support associations between short-term
4 increases in indoor NO2 exposures and increases in respiratory effects in children with
5 asthma (Sections 4.2.4.5 and 4.2.6.3). In schools in Ciudad Juarez, Mexico, correlations
6 between NO2 and copollutants such as BC, PM, and SO2 differed between the indoor and
7 outdoor environment, suggesting that NO2 was part of a different pollutant mixture
8 indoors and outdoors. Thus, the coherence of evidence for respiratory effects related to
9 indoor and outdoor NO2 exposure supports the independent effects of NO2 exposure.
10 Most epidemiologic studies assessed NO2 exposures from central monitoring stations;
11 however, personal, outdoor school, and near-road exposures also were associated with
12 respiratory effects in children and adults with asthma. Among studies that compared
13 exposure assessment methods, results did not consistently estimate stronger respiratory
14 effects for personal or outdoor school NO2 concentrations than indoor or central site NO2
15 (Greenwaldetal. 2013; Sarnatet al.. 2012; Delfino et al.. 2008a: 2006). Most evidence
16 for respiratory effects was related to multiday lags of NO2 of 2 to 5 days, but associations
17 also were found with single-day lags of 0 or 1 day. Comparisons among lags did not
18 clearly indicate a stronger association for a particular lag. Respiratory hospital admissions
19 and ED visits were associated with 24-h avg and 1-h max NO2 whereas pulmonary
20 inflammation and respiratory symptoms were associated primarily with 24-h avg NO2. In
21 the few studies that compared averaging times, no clear difference was found in the
22 magnitude of association for respiratory hospital admissions or ED visits. The
23 concentration-response relationship was analyzed for pediatric asthma ED visits in
24 Atlanta, GA and Detroit, MI, and neither a threshold nor deviation from linearity was
25 found in the distribution of 24-h avg or 1-h max ambient NO2 concentrations examined
26 (Lietal..2011b: Strickland etal.. 2010).
27 In conclusion, multiple lines of evidence support a relationship between short-term NO2
28 exposure and asthma morbidity, particularly epidemiologic evidence for increases in
29 asthma hospital admissions and ED visits and respiratory symptoms and pulmonary
30 inflammation in children with asthma. Findings also point to NO2-related effects on
31 impaired host defense, COPD, and respiratory mortality, but there is limited coherence
32 among various lines of evidence. The evidence for asthma is substantiated by additional
33 findings from recent multicity studies conducted in diverse geographic locations and
34 recent panel studies with strong exposure assessment characterized by monitoring of
35 personal, outdoor school, or outdoor near-road exposures. Associations are found
36 primarily for NO2, both 24-h avg and 1-h max concentrations. Respiratory effects were
37 found in association with multiday averages of ambient NO2 of 2 to 5 days and also
38 single-day lags of 0 or 1 day. Across studies finding NO2-associated respiratory effects,
November 2013 4-185 DRAFT: Do Not Cite or Quote
-------
1 the range of mean ambient NO2 concentrations was 11.7-36.9 ppb for 24-h avg NO2 and
2 22.0-44.4 ppb for 1-h max NO2. Associations of NO2 with respiratory effects persist with
3 adjustment for meteorological factors and for copollutants such as BC, EC, UFP, PM25,
4 PMio, and CO. Because of limited analysis of potentially correlated copollutants and
5 recognized limitations of copollutant models, the epidemiologic evidence combined with
6 the biological plausibility provided by findings of NO 2-induced increases in AHR in
7 adults with asthma in previous controlled human exposure studies provide compelling
8 evidence for the independent respiratory effects of NO2 exposure. NO2-associated
9 increases in oxidative stress and pulmonary inflammation, particularly allergic
10 inflammation, describe key events to inform the modes of action for AHR, lung function
11 decrements, and increases in asthma morbidity. The consistency and coherence of
12 evidence for increases in asthma morbidity, including biological plausibility and
13 copollutant-adjusted associations found for NO2, with more limited evidence for COPD,
14 impaired host defense, and respiratory mortality is sufficient to conclude that a causal
15 relationship exists between short-term NO2 exposure and respiratory effects.
November 2013 4-186 DRAFT: Do Not Cite or Quote
-------
Study
EMu,
Tolbertetal. (2007)
Samolietal. (2011)
Faustinietal. (2013)
Burnett etal. (1997)
Qianetal. (2009)
Qianetal. (2009)
Martinsetal. (2012)
Martiinsetal. (2012)
Peacocketal. (2011)
EM2.5
Lin etal. (2011)
Delfino etal. (2008)
Burnett etal. (1997)
Moshammeretal. (2006)
McCreanoretal. (2007)
Liu etal. (2009)
Adamkiewiczetal. (2004)
PMc
Mann etal. (2010)
Burnett etal. (1997)
Outcome
Respiratory ED visits
Asthma HospitalAd.
Respiratory Hospital Ad
Resp Hospital Ad.
eNO
PEF
EBCpH
FEV,
Symptomatic fall in PEF
eNO
FEV, % predicted
Resp Hospital Ad.
FEV,
FEV,
FEV,
Wheeze
Resp Hospital Ad.
Exposure metric
1-h maxNO2
1-h max NO2
24-h avg NO2
1 2-h avg NO2
24-h avg NO2
24-h avg NO2
1 -week avg NO2
1 -week avg NO2
1-h maxNO2
24-h avg NO2
24-h avg NO2 P
24-h avg NO2 CS
1 2-h avg NO2
8-h avg NO2
2-h avg NO2
24-h avg NO2
24-h avg NO
24-h avg NO2
1 2-h avg NO2
Correlation with
NO2orNO
0.53
' n '"'"
0.22-0.79
0.61
NR
NR
-0.72-0.55
-0.72-0.55
NR
0.30
0.38
0.36
0.45
0.54
0.60
0.71
NR
0.12
0.57
O-
— O
0 *
0
— 0
^
^^^-
0
o-
_^r
— o—
o
•-
0-
_^r
-10 -5 0 5 10 15 20 25 30 35 40
Percent change in outcome per 20,25, or 30 ppb increase in NO2 or NO (95% Cl)
Note: Magnitude and precision of effect estimates should not be compared among different outcomes. Results are organized by
copollutant analyzed, then in order of decreasing correlation between NO2 and copollutant. Studies not reporting correlations are
presented thereafter. Range of correlations refers to correlations across cities or different times of outcome assessment. Percent
change in FEV,, PEF, or EEC pH refers to percent decrease. Studies in Red = recent studies, Studies in Black = Studies reviewed
in the 2008 ISA for Oxides of Nitrogen. Effect estimates in Closed Circles = NO2 in a single-pollutant model, Effect estimates in
Open Circles = NO2 effect estimate adjusted for a copollutant. ED = Emergency department, Resp Hospital Ad. = Respiratory-
related Hospital Admission, eNO = Exhaled nitric oxide, PEF, Peak Expiratory Flow, EEC = Exhaled breath condensate, FEV, =
Forced Expiratory Volume in 1 second, P = Personal NO2, CS = Central site NO2, NR = Not reported.
aEffect estimates standardized to a 20-ppb increase for 24-avg or 1 -week average NO2 or NO, a 25-ppb increase for 8-h or 12-h avg
NO2, and a 30-ppb increase for 1 -h max or 2-h avg NO2. Quantitative data are reported in Section 4.2.7.3 and Table 4-7 and Table
4-14.
Figure 4-10 Associations of ambient NO2 or NO with respiratory outcomes
adjusted for PM™, PM2.5, or PM 10-2.5.
November 2013
4-187
DRAFT: Do Not Cite or Quote
-------
Study
Outcome
Exposure metric Correlation with NO2
orNOx
EC/BC
McCreanoretal. (2007) FEV,
Straketal. (2012)
Linetal. (2011)
Straketal. (2012)
eNO
eNO
eNO
2-h avg NO2
5-h avg NO2
5-h avg NOX
Peacocketal. (2011) Symptomatic Fall PEF 1-hmaxNO2
PNC/UFP
Straketal. (2012) eNO
Straketal. (2012) FVC
McCreanoretal. (2007) FEV,
von Klotet al. (2002) Wheeze
von Klotet al. (2002) Beta-agonist
Straketal. (2012) eNO
Straketal. (2012) FVC
5-h avg NO2
5-h avg NO2
2-h avg NO2
24-h avg NO2
24-h avg NO2
0.58
0.60
24-h avg NO2 0.68
0.87
NR
0.56
<«—
0.56
0.58
0.66
0.66
5-h avg NOX 0.75
5-h avg NOX 0.75
-30 -20 -10 0 10 20 30 40 50
Percent change in outcome per 20 or 30 ppb increase in NO2or
60 ppb increase in NOX (95% Cl)a
Note: Magnitude and precision of effect estimates should not be compared among different outcomes. Results are organized by
copollutant analyzed, then in order of decreasing correlation between NO2 and copollutant. Studies not reporting correlations are
presented thereafter. Percent change in FEV, refers to percent decrease. Studies in Red = recent studies, Studies in Black =
Studies reviewed in the 2008 ISA for Oxides of Nitrogen. Effect estimates in Closed Circles = NO2 in a single-pollutant model, Effect
estimates in Open Circles = NO2 effect estimate adjusted for a copollutant. FEV, = Forced Expiratory Volume in 1 second, PEF,
Peak Expiratory Flow, eNO = Exhaled nitric oxide, NR = Not reported.
"Effect estimates standardized to a 20-ppb increase for 24-avg NO2, a 30-ppb increase for 2-h avg or 5-h avg NO2, and a 60-ppb
increase for 2-h avg or 5-h avg NOX. Quantitative data are reported in Table 4-7. Table 4-14. and Table 4-18.
Figure 4-11 Associations of ambient NO2 or NOX with respiratory outcomes
adjusted for elemental carbon (EC) or black carbon (BC), ultrafine
particles (UFP), or particle number concentration (PNC).
November 2013
4-188
DRAFT: Do Not Cite or Quote
-------
Table 4-23 Summary of evidence supporting a causal relationship between
short-term NO2 exposure and respiratory effects.
Rationale for
Causal
Determination3 Key Evidence13 Key References'3
NO2
Concentrations
Associated with
Effects0
Asthma morbidity
Consistent Increases in asthma hospital
associations admissions, ED visits in diverse
from multiple, populations in association with
high quality 24-h avg and 1-h max NC>2, lags 0
epidemiologic and 3 to 5-day avg in additional
studies at recent studies of all ages and
relevant children.
concentrations No ass0ciation in recent Canadian
multicity study of all ages
Strickland et al. (2010), Villeneuve et
al. (2007), Ko et al. (2007b), Son et al.
(2013), Ito et al. (2007), Li et al.
(2011b)
Sections 4.2.7.3, Figure 4-4
Overall study
mean 24-h avg:
15.7-28.5 ppb
Overall study
mean 1-h max:
22.0-44.4 ppb
Stieb et al. (2009)
Mean 24-h avg:
22.7 ppb
Coherence with increases in
respiratory symptoms in children
with asthma in diverse populations
in association with 24-h avg, 2-4
avg NO2, 1-h max, lags 0, 3 to
6-day avg in previous and recent
studies
U.S. multicity studies: Mortimer et al.
(2002). Schildcrout et al. (2006)
Gent et al. (2003), Mann et al. (2010),
Section 4.2.6.1. Figure 4-3
Overall study
mean 24-h avg:
14.2-28.6 ppb
Overall study
mean 1-h max:
37.4-66 ppb
Decrements in lung function in
children with asthma in recent
studies; inconsistent results in
adults with asthma. Associations
with school outdoor, central site,
personal NO2.
O'Connor et al. (2008). Greenwald et
al. (2013). Holquin et al. (2007).
Delfino et al. (2008a). Section 4.2.3.1.
Figure 4-1
Outdoor school-
specific mean
1-week avg:
4.5-18.3 ppb
Overall study
mean 24-h avg
personal: 26.8 ppb
November 2013
4-189
DRAFT: Do Not Cite or Quote
-------
Table 4-23 (Continued): Summary of evidence supporting a causal relationship between
short-term NOi exposure and respiratory effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2
Concentrations
Associated with
Effects0
Chance,
confounding,
and other biases
can be ruled out
with reasonable
confidence in
part, by
epidemiologic
evidence
Across various outcomes and
locations, NC>2 associations found
with adjustment for weather, time
trends in previous and recent
studies
Across various outcomes and
locations, NC>2 associations persist
in copollutant models adjusted for
PM-io, PM25, EC, BC, BS, UFP,
CO, VOCs,O3
Recent studies expand on
evidence
Ko et al. (2007b), Strickland et al.
(2010), Villeneuve et al. (2007),
Jalaludin et al. (2008)
Delfino et al. (2008a): Delfino et al.
(2006), Martins et al. (2012), Qian et
al. (2009a), Gent et al. (2003), Mann
et al. (2010), von Klot et al. (2002)
Figure 4-10 and Figure 4-11
Same as above
Some associations were
attenuated with adjustment for
PM2.5, SO2, UFP
Samoli et al. (2011). Liu et al.
(2009b).
Copollutants weakly-moderately
correlated with NO2 in many
studies (r =-0.43 to 0.49)
Sarnat et al. (2012), Delfino et al.
(2006), Martins et al. (2012), Gent et
al. (2003)
Indoor NO2 associated with
increases in respiratory effects in
children with asthma in previous
and recent studies
Consistent results across various
lags of exposure and outcomes
examined in previous and recent
studies.
Previous and recent panel studies
of children examine representative
populations recruited from schools.
Sarnat et al. (2012). Greenwald et al.
(2013). Luetal. (2013). Hansel et al.
(2008)
Chance,
confounding,
and other biases
can be ruled out
with reasonable
confidence in
part, by
evidence from
controlled
human exposure
studies
NO2-induced increases in AHR in
adults with asthma exposed at rest
following nonspecific or allergen
challenge in several individual
previous studies and meta-
analyses.
Folinsbee (1992) 100ppbfor1h
Section 4.2.2, Table 4-3, Table 4-4, 200-300 ppb for
Table 4-5 30 min
November 2013
4-190
DRAFT: Do Not Cite or Quote
-------
Table 4-23 (Continued): Summary of evidence supporting a causal relationship between
short-term NOi exposure and respiratory effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2
Concentrations
Associated with
Effects0
Some evidence
describes key
events to inform
mode of action
Modification of
innate and
adaptive
immunity
Increases in eosinophil activation,
IgE, Th2 cytokines in adults and
rats and guinea pigs
Barck et al. (2005a): Barck et al.
(2002). Gilmouretal. (1996). Ohashi
etal. (1994)
Table 4-11, Table 4-12, Sections
3.3.2.6. and 4.2.4.3
Humans: 260 ppb
15-30 min
Rats/guinea pigs:
3,000 ppb for 2
weeks, 5,000 ppb
forSh
No consistent effect on pulmonary
clearance
Sections 4.2.5.2
1,500-3,500 ppb
for 2-6 h
Initiation of
inflammation
COPD
Inconsistent
epidemiologic
evidence
Increases in PMNs and
prostaglandins in healthy adults
Section 4.2.4.1
5,000 ppb for 3 h
Increases in eNO in children and
adults with asthma in association
with 2-h avg, 24-h avg NC>2
Delfino et al. (2006). Sarnat et al.
(2012). Martins et al. (2012). Qian et
al. (2009a)
24-h avg: 26.8 ppb
personal, 23.6 ppb
ambient
Oxidative
stress
Alteration of
epithelial
barrier
function
Recent studies expand on
evidence.
Changes in antioxidants in
experimental animals
Increases in LDH, BALF protein,
Clara cells in some but not all
studies of humans, rats, guinea
pigs
Section 4.2.4.4
Sections 3.3.2.3
Sections 4. 2. 4.1,
4.2.4.2, 3.3.2.4
1 -week avg
outdoor school:
4.5-18.3 ppb
Animal models:
400 or 2, 000 ppb
for 1-3 weeks
Increases in COPD hospital
admissions and ED visits in
additional recent studies
Lack of association with lung
function decrements and
symptoms in adults with COPD in
previous and recent studies
Faustini etal. (2013), Ko et al.
(2007b), Arbex et al. (2009)
Sections 4.2.7.2 and 4.2.7.5
Sections 4.2.3.1 and 4.2.6.1
Mean 24-h avg:
24.1-34.6 ppb
Mean 1-h max:
63.0 ppb
Inconsistent
evidence from
controlled
human exposure
studies
Increased inflammation in adults See above for asthma morbidity
Inconsistent evidence for
decreased lung function in adults
with COPD in previous studies
Decreases found in:
Morrow etal. (1992). Vaqaqqini et al.
(1996). Section 4.2.3.2
300 ppb
for 1 h> 4 h
November 2013
4-191
DRAFT: Do Not Cite or Quote
-------
Table 4-23 (Continued): Summary of evidence supporting a causal relationship between
short-term NOi exposure and respiratory effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2
Concentrations
Associated with
Effects0
Impaired Host Defense
Consistent Previous studies show mortality
animal from bacterial or viral infection in
toxicological animals with relevant NC>2
evidence exposures
Ehrlichetal. (1977), Ehrlich et al.
(1979). Ehrlich (1980). Graham et al.
(1987)
Section 4.2.5.1
1,500-5,000 ppb
for 1-8 h
Some Additional recent evidence for
epidemiologic associations with hospital
evidence admissions/ED visits for
respiratory infections and parental
reports of infection, particularly in
children. Some results have wide
95% CIs.
Zemek et al. (2010). Mehta et al.
(2013). Stieb et al. (2009). Faustini et
al. (2013). Just et al. (2002). Stern et
al. (2013)
Sections 4.2.5.1. 4.2.7. 4.2.8
Mean 24-h avg:
11.7-34.6 ppb
Some evidence
describes key
events to inform
mode of action
Initiation of
inflammation
See above for asthma morbidity
Modification of Diminished superoxide production
innate and in AM and bactericidal activity
adaptive |\|0 consistent effect on pulmonary
immunity clearance
Section 4.2.5.2
Respiratory Mortality
Consistent Recent multicity studies add to
epidemiologic evidence for associations of
evidence respiratory mortality with 24-h avg
NO2 at lag 0-1 days
NO2 results robust to adjustment
for PM-io, SO2, O3
Wong et al. (2008b). Chen et al.
(2012b). Chiusoloetal. (2011). Bellini
et al. (2007). Biqqeri et al. (2005)
Means across
cities for 24-h avg:
13.5-55.5 ppb
Limited Some evidence for asthma
biological morbidity in adults but limited
plausibility coherence among lines of
evidence for COPD and
respiratory infection.
Uncertainty regarding spectrum of
effects that can lead to respiratory
mortality.
"Based on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causal determination and,
where applicable, to uncertainties or inconsistencies. References to earlier sections indicate where full body of evidence is
described.
""Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, < 5,000 ppb).
November 2013
4-192
DRAFT: Do Not Cite or Quote
-------
4.3 Cardiovascular Effects
4.3.1 Introduction
1 The 2008 ISA for Oxides of Nitrogen concluded that the "available evidence on the
2 effects of short-term exposure to NO2 on cardiovascular health effects was inadequate to
3 infer the presence or absence of a causal relationship" (U.S. EPA. 2008c). Specifically,
4 the epidemiologic studies of heart rate variability (HRV), electrocardiographic markers of
5 cardiac repolarization, and arrhythmic events among patients with implanted cardioverter
6 defibrillators available at the time of the last review were inconsistent. Additionally,
7 while multiple studies reviewed had found associations between NO2 and rates of
8 hospital admission or ED visits for cardiovascular diseases (CVDs), it was unclear at that
9 time whether these results supported a direct effect of short-term NO2 exposure on
10 cardiovascular morbidity or were confounded by other correlated pollutants. Some
11 evidence from toxicological studies was reported for effects of NO2 on various
12 hematological parameters in animals, but these studies were limited, inconsistent, and
13 provided little biological plausibility for the cardiovascular effects observed in
14 epidemiologic studies.
15 This section reviews the published studies pertaining to the cardiovascular effects of
16 exposure to oxides of nitrogen in humans, animals, and cells. With the existing body of
17 evidence serving as the foundation, where available, emphasis was placed on studies
18 published since the 2008 ISA for Oxides of Nitrogen. As described in the following
19 sections, the recent epidemiologic and toxicological publications strengthen the evidence
20 for independent effects of NO2 exposure on cardiovascular morbidity and mortality.
4.3.2 Arrhythmia and Cardiac Arrest
21 The 2008 ISA for Oxides of Nitrogen found little epidemiologic evidence of an
22 association between short-term changes in ambient NO2 concentrations and cardiac
23 arrhythmias (U.S. EPA. 2008c). There continues to be limited epidemiologic evidence for
24 such an association, either from studies of patients with implantable cardioverter
25 defibrillators (ICDs), studies of arrhythmias detected on ambulatory electrocardiographic
26 (ECG) recordings, studies of out-of-hospital cardiac arrest, or studies of hospital
27 admission with a primary discharge diagnosis related to arrhythmias (Table 4-24).
November 2013 4-193 DRAFT: Do Not Cite or Quote
-------
1 In terms of studies of patients with ICDs, Ljungman et al. (2008) found that NO2 was
2 positively associated with increased risk of confirmed ventricular tachyarrhythmias in a
3 panel of patients with ICDs. The association with PM10 and PM2 5 was stronger than the
4 association for NO2. The authors observed no evidence of effect modification by city,
5 distance from the nearest ambient monitor at the time of the event, number of events,
6 type of event (ventricular fibrillation versus ventricular tachycardia), age, history of
7 ischemic heart disease (IHD), left ventricular ejection fraction, diabetes, body mass
8 index, or use of beta blockers. They did, however, report effect modification depending
9 on whether the patient was indoors or outdoors at the time of the event with a strong
10 association between NO2 and risk of ventricular tachyarrhythmias among the 22 subjects
11 that were outdoors at the time of ICD activation. Because the study authors accounted for
12 personal activity/behavior, exposure measurement error may have been reduced in the
13 effect modification analysis (Section 2.6.5.2). In a similar study, Anderson et al. (2010)
14 observed generally null associations between ICD activation and ambient NO, NO2, or
15 NOX concentrations. Anderson et al. (2010) were able to review the electrocardiograms
16 from only about 60% of ICD activations, potentially leading to greater misclassification
17 of the outcome than in the study by Ljungman et al. (2008). Recently, Link et al. (2013)
18 examined a panel of patients with dual chamber ICDs. They observed positive
19 associations between ICD-detected arrhythmias and atrial fibrillations > 30 seconds and
20 NO2 concentrations, though the effects associated with NO2 were smaller than those
21 observed for PM2 5. The NO2 associations were generally stronger when the authors used
22 a 2-h lag compared to a 2-day lag. Finally, Metzger et al. (2007) observed generally null
23 associations between NO2 concentrations and ventricular tachyarrhythmic events over a
24 10-year period in Atlanta, GA. Using a different approach, Barclay et al. (2009) generally
25 observed weak and inconsistent associations between NO2 or NO and incident
26 arrhythmias detected on ambulatory ECG recordings in a repeated-measures study of
27 non-smoking patients with stable heart failure.
28 The majority of out-of-hospital cardiac arrests are due to cardiac arrhythmias.
29 Dennekamp et al. (2010) observed generally positive, though weak, associations between
30 NO2 concentrations and risk of out-of-hospital cardiac arrest. A similar approach was
31 used by Silverman et al. (2010) using data from out-of-hospital cardiac arrests in New
32 York City and observed generally null associations with NO2 concentrations in all year
33 and cold season analyses, and a positive association in the warm season analysis.
34 In summary, there is currently inconsistent epidemiologic evidence for an association
35 between 24-h avg NO2 or NO and risk of cardiac arrhythmias as examined in patients
36 with ICDs, continuous ECG recordings, and out-of-hospital cardiac arrest. However,
37 existing studies have focused almost exclusively on ventricular arrhythmias and are
38 potentially limited by misclassification of the outcome.
November 2013 4-194 DRAFT: Do Not Cite or Quote
-------
Table 4-24 Epidemiologic studies of arrhythmia and cardiac arrest.
Study
Liunqman et
al. (2008)
Anderson et
al. (2010)
Linket al.
(2013)
Metzqer et al.
(2007)
Location
(Sample Size)
Gothenburg and
Stockholm,
Sweden
(n = 211
[266 events])
London, U.K.
(n = 705
[5,462 device
activations])
Boston, MA
(n = 176[328atrial
fibrillation episodes
> 30 seconds])
Atlanta, GA
(n = 518)
Mean NC>2 (ppb)
24-h avg NO2
Gothenburg: 11.8
Stockholm: 8.3
24-h avg NO2: 12.1
24-h avg NOX:
24.1
24-h avg NO: 19.4
24-h avg NO2:
16.1
1-h max NO2: 44.9
90th: 68
Max: 181
Exposure
assessment
Single monitor in
Gothenburg,
average of
2 monitors in
Stockholm
City-wide avg
City wide avg
Central Monitor
Selected Effect Estimates3
(95% Cl)
Ventricular Tachyarrhythmia (OR)
2-havg: 1.37(0.53, 3.64)
24-h avg: 1.26(0.49, 3.32)
ICD Activations (OR)
NO2; Lag 0-1: 0.93 (0.70, 1.24)
NOX; Lag 0-1: 0.92(0.86, 1.08)
NO: Lag 0-1: 0.96(0.93, 1.04)
ICD-detected Arrhythmias (OR)
24-h lag: 1.23(0.75,2.10)
2-hlag: 1.57(0.97,2.47)
All Arrhythmia events (OR)
All year: 1.00(0.95, 1.05)
Warm season: 1.00(0.93, 1.08)
Cold season: 1.01 (0.94, 1.08)
Events resulting in cardiac pacing or
defibrillation:
All year: 1.01 (0.94, 1.10)
Events resulting in defibrillation:
All year: 1.07(0.93, 1.23)
November 2013
4-195
DRAFT: Do Not Cite or Quote
-------
Table 4-24 (Continued): Epidemiologic studies of arrhythmia and cardiac arrest.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3
(95% Cl)
Barclay et al.
(2009)
Aberdeen,
Scotland
(n = 132)
24-h avg NO2:
30.1
NO: 14.7
Central Monitor
All Arrhythmias (regression
coefficients)
NO2: 3.193 (-3.600, 9.985)
NO: 3.524 (-3.059, 10.107)
Ventricular ectopic beats
NO2: 3.642 (-4.837, 12.121)
NO: 4.588 (-3.628, 12.803)
Ventricular couplets
NO2: 0.356 (-7.395, 8.106)
NO:-0.085 (-7.601, 7.431)
Ventricular runs
NO2: 2.443 (-2.537, 7.422)
NO: 2.219 (-2.618, 7.055)
Supraventricular ectopic beats
NO2: 2.888 (-4.833, 10.608)
NO:-2.688 (-10.170, 4.794)
Supraventricular couplets
NO2: 5.209 (-1.896, 12.313)
NO: 1.366 (-5.542, 8.274)
Supraventricular runs
NO2: 3.441 (-1.760, 8.641)
NO: 2.298 (-2.753, 7.348)
Dennekamp et Melbourne,
al. (2010) Australia
(n = 8,434)
24-h avg NO2:
12.0
75th: 15.16
Central Monitor
% Change in out-of-hospital
cardiac arrest
LagO: 3.23 (-10.19, 18.51)
Lag 1: 7.69 (-7.29, 25.11)
Lag 2:-4.51 (-16.48, 10.56)
Lag 3: 7.37 (-7.11, 24.13)
Lag 0-3: 9.28 (-7.54, 29.14)
Silverman et
al. (2010)
New York City, NY 24-h avg NO2
(n = 8,216) 50th: 27
75th: 32
95th: 43
City-wide avg No quantitative results presented for
NO2
a Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration
for 24- h and 1-h averaging times, respectively.
4.3.3
Heart Rate/Heart Rate Variability
HRV provides anon-invasive marker of autonomic nervous system function. Decreases
in indices of HRV have been associated with increased risk of cardiovascular events in
November 2013
4-196
DRAFT: Do Not Cite or Quote
-------
1 prospective cohort studies (La Rovere et al., 2003; Kikuya et al., 2000; Tsuji etal. 1996;
2 Tsuji et al.. 1994). The rhythmic variation in the intervals between heart beats can be
3 quantified in either the time domain or the frequency domain (Task Force of the
4 European Society of Cardiology and the North American Society of Pacing and
5 Electrophysiology. 1996). Common time-domain measures of HRV include the standard
6 deviation of all normal-to-normal intervals (SDNN, an index of total HRV) and the root-
7 mean-square of successive differences (rMSSD, an index influenced mainly by the
8 parasympathetic nervous system). In the frequency domain, HRV is usually divided into
9 the high frequency (HF, an index influenced mainly by the parasympathetic nervous
10 system) and low frequency (LF) components, as well as the ratio of the LF to HF
11 components (LF/HF, an index of relative sympathovagal balance) (Task Force of the
12 European Society of Cardiology and the North American Society of Pacing and
13 Electrophysiology. 1996). Sympathetic stimulation increases the firing rate of pacemaker
14 cells in the heart's sinoatrial node, thereby increasing heart rate (HR) as well as affecting
15 the LF/HF ratio. On the other hand, parasympathetic stimulation decreases the firing rate
16 of pacemaker cells and the HR and affects the HF component of HRV.
4.3.3.1 Epidemiologic Studies
17 The 2008 ISA for Oxides of Nitrogen found that there was insufficient evidence to
18 determine whether exposure to oxides of nitrogen was associated with changes in cardiac
19 autonomic control as assessed by indices of HRV (U.S. EPA, 2008c). Additional studies
20 are now available for review (Table 4-25) that provide evidence for an association
21 between exposure to NO2 and HRV among those with pre-existing disease, but not in
22 healthy individuals.
23 The multi-country ULTRA study assessed the longitudinal association between ambient
24 pollution and HRV among elderly participants with stable coronary artery disease in
25 Amsterdam, the Netherlands, Erfurt, Germany, and Helsinki, Finland (Timonen et al..
26 2006). In each participant, HRV was assessed multiple times over a 6 month period,
27 resulting in a total of 1,266 repeated measures. Pooling results across the three cities, the
28 authors found a 3.01 msec (95% CI: -5.94, -0.11) decrease in SDNN and a 18.8% (95%
29 CI: -28.4%, -3.0%) decrease in LF/HF associated with a 20-ppb increase in 24-h NO2
30 levels at lag 2. The magnitudes of these associations were somewhat larger in relation to
31 the 5-day moving average of NO2. The authors report that these effects were robust to
32 adjustment for other pollutants in two-pollutant models, but detailed results were not
33 provided. These results were reportedly similar in men and women and after exclusion of
34 those exposed to environmental tobacco smoke at home. Most associations with HF were
35 positive.
November 2013 4-197 DRAFT: Do Not Cite or Quote
-------
1 Huang etal. (2012a) measured HRV repeatedly in participants with pre-existing
2 cardiovascular disease in Beijing in the summer of 2007 and again in the summer of
3 2008, including one measurement period during the 2008 Beijing Olympics when city-
4 wide air pollution control measures substantially reduced ambient concentrations of most
5 criteria pollutants. In this study, NO2 concentrations during the Olympics were reduced
6 by close to 22% versus the previous month and 13% versus the same period the previous
7 summer (Huang et al., 2012a). Other ambient pollutants (except O3) were reduced by
8 similar or larger amounts. In single-pollutant models, a 30-ppb increase in 1-h max NO2
9 was associated with a 8.7% decrease (95% CI: -12.0%, -4.8%) in SDNN, a 20.5%
10 decrease (95% CI: -26.7%, -9.2%) in LF. The association with SDNN was stronger
11 among those with a higher CRP, women, and those without a history of diabetes, but
12 BMI did not appear to modify the association. Rich etal. (2012) also examined the
13 association between heart rate and NO2 concentrations before, during and after the 2008
14 Beijing Olympics. They observed increases in heart rate that were generally consistent in
15 magnitude across lags from 0 to 6 days.
16 Several studies (Weichenthal et al.. 2011; Laumbach et al. 2010; Suh and Zanobetti.
17 2010a) used exposure assessment techniques that would tend to reduce uncertainty in the
18 exposure when compared to the use of central site monitors. Suh and Zanobetti (2010a)
19 examined the association between HRV and short-term exposure to NO2 among people
20 that had either recently experienced an MI or had COPD. Same-day personal exposures
21 to NO2 were associated with decreased HRV. Decreases in PNN50 were the largest
22 among the individuals with COPD, while NO 2-associated decrements in HF were the
23 largest among individuals with a recent MI, but were less precise when all individuals or
24 individuals with COPD were included in the analysis. Laumbach et al. (2010) studied the
25 effects of in-vehicle exposure to traffic-related pollutants among a group of individuals
26 with diabetes. The authors observed decreases in HF HRV about 1 day after the in-
27 vehicle exposures, with effects that were similar, but smaller in magnitude, attributed to
28 NO2 concentrations. Weichenthal et al. (2011) carried out a cross-over trial with 42
29 healthy adults who cycled for 1 hour on high- and low-traffic routes as well as indoors.
30 Mean levels of NO2 measured at nearby stationary monitors were associated with
31 decreases in SDNN and increases in LF/HF.
32 In a repeated-measures study of Boston-area patients with clinically significant coronary
33 artery disease, Zanobetti et al. (2010) found that HF was inversely associated with
34 ambient NO2 concentrations. This association remained robust after adjustment for PM2 5
35 in a two-pollutant model. Among a population reporting a substantial prevalence of
36 cardiovascular risk factors (i.e., hypertension, diabetes, hyperlipidemia), Williams et al.
37 (2012a) observed a strong association between NO2 concentrations and reduced heart
38 rate. On the other hand, Barclay et al. (2009) found no association between NO2 or NO
November 2013 4-198 DRAFT: Do Not Cite or Quote
-------
1 and indices of HRV in a repeated-measures study of non-smoking patients with stable
2 heart failure. Also, Goldberg et al. (2008) followed 31 Montreal-area participants with
3 stable heart failure for 2 months and found no association between pulse rate and NO2
4 concentrations.
5 Infants are potentially at greater risk of pollution-related health effects. Peel et al. (2011)
6 examined data from 4,277 Atlanta-area infants prescribed home cardiorespiratory
7 monitors and found a slightly elevated risk of bradycardia linked to 1-h maximum NO2
8 concentrations over the past 2 days measured at a central site monitor. The clinical or
9 public health significance of this finding is unclear.
10 The above studies have all focused on infants or participants with a documented history
11 of heart disease. In contrast, HRV appears to be not associated with NO2 concentrations
12 in healthy participants. For example, a repeated-measures study of young healthy
13 participants in Taipei, Taiwan found no association between NO2 and HRV indices
14 (Chuang et al.. 2007a). In Beijing, Jiaetal. (2011) assessed HRV two times in each of 20
15 healthy participants and reported no association between oxides of nitrogen and HRV.
16 However, this study was quite small and detailed results were not shown.
17 Cross-sectional analyses of populations with or without a history of heart disease have
18 also tended to yield null results. In a cross-sectional analysis of 5,465 participants from
19 the multi-city Multi-Ethnic Study of Atherosclerosis (MESA), Parketal. (2010) found no
20 association between NO2 concentrations and indices of HRV. Participants in this study
21 were 45-84 years old and free of cardiovascular disease. A cross-sectional study from
22 Taipei also found no association between NO2 and HRV among 46 elderly participants
23 with cardiovascular disease risk factors (Chuang et al., 2007b). A cross-sectional study of
24 1,349 healthy participants in Taein Island, Korea by Min et al. (2008) found that NO2
25 was associated with decreases in the LF component of HRV, but not with changes in
26 SDNN or the HF component.
27 In summary, current evidence suggests that among participants with pre-existing or
28 elevated risk for cardiovascular disease, ambient NO2 concentrations are associated with
29 alterations in cardiac autonomic control as assessed by indices of HRV; however,
30 evidence for differential effects between populations with and without pre-existing
31 diseases and conditions is limited. In this specific subgroup of the population, HRV
32 seems to be associated with changes in HRV consistent with relative increases in
33 sympathetic nervous system activity and/or decreases in parasympathetic nervous system
34 activity. In contrast, this association has not been commonly apparent among healthier
35 participants.
November 2013 4-199 DRAFT: Do Not Cite or Quote
-------
Table 4-25 Epidemiologic studies of heart rate/heart rate variability.
Study
Location
(Sample Size)
Mean NC>2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Timonen et al. (2006)
Amsterdam,
the Netherlands;
Erfurt, Germany;
Helsinki, Finland
(n = 131)
24-h avg NO2
Amsterdam: 22.7
Erfurt: 15.4
Helsinki: 16.5
Central Monitor
SDNN (msec)
Lag 0:-1.05 (-3.50, 1.39)
Lag 1:-1.28 (-3.98, 1.43)
Lag 2:-3.01 (-5.94,-0.11)
Lag 3: -0.68 (-3.42, 2.07)
Lag 0-4:-4.59 (-9.32, 0.15)
HF (%)
Lag 0: 4.51 (-33.46, 18.42)
Lag 1:4.51 (-10.15, 19.55)
Lag 2: 1.88 (-13.91, 18.05)
Lag 3:-1.88 (-16.92, 13.16)
Lag 0-4: 4.51 (-22.18, 30.83)
LF/HF (%)
Lag 0:-3.01 (-15.41, 9.77)
Lag 1:-16.54 (-30.08,-3.01)
Lag 2:-17.67 (-31.95,-3.01)
Lag 3:-1.88 (-15.41, 11.65)
Lag 0-4:-25.94 (-50.00,-1.88)
Huang et al. (2012a) Beijing, China
(n = 40)
1-h max NC>2
2007, Visit 1: 33.8
2007, Visit 2: 26.3
2008, Visit 3: 29.2
2008, Visit 4: 22.9
Central Monitor
SDNN (% Change)
1-h:-3.75 (-6.71,-0.59)
4-h: -8.54 (-12.48, -4.82)
12-h:-9.91 (-15.14,-4.40)
r-MSSD (% Change)
1-h: 2.76 (-2.17, 7.70)
4-h: -4.82 (-12.48, 3.28)
12-h:-6.06 (-16.79, 5.50)
LF(% Change)
1-h:-10.66 (-18.36,-2.76)
4-h:-19.49 (-28.91,-9.42)
12-h:-21.74 (-35.23,-7.71)
HF (% Change)
1-h:-6.91 (-16.18,2.76)
4-h:-11.17 (-24.09, 2.85)
12-h:-10.18 (-28.62, 9.63)
Zanobetti et al. Boston, MA 2-h avg NO2 City-wide avg HF (% Change)
£°-l£) (n = 46) 50th: 21 2-h: -18.27 (-29.45, -6.82)
75th' 27
(aged 43-75 years) g5th; 36
72-h avg NO2
50th: 21
75th: 25
95th: 31
Lag 0-4: -47.00 (-70.50, -22.00)
All other results presented graphically, no quantitative results
November 2013
4-200
DRAFT: Do Not Cite or Quote
-------
Table 4-25 (Continued): Epidemiologic studies of heart rate/heart rate variability.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Barclay et al. (2009) Aberdeen,
Scotland, U.K.
(n = 132)
24-havgNO2: 30.1
24-h avg NO: 14.7
Central monitor
HR
NO2: 0.398 (-0.003, 0.799)
NO: 0.353 (-0.036, 0.742)
SDNN (msec)
NO2: 0.619 (-0,588, 1.826)
NO: 0.608 (-0.562, 1.778)
SDANN (msec)
NO2: 0.512 (-0.865, 1.890)
NO:0.570 (-0.766, 1.906)
rMSSD (msec)
NO2: 0.398 (-0.003, 0.799)
NO: 0.353 (-0.036, 0.742)
PNN 50%
NO2: 1.568 (-3.851, 6.986)
NO: 0.909 (-4.347, 6.165)
LF power
NO2: 2.353 (-1.052, 5.757)
NO: 1.940 (-1.328, 5.207)
LF normalized
NO2:-0.857 (-2.817, 1.102)
NO:-0.183 (-2.066, 1.699)
HF power
NO2: 3.365 (-1.169, 7.900)
NO: 2.895 (-1.457, 7.247)
HF normalized
NO2: 0.722 (-1.554, 2.998)
NO: 1.407 (-0.775, 3.558)
LF/HF ratio
NO2:-1.089 (-3.930, 1.753)
NO:-1.054 (-3.779, 1.672)
Goldberg et al.
(2008)
Chuanq et al. (2007a)
Jiaetal. (2011)
Montreal, Quebec,
Canada
(n = 31)
Taipei, Taiwan
(n = 76)
Beijing, China
(n = 41)
24-h avg NO2
17.9
Max: 54.1
24-h avg NO2
17.3
Max: 53.1
24-h avg NOX
35.0
City-wide avg Pulse Rate (mean difference)
Lag 0: -0.07 (-0.09, 0.80)
Lag 1: 0.78 (-0.14, 1.71)
Lag 0-2: 0.99 (-0.34, 2.32)
Central Monitor "We found no associations between HRV indices and NO2"
No quantitative results presented
Central Monitor "No significant effects are found between daily average... NOx on HRV
indices"
No quantitative results presented
November 2013
4-201
DRAFT: Do Not Cite or Quote
-------
Table 4-25 (Continued): Epidemiologic studies of heart rate/heart rate variability.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Parketal. (2010)
6 U.S. Communities:
Baltimore, MD;
Chicago, IL;
Forsyth County, NC;
Los Angeles, CA;
New York, NY;
St. Paul, MN
(n = 5,465)
24-h avg NO2
Lag 0-1:
23.5
City-wide avg "There were no significant associations of HRV with gaseous
pollutants (data not shown)"
No quantitative results presented
Chuanq et al. (2007b) Taipei, Taiwan 1-h max NO2
(n = 46) 38.4
Avg of monitors
within 1 km of
residence
"...NO2...exposures were not associated with any HRV indices in our
study participants (data not shown)."
No quantitative results presented
Min et al. (2008)
Taein Island,
South Korea
(n = 1,349)
24-h avg NO2
24
75th: 30
Max: 119
Central Monitor
SDNN (% Change)
6-h:-2.45 (-6.28, 1.53)
9-h:-3.89 (-8.31, 0.71)
12-h:-3.81 (-8.75, 1.34)
24-h:-1.72 (-6.71, 3.51)
48-h: 2.93 (-2.33, 8.42)
72-h: 1.20 (-3.81, 6.42)
LF (% Change)
6-h:-8.61 (-16.85, 0.31)
9-h:-12.24 (-21.48,-2.11)
12-h:-12.28 (-22.58,-0.88)
24-h: -5.71 (-16.58, 6.33)
48-h: 3.69 (-8.22, 16.92)
72-h:5.84(-6.19, 18.45)
HF (% Change)
6-h:-1.08 (-10.75, 9.47)
9-h: -3.31 (-14.32, 8.88)
12-h:-2.38 (-14.73, 11.45)
24-h:-4.53 (-16.58, 8.94)
48-h: 4.42 (-8.72, 19.14)
72-h: 4.18 (-8.52, 18.35)
November 2013
4-202
DRAFT: Do Not Cite or Quote
-------
Table 4-25 (Continued): Epidemiologic studies of heart rate/heart rate variability.
Study
Weichenthal et al.
(2011)
Location
(Sample Size)
Ottawa, Canada
(n = 42)
Mean NO2 (ppb)
1-h max NO2
4.8
Exposure
assessment
Central Monitor
Selected Effect Estimates3 (95% Cl)
ALF (ms2) ASDNN (msec)
1-h: -532.5 (-2872.5, 1807.5) 1-h: -18.75 (-112.50,
72.00)
2-h: 12.0 (-2467.5, 2490.0)
3-h: 577.5 (-2055.0, 3217.5)
4-h: -397.5 (-3532.5, 2025.0)
AHF (ms2)
1-h:-420.0 (-1785.0, 952.5)
2-h:-487.5 (-1612.5, 637.5)
3-h:-24.0 (-1020.0, 975)
4-h:-247.5 (-1417.5, 922.5)
ALF:HF
1-h: 5.70 (-2.10, 13.50)
2-h: 10.50(2.63, 18.75)
3-h: 12.75(4.20,21.75)
4-h: 7.50 (-1.80, 17.25)
2-h:-75.0 (-150.0,-2.55)
3-h:-39.75 (-120.0, 40.50)
4-h:-12.00 (-82.50, 61.50)
ArMSSD (msec)
1-h:-12.00 (-48.75, 24.75)
2-h:-12.00 (-41.25, 17.25)
3-h: 2.33 (-30.0, 34.5)
4-h:-2.10 (-33.0, 29.25)
ApNN50(5)
1-h:-3.30 (-31.50, 24.75)
2-h:-8.25 (-33.00, 15.75)
3-h: -3.23 (-29.25, 22.50)
4-h: 1.28 (-26.25, 29.25)
Peeletal. (2011)
Williams et al.
(2012b):
Atlanta, GA
(n = 4,277)
Detroit, Ml
(n = 65)
1-h max NO2
41.7
90th: 65.6
Max: 109.2
24-h avg NO2
24.0
75th: 28.0
Max: 100.0
Central Monitor Bradycardia (OR)
1.04(1.00, 1.08)
Personal Monitor HR (bpm)
bpm: -2.95 (-4.82,
-0.80)
November 2013
4-203
DRAFT: Do Not Cite or Quote
-------
Table 4-25 (Continued): Epidemiologic studies of heart rate/heart rate variability.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Suh and Zanobetti
(201 Oa)
Atlanta, GA
(n = 30)
24-h avg NO2 City-wide avg SDNN (% change)
Ambient: 17.1 Personal Ambient:-0.64 (-11.06, 10.43)
Personal: 11.6 Personal: -3.48 (-10.69, 3.89)
rMSSD (% change)
Ambient: -6.60 (-30.64, 20.85)
Personal:-14.52 (-29.87, 1.70)
pNNSO (% change)
Ambient: 0.30 (-38.28, 47.38)
Personal: -32.30 (-56.49, -5.65)
HF (% change)
Ambient:-1.49 (-37.09, 41.32)
Personal: -21.35 (-44.92, 4.48)
LF/HF (% change)
Ambient: 13.74 (-4.11, 33.13)
Personal: 9.69 (-2.34, 22.20)
Laumbach et al.
(2010)
Richetal. (2012)
New Brunswick, NJ
Beijing, China
(n = 125)
NO2: In-vehicle mean
50th: 25.9
75th: 32.8
Max: 61.1
24-h avg NO2 Central Monitor
Entire study: 27.0
Before: 26.0
During: 13.9
After: 41.4
HF (% Change)
-11. 92 (-104.64, 80.79)
LF/HF Ratio (% Change)
-107.28 (-298.01, 83.44)
No quantitative results presented; results presented graphically.
Positive, but non-statistically significant increase in heart rate,
generally consistent across lags from 0 to 6.
aEffect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration for 24- h and 1-h averaging times, respectively.
November 2013
4-204
DRAFT: Do Not Cite or Quote
-------
4.3.3.2 Controlled Human Exposure Studies
1 HRV was not evaluated in previous controlled human exposure studies of NO2, but was
2 evaluated in two recent studies (Table 4-29). Huang etal. (2012b) recently evaluated
3 changes in various HRV parameters following NO2 exposure in healthy adult volunteers
4 performing intermittent exercise. Exposure to 500 ppb NO2 did not alter HRV time
5 domain intervals, but did slightly increase LF/HF, however this change was not
6 statistically significant. The authors reported an 11.6% and 13% decrease in the HF
7 domain normalized for heart rate (HFn) 1 and 18 hours after exposure, respectively.
8 Combined exposure to NO2 and PM25 CAPs increased LF/HF (1 hour; p = 0.062), as
9 well as LFn (1 hour; p= 0.021) and cardiac t wave amplitude (18 hour; p = 0.057). CAPs
10 exposure alone did not induce such changes. Vagal modulation of cardiac activity
11 following NO2 exposure was also observed in both a controlled human exposure and
12 animal toxicological study (Section 3.3.2.8); however, at higher than ambient-relevant
13 concentrations. Epidemiologic studies found NO2-associated decreases in HRV primarily
14 in adults with or at risk for cardiovascular disease. However, a recent study of adults with
15 stable coronary heart disease and impaired left ventricular systolic function showed no
16 statistically significant changes in HRV with 400 ppb NO2 for exposure for 1 hour while
17 seated and without exercise (Scaife etal.. 2012); however, it should be noted that the
18 study had only 75% power to detect significant differences in the HF domain of 50% or
19 less.
20 The few studies reviewed in the previous assessments of oxides of nitrogen (U.S. EPA.
21 2008c. 1993) reported mixed effects of NO2 exposure on HR; a recent study shows no
22 effect. Folinsbee et al. (1978) and Drechsler-Parks (1995) exposed healthy adult males
23 and healthy older adults, respectively, to approximately 600 ppb NO2 for 2 hours and
24 reported no changes in HR. Changes in HR were also examined in potentially at-risk
25 populations exposed to NO2. Exposure to 400 ppb NO2 did not alter HR in adults with
26 coronary heart disease (Scaife etal.. 2012) and resulted in a statistically nonsignificant
27 increase in adults with COPD and healthy volunteers, (Gong et al.. 2005). Among healthy
28 volunteers and those with asthma, NO2 exposure resulted in no change in HR (Linn et al..
29 1985a).
4.3.3.3 Toxicological Studies
30 Toxicology studies examining HRV changes were not available for review in the 2008
31 ISA for Oxides of Nitrogen. Consistent with controlled human exposure studies, a recent
November 2013 4-205 DRAFT: Do Not Cite or Quote
-------
1 study in rats found mixed evidence for changes in HR and HRV (Table 4-30). Ramos-
2 Bonilla et al. (2010) examined body weight, HR, and HRV, following exposure of aged
3 inbred mice to an ambient mixture consisting of PM, CO, and NO2. Animals were
4 exposed to either filtered or unfiltered outdoor Baltimore air for 6 hours daily for 40
5 weekdays. Health effects associated with daily exposure to each pollutant were
6 ascertained with multipollutant models and lagged covariates. Significant declines in HR
7 were associated with NO2 at lag 3 and the 7 day cumulative concentration with
8 adjustment for PM and CO. However, HRV changes were not associated with NO2
9 exposure. The independent effects of a pollutant are difficult to distinguish in a
10 multipollutant model because of multicollinearity among pollutants.
4.3.4 ST-Segment Amplitude and QT-lnterval Duration
11 ST-segment changes (either ST-segment elevation or depression) on the
12 electrocardiogram are considered a nonspecific marker of myocardial ischemia. The QT
13 interval provides an electrocardiographic marker of ventricular repolarization.
14 Prolongation of the QT interval is associated with increased risk of life-threatening
15 ventricular arrhythmias.
4.3.4.1 Epidemiologic Studies
16 The 2008 ISA for Oxides of Nitrogen did not review any epidemiologic studies of
17 ambient oxides of nitrogen concentrations and markers of myocardial ischemia or
18 ventricular repolarization (U.S. EPA. 2008c). A few recent studies examined these
19 endpoints (Table 4-26). Chuang et al. (2008) conducted a repeated-measures study of
20 Boston-area patients with a history of coronary heart disease and examined the
21 association between ambient pollutants and ST-segment level changes. This study found
22 an odds ratio of 3.29 (95% CI: 1.82, 5.92) for ST-segment depression of > 0.1 mm per
23 20-ppb increase in 24-h avg NO2 concentrations over the previous 24 hours. This finding
24 was robust to additional adjustment for PM2 5 in a two-pollutant model.
25 Delfino et al. (2011) used a similar design to study 38 elderly, nonsmoking residents of 4
26 retirement homes in the Los Angeles area with a documented history of coronary artery
27 disease. A particular strength of this study is that the authors measured pollutant
28 concentrations outside of the residence to help address uncertainties related to exposure
29 assessment. This study found an odds ratio of 10.13 (95% CI: 1.37, 74.23) for ST-
30 segment depression > 1.0 mm per 30-ppb increase in mean 1-h NO2 concentrations
31 preceding measurement over the previous 3 days. Other averaging periods from 8 hours
November 2013 4-206 DRAFT: Do Not Cite or Quote
-------
1 to 4 days gave similar or slightly weaker results. NO2 was more strongly associated with
2 ST depression than was NOX.
3 Within the context of the Veterans Administration Normative Aging Study, Baia et al.
4 (2010) found that heart-rate corrected QT interval was not associated with the 10 hour
5 moving average of NO2 concentrations among older, generally white men, but was
6 associated with NO2 concentrations at lags 3 and 4 hours (longer lags or moving averages
7 were not considered). The only prior study available for comparison found that 24-h avg
8 NO 2 concentrations were positively associated with increased QT interval duration, but
9 this association was imprecise, and the 6-hour moving average of NO2 was not associated
10 with QT interval duration (Henneberger et al.. 2005).
11 In summary, a few available epidemiologic studies report an association between short-
12 term exposure to NO2 and ST-segment changes on the electrocardiogram of elderly
13 participants with a history of coronary artery disease, potentially indicating an association
14 between NO2 and increased risk of myocardial ischemia in this patient population. No
15 previous studies are available for comparison. Additionally, a recent study suggests a
16 potential link between NO2 and ventricular repolarization as assessed by QT interval
17 duration, in contrast to a previous study.
November 2013 4-207 DRAFT: Do Not Cite or Quote
-------
Table 4-26 Epidemiologic studies of ST-segment amplitude and QT-interval
duration.
Study
Chuanq et al.
(2008)
Delfino et al.
(2011)
Baiaetal. (2010)
Henneberqer et al.
(2005)
Location Mean NOz
(Sample Size) (ppb)
Boston, MA 24-h avg NO2
(n=48) 21.4
75th: 24.9
Max: 44.5
Los Angeles, CA 1-h NO2: 27.5
(n = 38) 1-h NOX: 46.6
Boston, MA 1-h max NO2
(n = 580) 19-21
Erfurt, Germany 24-h avg NO2:
(n = 56) 18-2
75th: 22.6
Max: 36.4
24-h avg NO:
19.4
75th: 24.2
Max: 110.1
Exposure
assessment Selected Effect Estimates3 (95% Cl)
City-wide avg ST segment change (mm):
12-h: -0.02 (-0.05, 0.00)
24-h: -0.08 (-0.12, -0.05)
RR for ST-segment depression
>0.1 mm:
12-h: 1.15(0.72, 1.82)
24-h: 5.97(2.45, 14.40)
Outdoor monitor at OR for ST-segment depression
retirement > 1.0 (mm)
community
1-h: 1.33(0.83,2.11)
8-h: 2.37 (1.14, 4.92)
24-h: 4.75 (1.51, 14.84)
2-day: 7.51 (1.49, 37.87)
3 day: 10.13(1.37, 74.23)
4 day: 5.47 (0.65, 45.89)
NOX:
1-h: 1.25(0.96, 1.64)
8-h: 1.49(0.94,2.40)
24-h: 1.88(0.83,4.20)
2-day: 2. 10 (0.67, 6.45)
3 day: 2.31 (0.45, 11.83)
4 day: 1.82(0.21, 15.84)
City-wide avg Change in QTc (msec)
10-hlag: 5.91 (-2.03, 13.85)
4-h lag: 6.28 (-0.02, 12.55)
City-wide avg QTc (msec)
NO2, lag 6-11 h: 9.77 (2.23, 17.33)
T-wave complexity (%)
NO, lag 0-23: 0.15(0.02, 0.28)
T-wave amplitude (uV)
NO, lag 0-5 h: -2.10 (-4.16, -0.03)
aEffect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration
for 24- h and 1-h averaging times, respectively.
November 2013
4-208
DRAFT: Do Not Cite or Quote
-------
4.3.4.2 Controlled Human Exposure Studies
1 Epidemiology studies report mixed results for an association between QT interval
2 changes and NO2 exposure. A recent controlled human exposure study (Huang et al..
3 2012b) also evaluated this endpoint (Table 4-29). The study found a borderline
4 statistically significant decrease in QT interval corrected for heart rate (QTc) at 1 and 18
5 hours after a 2-hour exposure to 500 ppb NO2 with exercise in healthy volunteers. In this
6 study, NO2 exposure also induced a 29.9% decrease (p = 0.001) in the QT variability
7 index (an increase has been associated with arrhythmia). However, when volunteers were
8 exposed to both PM2 5 and NO2 the QT variability index increased. Overall the various
9 cardiovascular parameters examined in this study were mixed.
4.3.5 Blood Pressure
4.3.5.1 Epidemiologic Studies
10 The 2008 ISA for Oxides of Nitrogen did not review any epidemiologic studies of
11 ambient oxides of nitrogen concentrations and blood pressure (BP) (U.S. EPA. 2008c).
12 Several studies are now available for review (Table 4-27). There is little evidence from
13 longitudinal studies of the association between NO2 and BP.
14 In the Detroit area, Williams et al. (2012b) measured BP up to 10 times in each of 65
15 adult participants and found no association between BP and either personal or ambient
16 NO2 concentrations. Huang et al. (2012a) measured BP repeatedly in participants with
17 pre-existing cardiovascular disease in Beijing before, during, and after the 2008 Beijing
18 Olympics when city-wide air pollution control measures substantially reduced ambient
19 levels of most criteria pollutants, as described in more detail in Section 4.3.3.1. above.
20 Despite these large changes in NO2 concentrations, this study found no association
21 between NO2 and either systolic or diastolic BP. Using a similar study design, Rich et al.
22 (2012) also found no association between NO2 and either systolic or diastolic BP among
23 healthy young participants assessed before, during, and after the 2008 Beijing Olympics.
24 In a repeated-measures study of pregnant women in France, Hampel et al. (2011) found
25 that a 20-ppb increase in 24-h avg NO2 at lag 1 was associated with a 0.8% decrease in
26 systolic BP (95% CI: -1.3%, -0.3%). Similar associations were reported for lags 1, 5, and
27 6 days, as well as for the 7 day moving average. The magnitude of this association tended
28 to be stronger in the cool months of the year and among non-smoking women. Diastolic
29 BP was not associated with NO2 levels.
November 2013 4-209 DRAFT: Do Not Cite or Quote
-------
1 Results of cross-sectional studies of the association between NO2 and BP measured on
2 the same day or with the NO2 measurement lagged 1-3 days before the BP measurement
3 have also been mixed. Cakmak et al. (201 la) used cross-sectional data from a national
4 population-based survey of children and adults in Canada and found a 1.1 mmHg (95%
5 CI: 0.2, 2.0 mmHg) increase in systolic BP and a 2.06 mmHg (95% CI: 1.11, 3.17
6 mmHg) increase in diastolic BP per 20-ppb increase in 24-h avg NO2. Chuang et al.
7 (2010) used cross-sectional data from a national population-based health screening of
8 adults in Taiwan and reported finding no association between BP and NO2 levels,
9 although quantitative results were not presented. On the other hand, subsequently, Chen
10 et al. (2012c) used cross-sectional data from a different population-based health screening
11 in adults across 6 townships in Taiwan and found a 4.20 mmHg decrease (95% CI: -5.22,
12 -3.17 mmHg) in systolic BP per 20-ppb increase in 24-h avgNO2 at lag 3 and a 1.54
13 mmHg increase (95% CI: 0.75, 2.32 mmHg) in diastolic BP per 20-ppb increase in
14 24-h avg NO2 at lag 2. Choi et al. (2007) observed positive associations between NO2
15 concentrations and systolic BP during the warm and cold seasons at lags 0 and 1, though
16 the associations for diastolic BP were generally null.
17 One available study has examined the association between NO2 concentrations and other
18 markers of vascular function. In an analysis of data from the U.S. EPA's Detroit
19 Exposure and Aerosol Research Study (DEARS), Williams et al. (2012b) found that
20 personal NO2 concentrations were associated with changes in brachial artery diameter
21 (positive association at lag 1 and negative association at lag 2) and positive (i.e.,
22 presumably beneficial) changes in flow mediated dilation. No associations were observed
23 in relationship to ambient measures of NO2.
24 In summary, there is little evidence from available epidemiologic studies to suggest that
25 short-term exposure to ambient NO2 is associated with increased BP in the population
26 overall. One large, repeated-measures study among pregnant women found pronounced
27 decreases in systolic BP associated with ambient NO2 concentrations (Hampel et al..
28 2011). In a recent epidemiologic study, Williams et al. (2012b) did not clearly indicate
29 whether or not short-term exposure to NO2 is associated with other markers of vascular
30 function such as flow mediated dilation.
November 2013 4-210 DRAFT: Do Not Cite or Quote
-------
Table 4-27 Epidemiologic studies of blood pressure.
Study
Location
(Sample Size)
Mean NC>2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Williams et al. (2012b) Detroit, Ml
(n = 65)
24-h avg NO2: 24.0
75th: 28.0
Max: 100.0
No quantitative results presented
Huang et al. (2012a) Beijing, China
(n = 40)
2007, Visit 1: 33.8
2007, Visit 2: 26.3
2008, Visit 3: 29.2
2008, Visit 4: 22.9
Central Monitor
Change in SBP (mmHg)
30-min: 3.73 (-1.04, 8.28)
2-h: 0.00 (-5.89, 5.89)
12-h: 1.69 (-7.73, 11.11)
24-h:-2.32 (-19.10, 14.47)
Change in DBP (mmHg)
30-min: 2.28 (-1.86, 6.221)
2-h:-0.19 (-4.45, 6.00)
12-h: 2.90 (-5.07, 10.87)
24-h: 4.34 (-9.84, 18.52)
Richetal. (2012)
Beijing, China
(n = 125)
24-h avg NO2:
Entire study: 27.0
Before: 26.0
During: 13.9
After: 41. 4
Central Monitor No quantitative results presented; results presented
graphically. Generally inconsistent results with SBP, including
both statistically significant positive and negative associations
across lags. Generally null and inconsistent associations with
DBP across lags 0-6.
Hampeletal. (2011)
Nancy and Poitiers, 24-h avg NO2: 10.0
France 75th: 142
(n = 1,500) Max: 38.0
Residence within 20 Change in SBP (mmHg)
km of one of 28
permanent
background-
monitoring sites
Lag 0:-0.58 (-0.86,-0.14)
Lag 0-6 days: -0.72 (-1.14, 0.28)
Change in DBP (mmHg)
No quantitative results presented; "We detected no clear
associations between air pollutants and diastolic BP"
November 2013
4-211
DRAFT: Do Not Cite or Quote
-------
Table 4-27 (Continued): Epidemiologic studies of blood pressure.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Cakmaket al. (2011a) Canada
(n = 5,604)
24-havgNO2: 12.6
City-wide avg Change in resting SBP (mmHg)
LagO: 1.76(0.35, 3.17)
Change in resting DBP (mmHg)
Lag 0:2.11 (1.12, 3.10)
Chuanq et al. (2010)
Taiwan
(n = 7,578)
24-h avg NO2: 22.4
Max: 65.5
Nearest monitor
(within 10 km)
No quantitative results presented for NO2
Chenetal. (2012c)
Taiwan 24-h avg NO2: 13.9 to 26.1 Central Monitor
(n = 9,238) Max: 34.3 to 49.1
Change in SBP (mmHg)
LagO: -0.81 (-2.16, 0.55)
Lag 0-1:-1.17 (-2.34,-0.01)
Lag 0-2:-4.20 (-5.22,-3.17)
Change in DBP (mmHg)
LagO: 1.03(0.11, 1.95)
Lag 0-1: 1.54(0.75,2.32)
Lag 0-2:-0.01 (-0.71, 0.68)
Pulse Pressure Change
LagO:-2.55 (-3.62,-1.48)
Lag 0-1:-2.09 (-3.02,-1.18)
Lag 0-2: -3.22 (-4.04, -2.40)
November 2013
4-212
DRAFT: Do Not Cite or Quote
-------
Table 4-27 (Continued): Epidemiologic studies of blood pressure.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Choi et al. (2007)
Incheon,
South Korea
(n = 10,459)
24-h avg NO2:
Warm season: 22.5
75th: 26.9
Max: 49.3
Cool season: 29.2
75th: 34.7
Max: 74.0
City wide avg
Warm Season
Change in SBP (mmHg)
Lag 0: 2.24 (p = 0.002)
Lag 1:2.40(p<0.001)
Lag 2: -0.04 (p = 0.534)
Change in DBP (mmHg)
Lag 0: 2.02 (p = 0.645)
Lag 1: 2.12 (p = 0.016)
Lag 2: -0.04 (p = 0.331)
Cool Season
Change in SBP (mmHg)
Lag 0: 2.06 (p = 0.181)
Lag 1:2.06(p = 0.195)
Lag 2: -0.06 (p = 0.223)
Change in DBP (mmHg)
Lag 0: -0.02 (p = 0.573)
Lag 1: 2.00 (p = 0.445)
Lag 2: 2.02 (p = 0.445)
' Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration for 24- h and 1-h averaging times, respectively.
November 2013
4-213
DRAFT: Do Not Cite or Quote
-------
4.3.5.2 Controlled Human Exposure Studies
1 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) reviewed controlled human
2 studies of cardiac output or BP (Table 4-29). Several of these studies also examined HR
3 as described in Section 4.3.3.2. and similarly found no effects of NO2 exposure on
4 increasing cardiac output or BP in healthy adults or those with COPD. These endpoints
5 have not been evaluated in recent controlled human exposure studies of NO2.
6 Cardiac output is the volume of blood pumped out by each of the two ventricles per
7 minute. It is directly related to HR, as the output of each ventricle is the product of the
8 HR (beats/minute) and the stroke volume (mL of blood/beat). BP is the product of
9 cardiac output and vascular resistance. Cardiac output, vascular resistance, and BP
10 interact moment-to-moment to ensure systemic circulatory demands are met.
11 Folinsbee et al. (1978) exposed three groups of 5 young healthy adult males to 620 ppb
12 NO2 for 2 hours with intermittent exercise. The authors reported no changes in HR,
13 cardiac output, or BP. Drechsler-Parks (1995) exposed 8 older healthy adults to FA, 600
14 ppb NO2, 450 ppb O3, and NO2 + O3 for 2 hours with intermittent exercise. There was no
15 change in HR, stroke volume, or cardiac output following exposure to NO2 or O3 alone
16 compared to FA; however, a decrease in cardiac output was observed following NO2 +
17 O3 exposure compared to O3 and FA exposures (p <0.05). Gong et al. (2005) reported a
18 statistically nonsignificant increase in HR, but no change in BP after exposure to 400 ppb
19 NO2 for 2 hours with intermittent exercise in volunteers with COPD and healthy
20 volunteers.
21 Exposures to higher concentrations of NO2 have also been examined. Linn et al. (1985b)
22 reported a small, but statistically significant decrease in BP after exposure to 4,000 ppb
23 NO2 for 75 minutes with exercise. In both healthy volunteers and those with asthma, the
24 mean BP decrease was about 5 mmHg relative to controls.
25 The vascular endothelium plays a fundamental role in the maintenance of vascular tone
26 that is involved in the regulation of blood pressure and blood flow. Langrish et al. (2010)
27 examined the effects of NO2 on vascular endothelial tone and fibrinolytic function. In a
28 random crossover double-blind study, healthy male volunteers were exposed to 4,000 ppb
29 of NO2 for 1 hour with intermittent exercise. This study employed infusion of
30 endothelial-dependent vasodilators, bradykinin and acetylcholine, and endothelial-
31 independent vasodilators, sodium nitroprusside and verapamil, to examine vascular
32 endothelial tone. The results demonstrated that NO2 did not attenuate the vasodilator
33 response to these xenobiotics.
November 2013 4-214 DRAFT: Do Not Cite or Quote
-------
1 In summary evidence suggests NO2 does not alter cardiac output or vascular function.
2 Collectively, a few observations from epidemiologic and controlled human exposure
3 studies indicate NO2-associated decreases in BP. However, most do not suggest an
4 association between NO2 exposure and increased BP.
4.3.6 Blood Biomarkers of Cardiovascular Effects
5 Several epidemiologic and toxicological studies have explored the potential association
6 between NO2 and biomarkers of cardiovascular risk. In particular, markers of
7 inflammation, cell adhesion, coagulation, and thrombosis have been evaluated in a
8 number of epidemiologic studies published since the 2008 ISA for Oxides of Nitrogen
9 (U.S. EPA. 2008c) (Table 4-28). Such effects also have been examined in controlled
10 human exposure and animal toxicological studies.
4.3.6.1 Epidemiologic Studies
11 Levels of some circulating systemic inflammatory markers appear to be related to NO2
12 concentrations among participants with a history of heart disease. Delfino et al. (2008b)
13 followed nonsmoking elderly subjects with a history of coronary artery disease living in
14 retirement communities in Los Angeles, California and measured plasma biomarkers
15 weekly over a 12 week period. They found that indoor and/or outdoor NO2
16 concentrations measured at the retirement homes were associated with increases in
17 interleukin-6 (IL-6) and the soluble tumor necrosis factor a receptor II (sTNFa-RII),
18 markers of systemic inflammation, but not associated with a number of other biomarkers
19 of inflammation and vascular injury including C-reactive protein (CRP), P-selectin, D-
20 dimer, TNFa, soluble intercellular adhesion molecule-1 (sICAM-1), or soluble vascular
21 adhesion molecule-1 (sVCAM-1). In subsequent analysis, Delfino et al. (2009) and
22 Delfino etal. (2010) found that NO2 and NOX were both associated with circulating
23 levels of IL-6. Delfino et al. (2009) also observed positive associations with P-selectin,
24 TNF-RII, and CRP. Similarly, Ljungman et al. (2009) repeatedly measured plasma IL-6
25 in 955 myocardial infarction survivors from 6 European cities and found that NO2 was
26 associated with increased levels of IL-6, and that the strength of the association varied in
27 individuals with specific variants of inflammatory genes. However, in studies conducted
28 among patients with stable chronic heart failure, no associations were observed between
29 any biomarkers (including hematological markers and markers of inflammation) and NO2
30 concentrations (Barclay et al.. 2009; Wellenius et al.. 2007).
November 2013 4-215 DRAFT: Do Not Cite or Quote
-------
1 In Augsburg, Germany, Briiske et al. (2011) measured lipoprotein-associated
2 phospholipase A2 (Lp-PLA2), a marker of vascular inflammation and an independent
3 predictor of coronary heart disease events and stroke, up to 6 times in 200 participants
4 with a history of myocardial infarction. They found that Lp-PLA2 was associated with
5 both NO and NO2. However, the association was negative at short lags and positive at
6 longer lags, making interpretation of these results difficult.
7 The results have been more heterogeneous in participants without a history of heart
8 disease. Among elderly men participating in the Veterans Administration Normative
9 Aging Study, Bind etal. (2012) found that NO2 was associated with fibrinogen,
10 sVCAM-1, and sICAM-1, but not CRP. In this same cohort, Ren etal. (2011) found that
11 NO2 was positively linked with urinary 8-hydroxy-29-deoxyguanosine (8-OHdG)
12 concentrations, a marker of oxidative stress resulting in DNA damage. Thompson et al.
13 (2010) analyzed the baseline data on IL-6 and fibrinogen from 45 nonsmoking subjects
14 that participated in a controlled-exposure study in Toronto, Canada. Importantly, the
15 blood samples used in this study were collected before participants entered the exposure
16 chamber. They found that NO2 concentrations were not associated with either IL-6 or
17 fibrinogen overall, but IL-6 was associated with NO2 in the winter months. In Rotterdam,
18 the Netherlands, Rudez et al. (2009) measured CRP, fibrinogen, and markers of platelet
19 aggregation and thrombin generation up to 13 times in 40 healthy participants. Both NO2
20 and NO were associated with markers of platelet aggregation and thrombin generation,
21 but neither NO2 nor NO was associated with CRP or fibrinogen. Increases in NO2
22 concentrations during the Beijing Olympics were associated with increases in biomarkers
23 indicative of the thrombosis-endothelial dysfunction mechanism (i.e., sCD62P) among
24 healthy young adults (Rich etal.. 2012). Among 3,659 individuals in Tel-Aviv, Steinvil
25 et al. (2008) found null association between NO2 levels and CRP, and a negative
26 association with fibrinogen and white blood cell counts. Baccarelli et al. (2007) observed
27 generally null associations between NO2 concentrations and total homocysteine among
28 subjects in Lombardia, Italy. Similarly, Chuang et al. (2007a) observed no association
29 between NO2 and any blood markers, including markers of systemic inflammation and
30 oxidative stress, as well as fibrinolytic and coagulation factors.
31 Other subgroups that might be at increased risk of pollution-related adverse health effects
32 have also been studied. In a repeated-measures study of male patients with chronic
33 pulmonary disease in Germany, Hildebrandt et al. (2009) reported that NO was positively
34 associated with fibrinogen levels, but not other markers of coagulation, but detailed
35 results were not presented in the paper. In a cross-sectional analysis of pregnant women
36 in Allegheny County, PA, there was no association between NO2 and CRP (Lee et al..
37 201 Ic). Among 374 Iranian children aged 10-18 years, Kelishadi et al. (2009) found that
38 NO2 was associated with CRP and markers of oxidative stress.
November 2013 4-216 DRAFT: Do Not Cite or Quote
-------
1 In summary, there is some epidemiologic evidence to suggest the presence of an
2 association between NO2 concentrations and some markers of systemic inflammation
3 among participants with a history of heart disease. This association is not consistently
4 observed in healthy individuals. Other potentially at-risk populations have not been
5 clearly identified.
November 2013 4-217 DRAFT: Do Not Cite or Quote
-------
Table 4-28 Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Location
(Sample Size)
Mean NC>2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Delfino et al.
Los Angeles, CA
(n = 29)
1-h NO2
Outdoor:
33.1
Max: 59.8
Indoor:
32.3
Max: 53.5
Indoor and
outdoor home
measurements
Outdoor:
CRP (ng/mL)
Lag 0: 1,124.73 (-313.63, 2,565.25)
Lag 0-2: 1,027.40 (-465.03, 2,519.83)
Fibrinogen (ug/mL)
Lag 0: -110.31 (-503.97, 283.35)
Lag 0-2: -110.31 (-501.80, 281.18)
IL-6 (pg/mL)
LagO: 1.32(0.48,2.18)
Lag 0-2: 1.17(0.28,2.08)
IL-6R (pg/mL)
Lag 0: -493.15 (-9,387.17, -248.74)
Lag 0-2: -3,211.97 (-7,788.75, 1,364.82)
TNF-a (pg/mL)
LagO: 0.13 (-0.26, 0.52)
Lag 0-2: 0.15 (-0.22, 0.54)
TNF Rll (pg/mL)
Lag 0:289.83 (-41.10, 622.93)
Lag 0-2: 240.09 (-82.19, 562.36)
P-selectin (ng/mL)
LagO: 5.13 (-1.02, 11.27)
Lag 0-2: 1.49 (-5.04, 8.02)
VCAM-1 (pg/mL)
Lag 0: 53,733.96 (-11,381.40, 118,849.32)
Lag 0-2: 18,266.04 (-45,532.08,
82,062.00)
ICAM-1 (pg/mL)
Lag 0: 5,381.40 (-8,987.02, 19,747.66)
Lag 0-2: 575.34 (-13,494.59, 14,643.11)
SOD (U/g Hb)
Lag 0: -540.74 (-1,020.91, -62.73)
Lag 0-2: -571.02 (-1,036.05, -105.98)
GPx (U/g Hb)
LagO:-1.99 (-3.68,-0.26)
Lag 0-2: 1.15 (-2.81, 0.58)
MPO (ng/mL)
LagO:-5.34 (-14.92, 4.33)
Lag 0-2:-1.15 (-10.81, 8.44)
November 2013
4-218
DRAFT: Do Not Cite or Quote
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Delfino RJ:
Staimer et al.
(2009)
Delfino et al.
(2010)
Ljunqman et al.
(2009)
Location
Exposure
(Sample Size) Mean NO2 (ppb) assessment Selected Effect Estimates3 (95% Cl)
Los Angeles, CA 1-h NO2
(n-60) Phase 1:26.4
Phase 2: 28.3
1-h NOX
Phase 1: 37.2
Phase 2: 53.9
Los Angeles, CA Warm season
(n = 60) 1-h NO2: 26.4
1-h NOX: 37.2
Cool Season
1-h NO2: 28.3
1-h NOX: 53.9
Six European 24-h avg NO2
cities 22.6
(n = 955)
(total n = 5,539
measurements)
Hourly outdoor NOX:
home air .. „ . . . .
measurements IL-6
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Bruske et al.
(2011)
Bindetal. (2012)
Renetal. (2011)
Thompson et al.
(2010)
Location
(Sample Size)
Augsburg,
Germany
(n = 200)
Boston, MA
(n = 704)
Boston, MA
(n = 320)
Toronto, Canada
(n = 45)
Mean NO2 (ppb)
24-h avg NO2
20.8
75th: 24.7
Max: 38.2
24-h NO
24.0
75th: 25.8
Max: 141.1
24-h avg NO2
18
95th: 35
24-h avg NO2
17.8
24-h avg NO2
23.8
Exposure
assessment
Central site
monitor
City-wide avg
Central site
Monitor
Central site
monitor
Selected Effect Estimates3 (95% Cl)
Lp-PLA2 (% Change)
NO2 Lag 4: 7.28(3.00, 11.56)
NO Lag 4: 2.74 (-0.21, 5.70)
"Inverse associations were observed for ... NO2 with Lp-PLA-2 at lag days 1-2 and
positive associations were estimated ...with Lp-PLA2 lagged 4 and 5 days."
Fibrinogen (percent change)
Lag 0-2: 8.18(4.73, 11.64)
8-OHdG (% change) Lag 0-13: 166.88 (28.75, 305.63)
Lag 0: 28.48 (-19.39, 76.36) Lag 0-20: 195.15 (44.85, 344.85)
Lag 0-6: 90.00 (-12.22, 191.67)
Quantitative results not presented.
November 2013
4-220
DRAFT: Do Not Cite or Quote
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Location
(Sample Size)
Exposure
Mean NO2 (ppb) assessment
Selected Effect Estimates3 (95% Cl)
Rudez et al.
(2009)
Rotterdam,
the Netherlands
(n = 40)
24-h avg NO2
50th: 19.7
75th: 25.5
Max: 43.1
24-h NO:
50th: 5.6
75th: 12
Max: 130.4
Central site Maximal platelet aggregation
monitor (% change) NO; NO2
Lag 0-6 h: 5.42 (-18.33, 29.58);
-4.11 (-13.04,4.82)
Lag 0-12 h: 2.92 (-22.50, 28.33);
-4.64 (-15.00, 5.89)
Lag 0-24 h: 7.92 (-12.50, 28.75);
-5.36 (-18.39, 7.68)
Lag 24-48 h: 5.00 (-17.08, 27.08);
-1.07 (-11.79, 9.46)
Lag 48-72 h: 25.42 (10.00, 40.42);
10.00(2.68, 17.32)
Late aggregation (% change) NO; NO2
Lag 0-6 h: 33.75 (-5.00, 72.08);
5.89 (-9.46, 21.07)
Lag 0-12 h: 35.42 (-2.92, 73.33);
13.39 (-4.11, 30.71)
Lag 0-24 h: 37.08(4.67,69.17);
17.68 (-4.46, 39.82)
Lag 24-48 h: 22.92 (-6.25, 51.67);
3.39 (-16.07, 22.68)
Lag 48-72 h: 32.92 (9.58, 55.83);
15.89(4.64,27.14)
Lag 72-96 h: 14.17 (-23.75, 52.50);
8.57 (-7.68, 24.82)
Lag 0-96 h: 54.17 (20.42, 87.92);
28.75(8.93,48.57)
Thrombin generation - Peak (% change)
NO; NO2
Lag 0-6 h:-1.67 (-15.00, 11.67);
-2.68 (-9.82, 4.46)
Lag 0-12 h:-1.67 (-12.92, 9.58);
-1.25 (-9.11, 6.61)
Lag 0-24 h: -2.50 (-16.25, 10.83);
-1.07 (-9.46, 7.32)
Lag 24-48 h: 17.08 (4.58, 30.00);
14.29(4.29,24.29)
Lag 48-72 h: 5.00 (-6.67, 16.67);
6.61 (-2.68, 16.07)
Lag 72-96 h: 14.58 (1.67, 27.92);
-0.36 (-8.57, 7.86)
Lag 0-96 h: 12.92 (-7.08, 32.50);
1.79 (-7.32, 10.71)
November 2013
4-221
DRAFT: Do Not Cite or Quote
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Location
(Sample Size)
Exposure
Mean NO2 (ppb) assessment
Selected Effect Estimates3 (95% Cl)
Rudez et al.
(2009)
(Continued)
Rotterdam,
the Netherlands
(n = 40)
24-h avg NO2
50th: 19.7
75th: 25.5
Max: 43.1
24-h NO:
50th: 5.6
75th: 12
Max: 130.4
Central site Thrombin generation - ETP (% change)
monitor NO; NO2
Lag 0-6 h:-1.67 (-9.58, 6.25);
-2.14 (-6.43, 2.14)
Lag 0-12 h:-1.67 (-8.33, 4.58);
-0.36 (-5.00, 4.29)
Lag 0-24 h:-1.25 (-9.17, 6.67);
0.54 (-4.46, 5.54)
Lag 24-48 h: 7.92(0.42, 15.42);
6.25(0.36, 12.14)
Lag 48-72 h: 1.43 (-3.39, 6.25);
7.08 (-4.58, 18.75)
Lag 72-96 h: 8.75(0.83, 16.67);
1.43 (-3.39, 6.25)
Lag 0-96 h: 7.08 (-4.58, 18.75);
1.96 (-3.04, 7.14)
Thrombin generation - Lag time (%
change) NO; NO2
Lag 0-6 h: -0.42 (-5.83, 4.58);
0.00 (-2.86, 2.86)
Lag 0-12 h: 0.00 (-4.58, 4.17);
0.00 (-3.21, 3.04)
Lag 0-24 h: 2.50 (-2.50, 7.50);
0.36 (-2.86, 3.57)
Lag 24-48 h: -7.50 (-12.08, -2.92);
-5.54 (-9.11,-1.79)
Lag 48-72 h:-3.33 (-7.50, 1.25);
-4.46 (-7.68, 1.07)
Lag 72-96 h:-5.83 (-10.83, -0.83);
0.00 (-3.21, 3.04)
Lag 0-96 h:-4.58 (-11.67, 2.08);
-1.25 (-4.46, 1.96)
Fibrinogen - Lag time (% change)
NO; NO2
Lag 24-48 h: 0.42 (-4.17, 5.42);
0.71 (-3.04, 4.46)
Lag 48-72 h: 1.25 (-3.33, 5.83);
2.50 (-1.07, 6.07)
Lag 72-96 h: 0.42 (-4.58, 5.83);
-0.71 (-4.11,2.50)
CRP - Lag time (% change) NO; NO2
Lag 24-48 h: 15.00 (-12.08, 41.67);
11.61 (-8.75, 31.79)
Lag 48-72 hrs: 0.42 (-27.08, 27.92);
-0.18 (-19.64, 19.29)
Lag 72-96 hrs: -19.17 (-50.00, 12.08);
-12.32 (-30.71, 6.25)
November 2013
4-222
DRAFT: Do Not Cite or Quote
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Steinvil et al.
(2007)
Hildebrandt et al.
(2009)
Chuanq et al.
(2007a)
Wellenius et al.
(2007)
Baccarelli et al.
(2007)
Location
(Sample Size)
Tel Aviv, Israel
(n = 3,659)
Erfurt, Germany
(n = 38)
Taipei, Taiwan
(n = 76)
Boston, MA
(n = 28)
Lombardia, Italy
(n = 1,213)
Exposure
Mean NO2 (ppb) assessment
24-h avg NO2 City-wide avg
19.5
75th: 25.3
24-h avg NO2 Central Monitor
13.5
24-h NO
10.7
24-h avg NO2 Central Monitor
17.3
Max: 53.1
24-h avg NO2 City-wide avg
20.7
24-h avg NO2 City-wide avg
Median: 22.7
75th: 33.7
Max: 194.2
Selected Effect Estimates3 (95% Cl)
CRP (% change) Men; Women WBC (% change) Men; Women
Lag 0: 0.31 (-7.87, 12.60); Lag 0: 22.05 (-155.91, 200.00);
-4.72 (-17.32, 9.45) -83.46 (-305.51, 138.58)
Lag 1: -7.87 (-17.32, 9.45); Lag 1: 39.37 (-146.46, 223.62);
-3.15 (-15.75, 11.02) -20.47 (-244.09, 203.15)
Lag 2: -1.57 (-11.02, 11.02); Lag 2: -36.22 (-226.77, 154.33);
0.00 (-12.60, 15.75) 18.90 (-218.90, 255.12)
Fibrinogen (% change) Men; Women
Lag 0: -9.92 (-15.59, -4.25);
-12.44 (-19.84, -5.20)
Lag 1: -7.87 (-13.86, -2.05);
-5.51 (-12.91, 1.89)
Lag 2: -7.09 (-13.07, -1.10);
-1.42 (-9.45, 6.46)
Increases in fibrinogen and prothrombin fragment 1 + 2 associated with NO
concentrations. A decrease in vWF was associated with NO2 concentrations. No
quantitative results presented for NO or NO2
"There was no association between... NO2... and any of the blood markers"
No quantitative results presented
"No significant associations were observed between [NO2] and BNP levels at any of
the lags examined"
No quantitative results presented
Homocysteine difference (% change) Homocysteine, postmethionine-load
Lag 24 h: 0.24 (-2.86, 3.57) (% change)
Lag 0-6 days: -2.21 (-6.01, 1.72) La9 24 h: °-°° (-2-86> 2-86)
Lag 0-6 days: 0.49 (-2.94, 4.17)
November 2013
4-223
DRAFT: Do Not Cite or Quote
-------
Table 4-28 (Continued): Epidemiologic studies of biomarkers of cardiovascular effects.
Study
Location
(Sample Size)
Mean NO2 (ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Barclay et al.
(2009)
Aberdeen,
Scotland
(n = 132)
24-h avg NO2:
30.1
24-h avg NO:
14.7
Central monitor
Hemoglobin
NO2: 0.035 (-0.291, 0.361)
NO:-0.011 (-0.331, 0.310)
Mean Corpuscular hemoglobin
NO2: 0.050 (-0.158, 0.257)
NO:-0.039 (-0.243, 0.165)
Platelets
NO2: -0.049 (-0.867, 0.768)
NO: 0.247 (-0.556, 1.050)
Hematocrit
NO2:-0.017 (-0.350, 0.316)
NO: 0.101 (-0.226, 0.428)
WBC
NO2:-0.722 (-1.670, 0.226)
NO:-0.708 (-1.640, 0.224)
CRP
NO2: 0.423 (-5.263, 6.108)
NO: 0.890 (-4.694, 6.473)
IL-6
NO2: 6.276 (0.594, 11.940)
NO: 2.767 (-2.810, 8.344)
vWF
NO2: 2.164 (-0.328, 4.655)
NO: 3.522(1.091, 5.954)
E-selectin
NO2: 1.162 (-0.372, 2.696)
NO: 0.483 (-1.022, 1.989)
Fibrinogen
NO2:-0.219 (-1.759, 1.322)
NO: 0.195 (-1.320, 1.709)
Factor VII
NO2: 0.273 (-1.441, 1.987)
NO: 0.335 (-1.348, 2.018)
d-dimer
NO2:-0.243 (-2.781, 2.294)
NO:-0.316 (-2.807, 2.175)
Rich et al. (2012) Beijing, China
(n = 125)
24-h avg NO2
Entire study: 27.0
Before: 26.0
During: 13.9
After: 41.4
Central Monitor No quantitative results presented; results presented graphically. Positive and
statistically significant increase in sCD62P, generally consistent across lags from 0
to 6. Generally null associations with sCD40L across lags from 0-6. Positive and
statistically significant increases in vWF and fibrinogen at early lags (lag 0, lag 1)
but null, or negative at later lags. Generally null or negative associations with WBC
across lags 0-6.
aEffect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration for 24- h and 1-h averaging times, respectively.
November 2013
4-224
DRAFT: Do Not Cite or Quote
-------
4.3.6.2 Controlled Human Exposure Studies
1 Markers of inflammation, oxidative stress, cell adhesion, coagulation, and thrombosis
2 have been evaluated in a few controlled human exposure studies published since the 2008
3 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) (Table 4-29). Similar to epidemiologic
4 studies, controlled human exposure studies also report evidence for increases in some
5 inflammatory markers, but not consistently across all studies. There is also evidence for
6 hematological changes following NO2 exposure, and a recent study found endothelial cell
7 activation.
8 In healthy adults, 500 ppb NO2 for 2 hours with intermittent exercise did not alter
9 circulating IL-8, a pro-inflammatory cytokine, or coagulation factors, but induced a
10 statistically nonsignificant increase in IL-6 (Huang et al.. 2012b). Lipid profile changes
11 were also reported. There was a 4.1% increase in blood total cholesterol (p = 0.059) and
12 5.9% increase in high density lipoprotein (HDL) cholesterol (p = 0.036) 18 hours after
13 exposure, but no changes in low density lipoprotein or very low density lipoprotein
14 cholesterol or triglycerides.
15 The controlled human exposure study by Langrish et al. (2010) examined the effects of
16 NO2 on fibrinolytic function. The endogenous fibrinolytic pathway was assessed by
17 sampling venous concentrations of tissue-plasminogen activator and plasminogen-
18 activator inhibitor type 1 at baseline and 4 and 6 hours post-exposure. Concentrations of
19 these proteins were not affected by exposure to NO2.
20 Atherosclerosis is a chronic inflammatory disease. Early stages of the disease include
21 inflammatory activation of endothelial cells and adhesion of leukocytes to the vascular
22 endothelium. Channell et al. (2012) reported endothelial cell activation following NO2
23 exposure. Plasma samples were collected from healthy volunteers exposed to FA or 500
24 ppb NO2 for 2 hours with intermittent exercise. Primary human coronary artery
25 endothelial cells (hCAECs) were then treated with a dilution of these plasma samples (10
26 or 30% in media) for 24 hours. Expression levels of endothelial cell adhesion molecules,
27 VCAM-1 and ICAM-1, from hCAECs were elevated for both post-exposure time points
28 compared to control. hCAECs treated with plasma (30%) collected immediately post
29 NO2 exposure had significantly greater release of IL-8, but not monocyte chemoattractant
30 protein-1 (MCP-1). In addition, plasma collected 24 hours post NO2 exposure had a
31 significant increase (30%) in soluble lectin-like oxLDL receptor (LOX-1) levels, a
32 protein recently found to play a role in the pathogenesis of atherosclerosis.
33 Riedl et al. (2012) reported on the cardiovascular effects of healthy volunteers and
34 individuals with asthma exposed to FA, diesel exhaust, or 350 ppb NO2 for 2 hours with
November 2013 4-225 DRAFT: Do Not Cite or Quote
-------
1 intermittent exercise. No statistically significant differences were found in IL-6, ICAM-1,
2 and blood coagulation factors, i.e., factor VII, fibrinogen, and von Willebrand factor
3 (vWF), the morning after NO2 exposure.
4 Studies from the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) reported
5 NO2-induced hematological changes. Frampton et al. (2002) reported decreases in
6 hematocrit, hemoglobin, and red blood cell count in healthy volunteers 3.5 hours after
7 exposure to 600 and 1,500 ppb NO2 for 3 hours with intermittent exercise. Results from
8 this study supports those of Posin et al. (1978). in which hematocrit and hemoglobin
9 levels were decreased in young males exposed to 1,000 and 2,000 ppb NO2 for 2.5-3
10 hours with intermittent exercise. However, a recent study reported no change in
11 hemoglobin levels 4 and 6 hours after exposure to 4,000 ppb NO2 for 1 hour (Langrish et
12 al..201Q).
13 To summarize results for biological markers of cardiovascular effects, also discussed in
14 Section 3.3.2.8. the few available controlled human exposure studies from the 2008 ISA
15 for Oxides of Nitrogen (U.S. EPA. 2008c) demonstrated that short-term NO2 exposure
16 causes a slight reduction in hematocrit and hemoglobin levels associated with a decrease
17 in RBC levels. The clinical significance of these findings is unknown (Section 3.3.2.8).
18 The recent available studies demonstrate that NO2 does not affect all measured
19 cardiovascular biomarkers. For instance, evidence has not shown NO2 to alter circulating
20 blood coagulation factors or modify the body's response to vasodilators. However, some
21 evidence suggests NO2 exposure increases inflammatory mediators and induces
22 endothelial cell activation, which has been linked to risk of atherosclerosis.
November 2013 4-226 DRAFT: Do Not Cite or Quote
-------
Table 4-29 Controlled human exposure studies of short-term NO2 exposure and
cardiovascular effects.
Study
Lifestage; Sex; n
Exposure Details
(Concentration; Duration)
Endpoints Examined
Channell et al.
(2012)
Adult
(25.3 ± 5.5 yr); M/F;
n = 7
Primary human
coronary artery
endothelial cells
(hCAECs)
Adults were exposed to 500 ppb
NO2; 2 h; intermediate
intermittent exercise (15 min
on/off; VE = 25 L/min per m2 of
BSA [body surface area]). Plasma
samples were collected before
exposures, immediately after, and
24-h post-exposure. hCAECs
were treated with a dilution of
these plasma samples (10 or 30%
in media) for 24 h.
LOX-1 protein measured from plasma
pre, immediately post, and 24-h post-
exposure ICAM-1 and VCAM-1 mRNA
from hCAECs and IL-8 and MCP-1
protein from cell supernatant
measured immediately post-exposure
to plasma.
Drechsler-Parks
(1995)
Folinsbee et al.
(1978)
Frampton et al.
(2002)
Gonq et al.
(2005)
Huanq et al.
(2012b)
Adult
(65.9±9yr);M/F;
n = 8
Adult (20-25 yr); M;
n = 5/group
Adult;
M (26.9 ± 4.5 yr
n = 12);
F(27.1 ±4.1 yr;
n = 9)
Elderly;
Healthy
nonsmokers;
68 ± 1 1 yr; n = 6;
Ex-smokers with
COPD; 72 ± 7 yr;
n = 18
Adult; M/F
(24.56 ± 4.28 yr);
n = 23
600 ppb; 2 h; intermittent exercise
(20 min on/off)
VE = 26-29 L/min
600 ppb; 2 h; exercise (15, 30, or
60 minutes; VE = 33 L/min)
600 and 1,500 ppb; 3 h;
intermittent exercise (10 min
on/20 min off); VE = 40 L/min
400 ppb NO2; 2 h; intermittent
exercise (15 min on/off);
VE = 22-26 L/min
500 ppb NO2 and
500 ppb NO2 + 73.4 ± 9.9 ug/m3
CAPs; 2 h; intermittent exercise
(15 min on/off); VE = 25 L/min per
m2 of BSA
HR was calculated throughout
exposure, cardiac output was
measured during the last two min of
each exercise period
HR, BP, and cardiac output were
measured during exposure
Venous blood collected for
hematocrit, hemoglobin, and red
blood cell count 3.5 h after exposure
HR and BP were measured
immediately post, 4-h post, and day 2
IL-6, coagulation factors, and lipid
panel in peripheral blood; HRV; and
HR measured 1 and 18 h after
exposure
November 2013
4-227
DRAFT: Do Not Cite or Quote
-------
Table 4-29 (Continued): Controlled human exposure studies of short-term NO2 exposure
and cardiovascular effects.
Study
Lifestage; Sex; n
Exposure Details
(Concentration; Duration)
Endpoints Examined
Lanqrish et al.
(2011)
Adult; M; (median
age24yr); n = 10
4,000 ppb NO2; 1 h; intermittent
exercise; (VE = 25 L/min)
4 h after exposure 5,10, and 20
ug/min acetylcholine; 100, 300,
and 1,000 pmol/min bradykinin;
2,4, and 8 ug/min sodium
nitroprusside; 10, 30, and 100
ug/min verapamil were infused in
the brachial artery for 6 min/dose
during forearm venous occlusion
plethysmography. Each
vasodilator administration was
separated by a 20 min washout
period.
Hemoglobin concentration was
measured 4 and 6 h after exposure,
Forearm blood flow and tissue-
plasminogen activator and
plasminogen-activator inhibitor type 1
were measured 4 h after exposure.
Linn et al.
Adults; M/F
w/ asthma
(18-34yr)n = 23,
w/o asthma
(20-36 yr) n = 25
3,850-4,210 ppb NO2; 75 min;
intermittent exercise; light and
heavy exercise VE = 25 and
50 L/min (15 min of each; light
minute ventilation 25 L/min and
heavy minute ventilation 50 L/min)
HR and BP was measured throughout
exposure
Posin et al. Adult; NR; NR; 1,000 or 2,000 ppb NO2; 2.5 h;
(1978) n = 8-10 light intermittent exercise (15 min
on/off)
Acetylcholinesterase, glutathione,
glucose-6-phosphate dehydrogenase,
lactate dehydrogenase, erythrocyte
glutathione reductase, erythrocyte
glutathione peroxidase, alpha-
tocopherol, TEARS, serum
glutathione reductase, 2,3
diphospoglycerate, hemoglobin,
hematocrit
Riedl et al.
Adult; M/F
(1) 37.33 ±
10.91 yr;
n=10M, 5F
(2) 36.13 ± 2.52 yr;
n=6M, 9F
(1-2)350ppbNO2;2h;
intermittent exercise (15 min
on/off); VE = 15-20 L/min*m2 BSA
(1) Methacholine challenge after
exposure
(2) Cat allergen challenge after
exposure
Serum levels of IL-6, ICAM-1,
fibrinogen, factor VII, and vWF.
Serum collected 22.5 h after
exposure.
Scaife et al.
Adult (median age
68 yr); with stable
coronary heart
disease or impaired
left ventricular
systolic function;
M/F; n = 18
400 ppb NO2; 1 h
HR and HRV monitored continuously
for 24 h after exposure.
November 2013
4-228
DRAFT: Do Not Cite or Quote
-------
4.3.6.3 Toxicological Studies
1 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) reported on various
2 hematological parameters in animals including oxidative stress, RBC turnover, and
3 methemoglobin levels. Similar to epidemiologic and controlled human exposure studies,
4 several recently published toxicological studies have examined the potential association
5 between short-term NO2 exposure and biomarkers of cardiovascular effects, including
6 markers of oxidative stress, inflammation, and cell adhesion (Table 4-30).
7 Recently, the effects of NO2 on markers of oxidative stress were examined by Li et al.
8 (201 la). Rats exposed to 2,660 or 5,320 ppb NO2 for 7 days had a small, but statistically
9 significant decrease in the activity of the antioxidant enzyme Cu/Zn-SOD and, at the
10 higher dose, an increase in MDA, an indicator of lipid peroxidation, in heart tissue. These
11 changes were accompanied by mild pathological changes in the heart. However, there
12 were no changes in Mn-SOD or GSH peroxidase activity or protein carbonyl (PCO)
13 levels at either exposure concentration. Campen et al. (2010) reported Apolipoprotein E
14 knockout mice (ApoE"7") exposed to 200 and 2,000 ppb NO2 had a concentration-
15 dependent decrease (significant linear trend) in the expression of the antioxidant enzyme
16 HO-1 in the aorta. Together these results demonstrate the ability of NO2 inhalation to
17 perturb the oxidative balance in the heart.
18 The effects of NO2 on antioxidant capacity were also examined in the context of diet (de
19 Burbure et al.. 2007). Rats were placed on low (Se-L) or supplemented (Se-S) selenium
20 (Se) diets and were exposed to 5,000 ppb NO2 for 5 days. Se is an integral component of
21 the antioxidant enzyme GSH peroxidase. GSH peroxidase levels in RBCs increased in
22 both groups immediately and 48 hours after exposure; however, plasma levels were
23 decreased in Se-L rats at both time points. RBC SOD activity also decreased in Se-L rats
24 at both time points, but increased in Se-S rats 48 hours after exposure. Overall, NO2
25 exposure stimulated oxidative stress protective mechanisms with high Se, but were mixed
26 with low Se.
27 The effects of NO2 on endothelial mediators, endothelin-1 (ET-1) and endothelial nitric
28 oxide synthase (eNOS) were recently examined in two studies. ET-1 is a potent
29 vasoconstrictor while the enzyme eNOS catalyzes the production of NO, which induces
30 vasodilation. Campen et al. (2010). described above, did not see a significant increase in
31 ET-1 expression level in the aorta after exposure of rats to 200 and 2,000 ppb NO2.
32 However, with exposure to higher NO2 concentrations, ET-1 increased significantly in
33 the heart at the mRNA (10,640 ppb) and protein level (5,320 and 10,640 ppb) (Li et al..
34 201 la). Furthermore, eNOS mRNA and protein levels were increased at both the 2,660
November 2013 4-229 DRAFT: Do Not Cite or Quote
-------
1 and 5,320 ppb doses and decreased to control levels at the 10,640 ppb dose. At more
2 relevant concentrations there was an increase in eNOS, while higher concentrations
3 elicited an increase in the vasoconstrictor, ET-1.
4 Studies have also reported changes in some inflammatory markers and adhesion
5 molecules after NO2 exposure in animals. Li etal. (201 la) observed a significant increase
6 in TNF mRNA levels in the heart at 5,320 ppb NO2. In addition, IL-1 expression and
7 protein levels were increased; however, this effect was in response to a higher NO2
8 concentration. ICAM-1 transcription and protein levels were increased in the heart after
9 both the 2,660 and 5,320 ppb NO2 exposures. These results are consistent with those
10 from a controlled human exposure study (Channell et al.. 2012) described in Section
11 4.3.6.2.
12 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) reported on several animal
13 studies examining hematological parameters. Three studies indicate elevated levels of a
14 younger population of RBC following NO2 exposure. RBC D-2,3-diphosphoglycerate
15 levels, important in hemoglobin-oxygen dissociation, were increased in guinea pigs
16 following a 7 day continuous exposure to 360 ppb NO2 (Mersch et al., 1973). Kunimoto
17 etal. (1984) reported an increase in RBC sialic acid after 24 hours of exposure to 4,000
18 ppb NO2. Similarly, Mochitate and Miura (1984) reported an elevation of the glycolytic
19 enzymes pyruvate kinase and phosphofructokinase after a 7 day continuous exposure to
20 4,000 ppb NO2. These results suggest an increase in RBC turnover after NO2 exposure.
21 Nakajima and Kusumoto (1968) reported that mice exposed to 800 ppb NO2
22 continuously for 5 days had no change in the oxygen-carrying metalloprotein
23 hemoglobin, methemoglobin.
24 In summary, there is some evidence from a few animal toxicological studies that short-
25 term NO2 exposure affects the cardiovascular system. Oxidative stress effects of NO2
26 were evident in RBC, the heart, and aorta of rodents. In addition, an increase in
27 inflammatory markers and adhesion molecules was also observed after exposure to NO2.
28 At higher concentrations, NO2 was found to induce the expression and production of the
29 vasoconstrictor ET-1. Other effects included changes in hematological parameters.
November 2013 4-230 DRAFT: Do Not Cite or Quote
-------
Table 4-30 Animal toxicological studies of short-term NO2 exposure and
cardiovascular effects.
Study
Campen et al.
(2010)
de Burbure et al.
(2007)
Kunimoto et al.
(1984)
Lietal. (2011 a)
Species
(Strain);
Lifestage;
Sex; n
Mice (ApoE"'");
8 weeks; M;
n = 5-10/group
Rats (Wistar);
8 weeks; M;
n = 8/group
Rats (Wistar);
16-20 weeks;
M; n = 6/group
Rats (Wistar);
Adults; M;
n = 6/group
Exposure Details
(Concentration; Duration)
High fat diet; 200 ppb, 2,000
ppb NC>2; 6 h/day for 7 days;
High (6 ug/day) or low
(1.3 ug/day) selenium;
1,000 ppb NO2, 28 day,
6 h/day, 5 days/week (Se+/Se-);
10,000 ppb NO2, 28 day,
6 h/day, 5 days/week;
5,000 ppb NO2, 5 days,
6 h/day;
50,000 ppb, 30 min
4,000 ppb NO2; continuously for
1-10 days
2,660, 5,320, and 10,640 ppb
NO2; 6 h/day for 7days
Endpoints Examined
ET-1, MMP-9, HO-1, and TIMP-1 mRNA
expression in aorta; TEARS in aorta;
MMP-2/9 activity in aorta. Endpoints
measured 18 h after exposure.
GPx in plasma and red blood cell lysate;
SOD activity in red blood cell lysate; GST
activity in red blood cell lysate; TEARS in
plasma. Endpoints examined immediately
and 48 h after exposure.
ATPase activity, sialic acid, and hexose in
red blood cell membranes were measured
after 1,4,7, and 1 0 days of exposure.
H&E staining of heart tissue; Cu/Zn-SOD,
Mn-SOD activity, GPx activity, MDA level,
and PCO level in heart tissue; ET-1,
eNOS, TNF-a, IL-1, and ICAM-1 mRNA
and protein levels in heart tissue; cardiac
myocyte apoptosis. Endpoints examined
18 h after exposure.
Mersch et al. Guinea pigs; 360 ppb NO2; continuously for 7
(1973) NR; n = 8 days
D-2,3-diphosphoglycerate content in red
blood cells; collection time NR.
Mochitate and
Miura(1984)
Rats (Wistar);
16-20 weeks;
M; n = 6
4,000 ppb NO2; continuously for
1-10 days
PK and PFK activity and hemoglobin
content in red blood cells was measured
after 1, 3, 5, 7, and 10 days of exposure.
Nakajima and
Kusumoto (1968)
Mice(ICR);
4 weeks; M;
n = NR
800 ppb NO2; continuously for 5 Meta-hemoglobin in blood from the heart
days
taken immediately after exposure.
Ramos-Bonilla et
al. (2010)
Mice (AKR/J); Low-pollution chamber
180 days; M; (21.2 ppb NO2, 465 ppb CO,
n = 3/group 11.5 ug/m3 PM);
High-pollution chamber
(36.1 ppbNO2, 744 ppb CO,
36.7 ug/m3 PM);
6 h/day, 5 days/week, 40 weeks
ECG (HR, SDNN, r-MSSD, TP, LF, HF,
LH:HF), BW;
Endpoints measured throughout the
exposure.
November 2013
4-231
DRAFT: Do Not Cite or Quote
-------
4.3.7 Hospital Admissions and Emergency Department Visits
4.3.7.1 All Cardiovascular Diseases
1 Many epidemiologic studies consider the composite endpoint of all cardiovascular
2 diseases, which typically includes all diseases of the circulatory system (e.g., heart
3 diseases and cerebrovascular diseases). Most studies reviewed in the 2008 ISA for
4 Oxides of Nitrogen found positive associations between ambient NO2 concentrations and
5 risk of hospital admissions or ED visits for all cardiovascular diseases (U.S. EPA. 2008c)
6 (Figure 4-12 and Table 4-31). However, it was unclear at that time whether these results
7 truly indicated effects of NO2 or were confounded by other correlated pollutants. Several
8 additional studies are now available with broadly consistent results.
9 Ito et al. (2011) found that risk of CVD hospitalization was associated with NO2
10 concentrations at lag 0 in New York City. Results from copollutant models were not
11 reported. In Beijing, China, Guo et al. (2009) found an association between ambient NO2
12 concentrations and risk of CVD hospital admissions at lag 0, but this association was
13 attenuated and less precise in copollutants models adjusting for either PM2 5 or SO2, or
14 both. In Shanghai, China, Chenetal. (2010b) found a 1.02% (95% CI: -2.0%, 4.0%)
15 increased risk of hospital admission for CVD per 20-ppb increase in 24-h avg NO2
16 concentrations (lag 0-1 days). This association was robust to additional adjustment for
17 PM10, but was attenuated after adjustment for SO2. A study in Sao Paulo, Brazil, also
18 found a positive association with some evidence that the association was stronger among
19 patients with a secondary diagnosis of diabetes mellitus (Filho et al., 2008). Studies from
20 Copenhagen, Denmark (Andersen et al.. 2008b). Madrid, Spain (Linares and Diaz. 2010).
21 and Taipei, Taiwan (Chan et al., 2008) reported null or negative associations between
22 NO2 concentrations and risk of hospital admission for CVD.
23 In summary, consistent evidence reported in the 2008 ISA for Oxides of Nitrogen
24 combined with recent epidemiologic data available continues to support the presence of
25 an association between ambient NO2 levels and risk of hospital admission for
26 cardiovascular diseases (Figure 4-12 and Table 4-31). Generally, the associations
27 observed in these studies are robust in copollutant models that adjust for PM or gaseous
28 pollutants.
November 2013 4-232 DRAFT: Do Not Cite or Quote
-------
Mean
Muay con
Guoet al. 2009
Chen etal. 2010
Itoetal. 2011
Son etal. 2013
Larrieuetal. 2007
Polonieckietal.1997
Chang etal. 2005
Yang etal. 2004
Linn etal. 2000
Wong etal. 1999
von Klot et al. 2005
Andersen ei ai. zuuo
Hinwood etal. 2006
Llorcaetal. 2005
Filhoetal. 2008
Atkinson etal. 1999
Peel etal. 2007
Tolbertetal. 2007
Funget al.2005
Jalaludin etal. 2006
Simpson et al. 2005
Ballesteretal. 2006
Morgan etal. 1998
centration
36.3
30.3
28.7
17.9-26.8
11.9-24.9
35
31.5
28.17
28-41
27.3
101 3 "7 0
lz.1 - 3 / .2.
10.3
11.3
9.76
61.1
50.3
45.9
43.2
38.9
23.2
16.3-23.7
12.4-40.5
61.8
15
29
Lag
0
1
0
1
0-1
0
1
0
1
1
0-2
0-2
0
0-1
00
-3
i
0
i
0-1
0
0-1
0
0-2
0-2
0-2
0
0-1
0
0-1
0
0-1
1
P,.,
0-1
0-1
0
0
0
Notes
—1
>20C
25+ C
<25C
> 35 yrs; Ml
survivors
All Ages
N02
NO
diabetic
diabetic
diabetic
non-diabetic
non-diabetic
non-diabetic
case-crossover
time series
-. f~ 1- . ... - m
65 yrs -
> 65 yrs —
> 65 yrs
> 65 yrs
> 65 yrs
24-h
1-h
24-h
24-h
-• —
0_
•—
•
i
+.
•
•-
••
2.54(2.27,2.84) >
•
-»-
-+-
— 9
— A
1-h
i
»
_»_
•*•
••
_
• —
•
_•_
»-
«-
•*•
24-h and 1-h
-• —
-•-
1-h -•-
0.5 0.75 1 1.25
Relative Risk(95% CD
1.5
1.75
Note: Studies in red are recent studies. Studies in black were included in the 2008 ISA for Oxides of Nitrogen. Relative risks are
standardized to a 20 ppb or 30-ppb increase in NO2 or NO concentration for 24-h and 1-h averaging times, respectively. Studies are
presented in descending order, with the largest mean concentration (ppb) at the top and the smallest at the bottom of the figure (by
averaging time and inclusion in previous ISA). Circles = NO2; Triangles = NO.
Figure 4-12 Results of studies of short-term exposure to oxides of nitrogen
and hospital admissions for all cardiovascular disease.
November 2013
4-233
DRAFT: Do Not Cite or Quote
-------
Table 4-31 Corresponding effect estimates for hospital admissions for all
cardiovascular disease studies presented in Figure 4-12.
Study
Guo et al. (2009)
Chenetal. (201 Ob)
Itoetal. (2011)
Sonetal. (2013)
Larrieu et al. (2007)
Poloniecki et al. (1997)
Chang et al. (2005)
Yang et al. (2004)
Linn et al. (2000)
Wongetal. (1999)
Von Klot et al. (2005)
Andersen et al. (2008b)
Hinwood et al. (2006)
Llorca et al. (2005)
Filho et al. (2008)
Atkinson et al. (1999)
Peel et al. (2007)
Tolbert et al. (2007)
Location
Beijing, China
Shanghai, China
New York City, NY
8 Korean Cities
8 French Cities
London, U.K.
Taipei, Taiwan
Kaohsiung, Taiwan
Los Angeles, CA
Hong Kong, China
Five European Cities
Copenhagen, Denmark
Perth, Australia
Torrelavega, Spain
Sao Paulo, Brazil
London, U.K.
Atlanta, GA
Atlanta, GA
Relative Risk3 (95% Cl)
Lag 0: 1.05(1.00, 1.11)
Lag 1: 1.03(0.985, 1.09)
LagO: 0.997(0.970, 1.025)
Lag 1:1.02(0.99, 1.05)
Lag 0-1: 1.01 (0.98, 1.04)
Lag 0: 1.04(1.03, 1.05)
Lag 1: 1.01 (1.00, 1.02)
LagO: 1.04(1.02, 1.06)
Lag 1: 1.00(0.98, 1.01)
1.02(1.00, 1.04)
Lag 1: 1.02(1.00, 1.04)
>20°C: 1.39(1.32, 1.45)
<20°C: 1.23(1.12, 1.37)
>25°C: 1.46(1.31, 1.62)
<25 °C: 2.54 (2.27, 2.84)
Lag 0: 1.03(1.02, 1.04)
Lag 0-1: 1.05(1.03, 1.08)
Lag 0: 1.16(1.07, 1.27)
Lag 0-3: 1.00(0.93, 1.10)
Lag 1: 1.08(1.04, 1.13)
NO2: 1.11 (1.05, 1.17)
NO: 1.13(1.07, 1.19)
Diabetics
LagO: 1.00(1.00, 1.00)
Lag 1: 1.00(0.99, 1.00)
Lag 0-1: 1.00(1.00, 1.00)
Non-diabetics
Lag 0: 1.00(1.00, 1.00)
Lag 1: 1.00(0.99, 1.00)
Lag 0-1: 1.00(1.00, 1.00)
Lag 0: 1.01 (1.00, 1.02)
Case-crossover; lag 0-2: 1.04 (1.02, 1.06)
Time-series; lag 0-2: 1.04 (1.02, 1.06)
Lag 0-2: 1.02(1.01, 1.03)
November 2013
4-234
DRAFT: Do Not Cite or Quote
-------
Table 4-31 (Continued): Corresponding effect estimates for hospital admissions for all
cardiovascular disease studies presented in Figure 4-12.
Study
Location
Relative Risk3 (95% Cl)
Fung et al. (2005)
Windsor, Ontario, Canada
<65yr
Lag 0: 0.99(0.90, 1.13)
Lag 0-1: 1.04(0.93, 1.16)
>65yr
LagO: 1.02(0.96, 1.07)
Lag 0-1: 1.02(0.98, 1.05)
Jalaludin et al. (2006)
Sydney, Australia
Lag 0: 1.06(1.02, 1.09)
Lag 1: 1.04(1.01, 1.08)
Lag 0-1: 1.01 (0.98, 1.05)
Simpson et al. (2005a) 4 Australian Cities
Lag 0: 1.07(1.05, 1.09)
Lag 0-1: 1.07(1.05, 1.09)
Ballester et al. (2006) Spain
24-hNO2, lagO: 1.01 (1.00, 1.03)
1-hNO2, lagO: 1.04(0.99, 1.09)
Morgan et al. (1998) Sydney Australia
24-hNO2, lagO: 1.09(1.06, 1.12)
1-hNO2, lagO: 1.06(1.04, 1.09)
Note: Studies correspond to studies presented in Figure 4-12.
"Relative Risks are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration
for 24- h and 1-h averaging times, respectively.
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
4.3.7.2 Cardiac Causes (Ml and Heart Failure)
The 2008 ISA for Oxides of Nitrogen found that the epidemiologic evidence consistently
supported the associations between short-term changes in NO2 concentrations and
hospital admissions or ED visits for cardiac diseases (U.S. EPA. 2008c). This hypothesis
continues to be supported by studies published since the 2008 ISA, as reviewed below
(Figure 4-13 and Table 4-32).
Generally, studies based on clinical registries are less susceptible to misclassification of
the outcome and exposure which may explain why they provide stronger evidence than
those based on administrative data. The most convincing evidence of an association
between ambient NO2 and risk of myocardial infarction (MI) comes from a study using
clinical registry data from the U.K.'s Myocardial Ischaemia National Audit Project
(Bhaskaran et al.. 2011). which found a 5.8% (95% CI: 1.7%, 10.6%) increase in risk of
MI per 30-ppb increase in 1-h max NO2 concentrations in the 6 hours preceding the
event. This study is unique because it included detailed data on the timing of MI onset in
more than 79,000 patients from 15 conurbations in England and Wales, which allowed
examination of association with ambient NO2 in the hours preceding MI. The association
with NO2 was strengthened in a multipollutant model adjusted for PMi0, CO, SO2, and
November 2013
4-235
DRAFT: Do Not Cite or Quote
-------
1 O3; however, multipollutant model results are more difficult to interpret because of
2 multicollinearity among pollutants. NO2 results were robust to a number of sensitivity
3 analyses that evaluated key aspects of study design and model specification (e.g., stricter
4 diagnosis criteria, different time strata). The findings for NO2 were more pronounced in
5 those aged between 60 and 80 years, among those with prior coronary heart disease, and
6 for events occurring in the autumn and spring. On the other hand, in a study of 429 MI
7 events, Turin et al. (2012) did not observe any association using data from the Takashima
8 County Stroke and AMI Registry in Central Japan.
9 A number of studies based on administrative data have also been published since the
10 2008 ISA for Oxides of Nitrogen. Using data from 14 hospitals in 7 Canadian cities,
11 Stieb et al. (2009) found a 3.0% (95% CI: 0.2%, 5.8%) increase in risk of ED visits for
12 the composite endpoint of angina or acute MI per 20-ppb increase in 24-h avg NO2 on
13 the previous day. However, the overall association was dominated by the association
14 observed in Edmonton, and exclusion of the data from Edmonton from analyses
15 attenuated the results. A related study in 6 Canadian cities found that NO2 concentrations
16 were associated with risk of ED visits for chest pain (Szyszkowicz. 2009). Larrieu et al.
17 (2007) observed a positive association between hospital admissions for IHD and NO2
18 concentrations in 8 French cities. The magnitude of the association was higher for older
19 adults (i.e., < 65 years) than for the general population. In 6 areas in central Italy,
20 Nuvolone et al. (2011) found a 8% (95% CI: 0, 15%) increase in risk of hospitalization
21 for MI per 20-ppb increase in 24-h avg NO2 on the previous day. Similar associations
22 were seen in relation to lags 1 to 4 days prior to hospital admission. The finding at lag 2
23 was robust to adjustment for PMi0 in a two-pollutant model but remained positive,
24 though somewhat attenuated, by adjustment for CO. The association with NO2 was
25 somewhat more pronounced among females and in the cold season. Szyszkowicz (2007b)
26 and Thach et al. (2010) found that NO2 was associated with increased risk of hospital
27 admission for IHD in Montreal, Canada, and Hong Kong, China, respectively.
28 In New Jersey, Rich et al. (2010) found a relative risk of 1.14 (95% CI: 0.96, 1.32) per
29 20-ppb increase in 24-h avg NO2 for hospitalization for transmural Mis. No results were
30 reported for all Mis or for non-transmural infarcts. Wichmann et al. (2012) found that
31 NO2 was positively associated with risk of hospital admission in Copenhagen, Denmark,
32 but only in the warm months of the year. NO2 was positively associated with hospital
33 admissions for MI (Hsieh et al.. 2010: Bell et al.. 2008) and for IHDs (Bell et al.. 2008) in
34 Taipei, Taiwan, and risk of hospital admissions for MI in Kaohsiung, Taiwan (Cheng et
35 al.. 2009a). The study by Bell et al. (2008) used three different exposure assessment
36 techniques aimed at reducing uncertainty related to the use of central site monitors. NO2
37 was not associated with risk of hospital admission for acute coronary syndrome in
38 Lithuania (Vencloviene et al.. 2011).
November 2013 4-236 DRAFT: Do Not Cite or Quote
-------
1 Two additional studies have considered hospital admissions or ED visits for heart failure.
2 In the study of 7 Canadian cities described above, Stieb et al. (2009) found a 5.1% (95%
3 CI: 1.3%, 9.2%) increase in risk of ED visits for heart failure per 20-ppb increase in
4 24-h avg NO2. In Taipei, Taiwan, Yang (2008) found that risk of hospital admission for
5 heart failure were associated with NO2 concentrations, but only on days where the mean
6 ambient temperature was > 20 °C.
7 In summary, the epidemiologic data available continue to support associations between
8 ambient NO2 concentrations and risk of hospital admission or ED visits for cardiac
9 causes, particularly MI and IHD. Generally, the associations observed in these studies are
10 robust in copollutant models that adjust for PM or other gaseous pollutants.
November 2013 4-237 DRAFT: Do Not Cite or Quote
-------
Study
Thach etal. 2010
Hsieh etal. 2010
Cheng et al. 2009
Bell etal. 2008
Szyszkowicz2009
Szyszkowicz 2007
Stieb etal 2009
Vencloviene et al 2011
Turin etal. 2012
Nuvol one etal. 2011
Wichmannetal 2012
Larrieuet al. 2007
Rich etal 2010
Mann etal. 2002
Polonieckietal. 1997
Linn etal. 2000
Wong etal. 1999
Ponka and Virtanen 1996
Barnett etal. 2006
von Klot et al 2005
Bhaska ran etal. 2011
Atkins on etal. 1999
Peel etal. 2007
Jalaludin etal. 2006
Simpson etal. 2005
Outcome
IHD
Ml
Ml
Ml
IHD
Angina
IHD
Ml
Ml
Ml
Ml
Ml
IHD
Ml
IHD
Ml
IHD
Ml
Angina
Ml
IHD
IHD
IHD
Ml
Ml
ngma
Ml
IHD
IHD
IHD
IHD
Concentration
31.2
29.88
-ICC
ZD.D
26.4
20.1
19.4
18.4
18 4
16
15.2-21.1
12
11.9-24.9
37.2
35
28-41
27.3
20 7
72.8
15.7-23.2
12 1 37 2
16.5
50.3
45.9
23.2
16.3-23.7
Lag
0-1
n ")
0-2
0-3
1
0-3
0
1-3
0
1
0
n i
ni
0
1
0
1
~
y
0-1
L
Q
0
0-1
0
1-1
I
1
1
1
0
0-1
Q
1
0
1
0-1
0-1
0-1
1-6 hrs
7-12 hrs
19-24 fir's
8
0-2
0-2
0
1-1
0-1
Notes
23+ C
<23C
<25 C
all monitors (13)
city monitors (5)
correlated monitors (8)
r~C-
*
'Vorrn 1
Cold
sARR
sCHF
sNO
4
NO2
NO
NO
> 65 yrs
15- 64 yrs
15 64yrs
_ 3D yrs, Ml survivors
— 0—
— <
-4
~9
0§6+4yrs
case-crossover
time series
> 65 yrs
24-h
*
t
*
* _
•
/~
0
0
±5^
•
— • —
t
9
—9 —
1
-£=
t
$
1
^
>
•-
.«-
A
^
• —
^L
1-h
1 —
t
.£_
•
_e —
•—
•+-
0.75
1 1.25
1.5
1.75
Note: Studies in red are recent studies. Studies in black were included in the 2008 ISA for Oxides of Nitrogen. Relative risks are
standardized to a 20 ppb or 30-ppb increase in NO2 or NO concentration for 24-h and 1-h averaging times, respectively. Studies are
presented in descending order, with the largest mean concentration (ppb) at the top and the smallest at the bottom of the figure (by
averaging time and inclusion in previous ISA). Circles = NO2; Triangles = NO.
Figure 4-13 Results of studies of short-term exposure to oxides of nitrogen
and hospital admissions for cardiac disease.
November 2013
4-238
DRAFT: Do Not Cite or Quote
-------
Table 4-32 Corresponding risk estimates for hospital admissions for cardiac
disease for studies presented in Figure 4-13.
Study Location Health Effect
Thach et al. (2010) Hong Kong, China IHD
Hsieh et al. (2010) Taipei, Taiwan Ml
Cheng et al. (2009a) Kaohsiung, Taiwan Ml
Bell et al. (2008) Taipei, Taiwan IHD
Szvszkowicz (2009) 6 Canadian Cities Angina
Szyszkowicz (2007b) Montreal, Canada IHD
Stieb et al. (2009) 7 Canadian Cities Ml
Vencloviene et al. (201 1 ) Kaunas, Lithuania Ml
Turin et al. (2012) Takashima County, Japan Ml
Nuvolone et al. (2011) Tuscany, Italy Ml
Wichmann et al. (2012) Copenhagen, Denmark Ml
Relative Risk3 (95% Cl)
Lag 0-1: 1.04(1.02, 1.05)
>23°C: 1.24(1.16, 1.35)
<23°C: 1.26(1.18, 1.35)
>25°C: 1.23(1.06, 1.44)
<25°C: 1.76(1.55,2.02)
All Monitors
LagO: 1.10(1.02, 1.18)
Lag 1: 1.05(0.98, 1.13)
Lag 0-3: 1.03(0.98, 1.21)
City Monitors
LagO: 1.09(1.02, 1.16)
Lag 1: 1.05(0.98, 1.12)
Lag 0-3: 1.08 (0.99, 1.20)
Correlated Monitors
LagO: 1.09(1.02, 1.17)
Lag 1: 1.05(0.98, 1.12)
Lag 0-3: 1.08(0.97, 1.20)
LagO: 1.04(1.03, 1.05)
Lag 1: 1.13(1.04, 1.22)
LagO: 1.03(1.00, 1.05)
Lag 1: 1.03(1.00, 1.06)
<65yrs, lag 0-1: 1.19(0.96, 1.48)
>65yrs, lag 0-1: 0.97(0.81, 1.17)
LagO: 1.14(0.92, 1.40)
Lag 1: 0.87(0.70, 1.08)
LagO: 1.04(0.97, 1.12)
Lag 1: 1.08(1.00, 1.15)
Warm Season
LagO: 0.99(0.86, 1.14)
Lag 1: 1.10(0.96, 1.26)
Lag 0-1: 1.07(0.93, 1.22)
Cool Season
LagO: 1.01 (0.91, 1.11)
Lag 1: 1.08(0.98, 1.19)
Lag 0-1: 1.06(0.96, 1.17)
November 2013
4-239
DRAFT: Do Not Cite or Quote
-------
Table 4-32 (Continued): Corresponding risk estimates for hospital admissions for cardiac
disease for studies presented in Figure 4-13.
Study Location
Larrieu et al. (2007) 8 French Cities
Rich et al. (2010) New Jersey, U.S.
Mann et al. (2002) Los Anqeles, CA
Poloniecki et al. (1997) London, U.K.
Linn et al. (2000) Los Anqeles, CA
Wong et al. (1999) Hong Kong, China
Ponka and Virtanen Helsinki, Finland
(1996)
Barnett et al. (2006) 7 Australian and New
Zealand Cities
Von Klot et al. (2005) 5 European Cities
Health Effect
IHD
Ml
IHD
Ml
IHD
Ml
Angina
Ml
IHD
IHD
IHD
Ml
Ml
Angina
Relative Risk3 (95% Cl)
1.07(1.03, 1.10)
LagO: 1.14(0.96, 1.32)
With secondary arrhythmia
LagO: 1.04(1.02, 1.06)
Lag 1: 1.02(1.00, 1.04)
Lag 0-1: 1.03(1.01, 1.05)
With secondary congestive heart failure
LagO: 1.05(1.01, 1.08)
Lag 1: 1.04(1.01, 1.08)
Lag 0-1: 1.05(1.02, 1.09)
With no secondary disease
LagO: 1.03(1.01, 1.04)
Lag 1: 1.03(1.01, 1.04)
Lag 0-1: 1.03(1.02, 1.05)
1.04(1.02, 1.06)
Lag1: 1.00(0.98, 1.02)
Lag 1: 1.02(1.01, 1.03)
Lag 1: 1.01 (1.00, 1.03)
LagO: 1.02(1.00, 1.04)
1.04(1.00, 1.08)
NO2
LagO: 1.17(0.96, 1.42)
Lag 1: 0.95(0.77, 1.16)
NO
LagO: 1.01 (0.96, 1.05)
Lag 1: 1.10(1.05, 1.15)
Lag 0-1, >65yrs: 1.10(1.04, 1
Lag 0-1, 15-64yrs: 1.03(0.96,
Lag 0-1, >65yrs: 1.18(1.04, 1
Lag 0-1, 15-64yrs: 1.07(0.96,
LagO: 1.14(0.99, 1.32)
LagO: 1.16(1.03, 1.31)
.17)
1.10)
.35)
1.20)
Bhaskaran et al. (2011) England and Wales
Ml
Lag 1-6 h: 1.06(1.02, 1.11)
Lag 7-12 h: 0.95(0.90, 0.99)
Lag 13-18h: 0.99(0.94, 1.05)
Lag 19-24h: 1.00(0.96(1.05)
Lag 0-2 days: 0.98(0.93, 1.02)
November 2013
4-240
DRAFT: Do Not Cite or Quote
-------
Table 4-32 (Continued): Corresponding risk estimates for hospital admissions for cardiac
disease for studies presented in Figure 4-13.
Study
Atkinson et al. (1999)
Peel et al. (2007)
Jalaludin et al. (2006)
Simpson et al. (2005a)
Location
London, U.K.
Atlanta, GA
Sydney, Australia
4 Australian Cities
Health Effect
IHD
IHD
IHD
IHD
Relative Risk3 (95% Cl)
Lag 0; 0-64 yrs: 1.01 (0.99, 1.04)
LagO; 65+ yrs: 1.03(1.01, 1.04)
Lag 0-2; case-crossover: 1.04 (1.00, 1.07)
Lag 0-2; time-series: 1.04 (1.01, 1.08)
LagO: 1.07(1.01, 1.13)
Lag 1: 1.02(0.97, 1.08)
Lag 0-1: 1.01 (0.99, 1.03)
Lag 0-1: 1.06(1.03, 1.09)
Note: Studies correspond to studies presented in Figure 4-13.
a Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration
tor 24- h and 1-h averaging times, respectively.
4.3.7.3
Cerebrovascular Diseases and Stroke
i
2
o
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
The 2008 ISA for Oxides of Nitrogen found that the epidemiologic evidence for
associations between short-term changes in NO2 levels and hospital admissions or ED
visits for cerebrovascular diseases was generally inconsistent and provided little evidence
for an NO2 effect (U.S. EPA. 2008c). Recent studies published since the 2008 ISA add to
the evidence (Figure 4-14 and Table 4-33).
Generally, studies based on clinical registries are less susceptible to misclassification of
the outcome and exposure, which may explain the stronger evidence provided by these
studies than those based on administrative data. Wellenius et al. (2012) reviewed the
medical records of 1,705 Boston-area patients hospitalized with neurologist-confirmed
acute ischemic stroke and found an odds ratio for ischemic stroke onset of 1.32 (95% CI:
1.08, 1.63) per 20-ppb increase in mean NO2 concentrations over the past 24 hours. A
unique strength of this study was the availability of information on the date and time of
stroke symptom onset in most patients, thereby substantially reducing misclassification of
the exposure. Copollutant models were not considered.
Andersen et al. (2010) obtained data on strokes in Copenhagen, Denmark from the
Danish National Indicator Project and found a positive association between ambient NOX
concentrations and risk of ischemic stroke, but not hemorrhagic stroke. The strongest
association was observed in relation to NOX levels 4 days earlier and for those suffering a
mild stroke, but the association seemed to be attenuated after adjustment for ultrafine
particles. Using data from a stroke registry in Como, Italy, Vidale etal. (2010) found that
NO2 was associated with risk of ischemic stroke hospitalization. On the other hand, Turin
November 2013
4-241
DRAFT: Do Not Cite or Quote
-------
1 et al. (2012) did not observe any association using data from the Takashima County
2 Stroke and AMI Registry in Central Japan. Similarly, Oudin et al. (2010) found no
3 association between modeled residential NOX concentration and risk of ischemic or
4 hemorrhagic stroke within the context of a Swedish quality register for stroke.
5 Additional studies based on administrative data are also available. Szyszkowicz (2008b)
6 observed a positive association between NO2 and emergency department visits for
7 ischemic stroke in Edmonton, Canada, but only within specific subgroups according to
8 sex, season, and age. Studies from Taipei, Taiwan (Bell et al., 2008) and Hong Kong
9 (Thach et al.. 2010) failed to find any associations with cerebrovascular disease.
10 In summary, the epidemiologic data available continues to support an association
11 between ambient NO2 concentrations and risk of hospital admission for cerebrovascular
12 disease and stroke. Generally, the associations observed in these studies are robust in
13 copollutant models that adjust for PM or other gaseous pollutants.
November 2013 4-242 DRAFT: Do Not Cite or Quote
-------
Study
Thach etal. 2010
Bell etal. 2008
Outcome
Stroke
CBV
Mean
Concentrations
31.2
26.4
Turin etal. 2012
Stroke 16
Cerebral Infarction
Intracerebral Hem
Subarachnoid hem
Welleniusetal. 2012
Andersen etal. 2010
Larrieuetal. 2007
Ischemic Stroke
Ischemic Stroke
Hemorrhagic Stroke
Mild Ischemic
Stroke
15.3
Lag
0-1
0
1
0-3
0
1
0-3
0
0-3
0
0
0
0
1
24hr
0
1
0-4
0
1
11.9-24.9
Ballesteretal. 2001
Peel etal. 2007
CBV
CBV
Jalaludin etal. 2006 Stroke
61.8
45.9
23.2
0-2
0-2
0
1
0-1
Notes
all monjtors
all monjtors
all monitors
cjty monjtors
cjty monjtors
city monitors
correlated monjtors
correlated monjtors
correlated monitors
all ages
> 65 yrs
Polonieckietal. 1997
Linn etal. 2000
Tsai et al. 2003
Wong etal. 1999
Villeneuve etal. 2006
Wellenius etal. 2005
Ponka and Virtanen 1996
CBV 35
CBV 28-41
Occ Stroke
Cerebral Stroke 28.17
Ischemic Stroke
Cerebral Stroke
Ischemic Stroke
CBV 27.3
Ischemic Stroke 24
Hemorrhagic Stroke
Cerebral Stroke
Ischemic Stroke 23.54
Hemorrhagic Stroke
CBV 20.7
1
0
0
0-2
0-2
0-2
0-2
0-1
0
1
0
1
0
1
0
0
0
1
20+ C
20+ C
<20C
<20C
65+ yrs
case-crossover
time series
> 65 yrs
> 65 yrs
> 65 yrs
24-h
1-h
0.5 1 1.5
Relative Risk (95% Cl)
2.5
Note: Studies in red are recent studies. Studies in black were included in the 2008 ISA for Oxides of Nitrogen. Relative risks are
standardized to a 20 ppb or 30-ppb increase in NO2 or NO concentration and 40 ppb or 60 ppb for NOX concentrations for 24-h and
1-h averaging times, respectively. Studies are presented in descending order, with the largest mean concentration (ppb) at the top
and the smallest at the bottom of the figure (by averaging time and inclusion in previous ISA). Circles = NO2; Diamonds = NOX.
Figure 4-14 Results of studies of short-term exposure to oxides of nitrogen
and hospital admissions for cerebrovascular disease and stroke.
November 2013
4-243
DRAFT: Do Not Cite or Quote
-------
Table 4-33 Corresponding risk estimates for hospital admissions for
cerebrovascular disease and stroke for studies presented in
Figure 4-14.
Study Location Health Effect
Thach et al. (2010) Hong Kong, China Stroke
Bell et al. (2008) Taipei, Taiwan Cerebrovascular disease
Turin et al. (2012) Takashima County, Stroke
JaPan Cerebral Infarction
Intracerebral Hemorrhage
Subarachnoid Hemorrhage
Wellenius et al. c--t-,- i\/r ,_.,._..—,._,._
f9n19^ bubtun, MA Ibcliemic otiuke
Andersen et al. Copenhagen, Ischemic Stroke
12010) Denmark Hemorrhagic Stroke
Selected Relative Risks3 (95% Cl)
Lag 0-1: 1.01 (1.00, 1.03)
All monitors:
LagO: 1.01 (0.95, 1.07)
Lag 1: 0.97(0.92, 1.03)
Lag 0-3: 1.04(0.96, 1.12)
City monitors:
LagO: 1.01 (0.96, 1.06)
Lag 1: 0.97(0.92, 1.02)
Lag 0-3: 1.04(0.96, 1.12)
Correlated monitors:
LagO: 1.01 (0.96, 1.06)
Lag 1: 0.97(0.92, 1.02)
Lag 0-3: 1.04(0.96, 1.12)
Stroke:
LagO: 0.98(0.89, 1.08)
Lag 1: 0.98(0.89, 1.08)
Cerebral Infarction:
LagO: 0.98(0.87, 1.10)
Lag 1: 1.00(0.89, 1.12)
Intracerebral Hemorrhage:
LagO: 1.06(0.85, 1.33)
Lag 1: 0.94(0.75, 1.16)
Subarachnoid Hemorrhage:
LagO: 1.12(0.80, 1.56)
Lag 1: 1.02(0.73, 1.42)
Lag 24 h: 1.32(1.08, 1.63)
NOX:
Ischemic Stroke:
Mild Ischemic Stroke
LagO: 1.20(0.96, 1.48)
Lag 1: 0.96(0.79, 1.20)
Lag 0-4: 1.36(0.96, 1.89)
Hemorrhagic Stroke:
LagO: 0.96(0.48, 1.81)
Lag 1: 1.14(0.59,2.20)
Lag 0-4: 0.43(0.13, 1.36)
Mild Ischemic Stroke :
Lag 0-4: 1.61 (0.79, 3.30)
November 2013
4-244
DRAFT: Do Not Cite or Quote
-------
Table 4-33 (Continued): Corresponding risk estimates for hospital admissions for
cerebrovascular disease and stroke for studies presented in
Figure 4-14.
Study
Larrieu et
Poloniecki
(1997)
Linn et al.
al.
_§t
(2007)
.li.
(2000)
Location
8 French Cities
London, U.K.
Los Angeles, CA
Health Effect
Stroke
Cerebrovascular disease
Cerebrovascular disease
Occlusive Stroke
Selected Relative Risks3
All ages:
> 65 yrs:
Lag 1: 0,
LagO: 1
LagO: 1
0.
1.
.99
.01
.04
99
01
(0
(0
(1
(0.96,
(0.97,
.98
.99
.02
, 1.
, 1.
, 1.
1.03)
1.05)
,00)
,02)
,06)
(95% Cl)
Tsai et al. (2003)
Kaohsiung, Taiwan Cerebral Stroke
Ischemic Stroke
Cerebral Stroke:
Lag 0-2; 20+ °C: 1.68 (1.38, 2.04)
Lag 0-2; <20 °C: 0.78 (0.44, 1.37)
Ischemic Stroke:
Lag 0-2; 20+°C: 1.67(1.48, 1.87)
Lag 0-2; <20 °C: 1.19(0.78, 1.84)
Wong et al. (1999) Hong Kong, China Cerebrovascular disease
Lag 0-1: 1.03(0.99, 1.07)
Villeneuve et al. Edmonton, Canada Ischemic Stroke
(2006a) Hemorrhagic Stroke
Cerebral Stroke
Ischemic Stroke:
LagO: 1.04(0.96, 1.15)
Lag 1: 1.07(0.97, 1.17)
Hemorrhagic Stroke:
LagO: 1.07(0.96, 1.21)
Lag 1: 1.06(0.94, 1.20)
Cerebral Stroke:
LagO: 0.99(0.90, 1.07)
Wellenius et al. 9 U.S. Cities
(2005)
Ponka and Virtanen Helsinki, Finland
(1996)
Ballester et al. (2001) Valencia, Spain
Ischemic Stroke
Hemorrhagic Stroke
Cerebrovascular disease
Cerebrovascular disease
Lag 1: 0.91 (0.84,
LagO: 1.05(1.03,
LagO: 1.01 (0.96,
Lag 0: 0.96 (0.87,
Lag 1: 0.98(0.87,
Lag 2: 1.22(1.04,
1.00)
1.07)
1.06)
1.07)
1.09)
1.44)
Peel et al. (2007) Atlanta, GA
Cerebrovascular disease
Lag 0-2; case-crossover:
1.05(1.01, 1.09)
Lag 0-2; time-series: 1.06 (1.02, 1.11)
Jalaludin et al.
(2006)
Sydney, Australia Stroke
LagO: 0.95(0.88, 1.02)
Lag 1: 0.96(0.90, 1.03)
Lag 0-1: 0.95(0.88, 1.02)
Note: Studies correspond to studies presented in Figure 4-14.
a Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or NO or 40 ppb or 60-ppb increase in NOX concentration
tor 24- h and 1-h averaging times, respectively.
November 2013
4-245
DRAFT: Do Not Cite or Quote
-------
4.3.7.4 Other Cardiovascular Causes of Hospital Admission or
ED Visit
1 A study covering the metropolitan region of Santiago, Chile, found a 9.7% (95% CI:
2 4.1%, 15.4%) and 8.4% (95% CI: 5.0%, 11.8%) increased rate of hospital admission for
3 venous thrombosis and pulmonary embolism, respectively, per 20-ppb increase in
4 24-h avg NO2 concentrations (Dales et al.. 2010). These associations were somewhat
5 attenuated in copollutant models, but the associations remained positive.
6 In Beijing, China, Guo etal. (2010) found that NO2 was associated with rates of ED
7 visits for hypertension, and the association remained relatively unchanged in copollutant
8 models adjusting for PMi0 or SO2. In contrast, in Edmonton, Canada, Szyszkowicz et al.
9 (2012) found that ED visits for hypertension were positively associated with NO2 in
10 single pollutant models. The association was attenuated in a multipollutant model
11 adjusting for SO2 and PM10, but results from these models are difficult to interpret given
12 the potential for multicollinearity among pollutants.
13 Using data from 14 hospitals in 7 Canadian cities, Stieb et al. (2009) found no association
14 between NO2 and risk of hospital admission for arrhythmias. However, Tsai et al. (2009)
15 reported finding an association in Taipei, Taiwan.
16 In summary, few studies from single locations have documented associations with
17 hospital admissions and ED visits for other cardiovascular causes including venous
18 thrombosis, pulmonary embolism, and hypertension.
4.3.8 Cardiovascular Mortality
19 Studies evaluated in the 2008 ISA for Oxides of Nitrogen that examined the association
20 between short-term NO2 exposure and cause-specific mortality consistently found
21 positive associations with cardiovascular mortality. Across studies, there was evidence
22 that the magnitude of the NO 2-cardiovascular mortality relationship was similar or
23 slightly larger than that for total mortality. Recent multi-city studies provide evidence
24 that is consistent with those studies evaluated in the 2008 ISA for Oxides of Nitrogen
25 (Section 4.4. Figure 4-17).
26 The NO2-cardiovascular mortality relationship was further examined in a few studies
27 through copollutants analyses. Chen etal. (2012b) in the 17 Chinese cities study
28 (CAPES) found that NO2 risk estimates for cardiovascular mortality were slightly
29 attenuated, but remained positive in copollutant models with PMi0 and SO2 (7.1% [95%
November 2013 4-246 DRAFT: Do Not Cite or Quote
-------
1 CI: 3.9, 10.3] for a 20-ppb increase in 24-h avg NO2 concentrations at lag 0-1; 4.7%
2 [95% CI: 1.2, 8.2] with PM10; 5.9% [95% CI: 2.6, 9.2] with SO2). Chiusolo et al. (2011)
3 also found evidence that associations between short-term NO2 exposure and
4 cardiovascular mortality remained robust in copollutant models in a study of 11 Italian
5 cities. In an all-year analysis, a 20-ppb increase in NO2 at lag 1-5 was associated with a
6 10.5% (95% CI: 6.0, 15.2) increase in cardiovascular mortality and a 10.3% (95% CI:
7 4.1, 16.8) increase adjusted for PM10. In a warm season analysis (April-September), the
8 NO2 effect estimate was 19.6% (95% CI: 11.7, 28.1) and 19.3% (95% CI: 11.0,28.2)
9 with adjustment for O3. Overall, the limited number of studies that have examined the
10 potential confounding effects on the NO2-cardiovascular mortality relationship, indicate
11 that associations remain robust.
12 Of the studies evaluated, only the studies conducted in Italy examined potential seasonal
13 differences in the NO2-cause-specific mortality relationship (Chiusolo et al.. 2011;
14 Bellini et al.. 2007). In a study of 15 Italian cities, Bellini et al. (2007) found that risk
15 estimates for cardiovascular mortality were dramatically increased in the summer from
16 1.6% to 7.4% for a 20-ppb increase in 24-h avg NO2 concentrations at lag 0-1,
17 respectively, with no evidence of an association in the winter. These results were
18 corroborated in a study of 10 Italian cities (Chiusolo et al.. 2011). which also observed an
19 increase in risk estimates for cardiovascular mortality in the warm season (i.e., April -
20 September) compared to all-year analyses. Chiusolo etal. (2011) did not conduct winter
21 season analyses. Although the cardiovascular mortality results are consistent with those
22 observed in the total mortality analyses conducted by Bellini et al. (2007) and Chiusolo et
23 al. (2011), as discussed in Section 4.4.3, studies conducted in Asian cities observed much
24 different seasonal patterns and it remains unclear if the seasonal patterns observed for
25 total mortality would be similar to those observed for cardiovascular mortality in these
26 cities.
4.3.9 Summary and Causal Determination
27 Evidence indicates that there is likely to be a causal relationship between short-term
28 exposure to oxides of nitrogen and cardiovascular health effects based primarily on
29 epidemiologic studies of adults that consistently demonstrate NO 2-associated
30 hospitalizations and ED visits for cardiovascular effects and mortality from
31 cardiovascular disease. This conclusion represents a change from the 2008 ISA for
32 Oxides of Nitrogen that concluded the "available evidence on the effects of short-term
33 exposure to NO2 on cardiovascular health effects was inadequate to infer the presence or
34 absence of a causal relationship at this time" (U.S. EPA. 2008c). Specifically, the
35 epidemiologic panel studies and toxicological studies available at the time of the last
November 2013 4-247 DRAFT: Do Not Cite or Quote
-------
1 review were inconsistent. Most epidemiologic studies reviewed in the 2008 ISA for
2 Oxides of Nitrogen found positive associations between ambient NO2 concentrations and
3 risk of hospital admissions or ED visits for all cardiovascular diseases (U.S. EPA.
4 2008c_). However, it was unclear at that time whether these results supported a direct
5 effect of short-term NO2 exposure on cardiovascular morbidity or were confounded by
6 other correlated pollutants. Recent high-quality epidemiologic studies have further
7 evaluated this uncertainty using copollutant models and comparing associations of NO2
8 with those of other criteria pollutants. These studies provide evidence for independent
9 associations of NO2 with cardiovascular effects, thus reducing the uncertainty from the
10 2008 ISA for Oxides of Nitrogen regarding the potential for NO2 to serve as an indicator
11 for another combustion-related pollutant or mixture. An uncertainty that remains from the
12 2008 ISA for Oxides of Nitrogen is the lack of mechanistic evidence to describe a role for
13 NO2 in the development of cardiovascular diseases, including key events that inform the
14 mode of action. The evidence for cardiovascular effects, with respect to the causal
15 determination for short-term exposure to oxides of nitrogen is detailed below using the
16 framework described in Table II of the Preamble. The key evidence, supporting or
17 contradicting, as it relates to the causal framework is summarized in Table 4-36.
18 Time-series studies of adults in the general population consistently report positive
19 associations between concentrations of oxides of nitrogen and hospital admissions and
20 ED visits for all cardiovascular disease and, specifically, IHD. High-quality single-city
21 studies from the U.S. (Ito etal.. 2011; Peel et al.. 2007; Tolbert et al.. 2007) and multicity
22 studies conducted in Europe and Australia and New Zealand (Larrieu et al.. 2007;
23 Ballester etal., 2006; Barnett et al., 2006; Von Klot et al., 2005) report positive
24 associations with all CVD hospitalizations in adults with adjustment for numerous
25 potential confounding factors, including weather and time trends. The strongest evidence
26 is for hospital admissions due to IHD (Stieb et al.. 2009; Larrieu et al.. 2007; Peel et al..
27 2007; Von Klot et al.. 2005; Mann et al.. 2002). Recent controlled human exposure and
28 animal toxicological studies provide weak evidence for a potential biologically plausible
29 mechanism leading to cardiovascular disease and IHD, with some studies reporting
30 induction of systemic inflammation and oxidative stress (Channell et al.. 2012; Huang et
31 al..2012b;Lietal..2011a).
32 The evidence for associations observed in time-series studies is coherent with positive
33 associations reported in epidemiologic studies of short-term NO2 exposure and
34 cardiovascular mortality in adults. These include studies reviewed in the 2008 ISA for
35 Oxides of Nitrogen and recent multicity studies that generally report a similar or slightly
36 larger magnitude for the NO2-cardiovascular mortality relationship compared to total
37 mortality.
November 2013 4-248 DRAFT: Do Not Cite or Quote
-------
1 There is limited evidence from epidemiologic and controlled human exposure studies to
2 suggest that NO2 exposure results in alterations of cardiac autonomic control, which may
3 trigger life-threatening cardiovascular events. Recent epidemiologic studies generally
4 reported associations between ambient NO2 levels and decreases in indices of HRV and
5 changes in ventricular repolarization among populations with pre-existing or at elevated
6 risk for cardiovascular disease. Since these effects are known indicators of myocardial
7 ischemia and other cardiovascular events, the consistently observed associations with
8 measures of altered autonomic control provide biological plausibility for the
9 hospitalizations and mortality associations observed in the studies. Although changes
10 were not observed across all endpoints, a recent controlled human exposure study
11 reported decreased HFn and QTVI in healthy exercising adults exposed to NO2,
12 indicating a potential disruption in the normal cardiac autonomic control (Huang et al..
13 2012b). However, similar measures of autonomic control in another controlled human
14 exposure study were inconsistent (Scaife et al., 2012). Evidence from epidemiologic and
15 experimental studies for other cardiovascular effects, including blood pressure and
16 arrhythmia, was mixed or limited in scope. Inconsistencies across studies and the limited
17 evidence available to suggest NO2-related subclinical and clinical cardiovascular effects
18 represent a lack of coherence across all lines of evidence to support the effects observed
19 in hospital admissions and ED visits, and cardiovascular mortality.
20 A key uncertainty discussed in the 2008 ISA for Oxides of Nitrogen was the potential for
21 confounding by other correlated pollutants. Recent studies have evaluated this
22 uncertainty using copollutant models and comparing associations of NO2 with those of
23 other pollutants. Figure 4-15 and Figure 4-16 present the single-pollutant and copollutant
24 model results from studies that examined associations between short-term exposure to
25 NO2 and cardiovascular disease adjusted for PM or CO, respectively. Additional figures
26 characterizing the copollutant models for NO2 with SO2 or O3 can be found online in
27 Supplemental Figures S4-2 and S4-3 (U.S. EPA. 2013c). Specifically, a number of
28 studies found that associations of NO2 and cardiovascular hospital admissions were
29 stronger in magnitude than associations with other pollutants in copollutant models,
30 including PM, O3, CO, and SO2 (Ito etal.. 2011: Larrieu et al. 2007: Peel et al. 2007:
31 Tolbert et al.. 2007). Other studies found that estimates for NO2 were robust to inclusion
32 of copollutants in the models (Ballester et al.. 2006: Von Klot et al.. 2005). In addition, a
33 limited number of studies that examined copollutant confounding on the
34 NO2-cardiovascular mortality relationship indicate that associations remain robust (Chen
35 et al., 2012b: Chiusolo et al., 2011). However, not all analyses reported NO2 as the
36 strongest predictor of cardiovascular effects. One study reported that associations with
37 cardiovascular hospitalizations were not robust in models matching on CO exposure
38 (Barnett et al.. 2006) and another reported associations with CO, total carbon, and EC and
39 OC components of PM25 that were stronger or similar in magnitude to those for NO2
November 2013 4-249 DRAFT: Do Not Cite or Quote
-------
1 (Tolbert et al. 2007). However, other (non-criteria) pollutants that may be potentially
2 correlated with NO2 were generally not examined in copollutant models, resulting in the
3 potential for unmeasured confounding. Finally, while copollutant models are a common
4 statistical tool used to evaluate the potential for copollutants confounding, their
5 interpretation can be limited (Section 1.5). Until more reliable methods to adjust for
6 multiple copollutants simultaneously become available, there is potential for residual
7 confounding due to unmeasured copollutants (Section 1.5). Without consistent and
8 reproducible experimental evidence that is coherent with the effects observed in
9 epidemiologic studies, some uncertainty still exists concerning the role of correlated
10 pollutants in the associations observed with oxides of nitrogen. However, recent
11 epidemiologic studies examining the extent to which NO2 is independently associated
12 with cardiovascular effects have decreased the uncertainty that these associations are a
13 result of NO2 serving as a marker for effects of another traffic-related pollutant or mix of
14 pollutants.
November 2013 4-250 DRAFT: Do Not Cite or Quote
-------
Study Outcome Copollutant Notes
Quo etal. 2009 CVD
PM2.5
Rich etal. 2010 Ml pM2J. _
Chen etal. 2010 CVD pM1Q
Chang etal. 2005 CVD >20C
Chang etal. 2005 CVD <20C _^
Yang etal. 2004 CVD 25+ C
Yang etal. 2004 CVD <25C
Nuvoloneet al. 2011 Ml piwim
Hsieh etal. 2010 Ml 23+ C
Hsieh etal. 2010 Ml <23C
Cheng etal. 2009 Ml 25+ C
Cheng etal. 2009 Ml <25C
Tsai et al. 2009 Arrhythmia 23+ C
Tsai et al. 2009 Arrhythmia PMln <23C
Guo etal. 2010 Hypertension pMln
Yang 2008 CHF pM1Q >20C
Yang 2008 CHF pM1Q <20C -
Tsai et al. 2003 Stroke PMIO Cerebral
Tsai et al. 2003 Stroke PMln Ischemic
Andersen etal. 2008 CVD . .
NC(tot) —
Turin etal. 2012 Ml
Turin etal. 2012 Stroke -*
Turin etal. 2012 Cerebral Infarc -<
Intracerebral
Turin et al. 2012 „„ —
Simpson etal. 2005 CVD all ages
BSP
Poloniecki etal. 1997 Ml Cool
BS
Poloniecki etal. 1997 Ml Warm
BS
«-
k-
-• —
*
»
h
-•-
-^
• —
-*
A
— •
* —
A
•
•
*
0 0.5 1 1.5 2 2.5 3
Relative Risks (95% Cl)
: Studies in red are recent studies. Studies in black were included in the 2008 ISA for Oxides of Nitrogen. Relative risks v
dardized to a 20 ppb or 30-ppb increase in NO2 concentration for 24-h and 1-h averaging times, respectively. Relative ris
Andersen et al. (2010) were standardized to a 40 ppb or 60-ppb increase in NOX concentration for 24-h and 1 -h averagi
s, respectively. Model estimates are presented as pairs with the top estimate (circles) for the single pollutant model and t
m estimate (triangles) for the copollutants model. Horizontal lines indicate 95% confidence intervals around the central
nate. Associated data presented in Table 4-34. BSP: black smoke particles; BS: black smoke; NC(tot): total number coun
Figure 4-15 Results of single-pollutant and copollutants models of short-term
exposure to NO2 or NOx with and without PM and hospital
admissions for cardiovascular disease.
November 2013
4-251
DRAFT: Do Not Cite or Quote
-------
Table 4-34 Corresponding risk estimates of ambient NO2 or NOX for hospital admissions for cardiovascular
disease in studies conducting copollutant models with PM for presented in Figure 4-15.
Study
Location
Notes
Cause
Single Pollutant Relative
Risk3 (95% Cl)
Copollutant Relative Risk3
(95% Cl)
Quo et al. (2009)
Richetal. (2010)
Chenetal. (201 Ob)
Chang et al. (2005)
Chanq et al. (2005)
Yang et al. (2004)
Beijing, China
New Jersey, U.S.
Shanghai, China
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung, Taiwan
CVD
Ml
CVD
> 20 °C CVD
<20 °C CVD
> 25 °C CVD
1.05(1.00,
1.14(0.96,
1.03(1.00,
1.39(1.32,
1.24(1.12,
1.46(1.32,
1.11)
1.32)
1.06)
1.45)
1.37)
1.62)
+PM2.5: 1.02(0.96, 1.09)
+PM2.5: 1.05(0.85, 1.28)
+PM10: 1.03(1.00, 1.05)
+PM10: 1.43(1.35, 1.52)
+PM10: 0.97(0.86, 1.10)
+PM10: 1.18(1.02, 1.36)
Yang et al. (2004)
Kaohsiung, Taiwan
<25°C
CVD
2.54 (2.27, 2.84)
+PM10: 2.74 (2.36, 3.17)
Nuvolone et al. (2011)
Hsiehetal. (2010)
Hsiehetal. (2010)
Cheng et al. (2009a)
Cheng et al. (2009a)
Tsai et al. (2009)
Tsai et al. (2009)
Guoetal. (2010)
Yang (2008)
Yang (2008)
Tsai et al. (2003)
Tsai et al. (2003)
Andersen et al. (201 0)b
Andersen et al. (2008b)
Tuscany, Italy
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung, Taiwan
Kaohsiung, Taiwan
Taipei, Taiwan
Taipei, Taiwan
Beijing, China
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung, Taiwan
Kaohsiung, Taiwan
Copenhagen, Denmark
Copenhagen, Denmark
>23°C
<23°C
>25°C
<25°C
>23°C
<23°C
>20°C
<20°C
Ml
Ml
Ml
Ml
Ml
Arrhythmia
Arrhythmia
Hypertension
CHF
CHF
Cerebral Stroke
Ischemic Stroke
Mild Ischemic Stroke
CVD
1.09
1.24
1.26
1.23
1.76
1.19
1.34
1.44
1.41
1.04
1.68
1.67
1.61
1.00
(1
(1
(1
(1
(1
(1
(1
(1
(1
(0
(1
(1
(0.
(0
.02,
.16,
.18,
.06,
.55,
.10,
.25,
.15,
.30,
.90,
.38,
.48,
79,
.93,
1
1
1
1
2
1
1
1
1
1
2
1
3.
1
.16)
.35)
.35)
.44)
.02)
.27)
.46)
.79)
.53)
.21)
.04)
.87)
30)b
.10)
+PM10:
+PMi0:
+PM-|0:
+PM10:
+PM-|0:
+PMi0:
+PM10:
+PM-|0:
+PM10:
+PMi0:
+ PM-|0:
1.10(1.00, 1
1.24(1.14, 1
1.22(1.12, 1
1.25(1.04, 1
1.62(1.36, 1
1.16(1.06, 1
1.32(1.21, 1
1.50(1.15, 1
1.37(1.23, 1
1.08(0.92, 1
1.37(1.04, 1
+PMi0:1.47(1.24, 1
+ UFP: 1
+NC(tot):
.09 (0.48, 2.
1.00(0.87,
.21)
.37)
.33)
.51)
.93)
.27)
.44)
.95)
.53)
.30)
.81)
.73)
56)b
1.10)
November 2013
4-252
DRAFT: Do Not Cite or Quote
-------
Table 4-34 (Continued): Corresponding risk estimates of ambient NO2 or NOX for hospital admissions for cardiovascular
disease in studies conducting copollutant models with PM for presented in Figure 4-15.
Study
Location
Notes
Cause
Single Pollutant Relative
Risk3 (95% Cl)
Copollutant Relative Risk3
(95% Cl)
Turin et al.
Turin et al.
Turin et al.
Turin et al.
Turin et al.
(2012)
(2012)
(2012)
(2012)
(2012)
Simpson et al. (2005a)
Poloniecki
Poloniecki
etal. (1997)
etal. (1997)
Shiga, Japan
Shiga, Japan
Shiga, Japan
Shiga, Japan
Shiga, Japan
4 Australian Cities
London, U.K. Cool
London, U.K. Warm
Ml
Stroke
Cerebral Infarction
Intracerebral Hemorrhage
Hemorrhage
CVD
Ml
Ml
1.
0.
0.
1.
1.
1.
1.
1.
14
98
98
06
12
07
00
00
(0.
(0,
(0.
(0.
(0,
(1
(1
(1
.92,
.89,
.87,
.85,
.80,
.05,
.00,
.00,
1.40)
1.08)
1.10)
1.33)
1.56)
1.09)
1.00)
1.00)
+TSP:
+TSP:
+TSP
+TSP:
+TSP:
+BSP:
+BS:
+BS:
1.16(0.91,
1.04(0.92,
:1. 04 (0.91,
1.04(0.82,
1.06(0.70,
1.04(1.02,
1.00(1.00,
1.00(1.00,
1.51)
1.16)
1.20)
1.33)
1.58)
1.07)
1.00)
1.00)
Note: Studies correspond to studies presented in Figure 4-15.
"Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 or 40 ppb or 60-ppb increase in NOX concentration for 24- h and 1 -h averaging times, respectively.
bEffect estimates from Andersen et al. (2010) were standardized to a 40 ppb or 60-ppb increase in NOX concentration for 24-h and 1 -h averaging times, respectively.
BSP: black smoke particles; BS: black smoke; NC(tot): total number count.
November 2013
DRAFT: Do Not Cite or Quote
-------
Study
Tolbertetal. 2007
Chang etal. 2005
Chang etal. 2005
Yang etal. 2004
Yang etal. 2004
Polonieckiet al. 1997
Polonieckietal. 1997
Nuvoloneet al. 2011
Hsiehetal. 2010
Hsiehetal. 2010
Cheng etal. 2009
Cheng etal. 2009
Tsai etal. 2009
Tsai etal. 2009
Yang 2008
Yang 2008
Tsai etal. 2003
Tsai etal. 2003
Outcome
CVD
CVD
CVD
CVD
CVD
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Arrhythmia
Arrhythmia
CHF
CHF
Stroke
Stroke
Notes
>20C
<20C
25+ C
<25C
Cool '
(
Warm <
t
23+ C
<23C
25+ C
— i
<25C
23+ C
<23C
>20C
<20C
— t
cerebral
Ischemic
»
_ +
+_
-A-
t
1
I
•-
fc-
_£
-*-
+
•
j-
-*-
•—
0
A
Relative Risks (95% CD
Note: Studies in red are recent studies. Studies in black were included in the 2008 ISA for Oxides of Nitrogen. Relative risks are
standardized to a 20 ppb or 30-ppb increase in NO2 concentration for 24-h and 1-h averaging times, respectively. Model estimates
are presented as pairs with the top estimate (Circles) for the single pollutant model and the bottom estimate (Triangles) for the
copollutants model. Horizontal lines indicate 95% confidence intervals around the central estimate. Associated data presented in
Table 4-35.
Figure 4-16 Results of single-pollutant and copollutants models of short-term
exposure to NO2 (withCO [triangles] and without CO [circles]) and
hospital admissions for cardiovascular disease.
November 2013
4-254
DRAFT: Do Not Cite or Quote
-------
Table 4-35 Corresponding risk estimates of ambient NO2 for hospital
admissions for cardiovascular disease in studies conducting
copollutant models with CO presented in Figure 4-16.
Study
Tolbert et al. (2007)
Chang et al. (2005)
Chanq et al. (2005)
Yang et al. (2004)
Yang et al. (2004)
Poloniecki et al. (1997)
Poloniecki et al. (1997)
Nuvolone et al. (2011)
Hsiehetal. (2010)
Hsiehetal. (2010)
Chenq et al. (2009a)
Cheng et al. (2009a)
Tsai et al. (2009)
Tsai et al. (2009)
Yang (2008)
Yang (2008)
Tsai et al. (2003)
Tsai et al. (2003)
Note:Studies correspond to
Location
Atlanta, GA
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung,
Taiwan
Kaohsiung,
Taiwan
London, U.K.
London, U.K.
Tuscany, Italy
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung,
Taiwan
Kaohsiung,
Taiwan
Taipei, Taiwan
Taipei, Taiwan
Taipei, Taiwan
Taipei, Taiwan
Kaohsiung,
Taiwan
Kaohsiung,
Taiwan
studies presented in
Notes
>20°C
<20°C
>25°C
<25°C
Cool
Warm
>23°C
<23°C
>25°C
<25°C
>23°C
<23°C
>20°C
<20°C
Figure 4-16,
Mortality
Cause
CVD
CVD
CVD
CVD
CVD
Ml
Ml
Ml
Ml
Ml
Ml
Ml
Arrhythmia
Arrhythmia
CHF
CHF
Cerebral
Stroke
Ischemic
Stroke
Single Pollutant
Relative Risk3 (95%
Cl)
1.02(1.01, 1.03)
1.39(1.32, 1.45)
1.24(1.12, 1.37)
1.46(1.32, 1.62)
2.45 (2.27, 2.84)
1.00(1.00, 1.00)
1.00(1.00, 1.00)
1.09(1.02, 1.16)
1.24(1.16, 1.35)
1.26(1.18, 1.35)
1.23(1.06, 1.44)
1.76(1.55,2.02)
1.19(1.10, 1.27)
1.34(1.25, 1.46)
1.41 (1.30, 1.53)
1.04(0.90, 1.21)
1.68(1.38,2.04)
1.67(1.48, 1.87)
Copollutant Relative
Risk3 (95% Cl)
0.99(0.97, 1.02)
1.31 (1.22, 1.41)
1.27(1.09, 1.47)
1.11 (0.91, 1.21)
2.89 (2.43, 3.42)
1.00(1.00, 1.00)
1.00(1.00, 1.00)
1.04(0.94, 1.14)
1.18(1.06, 1.31)
1.24(1.10, 1.42)
0.99(0.80, 1.23)
1.74(1.42,2.13)
1.14(1.02, 1.27)
1.32(1.16, 1.51)
1.39(1.21, 1.58)
0.96(0.76, 1.21)
1.73(1.30,2.32)
1.66(1.40, 1.98)
"Effect estimates are standardized to a 20 ppb or 30-ppb increase in NO2 concentration for 24- h and 1 -h averaging times,
respectively.
1
2
o
J
4
5
6
7
In conclusion, epidemiologic studies of adults consistently demonstrate NO2-associated
hospitalizations and ED visits for cardiovascular effects and mortality from
cardiovascular disease. These high-quality studies have been replicated by different
researchers in different locations and have adjusted for numerous potential confounding
factors, thus limiting the level of uncertainty for bias from confounding. Due to limited
analysis of potentially correlated non-criteria pollutants and recognized limitations of
copollutant models, some uncertainty remains regarding the extent to which oxides of
November 2013
4-255
DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
nitrogen are independently associated with cardiovascular outcomes or if NO2 serves as a
marker for the effects of another traffic-related pollutant or mix of pollutants. Animal
toxicological, controlled human exposure, and epidemiologic panel studies provide
limited evidence for changes in measures of autonomic nervous system function to
support the associations observed in epidemiologic studies of hospitalizations and
cardiovascular mortality. Thus, the combined evidence from epidemiologic and
experimental studies is sufficient to conclude that there is likely to be a causal
relationship between short-term NO2 exposure and cardiovascular effects.
Table 4-36 Summary of evidence supporting a likely to be a causal relationship
between short-term NO2 exposure and cardiovascular effects.
Rationale for
Causal
Determination3
Consistent
associations from
multiple, high-quality
epidemiologic
studies at relevant
NC>2 concentrations
Additional
epidemiologic
evidence help rule
out chance,
confounding, and
other biases with
reasonable
confidence
Consistent
associations from
multiple, high-quality
epidemiologic
studies at relevant
NO2 concentrations
and cardiovascular
mortality
Key Evidence13
Consistent evidence for hospital
admissions and ED visits for all CVD
in adults in multiple studies, including
multicity studies, in diverse locations.
Associations are strongest for
hospitalizations for IHD.
Associations with ED visits, hospital
admissions, and mortality found with
adjustment for numerous potential
confounding factors including
meteorological factors and time
trends.
Associations between ED visits and
hospital admissions and NC>2 are
generally robust in copollutant
models containing PM, CO.Os, or
SO2.
Consistent evidence for increased
risk of cardiovascular mortality in
adults applying differing model
specifications in diverse locations.
Key References'3
Larrieu et al. (2007);
Itoetal. (2011):
Peel et al. (2007);
Tolbert et al. (2007):
Von Klot et al. (2005):
Ballester et al. (2006):
Barnett et al. (2006)
Section 4.3.7
Larrieu et al. (2007):
Stieb et al. (2009):
Peel et al. (2007):
Von Klot et al. (2005);
Mann et al. (2002)
Section 4.3.7.2
Fiaure 4-15, Fiaure 4-16,
and Supplemental Figures
S4-2, and S4-3 (U.S. EPA,
201 3c)
Bellini etal. (2007):
Wong et al. (2008b):
Chen etal. (2012b):
Chiusoloetal. (2011)
Section 4.3.8
NO2 Concentrations
Associated with
Effects0
Mean 24-h avg:
11. 9-40.5 ppb
Mean 1-h max:
4? 9 4^ Q nnh
Mean 24-h avg:
11. 9 -37.2 ppb
Mean 1-h max:
45.9 ppb
Mean 24-h avg:
13.5 -35.4 ppb
November 2013
4-256
DRAFT: Do Not Cite or Quote
-------
Table 4-36 (Continued): Summary of evidence supporting a likely to be a causal
relationship between short-term NOi exposure and
cardiovascular effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NC>2 Concentrations
Associated with
Effects0
Uncertainty due to
limited coherence
with other lines of
evidence
Limited evidence from epidemiologic
panel studies and experimental
studies for subclinical and clinical
cardiovascular effects.
Limited evidence
from epidemiologic
panel studies
Limited epidemiologic evidence for
changes in HRV and ventricular
repolarization.
Stronger associations observed in
groups of individuals with
pre-existing cardiovascular disease.
HRV:
Timonen et al. (2006):
Suh and Zanobetti
(201 Oa):
Zanobetti et al. (2010)
Section 4.3.3.1
QT interval: Henneberqer
et al. (2005)
Section 4.3.4.1
Limited evidence
from animal
toxicological and
controlled human
exposure studies
Decrease in HF domain normalized
to HR (i.e., HRV) but decrease in
QTVI in controlled human exposure
study
Increase in a marker of endothelial
dysfunction (i.e., ICAM-1) in rats and
in cells treated with plasma from
adults exposed to NC>2.
Huang et al. (2012b)
Lietal. (201 1a)
Channelletal. (2012)
Healthy adults: 500 ppb
NO2
Rats: 2,660 and 5,320
ppb NC>2
Human cells exposed
to plasma from healthy
adults: 500 ppb NO2
Weak evidence to
describe key events
that inform the mode
of action
Oxidative stress
Inflammation
Section 3.3.2.8
Evidence of increased oxidative
stress in rats with relevant NC>2
exposures (i.e., MDA) and plasma
from NC>2-exposed humans
(i.e., LOX-1).
Lietal. (2011a)
Channelletal. (2012)
Toxicological evidence of increased
transcription of some inflammatory
mediators in vitro (i.e., IL-8) and in
rats (i.e., TNF-a).
Inconsistent epidemiologic evidence
for changes in CRP, IL-6, and
TNF-RII.
Channelletal. (2012)
Lietal. (2011a)
Section 4.3.6.1
Rats: 5,320 ppb NO2
Healthy adults: 500 ppb
NO2
Human cells exposed
to plasma from healthy
adults: 500 ppb NO2
Rats: 5,320 ppb NO2
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
""Describes the key evidence and references, supporting or contradicting, that contribute most heavily to causal determination.
References to earlier sections indicate where full body of evidence is described.
°Describes the NO2 concentrations with which the evidence is substantiated.
November 2013
4-257
DRAFT: Do Not Cite or Quote
-------
4.4 Total Mortality
4.4.1 Introduction and Summary of 2008 ISA for Oxides of Nitrogen
1 Prior to the 2008 ISA for Oxides of Nitrogen, epidemiologic studies had not been
2 identified that examined whether an association exists between short-term NO2 exposure
3 and mortality. The 2008 ISA for Oxides of Nitrogen evaluated a collection of studies,
4 including multicity studies, conducted in the U.S., Canada, and Europe, and a meta-
5 analysis (U.S. EPA. 2008c). All of these studies reported evidence of an association
6 between short-term NO2 exposure and mortality with estimates ranging from 0.5 to 3.6%
7 for a 20-ppb increase in 24-h avg or 30-ppb increase in 1-h max NO2 concentrations. A
8 limitation of this collection of studies was that the majority focused specifically on PM
9 and did not conduct extensive analyses to examine the relationship between short-term
10 NO2 exposure and mortality.
11 Multicity studies conducted in the U.S. (HEI. 2003), Canada (Brook et al.. 2007; Burnett
12 et al.. 2004) and Europe (Samoli et al.. 2006). as well as a large study conducted in the
13 Netherlands (Hoek. 2003). consistently reported positive associations between short-term
14 NO2 exposure and mortality, specifically at lag 1, with evidence that these associations
15 remain robust in copollutant models. These results were confirmed in a meta-analysis that
16 did not include any of the aforementioned multicity studies (Stieb et al.. 2002).
17 Of the studies evaluated in the 2008 ISA for Oxides of Nitrogen, a limited number
18 provided additional information on the NO2-mortality relationship. Initial evidence
19 indicated a larger NO2-mortality association during the warmer months (Brook et al..
20 2007; Burnett et al.. 2004; HEI. 2003). Additionally, an examination of total and cause-
21 specific mortality found associations similar in magnitude across mortality outcomes
22 (total, respiratory, and cardiovascular); however, some studies reported stronger NO2
23 associations for respiratory mortality (Biggeri et al.. 2005; Simpson et al.. 2005b).
24 Potential effect modifiers of the NO 2-mortality relationship were examined only within
25 the APHEA study, which found that within the European cities, geographic area and
26 smoking prevalence modified the NO2-mortality relationship. It is worth noting that
27 additional multicity European studies that focused on PM (Agaet al.. 2003; Katsouvanni
28 et al.. 2003) reported that cities with higher NO2 concentrations also had higher PM risk
29 estimates indicating that NO2 and PM may be potential effect modifiers of each other.
30 In summary, the multicity studies evaluated in the 2008 ISA for Oxides of Nitrogen
31 consistently observed positive associations between short-term NO2 exposure and
32 mortality. These studies indicated that associations were found to occur within the first
November 2013 4-258 DRAFT: Do Not Cite or Quote
-------
1 few days after exposure and are potentially influenced by season. However, uncertainties
2 remained in the NO2-mortality relationship which led to the 2008 ISA for Oxides of
3 Nitrogen (U.S. EPA. 2008c) concluding that the evidence "was suggestive but not
4 sufficient to infer a causal relationship." These uncertainties and data gaps included
5 whether: NO2 is acting as an indicator for another pollutant or a mix of pollutants; there
6 is evidence for potential copollutant confounding; specific factors modify the
7 NO2-mortality relationship; there is seasonal heterogeneity in mortality associations;
8 NO2 associations are stronger with specific mortality outcomes; and the shape of the
9 NO2-mortality concentration-response relationship is linear.
4.4.2 Associations between Short-term NO2 Exposure and Mortality
10 Since the completion of the 2008 ISA for Oxides of Nitrogen, the body of epidemiologic
11 literature that has examined the association between short-term NO2 exposure and
12 mortality has grown. However, similar to the collection of studies evaluated in the 2008
13 ISA for Oxides of Nitrogen, most of the recent studies did not focus specifically on the
14 NO2-mortality relationship. Of the studies identified, a limited number have been
15 conducted in the U.S., Canada, and Europe, with the majority being conducted in Asia
16 due to the increased focus on examining the effect of air pollution on health in
17 developing countries. Although these studies are informative in evaluation of the
18 relationship between oxides of nitrogen and mortality, the broad implications of these
19 studies in the context of results from studies conducted in the U.S., Canada, and Western
20 Europe are limited. This is because studies conducted in Asia encompass cities with
21 meteorological (Tsai etal. 2010; Wong et al., 2008b). outdoor air pollution
22 (e-g-, concentrations, mixtures, and transport of pollutants), and sociodemographic
23 (e-g-, disease patterns, age structure, and socioeconomic variables) (Kan et al., 2010)
24 characteristics that differ from cities in North America and Europe, potentially limiting
25 the generalizability of results from these studies to other cities.
26 Consistent with ISAs for other criteria pollutants, this section focuses primarily on
27 multicity studies because they examine the association between short-term NO2 exposure
28 and a health effect over a large geographic area using a consistent statistical methodology
29 (U.S. EPA. 2008c). Where applicable single-city studies are evaluated that encompass a
30 long study-duration, provide additional evidence indicating that a specific population
31 may be at increased risk of NO 2-related mortality, or address an uncertainty in the
32 NO2-mortality relationship not represented in other single-city or multicity studies. Other
33 recent studies of mortality are not the focus of this evaluation because of inadequate
34 study design or insufficient sample size. The full list of the studies can be found in
35 Supplemental Table S4-2 (U.S. EPA. 2013f). Overall this section evaluates studies that
November 2013 4-259 DRAFT: Do Not Cite or Quote
-------
1 examined the association between short-term NO2 exposure and mortality, and addresses
2 the key uncertainties in the NO2-mortality relationship identified in the 2008 ISA for
3 Oxides of Nitrogen: potential confounding of NO2 associations, effect modification
4 (i.e., sources of heterogeneity in risk estimates across cities or within a population),
5 seasonal heterogeneity in NO2 associations, and the NO2-mortality concentration-
6 response (C-R) relationship.
4.4.3 Associations between Short-term NO2 Exposure and Mortality in Ail-
Year Analyses
7 Multeity studies evaluated in the 2008 ISA for Oxides of Nitrogen reported consistent,
8 positive associations between short-tern NO2 exposure and mortality in all-year analyses
9 (U.S. EPA. 2008c). However, when focusing on specific causes of mortality, some
10 studies reported similar risk estimates across total (nonaccidental), cardiovascular, and
11 respiratory mortality (Samoli et al.. 2006; Burnett et al. 2004). while others indicated
12 larger respiratory mortality risk estimates compared to both total and cardiovascular
13 mortality (Biggeri et al., 2005; Simpson et al., 2005b). Although only a small number of
14 multicity studies have been conducted since the completion of the 2008 ISA for Oxides
15 of Nitrogen, these studies build upon and provide additional evidence for an association
16 between short-term NO2 exposure and total mortality along with potential differences by
17 mortality outcome. Air quality characteristics and study specific details for the studies
18 evaluated in this section are provided in Table 4-37.
November 2013 4-260 DRAFT: Do Not Cite or Quote
-------
Table 4-37 Air quality characteristics of studies evaluated in the 2008 ISA for
Oxides of Nitrogen and recently published multicity and select
single-city studies.
Study
Biqqeri et al.
(2005f
Brook et al. (2007f
Burnett et al.
(2004f
HEI (2003f
Hoek (2003f
Samoli et al.
(2006f
Simpson et al.
(2005bf
Stieb et al. (2003)a
Bellini etal. (2007)
Berqlind et al.
(2009)
Cakmak et al.
(2011 b)
Chen etal. (2012b)
Chiusolo et al.
(2011)
Kan etal. (2010):
Kan et al. (2008)
Location
8 Italian
cities
10
Canadian
cities
12
Canadian
cities
58 U.S.
cities'3
the
Netherlands
30
European
cities
4 Australian
cities
M eta-
analysis
15 Italian
cities
5 European
cities
7 Chilean
citiesd
17 Chinese
cities
10 Italian
cities'
Shanghai,
China
Years
1990-1999
1984-2000
1981-1999
1987-1994
1986-1994
1990-1997
1996-1999
—
1996-2002
1992-2002
1997-2007
1 996-20 10e
2001-2005
2001-2004
Mortality
Outcome(s)
Total,
Cardiovascular,
Respiratory
Total
Total,
Cardiovascular,
Respiratory
Total
Total
Total,
Cardiovascular,
Respiratory
Total,
Cardiovascular,
Respiratory
Total
Total,
Cardiovascular,
Respiratory
Total
Total
Total,
Cardiovascular,
Respiratory
Total,
Cardiovascular,
Cerebrovascular,
Respiratory
Total,
Cardiovascular,
Mean
Averaging Concentration
Time (ppb)
24-havg 30.1-55.0
94 h a\/n
94 h avn 100 9ft 4
24-h avg 9.2 - 39.4
24-h avg
1-hmaxc 24.0-80.5
1 h may 1fi 3 917
—
94 h a\/n
94 h avn 110 Ti 4
94 h avn 91 ^ 97 fl
24-havg 13.5-34.8
94 h avn 1 ^ ft Ti fl
24-h avg 35.4
Upper
Percentile
Concentrations
(PPb)
95th:
45.8-94.0
Max:
62.6-160.7
—
—
90th:
33.1 -132.5
Max:
96.0-111.5
—
—
Max:
55.1 -132.1
90th:
21.7-48.8
Respiratory
November 2013
4-261
DRAFT: Do Not Cite or Quote
-------
Table 4-37 (Continued): Air quality characteristics of studies evaluated in the 2008 ISA for
Oxides of Nitrogen and recently published multicity and select
single-city studies.
Study
Moolqavkar et al.
(2013)
Shinetal. (2012)
Stieb et al. (2008)
Wonqetal. (2010):
Wong et al. (2008b)
Location
72 U.S.
cities9
24
Canadian
cities
12
Canadian
cities
4 Asian
cities
Mortality Averaging
Years Outcome(s) Time
1987-2000 Total 24-h avg
1984-2004 Cardiopulmonary 24-h avg
1981-2000 Total 3-h max
1996-2004h Total, 24-h avg
Cardiovascular,
Respiratory
Mean
Concentration
(PPb)
8.7-25.0
1981-1990:
24.7-42.6
1991-2000:
16.3-39.2
91 9 ?4 fi
Upper
Percent! le
Concentrations
(PPb)
75th:
9K ^ 41 9
Max:
72.6-131.9
"Multicity studies evaluated in the 2008 ISA for Oxides of Nitrogen.
bOf the 90 cities included in the NMMAPS analysis only 58 had NO2 data
°Samoli et al. (2006) estimated 1-h max concentrations for each city by multiplying 24-h avg concentrations by 1.64.
dOf the 7 cities only 4 had NO2 data.
eStudy period varied for each city and encompassed 2 to 7 years. Hong Kong was the only city that had air quality data prior to
2000.
'Only 9 cities (Cagliari was excluded) were included in the formal analysis of examining potential factors that could increase the risk
of mortality due to short-term NO2 exposure.
9Of the 108 cities included in the analyses using NMMAPS data only 72 had NO2 data.
hThe study period varied for each city, Bangkok: 1999-2003, Hong Kong: 1996-2002, and Shanghai and Wuhan: 2001-2004.
1
2
o
J
4
5
6
7
8
9
10
11
12
As demonstrated in Figure 4-17 (Table 4-38). multicity studies evaluated in the 2008 ISA
for Oxides of Nitrogen and those recently published, consistently provide evidence of
positive associations between short-term NO2 exposure and total (nonaccidental)
mortality. In these multicity studies, the associations observed were in analyses that
primarily examined all ages, the exceptions being Chiusolo et al. (2011) and Berglind et
al. (2009), which both focused on the risk of mortality attributed to air pollution in the
population > 35 years of age. Across these studies, associations between short-term NO2
exposure and mortality were examined primarily in the total population; however,
Berglind et al. (2009) focused on a subset of the population, i.e., MI survivors. The large
effect estimate for Berglind et al. (2009) could be attributed to the larger mortality rate
for MI survivors, 30-day mortality rate of 14-15% and 1-year mortality rate of 22-24%,
compared to populations examined in the other multicity studies (Berglind et al.. 2009).
November 2013
4-262
DRAFT: Do Not Cite or Quote
-------
Study
Dominici et al. (2003)
Stieb et al. (2003)
Samoli et al. (2006)
Burnett et al. (2004)
Hoek(2003)
Simpson et al. (2005)
Brocket al. (2007)
Biggeri et al. (2005)
Stieb et al. (2008)
Moolgavkaret al. (2013)
Bellini et al. (2007)
Wong et al. (2008)
Cakmaketal. (2011)
Chenetal. (2012)
Chiusolo et al. (2011)
Berglind et al. (2009)
Location
58 U.S. cities
Meta-analysis
30 European cities
12 Canadian cities
Netherlands
4 Australian cities
10 Canadian cities
8 Italian cities
12 Canadian cities
72 U.S. cities
1 5 Italian cities
4 Asian cities
7 Chilean cities
17 Chinese cities
10 Italian cities
5 European cities
Lag
1
...
0-1
0-2
0-6
0-1
1
0-1
1
1
0-1
0-1
0-6
0-1
0-5
0-1
4 6 8 10
% Increase
Figure 4-17 Summary of multicity studies evaluated in the 2008 ISA for Oxides
of Nitrogen (black circles) and recently published (red circles)
that examined the association between short-term NO2 exposure
and total mortality.
November 2013
4-263
DRAFT: Do Not Cite or Quote
-------
Table 4-38 Corresponding percent increase in total mortality (95% Cl) for
Figure 4-17.
Study
Location
Age
Lag Averaging Time % Increase (95% Cl)
Dominici et al. (2003f
58 U.S. cities
All
1
24-h avg
0.50(0.09, 0.90)
Stieb et al. (2003)a
Meta-analysis
All
24-h avg
0.80(0.20, 1.5)
Samoli et al. (2006)a
30 European
cities
All
0-1
1-h max
1.8(1.3,2.2)
Burnett et al. (2004f
12 Canadian
cities
All
0-2
24-h avg
2.0(1.1,2.9)
Hoek (2003f
the Netherlands
All
0-6
24-h avg
2.6(1.2,4.0)
Simpson et al. (2005bf 4 Australian cities
All
0-1
1-h max
3.4(1.1, 5.7)
Brook et al. (2007f
10 Canadian
cities
All
24-h avg
1-h max
3.5(1.4, 5.5)
Bigger! et al. (2005f
8 Italian cities
All
0-1
3.6(2.3, 5.0)
Stieb et al. (2008)
12 Canadian
cities
All
3-h max
1.9(0.80,2.9)
Moolqavkar et al. (2013) 72 U.S. cities
All
24-h avg
2.1 (1.8,2.3)
Bellini etal. (2007)
15 Italian cities
All
0-1
24-h avg
2.3(1.0, 3.7)
Wong et al. (2008b)
4 Asian cities
All
0-1
24-h avg
4.8(3.3,6.4)
Cakmaketal. (2011b)
7 Chilean cities
All
0-6
24-h avg
6.0(5.3,6.7)
Chen etal. (2012b)
17 Chinese cities
All
0-1
24-h avg
6.4(4.3, 8.6)
Chiusoloetal. (2011)
10 Italian cities
>35
0-5
24-h avg
8.3(3.8, 13.1)
Berqlind et al. (2009)
5 European cities > 35
0-1
24-h avg
11.6 (-5.9, 32.4)
Note:Studies correspond to studies presented in Figure 4-17.
"Multicity studies evaluated in the 2008 ISA for Oxides of Nitrogen.
1
2
3
4
5
6
When focusing on cause-specific mortality, recent multicity studies have reported similar
patterns of associations to those evaluated in the 2008 ISA for Oxides of Nitrogen with
some evidence of larger respiratory mortality risk estimates (Figure 4-18). However, in a
study of 15 Italian cities, Bellini et al. (2007) observed smaller cardiovascular and
respiratory mortality risk estimates compared to total mortality, which contradicts the
results of Biggeri et al. (2005) of which Bellini et al. (2007) is an extension.
November 2013
4-264
DRAFT: Do Not Cite or Quote
-------
Study
Samoli et al. (2006)
Location
30 European cities
Burnett et al. (2004) 12 Canadian cities
Simpson et al. (2005) 4 Australian cities
Biggeri et al. (2005)
Bellini et al. (2007)
Wong et al. (2008)
Chenetal. (2012)
8 Italian cities
15 Italian cities
4 Asian cities
17 Chinese cities
ol8
0-1
0-1
0-2
n o
n o
n i
n i
n i
0-1
0-1
-1
n i ^
0-1
n i
n i
n i
n i
n ^
n ^
i ^
-•-
»,
A
A b
».
4.
A
A
A k
Chiusolo et al. (2011 )a 10 Italian cities
-20 2 4 6 8 10 12 14 16
% Increase
Black symbols = multicity studies evaluated in the 2008 ISA for Oxides of Nitrogen; Red symbols = recent studies.
Circle = total mortality; Diamond = cardiovascular mortality; Triangle = respiratory mortality.
a = Study focused on individuals > 35 years of age while the other studies focused on all ages.
Figure 4-18 Percent increase in total, cardiovascular, and respiratory mortality
from multicity studies for a 20-ppb increase in 24-h avg or 30-ppb
increase in 1-h max NO2 concentrations.
November 2013
4-265
DRAFT: Do Not Cite or Quote
-------
Table 4-39 Corresponding percent increase (95% Cl) for Figure 4-18.
Study
Samoli et al. (2006)a
Burnett et al. (2004f
Simpson et al. (2005bf
Bigger! et al. (2005f
Bellini etal. (2007)
Wong et al. (2008b)
Chen etal. (2012b)
Chiusoloetal. (2011)
Note:Studies correspond to
"Multicity studies evaluated
Averaging
Location Age Lag Time
30 European
cities All 0-1 1-h max
12 Canadian
cities All 0-2 24-h avg
4 Australian
cities All 0-1 1-h max
8 Italian
cities All 0-1 1-h max
15 Italian
cities All 0-1 24-h avg
4 Asian
cities All 0-1 24-h avg
17 Chinese
cities All 0-1 24-h avg
10 Italian
f^JtJpQ (}~O
cmes > 35 24-h avg
1-5
studies presented in Figure 4-18.
in the 2008 ISA for Oxides of Nitrogen.
Mortality
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
Total
Cardiovascular
Respiratory
% Increase
(95% Cl)
1.8(1.3,2.2)
2.3(1.7, 3.0)
2.2(0.98, 3.4)
2.0(1.1,2.9)
2.0(0.53, 3.9)
2.1 (-0.27, 3.9)
3.4(1.1, 5.7)
4.3(0.90, 7.8)
11.4(3.4, 19.7)
3.6(2.3, 5.0)
5.1 (3.0, 7.3)
5.6(0.2, 11.3)
2.3(1.0, 3.7)
1.6 (-1.8, 4.1)
1.5 (-2.4, 6.9)
4.8(3.3,6.4)
5.3(3.5, 7.2)
5.8(2.6, 9.1)
6.4(4.3, 8.6)
7.1 (3.9, 10.3)
10.1 (5.7, 14.5)
8.3(3.8, 13.1)
10.5(6.0, 15.2)
14.1 (2.9,26.4)
4.4.4 Potential Confounding of the NO2-Mortality Relationship
1 A key uncertainty of the NO 2 -mortality relationship identified in the 2008 ISA for
2 Oxides of Nitrogen (U.S. EPA. 2008c) was whether NO2 acts as a surrogate of another
3 unmeasured pollutant. As such, although the multeity studies evaluated in the 2008 ISA
4 for Oxides of Nitrogen reported consistent evidence of an association between short-term
5 NO 2 exposure and mortality that persisted in copollutant models, these studies often
6 concluded that the observed mortality effects could not be attributed solely to NO2.
7 Copollutant analyses conducted in recent studies further attempted to identify whether
8 NO2 has an independent effect on mortality. Additionally, recent studies have examined
November 2013
4-266
DRAFT: Do Not Cite or Quote
-------
1 whether the extent of temporal adjustment employed adequately controls for the potential
2 confounding effects of season on the NO2-mortality relationship.
Copollutant Confounding
3 In the examination of the potential confounding effects of copollutants on the
4 NO2-mortality relationship, it is informative to evaluate whether NO2 risk estimates
5 remain robust in copollutants models and whether NO2 modifies the effect of other
6 pollutants. Recent multicity studies examine the NO2-mortality relationship by taking
7 into consideration both of these aspects in different study designs and in different study
8 locations (i.e., U.S., Canada, Europe, and Asia).
9 In a study of 108 U.S. cities using data from the National Morbidity, Mortality, and Air
10 Pollution Study (NMMAPS) for 1987-2000 (of which 72 had NO2 data), Moolgavkar et
11 al. (2013) used a subsampling approach where a random sample of 4 cities was removed
12 from the 108 cities over 5,000 bootstrap cycles to examine associations between short-
13 term air pollution concentrations and mortality. This approach was used instead of the
14 two-stage Bayesian hierarchical approach employed in the original NMMAPS analysis,
15 which assumes that city-specific risk estimates are normally distributed around a national
16 mean (Dominici et al.. 2003). In a single-pollutant model using 100 degrees of freedom
17 (~7 df/yr, which is consistent with NMMAPS) to control for temporal trends,
18 Moolgavkar etal. (2013) reported a 2.1% (95% CI: 1.8, 2.3) increase in total
19 (nonaccidental) mortality at lag 1 day for a 20-ppb increase in 24-h avg NO2
20 concentrations. In a copollutant analysis, the NO2-mortality risk estimate remained robust
21 to the inclusion of PM10 (1.9% [95% CI: 1.2, 2.5]).
22 Stieb et al. (2008) reported results consistent with Moolgavkar et al. (2013) in a study
23 that focused on the development of a new air quality health index in Canada. Focusing on
24 lag day 1 and models using 10 df per year, Stieb et al. (2008) examined whether
25 copollutants confounded the single-pollutant results in both copollutant and
26 multipollutant models; the study did not clearly identify which results pertained to which
27 model. As stated previously in this ISA, it is important to note that multipollutant models
28 are difficult to interpret due to the multicollinearity often observed between pollutants.
29 However, the results of the copollutant and multipollutant analyses conducted by Stieb et
30 al. (2008) indicate that the NO2-mortality relationship remains robust when adjusted for
31 other pollutants (quantitative results not presented).
32 Additional studies conducted in Europe and Asia also provide evidence indicating that
33 NO2-mortality associations remain robust in copollutants models. In a multicity study
34 conducted with a time-stratified, case-crossover analysis of 10 Italian cities, which is part
November 2013 4-267 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
of the Italian Epi Air multicenter study "Air Pollution and Health: Epidemiological
Surveillance and Primary Prevention," Chiusolo et al. (2011) reported consistent, positive
associations for total and cause-specific mortality (i.e., cardiac, cerebrovascular, and
respiratory), ranging from an 8.3 to 14.1% increase for a 20-ppb increase in 24-h NO2
concentrations using an unconstrained distributed lag of 0-5 days (lag 1-5 days was used
for respiratory mortality). In copollutant analyses, NO2 risk estimates remained robust in
models with PM with all-year data and with O3 with data restricted to the summer season
(i.e., April -September) (Table 4-40V
Table 4-40
Mortality
All natural
Cardiac
Cerebrovascular
Respiratory
Percent increase in total and cause-specific mortality for a 20-ppb
increase in 24-h avg NO2 concentrations in single- and copollutant
models with PMio in all-year analyses or O3 in summer season
analyses.
Season
All-year
April-September
All-year
April-September
All-year
April-September
All-year
April-September
Model
NO2 (lag 0-5)
With PM-io (lag 0-5)
NO2 (lag 0-5)
With O3 (lag 0-5)
NO2 (lag 0-5)
With PM-io (lag 0-5)
NO2 (lag 0-5)
With O3 (lag 0-5)
NO2 (lag 0-5)
With PMio (lag 0-5)
NO2 (lag 0-5)
With O3 (lag 0-5)
NO2 (lag 0-5)
With PM-io (lag 0-5)
NO2 (lag 0-5)
With O3 (lag 0-5)
% Increase (95% Cl)
8.3(3.7, 13.1)
7.7(1.9, 13.9)
18.3(12.6,24.2)
18.7(13.4,24.2)
10.5(6.0, 15.2)
10.3(4.1, 16.8)
19.6(11.7,28.1)
19.3(11.0,28.2)
9.4 (-0.5, 20.2)
10.2 (-2.7, 24.8)
33.8(19.7,49.7)
30.9(14.2, 50.1)
14.1 (2.9,26.4)
13.9(3.0,25.5)
42.4(16.6, 73.9)
44.6(15.0, 81.9)
Note: Concentrations converted from |jg/m to ppb using the conversion factor of 0.532, assuming standard temperature (25 °C) and
pressure (1 atm).
Source: Reprinted from Chiusolo et al. (2011). Environmental Health Perspectives.
9
10
11
12
The Public Health and Air Pollution in Asia (PAPA) study as well as the China Air
Pollution and Health Effects Study (CAPES) collectively found that the NO2-mortality
association remains robust in copollutant models with other criteria air pollutants. The
PAPA study examined the effect of air pollution on mortality in four cities, one in
November 2013
4-268
DRAFT: Do Not Cite or Quote
-------
1 Thailand (i.e., Bangkok) and three in China (i.e., Hong Kong, Shanghai, and Wuhan)
2 (Wong etal.. 2010; Wong et al.. 2008b). In these study locations, PM10 and SO2
3 concentrations are much higher than those reported in the U.S.; however, NO2 and O3
4 concentrations are fairly similar. Copollutant analyses were only conducted in the
5 individual cities, a combined four city analysis was not conducted. In models using lag
6 0-1 days NO2 concentrations in the Chinese cities, NO2 mortality risk estimates were
7 robust to adjustment for copollutants (quantitative results not presented). However, in
8 Bangkok the NO2-mortality risk estimate was attenuated in models with PMi0.
9 The results from the Chinese cities in the PAPA study are consistent with those found in
10 CAPES (Chen et al.. 2012b). In a two-stage Bayesian hierarchical model, where the first
11 stage followed the PAPA protocol, Chen et al. (2012b) reported a 6.4% increase
12 (95% CI: 4.3, 8.6) in total mortality, 7.1% increase (95% CI: 3.9, 10.3) for cardiovascular
13 mortality, and 10.0% increase (95% CI: 5.7, 14.5) for respiratory mortality for a 20-ppb
14 increase in 24-h average NO2 concentrations at lag 0-1 days. Although NO2 was
15 moderately correlated with both PMi0 and SO2, 0.66 and 0.65, respectively, NO2-
16 mortality associations remained robust across total, cardiovascular, and respiratory
17 mortality with the percent increase in mortality ranging from 4.7-6.9% in copollutant
18 models with PMi0 and 5.3-7.2% in models with SO2 for a 20-ppb increase in 24-h
19 average NO2 concentrations.
20 In addition to examining whether copollutants confound the NO2-mortality relationship,
21 studies also conducted analyses to examine if there was any indication that NO2 modifies
22 the PM-mortality relationship. The Air Pollution and Health: A European and North
23 American Approach (APHENA) study, although focused specifically on examining the
24 PM10-mortality relationship, also conducted an analysis to identify whether NO2
25 modifies the PMi0-mortality relationship. In both the European and U.S. datasets, as
26 mean NO2 concentrations and the NO2/PM10 ratio increased, there was evidence that the
27 risk of PMio mortality increased. These results are consistent with Katsouvanni et al.
28 (2003) and Katsouvanni et al. (2001). who reported higher PM risk estimates in cities
29 with higher NO2 concentrations, suggesting that NO2 and PM may be effect modifiers of
30 each other.
Temporal Confounding
31 Recent studies have also examined whether the NO2-mortality relationship is subject to
32 temporal confounding. These studies have focused on examining the effect of increasing
3 3 the number of df employed per year to control for temporal trends on NO 2 -mortality risk
34 estimates. Using the entire dataset, which encompassed the years 1981-2000, Stieb et al.
35 (2008) examined the effect of using an alternative number of df to adjust for seasonal
November 2013 4-269 DRAFT: Do Not Cite or Quote
-------
1 cycles on NO2-mortality risk estimates. In analyses of single-day lags from 0 to 2 days in
2 single-pollutant models, the authors reported comparable risk estimates for each
3 individual lag day when using 6, 8, 10, 12, and 14 df per year. Similar to Stieb et al.
4 (2008). the PAPA study also examined the impact of alternative approaches to
5 controlling for temporal trends on mortality risk estimates. In models using 4, 6, 8, 10, or
6 12 df per year, Wong et al. (2010) also reported relatively similar results across the df per
7 year specified, with some evidence for a slight attenuation of the NO2-mortality
8 association in Wuhan as the df per year increased.
9 Unlike Stieb et al. (2008) and Wong etal. (2010). which conducted a systematic analysis
10 of the influence of increasing the df per year to control for temporal trends on the NO2-
11 mortality relationship, Moolgavkar et al. (2013) only compared models that used 50 df
12 (-3.5 df per year) or 100 df (~7 df peryear) in the statistical model. However, similar to
13 both Stieb et al. (2008) and Wong etal. (2010). the authors reported similar results
14 regardless of the number of df used, 2.0% (95% CI: 1.8, 2.3) for a 20-ppb increase in
15 24-h avg NO2 concentrations at lag Iday in the 50 df model and 2.1% (95% CI: 1.8, 2.3)
16 in the 100 df model.
4.4.5 Modification of the NO2-Mortality Relationship
17 To date, a limited number of studies have examined potential effect measure modifiers of
18 the NO2-mortality relationship. In the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
19 2008c). Samoli et al. (2006) provided evidence of regional heterogeneity in NO2-
20 mortality associations and higher NO2-mortality risk estimates in cities with a lower
21 prevalence of smoking as part of the Air Pollution and Health: A European Approach
22 (APHEA-2) study. Recent multicity studies conducted in Italy (Chiusolo et al.. 2011).
23 Chile (Cakmak et al.. 201 Ib). and Asia (Chen et al.. 2012b) conducted extensive analyses
24 of potential effect measure modifiers of the NO2-mortality relationship, and identified
25 specific factors that may characterize populations potentially at increased risk of NO2-
26 related mortality. Because these studies examine similar effect measure modifiers it
27 should be noted that demographic as well as socioeconomic differences between
28 countries may complicate the interpretation of results across these studies, and
29 subsequently the ability to make generalizations across locations regarding the factors
30 that may modify the NO2-mortality association.
Lifestage
31 Each of the evaluated multicity studies examined whether older adults were at increased
32 risk of mortality in response to short-term NO2 exposures. These studies collectively
November 2013 4-270 DRAFT: Do Not Cite or Quote
-------
1 found evidence of increased risk in the population over the age of 65 years, but the exact
2 age groups examined varied across studies. Comparing the risk of NO2-related mortality
3 across age groups in a multicity study in Italy, Chiusolo et al. (2011). reported that those
4 over 85 years of age were at greatest risk compared with individuals 35-64 years old.
5 Additionally, the authors observed that the risk of NO 2-associated mortality did not
6 follow a consistent trend as age increased over the age of 64. These results are similar to
7 those observed in a multicity study in Chile (Cakmak et al., 201 Ib). Cakmak et al.
8 (20lib) reported thatNO2-mortality risk estimates increased in age groups > 65 years of
9 age, with the magnitude of the effect increasing as age increased over 75 years. Chen et
10 al. (2012b) in CAPES did not examine individual age groups above 65 years of age, but
11 reported increased risk in individuals older than 65 years of age (6.7% [95% CI: 3.5,
12 10.0] for a 20-ppb increase in 24-h average NO2 concentrations at lag 0-1 days)
13 compared to those less than 65 years (3.3% [95% CI: 1.3, 5.3]).
Sex
14 Among the studies that examined potential differences in the risk of NO2-related
15 mortality by sex, both Cakmak et al. (201 Ib) and Chen et al. (2012b) reported some
16 evidence supporting increased risk of NO 2-associated mortality in females compared to
17 males. Cakmak et al. (201 Ib) found NO2 risk estimate for females (7.2%) were slightly
18 higher than those for males (4.7%). Chen et al. (2012b) reported risk estimates similar in
19 magnitude to those observed in Cakmak etal. (20lib) (males: 4.7% [95% CI: 2.3, 7.2];
20 females: 6.8% [95% CI: 2.9, 10.8] for a 20-ppb increase in 24-h avg NO2 concentrations
21 at lag 0-1). However, in Italy, Chiusolo et al. (2011) reported some evidence of effect
22 measure modification of the NO 2-mortality relationship by sex with a slightly larger risk
23 estimate found in males (9.4%) compared to females (6.7%).
Pre-existing Diseases
24 Of the multicity studies evaluated, only Chiusolo et al. (2011) examined whether
25 pre-existing diseases increased the risk of NO2-related mortality. The authors reported
26 the greatest risk in highly diseased individuals (i.e., individuals with 3 or more chronic
27 diseases or who had been hospitalized one month to 2 years prior to death for either
28 chronic or acute cardiopulmonary conditions). In analysis of individual chronic diseases
29 and individuals without the disease as the referent, the strongest evidence for increased
30 risk of NO2-related mortality was for individuals with pre-existing diseases that affect the
31 cardiovascular system (i.e., diabetes, cardiac ischemic disease, diseases of the pulmonary
32 circulation, heart conduction disorders, and heart failure). In a study conducted in 5
33 European cities, Berglind et al. (2009) reported additional evidence supporting this link
November 2013 4-271 DRAFT: Do Not Cite or Quote
-------
1 by demonstrating increased risk of NO2-associated mortality in survivors of myocardial
2 infarction. Although Chiusolo et al. (2011) examined only one pre-existing respiratory
3 condition (i.e., chronic pulmonary diseases), it provided evidence that individuals with
4 pre-existing respiratory conditions also may be at increased risk of NO2-related mortality.
Socioeconomic Status
5 The potential modification of the NO2-mortality relationship by SES was examined in a
6 number of studies, but each used slightly different indicators for income/employment and
7 educational attainment. Chiusolo etal. (2011) examined socioeconomic position and
8 income (both measured as the median of the census block of residence) and reported
9 inconsistent results: increased risk in the low and high socioeconomic position groups,
10 and in the low and middle income groups. Cakmak et al. (201 Ib) examined the effect of
11 income and employment on the risk of NO2-related mortality using community income
12 level and employment category (i.e., unemployed, blue-collar, and white-collar). Across
13 levels of each of these indicators, increased risk of NO2-related mortality was observed
14 for those that were not white-collar workers or had a lower income (Cakmak et al..
15 201 Ib). These results are consistent with Wong et al. (2008b). which found that the most
16 socially deprived areas of Hong Kong, as measured by a composite metric of
17 socioeconomic status, had higher NO2-mortality risks, especially for cardiovascular
18 mortality.
19 Studies that examined the effect of educational attainment on the NO2-mortality
20 relationship consistently reported that low educational attainment increased risk.
21 Although Cakmak et al. (201 Ib) divided educational attainment into more refined
22 categories (i.e., primary school not completed, primary school graduation, high school
23 graduation, some college, and university diploma), the authors only compared low versus
24 high education groups. The authors reported evidence of increased risk in the low
25 education group. Chen etal. (2012b) also reported that low educational attainment
26 increased the risk of NO 2-related mortality using the crude high and low categories,
27 where high corresponded to middle school education and above, and low corresponded to
28 illiterate or primary school.
4.4.6 Potential Seasonal Differences in the NO2-Mortality Relationship
29 Studies evaluated in the 2008 ISA for Oxides of Nitrogen indicated seasonal differences
30 in the NO2-mortality relationship with evidence of larger effects in the warm or summer
31 season. Recent multicity studies conducted in Canada (Shin etal.. 2012; Stieb et al..
32 2008) and Italy (Chiusolo etal.. 2011; Bellini et al.. 2007) further support these previous
November 2013 4-272 DRAFT: Do Not Cite or Quote
-------
1 findings, but also raise additional questions in light of the seasonal patterns in NO2
2 concentrations observed in the U.S. and Canada (i.e., higher concentrations in the winter
3 months compared to the summer months).
4 In the 12 Canadian city study, Stieb et al. (2008) reported that NO2-mortality risk
5 estimates were larger in the warm season (April-September) compared to the cool season
6 (October-March) (quantitative results not presented). These results are consistent with
7 those reported by Shin et al. (2012) in a study that examined year-to-year changes in the
8 association between short-term NO2 exposure and mortality (i.e., cardiopulmonary and
9 non-cardiopulmonary) across 24 Canadian cities 1984-2004. In seasonal analyses, NO2
10 associations with cardiopulmonary mortality at lag 0-2 days were observed to be stronger
11 in the warm season (April-September) compared to the cold season (October-March).
12 Shin etal. (2012) suggest that the larger NO2 mortality effects in the warm season could
13 be due to the role of NO2 in the atmospheric reactions that form O3, and subsequently
14 suggest that the relationship between NO2 and O3 does not allow for a clear assessment
15 of the independent effects of NO2. However, in Canada, as well as the U.S., NO2
16 concentrations are higher in the cold season compared to the warm season. Additionally,
17 NO2 and O3 are not well correlated during the summer (r=~0.35), which makes it less
18 likely O3 is a confounder of the NO2-mortality relationship (Section 4.4.6).
19 To date, U.S.-based studies have not examined whether the seasonal patterns of NO2-
20 mortality associations observed in Canadian multicity studies are similar in the U.S.
21 However, in a study conducted in New York City that examined the association between
22 short-term exposure to air pollution and cardiovascular mortality, Ito etal. (2011)
23 reported similar risk estimates in all-year (1.8% [95% CI: 0.17, 3.3] for a 20-ppb increase
24 in 24-h avg NO2 concentrations at lag 1 day) and seasonal (Warm: 1.8% [95% CI: -0.35,
25 3.9]; Cold: 2.3% [95% CI: 0.0, 4.7]) analyses.
26 Multicity studies conducted in Italy provide evidence consistent with that observed in the
27 Canadian multicity studies. In the MISA-2 study, Bellini et al. (2007) reported larger
28 NO 2 -mortality risk estimates in the summer (April-September) compared to the winter
29 (October-March) for total (6.54% versus 0.98% for a 20-ppb increase in 24-h NO2
30 concentrations at lag 0-1 days), respiratory (9.4% versus -0.04%), and cardiovascular
31 (7.4% versus -0.19%) mortality. In an analysis of 10 Italian cities, Chiusolo etal. (2011)
32 supports the results of Bellini et al. (2007) by indicating larger NO2-mortality risk
33 estimates in the warm season compared to all-year (Table 4-39) for total (non-accidental)
34 mortality and cause-specific mortality (i.e., cardiac, cerebrovascular, and respiratory).
35 The evidence for increased NO2-mortality associations in the warm season, as presented
36 in the Canadian and Italian multicity studies (Shin etal.. 2012; Stieb et al., 2008; Brook
37 et al.. 2007; Burnett et al.. 2004). differs from the seasonal patterns observed in a study
November 2013 4-273 DRAFT: Do Not Cite or Quote
-------
1 conducted in Shanghai as part of the PAPA study (Kan etaL 2010; Kan et al., 2008). The
2 authors reported evidence of increased NO 2 -mortality risk estimates in the cold season
3 compared to the warm for total (nonaccidental) mortality (Cold: 4.9% versus Warm:
4 1.8% for a 20-ppb increase in 24-h avg NO2 at lag 0-1 days), cardiovascular (Cold: 4.9%
5 versus Warm: 1.2%), and respiratory mortality (Cold: 10.6% versus Warm: -5.2%).
6 Across all of the gaseous pollutants examined, mortality risk estimates were double the
7 size or larger in the cool season, whereas PM10 mortality risk estimates were similar
8 across seasons except for respiratory mortality (larger in the cool season). The authors
9 speculate these seasonal differences could be due to seasonal exposure differences
10 specific to Shanghai, i.e., limited time spent outdoors and increased air conditioning use
11 in the warm season because of high temperature and humidity and heavy rain, versus
12 more time spent outdoors and open windows in the cool season (Kanetal.. 2010; Kan et
13 al.. 2008). The results of (Kan etal. 2010; Kan et al.. 2008) highlight the complexity of
14 clearly identifying seasonal patterns in NO2-mortality associations across locations with
15 drastically different seasonal weather patterns.
4.4.7 NO2-Mortality Concentration-Response (C-R) Relationship and Related
Issues
Lag Structure of Associations
16 The 2008 ISA for Oxides of Nitrogen found consistent evidence across studies indicating
17 that NO2-mortality effects occur within the first few days after exposure, with multiple
18 studies demonstrating the largest effect occurring the day after exposure (i.e., lag 1 day)
19 (U.S. EPA. 2008c). Recent multicity studies have conducted additional analyses
20 examining multiday lags, which further inform the lag structure of associations between
21 short-term NO2 exposure and mortality.
22 Chiusolo et al. (2011) in the analysis of 10 Italian cities examined the lag structure of
23 associations between mortality and short-term NO2 exposure through both single-day and
24 multiday lag analyses. Multiday analyses consisted of a priori defined lags (i.e., 0-1, 2-5,
25 and 0-5 days) examined using an unconstrained distributed lag model. In addition to
26 examining single-day lags of 0 to 5 days, the authors also explored the pattern of
27 associations observed over each individual day using a constrained polynomial
28 distributed lag model. It is important to note that the individual lag days of a constrained
29 distributed lag model are not directly interpretable; however, this analysis allowed
30 Chiusolo et al. (2011) to visually display the potential latency of the NO2 effect on
31 mortality. Collectively, the single- and multi-day lag analyses support an immediate
November 2013 4-274 DRAFT: Do Not Cite or Quote
-------
effect of NO2 on mortality, but also provide evidence for a delayed effect extending out
to 5 days for all mortality outcomes (Figure 4-19).
3.00
2.50
g 2.00
CO
£ 1.50
o
•5 1.00
5 0.50
2
QS n
w U
HB
-0.50
i fin
All natural mortality
.
C
I
0
-L
i
T '
J J
LT I
'
]
I
o Single-lag models
o Distributed-lag models
0
1 2
3
j
I 1
I
X T
1
"
'
i
i
i
.
Cardiac mortality
T T T T
1 T 1
IT tr Tl I ? L '
T li 1? 1 II
I? i i 4 i
c
"
i
1
3
J
1
_
1 5
450-12-50-5 0123450-1 2-5 0-5
Lag (days)
5.00
4.00
£
g 3.00
| 2.00
•™
S 1.00
§ 0
x
Q.
-1.00
200
Lag (days)
Cerebrovascular mortality
--
T
nl t
T
1
0
,T I 1
T nT
T IT c
-l rj
1 2
ij
™
3
)T C
1
I ji
I J-I
i ,
i
^
i
"
Respiratory mortality 1
i
T T
T -r 1
1 T i r
LL T IT rj
11 W i I
IT
i
i
C
]
C
i
450-12-50-5 012345 0-1 2-5 0-5
Lag (days)
Lag (days)
Source: Reprinted from Chiusolo et al. (2011) in Environmental Health Perspectives.
Figure 4-19 Percent increase in total and cause-specific mortality due to
short-term NOa exposure at single day lags, individual lag days of
a constrained polynomial distributed lag model, and multi-day
lags of an unconstrained distributed lag model.
3
4
5
Chenetal. (2012b) also conducted an extensive analysis of the lag structure of
associations for the NO2-mortality relationship as part of CAPES. Multiday lags were
examined by averaging multiple single lag days and using a constrained polynomial
distributed lag model of 0-4 days. Chenetal. (2012b) reported the largest effect at single
day lags of 0 and 1 and the average of lags 0-1 days indicating an immediate effect of
NO 2 on mortality (Figure 4-20). However, the similar or larger magnitude 0-4 day
average and distributed lag model results provide some evidence for a delayed NO2 effect
November 2013
4-275
DRAFT: Do Not Cite or Quote
-------
on total, cardiovascular, and respiratory mortality, which is consistent with the results of
Chiusolo et al. (2011) (Figure 4-19).
4.0 -i
S^ 3.5-
'Hs 3.0-
t±
o
E 2.5-
c
CD 2.0 -
C/5
CO
2 1.5-
u
c
~ 1.0-
CD
% °-5-
CD
Q_
0.0-
-0.5-
i
I
0
T
J_
1
I
f
2
_.
I
i
J_
3
Total
4
t
01
<
i
i
j_
04 PLIVJ
mortality
4
1
0
(
1
i
1
2
c
I
3
t
L
1
\
<
i
••
01
^
•
t
.
04 PLIVJ
Cardiovascular mortality
i
I
0
«
I
1
<
i
2
Respi
<
'
i
3
i
)
4
<
i
01
1
< i
i
••
04 piiv
ratory mortality
Percent increase (mean and 95% Cl) of daily mortality associated with a 10 |jg/m3 (5.3 ppb) increase of NO2 concentrations, using
different lag structures in the CAPES cities. Multi day average lags (01 [corresponds to 2-day moving average] and lag 04
[corresponds to 5-day moving average of NO2 concentration of the current and previous 4 days]). DIM: polynomial distributed lag
model, representing the cumulative effects of NO2.
Source: Reprinted with permission of Elsevier Ltd. Chen et al. (2012b)
Figure 4-20 Percent increase in total and cause-specific mortality due to
short-term NO2 exposure in single- and multi-day lag models.
3
4
5
6
7
8
9
10
11
12
13
14
Additional studies that examined the effect of NO2 on mortality at single-day lags or
multiday averages provide evidence that is consistent with those studies evaluated in the
2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). which demonstrated strong
associations between NO2 and mortality at lag 1. In the analysis of 12 Canadian cities,
Stieb et al. (2008) found the strongest association between short-term NO2 exposure and
mortality at lag 1 when examining single-day lags of 0-2 days. Wong et al. (2008b) and
Wong etal. (2010) examined single and multi-day lags in each individual city in the
PAPA study. In the 3 Chinese cities, similar to Stieb et al. (2008). the authors reported
evidence of immediate effects of NO2 on mortality; with the strongest association
occurring for a 0-1 day lag. However, in Bangkok, the lag structure of associations was
different and more in line with those observed in Chiusolo et al. (2011) and Chen et al.
(2012b). with the strongest association occurring at a lag of 0-4 days.
November 2013
4-276
DRAFT: Do Not Cite or Quote
-------
C-R Relationship
1 The studies evaluated in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) that
2 examined the association between short-term NO2 exposure and mortality did not
3 conduct formal analyses of the C-R relationship. Recent studies published since the
4 completion of the 2008 ISA for Oxides of Nitrogen have examined the NO2-mortality
5 C-R relationship in both multi- and single-city analyses, focusing on the shape of the C-R
6 curve and whether a threshold exists.
7 Using a subsampling approach, Moolgavkar et al. (2013) examined the shape of the C-R
8 relationship between short-term air pollution exposures and mortality in the NMMAPS
9 dataset by applying a nonlinear function (i.e., natural splines with 6 df) to each pollutant.
10 This analysis provides support for a linear relationship between short-term NO2
11 exposures and mortality (Figure 4-21). Although Moolgavkar et al. (2013) state that the
12 C-R relationship for NO2 "suggest[s] non-linearity and threshold like behavior" the
13 widening of the confidence intervals at the tails of the distribution prevent a clear
14 interpretation of the shape of the curve at the tails of the distribution where the data
15 density is low. It should be noted that the confidence intervals approach zero at the low
16 end of the NO2 distribution due to the way the model is structured.
November 2013 4-277 DRAFT: Do Not Cite or Quote
-------
0.06
0.04
0.02
-0.02
20
40
Lag-1 NO.
60
80
Source: Reprinted from Moolgavkar et al. (2013). Environmental Health Perspectives.
Figure 4-21 Flexible ambient C-R relationship between short-term NO2 (ppb)
exposure and mortality at lag day 1. Pointwise means and 95%
CIs adjusted for size of the bootstrap sample.
i
2
o
J
4
5
6
9
10
11
12
The evidence for a linear C-R relationship between short-term NO2 exposure and
mortality was further supported by Stieb et al. (2008) in a pooled analysis of 12 Canadian
cities. The authors examined three functional forms (i.e., linear, quadratic, and cubic
polynomial) and assessed the model fit using the sum of the Akaike Information Criterion
(AIC). Stieb et al. (2008) indicated that the linear function was the best fit of the
NO2-mortality relationship (quantitative results not presented).
Multicity studies conducted in Asia examined the NO2-mortality C-R relationship
through either a combined analysis using data from all cities or by examining the C-R
relationship in individual cities. Chen et al. (2012b) examined the shape of the
NO2-mortality C-R curve across all cities as part of CAPES for total, cardiovascular, and
respiratory mortality using 24-h NO2 concentrations at lag 0-1 days. To limit the
influence of extreme NO2 concentrations on the shape of the C-R curve, concentrations
November 2013
4-278
DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
7
greater than 120 (ig/m3 (62.4 ppb), which represented only 3% of the data, were
excluded. The authors used a cubic spline with two knots at different concentrations for
each of the mortality outcomes (40 (ig/m3 [20.8 ppb] and 70 (ig/m3 [36.4 ppb] for total
mortality, 50 (ig/m3 [26.0 ppb] and 70 (ig/m3 [36.4 ppb] for cardiovascular mortality, and
40 (ig/m3 [20.8 ppb] and 60 (ig/m3 [31.2 ppb] for respiratory mortality). Chen et al.
(2012b) found evidence of a linear relationship between short-term NO2 exposure and
total and cause-specific mortality (Figure 4-22). which was confirmed by the lack of a
statistically significant difference in the deviance between the spline and linear fit
models.
>\
"ro
o
E
_>
"ro
T3
W
Q.
Total mortality
Cardiovascular mortality
Respiratory mortality
I
20
I I I I
40 60 80 100
Two-day average N02 concentrations
i
120
Source: Reprinted with permission of Elsevier Ltd. (Chen et al.. 2012b).
Figure 4-22 CAPES C-R curve for the association between total and cause-
specific mortality and 24-h avg NO2 concentrations at lag 0-1
days. NO2 concentrations on the x-axis are in the unit of ug/m3.
10
11
12
The four-city PAPA study (Wong et al.. 2010; Wong et al., 2008b) also examined the
NO 2-mortality C-R relationship, but only focused on the shape of the C-R curve in each
individual city. The C-R curve for the NO2-mortality relationship was assessed by
November 2013
4-279
DRAFT: Do Not Cite or Quote
-------
1 applying a natural spline smoother with 3 df to NO2 concentrations. To examine whether
2 the NO2-mortality relationship deviates from linearity, the deviance between the
3 smoothed (nonlinear) pollutant model and the unsmoothed (linear) pollutant model was
4 examined. The C-R curves in the three Chinese cities further supports the results from
5 Stieb et al. (2008) and Chenet al. (2012b) by indicating a linear relationship between
6 short-term NO2 concentrations and mortality (Figure 4-23). Specifically, the evidence for
7 linearity was strongest between the 25th and 75th percentiles of the NO2 concentrations
8 in each city with some uncertainty in the shape of the C-R curve at lower concentrations
9 where the data density is low, generally below the 25th percentile. The results of the
10 analysis for Bangkok, which provides evidence for non-linearity, are consistent with what
11 has been observed in examinations of city-specific C-R curves for other air pollutants
12 (e.g., PM and O3). That is, the heterogeneity in city-specific risk estimates can translate
13 into heterogeneity in the shape of the C-R curve, which has often been hypothesized to be
14 due to city-specific exposure characteristics and demographics. The results from the
15 Bangkok analysis highlight the difficulty in interpreting a combined C-R curve across
16 cities, when there is evidence for city-to-city differences in the association between short-
17 term NO2 exposure and mortality.
November 2013 4-280 DRAFT: Do Not Cite or Quote
-------
-0.
0.3
0.2
0.1
0.0
-0.1
Hong Kong
20 40 60 80 100 120
N02 concentration (ug/m3)
20 40 60 80 100 120 140
N02 concentration (ug/m3)
0.3
0.2
M
V)
•= 0.1
ro
o
0.0
-0.1
Shanghai
0.3
0.2
0.1
0.0
-0.1
Wuhan
^
^^
50 100 150 200
N02 concentration (ug/m3)
20 40 60 80 100 120
N02 concentration (ug/m3)
Note: Thin vertical lines represent interquartile range of NO2 concentrations in each city. The thick line was included by Wong et al.
(2008b) to depict where the WHO 1-year averaging time standard for NO2 of 40 ug/m3 (20.8 ppb) could be found along the
distribution of NO2 concentrations in each city.
Source: Reprinted from Wong et al. (2008b). Environmental Health Perspectives.
Figure 4-23 C-R curve for association between total mortality and 24-h avg
NOa concentrations at lag 0-1 days in the four cities of the PAPA
study.
3
4
5
4.4.8 Summary and Causal Determination
Recent multicity studies evaluated since the completion of the 2008 ISA for Oxides of
Nitrogen continue to provide consistent evidence of positive associations between short-
term NO2 exposures and total mortality. This collective evidence indicates that there is
likely to be a causal relationship between short-term NO2 exposures and total mortality.
This conclusion represents a change from the 2008 ISA for Oxides of Nitrogen that
November 2013
4-281
DRAFT: Do Not Cite or Quote
-------
1 concluded the evidence "was suggestive but not sufficient to infer a causal relationship"
2 (U.S. EPA. 2008c). The recent multi-city studies evaluated inform key uncertainties and
3 limitations in the NO2-mortality relationship identified in the 2008 ISA for Oxides of
4 Nitrogen including confounding, modification of the NO 2 -mortality relationship,
5 potential seasonal differences in NO2 -mortality associations, and the shape of the
6 NO2-mortality C-R relationship. However, questions remain regarding whether NO2 is
7 independently associated with mortality. This section describes the evaluation of
8 evidence for total mortality, with respect to the causal determination for short-term
9 exposure to oxides of nitrogen using the framework described in Table II of the
10 Preamble.The key evidence, as it relates to the causal framework, is summarized in Table
11 4-41.
12 Collectively, the evidence from recent multicity studies of short-term NO2 exposures and
13 mortality consistently demonstrate the NO 2-mortality association is robust in copollutant
14 models. However, it should be noted that it is difficult to disentangle the independent
15 effects of NO2 from those of other measured or unmeasured pollutants that also
16 contribute to traffic-related pollution (Section 1.5). adding uncertainty to the
17 interpretation of the association between NO2 and total mortality. In addition, studies that
18 focused on PM and examined whether NO2 modified the PM-mortality relationship
19 reported that PM risk estimates increased as NO2 concentrations increased or the ratio of
20 NO2/PM increased. These results suggest that NO2 and PM may be effect modifiers of
21 each other. This is consistent with the conclusions of the 2008 ISA for Oxides of
22 Nitrogen (U.S. EPA. 2008c). Additionally, recent studies examined the potential
23 confounding effects of inadequate control for temporal trends and reported similar NO2-
24 mortality risk estimates across a range of degrees of freedom per year.
25 An examination of factors that may contribute to increased risk of NO2-related mortality
26 found evidence that older adults (> 65 years of age), females, individuals with
27 pre-existing cardiovascular or respiratory diseases, and individuals of lower SES,
28 specifically lower income and educational attainment, are at greater risk. Studies that
29 examined whether there are seasonal differences in the NO2-mortality relationship found
30 greater effects in the warm or summer months in multicity studies conducted in Canada
31 and Europe. However, these results are contradicted by a study conducted in Asia where
32 larger effects were observed in the cold season. These between-study differences in
33 seasonal associations are more than likely a reflection of the different seasonal weather
34 patterns observed between countries (Kanet al.. 2010; Kan et al. 2008).
35 Those studies that examined the lag structure of associations for the NO2-mortality
36 relationship observed that there continues to be evidence of an immediate effect (i.e., lag
37 0 to 1 day), which is consistent with studies evaluated in the 2008 ISA for Oxides of
November 2013 4-282 DRAFT: Do Not Cite or Quote
-------
1 Nitrogen. Recent studies also provided evidence for a delayed effect on mortality in
2 distributed lag models with lags ranging from 0-4 to 0-5 days (Chen et al.. 2012b:
3 Chiusolo et al., 2011). Multicity studies have examined the shape of the C-R relationship
4 and whether a threshold exists in both a multi- and single-city setting. These studies have
5 used different statistical approaches and consistently demonstrated a linear relationship
6 with no evidence of a threshold within the range of NO2 concentrations currently found
7 in the U.S. However, consistent with observations from C-R analyses conducted for other
8 criteria pollutants [e.g., PM (U.S. EPA. 2009a) and O3 (U.S. EPA. 2013b)1. an
9 examination of the C-R relationship in individual cities, specifically in China, has
10 demonstrated heterogeneity in the shape of the curve across cities (Wong et al., 2010;
11 Wong et al.. 2008R
12 In conclusion, the recent epidemiologic studies build upon and support the consistent
13 positive associations between short-term NO2 exposures and total mortality presented in
14 the 2008 ISA for Oxides of Nitrogen. These associations were found to remain generally
15 robust (i.e., positive, and almost unchanged compared to single-pollutant model results)
16 in the numerous studies that conducted copollutant analyses. However, studies that
17 focused on PM and examined whether NO2 modified the PM-mortality relationship
18 suggest that NO2 and PM may be effect modifiers of each other. Although the collective
19 evidence supports an independent effect of short-term NO2 exposures on all-cause
20 mortality, the biological mechanism that could lead to mortality as a result of short-term
21 NO2 exposures has not been clearly characterized. This is evident when evaluating the
22 underlying health effects (i.e., cardiovascular effects in Section 4.3 and respiratory effects
23 in Section 4.2) that could lead to cardiovascular (-35% of total mortality) and respiratory
24 (-9% of total mortality) mortality, the components of total mortality most thoroughly
25 evaluated (Hovert and Xu, 2012). An evaluation of epidemiologic studies that examined
26 the relationship between short-term NO2 exposure and cardiovascular effects found
27 evidence for increases in cardiovascular-related hospital admissions and ED visits,
28 specifically for IHD. However, there is limited evidence from experimental studies for
29 NO2-induced cardiovascular effects. Together, the evidence does not clearly describe a
30 biologically plausible mechanism to support NO2-induced cardiovascular mortality.
31 There is stronger evidence for NO 2-induced respiratory effects with toxicological and
32 controlled human exposure studies demonstrating AHR (Section 4.2.2) in response to
33 short-term NO2 exposures, as well as epidemiologic studies reporting respiratory-related
34 morbidity including hospital admissions and ED visits, specifically for asthma (Section
35 4.2.6). However, the biological mechanism that explains the continuum of effects that
36 could lead to respiratory-related mortality also remains unclear. The robust evidence
37 observed across various multeity studies in combination with the uncertainty in the
38 biological mechanism that could lead to NO2-induced mortality forms a collective body
November 2013 4-283 DRAFT: Do Not Cite or Quote
-------
1 of evidence that is sufficient to conclude there is likely to be a causal relationship
2 between short-term NO2 exposure and total mortality.
November 2013 4-284 DRAFT: Do Not Cite or Quote
-------
Table 4-41 Summary of evidence supporting a likely to be a causal relationship
between short-term NO2 exposure and total mortality.
Rationale for Causal
Determination3
Consistent associations
from multiple, high
quality epidemiologic
studies at relevant
concentrations
Key Evidence13 Key References'3
Increases in mortality in multicitv Section 4.4.3
studies conducted in the U.S., Table 4-37
Canada, South America, Europe, and
Asia.
NO2
Concentrations
Associated with
Effects0
Mean 24-h avg:
9.2-55.0 ppb
Mean 1-h max:
16.3-80.5 ppb
Mean 3-h max:
16. 3-42.6 ppb.
Table 4-37
Additional epidemiologic
evidence to reduce
chance, confounding,
and other biases
NC>2 associations remain robust with
adjustment for other combustion-
related pollutants in copollutant
models.
Moolqavkar et al. (2013):
Chenetal. (2012b):
Chiusolo et al. (2011):
Wonqetal. (2010): Stieb
et al. (2008): Wong et al.
(2008b)
NO2 and PM may be effect modifiers
of each other.
Katsouvanni et al. (2009):
Katsouvanni et al. (2003):
Katsouvanni et al. (2001)
Uncertainty due to
limited coherence with
morbidity evidence
Limited coherence and biological
plausibility for NO2-induced
cardiovascular mortality (-35% total
mortality1). Epidemiologic evidence
for increases in cardiovascular-
related hospital admissions and ED
visits, specifically for IHD, but
inconsistent epidemiologic and
experimental evidence for other
endpoints and weak biological
plausibility for cardiovascular effects,
including mode of action.
Section 4.3.9
Table 4-36
Limited coherence for NO2-induced
respiratory mortality (-8% total
mortality1).Consistent evidence for
increases in respiratory morbidity
across disciplines, particularly for
asthma morbidity. Less robust
evidence for COPD and respiratory
infection.
Section 4.2.9
Table 4-23
"Based on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causal determination and,
where applicable, to uncertainties or inconsistencies. References to earlier sections indicate where full body of evidence is
described.
""Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
dStatistics taken from American Heart Association (2011)
November 2013
4-285
DRAFT: Do Not Cite or Quote
-------
References for Chapter 4
Adamkiewicz. G: Ebelt. S: Syring. M; Slater. J; Speizer. FE; Schwartz. J; Suh. H; Gold. DR. (2004).
Association between air pollution exposure and exhaled nitric oxide in an elderly population. Thorax
59: 204-209. http://dx.doi.org/10.1136/thorax.2003.006445
Adams. WC: Brookes. KA; Schelegle. ES. (1987). Effects of NO2 alone and in combination with O3 on
young men and women. J Appl Physiol 62: 1698-1704.
Aga. E; Samoli. E; Touloumi. G: Anderson. HR; Cadum. E; Forsberg. B; Goodman. P; Goren. A;
Kotesovec. F; Kriz. B; Macarol-Hiti. M; Medina. S: Paldv. A; Schindler. C: Sunyer. J: Tittanen. P;
Wojtvniak. B; Zmirou. D; Schwartz. J: Katsouvanni. K. (2003). Short-term effects of ambient particles
on mortality in the elderly: Results from 28 cities in the APHEA2 project. Eur Respir J 21: 28s-33s.
http://dx.doi.org/10.1183/09031936.03.00402803
Ahmed. T; Danta. I; Dougherty. RL; Schreck. R; Sackner. MA. (1983a). Effect of NO2 (0.1 ppm) on
specific bronchial reactivity to ragweed antigen in subjects with allergic asthma [Abstract]. Am Rev
Respir Dis 127: 160.
Ahmed. T; Dougherty. R; Sackner. MA. (1983b). Effect of 0.1 ppm NO2 on pulmonary functions and non-
specific bronchial reactivity of normals and asthmatics [final report]. (CR-83/11/BI). Warren, MI:
General Motors Research Laboratories.
American Heart Association. (2011). Cardiovascular disease statistics. Available online at
http://web.archive.org/web/20110410093532/http://www.am ericanheart.org/presenter. jhtml?identifier=4
478 (accessed April 12, 2011).
Amoruso. MA; Witz. G: Goldstein. BD. (1981). Decreased superoxide anion radical production by rat
alveolar macrophages following inhalation of ozone or nitrogen dioxide. Life Sci28: 2215-2221.
Andersen. ZJ: Loft. S: Ketzel. M; Stage. M; Scheike. T; Hermansen. MN; Bisgaard. H. (2008a). Ambient
air pollution triggers wheezing symptoms in infants. Thorax 63: 710-716.
http://dx.doi.org/10.1136/thx.2007.085480
Andersen. ZJ: Olsen. TS: Andersen. KK; Loft. S: Ketzel. M; Raaschou-Nielsen. O. (2010). Association
between short-term exposure to ultrafine particles and hospital admissions for stroke in Copenhagen,
Denmark. Eur Heart J 31: 2034-2040. http://dx.doi.org/10.1093/eurheartj/ehql88
Andersen. ZJ: Wahlin. P: Raaschou-Nielsen. O; Ketzel. M; Scheike. T; Loft. S. (2008b). Size distribution
and total number concentration of ultrafine and accumulation mode particles and hospital admissions in
children and the elderly in Copenhagen, Denmark. Occup Environ Med 65: 458-466.
http://dx.doi.org/10.1136/oem.2007.033290
Anderson. HR; Armstrong. B; Hajat. S: Harrison. R; Monk. V; Poloniecki. J: Timmis. A; Wilkinson. P.
(2010). Air pollution and activation of implantable cardioverter defibrillators in London. Epidemiology
21: 405-413. http://dx.doi.org/10.1097/EDE.Ob013e3181d61600
Annesi-Maesano. I; Rouve. S: Desquevroux. H; Jankovski. R; Klossek. JM; Thibaudon. M; Demoly. P:
Didier. A. (2012b). Grass pollen counts, air pollution levels and allergic rhinitis severity. Int Arch
Allergy Immunol 158: 397-404. http://dx.doi.org/10.1159/000332964
Arbex. MA; de Souza Conceicao. GM; Cendon. SP; Arbex. FF; Lopes. AC: Moyses. EP; Santiago. SL;
Saldiva. PHN; Pereira. LAA; Braga. ALF. (2009). Urban air pollution and chronic obstructive
pulmonary disease-related emergency department visits. J Epidemiol Community Health 63: 777-783.
http://dx.doi.org/10.1136/iech.2008.078360
November 2013 4-286 DRAFT: Do Not Cite or Quote
-------
Atkinson. RW: Bremner. SA; Anderson. HR; Strachan. DP; Bland. JM; Ponce de Leon. A. (1999). Short-
term associations between emergency hospital admissions for respiratory and cardiovascular disease and
outdoor air pollution in London. Arch Environ Health 54: 398-411.
http://dx.doi.org/10.1080/00039899909603371
ATS (American Thoracic Society). (2000a). Guidelines for methacholine and exercise challenge testing-
1999. Am J Respir Crit Care Med 161: 309-329. http://dx.doi.Org/10.1164/airccm.161.l.atsll-99
ATSDR (Agency for Toxic Substances and Disease Registry). (2006). A study of ambient air contaminants
and asthma in New York City: Part A and B. Atlanta, GA: U.S. Department of Health and Human
Services, http://permanent.access.gpo.gov/lps88357/ASTHMA BRONX FINAL REPORT.pdf
Avol. EL; Linn. WS: Peng. RC: Valencia. G: Little. D; Hackney. JD. (1988). Laboratory study of asthmatic
volunteers exposed to nitrogen dioxide and to ambient air pollution. Am Ind Hyg Assoc J 49: 143-149.
http://dx.doi.org/10.1080/15298668891379530
Avol. EL; Linn. WS: Peng. RC: Whynot. JD: Shamoo. DA; Little. DE; Smith. MN; Hackney. JD. (1989).
Experimental exposures of young asthmatic volunteers to 0.3 ppm nitrogen dioxide and to ambient air
pollution. Toxicol Ind Health 5: 1025-1034.
Ayaz. KL; Csallany. AS. (1978). Long-term NO2 exposure of mice in the presence and absence of vitamin
E. II. Effect of glutathione peroxidase. Arch Environ Occup Health 33: 292-296.
Azadniv. M; Utell. MJ; Morrow. PE; Gibb. FR; Nichols. J: Roberts. NJ. Jr; Speers. DM; Torres. A; Tsai. Y;
Abraham. MK; Voter. KZ; Frampton. MW. (1998). Effects of nitrogen dioxide exposure on human host
defense. Inhal Toxicol 10: 585-601. http://dx.doi.org/10.1080/089583798197556
Baccarelli. A; Zanobetti. A; Martinelli. I; Grillo. P; Hou. L; Lanzani. G: Mannucci. PM; Bertazzi. PA;
Schwartz. J. (2007). Air pollution, smoking, and plasma homocysteine. Environ Health Perspect 115:
176-181. http://dx.doi.org/10.1289/ehp.9517
Bagheri Lankarani. N; Kreis. I; Griffiths. DA. (2010). Air pollution effects on peak expiratory flow rate in
children. Iran J Allergy Asthma Immunol 9: 117-126.
Baja. ES: Schwartz. JD: Wellenius. GA; Coull. BA; Zanobetti. A; Vokonas. PS: Suh. HH. (2010). Traffic-
related air pollution and QT interval: Modification by diabetes, obesity, and oxidative stress gene
polymorphisms in the Normative Aging Study. Environ Health Perspect 118: 840-846.
http://dx.doi.org/10.1289/ehp.0901396
Ballester. F; Rodriguez. P; Iniguez. C: Saez. M; Daponte. A; Galan. I; Taracido. M; Arribas. F; Bellido. J:
Cirarda. FB; Canada. A; Guillen. JJ; Guillen-Grima. F; Lopez. E; Perez-Hoyos. S: Lertxundi. A; Toro.
S.. (2006). Air pollution and cardiovascular admisisons association in Spain: Results within the
EMECAS project. J Epidemiol Community Health 60: 328-336.
http://dx.doi.org/10.1136/iech.2005.037978
Ballester. F; Tenias. JM; Perez-Hoyos. S. (2001). Air pollution and emergency hospital admissions for
cardiovascular diseases in Valencia, Spain. J Epidemiol Community Health 55: 57-65.
http://dx.doi.0rg/10.1136/iech.55.l.57
Barck. C: Lundahl. J: Hallden. G: Bylin. G. (2005 a). Brief exposures to NO2 augment the allergic
inflammation in asthmatics. Environ Res 97: 58-66. http://dx.doi.Org/10.1016/i.envres.2004.02.009
Barck. C: Lundahl. J: Holmstrom. M; Bylin. G. (2005b). Does nitrogen dioxide affect inflammatory
markers after nasal allergen challenge? Am J Rhinol 19: 560-566.
Barck. C: Sandstrom. T; Lundahl. J: Hallden. G: Svartengren. M; Strand. V: Rak. S: Bylin. G. (2002).
Ambient level of NO2 augments the inflammatory response to inhaled allergen in asthmatics. Respir
Med 96: 907-917.
Barclay. JL; Miller. BG: Dick. S: Dennekamp. M; Ford. I; Hillis. GS: Ayres. JG: Seaton. A. (2009). A panel
study of air pollution in subjects with heart failure: negative results in treated patients. Occup Environ
Med 66: 325-334. http://dx.doi.org/10.1136/oem.2008.039032
November 2013 4-287 DRAFT: Do Not Cite or Quote
-------
Barnes. PJ: Liew. FY. (1995). Nitric oxide and asthmatic inflammation [Review]. Immunol Today 16: 128-
130. http://dx.doi.org/10.1016/0167-5699(95)80128-6
Barnett. AG: Williams. GM; Schwartz. J; Best. TL; Neller. AH; Petroeschevsky. AL; Simpson. RW. (2006).
The effects of air pollution on hospitalizations for cardiovascular disease in elderly people in Australian
and New Zealand cities. Environ Health Perspect 114: 1018-1023. http://dx.doi.org/10.1289/ehp.8674
Barraza-Villarreal. A; Sunyer. J; Hernandez-Cadena. L; Escamilla-Nunez. MC: Sienra-Monge. JJ: Ramirez-
Aguilar. M; Cortez-Lugo. M; Holguin. F; Diaz-Sanchez. D; Olin. AC: Romieu. I. (2008). Air pollution,
airway inflammation, and lung function in a cohort study of Mexico City schoolchildren. Environ
Health Perspect 116: 832-838. http://dx.doi.org/10.1289/ehp.10926
Barth. PJ: Miiller. B. (1999). Effects of nitrogen dioxide exposure on Clara cell proliferation and
morphology. Pathol Res Pract 195: 487-493. http://dx.doi.org/10.1016/S0344-0338(99)80052-l
Barth. PJ: Muller. B; Wagner. U; Bittinger. A. (1995). Quantitative analysis of parenchyma! and vascular
alterations in NO2-induced lung injury in rats. Eur Respir J 8: 1115-1121.
http://dx.doi.org/10.1183/09031936.95.08071115
Bauer. MA; Utell. MJ; Morrow. PE; Speers. DM; Gibb. FR. (1986). Inhalation of 0.30 ppm nitrogen
dioxide potentiates exercise-induced bronchospasm in asthmatics. Am Rev Respir Dis 134: 1203-1208.
Belanger. K; Holford. TR; Gent. JF; Hill. ME; Kezik. JM; Leaderer. BP (2013). Household levels of
nitrogen dioxide and pediatric asthma severity. Epidemiology 24: 320-330.
http://dx.doi.org/10.1097/EDE.Ob013e318280e2ac
Bell. ML; Lew. JK; Lin. Z. (2008). The effect of sandstorms and air pollution on cause-specific hospital
admissions in Taipei, Taiwan. Occup Environ Med 65: 104-111.
http://dx.doi.org/10.1136/oem.2006.031500
Bellini. P; Baccini. M; Biggeri. A; Terracini. B. (2007). The meta-analysis of the Italian studies on short-
term effects of air pollution (MISA): Old and new issues on the interpretation of the statistical
evidences. Environmetrics 18: 219-229. http://dx.doi.org/10.1002/env.836
Berglind. N; Bellander. T; Forastiere. F; von Klot. S: Aalto. P; Elosua. R; Kulmala. M; Lanki. T; Lowel. H;
Peters. A; Picciotto. S: Salomaa. V: Stafoggia. M; Sunyer. J: Nyberg. F. (2009). Ambient air pollution
and daily mortality among survivors of myocardial infarction. Epidemiology 20: 110-118.
http://dx.doi.org/10.1097/EDE.Ob013e3181878b50
Berhane. K; Zhang. Y; Linn. WS: Rappaport. EB; Bastain. TM; Salam. MT; Islam. T; Lurmann. F;
Gilliland. FD. (2011). The effect of ambient air pollution on exhaled nitric oxide in the Children's
Health Study. Eur Respir J 37: 1029-1036. http://dx.doi.org/10.1183/09031936.00081410
Bhaskaran. K; Haiat. S: Armstrong. B; Haines. A; Herrett. E; Wilkinson. P; Smeeth. L. (2011). The effects
of hourly differences in air pollution on the risk of myocardial infarction: case crossover analysis of the
MINAP database. B MJ 343: d5531. http://dx.doi.org/10.1136/bmj.d5531
Biggeri. A; Baccini. M; Bellini. P; Terracini. B. (2005). Meta-analysis of the Italian studies of short-term
effects of air pollution (MISA), 1990-1999. Int J Occup Environ Health 11: 107-122.
Bind. MA; Baccarelli. A; Zanobetti. A; Tarantini. L; Suh. H; Vokonas. P: Schwartz. J. (2012). Air pollution
and markers of coagulation, inflammation, and endothelial function: Associations and epigene-
environment interactions in an elderly cohort. Epidemiology 23: 332-340.
http://dx.doi.org/10.1097/EDE.Ob013e31824523fO
Blomberg. A; Krishna. MT; Helledav. R; Soderberg. M; Ledin. MC: Kelly. FJ; Frew. AJ; Holgate. ST;
Sandstrom. T. (1999). Persistent airway inflammation but accomodated antioxidant and lung function
responses after repeated daily exposure to nitrogen dioxide. Am J Respir Crit Care Med 159: 536-543.
Boezen. M; Schouten. J: Rijcken. B; Vonk. J: Gerritsen. J: Van Per Zee. S: Hoek. G: Brunekreef. B;
Postma. D. (1998). Peak expiratory flow variability, bronchial responsiveness, and susceptibility to
ambient air pollution in adults. Am J Respir Crit Care Med 158: 1848-1854.
http://dx.doi.0rg/10.1164/airccm.158.6.9804072
November 2013 4-288 DRAFT: Do Not Cite or Quote
-------
Brannan. JD; Lougheed. MD. (2012). Airway hyperresponsiveness in asthma: Mechanisms, clinical
significance, and treatment. Front Physiol 3: 460. http://dx.doi.org/10.3389/fphys.2012.00460
Brook. JR: Burnett. RT; Dann. TF; Cakmak. S: Goldberg. MS: Fan. X; Wheeler. AJ. (2007). Further
interpretation of the acute effect of nitrogen dioxide observed in Canadian time-series studies. J Expo
Sci Environ Epidemiol 17: S36-S44. http://dx.doi.org/10.1038/sj.jes.7500626
Brusasco. V: Crimi. E; Gianiorio. P; Lantero. S: Rossi. GA. (1990). Allergen-induced increase in airway
responsiveness and inflammation in mild asthma. J Appl Physiol 69: 2209-2214.
Briiske. I; Hampel. R; Baumgartner. Z; Riickerl. R; Greven. S: Koenig. W: Peters. A; Schneider. A. (2011).
Ambient air pollution and lipoprotein-associated phospholipase A2 in survivors of myocardial
infarction. Environ Health Perspect 119: 921-926. http://dx.doi.org/10.1289/ehp.1002681
Burnett. RT; Smith-Doiron. M; Stieb. D; Cakmak. S: Brook. JR. (1999). Effects of particulate and gaseous
air pollution on cardiorespiratory hospitalizations. Arch Environ Health 54: 130-139.
http://dx.doi.org/10.1080/00039899909602248
Burnett. RT: Smith-Doiron. M; Stieb. D; Raizenne. ME; Brook. JR; Dales. RE; Leech. JA; Cakmak. S:
Krewski. D. (2001). Association between ozone and hospitalization for acute respiratory diseases in
children less than 2 years of age. Am J Epidemiol 153: 444-452. http://dx.doi.Org/10.1093/aie/153.5.444
Burnett. RT: Stieb. D: Brook. JR; Cakmak. S: Dales. R: Raizenne. M; Vincent. R: Dann. T. (2004).
Associations between short-term changes in nitrogen dioxide and mortality in Canadian cities. Arch
Environ Occup Health 59: 228-236. http://dx.doi.Org/10.3200/AEOH.59.5.228-236
Burra. TA; Moineddin. R; Agha. MM; Glazier. RH. (2009). Social disadvantage, air pollution, and asthma
physician visits in Toronto, Canada. Environ Res 109: 567-574.
http://dx.doi.0rg/10.1016/i.envres.2009.03.004
Bylin. G. (1993). Health risk evaluation of nitrogen oxide: Controlled studies on humans. Scand J Work
Environ Health 19: 37-43.
Bylin. G: Hedenstierna. G: Lindvall. T; Sundin. B. (1988). Ambient nitrogen dioxide concentrations
increase bronchial responsiveness in subjects with mild asthma. Eur Respir J 1: 606-612.
Bylin. G: Lindvall. T; Rehn. T; Sundin. B. (1985). Effects of short-term exposure to ambient nitrogen
dioxide concentrations on human bronchial reactivity and lung function. Eur Respir J 66: 205-217.
Cakmak. S: Dales. R: Leech. J: Liu. L. (2011 a). The influence of air pollution on cardiovascular and
pulmonary function and exercise capacity: Canadian Health Measures Survey (CHMS). Environ Res
111: 1309-1312. http://dx.doi.0rg/10.1016/i.envres.2011.09.016
Cakmak. S: Dales. RE; Angelica Rubio. M; Blanco Vidal. C. (20 lib). The risk of dying on days of higher
air pollution among the socially disadvantaged elderly. Environ Res 111: 388-393.
http://dx.doi.0rg/10.1016/i.envres.2011.01.003
Cakmak. S: Dales. RE; Judek. S. (2006). Respiratory health effects of air pollution gases: Modification by
education and income. Arch Environ Occup Health 61: 5-10. http://dx.doi.Org/10.3200/AEOH.61.l.5-10
Campen. MJ; Lund. AK; Doyle-Eisele. ML; McDonald. JD: Knuckles. TL; Rohr. AC: Knipping. EM;
Mauderly. JL. (2010). A comparison of vascular effects from complex and individual air pollutants
indicates a role for monoxide gases and volatile hydrocarbons. Environ Health Perspect 118: 921-927.
http://dx.doi.org/10.1289/ehp.0901207
Canova. C: Torresan. S: Simonato. L; Scapellato. ML; Tessari. R; Visentin. A; Lotti. M; Maestrelli. P.
(2010). Carbon monoxide pollution is associated with decreased lung function in asthmatic adults. Eur
Respir J 35: 266-272. http://dx.doi.org/10.1183/09031936.00043709
Carlsen. HK; Zoe'ga. H; Valdimarsdottir. U; Gislason. T; Hrafnkelsson. B. (2012). Hydrogen sulfide and
particle matter levels associated with increased dispensing of anti-asthma drugs in Iceland's capital.
Environ Res 113: 33-39. http://dx.doi.Org/10.1016/i.envres.2011.10.010
November 2013 4-289 DRAFT: Do Not Cite or Quote
-------
Carraro. S: Gottardi. G: Bonetto. G: Baraldi. E. (2007). Exhaled nitric oxide in children with asthma and
sinusitis [Review]. Pediatric Allergy and Immunology 18: 28-30. http://dx.doi.org/10. Ill 1/j. 1399-
3038.2007.00629.x
Casset. A; Purohit. A; Birba. E; Chenard. MP; Uring Lambert. B; Bahrain. S: Meyer. P; Pauli. G: De Slav.
E (2007). Bronchial challenge test in asthmatics sensitized to mites: Role of particle size in bronchial
response. J Aerosol Med 20: 509-518. http://dx.doi.org/10.1089/jam.2007.0591
Castro. HA; Cunha. MF; Mendonca. GA; Junger. WL; Cunha-Cruz. J; Leon. AP. (2009). Effect of air
pollution on lung function in schoolchildren in Rio de Janeiro, Brazil. Rev Saude Publica 43: 26-34.
Chan. CC: Chuang. KJ; Chen. WJ: Chang. WT; Lee. CT; Peng. CM. (2008). Increasing cardiopulmonary
emergency visits by long-range transported Asian dust storms in Taiwan. Environ Res 106: 393-400.
http://dx.doi.0rg/10.1016/i.envres.2007.09.006
Chang. CC: Tsai. SS: Ho. SC: Yang. CY. (2005). Air pollution and hospital admissions for cardiovascular
disease in Taipei, Taiwan. Environ Res 98: 114-119. http://dx.doi.Org/10.1016/j.envres.2004.07.005
Chang. YK; Wu. CC: Lee. IT; Lin. RS: Yu. YH; Chen. YC. (2012). The short-term effects of air pollution
on adolescent lung function in Taiwan. Chemosphere 87: 26-30.
http://dx.doi.0rg/10.1016/i.chemosphere.2011.ll.048
Channell. MM; Paffett. ML; Devlin. RB; Madden. MC: Campen. MJ. (2012). Circulating factors induce
coronary endothelial cell activation following exposure to inhaled diesel exhaust and nitrogen dioxide in
humans: Evidence from a novel translational in vitro model. Toxicol Sci 127: 179-186.
http://dx.doi.org/10.1093/toxsci/kfs084
Chen. BY: Chan. CC: Lee. CT: Cheng. TJ: Huang. WC: Jhou. JC: Han. YY: Chen. CC: Quo. YL. (2012a).
The association of ambient air pollution with airway inflammation in schoolchildren. Am J Epidemiol
175: 764-774. http://dx.doi.org/10.1093/aie/kwr380
Chen. R: Chu. C: Tan. J: Cao. J: Song. W: Xu. X: Jiang. C: Ma. W: Yang. C: Chen. B: Gui. Y: Kan. H.
(201 Ob). Ambient air pollution and hospital admission in Shanghai, China. J Hazard Mater 181: 234-
240. http://dx.doi.0rg/10.1016/j.jhazmat.2010.05.002
Chen. R; Samoli. E; Wong. CM; Huang. W: Wang. Z; Chen. B; Kan. H. (2012b). Associations between
short-term exposure to nitrogen dioxide and mortality in 17 Chinese cities: The China Air Pollution and
Health Effects Study (CAPES). Environ Int 45: 32-38. http://dx.doi.Org/10.1016/i.envint.2012.04.008
Chen. SY; Su. TC: Lin. YL; Chan. CC. (2012c). Short-term effects of air pollution on pulse pressure among
nonsmoking adults. Epidemiology 23: 341-348. http://dx.doi.org/10.1097/EDE.Ob013e3182452fld
Cheng. MF; Tsai. SS: Yang. CY. (2009a). Air pollution and hospital admissions for myocardial infarction in
a tropical city: Kaohsiung, Taiwan. J Toxicol Environ Health A 72: 1135-1140.
http://dx.doi.org/10.1080/15287390903091756
Chetta. A; Foresi. A; Del Donno. M; Consigli. GF; Bertorelli. G: Pesci. A; Barbee. RA; Olivieri. D. (1996).
Bronchial responsiveness to distilled water and methacholine and its relationship to inflammation and
remodeling of the airways in asthma. Am J Respir Crit Care Med 153: 910-917.
http://dx.doi.0rg/10.1164/airccm.153.3.8630572
Cheung. D; Dick. EC: Timmers. MC: de Klerk. EP; Spaan. WJ: Sterk. PJ. (1995). Rhinovirus inhalation
causes long-lasting excessive airway narrowing in response to methacholine in asthmatic subjects in
vivo. Am J Respir Crit Care Med 152: 1490-1496. http://dx.doi.Org/10.1164/airccm.152.5.7582282
Chimenti. L; Morici. G: Paterno. A; Bonanno. A; Vultaggio. M; Bellia. V: Bonsignore. MR. (2009).
Environmental conditions, air pollutants, and airway cells in runners: A longitudinal field study. J Sports
Sci 27: 925-935. http://dx.doi.org/10.1080/02640410902946493
Chiusolo. M; Cadum. E; Stafoggia. M; Galassi. C: Berti. G: Faustini. A; Bisanti. L; Vigotti. MA; Dessi.
MP; Cernigliaro. A; Mallone. S: Pacelli. B; Minerba. S: Simonato. L; Forastiere. F. (2011). Short term
effects of nitrogen dioxide on mortality and susceptibility factors in 10 Italian cities: The EpiAir Study.
Environ Health Perspect 119: 1233-1238. http://dx.doi.org/10.1289/ehp.1002904
November 2013 4-290 DRAFT: Do Not Cite or Quote
-------
Choi. JH; Xu. OS: Park. SY; Kim. JH; Hwang. SS: Lee. KH; Lee. HJ; Hong. YC. (2007). Seasonal
variation of effect of air pollution on blood pressure. JEpidemiol Community Health 61: 314-318.
http://dx.doi.org/10.1136/iech.2006.049205
Chuang. KJ; Chan. CC: Su. TC: Lee. CT; Tang. CS. (2007a). The effect of urban air pollution on
inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am J Respir
Crit Care Med 176: 370-376. http://dx.doi.org/10.1164/rccm.200611-1627OC
Chuang. KJ: Chan. CC: Su. TC: Lin. LY; Lee. CT. (2007b). Associations between paniculate sulfate and
organic carbon exposures and heart rate variability in patients with or at risk for cardiovascular diseases.
J Occup Environ Med 49: 610-617. http://dx.doi.org/10.1097/JOM.Ob013e318058205b
Chuang. KJ: Coull. BA; Zanobetti. A; Suh. H; Schwartz. J; Stone. PH; Litonjua. A; Speizer. FE; Gold. DR.
(2008). Paniculate Air Pollution as a Risk Factor for ST-Segment Depression in Patients With Coronary
Artery Disease. Circulation 118: 1314-1320.
http://dx.doi.org/10.1161/CIRCULATIONAHA.108.765669
Chuang. KJ: Yan. YH; Cheng. TJ. (2010). Effect of air pollution on blood pressure, blood lipids, and blood
sugar: A population-based approach. J Occup Environ Med 52: 258-262.
http://dx.doi.org/10.1097/JOM.Ob013e3181ceff7a
Cockcroft. DW. (2010). Direct challenge tests: Airway hyperresponsiveness in asthma: Its measurement
and clinical significance [Review]. Chest 138: 18S-24S. http://dx.doi.org/10.1378/chest.10-0088
Cockcroft. DW: Davis. BE. (2006a). Airway hyperresponsiveness as a determinant of the early asthmatic
response to inhaled allergen. J Asthma 43: 175-178. http://dx.doi.org/10.1080/02770900600566348
Cockcroft. DW: Davis. BE. (2006b). The bronchoprotective effect of inhaling methacholine by using total
lung capacity inspirations has a marked influence on the interpretation of the test result [Review]. J
Allergy Clin Immunol 117: 1244-1248. http://dx.doi.Org/10.1016/i.iaci.2006.02.038
Cockcroft DW: Davis. BE. (2006c). Mechanisms of airway hyperresponsiveness [Review]. J Allergy Clin
Immunol 118: 551-559. http://dx.doi.Org/10.1016/i.jaci.2006.07.012
Cockcroft. DW: Davis. BE; Todd. DC: Smycniuk. AJ. (2005). Methacholine challenge: Comparison of two
methods. Chest 127: 839-844. http://dx.doi.Org/10.1378/chest.127.3.839
Dales. R; Chen. L; Frescura. AM; Liu. L; Villeneuve. PJ. (2009a). Acute effects of outdoor air pollution on
forced expiratory volume in 1 s: A panel study of schoolchildren with asthma. Eur Respir J 34: 316-323.
http://dx.doi.org/10.1183/09031936.00138908
Dales. RE; Cakmak. S: Doiron. MS. (2006). Gaseous air pollutants and hospitalization for respiratory
disease in the neonatal period. Environ Health Perspect 114: 1751-1754.
http://dx.doi.org/10.1289/ehp.9044
Dales. RE; Cakmak. S: Vidal. CB. (2010). Air Pollution and hospitalization for venous thromboembolic
disease in Chile. J Thromb Haemost 8: 669-674. http://dx.doi.Org/10.llll/i.1538-7836.2010.03760.x
Darrow. LA; Klein. M; Sarnat JA; Mulholland. JA; Strickland. MJ; Sarnat SE; Russell. AG: Tolbert. PE.
(2011 a). The use of alternative pollutant metrics in time-series studies of ambient air pollution and
respiratory emergency department visits. J Expo Sci Environ Epidemiol 21: 10-19.
http://dx.doi.org/10.1038/ies.2009.49
Davis. JK; Davidson. M; Schoeb. TR. (1991). Murine respiratory mycoplasmosis: a model to study effects
of oxidants (pp. 1-29; discussion 31-43). Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=65
Dawson. SV: Schenker. MB. (1979). Health effects of inhalation of ambient concentrations of nitrogen
dioxide [Editorial]. Am Rev Respir Dis 120: 281-292.
de Burbure. CY; Heilier. JF; Neve. J: Becker. A; Albrecht. C: Borm. PJA; Gromadzinska. J: Wasowicz. W:
Rydzynski. K; Bernard. AM. (2007). Lung permeability, antioxidant status, and NO2 inhalation: A
selenium supplementation study in rats. J Toxicol Environ Health A 70: 284-294.
http://dx.doi.org/10.1080/15287390600884875
November 2013 4-291 DRAFT: Do Not Cite or Quote
-------
Delfino. RJ: Gillen. PL; Tjoa. T; Staimer. N; Polidori. A; Arhami. M; Sioutas. C: Longhurst. J. (2011).
Electrocardiographic ST-segment depression and exposure to trafficrelated aerosols in elderly subjects
with coronary artery disease. Environ Health Perspect 119: 196-202.
http://dx.doi.org/10.1289/ehp.1002372
Delfino. RJ: Staimer. N; Gillen. D; Tjoa. T; Sioutas. C: Fung. K; George. SC: Kleinman. MT. (2006).
Personal and ambient air pollution is associated with increased exhaled nitric oxide in children with
asthma. Environ Health Perspect 114: 1736-1743.
Delfino. RJ: Staimer. N; Tjoa. T; Arhami. M; Polidori. A; Gillen. PL; George. SC: Shafer. MM; Schauer.
JJ; Sioutas. C. (2010). Associations of primary and secondary organic aerosols with airway and
systemic inflammation in an elderly panel cohort. Epidemiology 21: 892-902.
http://dx.doi.org/10.1097/EDE.Ob013e3181f20e6c
Delfino. RJ: Staimer. N; Tjoa. T; Gillen. D; Kleinman. MT; Sioutas. C: Cooper. D. (2008a). Personal and
ambient air pollution exposures and lung function decrements in children with asthma. Environ Health
Perspect 116: 550-558. http://dx.doi.org/10.1289/ehp.10911
Delfino. RJ: Staimer. N; Tjoa. T; Polidori. A; Arhami. M; Gillen. PL; Kleinman. MT: Vaziri. ND;
Longhurst. J: Zaldivar. F: Sioutas. C. (2008b). Circulating biomarkers of inflammation, antioxidant
activity, and platelet activation are associated with primary combustion aerosols in subjects with
coronary artery disease. Environ Health Perspect 116: 898-906. http://dx.doi.org/10.1289/ehp.11189
Delfino RJ: Staimer. N; Tjoa. T; Gillen. PL; Polidori. A; Arhami. M; Kleinman. MT: Vaziri. ND;
Longhurst. J: Sioutas. C. (2009). Air pollution exposures and circulating biomarkers of effect in a
susceptible population: Clues to potential causal component mixtures and mechanisms. Environ Health
Perspect 117: 1232-1238.
Dennekamp. M; Akram. M; Abramson. MJ; Tonkin. A; Sim. MR; Fridman. M; Erbas. B. (2010). Outdoor
air pollution as a trigger for out-of-hospital cardiac arrests. Epidemiology 21: 494-500.
http://dx.doi.org/10.1097/EDE.Ob013e3181e093db
Desqueyroux. H; Pujet. JC: Prosper. M; Le Moullec. Y; Momas. I. (2002). Effects of air pollution on adults
with chronic obstructive pulmonary disease. Arch Environ Occup Health 57: 554-560.
http://dx.doi.org/10.1080/00039890209602088
Devlin. RB: Horstman. DP; Gerritv. TR: Becker. S: Madden. MC. (1999). Inflammatory response in
humans exposed to 2.0 PPM nitrogen dioxide. Inhal Toxicol 11: 89-109.
http://dx.doi.org/10.1080/089583799197195
Dominici. F: McDermott. A; Daniels. M; Zeger. SL; Samet. JM. (2003). Mortality among residents of 90
cities [HEI]. In Revised analyses of time-series studies of air pollution and health (pp. 9-24). Boston,
MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=4
Powell. AR; Kilburn. KH; Pratt. PC. (1971). Short-term exposure to nitrogen dioxide: Effects on
pulmonary ultrastructure, compliance, and the surfactant system. Arch Intern Med 128: 74-80.
http://dx.doi.org/10.1001/archinte.1971.00310190078009
Downie. SR; Salome. CM; Verbanck. S: Thompson. B; Berend. N; King. GG. (2007). Ventilation
heterogeneity is a major determinant of airway hyperresponsiveness in asthma, independent of airway
inflammation. Thorax 62: 684-689. http://dx.doi.org/10.1136/thx.2006.069682
Drechsler-Parks. DM. (1995). Cardiac output effects of O3 and NO2 exposure in healthy older adults.
Toxicol Ind Health 11: 99-109. http://dx.doi.org/10.1177/074823379501100109
Drvden. DM; Spooner. CH; Stickland. MK; Vandermeer. B; Tjosvold. L; Bialy. L; Wong. K; Rowe. BH.
(2010). Exercise-induced bronchoconstriction and asthma. (AHRQ Publication No. 10-E001).
Rockville, MD: Agency for Healthcare Research and Quality.
http ://www. ahrq. gov/research/Findings/evidence-based-reports/eibeia-evidence-report.pdf
November 2013 4-292 DRAFT: Do Not Cite or Quote
-------
Eenhuizen. E; Gehring. U; Wijga. AH; Smit. HA; Fischer. PH; Brauer. M; Koppelman. GH; Kerkhof. M; de
Jongste. JC: Brunekreef. B; Hoek. G. (2013). Traffic related air pollution is related to interrupter
resistance in four-year old children. Eur Respir J 41: 1257-1263.
http://dx.doi.org/10.1183/09031936.00020812
Ehrlich. R. (1980). Interaction between environmental pollutants and respiratory infections. Environ Health
Perspect35: 89-99.
Ehrlich. R; Findlav. JC: Fenters. JD; Gardner. DE. (1977). Health effects of short-term inhalation of
nitrogen dioxide and ozone mixtures. Environ Res 14: 223-231. http://dx.doi.org/10.1016/0013-
9351(77)90034-2
Ehrlich. R; Findlav. JC: Gardner. DE. (1979). Effects of repeated exposures to peak concentrations of
nitrogen dioxide and ozone on resistance to streptococcal pneumonia. J Toxicol Environ Health 5: 631-
642. http://dx.doi.org/10.1080/15287397909529775
Elsaved. NM; Mustafa. MG. (1982). Dietary antioxidants and the biochemical response to oxidant
inhalation: I: Influence of dietary vitamin E on the biochemical effects of nitrogen dioxide exposure in
rat lung. Toxicol Appl Pharmacol 66: 319-328. http://dx.doi.org/10.1016/0041-008X(82)90298-8
Erjefalt JS: Greiff. L; Andersson. M; Matsson. E; Petersen. H; Linden. M; Ansari. T; Jefferv. PK; Persson.
CG. (1999). Allergen-induced eosinophil cytolysis is a primary mechanism for granule protein release in
human upper airways. Am J Respir Crit Care Med 160: 304-312.
http://dx.doi.0rg/10.1164/ajrccm.160.l.9809048
Escamilla-Nunez. MC: Barraza-Villarreal. A; Hernandez-Cadena. L; Moreno-Macias. H; Ramirez-Aguilar.
M; Sienra-Monge. JJ: Cortez-Lugo. M; Texcalac. JL; del Rio-Navarro. B; Romieu. I. (2008). Traffic-
related air pollution and respiratory symptoms among asthmatic children, resident in Mexico City: The
EVA cohort study. Respir Res 9: 74. http://dx.doi.org/10.1186/1465-9921-9-74
Faustini. A; Stafoggia. M; Colais. P; Berti. G: Bisanti. L; Cadum. E; Cernigliaro. A; Mallone. S: Scarnato.
C: Forastiere. F. (2013). Air pollution and multiple acute respiratory outcomes. Eur Respir J 42: 304-
313. http://dx.doi.org/10.1183/09031936.00128712
Feo Brito. F; Mur Gimeno. P; Martinez. C: Tobias. A; Suarez. L; Guerra. F; Borja. JM; Alonso. AM.
(2007). Air pollution and seasonal asthma during the pollen season: A cohort study in Puertollano and
Ciudad Real (Spain). Allergy 62: 1152-1157. http://dx.doi.Org/10.llll/i.1398-9995.2007.01438.x
Ferin. J: Leach. LJ. (1975). The effects of selected air pollutants on clearance of titanic oxide particles from
the lungs of rats. Inhaled Particles 4 Pt 1: 333-341.
Filho. MAP: Pereira. LAA; Arbex. FF; Arbex. M; Conceicao. GM; Santos. UP; Lopes. AC: Saldiva. PHN;
Braga. ALF; Cendon. S. (2008). Effect of air pollution on diabetes and cardiovascular diseases in Sao
Paulo, Brazil. Braz J Med Biol Res 41: 526-532. http://dx.doi.org/10.1590/S0100-879X2008005000020
Flamant-Hulin. M; Caillaud. D; Sacco. P: Penard-Morand. C: Annesi-Maesano. I. (2010). Air pollution and
increased levels of fractional exhaled nitric oxide in children with no history of airway damage. J
Toxicol Environ Health A 73: 272-283. http://dx.doi.org/10.1080/15287390903249206
Folinsbee. LJ. (1992). Does nitrogen dioxide exposure increase airways responsiveness? Toxicol Ind Health
8: 273-283.
Folinsbee. LJ: Horvath. SM; Bedi. JF; Delehunt JC. (1978). Effect of 0.62 ppm NO2 on cardiopulmonary
function in young male nonsmokers. Environ Res 15: 199-205. http://dx.doi.org/10.1016/0013-
9351(78)90096-8
Forsberg. B; Stjernberg. N; Linne. R: Segerstedt. B; Wall. S. (1998). Daily air pollution levels and acute
asthma in southern Sweden. Eur Respir J 12: 900-905.
Fraenkel. DJ: Bardin. PG: Sanderson. G: Lampe. F; Johnston. SL; Holgate. ST. (1995). Lower airways
inflammation during rhinovirus colds in normal and in asthmatic subjects. Am J Respir Crit Care Med
151: 879-886. http://dx.doi.Org/10.1164/airccm/151.3 Pt 1.879
November 2013 4-293 DRAFT: Do Not Cite or Quote
-------
Frampton. MW: Boscia. J: Roberts. NJ. Jr; Azadniv. M; Torres. A; Cox. C: Morrow. PE; Nichols. J:
Chalupa. D; Frasier. LM; Gibb. FR; Speers. DM; Tsai. Y; Utell. MJ. (2002). Nitrogen dioxide exposure:
Effects on airway and blood cells. Am J Physiol 282: L155-L165.
Frampton. MW: Morrow. PE; Cox. C: Gibb. FR; Speers. DM; Utell. MJ. (1991). Effects of nitrogen
dioxide exposure on pulmonary function and airway reactivity in normal humans. Am Rev Respir Dis
143: 522-527. http://dx.doi.Org/10.1164/airccm/143.3.522
Frampton. MW: Smeglin. AM; Roberts. NJ. Jr; Finkelstein. JN; Morrow. PE; Utell. MJ. (1989). Nitrogen
dioxide exposure in vivo and human alveolar macrophage inactivation of influenza virus in vitro.
Environ Res 48: 179-192. http://dx.doi.org/10.1016/S0013-9351(89)80033-7
Freedman. S: Lane. R; Gillett. MK; Guz. A. (1988). Abolition of methacholine induced bronchoconstriction
by the hyperventilation of exercise or volition. Thorax 43: 631-636.
Fung. KY; Khan. S: Krewski. D; Chen. Y. (2006). Association between air pollution and multiple
respiratory hospitalizations among the elderly in Vancouver, Canada. Inhal Toxicol 18: 1005-1011.
http://dx.doi.org/10.1080/08958370600904538
Fung. KY; Luginaah. I; Gorev. KM; Webster. G. (2005). Air pollution and daily hospital admissions for
cardiovascular diseases in Windsor, Ontario. Can J Public Health 96: 29-33.
Gardner. DE; Miller. FJ; Blommer. EJ; Coffin. PL. (1979). Influence of exposure mode on the toxicity of
NO2. Environ Health Perspect 30: 23-29.
Gent. JF; Triche. EW: Holford. TR; Belanger. K; Bracken. MB; Beckett. WS: Leaderer. BP (2003).
Association of low-level ozone and fine particles with respiratory symptoms in children with asthma.
JAMA 290: 1859-1867. http://dx.doi.org/10.1001/iama.290.14.1859
Gillespie-Bennett. J: Pierse. N; Wickens. K; Crane. J: Howden-Chapman. P; Team. HHaHSR. (2011). The
respiratory health effects of nitrogen dioxide in children with asthma. Eur Respir J 38: 303-309.
http://dx.doi.org/10.1183/09031936.00115409
Gilmour. MI; Park. P; Selgrade. MK. (1996). Increased immune and inflammatory responses to dust mite
antigen in rats exposed to 5 ppm NO2. Fundam Appl Toxicol 31: 65-70.
Goings. SAJ; Kulle. TJ; Bascom. R; Sauder. LR; Green. DJ; Hebel. JR: Clements. ML. (1989). Effect of
nitrogen dioxide exposure on susceptibility to influenza A virus infection in healthy adults. Am Rev
Respir Dis 139: 1075-1081. http://dx.doi.Org/10.1164/ajrccm/139.5.1075
Goldberg. MS: Giannetti. N; Burnett. RT; Mayo. NE; Valois. MF; Brophy. JM. (2008). A panel study in
congestive heart failure to estimate the short-term effects from personal factors and environmental
conditions on oxygen saturation and pulse rate. Occup Environ Med 65: 659-666.
http://dx.doi.org/10.1136/oem.2007.034934
Goldstein. BD; Hamburger. SJ: Falk. GW: Amoruso. MA. (1977). Effect of ozone and nitrogen dioxide on
the agglutination of rat alveolar macrophages by concanavalin A. Life Sci 21: 1637-1644.
Goldstein. E; Eagle. MC: Hoeprich. PP. (1973). Effect of nitrogen dioxide on pulmonary bacterial defense
mechanisms. Arch Environ Health 26: 202-204. http://dx.doi.org/10.1080/00039896.1973.10666256
Goldstein. E; Warshauer. D; Lippert. W: Tarkington. B. (1974). Ozone and nitrogen dioxide exposure:
Murine pulmonary defense mechanisms. Arch Environ Health 28: 85-90.
http://dx.doi.org/10.1080/00039896.1974.10666444
Gong. H; Linn. WS: Clark. KW: Anderson. KR; Geller. MD; Sioutas. C. (2005). Respiratory responses to
exposures with fine particulates and nitrogen dioxide in the elderly with and without COPD. Inhal
Toxicol 17: 123-132. http://dx.doi.org/10.1080/08958370590904481
Goodman. JE; Chandalia. JK; Thakali. S: Seelev. M. (2009). Meta-analysis of nitrogen dioxide exposure
and airway hyper-responsiveness in asthmatics. Crit Rev Toxicol 39: 719-742.
http://dx.doi.org/10.3109/10408440903283641
November 2013 4-294 DRAFT: Do Not Cite or Quote
-------
Graham. JA; Gardner. DE; Blommer. EJ; House. DE; Menache. MG: Miller. FJ. (1987). Influence of
exposure patterns of nitrogen dioxide and modifications by ozone on susceptibility to bacterial
infectious disease in mice. J Toxicol Environ Health 21: 113-125.
http://dx.doi.org/10.1080/15287398709531006
Greenwald. R; Sarnat. SB; Ravsoni. AU; Li. WW: Johnson. BA; Stock. TH; Holguin. F; Sosa. T; Sarnat.
JA. (2013). Associations between source-indicative pollution metrics and increases in pulmonary
inflammation and reduced lung function in a panel of asthmatic children. Air Qua! Atmos Health 6: 487-
499. http://dx.doi.org/10.1007/sll869-012-0186-3
Gregory. RE; Pickrell. JA: Halm. FF; Hobbs. CH. (1983). Pulmonary effects of intermittent subacute
exposure to low-level nitrogen dioxide. J Toxicol Environ Health 11: 405-414.
http://dx.doi.org/10.1080/15287398309530354
Grineski. SE; Staniswalis. JG: Peng. Y; Atkinson-Palombo. C. (2010). Children's asthma hospitalizations
and relative risk due to nitrogen dioxide (NO2): Effect modification by race, ethnicity, and insurance
status. Environ Res 110: 178-188. http://dx.doi.Org/10.1016/i.envres.2009.10.012
Guo. Y; Jia. Y; Pan. X; Liu. L; Wichmann. HE. (2009). The association between fine paniculate air
pollution and hospital emergency room visits for cardiovascular diseases in Beijing, China. Sci Total
Environ 407: 4826-4830. http://dx.doi.Org/10.1016/i.scitotenv.2009.05.022
Guo. Y; Tong. S: Li. S: Barnett AG: Yu. W: Zhang. Y; Pan. X. (2010). Gaseous air pollution and
emergency hospital visits for hypertension in Beijing, China: a time-stratified case-crossover study.
Environ Health Global Access Sci Source 9: 57. http://dx.doi.org/10.1186/1476-069X-9-57
Guth. DJ; Mavis. RD. (1986). The effect of lung alpha-tocopherol content on the acute toxicity of nitrogen
dioxide. Toxicol Appl Pharmacol 84: 304-314. http://dx.doi.org/10.1016/0041-008X(86)90138-9
Hackney. JD; Thiede. FC: Linn. WS: Pedersen. EE; Spier. CE; Law. DC: Fischer. DA. (1978).
Experimental studies on human health effects of air pollutants. IV. Short-term physiological and clinical
effects of nitrogen dioxide exposure. Arch Environ Health 33: 176-181.
http://dx.doi.org/10.1080/00039896.1978.10667330
Hampel. R; Lepeule. J: Schneider. A; Bottagisi. S: Charles. MA; Ducimetiere. P; Peters. A; Slama. R.
(2011). Short-term impact of ambient air pollution and air temperature on blood pressure among
pregnant women. Epidemiology 22: 671-679. http://dx.doi.org/10.1097/EDE.Ob013e318226e8d6
Hansel. NN; Brevsse. PN; McCormack. MC: Matsui. EC: Curtin-Brosnan. J: Williams. DAL; Moore. JL;
Cuhran. JL; Diette. GB. (2008). A longitudinal study of indoor nitrogen dioxide levels and respiratory
symptoms in inner-city children with asthma. Environ Health Perspect 116: 1428-1432.
http://dx.doi.org/10.1289/ehp.11349
Hansel. NN; McCormack. MC: Belli. AJ; Matsui. EC: Peng. RD; Aloe. C: Paulin. L; Williams. PL; Diette.
GB; Brevsse. PN. (2013). In-home air pollution is linked to respiratory morbidity in former smokers
with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 187: 1085-1090.
http://dx.doi.org/10.1164/rccm.201211-1987OC
Harre. ESM; Price. PD; Ayrev. RB; Toop. LJ; Martin. IR; Town. GI. (1997). Respiratory effects of air
pollution in chronic obstructive pulmonary disease: a three month prospective study. Thorax 52: 1040-
1044.
Hatch. GE; Slade. R: Selgrade. MK; Stead. AG. (1986). Nitrogen dioxide exposure and lung antioxidants in
ascorbic acid-deficient guinea pigs. Toxicol Appl Pharmacol 82: 351-359.
http://dx.doi.org/10.1016/0041-008X(86)90212-7
Hazucha. MJ; Folinsbee. LJ: Seal. E; Bromberg. PA. (1994). Lung function response of healthy women
after sequential exposures to NO2 and O3. Am J Respir Crit Care Med 150: 642-647.
http://dx.doi.0rg/10.1164/airccm.150.3.8087332
Hazucha. MJ: Ginsberg. JF; McDonnell. WF; Haak. ED. Jr; Pimmel. RL; Salaam. SA; House. DE;
Bromberg. PA. (1983). Effects of 0.1 ppm nitrogen dioxide on airways of normal and asthmatic
subjects. J Appl Physiol 54: 730-739.
November 2013 4-295 DRAFT: Do Not Cite or Quote
-------
HEI (Health Effects Institute). (2003). Revised analyses of time-series studies of air pollution and health
[HEI]. (HEI Special Report). Boston, MA. http://pubs.healtheffects.org/view.php?id=4
HEI Collaborative Working Group (HEI Collaborative Working Group on Air Pollution, Poverty, and
Health in Ho Chi Minh City). (2012). Effects of short-term exposure to air pollution on hospital
admissions of young children for acute lower respiratory infections in Ho Chi Minh City, Vietnam
[HEI] (pp. 5-72; discussion 73-83). (ISSN 1041-5505 Research Report 169). Boston, MA: Health
Effects Institute. http://pubs.healtheffects.org/view.php?id=380
Helledav. R; Huberman. D; Blomberg. A; Stjernberg. N; Sandstrom. T. (1995). Nitrogen dioxide exposure
impairs the frequency of the mucociliary activity in healthy subjects. Eur Respir J 8: 1664-1668.
http://dx.doi.org/10.1183/09031936.95.08101664
Helledav. R; Sandstrom. T; Stjernberg. N. (1994). Differences in bronchoalveolar cell response to nitrogen
dioxide exposure between smokers and nonsmokers. Eur Respir J 7: 1213-1220.
Henneberger. A; Zareba. W: Ibald-Mulli. A; Riickerl. R; Cyrys. J: Couderc. JP; Mykins. B; Woelke. G:
Wichmann. HE; Peters. A. (2005). Repolarization changes induced by air pollution in ischemic heart
disease patients. Environ Health Perspect 113: 440-446. http://dx.doi.org/10.1289/ehp.7579
Hernandez-Cadena. L; Holguin. F; Barraza-Villarreal. A; Del Rio-Navarro. BE; Sienra-Monge. JJ: Romieu.
L (2009). Increased levels of outdoor air pollutants are associated with reduced bronchodilation in
children with asthma. Chest 136: 1529-1536. http://dx.doi.org/10.1378/chest.08-1463
Higgins. BG: Francis. HC: Yates. C: Warburton. CJ: Fletcher. AM; Pickering. CAC: Woodcock. AA.
(2000). Environmental exposure to air pollution and allergens and peak flow changes. Eur Respir J 16:
66.
Higgins. BG: Francis. HC: Yates. CJ: Warburton. CJ: Fletcher. AM; Reid. JA; Pickering. CAC: Woodcock.
AA. (1995). Effects of air pollution on symptoms and peak expiratory flow measurements in subjects
with obstructive airways disease. Thorax 50: 149-155. http://dx.doi.Org/10.1136/thx.50.2.149
Hildebrandt. K; Riickerl. R; Koenig. W: Schneider. A; Pitz. M; Heinrich. J: Marder. V; Frampton. M;
Oberdorster. G: Wichmann. HE; Peters. A. (2009). Short-term effects of air pollution: A panel study of
blood markers in patients with chronic pulmonary disease. Part Fibre Toxicol 6: 25.
http://dx.doi.org/10.1186/1743-8977-6-25
Hiltermann. TJN; Stolk. J: van der Zee. SC: Brunekreef. B; de Bruijne. CR; Fischer. PH; Ameling. CB;
Sterk. PJ; Hiemstra. PS: van Bree. L. (1998). Asthma severity and susceptibility to air pollution. Eur
Respir J 11: 686-693.
Hinwood. AL; De Klerk. N; Rodriguez. C: Jacobv. P; Runnion. T; Rye. P; Landau. L; Murray. F; Feldwick.
M; Spickett. J. (2006). The relationship between changes in daily air pollution and hospitalizations in
Perth, Australia 1992-1998: A case-crossover study. Int J Environ Health Res 16: 27-46.
http://dx.doi.org/10.1080/09603120500397680
Hoek. G. (2003). Daily mortality and air pollution in The Netherlands [HEI]. In Revised analyses of time-
series studies of air pollution and health (pp. 133-141). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/getfile.php?u=21
Holguin. F; Flores. S: Ross. Z; Cortez. M; Molina. M; Molina. L; Rincon. C: Jerrett. M; Berhane. K;
Granados. A; Romieu. I. (2007). Traffic-related exposures, airway function, inflammation, and
respiratory symptoms in children. Am J Respir Crit Care Med 176: 1236-1242.
http://dx.doi.org/10.1164/rccm.200611-1616OC
Hooftman. RN; Kuper. CF; Appelman. LM. (1988). Comparative sensitivity of histo-pathology and specific
lung parameters in the detection of lung injury. J Appl Toxicol 8: 59-65.
http://dx.doi.org/10.1002/iat.2550080110
Hovert. PL; Xu. J. (2012). Deaths: Preliminary data for 2011. US Department of Health and Human
Services National Center for Health Statistics National Vital Statistics Reports 61: 1-51.
November 2013 4-296 DRAFT: Do Not Cite or Quote
-------
Hsieh. YL; Yang. YH; Wu. TN; Yang. CY. (2010). Air pollution and hospital admissions for myocardial
infarction in a subtropical city: Taipei, Taiwan. J Toxicol Environ Health A 73: 757-765.
http://dx.doi.org/10.1080/15287391003684789
Huang. W: Zhu. T; Pan. X; Hu. M; Lu. SE; Lin. Y; Wang. T; Zhang. Y; Tang. X. (2012a). Air pollution and
autonomic and vascular dysfunction in patients with cardiovascular disease: Interactions of systemic
inflammation, overweight, and gender. Am J Epidemiol 176: 117-126.
http://dx.doi.org/10.1093/aie/kwr511
Huang. YC: Rappold. AG: Graff. DW: Ohio. AJ; Devlin. RB. (2012b). Synergistic effects of exposure to
concentrated ambient fine pollution particles and nitrogen dioxide in humans. Inhal Toxicol 24: 790-
797. http://dx.doi.org/10.3109/08958378.2012.718809
Hubbard. AK; Symanowicz. PT; Thibodeau. M; Thrall. RS: Schramm. CM; Cloutier. MM; Morris. JB.
(2002). Effect of nitrogen dioxide on ovalbumin-induced allergic airway disease in a murine model. J
Toxicol Environ Health A 65: 1999-2005. http://dx.doi.org/10.1080/00984100290071397
Ichinose. T; Arakawa. K; Shimojo. N; Sagai. M. (1988). Biochemical effects of combined gases of nitrogen
dioxide and ozone: II. Species differences in lipid peroxides and antioxidative protective enzymes in the
lungs. Toxicol Lett 42: 167-176.
Illing. JW: Miller. FJ; Gardner. DE. (1980). Decreased resistance to infection in exercised mice exposed to
NO2 and O3. J Toxicol Environ Health 6: 843-851. http://dx.doi.org/10.1080/15287398009529902
Inman. MD; Watson. RM; Killian. KJ; O'Byrne. PM. (1990). Methacholine airway responsiveness
decreases during exercise in asthmatic subjects. Am Rev Respir Dis 141: 1414-1417.
Ito. K; Mathes. R; Ross. Z; Nadas. A; Thurston. G: Matte. T. (2011). Fine paniculate matter constituents
associated with cardiovascular hospitalizations and mortality in New York City. Environ Health
Perspect 119: 467-473. http://dx.doi.org/10.1289/ehp.1002667
Ito. K; Thurston. GD; Silverman. RA. (2007). Association between coarse particles and asthma emergency
department (ED) visits in New York City. American Thoracic Society International Conference 2007,
San Francisco, CA.
Jaffe. DH; Singer. ME; Rimm. AA. (2003). Air pollution and emergency department visits for asthma
among Ohio Medicaid recipients, 1991-1996. Environ Res 91: 21-28.
Jakab. GJ. (1988). Modulation of pulmonary defense mechanisms against viral and bacterial infections by
acute exposures to nitrogen dioxide (pp. 1-38). (HEI Research Report no. 20). Cambridge, MA: Health
Effects Institute. http://pubs.healtheffects.org/view.php?id=238
Jalaludin. B; Khalaj. B; Sheppeard. V: Morgan. G. (2008). Air pollution and ED visits for asthma in
Australian children: A case-crossover analysis. Int Arch Occup Environ Health 81: 967-974.
http://dx.doi.org/10.1007/s00420-007-0290-0
Jalaludin. B; Morgan. G: Lincoln. D; Sheppeard. V: Simpson. R; Corbett. S. (2006). Associations between
ambient air pollution and daily emergency department attendances for cardiovascular disease in the
elderly (65+years), Sydney, Australia. J Expo Sci Environ Epidemiol 16: 225-237.
http://dx.doi.org/10.1038/sj.jea.7500451
Jalaludin. BB; O'Toole. BI; Leeder. SR. (2004). Acute effects of urban ambient air pollution on respiratory
symptoms, asthma medication use, and doctor visits for asthma in a cohort of Australian children.
Environ Res 95: 32-42. http://dx.doi.org/10.1016/80013-9351(03)00038-0
Jenkins. HS: Devalia. JL; Mister. RL; Sevan. AM; Rusznak. C: Davies. RJ. (1999). The effect of exposure
to ozone and nitrogen dioxide on the airway response of atopic asthmatics to inhaled allergen: dose- and
time-dependent effects. Am J Respir Crit Care Med 160: 33-39.
http://dx.doi.0rg/10.1164/ajrccm.160.l.9808119
Jia. X; Song. X; Shima. M; Tamura. K; Deng. F; Guo. X. (2011). Acute effect of ambient ozone on heart
rate variability in healthy elderly subjects. J Expo Sci Environ Epidemiol 21: 541-547.
http://dx.doi.org/10.1038/ies.2011.18
November 2013 4-297 DRAFT: Do Not Cite or Quote
-------
Jones. SL; Kittelson. J; Cowan. JO; Flannery. EM; Hancox. RJ; McLachlan. CR; Taylor. DR. (2001). The
predictive value of exhaled nitric oxide measurements in assessing changes in asthma control. Am J
Respir Crit Care Med 164: 738-743. http://dx.doi.Org/10.1164/ajrccm.164.5.2012125
Jorres. R; Magnussen. H. (1990). Airways response of asthmatics after a 30 min exposure, at resting
ventilation, to 0.25 ppm NO2 or 0.5 ppm SO2. Eur Respir J 3: 132-137.
Jorres. R; Magnussen. H. (1991). Effect of 0.25 ppm nitrogen dioxide on the airway response to
methacholine in asymptomatic asthmatic patients. Lung 169: 77-85.
Jorres. R: Nowak. D; Grimminger. F; Seeger. W: Oldigs. M; Magnussen. H. (1995). The effect of 1 ppm
nitrogen dioxide on bronchoalveolar lavage cells and inflammatory mediators in normal and asthmatic
subjects. Eur Respir J 8: 416-424. http://dx.doi.org/10.1183/09031936.95.08030416
Just. J: Segala. C: Sahraoui. F; Priol. G: Grimfeld. A; Neukirch. F. (2002). Short-term health effects of
paniculate and photochemical air pollution in asthmatic children. Eur Respir J 20: 899-906.
http://dx.doi.org/10.1183/09031936.02.00236902
Kan. H; Chen. B; Zhao. N; London. SJ: Song. G: Chen. G: Zhang. Y; Jiang. L. (2010). Part 1. A time-series
study of ambient air pollution and daily mortality in Shanghai, China. In Public Health and Air Pollution
in Asia (PAPA): Coordinated Studies of Short-Term Exposure to Air Pollution and Daily Mortality in
Four Cities (pp. 17-78). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=348
Kan. H; London. SJ: Chen. G: Zhang. Y; Song. G: Zhao. N; Jiang. L; Chen. B. (2008). Season, sex, age,
and education as modifiers of the effects of outdoor air pollution on daily mortality in Shanghai, China:
The Public Health and Air Pollution in Asia (PAPA) Study. Environ Health Perspect 116: 1183-1188.
http://dx.doi.org/10.1289/ehp.10851
Karakatsani. A; Analitis. A; Perifanou. D; Ayres. JG: Harrison. RM; Kotronarou. A; Kavouras. IG:
Pekkanen. J: Hameri. K; Kos. GP; de Hartog. JJ: Hoek. G: Katsouvanni. K. (2012). Paniculate matter
air pollution and respiratory symptoms in individuals having either asthma or chronic obstructive
pulmonary disease: A European multicentre panel study. Environ Health Global Access Sci Source 11:
75. http://dx.doi.org/10.1186/1476-069X-ll-75
Katsouvanni. K; Samet JM; Anderson. HR; Atkinson. R; Le Tertre. A; Medina. S: Samoli. E; Touloumi. G:
Burnett. RT; Krewski. D; Ramsay. T; Dominici. F; Peng. RD; Schwartz. J: Zanobetti. A. (2009). Air
pollution and health: A European and North American approach (APHENA) (pp. 5-90). (ISSN 1041-
5505 Research Report 142). Boston, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=327
Katsouvanni. K; Touloumi. G: Samoli. E; Grvparis. A; Le Tertre. A; Monopolis. Y; Rossi. G: Zmirou. D;
Ballester. F; Boumghar. A; Anderson. HR; Wojtyniak. B; Paldv. A; Braunstein. R; Pekkanen. J:
Schindler. C: Schwartz. J. (2001). Confounding and effect modification in the short-term effects of
ambient particles on total mortality: Results from 29 European cities within the APHEA2 project.
Epidemiology 12: 521-531.
Katsouvanni. K; Touloumi. G: Samoli. E; Petasakis. Y; Analitis. A; Le Tertre. A; Rossi. G: Zmirou. D;
Ballester. F; Boumghar. A; Anderson. HR; Woityniak. B; Paldv. A; Braunstein. R; Pekkanen. J:
Schindler. C: Schwartz. J. (2003). Sensitivity analysis of various models of short-term effects of
ambient particles on total mortality in 29 cities in APHEA2. In Revised Analyses of Time-series Studies
of Air Pollution and Health (pp. 157-164). Boston, MA: Health Effects Institute.
Kehrl. HR; Peden. DB; Ball. B; Folinsbee. LJ; Horstman. D. (1999). Increased specific airway reactivity of
persons with mild allergic asthma after 7.6 hours of exposure to 0.16 ppm ozone. J Allergy Clin
Immunol 104: 1198-1204.
Kelishadi. R: Mirghaffari. N; Poursafa. P; Gidding. SS. (2009). Lifestyle and environmental factors
associated with inflammation, oxidative stress and insulin resistance in children. Atherosclerosis 203:
311-319. http://dx.doi.0rg/10.1016/i.atherosclerosis.2008.06.022
November 2013 4-298 DRAFT: Do Not Cite or Quote
-------
Kelly. FJ; Blomberg. A; Frew. A; Holgate. ST; Sandstrom. T. (1996a). Antioxidant kinetics in lung lavage
fluid following exposure of humans to nitrogen dioxide. Am J Respir Crit Care Med 154: 1700-1705.
http://dx.doi.0rg/10.1164/ajrccm.154.6.8970358
Kharitonov. SA; Barnes. PJ. (2000). Clinical aspects of exhaled nitric oxide [Review]. Eur Respir J 16:
781-792.
Kikuya. M; Hozawa. A; Ohokubo. T; Tsuji. I; Michimata. M; Matsubara. M; Ota. M; Nagai. K; Araki. T;
Satoh. H; Ito. S: Hisamichi. S: Imai. Y. (2000). Prognostic significance of blood pressure and heart rate
variabilities: the Ohasama study. Hypertension 36: 901-906. http://dx.doi.Org/10.1161/01.HYP.36.5.901
Kinsella. J: Carter. R; Reid. WH; Campbell. D; Clark. CJ. (1991). Increased airways reactivity after smoke
inhalation. Lancet 337: 595-597.
Kjaergaard. SK; Rasmussen. TR. (1996). Clinical studies of effects of nitrogen oxides in healthy and
asthmatic subjects. Cent Eur J Public Health 4: 23-26.
Kleinman. MT; Bailey. RM; Linn. WS: Anderson. KR; Whynot JD; Shamoo. DA; Hackney. JD. (1983).
Effects of 0.2 ppm nitrogen dioxide on pulmonary function and response to bronchoprovocation in
asthmatics. J Toxicol Environ Health 12: 815-826. http://dx.doi.org/10.1080/15287398309530472
Ko. FW: Tarn. W: Wong. TW: Chan. DP: Tung. AH: Lai. CK: PS. H. (2007a). Temporal relationship
between air pollutants and hospital admissions for chronic obstructive pulmonary disease in Hong
Kong. Thorax 62: 780-785. http://dx.doi.org/10.1136/thx.2006.076166
Ko. FWS: Tarn. W: Wong. TW: Lai. CKW. (2007b). Effects of air pollution on asthma hospitalization rates
in different age groups in Hong Kong. Clin Exp Allergy 37: 1312-1319.
http://dx.doi.0rg/10.llll/i.1365-2222.2007.02791.x
Kobavashi. T; Shinozaki. Y. (1990). Effect of subacute exposure to nitrogen dioxide on the airway
responsiveness of guinea pig. Agents Actions Suppl 31: 71-74.
Koenig. JQ; Covert. PS: Marshall. SG: Van Belle. G: Pierson. WE. (1987). The effects of ozone and
nitrogen dioxide on pulmonary function in healthy and in asthmatic adolescents. Am J Respir Crit Care
Med 136: 1152-1157. http://dx.doi.Org/10.1164/airccm/136.5.1152
Kulle. TJ; Clements. ML. (1988). Susceptibility to virus infection with exposure to nitrogen dioxide.
Cambridge, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=243
Kunimoto. M; Mochitate. K; Kava. K; Miura. T; Kubota. K. (1984). Effects of nitrogen dioxide on red
blood cells of rats: Alterations of cell membrane components and populational changes of red blood
cells during in vivo exposure to NO2. Environ Res 33: 361-369.
La Rovere. MT; Pinna. GD; Maestri. R; Mortara. A; Capomolla. S: Febo. O: Ferrari. R; Franchini. M;
Gnemmi. M; Opasich. C: Riccardi. PG: Traversi. E; Cobelli. F. (2003). Short-term heart rate variability
strongly predicts sudden cardiac death in chronic heart failure patients. Circulation 107: 565-570.
Lagorio. S: Forastiere. F; Pistelli. R: lavarone. I; Michelozzi. P; Fano. V; Marconi. A; Ziemacki. G: Ostro.
BD. (2006). Air pollution and lung function among susceptible adult subjects: A panel study. Environ
Health 5: 11. http://dx.doi.org/10.1186/1476-069X-5-ll
Langrish. JP; Lundback. M; Barath. S: Soderberg. S: Mills. NL; Newbv. DE; Sandstrom. T; Blomberg. A.
(2010). Exposure to nitrogen dioxide is not associated with vascular dysfunction in man. Inhal Toxicol
22: 192-198. http://dx.doi.org/10.3109/08958370903144105
Langrish. JP; Unosson. J: Bosson-Damewood. J: Muala. A; Barath. S: Soderberg. S: Blackwell. S:
Pourazar. J: Megson. IL; Sandstrom. T; Mills. NL; Blomberg. A; Newbv. DE. (2011). Diesel exhaust
inhalation induced vascular dysfunction: The role of nitric oxide [Abstract]. J Am Coll Cardiol 57:
E1429. http://dx.doi.org/10.1016/S0735-1097(ll)61429-0
Larrieu. S: Jusot. JF; Blanchard. M; Prouvost. H; Declercq. C: Fabre. P; Pascal. L; Le Tertre. A; Wagner. V;
Riviere. S: Chardon. B; Borelli. D; Cassadou. S: Eilstein. D; Lefranc. A. (2007). Short term effects of
air pollution on hospitalizations for cardiovascular diseases in eight French cities: The PSAS program.
Sci Total Environ 387: 105-112. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.025
November 2013 4-299 DRAFT: Do Not Cite or Quote
-------
Last. JA; Warren. PL. (1987). Synergistic interaction between nitrogen dioxide and respirable aerosols of
sulfuric acid or sodium chloride on rat lungs. Toxicol Appl Pharmacol 90: 34-42.
http://dx.doi.org/10.1016/0041-008X(87)90303-6
Laube. BL; Norman. PS: Adams. GK. (1992). The effect of aerosol distribution on airway responsiveness
to inhaled methacholine in patients with asthma. J Allergy Clin Immunol 89: 510-518.
http://dx.doi.org/10.1016/0091-6749(92)90317-U
Laumbach. RJ; Rich. DQ; Gandhi. S: Amorosa. L; Schneider. S: Zhang. J; Ohman-Strickland. P; Gong. J;
Lelyanov. O; Kipen. HM. (2010). Acute changes in heart rate variability in subjects with diabetes
following a highway traffic exposure. J Occup Environ Med 52: 324-331.
http://dx.doi.org/10.1097/JOM.Ob013e3181d241fa
Laurent. O: Pedrono. G: Filleul. L; Segala. C: Lefranc. A; Schillinger. C: Riviere. E; D. B. (2009).
Influence of socioeconomic deprivation on the relation between air pollution and beta-agonist sales for
asthma. Chest 135: 717-723. http://dx.doi.org/10.1378/chest.08-1604
Lee. PC: Talbott. EO; Roberts. JM; Catov. JM; Sharma. RK; Ritz. B. (20lie). Particulate air pollution
exposure and C-reactive protein during early pregnancy. Epidemiology 22: 524-531.
http://dx.doi.org/10.1097/EDE.Ob013e31821c6c58
Lew. D; Sheppard. L; Checkowav. H; Kaufman. J: Lumlev. T; Koenig. J: Siscovick. D. (2001). A case-
crossover analysis of paniculate matter air pollution and out-of-hospital primary cardiac arrest.
Epidemiology 12: 193-199.
Li. H; Han. M; Guo. L; Li. G: Sang. N. (2011 a). Oxidative stress, endothelial dysfunction and
inflammatory response in rat heart to NO inhalation exposure. Chemosphere 82: 1589-1596.
http://dx.doi.0rg/10.1016/i.chemosphere.2010.ll.055
Li. S: Batterman. S: Wasilevich. E; Wahl. R; Wirth. J: Su. FC: Mukherjee. B. (20lib). Association of daily
asthma emergency department visits and hospital admissions with ambient air pollutants among the
pediatric Medicaid population in Detroit: Time-series and time-stratified case-crossover analyses with
threshold effects. Environ Res 111: 1137-1147. http://dx.doi.Org/10.1016/j.envres.2011.06.002
Lin. W: Huang. W: Zhu. T; Hu. M; Brunekreef. B; Zhang. Y; Liu. X; Cheng. H; Gehring. U; Li. C: Tang.
2L (2011). Acute respiratory inflammation in children and black carbon in ambient air before and during
the 2008 Beijing Olympics. Environ Health Perspect 119: 1507-1512.
http://dx.doi.org/10.1289/ehp.1103461
Linares. C: Diaz. J. (2010). Short-term effect of concentrations of fine paniculate matter on hospital
admissions due to cardiovascular and respiratory causes among the over-75 age group in Madrid, Spain.
Public Health 124: 28-36. http://dx.doi.Org/10.1016/i.puhe.2009.ll.007
Link. MS: Luttmann-Gibson. H; Schwartz. J: Mittleman. MA; Wessler. B; Gold. PR; Dockery. DW: Laden.
E (2013). Acute exposure to air pollution triggers atrial fibrillation. J Am Coll Cardiol 62.
http://dx.doi.0rg/10.1016/i.jacc.2013.05.043
Linn. WS: Shamoo. DA; Anderson. KR; Peng. RC: Avol. EL; Hackney. JD; Gong. H. Jr. (1996). Short-term
air pollution exposures and responses in Los Angeles area schoolchildren. J Expo Sci Environ
Epidemiol 6: 449-472.
Linn. WS: Shamoo. DA; Spier. CE; Valencia. LM; Anzar. UT; Venet. TG: Avol. EL; Hackney. JD. (1985a).
Controlled exposure of volunteers with chronic obstructive pulmonary disease to nitrogen dioxide. Arch
Environ Health 40: 313-317. http://dx.doi.org/10.1080/00039896.1985.10545939
Linn. WS: Solomon. JC: Trim. SC: Spier. CE; Shamoo. DA; Venet. TG: Avol. EL; Hackney. JD. (1985b).
Effects of exposure to 4 ppm nitrogen dioxide in healthy and asthmatic volunteers. Arch Environ Occup
Health 40: 234-239.
Linn. WS: Szlachcic. Y; Gong. H. Jr; Kinnev. PL; Berhane. KT. (2000). Air pollution and daily hospital
admissions in metropolitan Los Angeles. Environ Health Perspect 108: 427-434.
http://dx.doi.org/10.2307/3454383
November 2013 4-300 DRAFT: Do Not Cite or Quote
-------
Liu. L; Poon. R; Chen. L; Frescura. AM; Montuschi. P; Ciabattoni. G: Wheeler. A; Dales. R. (2009b).
Acute effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in
asthmatic children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehpll813
Ljungman. P; Bellander. T; Schneider. A; Breitner. S: Forastiere. F; Hampel. R: Illig. T; Jacquemin. B;
Katsouvanni. K; von Klot. S: Koenig. W: Lanki. T; Nyberg. F; Pekkanen. J; Pistelli. R: Pitsavos. C:
Rosenqvist. M; Sunyer. J; Peters. A. (2009). Modification of the interleukin-6 response to air pollution
by interleukin-6 and fibrinogen polymorphisms. Environ Health Perspect 117: 1373-1379.
Ljungman. PLS: Berglind. N; Holmgren. C: Gadler. F; Edvardsson. N; Pershagen. G: Rosenqvist. M;
Sjogren. B; Bellander. T. (2008). Rapid effects of air pollution on ventricular arrhythmias. Eur Heart J
29: 2894-2901. http://dx.doi.Org/l0.1093/eurheartj/ehn463
Llorca. J: Salas. A; Prieto-Salceda. D; Chinchon-Bengoechea. V: Delgado-Rodriguez. M. (2005). Nitrogen
dioxide increases cardiorespiratory admissions in Torrelavega (Spain). J Environ Health 68: 30-35.
Lu. KD; Brevsse. PN; Diette. GB; Curtin-Brosnan. J: Aloe. C: Williams. PL; Peng. RD; Mccormack. MC:
Matsui. EC. (2013). Being overweight increases susceptibility to indoor pollutants among urban
children with asthma. J Allergy Clin Immunol 131: 1017-1023.e!013.
http://dx.doi.org/10.1016/i.iaci.2Q12.12.1570
Madsen. C: Durand. KL; Nafstad. P: Schwarze. PE; R0nningen. KS: Haheim. LL. (2008). Associations
between environmental exposures and serum concentrations of Clara cell protein among elderly men in
Oslo, Norway. Environ Res 108: 354-360. http://dx.doi.Org/10.1016/i.envres.2008.07.017
Maestrelli. P: Canova. C: Scapellato. ML; Visentin. A; Tessari. R; Bartolucci. GB; Simonato. L; Lotti. M.
(2011). Personal exposure to paniculate matter is associated with worse health perception in adult
asthma. J Investig Allergol Clin Immunol 21: 120-128.
Magnussen. H; Reuss. G: Jorres. R. (1986). Airway response to methacholine during exercise induced
refractoriness in asthma. Thorax 41: 667-670.
Malm-Erjefalt. M; Persson. CG: Erjefalt JS. (2001). Degranulation status of airway tissue eosinophils in
mouse models of allergic airway inflammation. Am J Respir Cell Mol Biol 24: 352-359.
http://dx.doi.0rg/10.1165/ajrcmb.24.3.4357
Mann. JK; Balmes. JR; Bruckner. TA; Mortimer. KM; Margolis. HG: Pratt. B; Hammond. SK; Lurmann. F;
Tager. IB. (2010). Short-term effects of air pollution on wheeze in asthmatic children in Fresno,
California. Environ Health Perspect 118: 1497-1502. http://dx.doi.org/10.1289/ehp.0901292
Mann. JK; Tager. IB; Lurmann. F; Segal. M; Quesenberrv. CP. Jr; Lugg. MM; Shan. J: Van den Eeden. SK.
(2002). Air pollution and hospital admissions for ischemic heart disease in persons with congestive
heart failure or arrhythmia. Environ Health Perspect 110: 1247-1252.
Martins. PC: Valente. J: Papoila. AL; Caires. I; Araujo-Martins. J: Mata. P: Lopes. M; Torres. S: Rosado-
Pinto. J: Borrego. C: Annesi-Maesano. I; Neuparth. N. (2012). Airways changes related to air pollution
exposure in wheezing children. Eur Respir J 39: 246-253. http://dx.doi.org/10.1183/09031936.00025111
McCreanor. J: Cullinan. P; Nieuwenhuijsen. MJ; Stewart-Evans. J: Malliarou. E; Jarup. L; Harrington. R;
Svartengren. M; Han. IK; Ohman-Strickland. P: Chung. KF; Zhang. J. (2007). Respiratory effects of
exposure to diesel traffic in persons with asthma. N Engl J Med 357: 2348-2358.
http://dx.doi.org/10.1056/NEJMoa071535
Mehta. S: Ngo. LH; Dzung. DV; Cohen. A; Thach. TQ; Dan. VX; Tuan. ND; Giang. LT. (2013). Air
pollution and admissions for acute lower respiratory infections in young children of Ho Chi Minh City.
Air Qual Atmos Health 6: 167-179. http://dx.doi.org/10.1007/sll869-011-0158-z
Mersch. J: Dyce. BJ; Haverback. BJ; Sherwin. RP. (1973). Diphosphoglycerate content of red blood cells:
Measurements in guinea pigs exposed to 04 ppm nitrogen dioxide. Arch Environ Health 27: 94-95.
http://dx.doi.org/10.1080/00039896.1973.10666326
November 2013 4-301 DRAFT: Do Not Cite or Quote
-------
Metzger. KB; Klein. M; Flanders. WD; Peel. JL; Mulholland. JA; Langberg. JJ; Tolbert. PE. (2007).
Ambient air pollution and cardiac arrhythmias in patients with implantable defibrillators. Epidemiology
18: 585-592. http://dx.doi.org/10.1097/EDE.Ob013e318124ffOe
Min. KB; Min. JY; Cho. SI; Paek. D. (2008). The relationship between air pollutants and heart-rate
variability among community residents in Korea. Inhal Toxicol 20: 435-444.
http://dx.doi.org/10.1080/08958370801903834
Mochitate. K; Miura. T. (1984). In vivo effect of nitrogen dioxide on the activities of glycolytic enzymes in
red blood cells of rats. Toxicol Lett 22: 315-321.
Mochitate. K; Takahashi. Y; Ohsumi. T; Miura. T. (1986). Activation and increment of alveolar
macrophages induced by nitrogen dioxide. J Toxicol Environ Health 17: 229-239.
http://dx.doi.org/10.1080/15287398609530819
Mohsenin. V. (1987a). Airway responses to nitrogen dioxide in asthmatic subjects. J Toxicol Environ
Health 22: 371-380. http://dx.doi.org/10.1080/15287398709531080
Mohsenin. V. (1987b). Effect of vitamin C on NO2-induced airway hyperresponsiveness in normal
subjects: Arandomized double-blind experiment. Am Rev Respir Dis 136: 1408-1411.
http://dx.doi.0rg/10.1164/ajrccm/136.6.1408
Mohsenin. V. (1988). Airway responses to 2.0 ppm nitrogen dioxide in normal subjects. Arch Environ
Health 43: 242-246. http://dx.doi.org/10.1080/00039896.1988.9934941
Moolgavkar. SH. (2003). Air pollution and daily deaths and hospital admissions in Los Angeles and Cook
counties. In Revised Analyses of the National Morbidity, Mortality, and Air Pollution Study, Part II (pp.
183-198). Boston, MA: Health Effects Institute.
Moolgavkar. SH; Mcclellan. RO; Dewanji. A; Turim. J: Luebeck. EG: Edwards. M. (2013). Time-series
analyses of air pollution and mortality in the United States: A subsampling approach. Environ Health
Perspect 121: 73-78. http://dx.doi.org/10.1289/ehp.1104507
Moon. JS: Kim. YS: Kim. JH; Son. BS: Kim. PS: Yang. W. (2009). Respiratory health effects among
schoolchildren and their relationship to air pollutants in Korea. Int J Environ Health Res 19: 31-48.
http://dx.doi.org/10.1080/09603120802272201
Morgan. G: Corbett. S: Wlodarczyk. J. (1998). Air pollution and hospital admissions in Sydney, Australia,
1990 to 1994. Am J Public Health 88: 1761-1766. http://dx.doi.org/10.2105/AJPH.88.12.1761
Morrow. PE; Utell. MJ. (1989a). Responses of susceptible subpopulations to nitrogen dioxide [HEI] (pp. 1-
45). (ISSN 1041-5505 Research Report 23). Cambridge, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=235
Morrow. PE; Utell. MJ. (1989b). Responses of susceptible subpopulations to nitrogen dioxide: Appendix H.
Selected raw data [HEI]. (Research Report 23). Cambridge, MA: Health Effects Institute.
Morrow. PE; Utell. MJ: Bauer. MA; Smeglin. AM; Frampton. MW: Cox. C: Speers. DM; Gibb. FR. (1992).
Pulmonary performance of elderly normal subjects and subjects with chronic obstructive pulmonary
disease exposed to 0.3 ppm nitrogen dioxide. Am J Respir Crit Care Med 145: 291-300.
http://dx.doi.0rg/10.1164/ajrccm/145.2 Pt 1.291
Mortimer. KM; Neas. LM; Dockery. DW: Redline. S: Tager. IB. (2002). The effect of air pollution on
inner-city children with asthma. Eur Respir J 19: 699-705.
http://dx.doi.org/10.1183/09031936.02.00247102
Moshammer. H; Hutter. HP; Hauck. H; Neuberger. M. (2006). Low levels of air pollution induce changes
of lung function in a panel of schoolchildren. Eur Respir J 27: 1138-1143.
http://dx.doi.org/10.1183/09031936.06.00089605
Moss. OR; Oldham. MJ. (2006). Dosimetry counts: Molecular hypersensitivity may not drive pulmonary
hyperresponsiveness. J Aerosol Med 19: 555-564. http://dx.doi.org/10.1089/iam.2006.19.555
November 2013 4-302 DRAFT: Do Not Cite or Quote
-------
Miiller. B; Schafer. H; Barth. P; von Wichert. P. (1994). Lung surfactant components in bronchoalveolar
lavage after inhalation of NO2 as markers of altered surfactant metabolism. Lung 172: 61-72.
http://dx.doi.org/10.1007/BF00185077
Murray. AB; Ferguson. AC: Morrison. B. (1981). Airway responsiveness to histamine as a test for overall
severity of asthma in children. J Allergy Clin Immunol 68: 119-124. http://dx.doi.org/10.1016/0091-
6749(8 D90169-X
Mustafa. MG: Elsaved. NM; von Dohlen. FM; Hassett CM; Postlethwait. EM; Quinn. CL; Graham. JA;
Gardner. DE. (1984). A comparison of biochemical effects of nitrogen dioxide, ozone, and their
combination in mouse lung. Toxicol Appl Pharmacol 72: 82-90.
Nakaiima. T; Kusumoto. S. (1968). Effect of nitrogen dioxide exposure on the contents of reduced
glutathione in mouse lung. Food Sanitation Research (English) 6: 17-21.
Nuvolone. D; Balzi. D; Chini. M; Scala. D; Giovannini. F; Barchielli. A. (2011). Short-term association
between ambient air pollution and risk of hospitalization for acute myocardial infarction: Results of the
cardiovascular risk and air pollution in Tuscany (RISCAT) study. Am J Epidemiol 174: 63-71.
http://dx.doi.org/10.1093/aie/kwr046
O'Byrne. PM; Gauvreau. GM; Brannan. JD. (2009). Provoked models of asthma: What have we learnt?
[Review]. Clin Exp Allergy 39: 181-192. http://dx.doi.org/10.1111/i. 1365-2222.2008.03172.x
O'Connor. GT; Neas. L; Vaughn. B; Kattan. M; Mitchell. H; Grain. EF; Evans. R. Ill; Gruchalla. R;
Morgan. W: Stout. J: Adams. GK; Lippmann. M. (2008). Acute respiratory health effects of air pollution
on children with asthma in US inner cities. J Allergy Clin Immunol 121: 1133-1139.
http://dx.doi.0rg/10.1016/i.jaci.2008.02.020
Odajima. H; Yamazaki. S: Nitta. H. (2008). Decline in peak expiratory flow according to hourly short-term
concentration of paniculate matter in asthmatic children. Inhal Toxicol 20: 1263-1272.
http://dx.doi.org/10.1080/08958370802311151
Oftedal. B; Brunekreef. B; Nystad. W: Madsen. C: Walker. SE; Nafstad. P. (2008). Residential outdoor air
pollution and lung function in schoolchildren. Epidemiology 19: 129-137.
http://dx.doi.org/10.1097/EDE.Ob013e31815c0827
Ohashi. Y; Nakai. Y; Sugiura. Y; Ohno. Y; Okamoto. H; Tanaka. A; Kakinoki. Y; Havashi. M. (1994).
Nitrogen dioxide-induced eosinophilia and mucosal injury in the nose of the guinea pig. Acta
Otolaryngol 114: 547-551.
Orazzo. F; Nespoli. L; Ito. K; Tassinari. D; Giardina. D: Funis. M; Cecchi. A; Trapani. C: Forgeschi. G:
Vignini. M; Nosetti. L; Pigna. S: Zanobetti. A. (2009). Air pollution, aeroallergens, and emergency
room visits for acute respiratory diseases and gastroenteric disorders among young children in six
Italian cities. Environ Health Perspect 117: 1780-1785. http://dx.doi.org/10.1289/ehp.0900599
Orehek. J: Massari. JP; Gayrard. P; Grimaud. C: Charpin. J. (1976). Effect of short-term, low-level nitrogen
dioxide exposure on bronchial sensitivity of asthmatic patients. J Clin Invest 57: 301-307.
http://dx.doi.org/10.1172/JCI108281
Oudin. A; Stromberg. U; Jakobsson. K; Stroh. E; Bjork. J. (2010). Estimation of short-term effects of air
pollution on stroke hospital admissions in southern Sweden. Neuroepidemiology 34: 131-142.
http://dx.doi.org/10.1159/000274807
Pagani. P; Romano. M; Erroi. A; Ferro. M; Salmona. M. (1994). Biochemical effects of acute and subacute
nitrogen dioxide exposure in rat lung and bronchoalveolar lavage fluid. Arch Environ Contam Toxicol
27: 426-430. http://dx.doi.org/10.1007/BF00213182
Panel NAEaPPE (National Asthma Education and Prevention Program Expert Panel). (2007). Expert panel
reports: guidelines for the diagnosis and management of asthma. J Allergy Clin Immunol 120: S93.
http://dx.doi.0rg/10.1016/i.iaci.2007.09.043
November 2013 4-303 DRAFT: Do Not Cite or Quote
-------
Park. JW: Lim. YH; Kyung. SY; An. CH; Lee. SP; Jeong. SH; Ju. Y. -S. (2005). Effects of ambient
paniculate matter on peak expiratory flow rates and respiratory symptoms of asthmatics during Asian
dust periods in Korea. Respirology 10: 470-476. http://dx.doi.Org/10.llll/i.1440-1843.2005.00728.x
Park. SK; Auchincloss. AH; O'Neill. MS: Prineas. R; Correa. JC: Keeler. J; Barr. RG: Kaufman. JD; Diez
Roux. AV. (2010). Paniculate air pollution, metabolic syndrome, and heart rate variability: The Multi-
Ethnic Study of Atherosclerosis (MESA). Environ Health Perspect 118: 1406-1411.
http://dx.doi.org/10.1289/ehp.0901778
Parker. RF; Davis. JK; Cassell. GH; White. H; Dziedzic. D; Blalock. DK; Thorp. RB; Simecka. JW. (1989).
Short-term exposure to nitrogen dioxide enhances susceptibility to murine respiratory mycoplasmosis
and decreases intrapulmonary killing of Mycoplasma pulmonis. Am J Respir Crit Care Med 140: 502-
512. http://dx.doi.0rg/10.1164/ajrccm/140.2.502
Patel. MM; Chillrud. SN; Correa. JC: Hazi. Y; Feinberg. M; Kc. D; Prakash. S: Ross. JM; Lew. D; Kinnev.
PL. (2010). Traffic-related particulate matter and acute respiratory symptoms among New York City
area adolescents. Environ Health Perspect 118: 1338-1343. http://dx.doi.org/10.1289/ehp.0901499
Patel. MM; Chillrud. SN; Deepti. KC: Ross. JM; Kinnev. PL. (2013). Traffic-related air pollutants and
exhaled markers of airway inflammation and oxidative stress in New York City adolescents. Environ
Res 121: 71-78. http://dx.doi.Org/10.1016/i.envres.2012.10.012
Pathmanathan. S: Krishna. MT; Blomberg. A; Helledav. R; Kelly. FJ; Sandstrom. T; Holgate. ST; Wilson.
SJ: Frew. AJ. (2003). Repeated daily exposure to 2 ppm nitrogen dioxide upregulates the expression of
IL-5, IL-10, IL-13, and ICAM-1 in the bronchial epithelium of healthy human airways. Occup Environ
Med 60: 892-896. http://dx.doi.org/10.1136/oem.60.ll.892
Peacock. JL; Anderson. HR; Bremner. SA; Marston. L; Seemungal. TA; Strachan. DP; Wedzicha. JA.
(2011). Outdoor air pollution and respiratory health in patients with COPD. Thorax 66: 591-596.
http://dx.doi.org/10.1136/tfax.2010.155358
Peacock. JL; Symonds. P; Jackson. P; Bremner. SA; Scarlett. JF; Strachan. DP; Anderson. HR. (2003).
Acute effects of winter air pollution on respiratory function in schoolchildren in southern England.
Occup Environ Med 60: 82-89. http://dx.doi.Org/10.1136/oem.60.2.82
Peel. JL; Klein. M; Flanders. WD; Mulholland. JA; Freed. G: Tolbert. PE. (2011). Ambient air pollution
and apnea and bradycardia in high-risk infants on home monitors. Environ Health Perspect 119: 1321-
1327. http://dx.doi.org/10.1289/ehp.1002739
Peel. JL: Metzger. KB; Klein. M; Flanders. WD: Mulholland. JA: Tolbert. PE. (2007). Ambient air
pollution and cardiovascular emergency department visits in potentially sensitive groups. Am J
Epidemiol 165: 625-633. http://dx.doi.org/10.1093/aie/kwk051
Peel. JL: Tolbert. PE: Klein. M: Metzger. KB: Flanders. WD: Todd. K: Mulholland. JA: Ryan. PB:
Frumkin. H. (2005). Ambient air pollution and respiratory emergency department visits. Epidemiology
16: 164-174. http://dx.doi.org/10.1097/01.ede.0000152905.42113.db
Pilotto. LS: Nitschke. M; Smith. BJ; Pisaniello. D: Ruffin. RE; McElroy. HJ; Martin. J: Hiller. JE. (2004).
Randomized controlled trial of unflued gas heater replacement on respiratory health of asthmatic
schoolchildren. Int J Epidemiol 33: 208-214.
Pinkston. P; Smeglin. A; Roberts. NJ. Jr; Gibb. FR; Morrow. PE; Utell. MJ. (1988). Effects of in vitro
exposure to nitrogen dioxide on human alveolar macrophage release of neutrophil chemotactic factor
and interleukin-1. Environ Res 47: 48-58. http://dx.doi.org/10.1016/80013-9351(88)80020-3
Poloniecki. JD; Atkinson. RW: Ponce de Leon. A; Anderson. HR. (1997). Daily time series for
cardiovascular hospital admissions and previous day's air pollution in London, UK. Occup Environ Med
54: 535-540.
Ponka. A; Virtanen. M. (1996). Low-level air pollution and hospital admissions for cardiac and
cerebrovascular diseases in Helsinki. Am J Public Health 86: 1273-1280.
http://dx.doi.Org/10.2105/AJPH.86.9.1273
November 2013 4-304 DRAFT: Do Not Cite or Quote
-------
Posin. C: Clark. K; Jones. MP; Patterson. JV: Buckley. RD; Hackney. JD. (1978). Nitrogen dioxide
inhalation and human blood biochemistry. Arch Environ Occup Health 33: 318-324.
Postma. DS: Boezen. HM. (2004). Rationale for the Dutch hypothesis. Allergy and airway
hyperresponsiveness as genetic factors and their interaction with environment in the development of
asthma and COPD [Review]. Chest 126: 96S-104S; discussion 159S-161S.
http://dx.doi.0rg/10.1378/chest.126.2 suppl 1.96S
Povnter. ME; Persinger. RL; Irvin. CG: Butnor. KJ; van Hirtum. H; Slav. W: Heintz. NH; Robbins. J:
Hemenwav. D; Taatjes. DJ: Janssen-Heininger. Y. (2006). Nitrogen dioxide enhances allergic airway
inflammation and hyperresponsiveness in the mouse. Am J Physiol Lung Cell Mol Physiol 290: LI 44-
L152. http://dx.doi.org/10.1152/ajplung.00131.2005
Proust. B; Lacroix. G: Robidel. F; Marliere. M; Lecomte. A; Vargaftig. BB. (2002). Interference of a short-
term exposure to nitrogen dioxide with allergic airways responses to allergenic challenges in BALB/c
mice. Mediators Inflamm 11: 251-260. http://dx.doi.org/10.1080/096293502900000113
Purvis. MR; Ehrlich. R. (1963). Effect of atmospheric pollutants on susceptibility to respiratory infection: II
Effect of nitrogen dioxide. J Infect Dis 113: 72-76.
Qian. Z; Lin. HM; Chinchilli. VM; Lehman. EB; Duan. Y; Craig. TJ; Wilson. WE; Liao. D; Lazarus. SC:
Bascom. R. (2009a). Interaction of ambient air pollution with asthma medication on exhaled nitric oxide
among asthmatics. Arch Environ Occup Health 64: 168-176.
http://dx.doi.org/10.1080/19338240903240616
Qian. Z; Lin. HM; Chinchilli. VM; Lehman. EB; Stewart. WF; Shah. N; Duan. Y; Craig. TJ: Wilson. WE;
Liao. D: Lazarus. SC: Bascom. R. (2009b). Associations between air pollution and peak expiratory flow
among patients with persistent asthma. J Toxicol Environ Health A 72: 39-46.
http://dx.doi.org/10.1080/15287390802445517
Ramos-Bonilla. JP; Brevsse. PN; Dominici. F; Geyh. A; Tankerslev. CG. (2010). Ambient air pollution
alters heart rate regulation in aged mice. Inhal Toxicol 22: 330-339.
http://dx.doi.org/10.3109/08958370903349365
Ranzi. A; Gambini. M; Spattini. A; Galassi. C: Sesti. D: Bedeschi. M; Messori. A; Baroni. A; Cavagni. G:
Lauriola. P. (2004). Air pollution and respiratory status in asthmatic children: Hints for a locally based
preventive strategy AIRE study. Eur J Epidemiol 19: 567-576.
Rasmussen. TR; Kjaergaard. SK; Tarp. U; Pedersen. OF. (1992). Delayed effects of NO2 exposure on
alveolar permeability and glutathione peroxidase in healthy humans. Am Rev Respir Dis 146: 654-659.
http://dx.doi.0rg/10.1164/ajrccm/146.3.654
Reddel. HK; Taylor. PR; Bateman. ED; Boulet. LP; Boushev. HA; Busse. WW: Casale. TB; Chanez. P;
Enright. PL; Gibson. PG: de Jongste. JC: Kerstjens. HA; Lazarus. SC: Lew. ML; O'Byrne. PM;
Partridge. MR; Pavord. ID; Sears. MR; Sterk. PJ; Stoloff. SW: Sullivan. SD; Szefler. SJ: Thomas. MD;
Wenzel. SE; Exacerbations. ATSERSTFoACa. (2009). An official American Thoracic Society/European
Respiratory Society statement: Asthma control and exacerbations: Standardizing endpoints for clinical
asthma trials and clinical practice. Am J Respir Crit Care Med 180: 59-99.
http://dx.doi.org/10.1164/rccm.200801-060ST
Ren. C: Fang. S: Wright. RO: Suh. H; Schwartz. J. (2011). Urinary 8-hydroxy-2'-deoxyguanosine as a
biomarker of oxidative DNA damage induced by ambient pollution in the Normative Aging Study.
Occup Environ Med 68: 562-569. http://dx.doi.org/10.1136/oem.2010.056358
Rich. DO; Kipen. HM; Huang. W: Wang. G: Wang. Y; Zhu. P; Ohman-Strickland. P; Hu. M; Philipp. C:
Diehl. SR; Lu. SE; Tong. J: Gong. J: Thomas. D: Zhu. T; Zhang. JJ. (2012). Association between
changes in air pollution levels during the Beijing Olympics and biomarkers of inflammation and
thrombosis in healthy young adults. JAMA 307: 2068-2078. http://dx.doi.org/10.1001/jama.2012.3488
Rich. DO; Kipen. HM; Zhang. J: Kamat. L; Wilson. AC: Kostis. JB. (2010). Triggering of transmural
infarctions, but not nontransmural infarctions, by ambient fine particles. Environ Health Perspect 118:
1229-1234. http://dx.doi.org/10.1289/ehp.0901624
November 2013 4-305 DRAFT: Do Not Cite or Quote
-------
Riedl. MA; Diaz-Sanchez. D; Linn. WS: Gong. H. Jr; Clark. KW: Effros. RM; Miller. JW: Cocker. PR;
Berhane. KT. (2012). Allergic inflammation in the human lower respiratory tract affected by exposure to
diesel exhaust [HEI] (pp. 5-43; discussion 45-64). (ISSN 1041-5505 Research Report 165). Boston,
MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=373
Robison. TW: Duncan. DP; Forman. HJ. (1990). Chemoattractant and leukotriene B4 production from rat
alveolar macrophages exposed to nitrogen dioxide. Am J Respir Cell Mol Biol 3: 21-26.
http://dx.doi.0rg/10.1165/ajrcmb/3.l.21
Robison. TW: Forman. HJ. (1993). Dual effect of nitrogen dioxide on rat alveolar macrophage arachidonate
metabolism. Exp Lung Res 19: 21-36.
Robison. TW: Murphy. JK; Beyer. LL; Richters. A; Forman. HJ. (1993). Depression of stimulated
arachidonate metabolism and superoxide production in rat alveolar macrophages following in vivo
exposure to 0.5 ppm NO2. J Toxicol Environ Health 38: 273-292.
http://dx.doi.org/10.1080/15287399309531718
Rodriguez. C: Tonkin. R; Hevworth. J: Kusel. M; De Klerk. N; Sly. PD: Franklin. P; Runnion. T; Blocklev.
A; Landau. L; Hinwood. AL. (2007). The relationship between outdoor air quality and respiratory
symptoms in young children. Int J Environ Health Res 17: 351-360.
http://dx.doi.org/10.1080/09603120701628669
Roemer. W: Hoek. G: Brunekreef. B; Haluszka. J: Kalandidi. A; Pekkanen. J. (1998). Daily variations in air
pollution and respiratory health in a multicentre study: The PEACE project. Eur Respir J 12: 1354-1361.
Roger. LJ; Horstman. DH; McDonnell. W: Kehrl. H; Ives. PJ: Seal. E; Chapman. R; Massaro. E. (1990).
Pulmonary function, airway responsiveness, and respiratory symptoms in asthmatics following exercise
in NO2. Toxicol Ind Health 6: 155-171.
Rombout. PJ: Dormans. JA; Marra. M; Van Esch. GJ. (1986). Influence of exposure regimen on nitrogen
dioxide-induced morphological changes in the rat lung. Environ Res 41: 466-480.
Romieu. I; Barraza-Villarreal. A; Escamilla-Nunez. C: Almstrand. AC: Diaz-Sanchez. D; Sly. PD: Olin.
AC. (2008). Exhaled breath malondialdehyde as a marker of effect of exposure to air pollution in
children with asthma. J Allergy Clin Immunol 121: 903-909.
http://dx.doi.0rg/10.1016/i.jaci.2007.12.004
Romieu. I; Ramirez-Aguilar. M; Sienra-Monge. JJ: Moreno-Macias. H; Del Rio-Navarro. BE; David. G:
Marzec. J: Hernandez-Avila. M; London. S. (2006). GSTM1 and GSTP1 and respiratory health in
asthmatic children exposed to ozone. Eur Respir J 28: 953-959.
http://dx.doi.org/10.1183/09031936.06.00114905
Rose. RM; Fuglestad. JM; Skornik. WA; Hammer. SM; Wolfthal. SF; Beck. BD; Brain. JD. (1988). The
pathophysiology of enhanced susceptibility to murine cytomegalovirus respiratory infection during
short-term exposure to 5 ppm nitrogen dioxide. Am Rev Respir Dis 137: 912-917.
http://dx.doi.0rg/10.1164/ajrccm/137.4.912
Rose. RM; Pinkston. P; Skornik. WA. (1989b). Altered susceptibility to viral respiratory infection during
short-term exposure to nitrogen dioxide. (HEI Research Report no. 24). Cambridge, MA: Health Effects
Institute.
Rubinstein. I; Bigbv. BG: Reiss. TF; Boushev. HA. Jr. (1990). Short-term exposure to 0.3 ppm nitrogen
dioxide does not potentiate airway responsiveness to sulfur dioxide in asthmatic subjects. Am Rev
Respir Dis 141: 381-385. http://dx.doi.Org/10.1164/airccm/141.2.381
Rudez. G: Janssen. NA; Kilinc. E; Leebeek. FW: Gerlofs-Nijland. ME; Spronk. HM; ten Gate. H; Cassee.
FR; de Maat. MP (2009). Effects of ambient air pollution on hemostasis and inflammation. Environ
Health Perspect 117: 995-1001. http://dx.doi.org/10.1289/ehp.0800437
Salam. MT; Byun. HM; Lurmann. F; Breton. CV: Wang. X; Eckel. SP; Gilliland. FD. (2012). Genetic and
epigenetic variations in inducible nitric oxide synthase promoter, paniculate pollution, and exhaled
nitric oxide levels in children. J Allergy Clin Immunol 129: 232-239.e237.
http://dx.doi.0rg/10.1016/i.iaci.2011.09.037
November 2013 4-306 DRAFT: Do Not Cite or Quote
-------
Salonen. RO; Pennanen. AS: Vahteristo. M; Korkeila. P; Aim. S: Randell. JT. (2008). Health risk
assessment of indoor air pollution in Finnish ice arenas. Environ Int 34: 51-57.
http://dx.doi.0rg/10.1016/i.envint.2007.06.012
Samoli. E; Aga. E; Touloumi. G: Nisiotis. K; Forsberg. B; Lefranc. A; Pekkanen. J; Wojtyniak. B;
Schindler. C: Niciu. E; Brunstein. R; Dodic Fikfak. M; Schwartz. J; Katsouvanni. K. (2006). Short-term
effects of nitrogen dioxide on mortality: An analysis within the APHEA project. Eur Respir J 27: 1129-
1137. http://dx.doi.org/10.1183/09031936.06.00143905
Samoli. E; Nastos. PT; Paliatsos. AG: Katsouvanni. K; Priftis. KN. (2011). Acute effects of air pollution on
pediatric asthma exacerbation: Evidence of association and effect modification. Environ Res 111: 418-
424. http://dx.doi.0rg/10.1016/j.envres.2011.01.014
Sarnat. SE; Ravsoni. AU; Li. WW: Holguin. F; Johnson. BA; Flores Luevano. S: Garcia. JH; Sarnat. JA.
(2012). Air pollution and acute respiratory response in a panel of asthmatic children along the U.S.-
Mexico border. Environ Health Perspect 120: 437444. http://dx.doi.org/10.1289/ehp.1003169
Scaife. A; Barclay. J: Hillis. GS: Srinivasan. J: Macdonald. DW: Ross. JA; Ayres. JG. (2012). Lack of
effect of nitrogen dioxide exposure on heart rate variability in patients with stable coronary heart disease
and impaired left ventricular systolic function. Occup Environ Med 69: 587-591.
http://dx.doi.org/10.1136/oemed-2011-100126
Scarlett. JF; Abbott. KJ; Peacock. JL; Strachan. DP; Anderson. HR. (1996). Acute effects of summer air
pollution on respiratory function in primary school children in southern England. Thorax 51: 1109-1114.
Schildcrout. JS: Sheppard. L; Lumlev. T; Slaughter. JC: Koenig. JQ; Shapiro. GG. (2006). Ambient air
pollution and asthma exacerbations in children: An eight-city analysis. Am J Epidemiol 164: 505-517.
http://dx.doi.org/10.1093/aie/kwj225
Schindler. C: Kiinzli. N; Bongard. JP; Leuenberger. P; Karrer. W: Rapp. R; Monn. C: Ackermann-Liebrich.
LL (2001). Short-term variation in air pollution and in average lung function among never-smokers. The
Swiss Study on Air Pollution and Lung Diseases in Adults (SAPALDIA). Am J Respir Crit Care Med
163: 356-361. http://dx.doi.Org/10.1164/airccm.163.2.9911116
Schlesinger. RB. (1987a). Effects of intermittent inhalation exposures to mixed atmospheres of NO2 and
H2SO4 on rabbit alveolar macrophages. J Toxicol Environ Health 22: 301-312.
http://dx.doi.org/10.1080/15287398709531073
Schlesinger. RB. (1987b). Intermittent inhalation of nitrogen dioxide: Effects on rabbit alveolar
macrophages. J Toxicol Environ Health 21: 127-139. http://dx.doi.org/10.1080/15287398709531007
Schlesinger. RB: Driscoll. KE; Gunnison. AF; Zelikoff. JT. (1990). Pulmonary arachidonic acid metabolism
following acute exposures to ozone and nitrogen dioxide. J Toxicol Environ Health 31: 275-290.
http://dx.doi.org/10.1080/15287399009531456
Schwartz. J: Dockerv. DW: Neas. LM; Wvpij. D; Ware. JH; Spengler. JD; Koutrakis. P: Speizer. FE; Ferris.
BG. Jr. (1994). Acute effects of summer air pollution on respiratory symptom reporting in children. Am
J Respir Crit Care Med 150: 1234-1242. http://dx.doi.Org/10.1164/airccm.150.5.7952546
Segala. C: Fauroux. B; Just. J: Pascual. L; Grimfeld. A; Neukirch. F. (1998). Short-term effect of winter air
pollution on respiratory health of asthmatic children in Paris. Eur Respir J 11: 677-685.
Segala. C: Poizeau. D; Mesbah. M; Willems. S: Maidenberg. M. (2008). Winter air pollution and infant
bronchiolitis in Paris. Environ Res 106: 96-100. http://dx.doi.Org/10.1016/i.envres.2007.05.003
Selgrade. MK: Mole. ML: Miller. FJ: Hatch. GE: Gardner. DE: Hu. PC. (1981). Effect of NO2 inhalation
and vitamin C deficiency on protein and lipid accumulation in the lung. Environ Res 26: 422-437.
Sevanian. A; Hacker. AD; Elsaved. N. (1982b). Influence of vitamin E and nitrogen dioxide on lipid
peroxidation in rat lung and liver microsomes. Lipids 17: 269-277.
Sherwin. RP; Carlson. DA. (1973). Protein content of lung lavage fluid of guinea pigs exposed to 0.4 ppm
nitrogen dioxide. Arch Environ Health 27: 90-93.
November 2013 4-307 DRAFT: Do Not Cite or Quote
-------
Sherwin. RP; Dibble. J: Weiner. J. (1972). Alveolar wall cells of the guinea pig: Increase in response to 2
ppm nitrogen dioxide. Arch Environ Health 24: 43-47.
Sherwood. RL; Lippert. WE; Goldstein. E; Tarkington. B. (1981). Effect of ferrous sulfate aerosols and
nitrogen dioxide on murine pulmonary defense. Arch Environ Occup Health 36: 130-135.
Shin. HH; Stieb. D; Burnett. R; Takahara. G: Jessiman. B. (2012). Tracking national and regional spatial-
temporal mortality risk associated with NO(2) concentrations in Canada: A bayesian hierarchical two-
Level model. Risk Anal 32: 513-530. http://dx.doi.Org/10.llll/i.1539-6924.2011.01684.x
Silbaugh. SA; Mauderly. JL; Macken. CA. (1981). Effects of sulfuric acid and nitrogen dioxide on airway
responsiveness of the guinea pig. J Toxicol Environ Health 8: 31-45.
http://dx.doi.org/10.1080/15287398109530048
Silkoff. PE; Zhang. L; Dutton. S: Langmack. EL; Vedal. S: Murphy. J: Make. B. (2005). Winter air
pollution and disease parameters in advanced chronic obstructive pulmonary disease panels residing in
Denver, Colorado. J Allergy Clin Immunol 115: 337-344. http://dx.doi.Org/10.1016/i.jaci.2004.ll.035
Silverman. RA; Ito. K; Freese. J: Kaufman. BJ; De Claro. D; Braun. J: Prezant. DJ. (2010). Association of
ambient fine particles with out-of-hospital cardiac arrests in New York City. Am J Epidemiol 172: 917-
923. http://dx.doi.org/10.1093/aie/kwq217
Simpson. R; Williams. G: Petroeschevsky. A; Best. T; Morgan. G: Denison. L; Hinwood. A; Neville. G.
(2005a). The short-term effects of air pollution on hospital admissions in four Australian cities. AustN
Z J Public Health 29: 213-221.
Simpson. R; Williams. G: Petroeschevsky. A; Best. T; Morgan. G: Denison. L; Hinwood. A; Neville. G:
Neller. A. (2005b). The short-term effects of air pollution on daily mortality in four Australian cities.
Aust N Z J Public Health 29: 205-212. http://dx.doi.Org/10.llll/j.1467-842X.2005.tb00758.x
Sinclair. AH; Edgerton. ES: Wyzga. R: Tolsma. D. (2010). A two-time-period comparison of the effects of
ambient air pollution on outpatient visits for acute respiratory illnesses. J Air Waste Manag Assoc 60:
163-175. http://dx.doi.0rg/10.3155/1047-3289.60.2.163
Sinclair. AH; Tolsma. D. (2004). Associations and lags between air pollution and acute respiratory visits in
an ambulatory care setting: 25-month results from the Aerosol Research and Inhalation Epidemiological
Study. J Air Waste Manag Assoc 54: 1212-1218. http://dx.doi.org/10.1080/10473289.2004.10470979
Solomon. C: Christian. PL; Chen. LL; Welch. BS: Kleinman. MT; Dunham. E; Erie. DJ: Balmes. JR.
(2000). Effect of serial-day exposure to nitrogen dioxide on airway and blood leukocytes and
lymphocyte subsets. Eur Respir J 15: 922-928.
Son. JY; Bell. ML; Lee. JT. (2010). Individual exposure to air pollution and lung function in Korea: Spatial
analysis using multiple exposure approaches. Environ Res 110: 739-749.
http://dx.doi.0rg/10.1016/i.envres.2010.08.003
Son. JY; Lee. JT; Park. YH; Bell. ML. (2013). Short-term effects of air pollution on hospital admissions in
Korea. Epidemiology 24: 545-554. http://dx.doi.org/10.1097/EDE.Ob013e3182953244
Spira-Cohen. A; Chen. LC: Kendall. M; Lall. R: Thurston. GD. (2011). Personal exposures to traffic-related
air pollution and acute respiratory health among Bronx schoolchildren with asthma. Environ Health
Perspect 119: 559-565. http://dx.doi.org/10.1289/ehp.1002653
Steenhof. M; Mudwav. IS: Gosens. I; Hoek. G: Godri. KJ; Kelly. FJ; Harrison. RM; Pieters. RH; Cassee.
FR; Lebret. E; Brunekreef. BA; Strak. M; Janssen. NA. (2013). Acute nasal pro-inflammatory response
to air pollution depends on characteristics other than particle mass concentration or oxidative potential:
the RAPTES project. Occup Environ Med 70: 341-348. http://dx.doi.org/10.1136/oemed-2012-100993
Steerenberg. PA; Bischoff. EWM. A; de Klerk. A; Verlaan. APJ; Jongbloets. LMN; Van Loveren. H;
Opperhuizen. A; Zomer. G: Heisterkamp. SH; Hadv. M; Spieksma. FTM; Fischer. PH; Dormans. JAM.
A; van Amsterdam. JGC. (2003). Acute effect of air pollution on respiratory complaints, exhaled NO
and biomarkers in nasal lavages of allergic children during the pollen season. Int Arch Allergy Immunol
131: 127-137. http://dx.doi.org/10.1159/000070928
November 2013 4-308 DRAFT: Do Not Cite or Quote
-------
Steerenberg. PA; Nierkens. S: Fischer. PH; Van Loveren. H; Opperhuizen. A; Vos. JG: van Amsterdam. JG.
(2001). Traffic -related air pollution affects peak expiratory flow, exhaled nitric oxide, and inflammatory
nasal markers. Arch Environ Health 56: 167-174. http://dx.doi.org/10.1080/00039890109604069
Steinvil. A; Fireman. E; Kordova-Biezuner. L; Cohen. M; Shapira. I; Berliner. S: Rogowski. O. (2009).
Environmental air pollution has decremental effects on pulmonary function test parameters up to one
week after exposure. Am J Med Sci 338: 273-279. http://dx.doi.org/10.1097/MAJ.Ob013e3181adb3ed
Steinvil. A; Kordova-Biezuner. L; Shapira. I; Berliner. S: Rogowski. O. (2007). Short-term exposure to air
pollution and inflammation-sensitive biomarkers. Environ Res 106: 51-61.
Steinvil. A; Kordova-Biezuner. L; Shapira. I; Berliner. S: Rogowski. O. (2008). Short-term exposure to air
pollution and inflammation-sensitive biomarkers. Environ Res 106: 51-61.
http://dx.doi.0rg/10.1016/i.envres.2007.08.006
Stern. G: Latzin. P; Roosli. M; Fuchs. O: Proietti. E; Kuehni. C: Frev. U. (2013). A prospective study of the
impact of air pollution on respiratory symptoms and infections in infants. Am J Respir Crit Care Med
187: 1341-1348. http://dx.doi.org/10.1164/rccm.201211-2008OC
Stieb. DM; Burnett. RT; Smith-Doiron. M; Brion. O: Shin. HH; Economou. V. (2008). A new
multipollutant, no-threshold air quality health index based on short-term associations observed in daily
time-series analyses. J Air Waste Manag Assoc 58: 435-450.
Stieb. DM; Judek. S: Burnett. RT. (2002). Meta-analysis of time-series studies of air pollution and
mortality: Effects of gases and particles and the influence of cause of death, age, and season. J Air
Waste Manag Assoc 52: 470-484.
Stieb. DM; Judek. S: Burnett. RT. (2003). Meta-analysis of time-series studies of air pollution and
mortality: Update in relation to the use of generalized additive models. J Air Waste Manag Assoc 53:
258-261.
Stieb. DM; Szyszkowicz. M; Rowe. BH; Leech. JA. (2009). Air pollution and emergency department visits
for cardiac and respiratory conditions: A multi-city time-series analysis. Environ Health Global Access
Sci Source 8: 25. http://dx.doi.org/10.1186/1476-069X-8-25
Strak. M; Janssen. NA; Godri. KJ; Gosens. I; Mudwav. IS: Cassee. FR; Lebret. E; Kelly. FJ; Harrison. RM;
Brunekreef. B; Steenhof. M; Hoek. G. (2012). Respiratory health effects of airborne paniculate matter:
The role of particle size, composition and oxidative potential - The RAPTES project. Environ Health
Perspect 120: 1183-1189. http://dx.doi.org/10.1289/ehp.1104389
Strand. V: Rak. S: Svartengren. M; Bylin. G. (1997). Nitrogen dioxide exposure enhances asthmatic
reaction to inhaled allergen in subjects with asthma. Am J Respir Crit Care Med 155: 881-887.
http://dx.doi.0rg/10.1164/ajrccm.155.3.9117021
Strand. V: Salomonsson. P; Lundahl. J: Bylin. G. (1996). Immediate and delayed effects of nitrogen
dioxide exposure at an ambient level on bronchial responsiveness to histamine in subjects with asthma.
Eur Respir J 9: 733-740. http://dx.doi.org/10.1183/09031936.96.09040733
Strand. V: Svartengren. M; Rak. S: Barck. C: Bylin. G. (1998). Repeated exposure to an ambient level of
NO2 enhances asthmatic response to a nonsymptomatic allergen dose. Eur Respir J 12: 6-12.
http://dx.doi.org/10.1183/09031936.98.12010006
Strickland. MJ; Darrow. LA; Klein. M; Flanders. WD; Sarnat JA; Waller. LA; Sarnat. SE; Mulholland. JA;
Tolbert. PE. (2010). Short-term associations between ambient air pollutants and pediatric asthma
emergency department visits. Am J Respir Crit Care Med 182: 307-316.
http://dx.doi.org/10.1164/rccm.200908-1201OC
Strickland. MJ: Darrow. LA; Mulholland. JA: Klein. M; Flanders. WD: Winquist A; Tolbert. PE. (2011).
Implications of different approaches for characterizing ambient air pollutant concentrations within the
urban airshed for time-series studies and health benefits analyses. Environ Health Global Access Sci
Source 10: 36. http://dx.doi.org/10.1186/1476-069X-10-36
November 2013 4-309 DRAFT: Do Not Cite or Quote
-------
Suh. HH; Zanobetti. A. (2010a). Exposure error masks the relationship between traffic-related air pollution
and heart rate variability [Erratum]. J Occup Environ Med 52: 1138.
http://dx.doi.org/10.1097/JOM.Ob013e3181fd2632
Suzuki. T; Ikeda. S: Kanoh. T; Mizoguchi. I. (1986). Decreased phagocytosis and superoxide anion
production in alveolar macrophages of rats exposed to nitrogen dioxide. Arch Environ Contam Toxicol
15: 733-739. http://dx.doi.org/10.1007/BF01054920
Szyszkowicz. M. (2007b). Air pollution and emergency department visits for ischemic heart disease in
Montreal, Canada. Int J Occup Med Environ Health 20: 167-173. http://dx.doi.org/10.2478/vl0001-007-
0015-3
Szyszkowicz. M. (2008b). Ambient air pollution and daily emergency department visits for ischemic stroke
in Edmonton, Canada. Int J Occup Med Environ Health 21: 295-300. http://dx.doi.org/10.2478/vl0001-
008-0029-5
Szyszkowicz. M. (2009). Air pollution and ED visits for chest pain. Am J Emerg Med 27: 165-168.
http://dx.doi.0rg/10.1016/i.aiem.2008.01.010
Szyszkowicz. M; Rowe. BH; Brook. RD. (2012). Even low levels of ambient air pollutants are associated
with increased emergency department visits for hypertension. Can J Cardiol 28: 360-366.
http://dx.doi.0rg/10.1016/i.cjca.2011.06.011
Tashkin. DP; Simmons. MS: Chang. P; Liu. H; Coulson. AH. (1993). Effects of smoked substance abuse on
nonspecific airway hyperresponsiveness. Am Rev Respir Dis 147: 97-103.
http://dx.doi.0rg/10.1164/ajrccm/147.l.97
Task Force of the European Society of Cardiology and the North American Society of Pacing and
Electrophysiologv. (1996). Heart rate variability: Standards of measurement, physiological
interpretation and clinical use. Circulation 93: 1043-1065. http://dx.doi.Org/10.1161/01.CIR.93.5.1043
Thach. TO; Wong. CM; Chan. KP; Chau. YK; Neil Thomas. G: Ou. CO: Yang. L; Peiris. JSM; Lam. TH;
Hedlev. AJ. (2010). Air pollutants and health outcomes: Assessment of confounding by influenza.
Atmos Environ 44: 1437-1442. http://dx.doi.Org/10.1016/i.atmosenv.2010.01.036
Thaller. El; Petronella. SA; Hochman. D; Howard. S: Chhikara. RS: Brooks. EG. (2008). Moderate
increases in ambient PM2.5 and ozone are associated with lung function decreases in beach lifeguards. J
Occup Environ Med 50: 202-211. http://dx.doi.org/10.1097/JOM.Ob013e31816386b4
Thompson. AM; Zanobetti. A; Silverman. F; Schwartz. J: Coull. B; Urch. B; Speck. M; Brook. JR; Manno.
M; Gold. DR. (2010). Baseline repeated measures from controlled human exposure studies:
Associations between ambient air pollution exposure and the systemic inflammatory biomarkers IL-6
and fibrinogen. Environ Health Perspect 118: 120-124. http://dx.doi.org/10.1289/ehp.0900550
Timonen. KL; Hoek. G: Heinrich. J: Bernard. A; Brunekreef. B; De Hartog. J: Hameri. K; Ibald-Mulli. A;
Mirme. A; Peters. A; Tiittanen. P; Kreyling. WG: Pekkanen. J. (2004). Daily variation in fine and
ultrafine paniculate air pollution and urinary concentrations of lung Clara cell protein CC16. Occup
Environ Med 61: 908-914. http://dx.doi.org/10.1136/oem.2004.012849
Timonen. KL; Pekkanen. J. (1997). Air pollution and respiratory health among children with asthmatic or
cough symptoms. Am J Respir Crit Care Med 156: 546-552.
http://dx.doi.0rg/10.1164/ajrccm.156.2.9608044
Timonen. KL; Vanninen. E; De Hartog. J: Ibald-Mulli. A; Brunekreef. B; Gold. PR; Henrich. J: Hoek. G:
Lanki. T; Peters. A; Tarkiainen. T; Tiittanen. P; Kreyling. W: Pekkanen. J. (2006). Effects of ultrafine
and fine paniculate and gaseous air pollution on cardiac autonomic control in subjects with coronary
artery disease: The ULTRA study. J Expo Sci Environ Epidemiol 16: 332-341.
http://dx.doi.org/10.1038/sj.jea.7500460
Tolbert. PE; Klein. M; Metzger. KB; Peel. J: Flanders. WD; Todd. K; Mulholland. JA; Ryan. PB; Frumkin.
£L (2000). Interim results of the study of particulates and health in Atlanta (SOPHIA). J Expo Anal
Environ Epidemiol 10: 446-460.
November 2013 4-310 DRAFT: Do Not Cite or Quote
-------
Tolbert. PE; Klein. M; Peel. JL; Sarnat. SE; Sarnat. JA. (2007). Multipollutant modeling issues in a study of
ambient air quality and emergency department visits in Atlanta. J Expo Sci Environ Epidemiol 17: S29-
S35. http://dx.doi.org/10.1038/si.ies.7500625
Tsai. DH; Wang. JL; Chuang. KJ; Chan. CC. (2010). Traffic-related air pollution and cardiovascular
mortality in central Taiwan. Sci Total Environ 408: 1818-1823.
http://dx.doi.0rg/10.1016/i.scitotenv.2010.01.044
Tsai. SS: Chiu. HF; Wu. TN; Yang. CY. (2009). Air pollution and emergency room visits for cardiac
arrhythmia in a subtropical city: Taipei, Taiwan. Inhal Toxicol 21: 1113-1118.
http://dx.doi.org/10.3109/08958370902758939
Tsai. SS: Goggins. WB; Chiu. HF; Yang. CY. (2003). Evidence for an association between air pollution and
daily stroke admissions in Kaohsiung, Taiwan. Stroke 34: 2612-2616.
http://dx.doi.org/10.1161/01.STR.0000095564.33543.64
Tsuji. H; Larson. MG: Venditti. FJ. Jr; Manders. ES: Evans. JC: Feldman. CL; Lew. D. (1996). Impact of
reduced heart rate variability on risk for cardiac events The Framingham Heart Study. Circulation 94:
2850-2855.
Tsuji. H; Venditti. FJ: Manders. ES: Evans. JC: Larson. MG: Feldman. CL; Lew. D. (1994). Reduced heart
rate variability and mortality risk in an elderly cohort: The Framingham Heart Study. Circulation 90:
878-883.
Tunnicliffe. WS: Surge. PS: Ayres. JG. (1994). Effect of domestic concentrations of nitrogen dioxide on
airway responses to inhaled allergen in asthmatic patients. Lancet 344: 1733-1736.
http://dx.doi.org/10.1016/s0140-6736(94)92886-x
Turin. TC: Kita. Y; Rumana. N; Nakamura. Y; Ueda. K; Takashima. N; Sugihara. H; Morita. Y; Ichikawa.
M; Hirose. K; Nitta. H; Okavama. A; Miura. K; Ueshima. H. (2012). Short-term exposure to air
pollution and incidence of stroke and acute myocardial infarction in a Japanese population.
Neuroepidemiology 38: 84-92. http://dx.doi.org/10.1159/000335654
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park,NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for particulate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for ozone and
related photochemical oxidants (final report) [EPA Report]. (EPA/600/R-10/076F). Washington, DC.
http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=247492
U.S. EPA (U.S. Environmental Protection Agency). (2013c). Results of single-pollutant and copollutants
models of short-term exposure to NO2 or NOX with and without O3 and hospital admissions for
cardiovascular disease. Figure S4-2 and Figure S4-3 [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013d). Supplemental figure S4-1. Associations of
ambient NO2 with respiratory outcomes adjusted for various gaseous copollutants [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (20131). Supplemental table S4-2. Epidemiologic
studies of short-term exposure to oxides of nitrogen and mortality not included in the ISA [EPA Report].
Vagaggini. B; Paggiaro. PL; Giannini. D; Franco. AD; Cianchetti. S: Carnevali. S: Taccola. M; Bacci. E;
Bancalari. L; Dente. FL; Giuntini. C. (1996). Effect of short-term NO2 exposure on induced sputum in
normal, asthmatic and COPD subjects. Eur Respir J 9: 1852-1857.
November 2013 4-311 DRAFT: Do Not Cite or Quote
-------
van der Zee. SC: Hoek. G: Boezen. HM: Schouten. JP: van Wijnen. JH: Brunekreef. B. (1999). Acute effects of
urban air pollution on respiratory health of children with and without chronic respiratory symptoms. Occup
Environ Med 56: 802-813. http://dx.doi.org/10.1136/oem.56.12.802
van der Zee. SC: Hoek. G: Boezen. MH: Schouten. JP: Van Wijnen. JH: Brunekreef. B. (2000). Acute effects of
air pollution on respiratory health of 50-70 yr old adults. Eur Respir J 15: 700-709.
Van Schavck. CP; Dompeling. E; Van Herwaarden. CL; Wever. AM; Van Weel. C. (1991). Interacting
effects of atopy and bronchial hyperresponsiveness on the annual decline in lung function and the
exacerbation rate in asthma. Am Rev Respir Dis 144: 1297-1301.
http://dx.doi.0rg/10.1164/ajrccm/144.6.1297
Vencloviene. J: Grazuleviciene. R; Babarskiene. R; Dedele. A; Grazulevicius. T. (2011). Short-term
nitrogen dioxide exposure and geomagnetic activity interaction: contribution to emergency
hospitalization for acute coronary syndrome. Int J Environ Health Res 21: 149-160.
http://dx.doi.org/10.1080/09603123.2010.515671
Verbanck. S: Schuermans. D; Noppen. M; Vincken. W: Paiva. M. (2001). Methacholine versus histamine:
paradoxical response of spirometry and ventilation distribution. J Appl Physiol 91: 2587-2594.
Vidale. S: Bonanomi. A; Guidotti. M; Arnaboldi. M; Sterzi. R. (2010). Air pollution positively correlates
with daily stroke admission and in hospital mortality: a study in the urban area of Como, Italy. Neurol
Sci 31: 179-182. http://dx.doi.org/10.1007/sl0072-009-0206-8
Villeneuve. PJ: Chen. L; Rowe. BH; Coates. F. (2007). Outdoor air pollution and emergency department
visits for asthma among children and adults: A case-crossover study in northern Alberta, Canada.
Environ Health Global Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40
Villeneuve. PJ: Chen. L; Stieb. D; Rowe. BH. (2006a). Associations between outdoor air pollution and
emergency department visits for stroke in Edmonton, Canada. Eur J Epidemiol 21: 689-700.
http://dx.doi.org/10.1007/sl0654-006-9050-9
Villeneuve. PJ: Doiron. MS: Stieb. D: Dales. R: Burnett. RT; Dugandzic. R. (2006b). Is outdoor air
pollution associated with physician visits for allergic rhinitis among the elderly in Toronto, Canada?
Allergy 61: 750-758. http://dx.doi.Org/10.llll/i.1398-9995.2006.01070.x
Vollmuth. TA; Driscoll. KE; Schlesinger. RB. (1986). Changes in early alveolar particle clearance due to
single and repeated nitrogen dioxide exposures in the rabbit. J Toxicol Environ Health 19: 255-266.
http://dx.doi.org/10.1080/15287398609530925
Von Klot. S: Peters. A; Aalto. P; Bellander. T; Berglind. N; D'Ippoliti. D: Elosua. R: Hormann. A; Kulmala.
M; Lanki. T; Lowel. H; Pekkanen. J: Picciotto. S: Sunyer. J: Forastiere. F; Group. HEoPoSSHS. (2005).
Ambient air pollution is associated with increased risk of hospital cardiac readmissions of myocardial
infarction survivors in five European cities. Circulation 112: 3073-3079.
http://dx.doi.org/10.1161/CIRCULATIONAHA.105.548743
von Klot. S: Wolke. G: Tuch. T; Heinrich. J: Dockery. DW: Schwartz. J: Kreyling. WG: Wichmann. HE;
Peters. A. (2002). Increased asthma medication use in association with ambient fine and ultrafine
particles. Eur Respir J 20: 691-702. http://dx.doi.org/10.1183/09031936.02.01402001
Wang. JH; Devalia. JL; Puddle. JM; Hamilton. SA; Davies. RJ. (1995 a). Effect of six-hour exposure to
nitrogen dioxide on early-phase nasal response to allergen challenge in patients with a history of
seasonal allergic rhinitis. J Allergy Clin Immunol 96: 669-676.
Wang. JH: Devalia. JL; Rusznak. C: Bagnall. A; Sapsford. RJ: Davies. RJ. (1999). Effect of fluticasone
propionate aqueous nasal spray on allergen-induced inflammatory changes in the nasal airways of
allergic rhinitics following exposure to nitrogen dioxide. Clin Exp Allergy 29: 234-240.
http://dx.doi.0rg/10.1046/i.1365-2222.1999.00440.x
Wang. JH; Puddle. J: Devalia. JL: Davies. RJ. (1995b). Nitrogen dioxide increases eosinophil activation in
the early-phase response to nasal allergen provocation. Int Arch Allergy Immunol 107: 103-105.
November 2013 4-312 DRAFT: Do Not Cite or Quote
-------
Wanner. A; Brodnan. JM; Perez. J; Henke. KG: Kim. CS. (1985). Variability of airway responsiveness to
histamine aerosol in normal subjects. Role of deposition. Am Rev Respir Dis 131: 3-7.
http://dx.doi.org/10.1002/ppul.1950010433
Ward. DJ: Miller. MR; Walters. S: Harrison. RM; Ayres. JG. (2000). Impact of correcting peak flow for
nonlinear errors on air pollutant effect estimates from a panel study. Eur Respir J 15: 137-140.
Ward. DJ: Roberts. KT; Jones. N; Harrison. RM; Ayres. JG: Hussain. S: Walters. S. (2002). Effects of daily
variation in outdoor particulates and ambient acid species in normal and asthmatic children. Thorax 57:
489-502. http://dx.doi.0rg/10.1136/thorax.57.6.489
Weichenthal. S: Kulka. R; Dubeau. A; Martin. C: Wang. D; Dales. R. (2011). Traffic-related air pollution
and acute changes in heart rate variability and respiratory function in urban cyclists. Environ Health
Perspect 119: 1373-1378. http://dx.doi.org/10.1289/ehp. 1003321
Weinmavr. G: Romeo. E; De Sario. M; SK. W: Francesco Forastiere. F. (2010). Short-term effects of PM10
and NO2 on respiratory health among children with asthma or asthma-like symptoms: a systematic
review and meta-analysis [Review]. Environ Health Perspect 118: 449-457.
http://dx.doi.org/10.1289/ehp.0900844
Wellenius. GA; Burger. MR; Coull. BA; Schwartz. J: Suh. HH; Koutrakis. P; Schlaug. G: Gold. PR;
Mittleman. MA. (2012). Ambient air pollution and the risk of acute ischemic stroke. Arch Intern Med
172: 229-234. http://dx.doi.org/10.1001/archinternmed.2011.732
Wellenius. GA; Schwartz. J: Mittleman. MA. (2005). Air pollution and hospital admissions for ischemic
and hemorrhagic stroke among medicare beneficiaries. Stroke 36: 2549-2553.
http://dx.doi.org/10.1161/01.STR.0000189687.78760.47
Wellenius. GA: Yeh. GY: Coull. BA: Suh. HH: Phillips. RS: Mittleman. MA. (2007). Effects of ambient air
pollution on functional status in patients with chronic congestive heart failure: A repeated-measures
study. Environ Health 6: 26. http://dx.doi.org/10.1186/1476-069X-6-26
Wichmann. J: Ketzel. M; Ellermann. T; Loft. S. (2012). Apparent temperature and acute myocardial
infarction hospital admissions in Copenhagen, Denmark: A case-crossover study. Environ Health 11: 19.
http://dx.doi.org/10.1186/1476-069X-ll-19
Williams. R: Brook. R; Bard. R: Conner. T; Shin. H; Burnett. R. (2012a). Impact of personal and ambient-
level exposures to nitrogen dioxide and paniculate matter on cardiovascular function. Int J Environ
Health Res 22: 71-91. http://dx.doi.org/10.1080/09603123.2011.588437
Williams. R: Jones. P; Croghan. C: Thornburg. J: Rodes. C. (2012b). The influence of human and
environmental exposure factors on personal NO(2) exposures. J Expo Sci Environ Epidemiol 22: 109-
115. http://dx.doi.org/10.1038/ies.2011.20
Witten. A; Solomon. C: Abbritti. E; Arjomandi. M; Zhai. W: Kleinman. M; Balmes. J. (2005). Effects of
nitrogen dioxide on allergic airway responses in subjects with asthma. J Occup Environ Med 47: 1250-
1259.
Wiwatanadate. P: Liwsrisakun. C. (2011). Acute effects of air pollution on peak expiratory flow rates and
symptoms among asthmatic patients in Chiang Mai, Thailand. Int J Hyg Environ Health 214: 251-257.
http://dx.doi.0rg/10.1016/i.iiheh.2011.03.003
Wiwatanadate. P: Trakultivakorn. M. (2010). Air pollution-related peak expiratory flow rates among
asthmatic children in Chiang Mai, Thailand. Inhal Toxicol 22: 301-308.
http://dx.doi.org/10.3109/08958370903300327
Wong. CM; Vichit-Vadakan. N; Kan. H; Qian. Z. (2008b). Public Health and Air Pollution in Asia (PAPA):
Amulticity study of short-term effects of air pollution on mortality. Environ Health Perspect 116: 1195-
1202. http://dx.doi.org/10.1289/ehp. 11257
November 2013 4-313 DRAFT: Do Not Cite or Quote
-------
Wong. CM; Vichit-Vadakan. N; Vajanapoom. N; Ostro. B; Thach. TQ; Chau. PY; Chan. EK; Chung. RY;
Ou. CO: Yang. L; Peiris. JS: Thomas. GN; Lam. TH; Wong. TW: Hedlev. AJ; Kan. H; Chen. B; Zhao.
N; London. SJ: Song. G: Chen. G: Zhang. Y; Jiang. L; Qian. Z; He. Q: Lin. HM; Kong. L; Zhou. D;
Liang. S: Zhu. Z; Liao. D; Liu. W: Bentlev. CM; Dan. J; Wang. B; Yang. N; Xu. S: Gong. J; Wei. H;
Sun. H; Qin. Z. (2010). Part 5. Public health and air pollution in Asia (PAPA): A combined analysis of
four studies of air pollution and mortality. In Public Health and Air Pollution in Asia (PAPA):
Coordinated Studies of Short-Term Exposure to Air Pollution and Daily Mortality in Four Cities (pp.
377-418). Boston, MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=348
Wong. CM; Yang. L; Thach. TQ: Chau. PY; Chan. KP; Thomas. GN; Lam. TH; Wong. TW: Hedlev. AJ:
Peiris. JS. (2009). Modification by influenza on health effects of air pollution in Hong Kong. Environ
Health Perspect 117: 248-253. http://dx.doi.org/10.1289/ehp.11605
Wong. TW: Lau. TS: Yu. TS: Neller. A; Wong. SL; Tarn. W: Pang. SW. (1999). Air pollution and hospital
admissions for respiratory and cardiovascular diseases in Hong Kong. Occup Environ Med 56: 679-683.
Xuan. W: Peat. JK; Toelle. BG: Marks. GB; Berry. G: Woolcock. AJ. (2000). Lung function growth and its
relation to airway hyperresponsiveness and recent wheeze: Results from a longitudinal population study.
Am J Respir Crit Care Med 161: 1820-1824. http://dx.doi.Org/10.1164/ajrccm.161.6.9809118
Yamazaki. S: Shima. M; Ando. M; Nitta. H; Watanabe. H; Nishimuta. T. (2011). Effect of hourly
concentration of paniculate matter on peak expiratory flow in hospitalized children: A panel study.
Environ Health 10: 15. http://dx.doi.org/10.1186/1476-069X-10-15
Yang. CY. (2008). Air pollution and hospital admissions for congestive heart failure in a subtropical city:
Taipei, Taiwan. J Toxicol Environ Health A 71: 1085-1090.
http://dx.doi.org/10.1080/15287390802114428
Yang. CY; Chen. YS: Yang. CH; Ho. SC. (2004). Relationship between ambient air pollution and hospital
admissions for cardiovascular diseases in Kaohsiung, Taiwan. J Toxicol Environ Health A 67: 483-493.
http://dx.doi.org/10.1080/15287390490276502
Yang. Q; Chen. Y; Krewski. D; Burnett. RT; Shi. Y; McGrail. KM. (2005). Effect of short-term exposure to
low levels of gaseous pollutants on chronic obstructive pulmonary disease hospitalizations. Environ Res
99: 99-105. http://dx.doi.Org/10.1016/i.envres.2004.09.014
Yang. Q; Chen. Y; Shi. Y; Burnett RT; Mcgrail. KM; Krewski. D. (2003). Association between ozone and
respiratory admissions among children and the elderly in Vancouver, Canada. Inhal Toxicol 15: 1297-
1308. http://dx.doi.org/10.1080/08958370390241768
Zanobetti. A; Gold. PR; Stone. PH; Suh. HH; Schwartz. J: Coull. BA; Speizer. FE. (2010). Reduction in
heart rate variability with traffic and air pollution in patients with coronary artery disease. Environ
Health Perspect 118: 324-330. http://dx.doi.org/10.1289/ehp.0901003
Zemek. R; Szyszkowicz. M; Rowe. BH. (2010). Air pollution and emergency department visits for otitis
media: A case-crossover study in Edmonton, Canada. Environ Health Perspect 118: 1631-1636.
http://dx.doi.org/10.1289/ehp.0901675
Zora. JE; Sarnat. SE; Ravsoni. AU; Johnson. BA; Li. WW: Greenwald. R; Holguin. F; Stock. TH; Sarnat.
JA. (2013). Associations between urban air pollution and pediatric asthma control in El Paso, Texas. Sci
Total Environ 448: 56-65. http://dx.doi.Org/10.1016/i.scitotenv.2012.ll.067
November 2013 4-314 DRAFT: Do Not Cite or Quote
-------
CHAPTER 5 INTEGRATED HEALTH EFFECTS
OF LONG-TERM EXPOSURE TO OXIDES OF
NITROGEN
5.1 Introduction
1 This chapter summarizes, integrates, and evaluates the evidence for a broad spectrum of
2 health effects associated with long-term exposure (i.e., 1 month to years) to oxides of
3 nitrogen. This chapter comprises evaluations of the epidemiologic and toxicological
4 evidence for the effects of long-term exposure to oxides of nitrogen on health outcomes
5 related to respiratory effects (Section 5.2). cardiovascular effects (Section 5.3).
6 reproductive and developmental effects (Section 5.4). and mortality (Section 5.5).
7 Chapter 5 concludes with a discussion of the evidence for the cancer effects of oxides of
8 nitrogen (Section 5.6). In order to characterize the weight of evidence for the effects of
9 oxides of nitrogen on reproductive and developmental effects in a cohesive manner,
10 results from both short-term (i.e., up to 1 month) and long-term exposure studies are
11 included in this chapter and are identified according to their exposure duration in the text
12 and tables throughout Section 5.4.
13 Individual sections for major outcome categories (e.g., respiratory effects, cardiovascular
14 effects) begin with a summary of conclusions from the 2008 ISA for Oxides of Nitrogen
15 (U.S. EPA. 2008c) followed by an evaluation of recent (i.e., published since the
16 completion of the 2008 ISA for Oxides of Nitrogen) studies that is intended to build upon
17 evidence from previous reviews. Within each of these sections, results are organized into
18 smaller groups of endpoints (e.g., asthma incidence) then specific scientific discipline
19 (i.e., epidemiology, toxicology).
20 Sections for each of the major outcome categories (e.g., respiratory effects,
21 cardiovascular effects) conclude with an integrated summary of the assessment of
22 evidence and conclusions regarding causality. A determination of causality was made for
23 a major outcome category (e.g., respiratory effects) or smaller group of related outcomes
24 (e-g-, birth outcomes) by evaluating the evidence for each outcome category or group
25 independently with the causal framework (described in the Preamble to this ISA).
26 Judgments regarding causality were made by evaluating the evidence for the full range of
27 exposures to oxides of nitrogen or ambient concentrations in animal toxicological and
28 epidemiologic studies considered relevant to this ISA, i.e., NO2 concentrations up to
29 5,000 ppb as described in Section 1.1. Studies that examined higher NOX or NO2
30 concentrations were evaluated particularly to inform mode of action.
November 2013 5-1 DRAFT: Do Not Cite or Quote
-------
1 Judgments regarding causality were made by evaluating evidence for the consistency of
2 findings across multiple studies, the coherence of findings across related endpoints and
3 across disciplines, and the extent to which chance, confounding (i.e., bias due to a
4 correlation with NOX or NO2 exposures or ambient concentrations and relationship with
5 the outcome), and other biases could be ruled out with reasonable confidence. This
6 evaluation involved consideration of the strength of study design and analytical
7 methodology as well as the potential for selection bias, publication bias, and
8 confounding.
9 Epidemiologic studies of long-term NOX or NO2 exposure rely on between-subject
10 differences in exposure, as a result of residence in locations (spatial differences) or
11 examination in time periods that vary in long-term ambient NOX or NO2 concentrations,
12 for example. For the assessment of potential confounding, long-term exposure
13 epidemiologic studies were evaluated for the extent to which they considered other
14 factors associated with health outcomes and correlated with NOX or NO2 exposure that
15 vary between subjects. These potential confounding factors can include socioeconomic
16 status, diet, smoking or exposure to environmental tobacco smoke, medication use, and
17 copollutant exposures. Epidemiologic studies varied in the extent to which they
18 considered potential confounding. Because no single study considered all potential
19 confounding factors, and not all potential confounding factors were examined in the
20 collective body of evidence, residual confounding by unmeasured factors is possible.
21 Residual confounding also is possible by factors that are poorly measured. The evidence
22 was examined based on factors shown to be associated with NO2 exposure and health
23 outcomes. Epidemiologic studies present effect estimates for associations with health
24 outcomes scaled to various changes in concentrations, e.g., interquartile range for
25 exposures of the study population or an arbitrary unit such as 10 ppb. To increase
26 comparability among studies, the ISA presents effect estimates for a given averaging time
27 scaled to the same increment. Compared with short-term averages, long-term averages of
28 ambient concentrations are lower, less variable across time, and do not differ widely
29 among averages of multiple months, annual averages, or multiyear averages. Thus, for
30 long-term exposure, effect estimates are scaled to a 10-ppb increase in NO2 or NO and a
31 20-ppb increase in NOX. These increments were derived by calculating the U.S.
32 nationwide percentile distributions for annual average concentrations, and they represent
33 the approximate difference between the median and 95th percentile of annual average
34 concentrations among monitors in the network.
35 Animal toxicological studies can provide direct evidence for health effects related to or
36 NO2 exposures. Results from these studies were also used to address uncertainties in the
37 epidemiologic evidence, such as potential confounding. Experimental studies also
38 provide biological plausibility by describing key events to inform modes of action for
November 2013 5-2 DRAFT: Do Not Cite or Quote
-------
1 health effects. Thus, the integration of evidence across a spectrum of related endpoints,
2 including cause-specific mortality, and across disciplines was used to inform
3 uncertainties for any particular endpoint or discipline due to factors such as publication
4 bias, selection bias, confounding by copollutant exposures.
5.2 Respiratory Effects
5.2.1 Introduction
5 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) examined the epidemiologic
6 and toxicological evidence for effects of long-term exposure to NO2 on respiratory
7 morbidity and concluded that the evidence was suggestive but not sufficient to infer a
8 causal relationship. The key supporting epidemiologic evidence comprised positive
9 associations between NO2 and decrements in lung function and partially irreversible
10 decrements in lung function growth, although toxicological studies provided little
11 evidence for key events to inform the mode of action for these observations. Animal
12 studies did, however, demonstrate that long term exposure to NO2 resulted in permanent
13 morphologic changes to the lung, particularly the centriacinar region and bronchiolar
14 epithelium. Results from the Southern California Children's Health Study (CHS)
15 indicated that decrements in lung function in children were associated with increasing
16 ambient NO2 concentrations (Gauderman et al.. 2004). though similar associations were
17 also found for PM, O3, and proximity to traffic (<500 meters). Generally, the high
18 correlation among traffic-related pollutants in these long-term exposure studies made it
19 difficult to accurately estimate the independent effects of individual pollutants.
20 Additional uncertainty was related to the inconsistent evidence for associations between
21 long-term exposure to NO2 and increases in asthma prevalence and incidence. For
22 example, two cohort studies, the CHS in southern California (Gauderman et al.. 2005)
23 and a birth cohort study in the Netherlands (Brauer et al., 2007) observed positive
24 associations, while other studies did not find consistent associations between long-term
25 NO2 exposure and asthma outcomes. Epidemiologic studies conducted in both the U.S.
26 and Europe also reported inconsistent results regarding an association between long-term
27 exposure to NO2 and respiratory symptoms.
28 Recent prospective studies have evaluated the association between long-term exposure to
29 NO2 and various respiratory morbidity endpoints, including the reduced lung function
30 growth and lung function in children, the development of asthma and bronchitis, and the
31 role such exposure has in the aging respiratory system and the development of disease in
32 adults.
November 2013 5-3 DRAFT: Do Not Cite or Quote
-------
1 In this section, the current body of evidence for associations between long-term exposure
2 to NO2 and respiratory morbidity is characterized. This includes respiratory morbidity
3 studies in children such as asthma incidence, pulmonary function development,
4 respiratory symptoms, and various indicators of inflammation and allergy. For adults, the
5 endpoints evaluated include adult onset asthma, pulmonary function decrements,
6 respiratory symptoms, hospital admissions for respiratory disease, and indicators of
7 pulmonary inflammation. Individual studies are described and results are integrated
8 within the epidemiologic evidence base. Exposure-related concerns relevant to all
9 epidemiologic studies are discussed, including NO2 exposure assessment methods, the
10 potential role of NO2 as a surrogate measure for another pollutant or pollutant mixture,
11 and copollutant associations. Additionally, it should be noted that exposure measures
12 characterized in epidemiologic studies in this section are annual averages unless stated
13 otherwise. Epidemiologic study design characteristics are also considered in evaluating
14 the evidence and include study type (i.e., longitudinal, prospective study versus cross-
15 sectional study), the age at which exposure and/or respiratory endpoint is assigned, the
16 allergic versus nonallergic status, the follow-up interval, the concentration-response
17 relationship, and coherence and consistency within the epidemiologic evidence. No
18 recent animal toxicological studies evaluating respiratory effects of long-term NO2
19 exposure have been published since the release of the 2008 ISA for Oxides of Nitrogen,
20 but previous studies are evaluated for lung host defense; pulmonary inflammation, injury,
21 and oxidative stress; and respiratory morphology.
5.2.2 Asthma/Chronic Bronchitis Incidence
5.2.2.1 Children
22 Since the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). recent prospective
23 longitudinal cohort studies provide a larger evidence base to evaluate the relationship
24 between asthma incidence in children and long-term NO2 exposure. Details from these
25 prospective studies are presented in Table 5-1 and Figure 5-3 and in the text later in this
26 section. Cross-sectional studies were reviewed and are discussed as appropriate to inform
27 discussion of copollutants, genetic variants and other policy-relevant issues (Tung et al..
28 2011; AkinbamietaL 2010; Hwang and Lee. 2010; Kim et al., 2008; Morgenstern et al..
29 2008; Hwang et al.. 2005) and other studies found in the Annex tables of the 2008 ISA
30 for Oxides of Nitrogen (U.S. EPA. 2008c). Cross-sectional studies also reported
31 associations between NO2 and asthma. The prospective studies demonstrate a positive
32 relationship. The evidence base includes studies from North America, Europe and Asia
33 that use different designs and analyses.
November 2013 5-4 DRAFT: Do Not Cite or Quote
-------
1 The results of individual studies are supported by recent meta-analyses that report
2 positive associations (Gasana et al.. 2012; Braback and Forsberg. 2009). several of which
3 adjust for or examine the influence of publication bias (Anderson et al., 2013; Powers et
4 al.. 2012; Takenoue et al.. 2012). Some of these meta-analyses included children and
5 adults, and some included both cross-sectional and prospective studies.
6 In the majority of studies, asthma incidence was assessed using an annual respiratory
7 questionnaire that asked parents whether a doctor has ever diagnosed the child as having
8 asthma, without having fulfilled the definition of asthma at any previous time of follow-
9 up. Several studies assessed asthma incidence in a different manner. For example,
10 Carlsten et al. (201 la) used a pediatric allergist to assess asthma in the children when
11 they were 7 years old. Gruzievaet al. (2013) defined asthma incidence as children at 12
12 years of age having at least 4 episodes of wheeze in the last 12 months, or at least one
13 episode in combination with prescription of inhaled corticosteroids which would be
14 provided by a physician making a diagnosis of asthma.
15 The relationship between traffic-related air pollution and asthma onset in children was
16 evaluated in the prospective CHS cohort living in 11 communities with individual
17 exposure measurement (Jerrett et al.. 2008). The design of the study allowed the
18 examination of the independent contributions of local versus regional NO2 to the
19 associations with asthma by modeling the effects of the within- and between-community
20 variation in NO2. Similar results were obtained in analyses of within-community
21 variation and between-community variation in NO2, providing evidence that both
22 regional and local pollution contributed to the associations with asthma. However, these
23 results were based on a 6.2-ppb increase in NO2, which was the within-community
24 interquartile range for NO2. NO2 concentrations showed less variation within than
25 between communities. The hazard ratio (HR) for between-community effects increases to
26 3.25 if based on the average interquartile range across all measurements of 28.9 ppb for
27 annual NO2 and to 1.95 if based on the average within-community range of 16.4 ppb.
28 In a new CHS cohort recruited in 2002-2003 from 13 southern California communities,
29 McConnell et al. (2010) prospectively evaluated childhood incident asthma (n = 120) and
30 traffic-related pollutant (TRP) exposure based on estimates on a line source dispersion
31 model of traffic volume, distance from home and school, and local meteorology. Ambient
32 pollutants O3, NO2, and PM measured at central sites were also evaluated. Of the
33 regional community central-site pollutants, a 10-ppb increase in ambient NO2 measured
34 at a central site in each community was associated with an HR for new onset asthma or
35 1.39 (95% CI: 1.07, 1.80). In models with both NO2 and modeled TRP, there were
36 independent associations of asthma with TRP at school and home, whereas the estimate
37 forNO2 was attenuated (HR 1.14 [95% CI: 0.85, 1.53]).
November 2013 5-5 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
10
The GALA II (Chicago, IL; Bronx, NY; Houston, TX; San Francisco, CA; Puerto Rico)
and SAGE II (San Francisco, CA) cohorts of Latino and African American children and
young adults ages 8 to 21 were used to asses a relationship between air pollution and
development of asthma (Nishimura et al.. 2013a). A 10-ppb increase in average NO2
during the first year of life was associated with physician-diagnosed asthma with an OR
of 1.37 (95% CI: 1.08, 1.73). The different study regions had different concentrations and
mixtures of pollutants reflecting differing geography, weather, and pollutant sources
(Figure 5-1 for NO2). Most regions showed a nominally positive association with very
little between-study heterogeneity suggesting that the association between NO2 and
asthma is generalizable across geographic regions.
600-
^400-
8200
o-
60
NO2 exposure (ppb)
Study Region Fl Chicago New York Puerto Rico San Francisco (GALA) San Francisco (SAGE)HI Texas
Note: Abbreviations: NO2 = nitrogen dioxide, GALA = Genes-environments & Admixture in Latino Americans, SAGE = Study of
African Americans, Asthma, Genes & Environments. Dotted line = World Health Organization air quality guideline.
Source: Reprinted with permission of the American Thoracic Society, Nishimura et al. (2013a). specifically Supplemental Material:
(Nishimura etal., 2013b).
Figure 5-1
Distribution of NOa exposure in first year of life in GALA ll/ SAGE.
11
12
13
14
15
16
All children born in southwestern British Columbia in 1999 and 2000 (n = 37,401) were
assessed for incidence of asthma diagnosis up to 3-4 years of age using outpatient and
hospitalization records and evaluated for a relationship to air pollution (Clark et al..
2010). A total of 3,482 children (9%) were classified as asthma cases, and a 10-ppb
increase in NO2 was associated with asthma with an OR of 1.24 (95% CI: 1.14, 1.35).
Clougherty et al. (2007) used geographic information systems (GlS)-based models to
November 2013
5-6
DRAFT: Do Not Cite or Quote
-------
1 retrospectively estimate residential exposures to traffic-related pollution for 413 children
2 in a community-based pregnancy cohort, recruited in East Boston, Massachusetts,
3 between 1987 and 1993. Monthly NO2 measurements for 13 sites over 18 years were
4 merged with questionnaire data on lifetime asthma incidence. Univariate ORs for the
5 seven candidate NO2 exposure periods indicated that a 10-ppb increase in NO2 was
6 associated with asthma with an OR of 1.44 (95% CI: 0.87, 2.40).
7 A prospective study by Lee et al. (2012c) examined whether associations between high
8 pulmonary function indices and incident respiratory diseases in Taiwan are modified by
9 long-term ambient NO2 exposure. Questions regarding respiratory symptoms and
10 diseases were modeled after those used in the Children's Health Study in southern
11 California. They report loss of protective effects by each pulmonary function index
12 against bronchitis, chronic cough, and asthma in the communities with higher NO2
13 concentrations compared to those with lower NO2 concentrations. Per interquartile range
14 increase in forced expiratory flow over the mid-range of expiration, the RR for asthma is
15 0.73 (95% CI: 0.64, 0.83) in the low NO2 communities (<17.5 ppb) and 0.97 (95% CI:
16 0.85, 1.11) in the high NO2 communities (>17.5 ppb). These results are consistent with
17 the results of a similar analysis conducted by the Southern California CHS (Islam et al..
18 2007). For the purposes of presenting results in this ISA, the RRs reported above were
19 combined to calculate the main effect of NO2 exposure as RR: 1.33 (95% CI: 1.14, 1.52)
20 in the high NO2 community, with the low community serving as the reference. The main
21 effect of NO2 was calculated using the following formula:
November 2013 5-7 DRAFT: Do Not Cite or Quote
-------
RR for high NO2 community versus low NO2 community =
exp[log(high NO2 RR)-log(lovv NO2 RR)] =exp[log(0.97)-Iog(0.73)] = 1.33.
Standard error (s.e.) for the high NO2 community RR =
[log(l. 11 )-log(0.85)]/l .96x2 = 0.07
s.e. for the low NO? community RR =
[log(0.83)-log(0.64)]/l .96x2 = 0.07
Lower 95% confidence limit =
1.33 - 1.96 7(s. e, high N02)2 + (s. e. low N02)2 = 1.14
Upper 95% confidence limit =
1.33 + 1.96 7(s- e. high /V02)2 + (s. e. low N02}2 = 1.52
2 In the longitudinal analysis of the 8-year follow-up of the Dutch Prevention and
3 Incidence of Asthma and Mite Allergy (PIAMA) study in the Netherlands, Gehring et al.
4 (2010) studied the association between traffic-related air pollution and the development
5 of asthma, allergy and related symptoms. Preceding this analysis was Braueretal. (2002)
6 at the 2-year time period reviewed in the 2008 ISA for Oxides of Nitrogen and at the 4-
7 year time period (Brauer et al., 2007); both presented cross-sectional analysis. Individual
8 exposures to NO2, PM2 5, and soot at the birth address were estimated by land-use
9 regression models. This prospective birth cohort (n ~ 3,100) study (Gehring et al., 2010)
10 found positive associations to long-term NO2 exposure in adjusted analysis for prevalent
11 asthma, incident asthma, and wheeze. The results for incident asthma are shown in Figure
12 5-2. While the overall result is significant, the figure shows that for NO2 significance is
13 not achieved until age 8. Similar results are shown for PM2 5 and soot. In this study no
14 associations were found with atopic eczema, allergic sensitization, and bronchial
15 hyperresponsiveness.
November 2013 5-8 DRAFT: Do Not Cite or Quote
-------
•
2.8
2.0
1.2 '
n A
Incident asthma past 12 mo.
T
jc
>
4
J
4
I
IP L *v U
;>i
a
X
i
c
c
)
>
1
•:
T
:
4
i
1 345678 overall
Age (years)
NO
soot
Note: Results are presented as adjusted except study region odds ratios (ORs) with 95% confidence intervals. Because study
region is an important determinant of air pollution levels in the land use regression models that were used to estimate exposures,
the adjustment for region may be an over adjustment. ORs were calculated for an interquartile range increase in air pollution levels
of 5.5 ppb for NO2; blank circle NO2; Gray circle PM25; and Black circle soot.
Source: Reprinted with permission of American Thoracic Sociey, Gehringetal. (2010).
Figure 5-2 Adjusted overall and age-specific association between annual
average levels of air pollution at the birth address and asthma
during the first 8 years of life.
i
2
3
4
5
The Swedish population-based birth cohort, BAMSE, examined development of asthma
and related symptoms over a 12-year follow-up (Gruzieva et al.. 2013). A 20-ppb
increase in NOX during the first year of life was associated with development of incident
asthma at 12 years of age with an OR of 1.65 (95% CI: 1.0, 2.77). The Oslo Norway birth
cohort was used to investigate the associations of long-term traffic-related exposures in
early life and before onset with onset of doctor-diagnosed asthma assessed
retrospectively in current 9- to 10-year-old children (Oftedal et al., 2009). The effect of
annual averages pollutant levels over a 10-year period on asthmatic symptoms were
November 2013
5-9
DRAFT: Do Not Cite or Quote
-------
1 assessed in a prospective cohort study of 1,910 schoolchildren in 8 different communities
2 in Japan (Shima et al.. 2002). During the follow-up period, incidence rates of asthma
3 were associated with NO2.
4 A high-risk birth cohort for the risk of new onset asthma was considered in a population
5 of children with first-degree relatives with asthma in Vancouver, Canada, that was
6 studied by Carlsten et al. (201 Ic) to evaluate the association between early exposure to
7 traffic related air pollution using land use regression. The OR associated with a 10-ppb
8 increase in NO2 in the total group was 2.88 (95% CI: 0.76, 10.94). Combined early
9 exposure to dog (elevated Can-fl levels or dog ownership) plus elevated NO2 conferred
10 an increased risk of incident asthma (Carlsten et al.. 201 la).
11 In a cross-sectional analysis, Akinbami et al. (2010) examined the association between
12 chronic exposure to outdoor pollutants (12-month avg levels by county) and asthma
13 outcomes in a national sample of children ages 3-17 years living in U.S. metropolitan
14 areas (National Health Interview Survey, N = 34,073). An increase in annual NO2
15 concentration yielded a negative association both for children currently having asthma
16 and for children having at least 1 asthma attack in the previous year. Models in which
17 pollutant value ranges were divided into quartiles yielded adjusted odds for current
18 asthma for the highest quartile (30.8-40.2 ppb) of estimated NO2 exposure of 1.02 (95%
19 CI: 0.71, 1.47).
20 The exposure methods in the above asthma incidence studies are listed in Table 5-1.
21 Additionally, papers related to these studies that provide in depth discussion of the
22 exposure methods are referenced in supplemental Table S5-1 (U.S. EPA. 2013e). The
23 majority of these studies subjected methods to validation testing to ensure that the data
24 were of sufficient quality. Of the exposure assessment methods employed, Palmes tubes
25 are sometimes subject to positive biases in NO2 (Section 2.6.3.1). dispersion models not
26 employing NOX chemistry would potentially overestimate NO2 concentrations (Section
27 2.6.2.2). and central site NO2 monitors may fail to capture spatial variability in
28 concentrations and exposures influenced by time activity data (Section 2.6.5). All of
29 these issues would have the effect of biasing the health effect estimate towards the null,
30 so that reported effect estimates are conservative. LUR model predictions have been
31 found to correlate well with outdoor NO2 concentration measurements (Section 2.6.2.3)
32 and so wouldn't be anticipated to result in substantial bias.
33 These studies differ in exposure period evaluated, age of asthma diagnosis, and length of
34 follow-up time. Several involve birth cohorts up to an age of 8 to 12. Gruzieva et al.
35 (2012) note that in general, the strongest effect was observed in relation to exposure
36 during infancy, which may indicate that prenatal and early-life periods represent critical
37 windows for the effects of exposure on development of childhood asthma and related
November 2013 5-10 DRAFT: Do Not Cite or Quote
-------
1 symptoms. Other studies found larger magnitudes of association with NO2 exposure in
2 the first 3 years of life (Nishimura et al.. 2013a) or year of diagnosis (Cloughertv et al..
3 2007). indicating other time periods of exposure also are important. Other studies
4 evaluate children starting at age 8-10 and follow them until adolescence or young
5 adulthood and report generally positive results. Age of asthma diagnosis is also a factor.
6 Transient wheezing is common in infants and often resolves as the child ages (Martinez
7 et al.. 1995) and thus the reliability of asthma diagnosis in infants is a factor to consider.
8 The results of this NO2 asthma incidence evidence base are greater in magnitude and
9 generally stronger at later age evaluation and longer follow-up time.
10 While model adjustment varied by study, the collective body of evidence adjusted for
11 multiple potential confounding factors, including age, various SES indicators such as
12 maternal education and household income, health indicators such as family history of
13 asthma and atopic status, exposure to cigarette and wildfire smoke, housing
14 characteristics, presence of a gas stove in the child's home, and meteorological conditions
15 such as temperature and humidity.
November 2013 5-11 DRAFT: Do Not Cite or Quote
-------
Table 5-1 Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Results:
(Results are
standardized to
Exposure Pollutant Correlation Statistical Methods Comments 10 ppb)
Childrens Health Study (CHS), Southern California communities
Jerrett et al. (2008)
n = 217 children, 10-18
years of age enrolled in
1993 or 1996 in
11 Communities with 8
years of follow-up
Palmes tubes outside
home, 2 weeks summer
and winter to provide
annual and seasonal
levels
Moderate to high
correlations between the
measured residential
NO2 and various
measures of traffic
proximity or modeled
concentrations
Random-effects Cox
proportional hazards
models. SEP measures
of median household
income, proportion of
respondents with low
education (i.e., no high
school diploma), percent
of males unemployed
(as a marker for fulltime
income instability), and
percent living in poverty
were tested as potential
confounders.
Meteorological
conditions, temperature,
and humidity were
tested.
Within-community
effects indicative of
long-term local traffic
sources were similar to
effects of community
average NO2 across
communities,
suggesting that both
regional and local
pollution contributed to
associations with
asthma. The range of
variation of NO2 within
communities was
smaller than that
between communities,
and the HR values were
smaller when scaled to
the range within
communities
Adjusted HR of 1.29
(95% Cl: 1.07, 1.56) per
the average within-
community interquartile
range of 6.2 ppb in
annual residential NO2.
Based on the total
interquartile range for all
measurements of 28.9
ppb, the HR was 3.25
(95% Cl: 1.35, 7.85)
November 2013
5-12
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results:
(Results are
standardized to
10 ppb)
Childrens Health Study (CHS), Southern California communities (Continued)
McConnell et al. (2010)
n = 120 children 4.8 to
9.0 years of age
enrolled into a new
cohort during 2002-2003
in 13 communities with 3
years of follow-up
Community central site
pollutant measurements
and
Line source dispersion
model for residential and
school TRP
Not Reported
Multilevel Cox
proportional hazards
model.
Socio-demographic
characteristics,
exposure to cigarette
and wildfire smoke,
health insurance,
housing characteristics,
history of allergy, and
parental asthma were
assessed
Association of new-
onset asthma with
community central site
pollutant measurements.
HR: 1.39 (95% Cl: 1.07,
1.80); p = 0.01 per 10
ppb across the 13 study
communities at central
sites.
Modeled traffic-related
pollution exposure from
roadways near homes
(HR: 1.67 [95% Cl: 1.32,
2.12]) and near schools
(HR 1.88 [95% Cl: 1.10,
3.19]) per 10-ppb
increase.
November 2013
5-13
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Exposure
Pollutant Correlation
Statistical Methods
Comments
Results:
(Results are
standardized to
10 ppb)
Prevention and Incidence of Asthma and Mite Allergy (PIAMA) Study, the Netherlands
Gehrinq et al. (2010)
n = 3,863 children in
birth cohort aged 1 to 8
years
Braueretal. (2007):
Brauer et al. (2002)
Land-use regression
models used to estimate
annual concentrations
for birth address of each
participant
The estimated
exposures for the
various pollutants were
highly correlated (r =
0.93, 0.96, and 0.97 for
the correlation between
NO2 and PM2.5, NO2
and soot, and PM25 and
soot, respectively.
Generalized estimation
equations
Potential confounding
variables included sex,
study arm (intervention
or natural history), use
of mite-impermeable
mattress covers,
allergies of mother and
father, maternal and
paternal education,
maternal smoking during
pregnancy,
breastfeeding, presence
of a gas stove in the
child's home, presence
of older siblings, and
any smoking at home.
Adjusted OR of 1.37
(95% Cl: 1.09, 1.72) for
a 10-ppb increase
without adjustment for
study region.
Vancouver High Asthma Risk Birth Cohort
Carlsten et al. (2011c)
n = 184 children at
7 years of age
Carlsten et al. (2011 a):
Carlsten et al. (2011b)
Land use regression
used to estimate annual
concentrations of the
birth address of each
subject. Model
validated.
Pearson correlations
between pollutant
measures were as
follows:
NO-NO2, r = 0.8;
NO2-PM25, r=0.7;
NO-PM25, r=0.5;
BC-NO, r = 0.5;
BC-NO, r = 0.3; and
BC-PM25, r=0.2.
Multiple logistic
regression analysis,
potential confounders
(maternal education,
history of asthma (in
mother, father or
siblings), atopic status at
1 year)
High risk was defined as
a child having,
according to parental
report, at least one first-
degree relative with
asthma or two first-
degree relatives with
other IgE-mediated
allergic disease (atopic
dermatitis, seasonal or
perennial allergic
rhinitis, or food allergy).
The risks associated
with NO in both the total
and control groups was:
ORs:
1.8(95%CI: 1.1,2.9)
and2.5(95%CI:1.2,
5.2), respectively.
The risk associated with
NO2 in the total group
was:
OR 2.3 (95% Cl: 1.0,
5.1); p = 0.05).
November 2013
5-14
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Results:
(Results are
standardized to
Exposure Pollutant Correlation Statistical Methods Comments 10 ppb)
Taiwan Children Health Study (TCHS)
Leeetal. (2012c)
14 communities, n =
3,160, 12-14 years of
age cohort entry 2007,
mean follow-up 2 years
Average hourly levels of
NC>2 measured in 14
monitoring stations used
to compute annual
average of ambient NO2
levels between 2007
and 2009.
Poisson regression
models, potential
confounders, included in
utero exposure to
maternal smoking,
family history of asthma,
family history of atopy,
and community
The two NO2 strata
were defined as less
than and greater than
the median level of 17.5
ppb: the mean annual
ambient NO2 level was
22.1 ppb in higher NO2
communities and 14.0
ppb in lower NO2
communities
For each inter-quartile
range changes in
maximal mid-expiratory
flow (MMEF) the
incidence rate ratio of
asthma was lower in the
lower NO2 communities
(RR: 0.73[95%CI:0.64,
0.83]), compared with
the effect in the higher
NO2 communities
(RR: 0.97 [95% Cl: 0.85,
1.11] (p for interaction =
0.04)
Children, Allergy, Milieu, Stockholm, Epidemiology Survey (BAMSE)
Gruzieva et al. (2013)
Swedish birth cohort
followed up to 12 years
of age enrolled between
1994 and 1996 n =
3,633
Nordlinq et al. (2008)
Gruzieva et al. (2012)
Dispersion models were
used to calculate NOX
for all addresses in the
years 1994 to 2008
representing when the
first child was born until
the end of the 12-year
follow-up.
r = 0.96 between NOX
and PM-io exposure
levels during the first
year of life
Multinomial regression/
GEE
Potential confounders
adjusted for include
municipality, SES, year
the house was built, and
heredity.
Associations were
stronger for the oldest
children and for non
allergic asthma. The
authors investigated
several time aspects of
long-term exposure,
including early-life
exposure and current
exposure (during the
last year), and average
exposure between
follow-ups.
Association between
NOX during the first year
of life and development
of incident asthma at
12 years of age
OR: 1.87 (95% Cl: 1.0,
3.44).
46.8 ug/m3for NOX.
November 2013
5-15
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results:
(Results are
standardized to
10 ppb)
Oslo Norway birth cohort
Oftedal et al. (2009)
n = 2,329 children born
in Oslo in 1992-1993
were assessed when 9
to 10 years of age
NO2 exposure was
assessed by the
EPISODE dispersion
model and assigned at
updated individual
addresses during
lifetime.
PM-io and PM25 Cox proportional hazard
pollutants are correlated regression and logistic
with NO2 (r = 0.79-0.91 ). regression were used.
Potential confounders
considered included
sex, parental atopy,
maternal smoking in
pregnancy, paternal
education, and maternal
marital status at the
child's birth
Several long-term
exposures were
included: early exposure
in first year of life,
average exposure from
birth to asthma onset,
and previous year's
exposure before
completing the
questionnaire.
Adjusted RR and 95%
Cl (per IQR for NO2
varies between 14.5 and
10.4 ppb, decreasing
overtime)
For NO2 exposure 1st
year of life was 0.82
(0.67, 1.02); early onset
less than 4 years of age
was 0.78 (0.62, 0.98)
and for equal or later
than 4 years of age was
1.05(0.64, 1.72).
Chiba Prefecture, Japan Cohort
Shima et al. (2002)
n = 1,910 children in 8
communities at age 6
entered 1st graders
between 1989 and 1992
were followed to 6th
grade
The average annual
concentrations of air
pollutants for the 10-yr
period from 1988 to
1997 at ambient air
monitoring stations in
the vicinities of the study
schools were used.
Not Reported
Logistic Regression
Model Incidence rates
were adjusted for sex,
history of allergic
diseases, respiratory
diseases prior to age 2,
parental history of
allergic diseases,
maternal smoking
habits, type of heater
used in winter in the
home, and construction
elements of the house.
The incidence of asthma
was associated with
increasing NO2
concentrations
(OR: 1.71 [95%CI:1.04,
2.79 for a 10-ppb
increase).
November 2013
5-16
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Results:
(Results are
standardized to
Exposure Pollutant Correlation Statistical Methods Comments 10 ppb)
Genes-environment & Admixture in Latino Americans and the Study of African Americans, Asthma, Genes & Environments
Nishimura et al. (2013a)
The sample size
analyzed was 4,320.
Multi city study that
includes: (Chicago, IL;
Bronx, NY; Houston, TX;
San Francisco Bay
Area, CA) and Puerto
Rico. Participants were
8-21 years old.
Exposures over the first
three years of life were
calculated by averaging
all available pollutant
values from birth to age
three using residential
histories using regional
ambient pollutant data
using inverse distance-
squared weighted
average from the four
closest monitors within
50 km of the residence.
The different study
regions had different
levels and mixtures of
pollutants, reflecting
differing geography,
weather, and pollutant
sources.
Logistic regression
models adjusted for age,
sex, ethnicity, and
composite SES were
used. A sensitivity
analysis examining
additional potential
covariates for maternal
in utero smoking,
environmental tobacco
smoke in the household
between 0 and 2 years
old, and maternal
language of preference
(as an indicator of
acculturation).
Region-specific results
suggest that
susceptibility to asthma
due to air pollution may
not be uniform
throughout the nation
and could be dependent
on local characteristics,
such as varying
proportions of different
racial/ethnic groups and
differing pollution
sources and/or weather
patterns
After adjustment for
confounders, a 10-ppb
increase in average NO2
during the first year of
life was associated with
an OR of 1.37 for
physician-diagnosed
asthma (95% Cl: 1.08,
1.73) for the entire
study.
British Columbia Birth Cohort
Clark etal. (2010)
N = 2,801, children
mean age at follow-up
48SD7-mo, all 1999
and 2000 births in SW
BC
Estimated using
regulatory monitoring
data, Land use
regression models , and
point source derived
inverse distance-
weighed (IDW)
summation of emissions
Correlations between
different pollutants were
generally high, only the
O3 r= -0.7 to-0.9 was
provided.
Covariate-adjusted
conditional logistic
regression. Covariates
previously hypothesized
to have an effect on
asthma status (native
status, breast-feeding,
maternal smoking,
income quartile,
maternal age, birth
weight, and gestational
length were included.
The potential limitation
of the young age of the
children were wheezing
is more common was
addressed by restricting
asthma cases to
children with a hospital
admission or at least
two outpatient diagnosis
of asthma which indicate
severe ongoing
symptoms.
An increased risk of
asthma diagnosis with
increased early life
exposure to CO, NO,
NO2, PM-io, SO2, and
black carbon (BC) and
proximity to point
sources was found.
Traffic-related pollutants
were associated with the
highest risks:
Adjusted ORs:
1.15(95%CI: 1.08,
1.24) for a 10-ppb
increase of NO, and
1.24 (95% Cl: 1.14,
1.35) for a 10-ppb
increase in NO2.
November 2013
5-17
DRAFT: Do Not Cite or Quote
-------
Table 5-1 (Continued): Prospective long-term NO2 exposure new onset asthma in children cohort studies.
Study
Results:
(Results are
standardized to
Exposure Pollutant Correlation Statistical Methods Comments 10 ppb)
Maternal-Infant Smoking Study of East Boston
Clouqhertvet al. (2007)
n = 413 full cohort/255
lifetime residents
Pregnant women were
recruited from East
Boston between 1987
and 1993 and a
questionnaire
administrated in 1997
when the childrens
average age was 6.8.
Estimated exposure
using Land use
regression based on
predictive model using
passive monitors, sites
and traffic
Not Reported
Regression model
examined for the main
effects of NO2. Potential
confounders included
maternal asthma,
education, smoking
before and after
pregnancy, child's sex
and age.
Found an association
between traffic-related
air pollution and asthma
solely among urban
children exposed to
violence.
Univariate ORs for the
seven candidate NC>2
exposure periods
indicated that a 10-ppb
increase showed the
following association
with asthma for the full
cohort: OR: 1.44;
(95% Cl:0.87, 2.40).
November 2013
5-18
DRAFT: Do Not Cite or Quote
-------
Study
Clark et at 201 0
Southwestern
British Columbia
Nested case control
Gehring et at 2010
Communities sn
The Netherlands
Carlsten et al , 2011
Southwestern
British Columbia
Gruzieva et al , 2013
Stockholm, Sweden
Shimaetal,, 2002
Exposure
Characteristics
inverse weighted summation of
emissions from pt sources within 10 miles
land use regression
NO22 at updated residence
land use regression estimate
of birth address, NO2
without adjustment for study region
Land use regression
estimate of residence, NO2
Land use regression
estimate of residence, NO
Dispersion modeled NOZ
at birth address, school, & daycare
NO2 monitors
CWba Perfecture, Japanwithin 3 km of school
Nishimura et al , 2013
6 U.S. cites
Clougherty ey al, 2007
Boston
Lee etaL 2012
Taiwan Children's
Health Study
McConnell et al ., 2010
Southern California
Children's Health Study
enrolled 2002-03
Oftedaletal.. 2009
Oslo, Norway
Jerrett ©t 3l 2008
Southern California
Children's Health Study
enrolled before 2002
inverse distance weighted
NO; from central site monitors
regression of NO2 from passive
samplers on traffic, land use, and others
Period risk
of Exposure type
birth year OR
birth year OR
birth address OR
OR
E>irfh ysar OR
OR
lifetime OR
6to11 OR
years old
3irth year OR
year of OR
diagnosis
ave hourly NOjfrom 14 12-14 years old RR
monitoring stations, mediated through FVC
mediated through MMEF
mediated through FEV1
modeled traffic related
NQj near homes
model traffic related
NQ2 near schools
Community central site NO,
Community central srte NO2
dispersion model estimate of
updated residence NO2
Palmes tubes outside home NO?
RR
RR
continuously HR
during followup
continuously HR
during followup
birth year HR
From birth HR
to onset
from birth HR
to 10 years old
14 or 17 HR
year olds
age yrs of
in years follow-up N
4 3-4 37401
1-8 8 3143
I m ~~J 1 RA m
Up / !o*l "
to age 7
up to 12 3633
age 12
6-12 5 3049
8-21 2-7 4320
Kir+H A ft ^"J1^ —
Dinn o.o *t u
to 18 SO 1.6
12-14 2 3160
4.8-9 3 3372
120
up to 9-10 2329 *
9 or 10
2329 fl
M1 ft A 1 Q&
" ! O O I WO
I
1
§
»-
»
• —
»•
* \
^
~*~
—•—
•9 —
0
• \
— «
I I 1
,5 1 1,5 2 2.5
risk ratio in 10 ppb for NO2, NO, NOX or modeled traffic pollution
Note: These studies are presented by the statistical method used: first by odds ratio (OR), then relative risk (RR), and finally by
hazard rate (HR).Then within the statistical method they are arranged by age.
Figure 5-3 Risk ratio estimates of asthma incidence from prospective
studies.
5.2.2.2
Adults: Asthma-Chronic Bronchitis
The European Community Respiratory Health Survey (ECRHS), a European adult cohort
study, assessed the relationship to long-term ambient pollution exposure and onset
November 2013
5-19
DRAFT: Do Not Cite or Quote
-------
1 asthma both via positively responding to the question "Have you ever had asthma?"
2 (Jacquemin et al.. 2009b) and via a continuous asthma score (Jacquemin et al.. 2009a).
3 Jacquemin et al. (2009b) analyzed 4,185 adults. Subjects' home addresses were geocoded
4 and linked to outdoor NO2 estimates, as a marker of local traffic-related pollution using
5 information from the 1-km background NO2 surface modeled in APMoSPHERE (Air
6 Pollution Modelling for Support to Policy on Health and Environmental Risk in Europe).
7 Asthma incidence was defined as reporting asthma in the follow-up (1999 to 2001) but
8 not in the baseline (1991 to 1993). They adjusted for center effects. A set of predefined
9 covariates was included (sex, age, socioeconomic status, atopy, family history of asthma
10 or atopy, and smoking). When assessing the risk of developing asthma using the 186 new
11 cases, the adjusted OR for asthma onset was 1.96 (95% CI: 1.04, 3.70) for a 10 ppb
12 change in NO2. All the stratified associations between asthma incidence and NO2 were
13 higher than 1. Results were homogeneous among centers in both crude and the adjusted
14 analyses (Figure 5-4) (Jacquemin et al.. 2009b).
November 2013 5-20 DRAFT: Do Not Cite or Quote
-------
Umea
Uppsala
Goteburg
Norwich-
Ipswich
Antwerp
Paris
Grenoble
Verona
Oviedo
Galdakao
Barcelona
Albacete
Huelva
Combined,
J
P
L
0.01
1 rj
Lj
"i
r
9
M
1 1
L4J 6228
OR for each 10 jig/m3 increase in NO2
Note: P value for heterogeneity 0.594. Erfurt in Germany, and Pavia and Torino in Italy, were automatically dropped from the
analysis due to empty cells. Boxes represent the OR percenter, where the size of the box is proportional of the sample size of such
center. Lines represent the 95% Cl of the respective OR. Diamond represents the combined OR.
Locations in the Figure: Umea, Uppsla, and Goteburg, Sweden; Norwich and Ipswich, U.K.; Antwerp, Belgium; Paris and Grenoble,
France; Verona, Italy; Oviedo, Galdakao, Barcelona, Albacte, and Huelva, Spain.
Source: Reprinted with permission of Wolters Kluwer Health, Jacguemin et al. (2009b).
Figure 5-4 Adjusted ratios of new asthma in ECRHS II (1999-2001) for every
10 ug/m3 (5.3 ppb) NO2 increase by center in subjects with no
asthma in ECRHS I (1991-1993).
i
2
3
4
5
6
7
Defining asthma as a continuous trait using a grading scheme based on reported
symptoms was used by Jacquemin et al. (2009a) to evaluate the relationship between
long-term pollution exposure and asthma among adults with this novel measure in the
ECRHS cohort. The asthma score was developed from the five symptoms: wheeze and
breathlessness, feeling of chest tightness, attack of shortness of breath at rest, attack of
shortness of breath after exercise, and woken by attack of shortness of breath during the
last 12 months. Covariates evaluated included: sex, age, social class (in five groups based
November 2013
5-21
DRAFT: Do Not Cite or Quote
-------
1 on the International Standard Classification of Occupations coding of the occupational
2 history at ECRHS II and derived from the longest-held job during the follow-up period
3 between ECRHS I and II), family history of asthma or atopy, smoking (no, former or
4 current), pack-years, exposure to second-hand tobacco smoke, any exposure to dust, fume
5 or gases at work, gas cooking and season of the interview. The results are expressed as
6 ratios of the mean asthma scores (RMS). A multivariate model and data pooling was
7 adopted to analyze the association between the score and NO2 concentration. Both were
8 defined at follow-up, in a subpopulation reporting neither symptoms nor asthma at
9 baseline. This population may be considered a sample being in all likelihood free of
10 asthma at baseline. Thus, the occurrence of symptoms at follow-up may be interpreted as
11 new onset of symptoms, which may ultimately reflect incidence of asthma. In the
12 multivariate analysis, the RMS for each increase of 10 ppb of NO2 was 1.48 (95% CI:
13 1.18, 1.85). The association was homogeneous among centers after excluding Turin,
14 which had very large confidence intervals; the p-value for heterogeneity was still not
15 significant. In participants with no asthma and no symptoms at baseline, the associations
16 between NO2 and asthma score were positive (RMS 1.25 [95% CI: 1.05, 1.50]). A high
17 symptom score was shown to be strongly associated with doctor diagnosed asthma
18 (Sunyer et al., 2007); thus, this study's finding may indicate a role of pollution in new
19 onset of asthma in adults. This interpretation is consistent with previous finding based on
20 the more traditional definition of' 'asthma incidence'', using asthma at follow-up among
21 those free of the disease at baseline (Jacquemin et al.. 2009b) discussed in the above text.
22 In the prospective Respiratory Health in Northern Europe (RHINE) cohort study, Modig
23 et al. (2009) assessed the relationship between traffic-related air pollution levels and two
24 definitions of asthma among adults aged 20-44 at inclusion: the cumulative number of
25 onset onset cases of asthma and incident cases of asthma. The RHINE cohort is based on
26 the random sample of people receiving the first screening questionnaire sent out within
27 the ECRHS stage 1 (Toren et al., 2004). The questionnaire was sent out in 1990 and
28 included questions regarding respiratory symptoms such as wheezing, attacks of asthma
29 and current use of asthma medication. All participants who answered the first
30 questionnaire received the follow-up questionnaire in 1999. In contrast, in the ECRHS
31 cohort, only a subsample of the participants that received the first survey was included in
32 the follow-up. The study population consisted of 10,800 participants, born between 1945
33 and 1973. In order to be defined as an onset case of asthma observed during the follow-
34 up period, the participant had to have negative answers to the questions on attacks of
35 asthma during the last 12 months and current use of asthma medication in the first survey,
36 followed by a positive answer to at least one of these questions at the follow-up: ' 'Do you
37 have or have you ever had asthma?" or "Have you ever had asthma diagnosed by a
38 doctor?" Adjustment was made for a predetermined set of potential confounding
39 variables: body mass index (BMI), sex, age, smoking, water damage or mold in the home
November 2013 5-22 DRAFT: Do Not Cite or Quote
-------
1 at any time during the last 8 years, and city, simultaneously in the main analysis.
2 Socioeconomic index (SEI) based on job title was used for 80% of the participants to
3 classify five categories, and was only used for sensitivity analysis. The overall winter
4 half-year mean NO2 concentration was 9.6 ppb in total for Gothenburg and Umea. The
5 indoor/outdoor ratio for NO2 in Umea was 0.4/0.7. In the 3,824 participants, the analysis
6 of the relationship between onset and incident cases of asthma and the levels of NO2
7 outside the home showed a positive coefficient, indicating an increased risk of
8 developing asthma among adults with increasing levels of NO2 outside the home. The
9 OR in the fully adjusted model was 2.04 (95% CI: 1.14, 3.65) for the onset definition and
10 2.25 (95% CI: 1.00, 5.07) forthe incident definition of cases per 10-ppb increase in NO2
11 levels. When NO2 was grouped into tertiles there was a dose-response pattern, with
12 higher estimates for the third tertile (ORonSet 1.58 [95% CI: 0.96, 2.6]; ORmcident 2.06
13 [95% CI: 0.98, 4.32]) than forthe second tertile (ORonset 1.17 [95% CI: 0.70, 1.94];
14 ORmcident 1.77 [95% CI: 0.86, 3.64]), and with the first tertile used as a reference.
15 In 13 European cities, Castro-Giner et al. (2009) prospectively identified interactions
16 between genetic variants and traffic-related pollution on asthma incidence and prevalence
17 in the large (2,577 subjects at follow-up) adult population-based cohort - ECRHS. The
18 genetic aspect of this study is discussed in Section 5.2.11. In an analysis of longitudinal
19 data to evaluate the effect of NO2 on new-onset asthma they observed an association
20 between new-onset asthma and NO2 levels for the 120 subjects who developed asthma
21 during the follow-up period (OR = 2.20 [95% CI: 1.17, 4.10]). The authors restricted the
22 analysis to those subjects who lived in the same home during follow-up (n = 1,348) to
23 reduce exposure misclassification. Compared with subjects who changed homes, this
24 group had an increased risk for main effects of exposure to NO2 on asthma prevalence
25 (movers OR: 2.53 [95% CI: 1.16, 5.56]; non-movers OR:1.04 [95% CI: 0.66, 1.64];/>-
26 value for interaction = 0.03), whereas the effect on new-onset asthma was not different
27 between movers and non-movers (movers OR: 2.09 [95% CI: 0.70-6.12], non-movers
28 OR: 2.39 [95% CI: 1.10, 5.22], p-value for interaction = 0.81).
29 A separate analysis of the ECRHS cohort examined the relationship of chronic bronchitis
30 and air pollution. Sunyeretal. (2006) used two definitions for symptoms of chronic
31 bronchitis: (1) productive chronic cough for chronic cough and chronic phlegm (more
32 than three months each year), and (2) chronic phlegm alone. Since the two definitions
33 yielded similar results and given the higher frequency only the latter was used. The
34 prospective nature of the study design allowed consideration of two potentially different
35 symptomatic groups, namely those with symptoms at baseline, and those with symptoms
36 only at follow up ("new onset"). The follow-up time period was 8.9 years. Potential
37 confounding factors evaluated included smoking, age at end of education, occupational
38 groups, and occupational exposures, respiratory infections during childhood, rhinitis,
November 2013 5-23 DRAFT: Do Not Cite or Quote
-------
1 asthma, and traffic intensity at household level. A home based measurement of NO2
2 using Palmes tubes as a marker for local tail pipe emissions was implemented. At this
3 individual level, outdoor (at the kitchen window) and kitchen indoor NO2 concentrations
4 were collected during a 14 day period in 16 centers involving 1,634 households of
5 subjects who did not move house during the follow up. After about six months this
6 procedure was repeated in 659 households (45%) who volunteered to repeat the
7 measurement. A dose-response relationship with NO2 using the GAM modeling after
8 adjusting for variables was reported by the authors. The dose-response with NO2 was
9 linear in females but not in males (p for gain of non-linearity 0.15 and 0.03, respectively).
10 Among females, the association remained when the outcome was chronic productive
11 cough instead of chronic phlegm (the OR for a change of 10 ppb being 1.48 [95% CI:
12 0.99,2.20]).
5.2.3 Pulmonary Function
5.2.3.1 Epidemiologic Studies
Children
13 Recent prospective cohort studies add to the evidence base that evaluates the relationship
14 between supervised pulmonary function tests and long-term NO2 exposure. These
15 longitudinal prospective studies are summarized in Table 5-2 and the following text.
16 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) characterized the long-term
17 prospective studies as observing a positive association for the relationship between NO2
18 concentrations and decrements in lung function and reduction in growth. A key
19 uncertainty associated with these studies was the high correlation of NO2 concentrations
20 with other ambient pollutants. In general the studies depicted in Figure 5-3 continue to
21 show the relationship observed earlier, that is consistent decrements in lung function,
22 especially as children reach later ages.
23 A recent study of the long-term relationship between exposure to air-pollution and lung
24 function was examined in 1,924 school-age children in the Swedish birth cohort BAMSE
25 (Schultz et al.. 2012) that were followed with repeated questionnaires, dynamic
26 spirometry and IgE measurements until 8 years of age. Exposure during the first year of
27 life was associated with a deficit of 28.0 mL (95% CI: -64.0, 8.0) in FEVj for a 20 ppb
28 difference in time-weighted exposure to traffic-NOx, while the corresponding deficit was
29 -79.2 mL (95% CI: -135.0, -22.8) among those sensitized at 8 years. The odds ratios
November 2013 5-24 DRAFT: Do Not Cite or Quote
-------
1 associated with 80% and 85% of predicted FEVi were 1.69 (95% CI: 0.67, 4.2), and 2.6
2 (95% CI: 1.4, 4.5), respectively, for first year exposure to traffic-NOx. No impact of
3 short-term air pollution exposure on the estimates of the long-term effects of air pollution
4 was found, but these analyses were limited to PMi0.
5 The long term effects of PMi0 and NO2 exposure on specific airway resistance (sRaw)
6 and forced expiratory volume (FEVi) before and after bronchodilator treatment was
7 examined within the Manchester Asthma and Allergy Study (MAAS) birth cohort
8 (N = 1,185) (Molter et al.. 2013). Lifetime exposure to NO2 was associated with less
9 growth in FEVi (% predicted) over time, both before (16.0% [95% CI: -26.0, -0.5]) for a
10 10-ppb increase in NO2) and after bronchodilator treatment (23% [95% CI: -37.0, -9.0]).
11 In the recent CHS study, Breton et al. (2011) examined the association between ambient
12 air pollutant exposures and lung function growth in children. This study involves a
13 second cohort of children in the CHS study started in 1996. Results are shown in Table
14 5-2. Gauderman et al. (2004) examined the 1993 CHS cohort. The results of Breton etal.
15 (2011) are consistent with earlier results from the cohort (Gauderman et al.. 2004).
16 In Mexico City, Mexico, Rojas-Martinez et al. (2007a. b) evaluated the association
17 between long-term pollutant exposure and lung function growth in a prospective analysis
18 in a cohort of 3,170 children aged 8 years at baseline. In multipollutant models presented
19 in Section 1.5.2. the negative association of O3, PMi0, and NO2 with lung function
20 growth persisted: 10-ppb increase in NO2 was associated with an annual deficit in FEVi
21 of 25 mL in girls and 20.8 mL in boys. A deficit in FVC and FEVi growth was observed
22 for O3, PMio, and NO2 after adjusting for the acute effect of these pollutants (previous-
23 day concentrations) and for confounding factors. A cohort study in Oslo, Norway,
24 examined short- and long-term NO2 and other pollutant exposure effects on lung function
25 (PEF, forced expiratory flow at 25% of forced vital capacity [FEF25o/0], and forced
26 expiratory flow at 50% of forced vital capacity [FEF50o/0]) in 2,307 nine- and ten-year-old
27 children (Oftedal et al.. 2008). Examining short- and long-term NO2 exposures
28 simultaneously yielded only the long-term effects. Adjusting for a contextual
29 socioeconomic factor diminished the association. The association between long-term
30 exposure to NO2 and decreased PEF was comparable to that found in the CHS, but
31 associations with forced volumes were considerably weaker.
32 Generally, these findings provide further support that early life exposure has long-lasting
33 impact on the lung function development. Effects were mainly on FEVi and FEV0.5,
34 which reflect the mechanical properties of the airways and not as much on FVC,
35 representing lung size.
November 2013 5-25 DRAFT: Do Not Cite or Quote
-------
1 Additionally, cross-sectional studies (Gao et al.. 2013; Svendsen et al., 2012; Lee et al.,
2 201 Id; Rosenlund et al.. 2009b; Tageretal.. 2005: Sekine et al.. 2004; Moseler et al.
3 1994) report associations between NO2 exposure and lowered lung function in children.
November 2013 5-26 DRAFT: Do Not Cite or Quote
-------
Table 5-2 NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Children's Health Study (CHS) California Communities
Breton et al. (2011)
Two cohorts of fourth-
grade children, one in
1993 (cohort 1, n =
1,759) and the second in
1996 (cohort 2, n =
2,004), mean age at
baselinelO.O, were
enrolled and monitored
for 8 years, through
twelfth grade
Central monitoring
stations in each of the
original 12 study
communities from
1994 to the present,
average hourly levels of
NO2 used to compute
annual averages
NO2 and PM25 Hierarchical mixed effects with
= 0.79; adjustments for height, height
NO2 and O3 = squared, body mass index (BMI),
_g n BMI squared, current asthma
status, exercise or respiratory
illness on the day of the test, any
tobacco smoking by the child in
the last year, glutathione S-
transferase mu 1 (GSTM1)
genotype, and indicator variables
for the field technician
Main purpose was to
determine whether
sequence variation in
genes in the glutathione
synthesis pathway alters
susceptibility to air
pollution effects on lung
function
Haplotype "0100000"
was associated with a
39.6 mL, 29.1 mL, and
51.0 mL/sec reduction in
8-year growth
FVC, and MMEF,
respectively
Main study without genetic
effect
NO2 (10ppb)
Change (95% Cl); p=value
FEV-,
-29.83 (-50.10,-9.57)0.004;
FVC
-25.35 (-46.50, -4.20) 0.02;
MMEF
-54.38 (-90.80, -17.96); 0.003
November 2013
5-27
DRAFT: Do Not Cite or Quote
-------
Table 5-2 (Continued): NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Gauderman et al. (2004)
1757 children aged 10 to
18, 8 year follow-up see
cohort one in Breton et
al. (2011)
Air pollution monitoring
stations in 12 study
communities, beginning
in 1994, measured
average hourly levels
NO2 computed annual
averages on the basis of
average levels in a 24-h
period calculated long-
term mean pollutant
levels (from 1994
through 2000) for use in
the statistical analysis of
the lung-function
outcomes
Not Reported 2 stage linear regression
adjusted for log values for height;
body-mass index (the weight in
kilograms divided by the square
of the height in meters); the
square of the body-mass index;
race; the presence or absence of
Hispanic ethnic background,
doctor-diagnosed asthma, any
tobacco smoking by the child in
the preceding year, exposure to
environmental tobacco smoke,
and exercise or respiratory tract
illness on the day of the test; and
indicator variables for the field
technician and the spirometer.
NO2 (10ppb)
Change (95% Cl); p=value
FVC
-27.5 (-54.7, -0.2); 0.05
FEV-,
-29.3 (-47.5, -11.1); 0.005
MMEF
-61.0 (-109.1,-12.8); 0.02
November 2013
5-28
DRAFT: Do Not Cite or Quote
-------
Table 5-2 (Continued): NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Mexico City School Children Cohort
Roias-Martinez et al.
(2007a, b)
Cohort of 3,170 children
aged 8 years at baseline
in 31 schools from April
1996 through May 1999.
Ten air-quality
monitoring stations
within 2 km of the
schools provided
exposure data.
24-h avg NO2
and 8-h avg
O3: 0.166
(p<0.001)
NO2 and
24-h avg PMi0:
0.250
(p = 0.001)
General linear mixed models
were used to evaluate the
association between air pollutant
concentrations and deficits in
lung function growth overtime.
Potential confounding factors
adjusted for include age; body
mass index; height; height by
age; weekday time spent in
outdoor activities; and
environmental tobacco smoke.
A deficit in FVC and FEV1
growth was observed for
O3, PM-io, and NO2 after
adjusting for the acute
effect of these pollutants
(previous-day
concentrations) and for
confounding factors
Annual change as percentage
values per 10-ppb increase in
6-mo mean NO2.
Change (95% Cl); p=value
Multi-pollutant model
(03, PM10, N02)
Girls
FEV-, -0.71 (-1.00,-0.42);
<0.0001
Boys
FEV-, -0.64 (-0.92, -0.37);
<0.0001
Girls
FVC-1.05 (-1.32,-0.77);
<0.0001
Boys
FVC-1.09 (-1.36,-0.82)
<0.0001
FEF25-75% for 12-ppb increase
in 6-mo concentration
Girls
2.5 (-17.5, 12.5)
Boys
6.7 (-20.8, 7.5)
November 2013
5-29
DRAFT: Do Not Cite or Quote
-------
Table 5-2 (Continued): NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Children, Allergy, Milieu, Stockholm, Epidemiology Survey (BAMSE)
Schultzetal. (2012)
In Swedish birth cohort,
1924 school-age children
followed from birth until 8
years of age.
Nordlinq et al. (2008)
The lifetime residential,
day care, and school
addresses were
geocoded, and time-
weighted average
outdoor levels for the
different time windows
were calculated using
emission inventories and
a Gaussian air
dispersion model. Short-
term exposure was
estimated using daily air
quality measurements
and meteorological data
from urban background
and rural monitoring
stations
Linear regression. In addition to
the chosen adjustment
covariates (municipality, sex,
age, height and heredity for
asthma and/or allergy) the
following potential confounders
were evaluated: gestational age,
birth weight, birth length, current
passive smoking, maternal
smoking during pregnancy or at
birth of child, socioeconomic
status of parents, possession of
furred pets at birth and current,
ethnicity, mold and moist in
house during first year of life,
year the house was built, but
none of these were found to
have any influence on the effect
of air pollution.
The odds ratios
associated with 80% and
85% of predicted FEV1
were 2.1 (95% Cl: 0.6,
8.1) and 3.4 (95% Cl: 1.6,
7.4), respectively, for first
year exposure to traffic-
NOX.
Additional adjustment for
temperature, relative
humidity, ozone, and
PM-io levels during 3-7
days before each child's
pulmonary function test
showed little effect on the
estimates of the long-term
effects of air pollution.
Specific adjustment for
NC>2 was not discussed.
No impact of short-term
air pollution exposure on
the estimates of the long-
term effects of air
pollution was found, but
these analyses were
limited to PM-io (in
Stockholm usually
coarse) and O^.
Exposure during the first year
of life was associated with a
deficit of-14.0 mL (95% Cl:
-32.0, 4.1) in FEV-, fora 10
ppb difference in time-
weighted exposure to traffic-
NOx, whereas the
corresponding deficit was 39.6
mL (95% Cl:-67.8,-11.4)
among those sensitized
against any common inhalant
or food allergens, and those
with asthma at 8 years. No
clear effects on lung function
seen after infancy.
November 2013
5-30
DRAFT: Do Not Cite or Quote
-------
Table 5-2 (Continued): NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Oslo Birth Cohort
Oftedal et al. (2008)
In 2001-2002, spirometry
was performed in 2,307
9- and 10-year-old
children who had lived in
Oslo, Norway, since birth.
SI0rdal et al. (2003)
Outdoor air pollution was
modeled by EPISODE, a
dispersion model based
on emissions,
meteorology,
topography, and
background air pollution
concentrations
measured at regional
background stations for
which evaluation
concluded that the
modeled values
represent exposure
reasonably well.
NO2 and PM Multiple linear regression.
r = 0.83 - 0.95 Adjusted for height, age, body
mass index, birth weight,
temperature lags 1-3 days
before the lung function test,
current asthma, indicator for
participation in the Oslo Birth
Cohort, maternal smoking in
early lifetime, parental ethnicity,
education, and smoking. Models
for all children are also adjusted
for sex.
Early and lifetime
exposures to outdoor air
pollution were associated
with reduced peak
expiratory flow and
reduced forced expiratory
flow at 25% and 50% of
forced vital capacity,
especially in girls.
Including short- and long-
term NO2 exposures
simultaneously yielded
only the long-term effect.
No effect on forced
volumes was found.
NO2 in first year of life
Change (95% Cl) per 10 pb
All children:
FEV-,
6.0 (-18.0, 6.1)
FVC
1.4 (-14.6, 11.7)
PEF
57.9 (-92.5, -22.3)
FEF 50%
-37.3 (-71.1,-3.5)
November 2013
5-31
DRAFT: Do Not Cite or Quote
-------
Table 5-2 (Continued): NO2 long-term pulmonary function growth children cohort prospective studies.
Study
Pollutant
Exposure Correlation Statistical methods Comments
Results:
FEVi and FVC (mL)
MMEF, PEF, and FEF
(mL/sec)
Manchester Asthma and Allergy Study (MAAS)
Molteretal. (2013)
Molteretal. (201 Oa)
Molteretal. (201 Ob)
Molteretal. (2012)
N = 1185.
Manchester, U.K.
Birth cohort recruited
between 1995 and 1997
and evaluated at ages 3.
5. 8 and 11 years.
Land use regression
model with modeled
exposure at the
individual level rather
than community level.
Using the
Microenvironmental
Exposure Model that
incorporates children's
time-activity patterns to
predict personal
exposure, the
performance of which
was found to be in
agreement between
modeled and measured
NO2 concentrations.
Not reported Generalized estimating
equations were used to analyze
the association between lifetime
exposure and the development
of lung function. Potential
confounding variables and
covariates evaluated included
sex, age, ethnicity, older siblings,
sensitization, asthma or current
wheeze, family history of asthma,
parental smoking, parental atopy,
daycare attendance during the
first two years of life,
hospitalization during the first two
years of life, presence of a gas
cooker in the home, presence of
a dog or cat in the home, visible
signs of dampness or mould in
the home, body height, body
weight, body mass index,
maternal age at birth, gestational
age, duration of breast feeding,
Tanner stage (age 11 only) and
socioeconomic status (paternal
income).
A negative association
between post
bronchodilator FEVi and
PM-io and NC>2 exposure
overtime:
(PM10: (3 = -3.59 [95% Cl:
-S.36,-1.83];
NO2: (3 = -1.20,
[95%CI:-1.97, -0.43] per
1 ug/m3).
Based on the average
predicted FEV-i of 1.65 L
(see above), these would
be equivalent to a growth
deficit in post
bronchodilator FEVi of
59 mL from age 5 years
to 11 years per unit
increase in PM-io, and a
growth deficit of 20 mL
from age 5 years to 11
years per unit increase in
NO2.
ForFEV-,
NC>2 exposures were
associated with poorer lung
function overtime:
(PMi0: (3 = -1.37[95%CI:
-2.52, -0.23];
NO2: (3 = -0.83[95%CI:
-1.39,-0.28] perl ug/m3).
Based on the average
predicted FEVi within MAAS
at ages 5, 8 and 11 of 1.65 L,
the model estimated that for
each unit increase (1 ug/m3) in
PM-io exposure, the growth in
FEVi from age 5 years to 11
years was 23 mL smaller, and
for each 10-ppb increase in
NO2 exposure, the growth in
FEVi was 263 mL smaller
[AFEV-i (mL)
= P/100*1.65*1000].
November 2013
5-32
DRAFT: Do Not Cite or Quote
-------
exposure
Study method
Breton et al., central
201 1 Southern California monitoring stations
Gauderman etal., central
2004 Southern California monitoring stations
Rajas-Martinez et al . monitoring stations
2007a,b Mexico City. Mexico within 2 km of schools
SchultZ et al . , Modeled lifetime
2012 Stockholm. Sweden alresidence
Oftedal et al.. outdoor modeled
2008 Oslo, Norway °y dispercron model
Molter et al., dispersion model
2013 Manchester, England estimates of microenwonment
age in type of lung
comment years function metric
18 FEV, 0
(0.6)
FVC +
— •—
10 FEV, «
to 18
FVC — «—
MMEF ( »
girls 11 FEV, *
boys FEV, *•
gills FVC »
boys FVC +
girts FEF,,,, -
boys FEF^.,,
all 8 FEV, (
children
sensitized 8 FEV, •
at age 8
FEV, (
(rat)
FVC (mL) 1
PEF (ml/sec) •••
(mUsec)
3 FEV, 4
to 11
I I
«-
>-
i
1 1
-100 -50 0 50 100
lung function metric slopes per 10 ppb NO2 or NOX
Note The studies are presented by age. Follow-up was 3 years for Roias-Martinez et al. (2007a). and 10 years for Oftedal et al.
(2008). All other studies were 8 years of follow-up.
Figure 5-5 Associations of NO2 or NOX with lung function indices from
prospective studies.
November 2013
5-33
DRAFT: Do Not Cite or Quote
-------
Adults
1 Prospective studies evaluating long-term NO2 exposure and pulmonary function consist
2 of a European multi city study and a study in the U.K. that are discussed next. Cross-
3 sectional studies were reviewed but not presented here are (Forbes et al., 2009b; Sekine et
4 al.. 2004).
5 Gotschi et al. (2008) examined the relationship between air pollution and lung function in
6 adults in the European Community Respiratory Health Survey (ECRHS). FEVi and FVC
7 were assessed at baseline and after 9 years of follow-up from 21 European centers
8 (followed-up sample n = 5,610). No statistically significant associations were found
9 between city-specific annual mean NO2 and average lung function levels which is in
10 contrast to the results seen by Ackermann-Liebrich et al. (1997) (SAPALDIA) and
11 Schikowski et al. (2005) (SALIA) which compared across far more homogenous
12 populations than for the population assessed in the ECRHS. Misclassification and
13 confounding may partially explain the discrepancy in findings.
14 A recent study Boogaard et al. (2013) evaluates the impact on pulmonary function of a
15 reduction in outdoor pollution concentrations resulting from the policy implementation of
16 forbidding old heavy duty vehicles in all inner cites and other related policies. At 12
17 locations in the Netherlands, air pollution concentrations and respiratory health were
18 measured in 2008 and 2010, indicating a reduction over that time period. NO2 and NOX
19 concentrations were measured with Ogawa passive samplers. In regression analyses
20 adjusted for important covariates, reductions in concentrations of soot, NO2, NOX, Cu,
21 and Fe were associated with increases in forced vital capacity (FVC) (increase per
22 interquartile range [IQR] decline). Airway resistance decreased with a decline in
23 particulate matter ([PM10] and PM25 [per IQR]), although these associations were
24 somewhat less consistent. No associations were found with exhaled NO. The response
25 rate for participation in the study was around 10%. Over the two time periods 585
26 subjects were reevaluated for spirometry. This was a heterogeneous study population
27 with both children and adults were 84% were above 30 years of age at baseline. Results
28 were driven largely by the small group of residents living at the one urban street where
29 traffic flow as well as air pollution were drastically reduced.
30 Both (1) cross-sectional associations with bronchial hyperresponsiveness, FEVi,
31 spirometry defined COPD, skin test positivity, total IgE and questionnaire-reported
32 wheeze, asthma, eczema and hay fever in 2,599 subjects, and (2) longitudinal
33 associations with decline in FEVi in 1,329 subjects followed-up nine years later in 2000
34 were evaluated in a Nottingham U.K. cohort of adults aged 18-70 in relation to modeled
35 outdoor NO2 concentrations (Pujades-Rodriguez et al., 2009). There were no significant
36 cross-sectional associations between home proximity to the roadside or NO2 level with
November 2013 5-34 DRAFT: Do Not Cite or Quote
-------
1 any of the outcomes studied. Also, neither exposure was associated with a significantly
2 greater decline in FEVi over time. Insufficient contrast in exposure may be a factor why
3 this study was unable to detect any effects of localized traffic pollution markers in this
4 study population were the modeled NO2 variation ranged from values for IQR of 18.1 to
5 19.1ppb.
5.2.3.2 lexicological Studies
6 Studies included in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) showed
7 minimal evidence of changes in lung function in animals after long-term exposure to
8 concentrations of NO2 relevant to ambient exposure .No recent studies were available.
9 Arner and Rhoades (1973) published early studies that exposed rats to 2,900 ppb NO2
10 continuously for 5 days per week for 9 months and reported changes in lipid composition
11 in the airway and that could be related to observed functional consequences, including
12 decreased lung volume and compliance and increased surface tension, though these
13 changes have not been consistently observed in animals studies.
14 Tepperetal. (1993) exposed rats to a background concentration of 500 ppb NO2 for 16
15 hours per day followed by a 6 hour peak of 1,500 ppb and 2 hours of downtime for up to
16 78 weeks. Frequency of breath was significantly slower in these animals and was
17 paralleled by a trend toward increased tidal volume, expiratory resistance, and inspiratory
18 and expiratory time, though changes were not significant. Mercer etal. (1995) and Miller
19 et al. (1987) published studies with similar exposures in rats and mice, respectively, and
20 also reported that NO2 exposure did not alter lung function, though mice tended to have
21 slightly decreased vital capacity from 16 to 52 weeks of exposure.
22 Minimal effects were also described in studies of shorter, yet still long-term, NO2
23 exposure. Stevens etal. (1988) exposed 1 day and 7 week old rats to 500, 1,000, and
24 2,000 ppb NO2 continuously with two daily peaks at three times the baseline
25 concentration (1,500, 3,000, and 6,000 ppb) for 1-7 weeks and observed different results
26 among age groups. Rats exposed from 1 day of age had increased compliance after 3
27 weeks of exposure that returned to control levels by 6 weeks (1,000 ppb with 3,000 ppb
28 peaks). In rats exposed from 7 weeks of age, compliance was decreased after 6 weeks of
29 exposure at 1,000 and 2,000 ppb NO2. In an 8 week study, Lafumaetal. (1987). reported
30 increased lung volumes in animals exposed to 2,000 ppb (8 hours per day, 5 days per
31 week), but vital capacity and compliance were not affected.
32 Overall, these animal studies demonstrate minimal effects of long-term NO2 exposure on
33 pulmonary function in animals which is consistent with results from short-term
November 2013 5-35 DRAFT: Do Not Cite or Quote
-------
1 exposures; however, age may be an important factor that has not been adequately
2 addressed by the existing body of toxicological evidence.
5.2.4 Hospital Admissions
3 Recent studies represent the first evaluation of long-term NO2 exposure for respiratory
4 hospital admissions. Studies include hospitalization for chronic obstructive pulmonary
5 disease (COPD), asthma, and community-acquired pneumonia. The relationship between
6 long-term pollutant exposures on the development of COPD was evaluated by Andersen
7 et al. (2011) in a prospective cohort study. COPD incidence in relation to estimated
8 modeled outdoor annual means of NO2 and NOX from residential address history since
9 1971 was studied with respect to the first admission to hospital for COPD in 57,053
10 participants (median age 56.1) in the Danish Diet, Cancer, and Health cohort in the
11 Hospital Discharge Register between 1993 and 2006. The incidence (date of first
12 admission) of COPD between baseline and 27 June 2006 was the main outcome. After
13 adjustment for occupation, educational level, body mass index, and fruit intake,
14 associations with COPD incidence remained for the 35- and 25-year mean levels of NO2
15 (HR: 1.28; [95% CI: 1.07, 1.54] and 1.22 [95% CI: 1.03, 1.45], per 10-ppb increase) and
16 35-year mean level of NOX (1.16 [95% CI: 1.04, 1.31], per 20-ppb increase), whereas
17 weak positive associations were observed with the 25-year mean level of NOX, 15-year
18 mean levels of NO2 and NOX, and baseline residence traffic proxies (major road within
19 50 meters, traffic load within 200 meters). The associations with NO2 were stronger than
20 those with NOX. Additional adjustment for sex, physical activity, alcohol consumption,
21 and vegetable consumption did not substantially change the risk estimates. The effect was
22 stronger in people with diabetes and asthma compared to the rest of the cohort. The
23 strongest association with COPD incidence was found with the longest exposure,
24 reinforcing the conclusion that exposure over a long period, perhaps over a whole life, is
25 relevant for the development of chronic lung diseases such as COPD. No evidence was
26 found that the effect was modified by smoking or occupational exposure.
27 A study also examining long-term NO2 exposure and COPD hospitalization was
28 conducted in Canada. Gan et al. (2013) evaluated a population-based cohort that included
29 a 5-year exposure period and a 4-year follow-up period. All residents aged 45-85 years
30 who resided in Metropolitan Vancouver, Canada, during the exposure period and did not
31 have known COPD at baseline were included in this study (N = 467,994). Residential
32 exposures to traffic-related air pollutants (black carbon [BC], PM2 5, NO2, and NO) and
33 wood-smoke were estimated using land-use regression models and integrated changes in
34 residences during the exposure period. COPD hospitalizations during the follow-up
35 period were identified from provincial hospitalization databases. Mortality data was also
November 2013 5-36 DRAFT: Do Not Cite or Quote
-------
1 studied and is discussed in Section 4.2. In unadjusted single-pollutant models, NO2 and
2 NO were associated with COPD hospitalization. However, after adjustment for
3 covariates, the association of these air pollutants with COPD hospitalization was
4 attenuated which may reflect the lack of spatial variability of these pollutants in this intra-
5 urban study. Additionally, the shorter exposure period examined in this study compared
6 with that in Andersen et al. (2011) may be a factor in their different results.
7 The relationship between long-term pollutant exposures on the risk for asthma
8 hospitalizations in older people was also evaluated in the Danish Diet, Cancer and health
9 cohort (Andersen et al.. 2012). Associations between NO2 level and first hospital
10 admission were found in the full cohort (HR per IQR 10 ppb: 1.44 [95% CI: 1.14, 1.84]),
11 which was insensitive to choice of potential confounders. The associations were similar
12 for the first asthma hospitalization (1.36 [95% CI: 1.03, 1.80]), but markedly higher for
13 re-hospitalization in people with a previous asthma hospitalization (3.05 [95% CI:
14 1.57-5.90], p = 0.05, Wald test for interaction). The risk for asthma hospitalization
15 associated with NO2 was four times higher in people with previous COPD
16 hospitalizations (2.34 [95% CI: 1.25, 4.40], with effect modification [p = 0.04]). Some of
17 the observed effects could possibly be ascribed to the short-term effects of increases in air
18 pollution on the days prior to asthma admission. The dispersion model estimating NO2
19 levels may have been poorer longer back in time, owing to the higher uncertainty of
20 model input data such as emission factors and traffic counts. Thus, the stronger
21 associations with NO2 levels at follow-up, when compared with the 35-, 15- and 1-year
22 means at baseline may reflect the relevance of more recent exposures for asthma
23 hospitalization or merely the better performance of the dispersion model in more recent
24 years.
25 In an ecological time series study, Delamater et al. (2012) explored the relationship
26 between asthma morbidity, ambient levels of air pollutants, and weather conditions at a
27 county level using a monthly time series analysis. In LA County, they found that asthma
28 hospitalizations were associated with CO, NO2, and PM2 5 concentrations in single
29 variable regression models and NO2 + relative humidity, PM2 5 +relative humidity,
30 PM10 + relative humidity, and NO2 + maximum temperature in multi-variable models. In
31 a cross-sectional study, Meng et al. (2010) examined associations between air pollution
32 and asthma morbidity in the San Joaquin Valley in California using the 2001 California
33 Health Interview Survey data from subjects ages 1 to 65+ who reported physician-
34 diagnosed asthma (n = 1,502). Subjects were assigned annual average concentrations for
35 NO2 based on residential ZIP code and the closet air monitoring station within 8 km but
36 did not have data on duration of residence. No associations were found between average
37 annual concentrations of NO2 and odds of asthma-related ED visits or hospitalizations.
38 No quantitative results were shown for NO2.
November 2013 5-37 DRAFT: Do Not Cite or Quote
-------
1 In a population-based case-control study in Hamilton, Ontario, Canada, Neupane et al.
2 (2010) examined the relationship between the previous 12 months exposure to ambient
3 pollutants (NO2, PM2 5, SO2) and hospitalization for community-acquired pneumonia in
4 345 hospitalized patients aged 65 years or more and 494 control participants aged 65
5 years and more. Control participants were randomly selected from the same community
6 as cases from July 2003 to April 2005. Three methods were used to estimate the annual
7 average NO2 levels: daily ambient data, LUR models and inverse distance weighting.
8 Participants had to present to the emergency room with at least two signs and symptoms
9 for pneumonia and have a new opacity on a chest radiograph interpreted by a radiologist
10 as being compatible with pneumonia. Various interaction terms were evaluated but none
11 significantly interacted with any air pollutant variant and therefore were not included in
12 the logistic regression model. Covariates evaluated included: age, male sex, education,
13 smoking, and history of exposure to fumes at work. NO2 and PM25 were associated with
14 hospitalization for community-acquired pneumonia but SO2 was not. For NO2, all three
15 exposure estimate methods indicated an association. While no mention of control or
16 adjustment for short-term exposure effects was discussed, two recent unrelated short-term
17 exposure studies of hospitalization for pneumonia (Chiu et al.. 2009; Cheng et al.. 2007)
18 report smaller effect associations with NO2 and pneumonia hospitalization than this long-
19 term study which is as expected.
5.2.5 Respiratory Symptoms
5.2.5.1 Children
20 In the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). there was limited evidence,
21 consisting of a single prospective cohort study and several cross-sectional studies, to
22 support an association between long-term exposure to NO2 and respiratory symptoms.
23 That review considered the results to be inconsistent, mainly due uncertainties inherent to
24 the cross-sectional studies. Recent prospective cohort studies evaluating the relationship
25 between respiratory symptoms and long-term exposure to NO2 are summarized in Table
26 5-3 and the following text. Cross-sectional studies were reviewed and are discussed in
27 other sections as appropriate (Annesi-Maesano et al.. 2012a: Ghosh etal.. 2012b: Dong et
28 al., 2011; Mi et al., 2006; Pattenden et al., 2006; Nicolai etal.. 2003; Brauer etal.. 2002;
29 Gehring et al.. 2002; Zempetal. 1999) and in the Annex Table AX6.3-17 of the 2008
30 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
31 A number of studies have observed an association between different respiratory
32 symptoms in children with asthma and long-term exposure to NO2. McConnell et al.
November 2013 5-38 DRAFT: Do Not Cite or Quote
-------
1 (2003) examined children with asthma for bronchitic symptoms, including daily cough
2 for 3 months in a row, congestive phlegm 3 months in a row, or bronchitis, all
3 representing chronic indolent symptoms. The remaining studies focused on asthmatic
4 symptoms. Belanger et al. (2013) recorded wheeze, night symptoms, rescue medication
5 use and an asthma severity score which consist of symptoms and medication use based on
6 the Global Initiative for Asthma (NHLBI. 2002). Hansel et al. (2008) examined
7 wheezing, coughing, chest tightness (activity limiting or while running), wheezing so bad
8 that speaking is difficult, and nocturnal cough without a cold. Gehring etal. (2010)
9 examined asthma symptoms (one or more attacks of wheeze, SOB, prescription of
10 inhalation steroids), wheeze (transient, late onset, persistent), sneezing, hay fever, atopic
11 eczema, and prevalence of asthma. Gruzieva et al. (2013) examined wheeze, categorized
12 as either one or more episodes or three or more episodes in the past year. These
13 approaches to examining respiratory symptoms in asthmatics provide different
14 information from which to consider the impact of long-term NO2 exposure on respiratory
15 health.
16 It is clear that the respiratory symptoms examined vary among these studies and are
17 measured in a different ways. McConnell et al. (2003) is unique in that it is the only
18 prospective study examining bronchitic symptoms in children with asthma for which they
19 report positive results that are stronger in their within community analysis as compared to
20 the between community analysis. Hansel et al. (2008) finds positive results for all of the
21 symptoms evaluated, and noted that symptoms of a more severe nature may produce
22 stronger relationships to NO2 exposure. In a threshold model, Belanger et al. (2013)
23 report positive associations for the asthma severity score, wheeze, night symptoms, and
24 rescue medication use. These results were generally larger than those reported in the
25 other studies. Gehring etal. (2010) observed positive results for prevalent asthma, asthma
26 symptoms, and wheeze; but not for hay fever or atopic eczema. Gruzieva etal. (2013)
27 report a weak association with wheeze.
28 The respiratory symptoms evaluated in this group of studies vary and some health effect
29 indicators involve several components. The ability of these various indicators to serve as
30 a measure for respiratory health effects may differ. Some indicators may be better than
31 others.
32 The collective evidence from this group of prospective studies provides results that could
33 be cautiously viewed to be supportive of a relationship of long-term exposure to NO2 and
34 increased respiratory symptoms using various indicators in children with asthma. This is
35 especially so, for studies that involve a group of health indicators, as they may be more
36 able to inform the asthmatic relation as opposed to a more sparse approach, i.e., just
November 2013 5-39 DRAFT: Do Not Cite or Quote
-------
1 wheeze. Thus weighing the multiple approaches as potentially being more informative
2 might indicate a stronger less uncertain relationship in this evidence base.
3 Some studies examined respiratory health to include symptoms and respiratory infection
4 in infants and young children up to three years of age (Aguilera et al., 2013;
5 Sonnenschein-Van der Voort et al.. 2012; Raaschou-Nielsen et al.. 201 Ob: Sunyer et al..
6 2004). Both Sunver et al. (2004) and Raaschou-Nielsen et al. (201 Ob) found no
7 associations in infants between indoor NO2 exposure and respiratory infections and
8 wheezing respectfully. Sonnenschein-Van der Voort et al. (2012) only found associations
9 of increased risk in wheezing in children up to age three years exposed to tobacco smoke.
10 Aguilera et al. (2013) related NO2 exposure to increased risk of upper and lower
11 respiratory tract infections in infants.
November 2013 5-40 DRAFT: Do Not Cite or Quote
-------
Table 5-3 Long-term NO2 exposure prospective children studies: respiratory symptoms.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results
Childrens Health Study (CHS) Southern California communities
McConnell et al. (2003)
Children with a history of
asthma at study entry, who
completed two or more
follow-up questionnaires
any time during the years
1996to1999(n = 475)
were included in the
current analysis. Yearly
follow-up for 4 years.
Air pollution monitoring
stations were
established in each of
the 12 communities.
Annual averages were
computed of the 24-h
NO2.
Four-year mean levels
(1996-1999) in each
community were
computed for each
pollutant metric.
The correlations of 4-
year average pollution
across the 12
communities were
relatively high with each
other (R >0.65) except
with Os which was low.
The within-community
correlations differed in
that NO2 could be
distinguished from most
other major pollutants
except OC and
inorganic acid.
Examined the
associations of
bronchitic symptoms
both with yearly
variation in air pollution
within communities and
with the 4-year average
of air pollutants across
the 12 study
communities.
The modeling strategy
can be conceptualized
as a three-stage
regression.
OR adjusted for age,
maternal and child's
smoking history, sex,
and race; within-
community estimates
were adjusted for
between-community
effects of the pollutant,
and vice versa.
In two pollutant models,
the effects of yearly
variation in OC and NO2
were only modestly
reduced by adjusting for
other pollutants, except
in a model containing
both OC and NO2;
McConnell et al. (2006)
further found that this
cohort result was
modified by dog or cat
ownership indicators or
allergen and endotoxin
exposure as the odds
ratio for NO2 was 1.49
(95% Cl: 1.14, 1.95),
indicating that dog
ownership may worsen
the relationship between
air pollution and
respiratory symptoms in
asthmatic children,
Within communities
NO2(OR: 1.97 [95%
Cl: 1.22, 3.18] per
10ppb);
ORs associated with
yearly within-community
variability in air pollution
were larger than the
effect of the between-
community 4-year
average concentrations.
Between communities
(NO2OR: 1.22[95%CI:
1.00, 1.49] per 10 ppb).
November 2013
5-41
DRAFT: Do Not Cite or Quote
-------
Table 5-3 (Continued): Long-term NO2 exposure prospective children studies: respiratory symptoms.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results
Prevention and Incidence of Asthma and Mite Allergy (PIAMA) study, the Netherlands
Gehrinq et al. (2010)
n = 3,863 Children in birth
cohort aged 1 to 8 years.
Land-use regression
models used to provide
annual levels for birth
address of each
participant.
The estimated
exposures for the
various pollutants were
highly correlated (r =
0.93, 0.96, and 0.97 for
the correlation between
NO2 and PM2.5, NO2
and soot, and PM2.5
and soot, respectively.
Generalized estimation
equations.
Potential confounding
variables included sex,
study arm (intervention
or natural history), use
of mite-impermeable
mattress covers,
allergies of mother and
father, maternal and
paternal education,
maternal smoking
during pregnancy,
breastfeeding,
presence of a gas stove
in the child's home,
presence of older
siblings, smoking, signs
of dampness and pets
in the child's home,
day-care attendance,
and Dutch nationality.
No associations were
found with atopic
eczema, allergic
sensitization, and
bronchial
hyperresponsiveness.
Adjusted OR for asthma
symptoms of 1.17 (95%
Cl: 0.98, 1.39) per
10-ppb increase without
adjustment for study
region.
For wheeze OR: 1.27
(95% Cl: 1.07, 1.50)
without adjustment for
study region.
Children, Allergy, Milieu, Stockholm, Epidemiology Survey (BAMSE)
Gruzieva et al. (2013)
Swedish birth cohort
followed up to 12 years of
age enrolled between 1994
and 1996 n = 3,633.
Nordlinq et al. (2008).
Melen et al. (2008)
Dispersion models were
used to calculate for all
addresses in the years
1994 to 2008
representing when the
first child was born until
the end of the 12-year
follow-up.
r= 0.96 between NOX
and PM-io exposure
levels during the first
year of life.
Multinomial regression/
GEE.
Potential confounders
adjusted for include
municipality, SES, year
the house was built,
and heredity.
No overall association
was observed between
air pollution exposure
after the first year of life
and development of
asthma symptoms.
Wheeze at 12 years of
age OR 3 or more
episodes: 1.35 (95% Cl:
0.79, 2.29) per 20 ppb
for NOX.
November 2013
5-42
DRAFT: Do Not Cite or Quote
-------
Table 5-3 (Continued): Long-term NO2 exposure prospective children studies: respiratory symptoms.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results
Longitudinal New England Indoor Children's Asthma Study
Belanqeret al. (2013)
Connecticut/Massachusetts
cohort, n = 1,642, age
5-10 followed for one year,
asthma severity score from
2006 through 2009
Palmes tubes in
bedrooms and dayroom
for 4 weeks for 4
seasons
Not Reported
Adjusted, hierarchical
ordered logistic
regression models used
Adjustments included
age, sex, atopy, season
of monitoring,
race/ethnicity, mother's
education, smoking in
the home, and all five
variables for combined
specific sensitization
and exposure to indoor
allergens
Included maintenance
medication use as a
covariate in models
exploring associations
between symptoms and
NC>2 exposure. Because
use of maintenance
medication is also
associated with
socioeconomic status
An alternative model
that adds only "residual"
amounts above what is
measured indoors was
considered. In this
alternative model, where
only "extra" NO2 not
accounted for in the
indoor measurement is
added, the OR for
indoor NC>2 exposure on
the asthma severity
score is 1.52 (1.06-
2.18), and the OR for
outdoor NO2 exposures
is 1.20
(0.98-1.46).
Every 5-fold increase in
NO2 exposure above a
threshold of 6 ppb was
associated with a dose-
dependent increase in
risk of higher asthma
severity score
(ORs: 1.37 [95% Cl:
1.01, 1.89]; Wheeze:
1.49 [95% Cl: 1.09,
2.03]; Night symptoms:
1.52 [95% Cl: 1.16,
2.00]; and Rescue
Medication Use: 1.78
[95% Cl: 1.33,2.38]).
November 2013
5-43
DRAFT: Do Not Cite or Quote
-------
Table 5-3 (Continued): Long-term NO2 exposure prospective children studies: respiratory symptoms.
Study
Exposure
Pollutant Correlation Statistical Methods
Comments
Results
African-American Baltimore School
Hansel et al. (2008)
n = 150, 2-6 years of age
with physician diagnosed
asthma.
Indoor air was
monitored over a 72-h
period in the children's
bedrooms at baseline
and 3- and 6-mo.
There was minimal
correlation (R2 = 0.056,
p <0.01) between
ambient and indoor
NO2 concentrations.
Logistic regression
models and GEE
Multivariate models to
adjust for potential
confounders, including
age, sex, race,
caregiver education
level, season of
sampling, PlVb.s,
secondhand smoke
(SHS) exposure
[defined as caregiver
report of presence of a
smoker in the home
(yes vs. no)], distance
from curb, and type of
street in front of house.
This longitudinal study
with repeated measures
of NO2 concentrations
and respiratory
symptoms improves on
the ability to directly
model individual
response to changing
NC>2 concentrations
accounting forwithin-
person correlations of
asthma severity. The
link between indoor NC>2
concentrations and
asthma symptoms
appears to be robust,
because the
associations were not
affected by the potential
confounders studied.
This study was
strengthened by its
ability to adjust for other
relevant copollutants.
Although PlVb.s was
associated with
increased asthma
symptoms [(McCormack
et al.. 2008). data not
shown], adjusting for
other copollutants did
not meaningfully alter
the association between
indoor NC>2
concentrations and
asthma symptoms.
Adjusted IRR per
10-ppb increase in NC>2
exposure;
Daytime wheezing,
coughing, or chest
tightness 1.02 (0.98,
1.06);
Slowing activity due to
asthma, wheeze, chest
tightness, or cough:
1.04(0.97, 1.11);
Limited speech due to
wheeze:
1.08(1.04, 1.12);
Wheeze, cough, or
chest tightness while
running:
1.04(1.00, 1.08*);
Coughing without a cold
1.07(1.03, 1.11);
Nocturnal awakenings
due to cough, wheeze,
shortness of breath, or
chest tightness:
1.06(1.02, 1.10)
November 2013
5-44
DRAFT: Do Not Cite or Quote
-------
5.2.5.2 Adults
1 Studies examining the relationship between long-term NO2 exposure and respiratory
2 symptoms in adults include prospective studies discussed under asthma incidence.
3 Jacquemin et al. (2009b) report that all the associations between NO2 and asthma
4 symptoms at ECHRS II were positive; the strongest was for waking "with a feeling of
5 tightness in the last 12 months." Results were homogeneous among the centers in both
6 the crude and the adjusted analyses. Symptoms in the last 12 months at ECRHS II among
7 people without asthma at baseline were also associated with NO2. The authors note that
8 these observations are, on the one hand, complementary to and in strong support of main
9 findings on asthma incidence. On the other hand, due to the study design, the symptom
10 results also call for a partly different interpretation. The standard questionnaire asks about
11 symptoms during the last 12 months only. Thus, people with new asthma onset who did
12 not suffer symptoms during the last 12 months (e.g., due to treatment) would not be
13 captured. Instead, a subject without asthma who reported symptoms during the last 12
14 months (e.g., due to some infection) would be identified as an incident case. Moreover,
15 air pollution is a known trigger of several asthma-related symptoms. Thus, reporting of
16 symptoms at ECRHS II (but not at ECHRS I) may not necessarily reflect the onset of
17 asthma due to air pollution, but represent the acute effects of air pollution exposure
18 during the past 12 months. The main approach using asthma incidence is less affected by
19 these methodological issues. Cases diagnosed during the entire follow-up period
20 contributed to these findings, independent of symptom status during the last 12 months.
21 Zemp et al. (1999) report, in a cross-sectional study (SAPALDIA), an NO2 association
22 with high prevalence of respiratory symptoms in adults. Bentayeb et al. (2010) report the
23 first baseline cross-sectional study results examining the respiratory symptoms cough and
24 phlegm in adults (> 65 years old, in Bordeaux, France) in relation to NO2 exposure to be
25 of borderline significance.
November 2013 5-45 DRAFT: Do Not Cite or Quote
-------
5.2.6 Allergic sensitization
5.2.6.1 Epidemiologic Studies
Children
1 Recent cross-sectional studies evaluate aspects of allergic responses and long-term
2 exposure to NO2. In the Munich metropolitan area, a study population consisting of two
3 prospective birth cohort studies (German Infant Nutritional Intervention -GINI and
4 Lifestyle-Related Factors on the Immune System and the Development of Allergies in
5 Childhood study-LISA) was used to evaluate the relationship between individual-based
6 exposure to traffic-related air pollutants and allergic disease outcomes during the first
7 6 years of life (Morgenstern et al., 2008). Exposure assessment was calculated at three
8 different time points (birth, 2 or 3 years, and 6 years). Positive, but imprecise,
9 associations were observed between doctor-diagnosed asthma and parental reporting of
10 symptoms and long-term exposure to NO2. Previous analyses of the LISA and GINI
11 cohorts at age 1 reported positive associations between PM25, PM25 absorbance, and
12 NO2, and cough without infection, and dry cough at night. At age 2 years, these effects
13 were attenuated (Morgenstern et al., 2007; Gehring et al., 2002). NO2 was positively
14 associated with eczema. For a 10-ppb increase in NO2, the association for allergic
15 sensitization at 6 years of age was 1.07 (95% CI: 0.72, 1.59) for any inhalant, 1.00 (95%
16 CI: 0.72, 1.59) for outdoor allergens, and 0.89 (95% CI: 0.40, 2.00) for indoor allergens.
17 Annesi-Maesano et al. (2007) related individual data on asthma and allergy from 5,338
18 school children (10.4 ± 0.7 years) attending 108 randomly chosen schools in six French
19 cities to the concentration of NO2 measured in school yards with passive diffusion
20 samplers and at the city level at fixed-site monitoring stations. NO2 was positively
21 associated with exercise-induced bronchial (EIB) reactivity, flexural dermatitis and skin
22 prick test (SPT) to indoor allergens. Hwang et al. (2006) report the prevalence of allergic
23 rhinitis (adjusted OR per 10 ppb NO2 = 1.11 [95% CI: 1.08-1.15]) in a large cross-
24 sectional study of school children in Taiwan. Parker et al. (2009) evaluated the
25 association between air pollutants and childhood respiratory allergies in the U.S. using
26 the 1999-2005 National Health Interview Survey of approximately 70,000 children and
27 found no associations between NO2 and the reporting of respiratory allergy/hay fever.
28 Strong positive associations were found for O3.
29 Nasal eosinophils, which participate in allergic disease, were observed to decrease by
30 fourfold in 37 atopic mildly asthmatic children 7 days after relocation from a highly
31 polluted urban area (NO2 51.8 ± 6.3 ppb) in Italy to a rural location with lower NO2
November 2013 5-46 DRAFT: Do Not Cite or Quote
-------
1 levels (NO2 3.5 ± 0.27 ppb) (Renzetti et al.. 2009). Living in a less-polluted environment
2 was also associated with reduced eosinophilic inflammation of the lower airways,
3 reflected by a decrease in mean FENO (fraction of NO in exhaled air) concentration and
4 with consistent improvement in lower function reflected by an increase in mean PEF.
5 Nordling et al. (2008). discussed in Table 5-1. Table 5-2. and Table 5-3. reported that
6 exposure to NO2 from traffic during the first year of life was associated with IgE-
7 antibodies for pollens (OR = 1.24 (1.04-1.49), per 10-ppb increase in NO2) but other
8 measures were not. The relationship between the development of allergic sensitization in
9 children during the first 8 years of life and long-term exposure to NO2 was evaluated in a
10 prospective analysis of the BAMSE cohort (Gruzieva et al.. 2012). There was no overall
11 risk of sensitization at 4 years of age associated with traffic-related air pollution
12 exposure. However, exposure during the first year of life was associated with an
13 increased risk of sensitization to pollen (OR: 1.64[95%CI: 1.02, 2.63, per 20-ppb
14 increase) for traffic-related NOX. On the other hand, there was no apparent effect of
15 exposure to air pollution after the first year of life on the development of sensitization.
16 In a cross-sectional analysis in six French cites, Annesi-Maesano et al. (2012a) evaluated
17 the relationship between indoor air quality in schools and the allergic and respiratory
18 health of schoolchildren (mean age 10.4). For each pollutant, a 5-day mean concentration
19 in the classroom was computed and a three-class variable of exposure (high, medium, or
20 low) was defined with respect to the tertiles of concentration in the class room,
21 independent of the city. Between-school and within-school variability of the measured
22 indoor pollutants were estimated using linear mixed models for longitudinal data. Among
23 atopic children (n = 1,719) NO2 was related to past year allergic asthma after adjusting
24 for the potential confounders age, sex, passive smoking and parental or maternal history
25 of asthma and allergic diseases (OR: 1.40, p = 0.0514).
26 In a cross-sectional analysis of 30,139 Chinese children aged 3-to-12 years, Dong et al.
27 (2011) evaluated the relationship between 3-year averages of ambient pollutants (PM10,
28 SO2, NO2, CO, and O3) and asthmatic symptoms (persistent cough, persistent phlegm,
29 doctor-diagnosed asthma, current asthma, current wheeze, and allergy rhinitis). The study
30 examined 25 districts of seven cities in northeast China in 2009 and also investigated
31 whether allergic predisposition modifies this relationship. The data confirm that ambient
32 compound air pollution was associated with respiratory symptoms and diseases in young
33 children. Among children without an allergic predisposition, males might be more
34 susceptible to ambient air pollution than females; whereas among children with an
35 allergic predisposition, more associations were detected in females in this group. Among
36 children without allergic predisposition (n = 26,004) several NO2 effects were obtained
37 mainly in males in single pollutant models. In multipollutant models that evaluated the
November 2013 5-47 DRAFT: Do Not Cite or Quote
-------
1 five pollutants, persistent cough effects increased for NO2 for both males and females;
2 however, because of potential multicollinearity among more than two pollutants, these
3 resutlts are difficult to interpret. Among children with allergic predisposition (n = 4,135),
4 Several NO2 analyses that were positive in the single pollutant models for both males and
5 females were attenuated in the five pollutant model for those with allergic predisposition.
6 This differs from the results without allergic predisposition as further discussed in
7 Section 5.2.13.
Adults
8 Several recent studies examined the association between allergic responses and exposure
9 to NO2, and report generally inconsistent results. Castro-Giner et al. (2009). discussed in
10 adult asthma incidence (Section 5.2.2.2). noted that stratification by atopic status showed
11 that interaction between NO2 and NQO1 rs2917666 was more pronounced among
12 carriers ofNQOl rs2917666 C/C without atopy (OR: 5.10 [95% CI: 1.26, 20.70];/?-value
13 for interaction = 0.01, per 10-ppb increase) compared with subjects with atopy (OR: 1.50
14 [95% CI: 0.72, 3.12];/?-value for interaction = 0.45). Puiades-Rodriguez et al. (2009)
15 examined a cohort of 2,644 adults aged 18-70 living in Nottingham, U.K. to evaluate the
16 relationship between NO2 exposure and respiratory outcomes. They found no
17 associations between NO2 level and bronchial hyperresponsiveness, FEVi, skin test
18 positivity, total IgE and questionnaire-reported wheeze, asthma, eczema or hay fever in
19 cross-sectional analyses. Further, they found no associations with decline in FEVi
20 followed-up over nine years when the data were analyzed longitudinally. Total IgE levels
21 were not related to NO2 concentrations in 369 adult asthmatics in five French centers
22 using generalized estimated equations (GEE) as part of the EGEA study (Rage et al..
23 2009) but were related to O3 concentrations.
5.2.6.2 lexicological Studies
24 Toxicological studies provide some experimental data which is coherent with the
25 development of allergic responses seen in epidemiologic studies. One subchronic
26 toxicological study showed that exposure to 4,000 ppb NO2 for 12 weeks led to enhanced
27 IgE-mediated release of histamine from mast cells isolated from guinea pigs (Fujimaki
28 andNohara. 1994). This response was not found in mast cells from rats similarly exposed
29 in the same study. Furthermore, two shorter term studies provide evidence that exposure
30 to NO2 leads to Th2 skewing and/or allergic sensitization, as discussed in Sections
31 3.3.2.6 and 4.2.4.3 (Pathmanathan et al.. 2003; Ohashietal.. 1994). Findings of increased
32 histamine release from mast cells, increased nasal eosinophils and increased Th2
November 2013 5-48 DRAFT: Do Not Cite or Quote
-------
1 cytokines seen in humans and animal models exposed to NO2 provide support for
2 epidemiologic evidence of the association of NO2 exposure with the development of
3 allergic responses.
5.2.7 Pulmonary Inflammation and Oxidative Stress
5.2.7.1 Epidemiologic Studies
Children
4 Inflammatory markers and peak expiratory pulmonary function were examined in 37
5 allergic children with physician-diagnosed mild persistent asthma in a highly polluted
6 urban area in Italy. These 37 allergic children were evaluated again 7 days after
7 relocation to a rural location with lower pollutant levels (Renzetti et al., 2009). The
8 authors observed a 4-fold decrease in nasal eosinophils and a decrease in fractional
9 exhaled nitric oxide along with an improvement in lower airway function. Several
10 pollutants were examined, including PMi0, NO2, and O3, though pollutant-specific
11 results were not presented. Exhaled NO (eNO) has been shown to be a useful biomarker
12 for airway inflammation in large population-based studies (Linn et al. 2009). Thus, while
13 the time scale of 7 days between examinations for eNO needs to be evaluated for
14 appropriateness, the results suggest that inflammatory responses are reduced when NO2
15 levels are decreased.
16 One epidemiologic study examined the relationship of airway inflammation (eNO) and
17 pulmonary function and NO2 in Windsor, Ontario (Dales et al.. 2008). This cohort of
18 2,402 school children estimated NO2 for each child's residence at the postal code level.
19 The FEVi and FVC were approximately 40 mL less in the highest compared with the
20 lowest tertiles of NO2, but these differences were weak. NO2 showed positive but weak
21 associations with eNO. The LUR estimates for Windsor did not have a large degree of
22 variability and thus may have had insufficient small-scale spatial variability. An eNO-
23 roadway density association persisted after adjustment for air pollutant levels (NO2, SO2,
24 PM2 5) within the previous 24 and 48 hours of the eNO measure, indicating that it was not
25 confounded by an unmeasured acute effect.
Adults
26 In a cross-sectional study, Wood et al. (2009) examined the association of outdoor air
27 pollution with respiratory phenotype (PiZZ type) in alpha 1-antitrypsin deficiency
November 2013 5-49 DRAFT: Do Not Cite or Quote
-------
1 (a-ATD) from the U.K. a-ATD registry. This deficiency leads to exacerbated responses
2 to inflammatory stimuli. In total, 304 PiZZ subjects underwent full lung-function testing
3 and quantitative high-resolution computed tomography to identify the presence and
4 severity of COPD - emphysema. Mean annual air pollution data for 2006 were matched
5 to the location of patients' houses and used in regression models to identify phenotypic
6 associations with pollution controlling for covariates. Regression models showed that
7 estimated higher exposure to O3 exposure was associated with worse gas transfer and
8 more severe emphysema, albeit accounting for only a small proportion of the lung
9 function variability. The positive association observed for NO2, SO2, and particles may
10 most likely be attributable to an inverse correlation of their concentrations with those of
11 O3 or they may be insufficiently representative of long-term exposure to detect effects
12 reliably.
5.2.7.2 lexicological Studies
13 Similar to studies of short-term (minutes to weeks) NO2 exposure, animal toxicological
14 studies of long-term (months to years) exposure show increases in pulmonary
15 inflammation and oxidative stress (Section 4.2.4.2). Compared with short-term exposure
16 studies, long-term studies provide more evidence of NO2-induced airway injury. Details
17 from these studies, all of which were reviewed in the 2008 ISA for Oxides of Nitrogen
18 (U.S. EPA. 2008c). are presented in Table 5-4.
19 Many studies investigating NO2-induced injury and oxidative stress in the airway
20 measure changes in lipid content, which is necessary for both lung function and defense.
21 Sagai etal. (1982) and Ichinose etal. (1983) published studies showing that rats exposed
22 to 40 or 120 ppb NO2 for 9 or 18 months had increased ethane exhalation and exposure
23 to 40 ppb for 9 months resulted in increased lipid peroxidation. Arner and Rhoades
24 (1973) showed that rats exposed to 2,900 ppb NO2 for 9 months had decreased lipid
25 content leading to increased surface tension and altered lung mechanics.
26 Histopathological assessment of lung tissue showed that long-term exposure to NO2
27 resulted in alveolar macrophage accumulation and areas of hyperinflation (Gregory et al..
28 1983). Kumae and Arakawa (2006) exposed rats to 200, 500, or 2,000 ppb NO2 from
29 birth or the weanling period (5 weeks old) and assayed BALF at 8 and 12 weeks of age.
30 Lymphocytes increased at 8 weeks with exposure to 500 ppb NO2 in the embryonic
31 group and macrophages and neutrophils were increased at 12 weeks with exposure to 500
32 ppb NO2. No changes in differential cell counts were observed in the weanling group at
33 8 weeks of age, but at 12 weeks of age, lymphocytes were increased with exposures
34 above 500 ppb and neutrophils were increased at 2,000 ppb. The embryonic group also
November 2013 5-50 DRAFT: Do Not Cite or Quote
-------
1 had increased TNF-a and IFN-y at 8 weeks but not at 12 weeks, while in the weanling
2 group, IFN-y was increased only at 12 weeks.
3 Oxidative stress resulting from NO2 exposure has been further characterized in a number
4 of studies, and the varying effects of NO2 on antioxidant levels and enzyme activity were
5 presented in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). After NO2
6 exposure, studies have reported both increased and decreased activity of enzymes
7 involved in the glutathione cycle (Sagai et al.. 1984; Gregory et al.. 1983; Ayaz and
8 Csallany. 1978). Sagai et al. (1984) reported increased non-protein sulfhydryl levels and
9 glutathione S-transferase activity in adult male rats after 9 and 18 months of exposure to
10 400 ppb NO2, and decreased glutathione peroxidase activity while glucose-6-phosphate
11 dehydrogenase activity increased after exposure to 4,000 ppb NO2. There were no
12 changes in the activity of 6-phosphogluconate dehydrogenase, superoxide dismutase, or
13 disulfide reductase after 400 ppb NO2. Gregory et al. (1983) reported increased
14 glutathione peroxidase activity in BALF after 6 weeks of exposure to 5,000 ppb NO2;
15 however, at 15 weeks, enzyme activity returned to control levels though slight changes in
16 pathology were reported. Ayaz and Csallany (1978) showed that continuous exposure to
17 1,000 ppb NO2 for 17 months decreased GPx activity in Vitamin E-deficient mice, while
18 Vitamin E-supplemented mice had increased glutathione peroxidase activity.
19 These studies demonstrate that long-term NO2 exposure modifies oxidant balance in the
20 airway and can initiate inflammation; however, the observations from these studies at
21 concentrations relevant to ambient exposures across species do not consistently show this
22 to be the case. Antioxidant enzymes are involved in response to NO2 exposure, but this
23 response is variable and transient. Overall, these findings are consistent with the reported
24 effects from short-term exposures (Section 4.2.4.2).
November 2013 5-51 DRAFT: Do Not Cite or Quote
-------
Table 5-4 Animal Toxicological Studies of the Respiratory Effects of Long-term NO2 Exposure.
Study
Arner and
Rhoades(1973)
Aranvi et al.
(1976)
Avaz and
Csallanv(1978)
Blair etal. (1969)
Chanq et al.
(1986)
Crapo et al.
(1984)
Ehrlich and Henry
Species (Strain);
Lifestage; Sex; n
Rats (Long Evans); Male
Mice
Mice (C57BL/6J); Female; n = 120
Mice; n = 4/group
Rat (Fisher 344); 1-day or 6 weeks;
Male, n = 8/group
Rat (CD, Fisher 344); 6 week;
Male
Mice (Swiss albino); Female; n =
Exposure Details (Concentration; Duration)
2,900 ppb 5 days/weeks for 9 mo
500 ppb continuously, 2,000 ppb continuously, 100 ppb
continuously with daily 3-h peaks of 1 ,000 ppb, or 500 ppb
with daily 1-h peaks of 2,000 ppb for 4, 12, 21, 24, 28, or 33
weeks.
500 ppb or 1,000 ppb continuously for 17 mo
500 ppb for 6, 18, or 24 h/day, 7 days/week for 3-12 mo
(1) 500 ppb continuously with two, daily 1-h spikes of 1,500
ppb, 5 days/week for 6 weeks,
(2) 2,000 ppb continuously for 7 days/week for 6 weeks; Two
1 h spikes daily to 6,000 ppb (6-week rats only)
2,000 ppb for 23 h/day; two daily 30 min spikes of 6,000 ppb
(1) 500 ppb continuously,
Endpoints Examined
Histopathologic evaluation and
morphometry
Morphometry
Morphometry
Histopathologic evaluation
Histopathologic evaluation and lung
morphometry
Morphometric analysis of proximal
alveolar and distal alveolar regions
Mortality, hematology, serum LDH,
(1968) > 30/group, n = 4-8/group
(2) 500 ppb for 6 h/day,
(3) 500 ppb for 18 h/day;
(1-3) for 1 to 12 mo; Challenged with Klebsiella pneumoniae
after exposure
body weight, bacterial clearance
Fujimaki and Rats (Wistar); 8 weeks; Male; n =
Nohara (1994) 10/group;
Guinea pigs (Hartley); 8 weeks; n
10/group
1,000, 2,000, or 4,000 ppb continuously for 12 weeks
Mast cell counts and histamine release
November 2013
5-52
DRAFT: Do Not Cite or Quote
-------
Table 5-4 (Continued): Animal Toxicological Studies of the Respiratory Effects of Long-term NO2 Exposure.
Study
Species (Strain);
Lifestage; Sex; n
Exposure Details (Concentration; Duration)
Endpoints Examined
Furiosi et al. Monkey (Macaca speciosa), Rat
(1973) (Sprague-Dawley);
Maturing (Monkey), Weanling (Rat);
Male/female (Monkey), Male (rat), n
= 4-5/group (Monkey), n =
15-25/group (Rat)
(1) 2,000 ppb NO2 continuously,
(2) 330 ug/m3 NaCI continuously,
(3) 2,000 ppb NO2 + 330 ug/m3 NaCI continuously
(1-3) for 14 mo
Histopathologic evaluation, hematology
Greene and
Schneider (1978)
Gregory et al.
(1983)
Havashi et al.
(1987)
Henry et al.
(1970)
Ichinose et al.
(1983)
Kumae and
Arakawa (2006)
Kubota et al.
(1987)
Lafuma et al.
(1987)
Mercer et al.
(1995)
Baboons; 3 to 4 years; Males and
Females; n = 6
Rat (Fischer 344); 14-16 weeks;
n = 4-6/group
Rat (Wistar); Male, n =
18-160/group
Squirrel Monkeys; Male; n = 37
Rats (JCL, Wistar); Sand 13
weeks; Male
Rats (Brown-Norway); prenatal
exposure; Female; n = 201
Rat (JCL Wistar);
2 mo; Male, n = 3-4/group
Hamster (Golden Syrian);
Male, n = 7-9/group
Rats (Fisher 344);
7 weeks; Male,
n = 5/group
2,000 ppb 8 h/day 5 days/week for 6 mo
(1) 1,000 ppb,
(2) 5,000 ppb,
(3) 1 ,000 ppb with two daily, 1 .5 h spikes of 5,000 ppb;
(1-3) 7 h/day for 5 days/week for up to 15 weeks
500 ppb or 5,000 ppb continuously for up to 19 mo
5,000 ppb continuously for 2 mo;
Challenge with Klebsiella pneumonia or influenza after
exposure
(1) 10,000 ppb continuously for 2 weeks;
(2)400, 1,200, or 4, 000 ppb continuously for 1, 2, 4, 8, 12, or
16 weeks;
(3) 40, 400, or 4,000 ppb continuously for 9, 1 8, or 27 mo
200, 500, or 2,000 ppb pre- and postnatal for up to 12
postnatal weeks
40 ppb, 400 ppb, or 4,000 ppb continuously for 9, 18, and 27
mo
2,000 ppb NO2 for 8 h/day for 5 days/week for 2 mo
500 ppb continuously with 2 daily, 1 h peaks of 1 ,500 ppb for
9 weeks
Immunologic and histopathologic
evaluation
Histopathological evaluation, BALF
analysis (LDH, ALKP, glutathione
peroxidase), antioxidant enzymes in
lung homogenates
Morphological changes, histology
Infection resistance, mortality,
peripheral blood markers, and
respiratory function
Histopathologic evaluation and
morphometry
Immunologic evaluation (Alveolar
macrophage activity)
Serological examination and lung
morphometry
Lung histopathology and morphometry,
lung mechanics, serum elastase
activity and protease inhibitor capacity
Histopathologic evaluation and
morphometry
November 2013
5-53
DRAFT: Do Not Cite or Quote
-------
Table 5-4 (Continued): Animal Toxicological Studies of the Respiratory Effects of Long-term NO2 Exposure.
Study
Miller et al.
(1987)
Saqai et al.
(1982)
Saqai et al.
(1984)
Sherwin and
Richters(1982)
Stevens et al.
(1988)
Tepper et al.
(1993)
Species (Strain);
Lifestage; Sex; n
Mice(CD-l);
4-6 weeks; Female;
n = 18-21/treatment group
Rats (JCL, Wistar);
8 weeks; Male
Rats (JCL Wistar);
8 weeks; Male; n = 4-6/group
Mice (Swiss Webster);
Young adults;
Male, n = 30/group
Rat (Fischer 344);
Young adult, neonate;
Male, n = 1 ore/group
Rats (Fischer 344);
60 days; Male;
n = 11-16/group
Exposure Details (Concentration; Duration)
(1)200ppb,
(2) 200 ppb daily continuously for 7 days/week with 2 daily, 1
h peaks of 780 ppb 5 days/week;
(1-2) 16, 32, or 52 weeks
10,000 ppb continuously for 2 weeks
40, 400, or 4, 000 ppb continuously for 9, 18, or 27 mo
340 ppb for 6 h/day for 5 days/week for 6 weeks
500, 1 ,000, or 2,000 ppb continuously with two daily, 1 h
spikes at 1 ,500, 3,000, or 6,000 ppb for 5 days/week for 6
weeks
500 ppb continuously 7 days/week with two daily, 2-h spikes
of 1,500 ppb, 5days/week for up to 78 weeks
Endpoints Examined
Histopathologic evaluation, pulmonary
function and antibacterial host
defenses
Histopathologic evaluation and
morphometry
Histopathologic evaluation and
morphometry
Type 2 pneumocytes in the lungs
alveolar wall area
Pulmonary function
and
Pulmonary function and lung disease
November 2013
5-54
DRAFT: Do Not Cite or Quote
-------
5.2.8 Toxicological Studies of Airway Hyperresponsiveness
1 Animal toxicological studies have demonstrated that NO2 exposure enhances
2 responsiveness of airways to nonspecific and specific challenges. A subchronic study
3 demonstrated dose-dependent increases in AHR to histamine in NO2-exposed guinea pigs
4 (U.S. EPA. 2008c; Kobayashi and Miura. 1995). In this study, one experiment
5 demonstrated AHR after 6 weeks of exposure to 4,000 ppb, but not 60 or 500 ppb NO2.
6 In another experiment, AHR was observed in guinea pigs exposed to 4,000 ppb NO2 for
7 6 weeks and to 2,000 ppb for 6 and 12 weeks and to 1,000 ppb for 12 weeks. Specific
8 airways resistance in the absence of a challenge agent was increased in guinea pigs
9 exposed to 2,000 and 4,000 ppb NO2 for 12 weeks, which indicates the development of
10 airways obstruction. Another subchronic exposure study found delayed bronchial
11 responses, measured as increased respiration rate, in guinea pigs sensitized and
12 challenged with C. albicans and exposed to NO2 (4,760 ppb, 4 hours per day, 5 days per
13 week, 6 weeks) (Kitabatake etal.. 1995). However, NO2 exposure (4,000 ppb, 2 hours
14 per day, 3 months) failed to alter airway responsiveness to a nonspecific challenge in
15 rabbits sensitized at birth with house dust mite antigen (Douglas et al., 1995). Studies of
16 acute exposures to NO2 are discussed in Section 4.2.2.2. Mechanisms underlying these
17 responses are discussed in Section 3.2.5.
5.2.9 Toxicological Studies of Host Defense
18 Decrements in the host defense mechanisms can increase susceptibility to bacterial and
19 viral infection, and toxicological studies have demonstrated that experimental animals
20 exposed to concentrations of NO2 relevant to ambient exposure for periods greater than 6
21 weeks have modulated lung host defense ranging from characteristics of alveolar
22 macrophages (AMs) to increased infection-induced mortality. Details from these studies,
23 which were reviewed in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). are
24 presented in Table 5-4.
25 Alveolar macrophages play a critical role in removing pathogens from the airways and
26 impaired function can increase susceptibility to infection and injury. Aranyi etal. (1976)
27 found that AM morphology was abnormal after 21 weeks of continuous exposure to
28 2,000 ppb NO2 and/or a base of 500 ppb NO2 with 3 hour peaks of 2,000 ppb, though
29 exposures at lower concentrations had no effects on AM morphology. Chang etal. (1986)
30 showed that exposure to 500 ppb NO2 continuously with 1,500 ppb one hour peaks twice
31 daily for 6 weeks increased the number of macrophages in the alveoli and their cellular
November 2013 5-55 DRAFT: Do Not Cite or Quote
-------
1 volume. Gregory etal. (1983) reported similar findings and observed AM accumulation
2 in lung sections by light microscopy after exposure to 5,000 ppb NO2 or a base of 1,000
3 ppb NO2 with 5,000 ppb spikes twice each day for 15 weeks.
4 Greene and Schneider (1978) investigated the functional effects of NO2 exposure on
5 AMs isolated from antigen-sensitized baboons exposed to 2,000 ppb NO2 for 8 hours per
6 day, 5 days per week for 6 months and found that they had diminished response to
7 migration inhibitory factor obtained from antigen-stimulated lymphocytes. However,
8 sample size in this study was small: 3 exposed to NO2 and antigen, 1 exposed to NO2
9 alone, 1 exposed to antigen alone, and 1 air control. Other studies have not reported on
10 this endpoint.
11 In addition to AMs, mast cells also play an important role in host defense and
12 inflammatory processes, and Fujimaki and Nohara (1994) investigated the effects of a
13 12 week continuous exposure to 1,000, 2,000, and 4,000 ppb NO2 in both rats and guinea
14 pigs. Although the number of mast cells in the airway increased after exposure to 2,000
15 and 4,000 ppb, these changes were not significant. Histamine, released by mast cells, was
16 reduced in rats at 2,000 ppb NO2 and increased in guinea pigs at 4,000 ppb. This
17 observation suggests species differences in response to NO2 exposure.
18 Effects of NO2 on impaired host defense would be expected to contribute to increased
19 susceptibility to infection. Henry etal. (1970) published a study showing that squirrel
20 monkeys exposed to 5,000 ppb NO2 for a period of 2 months and then exposed to
21 Klebsiella pneumonia or Influenza had increased markers of infection, white blood cell
22 counts and erythrocyte sedimentation rate (ESR), 3 days post infection. Furthermore, 2 of
23 the 7 monkeys exposed to NO2 died at 3 and 10 days post infection. When Influenza
24 virus was given 24 hours prior to NO2 exposure and after NO2 exposure, tidal volume
25 and respiratory rate increased and the ESR increased. One of the 3 exposed monkeys died
26 5 days post infection. Ehrlich and Henry (1968) and Ehrlich (1980) also studied the
27 effects of NO2 on Klebsiella pneumonia infection in mice. Exposures were either
28 continuous or intermittent (6 or 18 hours per day) at a concentration of 500 ppb NO2 and
29 bacterial challenge was done at 1, 3, 6, 9, and 12 months. Continuous exposure to NO2
30 for 3 months or longer resulted in increased mortality rates after infection, whereas
31 intermittent exposures led to increased mortality at 6, 9, and 12 months. Likewise, Miller
32 et al. (1987) showed increased mortality in mice exposed to a base of 200 ppb NO2 with
33 two daily 1 hour peaks of 800 ppb and subsequent challenge with Streptococcus
34 zooepidemicus at 16, 32, and 52 weeks.
35 Taken together, these studies show that animals exposed to concentrations of NO2
36 relevant to ambient exposure for periods greater than 6 weeks have modulated lung host
November 2013 5-56 DRAFT: Do Not Cite or Quote
-------
1 defense ranging from altered AM morphology and function to increased infection-
2 induced mortality.
5.2.10 lexicological Studies of Respiratory Morphology
3 While no recent studies are available, the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
4 2008c) reported that animal toxicological studies demonstrate morphological changes to
5 the respiratory tract resulting from exposure to NO2. Details from the available studies
6 are presented in Table 5-4. Studies have examined long-term exposures to NO2 to
7 determine effects on lung structure and morphology and report variations in response to
8 concentrations below 5,000 ppb. Wagner et al. (1965) exposed dogs, rabbits, guinea pigs,
9 rats, hamsters, and mice to 1,000, 5,000 or 25,000 ppb NO2 for up to 18 months and
10 found enlarged air space and edema and areas of mild to moderately thickened septae
11 with chronic inflammatory cells. However, some of these observations were also made in
12 control animals and were not considered to be significant in any species. Importantly, this
13 study demonstrated differences in sensitivity to NO2 across species. Furiosi etal. (1973).
14 exposed monkeys and rats to 2,000 ppb NO2 continuously for 14 months and also found
15 species-specific responses; monkeys experienced hypertrophy of the bronchiolar
16 epithelium that was most notable in the respiratory bronchioles in addition to
17 development of a cuboidal phenotype in the squamous proximal bronchiolar epithelium.
18 In rats, these effects were more occasional under identical exposure conditions.
19 The majority of other morphologic studies have employed rodent models to evaluate
20 effects of NO2 exposure. Chang et al. (1986) compared responses in mature and juvenile
21 rats to an urban exposure pattern of NO2 for 6 weeks (500 ppb continuously with two
22 daily peaks at 1,500 ppb). Mature rats were more sensitive to NO2 exposure and
23 exhibited increased surface density of the alveolar basement membrane and decreased air
24 space in the proximal alveolar regions, accompanied by an increase in lung volume
25 attributable to Type 2 cell hyperplasia and increases in fibroblasts, alveolar macrophages,
26 and extracellular matrix. In the juvenile rats, effects of exposure were limited to thinning
27 of Type 2 cells that were spread over more surface area compared to controls. Mercer et
28 al. (1995) found more subtle effects in rats with this exposure; lungs did not appear to
29 have differences in alveolar septal thickness, parenchymal cell populations, or cellular
30 size and surface area after 9 weeks of exposure. Although the frequency of fenestrae was
31 increased in the alveolar epithelium, there were no changes found in the extracellular
32 matrix or interstitial cells. Crapo etal. (1984) conducted a 6 week study in rats with a
33 similar exposure pattern at higher concentrations (2,000 ppb NO2 for 23 hours per day
34 with two 30 minutes peaks of 6,000 ppb) and reported hypertrophy and
35 hyperproliferation of the alveolar epithelium. In another study, rats were exposed to a
November 2013 5-57 DRAFT: Do Not Cite or Quote
-------
1 similar urban exposure pattern in addition to a single high concentration for up to 15
2 weeks; these animals had subpleural alveolar macrophage accumulation and areas of
3 focal hyperinflation, though the mean linear intercept (MLI), a measure of free distance
4 in the air space, was not changed (Gregory et al. 1983). Conversely, Lafumaetal. (1987)
5 reported that hamsters exposed to 2,000 ppb NO2 for 8 hours per day, 5 days per week
6 for 8 weeks had increased MLI and decreased internal surface area, but no lesions were
7 found in the bronchiole or bronchiolar epithelium, alveolar ducts, or alveolar epithelium.
8 Kubota et al. (1987) conducted a 27-month study in rats that included pathological
9 assessments of the airways after continuous exposure to 40, 400, or 4,000 ppb NO2. At
10 the highest exposure, rats had increased bronchial epithelial proliferation after 9 and 18
11 months, and by 27 months, proliferation and edema resulted in fibrosis. Exposure to 400
12 ppb produced similar morphological changes in the bronchial epithelium that was not
13 apparent until 27 months. Exposure to 40 ppb NO2 did not yield morphological changes
14 that could be identified by microscopic techniques. Studies conducted at similar
15 concentrations and duration have reported analogous effects. Blair et al. (1969) and
16 Hayashi etal. (1987) exposed mice and rats, respectively to 500 ppb for up to 19 months.
17 Blair etal. (1969) described an increase in alveolar size after 3 months of exposure with
18 loss of cilia in respiratory bronchioles, which persisted at 12 months. After 4 months of
19 exposure, Hayashi et al. (1987) reported type 2 cell hypertrophy and interstitial edema
20 leading to thickened alveolar septa at 6 months and fibrous pleural thickening at 9
21 months. Similarly, exposure to 500 ppb for 7 months resulted in interstitial edema and
22 type 2 cell hyperplasia in rats, and additional injury at 1,000 ppb included loss of cilia in
23 the terminal bronchioles (Yamamoto and Takahashi. 1984). Type 2 cell hyperplasia was
24 also documented by Sherwin and Richters (1982) as well as an increase in the MLI.
25 These studies demonstrate that long-term exposure to high ambient levels of NO2 can
26 result in subtle changes in lung morphology including type 2 cell hyperplasia, loss of cilia
27 in the bronchiolar region, and enlarged airspace.
5.2.11 Gene-Environment Interactions
28 Several recent studies evaluate long-term NO2 exposure and health effects and consider
29 the role that genetic variants may play in modifying risk of NO2-associated respiratory
30 effects. Such discussion provides information related to potentially at-risk populations.
31 One used the indoor cohort study by Belanger et al. (2013) discussed in Section 5.2.5 to
32 examine the association between NO2 exposure, childhood asthma severity, and levels of
33 methylation in the promoter region of the beta-adrenergic receptor (ADRB2), a target of
34 beta-agonist bronchodilators, and found that higher NO2 exposure and increased ADRB2
November 2013 5-58 DRAFT: Do Not Cite or Quote
-------
1 methylation in blood were associated with higher odds of asthma severity in children (Fu
2 etal.. 2012a).
3 Several studies examined interactions between variants in multiple genes and long term
4 outdoor exposure to NO2, increasing the possibility of finding associations by chance. In
5 a separate analysis of the BAMSE cohort (discussed earlier in Section 5.2.2). Melen et al.
6 (2008) assessed gene-environment interaction on multiple respiratory effects using
7 exposure to traffic NOX during the first year of life. Among multiple variants of ADRB2,
8 TNF, and GSTP1, effect measure modification was only found for GSTP1 and allergic
9 sensitization and PEF. Associations of these outcomes with NOX were larger among
10 children with Ile/Val or Val/Val genotypes for codon 105 (encodes enzyme with reduced
11 oxidative metabolism) and Ala/Val or Val/Val genotypes for codon 114 (functional
12 difference unknown). A three-way interaction was found with variants in TNF
13 (inflammatory cytokine), but the odds of sensitization was estimated with large
14 imprecision.
15 In the CHS (discussed earlier in Section 5.2.1). among the multiple glutathione genes
16 (i.e., GSS, GSR, GCLC, and GCL) investigated, Breton etal. (2011) found that NO2-
17 associated lung function growth deficits were modified only by variants in GSS
18 haplotype (combination of multiple alleles). Compared with children with other GSS
19 haplotypes, children with the HO0100000 haplotype (48% of study population) had
20 larger NO 2-associated decrement in growth of FEVi and MMEF but similar association
21 with FVC growth. These interactions were found for NO2 after adjusting for O3,
22 providing evidence for an independent association for NO2. In the CHS, examination of
23 community NO2 concentrations as a modifier of associations between TNF-a 308
24 variants and bronchitic symptoms among children with asthma found no difference
25 between children living in low (concentrations not reported) and high NO2 communities
26 (Lee et al.. 2009). A study of children in Taiwan also examined NO2 exposure as an
27 effect measure modifier, and found the associations of the EPHX1 His/Arg or Arg/Arg
28 genotype with lifetime asthma, early-onset asthma, and wheeze were larger among
29 children who resided in higher NO2 communities (Tung etal.. 2011). Risk was elevated
30 further if those children also had the GSTP1 Ile/Val or Val/Val genotype (reduced
31 oxidative metabolism activity), but results were based on <1% of the study population. A
32 three-way interaction was not consistently found with GSTM1 genotype variants.
33 Castro-Giner et al. (2009) (Section 5.2.2.2) examined many variants of the same genes as
34 the aforementioned studies but in adults. Associations between NO2 and asthma
35 prevalence were not modified by variants in GSTM1, TLR4 (involved in innate
36 immunity), or most of the ADRB loci examined, although one TNF variant and several
37 GSTP1 variants were found to modify the association between NO2 and asthma
November 2013 5-59 DRAFT: Do Not Cite or Quote
-------
1 prevalence. In contrast, Melen et al. (2008) did not observe effect measure modification
2 by variants in TNF, GSTP1, or ADRB2 for the association between NOX and asthma in
3 children. Also in contrast with Melen et al. (2008). a larger NO2 effect was estimated for
4 the GSTP1 lie/lie genotype. Castro-Giner et al. (2009) also found that associations
5 between prevalence of asthma and concurrent NO2 exposure were limited to adults with
6 the C/C genotypes of various polymorphisms of NQO1, which also encodes an enzyme
7 involved in oxidative metabolism. For example, adults with the C/C genotype for NQO1
8 rs291766 had a higher odds of asthma (OR: 3.75 [95% CI: 1.32, 10.64] per 10-ppb
9 increase in NO2) compared with those with CG/GG genotypes (OR: 1.54 [95% CI: 0.70,
10 3.38]).
11 Multiple recent studies examined genetic variants in ADRB2, GSTP1, and TNF, and
12 results were not consistent in showing modification of the respiratory effects of long-term
13 NO 2 exposure. A limitation of the collective body of evidence is the potentially increased
14 probability of finding association by chance with multiple comparisons. Several studies
15 examined genetic variants in oxidative metabolism enzymes, particularly those involving
16 glutathione metabolism. For many variants, no interaction was found. However, there
17 were a few observations of effect measure modification of lung function growth in
18 children and asthma in adults by variants in GSS and NQO1.
5.2.12 Concentration-Response
19 Several studies report information examining the NO2 exposure-response function. For
20 the Connecticut/Massachusetts Childrens Asthma cohort discussed by Belanger et al.
21 (2013). Figure 5-6 illustrates, for fully adjusted models, the exposure-response
22 relationships between indoor NO2 and health outcomes using a constrained, natural
23 spline function of ln(NO2) and 95% confidence limits as well as threshold functions for
24 each outcome. In adjusted models examining quintiles of NO2 exposure, levels >14.3 ppb
25 compared with the reference level (< 6 ppb, the threshold value) resulted in an increased
26 risk of a one-level increase in asthma severity score (OR: 1.43 [95% CI: 1.08, 1.88]).
27 Wide CI's were observed. These same exposures were also associated with increased
28 risks of wheeze (1.53 [95% CI: 1.16, 2.02]), night symptoms (1.59 [95% CI: 1.24, 2.01]),
29 and rescue medication use (1.74 [[95% CI: 1.34, 2.26]). In the fully adjusted threshold
30 models, every 5-fold increase in NO2 exposure >6 ppb was associated with a dose-
31 dependent increase in asthma severity score (1.37 [[95% CI: 1.01, 1.89]) and asthma
32 morbidity measured by wheeze (1.49 [[95% CI: 1.09, 2.03]), night symptoms (1.52
33 [[95% CI: 1.16, 2.00]), and rescue medication use (1.78 [[95% CI: 1.33, 2.38]).
November 2013 5-60 DRAFT: Do Not Cite or Quote
-------
A Asthma Severity Score
B wheeze
2.0-,
S. -i
0.5.-'
0.3
6 ppb /
32 ---
128
128
NO2 {ppb, log scale)
0.3
C Night Symptoms
NO2 (ppb, log scale)
D Rescue Medication Use
2.0
1-1.0
0.5
0.3 J
NO2 (ppb, log scale)
2 4 8 16 32 64 128
NO, (ppb, log scale)
Note: (A) asthma severity score, (B) wheeze, (C) night symptoms, and (D) rescue medication use. Also shown is a histogram of
NO2 levels measured in subjects' homes (lower portion of panel D) for all observations (thin border) and observations taken in
homes of gas stove users (bold border).
Source: Reprinted with permission of Wolters Kluwer Health, Belanger et al. (2013).
Figure 5-6 Concentration-response relationships between health outcome
and NO2 (log concentration as a continuous variable) illustrated
with constrained, natural spline functions (solid lines) with 95%
confidence limits (small dashed lines) and threshold function
(bold dashed line) from fully adjusted, hierarchical ordered
logistic regression models.
i
2
3
4
5
6
Gauderman et al. (2004) discussed in Section 5.2.1 states that although the average
growth in FEVi was larger in boys than in girls, the correlations of growth with air
pollution did not differ significantly between the sexes, as shown for NO 2 in Figure 5-7.
The sex-averaged analysis, depicted by the regression line in the figure, demonstrated a
significant negative correlation between the growth in FEVi over the eight-year period
and the average NO2 level (p = 0.005). The estimated difference in the average growth in
November 2013
5-61
DRAFT: Do Not Cite or Quote
-------
1
2
3
over the eight-year period from the community with the lowest NO2 level to the
community with the highest NO2 level, represented by the slope of the plotted regression
line in the Figure 5-7. was -101.4 mL.
1450-
^ 1420-
•£ 1390-
w> 1360-
o
I 1330H
i—i
? 1300H
1270-
1240-
1210-
0
0
O Girls
• Boys
Q>
O
I
5
10 15 20 25
N02 (ppb)
30
35
-2485
-2455
-2425
-2395
-2365
40
W>
o
n
-2335 ^
-2305
-2275
-2245
0
2
Source: Reprinted with permission of the Massachusetts Medical Society, Gauderman et al. (2004).
Figure 5-7 Community-specific average growth in FEVi (mL) among girls
and boys during the eight-year period from 1993 to 2001, plotted
against average NO2 levels from 1994 through 2000.
4
5
Gauderman et al. (2004) further observed that pollution-related deficits in the average
growth in lung function over the eight-year period resulted in clinically important deficits
in attained lung function at the age of 18 years (Figure 5-8). Across the 12 communities,
a clinically low FEVi was positively correlated (r = 0.75) with the level of exposure to
NO2.
November 2013
5-62
DRAFT: Do Not Cite or Quote
-------
10-.
8-
6-
4-
2-
0
0
R=0.75
P=0.005
• UP
ML
LA
10
» LN,
20
N02(ppb)
i
30
40
Note: The correlation coefficient (R) and P value are shown for each comparison. AL = Alpine, AT = Atascadero, LE= Lake Elsinore,
LA = Lake Arrowhead, LN = Lancaster, LM = Lompoc, LB = Long Beach, ML = Mira Loma, RV = Riverside, SD = San Dimas,
SM = Santa Maria, and UP = Upland. NO2: nitrogen dioxide.
Source: Reprinted with permission of the Massachusetts Medical Society, Gauderman et al. (2004).
Figure 5-8 Community-specific proportion of 18-year-olds with a FEVi below
80 percent of the predicted value, plotted against the average
levels of NO2 from 1994 through 2000.
i
2
3
4
5
6
Additional studies that demonstrate associations between long-term exposure to NO2 and
respiratory health outcomes have generally observed a linear relationship in the range of
ambient NO2 concentrations examined (Andersen et al.. 2012; Lee et al.. 2012c: Carlsten
etal.. 20lie: Modig et al.. 2009; Islam et al.. 2007; Roias-Martinez et al.. 2007a. b;
Shima et al.. 2002). Most of these studies did not conduct analyses to evaluate whether
there is a threshold for effects.
5.2.13 Analysis of copollutants
9
10
Various studies provide information that informs the concept of copollutant aspects of
effects that may be attributed to various NO2 measures. Table 5-5 presents this
information for the copollutants PM, O3, SO2, and CO showing the results of two
pollutant models and those with three or more. Generally, these studies reported that
November 2013
5-63
DRAFT: Do Not Cite or Quote
-------
estimates from two-pollutant models were not substantially different from the estimates
from models that just included NO2 (Lee et al.. 2012c: Roias-Martinez et al.. 2007a. b),
(Hwang and Lee. 2010; Hansel et al.. 2008; Hwang et al.. 2005; McConnell et al.. 2003).
Table 5-5 Studies that provide evidence for NO2 and also provide analysis of
copollutants (PM, O3, SO2, CO).
Study
Design
Health Effect Endpoint
Increament of
NO2 Exposure
Examined
NO2 single pollutant effect
estimate
NO2 with other pollutant
effect estimate
Roias-Martinez et al.
(2007a)
Prospective
Pulmonary Function
10 ppb
FEVi,FVC(inmL)
NO2:
Girls
FVC-1.26(-1.51,-1.01)*V
FEV1 -0.70 (-0.96,-0.44)'
Boys
FVC-1.18(-1.42, -0.94) *v
FEV1 -0.56 (-0.81,-0.32)'
!, FVC-mL
NO2 and O3:
Girls
FVC-1.20 (-1.46,-0.94)**
FEV1 -0.81 (-1.07, -0.54)*
Boys
FVC-1.16 (-1.41,-0.91)**
FEV1 -0.75 (-1.01, -0.50)*
NO2 and O3 and PM10:
Girls
FVC-1.05 (-1.32,-0.77)**
FEV-, -0.71 (-1.00, -0.42)*
Boys
FVC-1.09 (-1.36,-0.82)**
FEV-, -0.64 (-0.92, -0.37) *
November 2013
5-64
DRAFT: Do Not Cite or Quote
-------
Table 5-5( Continued): Studies that provide evidence for NO2 and also provide analysis of
copollutants (PM, O3, SO2, CO).
Study
Design
Health Effect Endpoint
Increament of
NO2 Exposure
Examined
NO2 single pollutant effect
estimate
NO2 with other pollutant
effect estimate
McConnell et al. (2003)
Prospective
Bronchitic symptoms in
asthmatics in CHS
OR
1 ppb
Within Communities
NO2:
1.07(1.02, 1.13)**
Within Communities
NO2 andO3: 1.06 NS
NO2 and PMi0: 1.07*
NO2 and PM2.5: 1.05NS
NO2 and PMio-2.5: 1.08**
NO2 and inorganic: 1.09*
NO2 and organic: 1.07*
NO2andEC: 1.05*
NO2andOC: 1.04 NS
Between Communities
NO2:
1.02(1.00, 1.03)*
Between Communities
NO2andO3: 1.02*
NO2 and PMi0: 1.01 NS
NO2 and PM2.5:1.01 NS
NO2andPMio-25.: 1.02*
NO2 and inorganic: 1.02NS
NO2 and organic: 1.02 NS
NO2andEC: 1.01 NS
NO2andOC: 1.01 NS
Donqetal. (2011)
Cross-sectional
Respiratory Symptoms
OR
5.3 ppb
NO2:
Without allergic predisposition
Persistent cough
Males
1.28(1.16-1.41)
Females
1.21 (1.09-1.33)
Persistent Phlegm
Males
1.16(1.02-1.33)
Females
1.13(0.98-1.30)
Doctor-diagnosed asthma
Males
1.19(1.06-1.34)
Females
1.14(0.99-1.30)
NO2 and
, SO2, O3, CO:
Persistent cough
Males
1.36(1.19-1.56)
Females
1.30(1.13-1.50)
Persistent Phlegm
Males
1.26(1.05-1.51)
Females
1.15(0.94-1.42)
Doctor-diagnosed asthma
Males
0.97(0.80-1.16)
Females
0.97(0.74-1.26)
November 2013
5-65
DRAFT: Do Not Cite or Quote
-------
Table 5-5( Continued): Studies that provide evidence for NO2 and also provide analysis of
copollutants (PM, O3, SO2, CO).
Study
Design
Health Effect Endpoint
Increament of
NO2 Exposure
Examined
NO2 single pollutant effect
estimate
NO2 with other pollutant
effect estimate
Hwang and Lee (2010)
Cross-sectional
Bronchitic Symptoms in
children with asthma
OR
8.79 ppb
NO2:
Bronchitis
1.83(1.07, 3.14)
Chronic Phlegm
1.52(0.83,2.78)
Chronic Cough
1.12(0.53,2.4)
Bronchitic symptoms
1.81 (1.14,2.86)
NO2 and SO2:
Bronchitis
1.99(1.02, 3.87)
Chronic Phlegm
1.27(0.59,2.74)
Chronic cough
1.25(0.49, 3.23)
Bronchitic symptoms
1.76(0.99, 3.14)
NO2 and PM2.5:
Bronchitis
2.04(1.15, 3.63)
Chronic Phlegm
1.51 (0.79,2.89)
Chronic cough
1.26(0.57,2.80)
Bronchitic symptoms
2.01 (1.20, 3.36)
NO2 and O3:
Bronchitis
1.81 (1.06, 3.10)
Chronic Phlegm
1.49(0.81,2.73)
Chronic cough
1.10(0.58,2.07)
Bronchitic symptoms
1.79(1.12,2.85)
Hwang et al. (2005)
Cross-sectional
Physician-diagnosed
asthma
OR
10 ppb
NO2:
1.005(0.954, 1.060)
NO2 and SO2:
1.048(0.983, 1.117)
NO2 and PMi0:
1.065(1.009, 1.123)
NO2 and O3:
1.029(0.973, 1.089)
NO2, SO2, and O3:
1.113(1.038, 1.194)
NO2, PMio and O3:
1.152(1.082, 1.227)
November 2013
5-66
DRAFT: Do Not Cite or Quote
-------
Table 5-5( Continued): Studies that provide evidence for NO2 and also provide analysis of
copollutants (PM, O3, SO2, CO).
Study
Design
Health Effect Endpoint
Leeetal. (201 2c)
Prospective
Incident
bronchitis/Pulmonary
function differences
Hansel et al. (2008)
Prospective
Asthma symptoms
Increament of
NO2 Exposure
Examined
Two NO2 strata
defined as less
than and greater
than the median
level of 17.5 ppb
20 ppb
NO2 single pollutant effect
estimate
Incidence rate ratio of bronchitis
was 0.56 (95% Cl: 0.49, 0.65) in
the lower NO2 communities,
whereas the effect was 1.10
(95% Cl: 0.96, 1 .24) in the higher
NO2 communities
(p for interaction = 0.005).
In example
Cough without cold:
OR 1.10 (95% Cl: 1.02, 1.18)**
NO2 with other pollutant
effect estimate
Found no statistical significant
differences on the effects of
pulmonary function indices for
incident bronchitis in relation to
exposure to the other air
pollutants in TCHS (PM2.s,
PM-io and 8-h O3).
Adjusting for other copollutants
did not meaningfully alter the
association between indoor
NO2 concentrations and
asthma symptoms.
***p<0.0001,
**p<0.01,
*p <0.05,
NSp>0.05).
5.2.14 Mixtures: Traffic-related Pollutants
1 Studies that inform measures of NO2 and estimates for traffic related pollutant are
2 discussed next and are presented in Table 5-6. Several studies conducted in California
3 evaluated the effects of exposure to NO2 and an indicator of traffic related pollution on
4 respiratory health effects. Each of the studies observed an independent effect between
5 NO2 and respiratory health effects; however it was not always simple to disentangle this
6 effect from that observed for traffic related pollution. For example, McConnell et al.
7 (2010) observed an association between ambient NO2 measured at a central site and new-
8 onset asthma. In models with both NO2 and modeled traffic exposures, there were
9 independent associations of asthma with traffic-related pollution (TRP) at school and
10 home, whereas the estimate for NO2 was attenuated. Associations with asthma were
11 positive for TRP exposure estimates modeled from local non-freeway roadway
12 proximity, traffic volume, and meteorology. There was little evidence for an effect of
13 major roadway proximity alone, for traffic density, or for pollution from freeways. An
14 important distinction between the TRP and simpler traffic metrics is the inclusion of
15 meteorology (average annual wind speed and direction and height of the mixing layer) in
16 addition to proximity and volume. The contribution that NO2 may or may not make to
17 TRP measures directly or indirectly is not clear. The modeled TRP may reflect the
18 mixture of multiple pollutants from nearby traffic, and the high correlation of pollutants
19 in the mixture may preclude identifying the effect of any specific pollutant in the mixture
November 2013
5-67
DRAFT: Do Not Cite or Quote
-------
1 as a causative agent. Similarly, Gauderman et al. (2007) noted that reduced lung-function
2 growth was independently associated with both freeway distance and with regional air
3 pollution. Positive associations were observed in joint models of regional pollution with
4 distance to freeway and NO2, acid vapor, EC, and PMi0 and PM2 5. Ozone was not
5 associated with reduced lung-function growth. There was no evidence of effect
6 modification (interaction) of local traffic effects with any of the regional pollutants. In a
7 cross-sectional analysis of residential traffic and children's respiratory health (current
8 asthma) in San Francisco, Kim et al. (2008) found that when outdoor school-based NO
9 concentration was added to multivariate models containing residential-based traffic, the
10 effect estimate for residential traffic was mildly attenuated. Effect estimates for
11 residential traffic were essentially unchanged with the addition of NO2, PMi0, or PM2 5.
12 Traffic density and maximum annual average daily traffic (AADT) were correlated with
13 pollutants and explained between 35% and 60% of the variability in NOX and NO. The
14 traffic metrics used in these studies are surrogates for a complex mixture of traffic
15 pollutants composed of reactive gases and PM, not just NOX. Many constituents of traffic
16 exhaust may contribute to toxicity.
November 2013 5-68 DRAFT: Do Not Cite or Quote
-------
Table 5-6 Studies reporting NO2 results and results for traffic measures.
Study Design;
Health effect
McConnell et al. (2010)
NO2 Exposure
Range of exposure
NC>2 Single pollutant
Effect Estimate;
Other TRP Measures
Single NO2
NO2 and Indicator of
Traffic
NO2 adjusted for
Prospective
Incident Asthma
over the 13 |_|p
communities for central 217(1 18 400)
siteNO2 ' '
23.6 ppb
Other TRP measures:
Non-freeway TRP Combined
Home and School
1.61 (1.29,2.00)
Freeway TRP combined Home
and School
1.12(0.94, 1.35)
Total TRP Combined
1.34(1.07, 1.68)
Traffic Density
Combined
1.09(0.99, 1.19)
Distance to Major Road
Combined 0.85 (0.68, 1.07)
Distance to Freeway
Combined 0.89 (0.76, 1.05)
nonfreeway TRP at Home
and School
HR
1.32(0.69,2.71)
HR for NO2 adjusted for
Total TRP at Home and
School
1.79(0.91, 3.52)
Gauderman et al. (2007)
Prospective
Pulmonary function
Range of 12 community FEV-i growth
means 34.6 ppb ^Og m|_ p 0.003
FEV-i Growth
Local freeway distance
(in meters)
<500:-80 p 0.012
500-1000:-41 (p = 0.166)
1000-1500:-33 (p= 0.279)
Kim et al. (2008)
Cross-sectional
Current asthma and
Bronchitis
Quantitative data not
reported.
No direct NO2 or NOX effect
reported
When school-based
concentration of NO was
added to multivariate
models containing
residential-based traffic, the
effect estimate for
residential traffic was mildly
attenuated.
School-based NO had
borderline significance in
the models (p <0.12). Effect
estimates for residential
traffic were essentially
unchanged with the
addition of the school
pollutant NO2, PM-io, or
PM2.5.
November 2013
5-69
DRAFT: Do Not Cite or Quote
-------
5.2.15 Indoor Studies
1 In the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). the intervention study by
2 Pilotto et al. (2004) found that exposure to NO2 from an indoor combustion source is
3 associated with respiratory effects. In this study NO2 effects would not be confounded by
4 other motor vehicle emission pollutants, though potential confounding by other pollutants
5 from gas stove emissions, such as UFP, could occur. This was an important study that
6 helped reduce the uncertainty for NO2 providing direct health effects. The two recent
7 indoor studies (Belanger et al.. 2013; Hansel et al.. 2008) discussed in Section 5.2.5.1
8 provide evidence that supports this notion in that they are not the same air mixtures as in
9 the ambient air and indicate strong positive relationships for long-term NO2 exposure and
10 respiratory symptoms in asthmatic children.
5.2.16 Surrogate for ambient NO2 or other pollutants as a mixture
11 The above discussion of studies of potential health effects related to measures of ambient
12 NO2 used various methods to estimate levels of NO2: (1) Palmes tubes outside the homes
13 of the study subjects; (2) community pollutant monitoring cites; (3) data from indoor
14 monitoring and activity patterns; and (4) various land use regression models most with an
15 element of validation of the model. It is with this backdrop that we discuss NO2 as an
16 independent agent related to health effects observed or as discussed next a surrogate for a
17 mixture or other role related to the observed health effects.
18 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) concluded that "The errors and
19 uncertainties associated with the use of ambient NO2 concentrations as a surrogate for
20 personal exposure to ambient NO2 generally tend to reduce rather than increase effect
21 estimates, and therefore are not expected to change the principal conclusions from NO2
22 epidemiologic studies." and further concluded that "It is difficult to determine from these
23 new studies the extent to which NO2 is independently associated with respiratory effects
24 or if NO2 is a marker for the effects of another traffic-related pollutant or mix of
25 pollutants".
26 In evaluating the potential relationships between long-term exposure to NO2 and
27 respiratory effects, it is important to note the interrelationships between NO2 and other
28 pollutants, and the potential for NO2 to serve as a marker for a pollutant mixture,
29 particularly traffic-related pollution. This includes consideration of potential pathways,
30 such as the direct causal pathway for effects, mediation of effects, the pollutant acting as
31 a surrogate for a pollutant mixture, mixtures that share the same source (e.g., motor
32 vehicles, electricity generation), or confounding between pollutants. As observed above,
November 2013 5-70 DRAFT: Do Not Cite or Quote
-------
1 associations with NO2 were often robust to adjustment for traffic-related pollutants (e.g.,
2 PM and CO),
3 Although this complicates the efforts to disentangle specific long-term NO 2 -related
4 health effects, the evidence summarized in this assessment indicates that NO2
5 associations generally remain robust in multipollutant models as discussed above in
6 Section 5.2.13 which supports a direct effect of long-term NO2 exposure on respiratory
7 morbidity at ambient concentrations. The robustness of epidemiologic findings to
8 adjustment for copollutants, coupled with limited data from animal and human
9 experimental studies, inform the strength of evidence for a relationship between NO2 and
10 respiratory morbidity. In addition, the short-term exposure intervention study of indoor
11 NO2 exposures by Pilotto et al. (2004) discussed in the 2008 ISA for Oxides of Nitrogen
12 (U.S. EPA. 2008c) found that exposure to NO2 from indoor combustion sources were
13 associated with respiratory effects. In Pilotto et al. (2004) NO2 effects would not be
14 confounded by other motor vehicle emission pollutants, though potential confounding by
15 other pollutants from gas stove emissions, such as UFP, could hypothetically occur but
16 data is limited in support of this notion. Support for long-term NO2 exposure effects of
17 indoor NO2 are provided by the prospective studies Hansel et al. (2008) and Belanger et
18 al. (2013). which demonstrate respiratory morbidity effects related to long-term NO2
19 exposure which are unlikely to be confounded by other motor vehicle emission
20 pollutants. Hansel et al. (2008) noted that adjusting for copollutants did not alter the
21 relationship for indoor NO2 and respiratory symptoms and Belanger et al. (2013) reported
22 concentration response data.
23 Human clinical and toxicological study findings also provide support for independent
24 effects of NO2 on respiratory health. Limited short-term exposure evidence from human
25 clinical studies indicated that NO2 may increase susceptibility to injury by subsequent
26 viral challenge; toxicological studies show that lung host defenses are sensitive to NO2
27 exposure. The epidemiologic and experimental evidence together show coherence for
28 effects of NO2 exposure on host defense or immune system effects providing plausibility
29 and mechanistic support for respiratory morbidity.
30 In regard to the question on surrogates, the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
31 2008c) evaluated the available evidence base. Recently, Meng et al. (2012b) further
32 inform this question and note that it is necessary to examine personal-ambient
33 associations of NO2 in a multipollutant environment. The issue raised in the
34 epidemiologic studies is whether ambient NO2 is a surrogate of personal exposure to
35 ambient NO2 or a surrogate of personal exposure to other ambient pollutants, such as fine
36 particles (PM2 5). Meng et al. (2012b) conducted a quantitative research synthesis on
37 studies of the associations between personal exposures and ambient concentrations of
November 2013 5-71 DRAFT: Do Not Cite or Quote
-------
1 NO2 reported in peer-reviewed publications. Random-effects meta-analysis was
2 conducted to estimate the strength of the associations between personal exposures and
3 ambient concentrations of NO2 across the studies. Ambient NO2 was found to be
4 significantly associated with personal NO2 exposures, with overall correlation coefficient
5 estimates of 0.42, 0.16, and 0.72 for pooled, longitudinal, and daily average correlation
6 coefficients across studies. This conclusion was robust to correction for publication bias.
7 Random effects meta-regressions were also conducted to examine factors affecting the
8 heterogeneity in the reported correlation coefficients across studies. They reported that
9 personal-ambient associations of NO2 depend on various factors, including season, age of
10 the study population, pre-existing disease, and possibly indoor and local sources and
11 sampling aspects. The dependence of the personal-ambient associations on these factors
12 complicates the interpretation of ambient NO2 as a surrogate of personal NO2 exposure
13 of ambient origin. Ambient NO2 might be a good surrogate for personal NO2 exposure of
14 ambient origin for some subpopulations but not for others, even though the collective
15 evidence suggests that ambient NO2 is a good surrogate for personal exposure of ambient
16 origin. Furthermore, measured personal exposures could be influenced by sampling
17 artifacts. More accurate real-time exposure measurements will help improve the
18 interpretation of personal-ambient associations. Caution needs to be exercised when
19 comparing personal ambient associations obtained with different study designs. It should
20 be noted that the number of studies in this analysis was relatively small. More meaningful
21 and rigorous comparisons would be possible if greater detail were published on study
22 design and data quality.
23 The HEI (2010) review of traffic-related pollutants viewed the alternative hypothesis
24 surrogate issue from a different perspective: surrogates for TRP. They noted that none of
25 the surrogates considered met all the criteria for an ideal surrogate. Data are not available
26 on the ratios of the surrogates to the complex pollutant mixtures emitted by traffic and
27 how these ratios have varied over time. CO, benzene, and NOX (in this case NO2), found
28 in on-road vehicle emissions, are components of emissions from all sources. All also
29 have significant ambient and microenvironmental sources, making it difficult to
30 disentangle the contributions from motor vehicles. Primary on-road emissions of PM
31 represent a small contribution to emissions from all sources. The quality of the surrogates
32 (i.e., their degree of association with "true" traffic exposure) therefore depends very
33 much on the understanding of the contributions from other sources. Thus in the
34 discussion of NO2 representing various mixtures here, a surrogate, one should be
35 cautious in considering the ideal surrogate that may represent the alternative hypothesis
36 under consideration.
November 2013 5-72 DRAFT: Do Not Cite or Quote
-------
5.2.17 Summary and Causal Determination
1 Evidence indicates that there is likely to be a causal relationship between long-term NO2
2 exposure and respiratory effects based on multiple lines of evidence indicating increases
3 in asthma incidence in children, decrements in lung function, and partially irreversible
4 decrements in lung function growth in children. There is supporting evidence for
5 increases in respiratory symptoms in children with asthma, increases in asthma incidence
6 in adults. Evidence from toxicological studies provide biological plausibility for the
7 associations observed between long-term exposure to NO2 and asthma incidence and
8 demonstrate impaired lung host defense and increased infection mortality. This
9 conclusion represents a change from the "suggestive but not sufficient to infer a causal
10 relationship" determined in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
11 Consistent with previous findings, recent epidemiologic results continue to support
12 associations between increases in ambient NO2 concentrations and pulmonary function
13 decrements. The recent epidemiological evidence base evaluating long-term NO2
14 exposure and asthma incidence in children now includes several prospective longitudinal
15 studies examining asthma incidence. Incontrast, the 2008 ISA for Oxides of Nitrogen
16 which had a limited number of cross-sectional studies available to consider. The key
17 uncertainty identified in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) (related
18 to the high correlations among traffic-related pollutants which made it difficult to
19 accurately estimate the independent effects of long-term NO2 exposures) was the
20 potential for NO2 to serve as mainly an indicator for another combustion-related pollutant
21 or mixture. This uncertainty is informed with recent studies, but remains as a key
22 uncertainty. The evidence for respiratory effects with respect to likely to be a causal
23 relationship with long-term NO2 exposure is detailed below using the framework
24 described in Table II of the Preamble to this ISA. The key evidence, supporting or
25 contradicting, as it relates to the causal framework is presented in Table 5-9.
26 The strongest evidence is provided by recent studies of asthma incidence in children
27 where previous evidence was inconsistent. Multiple longitudinal, prospective studies
28 (Table 5-9) have demonstrated associations between higher ambient NO2 concentrations
29 measured in the first year of life, in the year of diagnosis, or over a lifetime and asthma
30 incidence in children. Results have been replicated by investigators in different locations
31 using various study designs and cohorts (Gruzieva et al.. 2013: Nishimura et al.. 2013a:
32 Lee etal.. 2012c: Carlsten etal.. 20 lie: Clark et al.. 2010: Gehring etal.. 2010:
33 Mcconnell etal..2010: Oftedal et al.. 2009: Jerrett et al.. 2008: Cloughertv et al.. 2007:
34 Shima et al.. 2002).
35 Another line of evidence supporting NO2-related respiratory effects is multiple, high-
36 quality longitudinal studies finding associations between long-term NO2 exposure and
November 2013 5-73 DRAFT: Do Not Cite or Quote
-------
1 decrements in lung function and partially irreversible decrements in lung function growth
2 in children which has added additional longitudinal prospective studies since the 2008
3 ISA (Figure 5-5) (Schultz etal.. 2012; Breton etal. 2011; Oftedal et al.. 2008; Roias-
4 Martinez et al.. 2007a; Gauderman et al.. 2004). Some studies found an NO2
5 concentration-dependent decrement in lung function and lung function growth (Rojas-
6 Martinez et al.. 2007a; Gauderman et al.. 2004). The toxicological evidence regarding
7 NO2 -induced changes in lung function and growth is limited and thus does not provide
8 clear biological plausibility for epidemiologic observations. A few studies demonstrated
9 effects of long-term NO2 exposure on lung function in rats exposed to a base 500-2,000
10 ppb for 6-78 weeks (Tepper et al., 1993; Lafumaetal.. 1987). and slight decrements in
11 lung function resulting from short-term exposure were found to resolve with continued
12 NO2 exposure (1,000-3,000 ppb for 6 weeks) (Stevens et al.. 1988). A number of studies
13 have demonstrated that long-term exposure to NO2 alters lung morphology in
14 experimental animals, though these changes do not appear to contribute to altered lung
15 function (Hayashi et al.. 1987; Kubotaetal.. 1987). Additionally, these effects were not
16 clearly demonstrated in juvenile animals (Chang etal.. 1986; Furiosi etal.. 1973).
17 These observations are supported by evidence from several longitudinal studies
18 consistently demonstrating increases in respiratory symptoms in children with asthma
19 with increasing ambient NO2 concentrations (Table 5-3) (Belanger et al., 2013; Gruzieva
20 etal.. 2013; Gehring etal.. 2010; Hansel et al.. 2008; McConnell et al.. 2003). Also
21 supporting a relationship between long-term NO2 exposure and respiratory effects is new
22 evidence from several multicity studies demonstrating increases in asthma incidence in
23 adults (Table 5-8) (Castro-Giner et al.. 2009; Jacquemin et al.. 2009a; Jacquemin et al..
24 2009b; Modig et al.. 2009; Sunver et al.. 2006).
25 Evidence for asthma incidence, pulmonary function, and respiratory symptoms in
26 children is substantiated by the prospective design of studies, which better characterizes
27 the directionality between exposure and development of respiratory morbidity. Also,
28 associations were found with adjustment for several well-characterized potential
29 confounding factors such as SES, smoking exposure, housing characteristics, and
30 meteorological conditions. In addition, studies characterized the concentration-response
31 for the relationship between NO2 and respiratory outcomes, generally observed a linear
32 relationship (Andersen et al.. 2012; Carlsten et al.. 20lie; Modig et al.. 2009; Islam et al..
33 2007; Rojas-Martinez et al.. 2007a. b; Shima et al.. 2002). Evidence for an increase in
34 respiratory symptoms related to NO2 indoors and the related indoor mixture is provided
35 by (Belanger et al.. 2013). These observations for long-term exposure are supported by
36 evidence for short-term NO2 increases in exposure increasing pulmonary inflammation
37 and respiratory symptoms in children in the general population and increasing respiratory
38 symptoms in children with asthma (Section 4.2). Recent meta-analysis of asthma
November 2013 5-74 DRAFT: Do Not Cite or Quote
-------
1 incidence also informs the evidence base (Anderson et al.. 2013; Gasanaet al.. 2012;
2 Powers et al.. 2012; Takenoue et al.. 2012; Braback and Forsberg. 2009).
3 Evidence from toxicological studies provides biological plausibility for the associations
4 observed between long-term exposure to NO2 and asthma incidence. Increases in AHR
5 were reported following 6-12 weeks of exposure to NO2 (1,000-4,000 ppb) (Kobavashi
6 and Miura. 1995). Studies of short-term exposure and long-term exposure provide
7 evidence of key events to inform the mode of action for development of asthma,
8 including oxidative and nitrative stress, altered regulation of inflammation (Th2
9 cytokines), airway remodeling, and enhanced allergic sensitization (Sections 3.3.2 and
10 5.2.6). Associations between long-term NO2 exposure and allergic sensitization in
11 children and adults have been described (Section 5.2.6.1). There is additional evidence
12 indicating oxidative stress may underlie the observed associations between long-term
13 ambient NO2 exposure and asthma incidence as some studies have found that individuals
14 with variant genotypes for enzymes with antioxidant activity (i.e., NQO1, EPHX), are at
15 greater risk for asthma incidence and symptoms. Long-term exposure of rodents to
16 ambient-relevant concentrations of NO2 resulted in increased lipid peroxidation in lung
17 tissue and exhaled ethane (Kumae and Arakawa. 2006). though only slight, transient
18 effects of NO2 on antioxidant enzymes were observed (Sagai et al.. 1984; Gregory et al..
19 1983).
20 The strongest evidence of effects of long-term NO2 exposure (500-2,000 ppb for 1 month
21 up to 1 year) in toxicological studies demonstrates impaired lung host defense and
22 increased infection mortality. Alterations to alveolar macrophage function and
23 morphology have been reported (Gregory et al.. 1983; Aranyi et al.. 1976) and increases
24 in mortality following bacterial challenge have been documented in rodents and squirrel
25 monkeys exposed to NO2 compared to air controls (Miller etal.. 1987; Henry et al..
26 1970). These findings provide support for the limited available epidemiologic evidence
27 showing increases in respiratory infections, respiratory hospital admissions, or mortality
28 in association with long-term NO2 exposure. Neupane et al. (2010) demonstrated that
29 NO2 was associated with increases in community-acquired pneumonia using 3 different
30 models to estimate annual NO2 exposure. Although the models included covariates, they
31 did not adjust for short-term NO2 exposure. In addition, there is evidence for associations
32 of long-term ambient NO2 exposure and respiratory mortality as discussed in Section
33 5.5.2.
34 Several lines of epidemiologic evidence support a relationship between long-term
35 ambient NO2 exposure and increased asthma incidence and respiratory symptoms, and
36 decrements in lung function growth in children. However, uncertainty remains from
37 limited supporting toxicological evidence to provide biological plausibility and the
November 2013 5-75 DRAFT: Do Not Cite or Quote
-------
1 possibility that the observed effects could result from exposure to other pollutants or a
2 mixture of NO2 and other pollutants. Rojas-Martinez et al. (2007a) and McConnell et al.
3 (2003) found a robust relationship for NO2 with decrements in pulmonary function and
4 bronchitic symptoms, respectively, with adjustment for O3, PMi0, PM2 5, or EC.
5 However, analysis of copollutant models is limited. Several studies show associations
6 with copollutants such as PM25, CO, BC, and SO2 , which tend to show high correlations
7 with NO2. Studies providing results for pollutants other than NO2 show similar results
8 for the other pollutants or in the case of Nishimura et al. (2013a) and McConnell et al.
9 (2010) show NO2 effects that are larger than those for the other pollutants.
10 Taken together, the recent epidemiologic studies of asthma incidence, decrements in lung
11 function growth, increased respiratory symptoms, and toxicological studies of lung host
12 defense and increased susceptibility to respiratory infections provide evidence that there
13 is likely to be a causal relationship between long-term NO2 exposure and respiratory
14 effects (Table 5-9). The strongest evidence is provided by studies that demonstrated
15 increases in asthma incidence with NO2 exposure in children and decrements in
16 pulmonary function in children. Supporting evidence is provided by evidence of asthma
17 incidence in adults and increases in respiratory symptoms in children with asthma.
18 Biological plausibility for epidemiologic evidence is provided by toxicological evidence
19 for development of AHR and Th2 phenotype. Several studies characterize a linear
20 relationship. The majority of studies adjust for potential confounding by SES, smoking
21 exposure, housing characteristics, and meteorological conditions. While studies show
22 associations with NO2 to be robust with adjustment for O3, PMi0, PM2 5, or EC, analysis
23 of potential copollutant confounding is limited, and NO2 is often highly correlated with
24 copollutants. Based on this small group of studies and limited evidence from
25 experimental studies to provide biological plausibility, it is difficult to determine the
26 extent to which long-term NO2 exposure is independently associated with respiratory
27 effects or if NO2 is a marker for the effects of another traffic-related pollutant or mix of
28 pollutants. Overall, the evidence for asthma incidence, respiratory symptoms, and
29 decrements in lung function and lung function growth and epidemiologic and
30 toxicological evidence for impaired host defense but some uncertainty regarding the
31 independent effects of ambient NO2 exposure is sufficient to conclude that there is likely
32 to be a causal relationship between long-term NO2 exposure and respiratory effects.
33
November 2013 5-76 DRAFT: Do Not Cite or Quote
-------
Table 5-7 Annual ambient NO2 concentrations in prospective studies
examining relationships in children with respiratory health effects in
children.
Study-Cohort or Location
Annual Mean NO2 Concentration, and/or Range, and/or IQR
(PPb)
Asthma Incidence
Jerrett et al. (2008)
CHS
Within community IQR 6.2
Between community IQR 28.9
Annual mean range across communities 9.6 to 51.3
McConnell et al. (2010)
CHS
Mean 20.4
Range 23.6
Gehrinq et al. (2010)
PIMA/the Netherlands
Mean 13.5
Range 6.7 to 31.1
IQR 5.53*
Carlsten et al. (2011c)
Vancouver High Risk Birth Cohort
Mean (SD) 17.3 (3.1)*
Leeetal. (201 2c)
TCHS
Gruzieva et al. (2013)
BAMSE
Oftedal et al. (2009)
Oslo Norway cohort
Shima et al. (2002)
Clark etal. (2010)
Cloughertv et al. (2007)
Nishimura etal. (201 3a)
High Mean 22.1
Low mean 14.0
Approximate overall range 10 to 25
Mean NO2 decrease over time from 1 1 .4 to 4. 1
5thto95th%24.9*
1st year of Life mean 20.9
Min/Max0.8to44.7
IQR decrease over time from 1 55.5 to 1 0.4*
Mean range over study communities 7.3 to 31 .4
First year: 15.9 (2.9) ppb; 25% to 75%: 13.9 to 17.6 ppb
Full cohort; 14.6 (2.3) ppb
Mean range across all communities 9.9 to 32.1 ppb; All 19.3 (8.0) ppb
Pulmonary Function
Breton etal. (2011)
CHS
Gauderman et al. (2004)
CHS
Roias-Martinez et al. (2007a)
Mexico City School children
Range 33.9
Over approximately 3 to 40
Range 34.6
Mean (SD) range across communities 27.2 (10.9) to 42.6 (13.2)
Overall mean 34.4
IQR 12
November 2013
5-77
DRAFT: Do Not Cite or Quote
-------
Table 5-7 (Continued): Annual ambient NO2 concentrations in prospective studies examining relationships in
children with respiratory health effects in children.
Annual Mean NO2 Concentration, and/or Range, and/or IQR
Study-Cohort or Location (ppb)
Schultzetal. (2012) 5th to 95th percentile: 25*
BAMSE
Molteretal. (2013) Range 9.0 to 11.7
Oftedal et al. (2009) Mean 1st year life 20.8
Oslo Norway Cohort Mean lifetime 15.4
Annual mean 14.4
IQR 14.6*
Respiratory Symptoms
McConnell et al. (2003) Between Communities Mean(SD) 19.4(11.3)
CHS Range 4.2 to 38.0
Within Communities
Mean (SD) 4.9 (4.0)
Range 1.1 to 12.8
Gehrinq et al. (2010) Annual mean 13.5
PIMA Range 6.7 to 31.0
IQR 5.53*
Gruzieva et al. (2013) 5 to 95 % 24.9
BAMSE Mean decrease 11.4 to 4.1*
Hansel et al. (2008) Indoor mean (SD) 30.0 (33.7)
African-American Baltimore School Range 2.9 to 394.0
Study mean 25.7
Belanqeret al. (2013) Indoor mean 10.6 (SD = 9.4) ppb
New England Indoor Children's Asthma Study IQR 4.5 - 12.5 ppb
'Estimated using land use regression.
November 2013 5-78 DRAFT: Do Not Cite or Quote
-------
Table 5-8 Annual ambient NO2 concentrations prospective studies examining
relationships with respiratory health effects in adults.
Study
Health Endpoint
Cohort or Location
Annual Mean NO2 Concentration,
and/or Range,
and /or IQR (ppb)
Jacquemin et al. (2009b)
Asthma incidence
ECRHS Europe
Medians by Center range from 6.4 to 30.3
Effects determined for a 5.3 change*
Jacquemin et al. (2009a)
Asthma Incidence
ECHRS
Medians by Center range from 6.4 to 30.3
Effects determined for a 5.3 change*
Modiq et al. (2009)
Asthma Incidence
Rhine cohort
Overall mean Winter 9.6 for 3 cities
Effects determined for a 5.3 change*
Castro-Giner et al. (2009)
Asthma Incidence
ECRHS
Median by center ranges from 6.4 to 30.3
Effects determined for a 5.3 change*
Sunyeret al. (2006)
Chronic Bronchitis
ECRHS
Mean 20.2
Range by center 3.28 to 40.5
Effects per increase of 15.9*
Gotschi et al. (2008)
Pulmonary Function
ECRHS
Medians by Center range from 6.4 to 30.3
Effects determined for a 5.3 change*
Pujades-Rodriquez et al. (2009)
Pulmonary Function
Nottingham U.K.
IQR Differences compared 18.1 to 19.1*
Andersen et al. (2011)
COPD First hospital admission
Danish cohort
Range <5.3 to 21.3*
Median 8.1
IQR 3.1
Ganetal. (2013)
COPD First hospital admission
Vancouver Canada Cohort
Mean 17(SD4.3)
IQR 4.47
Range 8.0 to 30.1
Andersen et al. (2012)
Asthma Hospital Admission
Danish cohort
Range <5.3 to 21.3*
Median 8.1
IQR 3.1
Neupane et al. (2010)
Hospitalization for community acquired pneumonia
Hamilton, Ontario
Mean(SD) 15.25(2.7)
Range 8.79 to 25.67
'Estimated with land use regression
November 2013
5-79
DRAFT: Do Not Cite or Quote
-------
Table 5-9 Summary of evidence supporting a likely to be a causal relationship
between long-term NO2 exposure and respiratory effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2
Concentrations
Associated with
Effects0
Consistent
associations from
multiple, high
quality
epidemiologic
studies with
relevant
exposures
Consistent evidence for increases in
asthma incidence in children in U.S.
multicity cohort and diverse cohorts in
Europe, Canada, and Asia.
Asthma ascertainment by parental report
of doctor diagnosis.
Associations found with NO2 measured
outside children's homes, at central
monitoring sites, modeled using land-use
regression, inverse distance-squared we
Epidemiologic evidence for decrements
in lung function and partially irreversible
decrements in lung function growth in
children
Coherence with evidence for increases in
respiratory symptoms in children with
asthma.
Supporting epidemiologic evidence for
asthma incidence in adults and
respiratory hospital admissions,
respiratory symptoms, respiratory
mortality.
Jerrett et al. (2008).
McConnell et al. (2010).
Gehrinq et al. (2010).
Carlsten et al. (2011 c).
Gruzieva et al. (2013),
Oftedal et al. (2009),
Shima et al. (2002),
Nishimura et al. (2013a),
Clark etal. (2010).
Clouqhertv et al. (2007)
Table 5-1, Figure 5-3.
Breton etal. (2011).
Gauderman et al. (2004),
Rojas-Martinez et al.
(2007a),
Schultzetal. (2012),
Oftedal et al. (2008)
McConnell et al. (2003).
Gehrinq etal. (2010).
Gruzieva etal. (2013).
Hansel et al. (2008),
Belanqeret al. (2013)
Table 5-2. Figure 5-5.
Table 5-3.
Jacquemin et al. (2009b),
Jacquemin et al. (2009a),
Modiq et al. (2009),
Castro-Giner et al.
(2009).
Sunver et al. (2006)
Hospital admissions:
COPD Andersen et al.
(2011). Ganetal. (2013)
Asthma
Andersen etal. (2012)
And community-
acquired pneumonia
Neupane etal. (2010)
Overall annual
means from studies:
13.5-20.0 ppb
Community-specific
means:
9.6-51.3 ppb.
Table 5-7
Overall annual
means from studies:
8-20 ppb
Community-specific
annual means:
3-40 ppb
Table 5-8
November 2013
5-80
DRAFT: Do Not Cite or Quote
-------
Table 5-9 (Continued): Summary of Evidence Supporting a Likely to be Causal Relationship between Long-
term NO2 Exposure and Respiratory Effects.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2
Concentrations
Associated with
Effects0
Uncertainty
remains regarding
independent
effects of NO2
Associations with respiratory symptoms,
pulmonary function remain robust with
adjustment for O3, SO2, PM-io, PM2.s,
PM-io-2.5, EC, or OC but analysis is limited
When reported correlations between
copollutants were often high
Across outcomes, associations found
with adjustment for various SES
indicators, family history of asthma,
smoking exposure, housing
characteristics, presence of gas stove,
temperature, and humidity
Several studies indicate increases in
respiratory symptoms related to indoor
NO2 exposure
McConnell et al. (2003).
Hwang and Lee (2010),
Dong etal. (2011),
Hansel et al. (2008):
Hwang et al. (2005),
Lee etal. (2012c).
Roias-Martinez et al.
(2007b)
See Table 5-5
Belanqeret al. (2013).
Hansel et al. (2008)
Coherence with
respiratory effects
of short-term
exposure
Across disciplines, results consistently
demonstrate increases in asthma
morbidity.
Particular coherence with effects in
children with asthma.
Table 4-23
Limited biological
plausibility for
effects on asthma
provided by
toxicological
evidence
Increased AHR in a few studies of guinea
pigs with long-term or short-term NO2
exposure
Kobavashi and Miura
(1995)
Kobavashi and Shinozaki
(1990)
1,000-4,000 ppbfor
6-12 weeks
4,000 ppb for 7 days
Toxicological
evidence for
impaired host
defense
Increased mortality of mice and monkeys
with NO2 exposure and challenge with
bacterial or viral infection.
Henry etal. (1970).
Ehrlich and Henry (1968),
Ehrlich(1980),
Miller et al. (1987)
500 ppb for 3 mo,
5,000 ppb for 2 mo,
200 ppb base plus
daily spike of 800
ppb for 16-52 weeks
Toxicological Increases in edema, hypertrophy of lung
evidence for epithelium, fibrotic changes in adult not
changes in lung juvenile animals. Uncertain relevance to
morphology epidemiologic findings
Kubotaetal. (1987).
Havashietal. (1987)
500 ppbfor 19 mo,
4,000 ppb for 9-27
mo
November 2013
5-81
DRAFT: Do Not Cite or Quote
-------
Table 5-9 (Continued): Summary of Evidence Supporting a Likely to be Causal Relationship between Long-
term NO2 Exposure and Respiratory Effects.
Rationale for
Causal
Determination3
Some evidence
for key events to
inform mode of
action
Modification of
innate and
adaptive
immunity
Key Evidence13
Increased IgE-mediated mast cell
histamine release in guinea pigs with
long-term exposure
Increased macrophage infiltration to lung
tissue or increased # of lymphocytes in
Key References'3
Fujimaki and Nohara
(1994).
Gregory et al. (1983).
Kumae and Arakawa
(2006)
NO2
Concentrations
Associated with
Effects0
4,000 ppb for 12
weeks
200, 500, 2,000 pp
BALF of experimental animals
Epidemiologic evidence for allergic
sensitization in children
With short-term exposure: Upregulation
of inflammatory cytokines (IL-10, IL-5,
IL-13) and inflammatory adhesion
molecule ICAM-1 in healthy humans and
increased numbers of eosinophils in nose
in animals.
Pathmanathan et al.
(2003), Ohashi et al.
(1994)
2,000 ppb over 4
consecutive days;
3,000 ppb NO2 for 2
weeks
Inflammation Epidemiologic evidence for increases in
increases in eNO, elevated number of
nasal eosinophils)
(Renzettietal..20Q9).
(Dales etal., 2008)
Oxidative Stress
Animal toxicology models: Increased lipid
peroxidation
Alterations in the glutathione antioxidant
pathway
Arner and Rhoades
(1973), Saqai et al.
(1984). Gregory et al.
(1983). Avaz and
Csallanv(1978)
2,900 ppb
400-4,000 ppb
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table I
Preamble.
""Describes the key evidence and references contributing most heavily to causal determination and, where applicable, to
uncertainties or inconsistencies. References to earlier sections indicate where full body of evidence is described.
°Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
ofthe
November 2013
5-82
DRAFT: Do Not Cite or Quote
-------
5.3 Cardiovascular Effects
5.3.1 Introduction
1 The 2008 ISA for Oxides of Nitrogen concluded that "the available epidemiologic and
2 toxicological evidence was inadequate to infer the presence or absence of a causal
3 relationship" for cardiovascular effects related to long-term NO2 exposure.
4 This section reviews the published studies pertaining to the cardiovascular effects of
5 NOX exposure in humans, animals, and cells; study details can be found in Table 5-10
6 and Table 5-11. With the limited existing body of evidence serving as the foundation,
7 emphasis was placed on studies published since the 2008 ISA for Oxides of Nitrogen.
8 The recent epidemiologic and toxicological publications add to the evidence for
9 independent effects of NOX exposure on cardiovascular morbidity. For epidemiologic
10 studies, emphasis was placed on longitudinal analyses of incident cardiovascular diseases
11 with consideration of multiple potential confounding factors. With regard to animal
12 toxicological studies, studies with relevant NO2 exposure concentrations (i.e., less than
13 5,000 ppb) were included.
5.3.2 Cardiovascular Diseases
14 At the completion of the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) only one
15 epidemiologic study had examined the association of cardiovascular disease with long-
16 term exposure to NO2. In this study, Miller et al. (2007) (see Table 5-10 for further
17 details about this and other epidemiologic studies) included 65,893 postmenopausal
18 women, between the ages of 50 and 79 years, without previous CVD residing in 36 U.S.
19 metropolitan areas from 1994 to 1998. Exposures to air pollution were estimated by
20 assigning the annual mean levels of air pollutants in 2000 measured at the monitor
21 nearest the subject's residence, based on its five-digit ZIP Code centroid. In the single-
22 pollutant model results, PM2 5 showed the strongest associations with the CVD events
23 (MI, revascularization, angina, CHF, CHD death) among pollutants, followed by SO2.
24 NO2 was not associated with the overall CVD events (hazard ratio [HR]: 0.98 [95% CI:
25 0.89, 1.08] per 10-ppb increase in the annual average) when the dataset was restricted to
26 those who were not missing exposure data. Several recent studies of the association of
27 long term NO2 exposure with preclinical and clinical cardiovascular outcomes add to the
28 available body of evidence.
November 2013 5-83 DRAFT: Do Not Cite or Quote
-------
1 Lipsettetal. (2011) used Cox proportional hazards regression to analyze the association
2 of incident stroke and MI with long-term exposure to NO2, NOX, other gases (CO, O3,
3 SO2) and PM. These authors studied a cohort of California public school teachers aged
4 20-80 years old (n = 124,614). Approximately 40% of those initially contacted
5 participated in the study, with approximately 80% continuing their participation during
6 subsequent follow-up contacts (Bernstein et al.. 2002). Each participant's geocoded
7 residential address was linked to pollutant surfaces that were determined by inverse
8 distance weighted (IDW) interpolation of pollutant concentrations measured at fixed site
9 monitors during the period 1996-2005. The average of monthly NO2 concentrations was
10 modeled as a time dependent function for subjects with at least 12 months of exposure.
11 Those living outside the radial range for which the monitor was intended to provide
12 representative data were excluded from the analysis. This "representative range" was 3
13 km for neighborhood NOX and NO2 monitors and 5 km for the urban/regional NO2 and
14 NOX monitors. The authors observed a positive association between NO2 and incident
15 MI (HR: 1.06 [95%CI: 0.88, 1.27] per 10-ppb, respectively) and a weak association with
16 incident stroke (HR: 1.02 [95%CI: 0.90, 1.115] per 10-ppb, respectively). Point estimates
17 for the association of other pollutants (PM2 5, SO2 and O3) with incident stroke were
18 increased and the association between PMi0 and incident stroke was significantly
19 increased. Fewer observations were available for the NOX compared to PM analyses
20 because the requirements for the participants' proximity to the monitor were more
21 stringent for NOX (residing within 5 km as opposed to 20 km for PM).
22 Gan et al. (2011) used Cox proportional hazards regression to examine the association of
23 long-term exposure to black carbon, PM2 5 NO2, and NO with CHD hospitalization and
24 mortality among participants (45-85 year-olds) in the universal health insurance system
25 residing in Vancouver, Canada (n = 418,826). In this study, land use regression was used
26 to predict 5-year average concentrations at a resolution of 10 meters. These predicted
27 concentrations were adjusted using factors derived from regulatory monitoring data and
28 linked to each participant's postal code of residence. After adjustment for sex, age,
29 comorbidity, and SES, NO2 and NO were inversely associated with CHD hospitalization
30 (HR: 0.93 [95%CI: 0.89, 0.98] and HR: 0.96 [95% CI: 0.92, 1.00] per 10 ppb); however,
31 positive associations of NO2 and NO with CHD mortality were observed (Section 5.5.2).
32 Atkinson et al. (2013) applied Cox proportional hazards regression to examine the
33 association of incident cardiovascular disease with NO2. These authors studied patients
34 (aged 40-89 years) registered with 205 general practices across the U.K. The authors
35 report that approximately 98% of the population is registered with a general practitioner
36 minimizing the potential for selective participation. Predicted annual average NO2
37 concentrations within 1 km by 1 km grids, estimated using dispersion models, were
38 assigned to participants based on their residential postal code. Cardiovascular disease
November 2013 5-84 DRAFT: Do Not Cite or Quote
-------
1 outcomes included in the analysis were MI, stroke, arrhythmias, and heart failure.
2 Authors reported a positive association between NO2 and heart failure in fully adjusted
3 models (HR: 1.11 [95% CI: 1.02, 1.21] per 10 ppb). Incident MI, stroke and arrhythmia
4 were not associated with NO2 concentration in this analysis. A similar pattern of findings
5 were observed for the associations between PM and these outcomes (associations with
6 CHD, MI and stroke were null while the association of PMi0 with heart failure was
7 increased). Rosenlund et al. (2009a) conducted a case control study of first MI reported
8 between 1985 and 1996 using the registry of hospital discharges and deaths for
9 Stockholm County, Sweden and randomly selected population-based controls. Predicted
10 5-year average NO2 concentration was determined and linked to each participant's
11 geocoded address using dispersion models. The resolution of the predicted concentrations
12 corresponded to 500 meters in the countryside, 100 meters in urban areas and 25 meters
13 in the inner city. Multinomial logistic regression was performed to obtain association of
14 most 5-year average NO2 concentration with incident MI. This metric, which was
15 designed to capture traffic exposure, was associated with fatal MI (OR: 1.14 [95%CI:
16 1.09, 1.19] per 10 ppb) but not with non-fatal MI (OR: 0.96 [95%CI: 0.93, 1.00] per 10
17 ppb). CO and PMi0 were also associated with fatal cases of MI in this population.
18 Epidemiologic studies using a variety of exposure assessment methods (e.g., dispersion
19 modeling, land use regression and central site monitor concentration nearest to the
20 subject's residence) provide some evidence that long term exposure to NO2 may be
21 associated with the risk of cardiovascular diseases including MI and heart failure. The
22 association with heart failure is reported in one large, well conducted study and was
23 robust to adjustment for multiple potential confounding factors including age, sex,
24 smoking, BMI, and pre-existing medical conditions (Atkinson et al.. 2013). The
25 association of long term NO2 exposure with incident MI is also reported in one well
26 conducted study after adjustment for a similar array of confounding factors (Lipsett et al..
27 2011); however associations of NO2 with MI were not observed consistently across
28 generally comparable studies (Atkinson et al.. 2013; Rosenlund et al.. 2009a). Further,
29 the observed associations of MI and heart failure with long term NO2 exposure are not
30 supported by a study of CHD hospital admissions (Gan et al.. 2011) nor are they
31 supported by Miller et al. (2007) who found a null association between NO2 and
32 cardiovascular events combined (MI, revascularization, angina, CHF, CHD death) among
33 post-menopausal women (a positive association with PM2 5 was observed in this study).
November 2013 5-85 DRAFT: Do Not Cite or Quote
-------
Table 5-10 Epidemiologic studies of long-term exposure to NO2 or NOX and
effects on the cardiovascular system.
Study
Miller et al.
(2007)
Cohort (location)
Study Period Mean (ppb)
WHI Cohort (U.S.) NR
1994-1998
Exposure Assessment
Annual avg (2000):
nearest monitor to
residence ZIP code
centroid
Effect Estimates
CVD Events
HR*: 0.98(0.89, 1
per 10 ppb
(95% Cl)
.08)
Covariates: age, ethnicity,
education, household income,
smoking, diabetes, hypertension,
systolic blood pressure, BMI, and
hypercholesterolemia
*HR for subjects with non-missing
exposure data
Lipsett et al.
(2011)
CIS Cohort
(CA, U.S.)
June 1996-
Dec 2005
NO2
IQR: 10.29
Mean: 33.59
NOX
IQR: 58.31
Mean: 95.6
Geocoded residential
address linked to
pollutant surface
developed using IDW
(fixed site monitors
concentrations from
1995-2005 used to
model exposure as a
time dependent
function)
Ml Incidence
NOX: HR 1.01 (0.01, 1.11)
NO2: HR 1.06 (0.88, 1.28)
Stroke Incidence
NOX: HR 1.02(0.96, 1.09)
NO2: HR 1.02 (0.90, 1.16)
per 10 ppb NO2 and 20 ppb NOX
f^rtw^rio+^e- • o«^ r^r*c\ c- t~t~>r\\s'mn
second-hand smoke, BMI, lifetime
physical activity, nutritional
factors, alcohol, marital status,
menopausal status, I hormone
therapy, hypertension medication
and aspirin, family history of
Ml/Stroke
Atkinson et al. National GP Patient IQR: 5.7 ppb
(2013) Cohort Mean (SD):
(U.K.) 12.0
2003
Annual average NO2
concentration for 2002
at a 1 by 1 km resolution
derived from emission-
based models and
linked to residential post
codes.
Ml Incidence
HR: 0.97(0.90, 1.04)
Stroke Incidence
HR: 0.98(0.91, 1.06)
Arrhythmia Incidence
HR: 0.98(0.91, 1.04)
Heart Failure Incidence
HR: 1.11 (1.02, 1.21)
per 10 ppb
Covariates: age, sex, smoking
BMI, diabetes, hypertension,
index of multiple deprivation
Rosenlund et
(2009a)
_ai SHEEP Study
(Stockholm,
Sweden)
1985-1996
5th-95th 15.9
cases: Median
(cases); 6,
Median: 6
(controls)
.9
3
5-yr average NO2
concentration
by dispersion
assessed
modeling
First Nonfatal Ml
OR: 0.96 (0
per 10 ppb
Covariates:
year and SE
.93,
age
:S
1.00)
, sex,
calendar
November 2013
5-86
DRAFT: Do Not Cite or Quote
-------
Table 5-10 (Continued): Epidemiologic studies of long-term exposure to NOX and effects on the
cardiovascular system.
Ganetal. (2011)
Population based
cohort
(Vancouver,
Canada)
1999-2002
Mean (NO2): LUR, 5-yr average
16.3 concentration
IQR(N02):4.5 (1995-1998) and 4 yr
avg concentration
Mean (NO): (1999-2002)
26.1
IQR(NO): 10.8
CHD Hospitalization:
(ICD9 410-414)
RR(NO2): 0.93(0.89, 0.98)
RR(NO): 0.96(0.92, 1.00)
per 10 ppb NO2 and NO
Covariates: age, sex, preexisting
diabetes, COPD, hypertension,
WHI = Women's Health Initiative;
CIS = California Teachers Study;
GP=General Practice;
IDW= Inverse Distance Weighted;
SHEEP=Stockholm Heart Epidemiology Program
5.3.3 Markers of Cardiovascular Disease Risk
1 Epidemiologic and toxicological studies have also investigated the effects of long-term
2 NO 2 exposure on risk factors and markers of cardiovascular disease risk, such as arterial
3 stiffness, circulating lipids, and HRV. A recent experimental animal study also reported
4 changes in markers that are characteristic of vascular disease and progression (see Table
5 5-11 for toxicology study details). Mice were exposed for 50 days to various
6 multipollutant atmospheres (diesel or gasoline exhaust, wood smoke, or simulated
7 "downwind" coal emissions) comprising varying concentrations of NO2 (0-3,670 ppb). A
8 data mining technique known as Multiple Additive Regression Trees analysis was
9 employed to identify associations between the 45 different exposure component
10 categories, including NO2, and various effects [markers of oxidative stress (discussed in
11 Section 5.3.4) and cardiovascular disease stability and progression (endothelin-1 (ET-1),
12 matrix metalloproteinase (MMP)-3, MMP-7, MMP-9, tissue inhibitor of
13 metalloproteinase-2 (TIMP-2))]. The results demonstrated NO2 was among one of the
14 strongest predictors of responses. More specifically NO2 ranked among the top 3
15 predictors for ET-1 and TIMP-2 (Seilkop et al.. 2012).
16 Hyperlipidemia is recognized as a risk factor for cardiovascular disease. Takano et al.
17 (2004) reported that obese rats (Otsuka Long-Evans Tokushima Fatty) had elevated
18 levels of triglycerides and decreased HDL and HDL/total cholesterol levels after long-
19 term exposure to 160 ppb NO2 compared to clean air. HDL levels were also decreased
20 after 800 ppb NO2 exposure in the obese strain and in the non-obese rats (Long-Evans
21 Tokushima). The authors suggested that obese animals were at greater risk of
22 dyslipidemia following NO2 exposure.
November 2013
5-87
DRAFT: Do Not Cite or Quote
-------
1 The effects of NO2 in relation to autonomic function in a random selection of Swiss
2 cohort study participants have also been examined. In this study, Felber Dietrich et al.
3 (2008) linked measures of HRV to annual NO2 concentration at the participant's
4 residential address using dispersion model predictions supplemented with land use and
5 meteorological data. Annual average NO2 concentration was associated with decreased
6 SDNN, nighttime LF, and LF/HF ratio in women. No associations with other parameters
7 of HRV were observed in these data.
8 The 1993 NOX Air Quality Criteria Document (U.S. EPA. 1993) reported a significant
9 reduction in HR in rats exposed to 1,200 and 4,000 ppb NO2 for 1 month, but not after
10 lower concentration or longer durations of exposure (Suzuki etal.. 1981). There were no
11 changes in vagal responses in rats exposed to 400 ppb NO2 for 4 weeks (Tsubone and
12 Suzuki. 1984).
13 A number of null findings related to changes in hematological parameters were reported
14 in the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). Hematocrit and hemoglobin
15 levels were unchanged in squirrel monkeys (Tenters et al.. 1973). rats (Suzuki et al..
16 1981). or dogs exposed to < 5,000 ppb NO2 (Wagner et al.. 1965). However, Furiosi et al.
17 (1973) did report polycythemia due to reduced mean corpuscular volume and an
18 increased trend in the ratio of neutrophil to lymphocytes in the blood of NO2-exposed
19 monkeys and similar increases in erythrocyte counts in NO2-exposed rats.
20 Overall, a limited number of epidemiologic and toxicological studies have evaluated
21 long-term NOX exposure on risk factors and markers of cardiovascular disease. There is
22 some evidence for increased arterial stiffness, increased markers for cardiovascular
23 disease stability and progression, dyslipidemia, decreased HRV, and reduced HR;
24 however, these effects have only been reported in one study each.
5.3.4 Inflammation and Oxidative Stress
25 Inflammation and oxidative stress have been shown to play a role in the progression of
26 chronic cardiovascular disease. A limited number of studies have evaluated markers of
27 inflammation and oxidative stress. Forbes et al. (2009a) examined the association of
28 predicted annual average NO2 concentration with CRP and fibrinogen among the English
29 population. Multilevel linear regression models were used to determine pooled estimates
30 across three cross-sectional surveys conducted during different years. Each participant's
31 postal code of residence was linked to predicted annual average NO2 concentration
32 derived from dispersion models. NO2 was not associated with increased CRP or
33 fibrinogen in these data nor were PMi0, SO2, or O3. A study conducted among men and
34 women (45-70 year-olds) in Stockholm reported an association of 30-year average
November 2013 5-88 DRAFT: Do Not Cite or Quote
-------
1 traffic-related NO2 concentration estimated using dispersion models with increases in
2 IL-6 and CRP; however, NO2 was not associated with TNF-a, fibrinogen or PAI-1 in this
3 population (Panasevich et al.. 2009). Associations between several metrics of SO2
4 exposure and increased IL-6 and CRP were observed in this study.
5 de Burbure et al. (2007) examined oxidative stress markers in rats on a low (Se-L) or
6 sufficient selenium (Se-S) diet exposed to 1,000 ppb NO2 for 28 days. Blood Se levels
7 decreased significantly in both groups immediately after the 28-day exposure and
8 continued to decrease in the Se-S following a 48 hour recovery period. Glutathione
9 peroxidase (GPx), of which Se is an integral component, also decreased immediately and
10 48 hours after exposure only in the plasma of Se-S rats. However, GPx levels increased
11 in RBC of Se-L rats immediately after the 28-day exposure and increased in both groups
12 48 hours later. RBC SOD activity increased in both groups immediately after the
13 exposure and decreased in Se-L rats 48 hours later. GST was increased for both groups
14 immediately after the 28-day exposure and continued to increase after the 48 hour
15 recovery period potentially compensating for the increase in TEARS immediately after
16 exposure.
17 As discussed in Section 5.5.3. Seilkop etal. (2012). examined the effects of NO2
18 exposure, in a multipollutant context, on markers of oxidative stress (heme oxygenase-1
19 [HO-1] expression and thiobarbituric acid reactive substances [TBARs], indicator of lipid
20 peroxidation) in ApoE"7" mice fed a high-fat diet. Mice were exposed to various
21 atmospheres (diesel or gasoline exhaust, wood smoke, or simulated "downwind" coal
22 emissions) with varying concentrations of NO2 (0-745 ppb) for 50 days. Associations
23 between the oxidative stress indicators and the 45 different exposure component
24 categories were determined using a data mining technique known as Multiple Additive
25 Regression Trees analysis. The results demonstrated NO2 was among one of the strongest
26 predictors of responses. Although HO-1 was not highly correlated with NO2; NO2, SO2,
27 and NO ranked among the top 3 predictors for TBARs.
November 2013 5-89 DRAFT: Do Not Cite or Quote
-------
Table 5-11 Study details for toxicological studies examining cardiovascular
effects from long-term NO2 exposure
Study
de Burbure et al.
(2007)
Renters et al.
(1973)
Furiosi et al.
(1973)
Seilkop et al.
(2012)
Suzuki et al.
(1981)
Takano et al.
(2004)
Tsubone and
Suzuki (1984)
Wagner et al.
(1965)
Species
(Strain);
Lifestage;
Sex; n
Rats(Wistar);
8 weeks; M;
n = 8/group
Squirrel
monkeys;
Adult; M;
n = 4
Monkeys
(Macaca
spec/osa);
Adult; M/F;
n = 4
Rats(Sprague
-Dawley);
4 weeks; M;
n = 8
Mice
(ApoE"'"); 10
weeks; M;
n = 8-10
Rats; NR;
NR; n = 6
Rats (OLETF
and LETO);
4 weeks; M;
n = 10-14
Rats(Wistar);
9-13 weeks;
M; n = 6
Dogs; Adult;
M; n = NR
Exposure Details
(Concentration; Duration)
High (6 ug/day) or low
(1.3 ug/day) selenium;
(1)1,000ppb, 28 days, 6 h/day,
5 days/week (Se+/Se-);
(2) 10,000 ppb, 28 days, 6 h/day,
5 days/week;
(3) 5,000 ppb, 5 days, 6 h/day;
(4) 50,000 ppb, 30 min
1,000 ppb NC>2, continuously for
16 mo; challenged with influenza
virus
(1) 2,000 ppb NO2, continuously
for 14 mo
NO2 (along with 700 other
components) Fed a high-fat diet;
260, 745, and 3,670 ppb (along
with dilutions of 1/3 and 1/10);
6 h/day, 7 days/week for 50 days
400, 1,200, and 4,000 ppb NO2;
1, 2, and 3 mo
160, 800, or 4,000 ppb NO2;
continuously for 32 weeks
400 and 4,000 ppb NO2;
continuously for 1 and 4 weeks,
respectively; Immediately after
exposure animals were injected
with 5 ug/kg bw phenyl diguanide
1,000 or 5, 000 ppb NO2;
continuously for 18 mo
Endpoints Examined
GPx in plasma and red blood cell lysate;
SOD activity in red blood cell lysate; GST
activity in red blood cell lysate; TEARS in
plasma. Endpoints examined immediately
and 48 h after exposure
Hemoglobin and hematocrit levels were
measured throughout the study.
Erythrocyte, hematocrit, and hemoglobin
levels were measured throughout the
study
ET-1, VEGF, MMP3, MMP7, MMP9,
TIMP2, HO-1, TEARS in proximal aorta
18-h after exposure
HR and hemoglobin levels measured after
1, 2, and 3 mo exposures.
BW, Triglyceride, HDL, total cholesterol,
HDL/total cholesterol, sugar measured 8
weeks following exposure
HR was measured 10 sec after injection
Hemoglobin and hematocrit levels were
measured quarterly throughout exposure.
November 2013
5-90
DRAFT: Do Not Cite or Quote
-------
5.3.5 Cardiovascular Mortality
1 Results of studies of long-term exposure to NO2 and cardiovascular diseases are coherent
2 with findings reporting associations of long-term NO2 exposure with all cause and
3 cardiovascular mortality. Consistent, positive associations with total mortality, as well as
4 deaths due to cardiovascular disease have been observed in cohort studies conducted in
5 the U.S. and Europe (Section 5.5.2. Figure 5-10. and Table 5-17). Specifically, the
6 strongest evidence comes from a number of recent studies that have observed positive
7 associations between exposure to NOX and NO2 and IHD mortality (Cesaroni et al..
8 2013; Chen etal.. 2013; Lipsett et al.. 2011; Yorifuji etal.. 2010). mortality due to
9 coronary heart disease (Gan etal.. 2011; Rosenlund et al.. 2008b). and circulatory
10 mortality (Yorifuji et al.. 2010; Jerrett et al.. 2009). Coherence is also provided for the
11 effect of long-term exposure and cardiovascular effects by the evidence from studies of
12 short-term cardiovascular mortality and morbidity (Section 4.3).
5.3.6 Summary and Causal Determination
13 Overall, the evidence is suggestive of a causal relationship between long-term exposure
14 to NO2 and cardiovascular effects. This conclusion is based on the consideration of recent
15 prospective epidemiologic studies reporting associations of NO2 or NOX with CHF, MI
16 and stroke although associations with these cardiovascular outcomes were not
17 consistently observed across studies. This current conclusion represents a change from
18 the conclusion drawn in the 2008 ISA for Oxides of Nitrogen, which stated that the
19 evidence was inadequate to infer the presence or absence of a causal relationship.
20 Although cardiovascular morbidity effects are not consistently observed across
21 epidemiologic studies, some support is provided by a limited body of evidence
22 demonstrating biological plausibility, as well as consistent associations between long-
23 term NO2 exposure and cardiovascular mortality. The evidence for cardiovascular effects
24 with respect to the causal determination for long-term NO2 exposure is detailed below
25 using the framework described in Table II of the Preamble to this ISA. The key evidence
26 as it relates to the causal framework is summarized in Table 5-12.
27 The 2008 ISA for Oxides of Nitrogen concluded that the available evidence was
28 inadequate to infer the presence or absence of a causal relationship between long-term
29 NO2 exposure and cardiovascular disease. Miller et al. (2007) found no association
30 between long-term NO2 exposure and cardiovascular events among post-menopausal
31 women enrolled in the WHI study, although an association with PM2 5 was observed.
32 Several studies evaluating hematological parameters reported mixed results that included
33 no changes in hematocrit or hemoglobin and increased erythrocyte count.
November 2013 5-91 DRAFT: Do Not Cite or Quote
-------
1 Recent, large and well conducted prospective epidemiologic studies provide some
2 evidence that long-term exposure to NO2 is associated with heart failure, MI and stroke.
3 The association with heart failure was reported in a large study with high participation
4 rates using a validated database of doctor diagnosed cardiovascular outcomes and
5 persisted after adjustment for multiple potential confounding factors including age, sex,
6 smoking BMI, SES and pre-existing medical conditions (Atkinson et al.. 2013). The
7 association of long term NO2 exposure with incident MI and long-term NOX exposure
8 with stroke was reported in another large study with adjustment for a similar array of
9 potential confounding factors and which employed a refined exposure assessment
10 strategy (i.e., residence with 3-5 km of a monitor) (Lipsett et al., 2011). An association
11 between NO2 exposure and MI were not consistently reported across studies of generally
12 comparable quality (Atkinson et al., 2013; Rosenlund et al.. 2009a). Additionally, studies
13 examining associations of cardiovascular effects with other pollutants (i.e., PMi0, PM2 5,
14 CO, SO2, O3), in addition to NO2 often reported associations with these other pollutants.
15 Because IHD includes MI and can lead to heart failure there is some coherence across the
16 cardiovascular endpoints associated with NO2 exposure in the epidemiologic studies
17 described above. Additionally, consistent associations from multiple, high-quality
18 epidemiologic studies of cardiovascular mortality support findings for cardiovascular
19 morbidity effects. Studies in human populations offer limited support that long term NO2
20 may be associated with arterial stiffness (Lenters etal.. 2010) and decreased HRV (Felber
21 Dietrich et al., 2008). Experimental animal studies report that NO2 exposure was
22 associated with dyslipidemia and increases in some markers of oxidative stress and
23 vascular function, but each has been evaluated each in one study. There is weak evidence
24 to describe a biologically plausible mechanism for the NO2 related cardiovascular effects
25 observed, potentially through the induction of oxidative stress and systemic
26 inflammation. Evidence for these events is provided by both animal toxicological studies
27 that report increased markers of oxidative stress and inflammatory markers as well as
28 cytokines after short-term NO2 exposure and human epidemiologic studies that report
29 increased CRP and IL-6 after long-term NO2 exposure. Overall, the evidence from some
30 epidemiologic studies of cardiovascular morbidity, in consideration of the consistent
31 associations observed between NO2 exposure and cardiovascular mortality and the
32 limited evidence demonstrating biological plausibility, is suggestive of a causal
33 relationship between long term NO2 exposure and cardiovascular morbidity.
November 2013 5-92 DRAFT: Do Not Cite or Quote
-------
Table 5-12 Summary of evidence supporting a suggestive relationship between
long-term NO2 exposure and cardiovascular effects.
Rationale for
Causal
Determination3
Key Evidence13
Key References'3
NO2 Concentrations
Associated with
Effects0
Cardiovascular Morbidity - Suggestive
At least one well
conducted study
reporting
associations with
heart failure and Ml
Evidence from prospective studies in
New England and California for heart
failure, Ml, and stroke in association
with long term modeled NO2
concentration.
Atkinson et al. (2013)
Lipsettetal. (2011)
But, positive associations with Ml not Atkinson et al. (2013)
observed consistently
Rosenlund et al.
(2009a)
Mean annual avg
(2002): 12.0 ppb
Mean avg (1996-2005):
33.59 ppb
Mean annual avg
(2002): 12.0 ppb
Median of 5-yr avg: 6.9
ppb (cases) 6.3 ppb
(controls)
Consistent
associations from
multiple, high-quality
epidemiologic
studies of
cardiovascular
mortality
Supporting evidence
of decreased
markers of
cardiovascular
disease risk
Adequate
consideration of
potential
confounding
Consistent evidence for increases in
risk of cardiovascular mortality in
adults in diverse populations and
applying diverse methods
Strongest evidence of mortality from
IHD, CHD, and circulatory diseases
Increased arterial stiffness measured
by pulse wave velocity and
augmentation index among U.S.
young adults.
Decreased HRV (SDNN, LF) among
Swiss cohort study participants.
Although epidemiologic studies
generally considered important
confounders, studies with similar
designs and approaches for control of
confounding did not consistently
report associations with
cardiovascular effects
Section 5.5.2
Lenters et al. (2010)
Felber Dietrich et al.
(2008)
Miller etal. (2007)
Atkinson et al. (2013)
Lipsettetal. (2011)
Rosenlund et al.
(2009a)
Mean: 18.3 ppb
Mean: 12.1 ppb
Annual svg (2000): NR
Mean annual avg
(2002): 12.0 ppb
Mean avg (1996-2005):
33.59 ppb
Median of 5-yr avg: 6.9
ppb (cases) 6.3 ppb
(controls)
Limited toxicological
evidence with
relevant exposures
Increased triglycerides and decreased
HDL in rats.
Takano et al. (2004)
160 ppb
November 2013
5-93
DRAFT: Do Not Cite or Quote
-------
Table 5-12 (Continued): Summary of evidence supporting a suggestive relationship
between long-term NOi exposure and cardiovascular effects.
Rationale for
Causal
Determination3
Key Evidence13
Key References'3
NC>2 Concentrations
Associated with
Effects0
Weak evidence of
key events that
inform mode of
action
Oxidative Stress
Evidence of increased oxidative
stress in rats with relevant NC>2
exposures (i.e., MDA, TEARS, GPx,
GST).
Evidence of increased oxidative
stress in plasma from NO2-exposed
humans (i.e., LOX-1).
Lietal. (2011 a) Rats: 5,320 ppb NO2
de Burbure et al. (2007) Rats: 1,000 ppb NO2
Channelletal. (2012) Healthy adults: 500 ppb
NO2
Inflammation
Toxicological evidence of increased
transcription of some inflammatory
mediators in vitro (i.e., IL-8) and in
rats (i.e., TNF-a).
Limited epidemiologic support for
increases in CRP and IL-6 in adults
Channelletal. (2012)
Human cells exposed to
plasma from healthy
adults: 500 ppb NO2
Lietal. (2011 a) Rats: 5,320 ppb NO2
Panasevich et al. (2009) Mean: 12.6 ppb
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
""Describes the key evidence and references that contribute most heavily to causal determination, and where applicable, to
uncertainties and inconsistencies. References to earlier sections indicate where full body of evidence is described.
""Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
5.4 Reproductive and Developmental Effects
5.4.1
Introduction
i
2
3
4
5
6
7
8
9
10
The body of literature characterizing the health effects associated with exposure to NO2
is large and continues to grow, including research focusing on birth outcomes, for which
the body of evidence has grown considerably since the 2008 ISA for Oxides of Nitrogen
(U.S. EPA. 2008C). Among the epidemiologic studies, various measures of birth weight
and fetal growth, such as low birth weight (LEW), small for gestational age (SGA),
intrauterine growth restriction (IUGR), and preterm birth (<37-week gestation; [PTB])
have received more attention in air pollution research, while congenital malformations
are less studied. The toxicological studies of similar outcomes measured litter size and
birth weight. Nervous system and respiratory outcomes after early life exposures to NO2
are examined in the developmental toxicology and epidemiology literature.
November 2013
5-94
DRAFT: Do Not Cite or Quote
-------
1 A major issue in studying environmental exposures and reproductive and developmental
2 effects (including infant mortality) is selecting the relevant exposure period, since the
3 biological mechanisms leading to these outcomes and the critical periods of exposure are
4 poorly understood. To account for this, many epidemiologic studies evaluate multiple
5 exposure periods, [including long-term (months to years) exposure periods, such as the
6 entirety of pregnancy, individual trimesters or months of pregnancy; or short-term (days
7 to weeks) exposure periods, such as the days and weeks immediately preceding birth].
8 Due to the length of gestation in rodents (18-24 days, on average), animal toxicological
9 studies investigating the effects of NO2 on pregnancy generally utilize short-term
10 exposure periods, which cover an entire lifestage. Thus, an epidemiologic study that uses
11 the entire pregnancy as the exposure period is considered to have a long-term exposure
12 period (about 40 weeks, on average), while a toxicological study conducted with rats that
13 also uses the entire pregnancy as the exposure period is considered to have a short-term
14 exposure period (about 18-24 days, on average). In order to characterize the weight of
15 evidence for the effects of NO2 on reproductive and developmental effects in a
16 consistent, cohesive and integrated manner, results from both short-term and long-term
17 exposure periods are included in this section and are identified accordingly in the text and
18 tables throughout this section.
19 Due to the poorly understood biological mechanisms and uncertainty regarding relevant
20 exposure periods, all of the studies of reproductive and developmental outcomes,
21 including infant mortality, are evaluated in this section. Exposures proximate to death
22 may be most relevant if exposure causes an acute effect. However, exposure occurring in
23 early life might affect critical growth and development, with results observable later in
24 the first year of life, or cumulative exposure during the first year of life may be the most
25 important determinant. In dealing with the uncertainties surrounding these issues, studies
26 have considered several exposure metrics based on different periods of exposure,
27 including both short- and long-term exposure periods. These studies are characterized
28 here as they contribute to the weight of evidence for an effect of NO2 on reproductive
29 and developmental effects.
30 Although the physical mechanisms are not fully understood, several hypotheses have
31 been proposed for the effects of NO2 on reproductive and developmental effects; these
32 include: oxidative stress, systemic inflammation, vascular dysfunction, and impaired
33 immune function. Study of these outcomes can be difficult given the need for detailed
34 exposure data and potential residential movement of mothers during pregnancy. Air
35 pollution epidemiologic studies reviewed in the 2008 ISA for Oxides of Nitrogen (U.S.
36 EPA, 2008c) examined impacts on birth-related endpoints, including intrauterine,
37 perinatal, postneonatal, and infant deaths; premature births; intrauterine growth
38 retardation; very low birth weight (weight <1,500 grams) and low birth weight (weight
November 2013 5-95 DRAFT: Do Not Cite or Quote
-------
1 <2,500 grams); and birth defects. However, in the limited number of studies included in
2 the 2008 ISA for Oxides of Nitrogen, no associations were found between NO2 and birth
3 outcomes, with the possible exception of birth defects.
4 Several recent articles have reviewed methodological issues relating to the study of
5 outdoor air pollution and adverse birth outcomes (Chen etal.. 2010a: Woodruff etal.
6 2009; Ritz and Wilhelm. 2008; Slama et al.. 2008). Some of the key challenges to
7 interpretation of these study results include the difficulty in assessing exposure as most
8 studies use existing monitoring networks to estimate individual exposure to ambient air
9 pollution; the inability to control for potential confounders such as other risk factors that
10 affect birth outcomes (e.g., smoking); evaluating the exposure window (e.g., trimester) of
11 importance; and limited evidence on the physiological mechanism of these effects (Ritz
12 and Wilhelm. 2008: Slama et al.. 2008).
13 Overall, the number of studies examining the association between exposure to ambient
14 NO2 and reproductive and developmental outcomes has increased tremendously, yet
15 evidence for an association with these outcomes remains relatively uncertain. Recently,
16 an international collaboration was formed to better understand the relationships between
17 air pollution and adverse birth outcomes and to examine some of these methodological
18 issues through standardized parallel analyses in datasets from different countries
19 (Woodruff et al.. 2010). Initial results from this collaboration have examined PM and
20 birth weight (Parker et al., 2011); work on NO2 has not yet been performed. Although
21 early animal studies Shalamberidze and Tsereteli (197la. b) found that exposure to NO2
22 during pregnancy in rats led to some abnormal birth outcomes, human studies to date
23 have reported inconsistent results for the association of ambient NO2 concentrations and
24 birth outcomes.
5.4.2 Fertility, Reproduction, and Pregnancy
5.4.2.1 Effects on Sperm
25 A limited amount of research has been conducted to examine the association between air
26 pollution and male reproductive outcomes, specifically semen quality. To date, the
27 epidemiologic studies have considered various exposure durations before semen
28 collection that encompass either the entire period of spermatogenesis (i.e., 90 days) or
29 key periods of sperm development that correspond to epididymal storage, development of
30 sperm motility, and spermatogenesis.
November 2013 5-96 DRAFT: Do Not Cite or Quote
-------
1 An occupational study of male motorway company employees reported that men with the
2 highest exposures (-160 ppb) had lower sperm motility, but no difference in sperm count,
3 compared to men with lower exposures (~80 ppb) (Boggia et al., 2009). Two
4 epidemiologic studies evaluated the relationship between ambient concentrations of NO2
5 and sperm quality and observed no associations (Rubes et al.. 2010; Sokol et al. 2006).
6 Overall, there is no epidemiologic evidence for an association between exposure to
7 ambient NO2 concentrations and effects on sperm.
8 No recent toxicological studies have examined the effect of NO2 exposure on male
9 reproductive outcomes, specifically semen quality. Kripke and Sherwin (1984) found no
10 significant effects on spermatogenesis, or on germinal and interstitial cells of the testes of
11 a small group of male LEW/f mai rats (n = 6) after 21 days of exposure to a single
12 concentration of NO2, 1,000 ppb 7 h/day, 5 days/week. Overall, there is no toxicological
13 evidence of effects of NO2 exposure on sperm or semen quality.
5.4.2.2 Effects on Reproduction
14 Evidence suggests that exposure to air pollutants during pregnancy may be associated
15 with the ability to reproduce. Gametes (i.e., ova and sperm) may be even more at-risk,
16 especially outside of the human body, as occurs with assisted reproduction. Smokers
17 require twice the number of in vitro fertilization (IVF) attempts to conceive as non-
18 smokers (Feichtinger et al.. 1997). suggesting that a preconception exposure can be
19 harmful to pregnancy. A recent study estimated daily concentrations of criteria pollutants
20 at addresses of women undergoing their first IVF cycle and at their IVF labs from 2000 to
21 2007 in the northeastern U.S. (Legro et al.. 2010). Increasing NO2 concentration at the
22 patient's address during ovulation induction (short-term exposure, -12 days) was
23 associated with a decreased chance of live birth (OR: 0.80, [95% CI: 0.71, 0.91] per
24 10-ppb increase). Similar risks were observed when the exposure period was the daily
25 concentration from oocyte retrieval to embryo transfer, and embryo transfer to pregnancy
26 test (14 days). The authors also observed a decreased odds of live birth when exposed
27 from embryo transfer to live birth (long-term exposure, -200 days) (OR: 0.76, [95% CI:
28 0.56, 1.02] per 10-ppb increase). After adjusting for O3 in a copollutant model, NO2
29 continued to be significantly associated with IVF failure. The results of this study suggest
30 that both short- and long-term exposure to NO2 during ovulation and gestation was
31 detrimental, and reduced the likelihood of a live birth.
32 In contrast, NO2 exposure has not been shown to induce such effects in animals.
33 Breeding studies by Shalamberidze and Tsereteli (197 la, b) with exposures of animals to
34 67 ppb or 130 ppb NO2 12h/day for 3 months found that long-term NO2 exposure had no
November 2013 5-97 DRAFT: Do Not Cite or Quote
-------
1 effect on fertility; NO2 exposure produced no change in the number of dams that became
2 pregnant after mating with an un-exposed male. At the higher dose, Shalamberidze and
3 Tsereteli (197la. b) did see impaired estrous cyclicity (cycle prolongation, increased
4 duration of diestrus, decreased number of normal and total estrus cycles) and the exposed
5 females had a decreased number of ovarian primordial follicles.
5.4.2.3 Effects on Pregnancy
Epidemiologic Evidence
6 Evidence suggests that exposure to air pollutants may affect maternal and fetal health
7 during pregnancy. One such health effect, systemic inflammation, has been proposed as a
8 potential biological mechanism through which air pollution could result in other adverse
9 pregnancy outcomes (Slama et al. 2008; Kannan et al. 2006). Recent studies have
10 investigated the relationship between C-reactive protein (CRP), a marker for systemic
11 inflammation, measured in maternal blood during early pregnancy and in umbilical cord
12 blood (as a measure of fetal health) and the association with NO2 concentrations, van den
13 Hooven et al. (2012a) observed generally null associations between exposure to NO2 and
14 elevated maternal CRP levels, but did observe a positive, linear relationship between
15 quartiles of NO2 exposure and elevated fetal CRP levels. This association was evident
16 when exposure was measured 1 week, 2 weeks, and 4 weeks prior to delivery, but was
17 strongest when exposure to NO2 was measured over the entire pregnancy. Similarly, Lee
18 et al. (20lie) observed generally null associations between exposure to NO2 and elevated
19 maternal CRP levels.
20 Pregnancy-associated hypertension is a leading cause of perinatal and maternal mortality
21 and morbidity. A large body of research has linked changes in blood pressure to ambient
22 air pollution; however, evidence is inconsistent for NO2 (see Section 4.3.5). A few recent
23 studies have examined whether increases in NO2 concentrations are associated with
24 blood pressure changes in women who are pregnant. The results of these studies were not
25 consistent. Hampel et al. (2011) observed that increases in NO2 were associated with
26 decreases in systolic blood pressure, but found no clear associations between NO2
27 concentrations and diastolic blood pressure. Lee etal. (2012b) observed associations
28 between exposure to NO2 and changes in blood pressure that were null for the entire
29 population and when the population was restricted to nonsmokers. van den Hooven et al.
30 (2011) observed small increases in systolic blood pressure associated with increases in
31 NO2 concentrations across all three trimesters of pregnancy, but did not observe a similar
32 association with diastolic blood pressure. (Mobasher et al.. 2013) observed a positive
33 association between exposure to NO2 during the first trimester and hypertensive disorders
November 2013 5-98 DRAFT: Do Not Cite or Quote
-------
1 of pregnancy, though the association was very imprecise and was reduced when exposure
2 was averaged over the second and third trimesters. The same pattern was observed when
3 analyses were restricted to non-obese women, but among obese women, the effect
4 estimate was below 1.00 for each trimester.
5 New-onset gestational hypertension can contribute to pre-eclampsia, a common
6 complication of pregnancy diagnosed after 20 weeks of pregnancy. Wu et al. (2009)
7 observed a 44% increase in the risk of pre-eclampsia associated with a 20-ppb increase in
8 NOX measured over the entire pregnancy; when the exposure was examined
9 categorically, the association between pre-eclampsia risk and NOX concentration was
10 consistent with a linear concentration-response relationship. Similarly, NO2
11 concentrations during pregnancy were associated with an increased risk of pre-eclampsia
12 among a cohort of Australian women (Pereiraet al.. 2013). with the strongest association
13 observed when exposure was limited to the third trimester. Malmqvist et al. (2013) also
14 observed a positive association between NOX concentrations in the third trimester of
15 pregnancy and pre-eclampsia consistent with a linear concentration-response relationship
16 in a Swedish cohort, van den Hooven et al. (2011) did not observe an association for NO2
17 exposure and risks of pregnancy-induced hypertension or pre-eclampsia.
18 Other pregnancy complications that have recently been evaluated and found to be
19 associated with NO2 include gestational diabetes (Malmqvist et al.. 2013) and markers of
20 placental growth and function (van den Hooven et al.. 2012c). Key studies examining the
21 association between exposure to NO2 and pregnancy-related effects can be found in
22 Table 5-13. A supplemental Table S5-2 (U.S. EPA. 2013g) provides an overview of all of
23 the epidemiologic studies of pregnancy-related health effects.
lexicological Evidence
24 Evidence from animal toxicological studies suggests that exposure to NO2 may affect
25 pregnancy. Maternal toxicity was reported in pregnant rats with inhalation exposure to
26 5,300 ppb NO2 for 6 h/day throughout gestation (21 days). Deficits in maternal weight
27 gain during gestation were also reported at 5,300 ppb NO2 (Tabacova et al.. 1984).
28 Fetal lethality in toxicological studies is measured by counting pup loss or resorption
29 sites. This directly affects litter size, or number of live pups born. Mechanisms related to
30 oxidative stress have been proposed (Section 3.3.2.8). but not specifically in relation to
31 reproductive effects. Shalamberidze and Tsereteli (197la. b) reported decreased litter
32 sizes (fewer pups born) to dams that received 1,300 ppb NO2 12h/day for 3 months
33 during pregnancy.
November 2013 5-99 DRAFT: Do Not Cite or Quote
-------
5.4.3 Birth Outcomes
5.4.3.1 Fetal Growth
1 Fetal growth is influenced by maternal, placental, and fetal factors. The biological
2 mechanisms by which air pollutants may influence the developing fetus remain largely
3 unknown. Low birth weight (LEW) has often been used as an outcome measure because
4 it is easily available and accurately recorded on birth certificates. However, LEW may
5 result from either short gestation or inadequate growth in utero. Most of the studies
6 investigating air pollution exposure and LEW limited their analyses to term infants to
7 focus on inadequate growth. A number of studies were identified that specifically
8 addressed growth restriction in utero by identifying infants who failed to meet specific
9 growth standards. Usually, these infants had birth weight less than the 10th percentile for
10 gestational age, using an external standard.
11 A limitation of environmental studies that use birth weight as a proxy measure of fetal
12 growth is that patterns of fetal growth during pregnancy cannot be assessed. This is
13 particularly important when investigating pollutant exposures during early pregnancy as
14 birth weight is recorded many months after the exposure period. The insult of air
15 pollution may have a transient effect on fetal growth, where growth is hindered at one
16 point in time but catches up at a later point. For example, maternal smoking during
17 pregnancy can alter the growth rate of individual body segments of the fetus at variable
18 developmental stages, as the fetus experiences selective growth restriction and
19 augmentation (Lampl and Jeanty. 2003).
20 The terms small-for-gestational-age (SGA), which is defined as a birth weight <10th
21 percentile for gestational age (and often sex and/or race), and intrauterine growth
22 retardation (IUGR) are often used interchangeably. However, this definition of SGA does
23 have limitations. For example, using it for IUGR may overestimate the percentage of
24 "growth-restricted" neonates as it is unlikely that 10% of neonates have growth
25 restriction (Wollmann. 1998). On the other hand, when the 10th percentile is based on the
26 distribution of live births at a population level, the percentage of SGA among PTB is
27 most likely underestimated (Hutcheon and Platt. 2008). Nevertheless, SGA represents a
28 statistical description of a small neonate, whereas the term IUGR is reserved for those
29 with clinical evidence of abnormal growth. Thus all IUGR neonates will be SGA, but not
30 all SGA neonates with be IUGR (Wollmann. 1998). In the following section the terms
31 SGA and IUGR are referred to as each cited study used the terms.
32 The 2008 ISA for Oxides of Nitrogen reviewed three studies that evaluated the
33 relationship between exposure to NO2 and fetal growth (Mannes et al.. 2005; Salam et
November 2013 5-100 DRAFT: Do Not Cite or Quote
-------
1 al.. 2005; Liu et al.. 2003). and concluded that they "did not consistently report
2 associations between NO2 exposure and intrauterine growth retardation" [(U.S. EPA.
3 2008c).p. 3-731.
4 In recent years, a number of studies have examined various metrics of fetal growth
5 restriction. Several of these more recent studies have used anthropometric measurements
6 (e.g., head circumference, abdominal circumference) measured via ultrasound at different
7 periods of pregnancy in order to evaluate patterns of fetal growth during pregnancy and
8 to detect potentially transient effects of early exposure on fetuses. In a mother and child
9 cohort study conducted in Spain, ultrasound measurements were recorded at 12, 20, and
10 32 weeks of gestation, and these anthropometric measurements were recorded again at
11 birth (Iniguez et al.. 2012; Aguileraet al.. 2010). Aguilera et al. (2010) observed that
12 exposure to NO2 early in pregnancy was associated with impaired growth in head
13 circumference from weeks 12 to 20 of gestation and abdominal circumference and
14 estimated fetal weight from weeks 20 to 32. Similarly, Iniguez et al. (2012) reported
15 decreased fetal length and decreased biparietal diameter measured by ultrasound in
16 association with exposure to NO2 during weeks 12-20 of gestation. Decreased birth
17 length and head circumference measured at birth were also associated with exposure to
18 NO2 during this same period. Examining fetal growth characteristics assessed by
19 ultrasound during each trimester of pregnancy, van den Hooven et al. (2012b) observed
20 decreases in head circumference and fetal length in the second and third trimesters
21 associated with exposure to NO2. Hansen et al. (2008) used ultrasound measurements
22 during weeks 13-26 of pregnancy and did not observe associations between exposure to
23 relatively low concentrations of NO2 (mean = 9.8 ppb) and head circumference,
24 biparietal diameter, abdominal circumference, or fetal length.
25 Several studies made use of anthropometric measurements made immediately after birth
26 to evaluate fetal growth. Estarlich et al. (2011). Ballester et al. (2010). and Hansen et al.
27 (2007) observed decreases in body length associated with exposure to NO2. This
28 association persisted when NO2 exposure was estimated for each trimester of pregnancy
29 in the study by Estarlich et al. (2011). Ballester et al. (2010) observed the strongest
30 association with NO2 exposure during the first trimester, while Hansen et al. (2007)
31 reported that the association was strongest for NO2 exposure measured at the end of the
32 pregnancy.
33 When using SGA as an indicator of fetal growth restriction, several studies observed
34 associations with exposure to NO2, NOX or NO (Pereira et al.. 2012; Malmqvist et al..
35 2011; Ballester etal.. 2010; Rich et al.. 2009; Brauer et al.. 2008; Mannes et al.. 2005).
36 These associations were most often observed for exposure to NO2 during the second
37 trimester (Pereira et al.. 2012; Ballester etal.. 2010; Rich et al.. 2009; Mannes et al..
November 2013 5-101 DRAFT: Do Not Cite or Quote
-------
1 2005). Gehring etal. (201 la). Hansen et al. (2007). and Kashimaetal. (2011) did not
2 observe an increased risk of SGA associated with exposure to NO2. All of the studies that
3 used IUGR as an indicator of fetal growth restriction observed an association with
4 exposure to NO2, and this association was strongest for exposures at the beginning of
5 pregnancy (i.e., first month or first trimester) (Liu et al.. 2007; Salam etal.. 2005; Liu et
6 al.. 2003).
7 When evaluating the association between fetal growth and exposure to NO2, many
8 studies relied on modeled concentrations of NO2 coming from land use regression
9 models (Iniguez et al.. 2012; Pereira et al.. 2012; Estarlich et al.. 2011; Gehring et al..
10 201 la; Aguilera et al.. 2010; Ballester et al.. 2010; Brauer et al.. 2008) and emissions or
11 dispersion models (van den Hooven et al.. 2012b; Malmqvist et al.. 2011). Generally, the
12 results of studies that relied on modeled estimates of NO2 were not substantially different
13 from those that used measured NO2 concentrations. However, in a study that assigned
14 exposure to NO2 using both a land use regression model and inverse distance weighting
15 of measured NO2 concentration from monitors Brauer et al. (2008) found higher risks for
16 SGA using the monitoring data (OR: 1.28 [95%CI: 1.18,1.36]) compared to the risks
17 observed with the NO2 estimates from the land use regression model (OR 0.94 [95%CI:
18 0.80,1.10]).
19 Several studies were able to incorporate data on activity patterns in order to help reduce
20 uncertainty related to exposure assessment. Some analyses attempted to decrease the
21 potential exposure measurement error associated with exposure to ambient NO2 by
22 limiting inclusion to subjects that spent 15 or more hours per day at home or subjects that
23 spent less than 2 hours a day in an outdoor environment other than at their primary
24 residence. In such analyses, Aguilera et al. (2010) and Estarlich et al. (2011) found
25 stronger associations between measures of decreased fetal growth and exposure to NO2.
26 In contrast, when Gehring et al. (201 la) limited their analyses to participants that did not
27 move during pregnancy or did not have paid employment outside of the home, there were
28 no consistent associations between SGA and exposure to NO2.
29 In summary, there is generally consistent evidence for an association between exposure
30 to NO2 and fetal growth restriction, including recent evidence from studies that have used
31 fetal anthropometric measurements made via ultrasound and anthropometric
32 measurements made immediately after birth. These are consistent with the studies of the
33 clinical measurement of IUGR and the statistical definition of SGA. The evidence is less
34 certain when it comes to assessing the time period of pregnancy when exposure to NO2 is
35 associated with the highest risks. Some studies find the highest risks associated with NO2
36 when NO2 is measured in early pregnancy, while in other studies the time period
37 associated with the greatest risk is toward the end of pregnancy. Others find the greatest
November 2013 5-102 DRAFT: Do Not Cite or Quote
-------
1 risk when exposure is assigned for the entire pregnancy period. Key studies examining
2 the association between exposure to NO2 and fetal growth effects can be found in Table
3 5-13. A supplemental Table S5-3 (U.S. EPA. 2013h) provides an overview of all of the
4 epidemiologic studies of fetal growth effects.
5.4.3.2 Preterm Birth
5 Preterm birth (PTB) is a syndrome (Romero et al. 2006) that is characterized by multiple
6 etiologies. It is therefore unusual to be able to identify an exact cause for each PTB. In
7 addition, PTB is not an adverse outcome in itself, but an important determinant of health
8 status (i.e., neonatal morbidity and mortality). Although some overlap exists for common
9 risk factors, different etiologic entities related to distinct risk factor profiles and leading
10 to different neonatal and postneonatal complications are attributed to PTB and measures
11 of fetal growth. Although both restricted fetal growth and PTB can result in LEW,
12 prematurity does not have to result in LEW or growth restricted babies.
13 A major issue in studying environmental exposures and PTB is selecting the relevant
14 exposure period, since the biological mechanisms leading to PTB and the critical periods
15 of vulnerability are poorly understood (Bobak. 2000). Short-term exposures proximate to
16 the birth may be most relevant if exposure causes an acute effect. However, exposure
17 occurring in early gestation might affect placentation, with results observable later in
18 pregnancy, or cumulative exposure during pregnancy may be the most important
19 determinant. The studies reviewed have dealt with this issue in different ways. Many
20 have considered several exposure metrics based on different periods of exposure. Often
21 the time periods used are the first month (or first trimester) of pregnancy and the
22 last month (or 6 weeks) prior to delivery. Using a time interval prior to delivery
23 introduces an additional problem since cases and controls are not in the same stage of
24 development when they are compared. For example, a preterm infant delivered at
25 36 weeks is a 32-week fetus 4 weeks prior to birth, while an infant born at term
26 (40 weeks) is a 36-week fetus 4 weeks prior to birth.
27 Recently, investigators have examined the association of PTB with both short-term (i.e.,
28 hours, days, or weeks) and long-term (i.e., months or years) exposure periods. Time-
29 series studies have been used to examine the association between air pollution
30 concentrations during the days immediately preceding birth. An advantage of these time-
31 series studies is that this approach can remove the influence of covariates that vary across
32 individuals over a short period of time. Retrospective cohort and case-control studies
33 have been used to examine long-term exposure periods, often averaging air pollution
34 concentrations over months or trimesters of pregnancy.
November 2013 5-103 DRAFT: Do Not Cite or Quote
-------
1 Studies of PTB fail to show consistency in the periods during pregnancy when pollutants
2 are associated with an effect. For example, while some studies find the strongest effects
3 associated with exposures early in pregnancy, others report effects when the exposure is
4 limited to the second or third trimester. However, the effect of air pollutant exposure
5 during pregnancy on PTB has a biological basis. There is an expanding list of possible
6 mechanisms that may explain the association between NO2 exposure and PTB.
7 Many studies of PTB compare exposure in quartiles, using the lowest quartile as the
8 reference (or control) group. No studies use a truly unexposed control group. If exposure
9 in the lowest quartile confers risk, than it may be difficult to demonstrate additional risk
10 associated with a higher quartile. Thus negative studies must be interpreted with caution.
11 Preterm birth occurs both naturally (idiopathic PTB), and as a result of medical
12 intervention (iatrogenic PTB). Ritz et al. (2000) excluded all births by Cesarean section
13 to limit their studies to idiopathic PTB. No other studies attempted to distinguish the type
14 of PTB, although air pollution exposure maybe associated with only one type. This is a
15 source of potential effect misclassification.
16 A number of recent studies have evaluated the association between exposure to NO2 and
17 PTB, and the results have generally been inconsistent. The body of literature that has
18 observed an association between NO2 and PTB (Trasande et al.. 2013; Olsson et al..
19 2012; Wu etal..201 la; Llop etal.. 2010; Darrow et al.. 2009; Wu et al.. 2009; Jiang et
20 al.. 2007; Leem et al.. 2006; Maroziene and Grazuleviciene. 2002; Bobak. 2000) is
21 generally the same (in both the quantity and quality of studies) to those that find no
22 consistent pattern in the association between NO2 and PTB (Gehring et al.. 201 la:
23 Gehring etal. 20lib; Kashimaet al.. 2011; Basu et al.. 2010; Brauer et al.. 2008;
24 Jalaludin et al.. 2007; Ritz et al.. 2007; Hansen et al.. 2006; Liu etal.. 2003; Ritz et al..
25 2000). Among the studies that observe an association between exposure to NO2 and PTB,
26 the association seems to be strongest for exposure to NO2 late in pregnancy, including
27 the third trimester (Llop etal.. 2010; Leem et al.. 2006; Bobak. 2000). the last 8 weeks of
28 pregnancy (Jiang et al.. 2007). the last six weeks of pregnancy (Darrow et al.. 2009).
29 month of birth (Trasande et al.. 2013). or the last week of pregnancy (Olsson et al.. 2012).
30 Several studies examined very preterm birth (VPTB, <30 weeks gestation), and observed
31 positive associations with NO2 for VPTB when none were observed for PTB (Brauer et
32 al.. 2008). or observed stronger associations for VPTB compared to those for PTB (Wu et
33 al.. 201 la; Wu etal.. 2009V
34 When evaluating the association between PTB and exposure to NO2, several studies
35 relied on modeled concentrations of NO2 coming from land use regression models
36 (Gehring etal.. 201 la; Gehring et al.. 201 Ib; Kashimaet al.. 2011; Wu et al.. 201 la;
November 2013 5-104 DRAFT: Do Not Cite or Quote
-------
1 Llop etal.. 2010; Brauer et al.. 2008) and dispersion models (Wu etal.. 201 la; Wu et al..
2 2009). Generally, the results of studies that relied on modeled estimates of NO2 were
3 similarly inconsistent, and not substantially different from those that used measured NO2
4 concentrations. In a study that assigned exposure to NO2 using both a land use regression
5 model and inverse distance weighting of measured NO2 concentration from monitors
6 Brauer et al. (2008) found generally comparable risk estimates for VPTB using the
7 monitoring data (OR: 1.24, [95%CI: 0.80, 1.88]) and NO2 estimates from the land use
8 regression model (OR: 1.16 [95%CI: 0.93, 1.61]).
9 In summary, the evidence is generally inconsistent, with some studies observing
10 associations between NO2 exposure and PTB while other studies observe no consistent
11 pattern of association. These studies are characterized in supplemental Table S5-4 (U.S.
12 EPA. 20130.
5.4.3.3 Birth Weight
13 With birth weight routinely collected in vital statistics and being a powerful predictor of
14 infant mortality, it is the most studied outcome within air pollution-birth outcome
15 research. Air pollution researchers have analyzed birth weight as a continuous variable
16 and/or as a dichotomized variable in the form of LEW (<2,500 g [5 Ibs, 8 oz]).
17 Birth weight is primarily determined by gestational age and intrauterine growth, but also
18 depends on maternal, placental, and fetal factors as well as on environmental influences.
19 In both developed and developing countries, LEW is the most important predictor for
20 neonatal mortality and is a significant determinant of postneonatal mortality and
21 morbidity. Studies report that infants who are smallest at birth have a higher incidence of
22 diseases and disabilities, which continue into adulthood (Hack and Fanaroff. 1999).
23 A number of recent studies have evaluated the association between exposure to NO2 and
24 birth weight, and the results have generally been inconsistent. When examining birth
25 weight as a continuous variable, several studies have observed decreases in birth weight
26 associated with increases in NO2 exposure (Darrow et al., 201 Ib; Estarlich et al.. 2011;
27 Ballesteretal.. 2010; Morello-Frosch et al.. 2010; Bell et al.. 2007). Generally, these
28 studies observed the largest decreases in birth weight when exposure to NO2 was
29 averaged over the entire pregnancy. There were also a number of studies that examined
30 birth weight as a continuous variable that found no consistent decreases in birth weight
31 associated with increases in NO2 exposure averaged over the entire pregnancy or specific
32 trimesters of pregnancy (Geer et al.. 2012; Rahmalia et al., 2012; Gehring et al., 201 la;
33 Gehring etal. 20lib; Kashimaet al.. 2011; Lepeule etal.. 2010: Aguileraetal.. 2009:
34 Hansen et al.. 2007; Salam etal.. 2005; Gouveia et al.. 2004). When evaluating the risk of
November 2013 5-105 DRAFT: Do Not Cite or Quote
-------
1 having a baby weighing less than 2,500 g, the study results remained inconsistent, with
2 some authors observing an association between LEW and exposure to NO2 (Ebisu and
3 Bell. 2012; Ghosh etal.. 2012a: Wilhelm etal.. 2012; Morello-Frosch et al.. 2010; Brauer
4 et al.. 2008; Bell et al.. 2007; Lee et al.. 2003). while others reported no consistent
5 association (Kashima et al.. 2011; Slamaetal.. 2007; Salam et al.. 2005; Wilhelm and
6 Ritz. 2005; Gouveia et al.. 2004; Liu et al.. 2003; Maroziene and Grazuleviciene. 2002;
7 Bobak. 2000). Generally, the studies that observed the largest risks for LEW averaged
8 exposure to NO2 over the entire pregnancy.
9 Several studies were able to incorporate data on activity patterns in order to help reduce
10 uncertainty related to exposure assessment. In analyses limited to subjects that spent 15
11 or more hours per day at home or subjects that spent less than 2 hours a day in an outdoor
12 environment other than at their primary residence, Estarlich et al. (2011) found stronger
13 associations between birth weight and exposure to NO2. These sensitivity analyses did
14 not consistently change the associations observed by (Aguilera et al.. 2009). When
15 Gehring et al. (201 la) limited their analyses to participants that did not move during
16 pregnancy, or did not have paid employment outside of the home, they continued to
17 observe no consistent associations between birth weight and exposure to NO2.
18 When evaluating the association between birth weight and exposure to NO2, several
19 studies relied on modeled concentrations of NO2 coming from land use regression
20 models (Ghosh etal.. 2012a; Wilhelm etal.. 2012; Estarlich etal.. 2011; Gehring et al..
21 201 la; Gehring et al.. 201 Ib; Kashima et al.. 2011; Ballester et al.. 2010; Lepeule et al..
22 2010; Aguilera et al.. 2009; Brauer et al.. 2008; Slama et al.. 2007) and dispersion models
23 (Rahmalia et al.. 2012; van den Hooven et al.. 2012c; Madsen et al.. 2010). Generally, the
24 results of studies that relied on modeled estimates of NO2 were similarly inconsistent,
25 and not substantially different from those that used measured NO2 concentrations.
26 Several studies compared the use of a statistical models and the use of routinely collected
27 monitoring data to assign exposure to NO2, and concluded that while the monitoring data
28 may include larger errors in estimated exposure, these errors had little impact on the
29 association between exposure to NO2 and birth weight calculated using the two different
30 methods for exposure assessment (Lepeule etal.. 2010; Madsen etal.. 2010).
31 In animal toxicological studies by Shalamberidze and Tsereteli (197la. b), albino rats
32 with exposures to 67 or 130 ppb NO2 12h/day for 3 months prior to breeding produced
33 pups with significantly decreased birth weights. These body weight decrements continued
34 to be significantly decreased at PND 4 and PND 12.
35 In summary, the evidence is generally inconsistent, with some studies observing
36 associations between NO2 exposure and birth weight while other studies observing no
37 consistent pattern of association. Key studies examining the association between
November 2013 5-106 DRAFT: Do Not Cite or Quote
-------
1 exposure to NO2 and birth weight can be found in Table 5-13. A supplemental Table
2 S5-5 (U.S. EPA. 2013J) provides an overview of all of the epidemiologic studies of birth
3 weight.
5.4.3.4 Birth Defects
4 Despite the growing body of literature evaluating the association between ambient air
5 pollution and various adverse birth outcomes, relatively few studies have investigated the
6 effect of temporal variations in ambient air pollution on birth defects. Heart defects and
7 oral clefts have been the focus of the majority of these recent studies, given their higher
8 prevalence than other birth defects and associated mortality. Mechanistically, air
9 pollutants could be involved in the etiology of birth defects via a number of key events.
10 A recent study investigated the association between NO or NO2 and cardiac birth defects
11 (Padula et al., 2013a) and other non-cardiac birth defects (Padula et al., 2013b) in the San
12 Joaquin Valley in California. The authors observed no associations between heart defects
13 and NO or NO2, but did observe an association between neural tube defects and both NO
14 and NO2. In general, however, studies of birth defects have focused on cardiac and oral
15 cleft defects, and the results from these studies are not entirely consistent. This
16 inconsistency could be due to the absence of true associations between NO2 and risks of
17 cardiovascular malformations and oral cleft defects; it could also be due to differences in
18 populations, pollution concentrations, outcome definitions, or analytical approaches. The
19 lack of consistency of associations between NO2 and cardiovascular malformations or
20 oral cleft defects might be due to issues relating to statistical power or measurement
21 error. A recent meta-analysis of air pollution and congenital anomalies concluded that
22 there was no statistically significant increase in risk of congenital anomalies and NO2
23 fVrijheid et al.. 2011). These authors note that heterogeneity in the results of these studies
24 may be due to inherent differences in study location, study design, and/or analytic
25 methods, and comment that these studies have not employed some recent advances in
26 exposure assessment used in other areas of air pollution research that may help refine or
27 reduce this heterogeneity. These studies are characterized in supplemental Table S5-6
28 (U.S. EPA. 2013k).
5.4.4 Postnatal Development
29 The issue of prenatal exposure has assumed increasing importance since ambient air
30 pollution exposures of pregnant women have been shown to lead to adverse pregnancy
31 outcomes, as well as to respiratory morbidity and mortality in the first year of life.
November 2013 5-107 DRAFT: Do Not Cite or Quote
-------
1 Extensive growth and development of the nervous and respiratory systems take place
2 during the prenatal and early postnatal periods. This early developmental phase is thought
3 to be very important in determining long-term lung growth. Shalamberidze and Tsereteli
4 (1971a. b) showed decrements in postnatal body weight at PND 4 and 12 in albino rats
5 with prenatal exposures to 67 ppb or 1,300 ppb NO2 12h/day for 3 months prior to
6 breeding.
5.4.4.1 Developmental Nervous System Effects
7 Central nervous system effects were not evaluated in the 2008 ISA for Oxides of
8 Nitrogen (U.S. EPA. 2008c). Several recent studies have been performed examining the
9 NO 2 effects on the central nervous system in children, with a more extensive examination
10 of cognitive function and additional studies on attention-related behaviors, motor
11 function, psychological distress, and autism. This section is organized by outcome
12 category, and key studies examining the association between exposure to NO2 and
13 developmental nervous system effects can be found in Table 5-13. Supplemental Table
14 S5-7 (U.S. EPA. 20131) provides an overview of all of the epidemiologic studies of
15 developmental nervous system effects.
Cognitive Function
16 Most of the studies examining cognitive function included here are on school children,
17 employed widely used structured neuropsychological tests, assessed ambient exposure by
18 modeling, and examined potential confounding by multiple SES indicators. While some
19 studies considered birth outcomes and noise exposure, none of the studies considered
20 polycyclic aromatic hydrocarbons or lead (Pb), both of which are well-characterized risk
21 factors for neurodevelopmental decrements.
School Children
22 Studies of schoolchildren examined the effects of concurrent exposure on cognitive
23 function were studied; other exposure periods were not examined.
24 van Kempen et al. (2012) and Clark et al. (2012) studied children who were part of the
25 Road Traffic and Aircraft Noise Exposure and Children's Cognition and Health
26 (RANCH) project to determine if there was a relationship between cognitive performance
27 and air pollution, road traffic noise, and aircraft noise in home and school settings, van
28 Kempen et al. (2012) used Neurobehavioral Evaluation System (NES) tests to evaluate
29 485 Dutch children at home and school (9-11 years old), while Clark etal. (2012) tested
30 719 children (9-10 years old) at school using the Suffolk Reading Scale, Child Memory
November 2013 5-108 DRAFT: Do Not Cite or Quote
-------
1 Scale, and a modified version of the Search and Memory Task. Both studies used
2 modeling to predict annual mean ambient NO2 concentrations, then linked the exposures
3 to home and school addresses. A strength of the van Kempen et al. (2012) study that
4 Clark etal. (2012) did not have was that they looked at exposure assessment both in the
5 home and school settings, van Kempen et al. (2012) found a negative association between
6 NO 2 exposure at school and memory (digit span length), but perceptual coding was not
7 associated with school NO2 after additional adjustment for road traffic noise and aircraft
8 noise. No associations were found between home NO2 exposure and cognition before or
9 after noise adjustment. Home and school noise exposure were not associated with
10 memory and perceptual coding before or after NO2 adjustment. Cognitive function was
11 not associated with combined exposure to air pollution and either road or aircraft noise
12 and memory and perceptual coding at home or school. Clark etal. (2012) found that NO2
13 at school was not associated with reading comprehension, recognition memory,
14 information recall, conceptual recall, and working memory before or after adjustment for
15 noise.
16 Freire et al. (2010) evaluated 210 four-year old boys (mean age: 51.3 months; part of the
17 Environment and Childhood [IMNA] study) in the Granada province of southern Spain
18 using a standardized Spanish adaptation of the McCarthy Scales of Children's abilities
19 (MSCA) to determine if there was an association between NO2 exposure (surrogate for
20 traffic-related air pollution) and cognitive development. Using land use regression to
21 assign exposure, predicted NO2 levels were higher in urban settings (15.8 ppb) than non-
22 urban settings (4.9 ppb). In the fully adjusted model, a negative although imprecise
23 association was found between general cognitive score and NO2 ((3 [95% CI]: 8.2-13.2
24 ppb: -1.07 [-9.99, 7.85] points; >13.2 ppb: -4.19 [-14.02, 5.64] points, <8.2 ppb as
25 reference). Negative fully adjusted associations (sometimes imprecise) were also found in
26 both the lower and higher predicted NO2 exposure categories for perceptual-performance,
27 verbal score, quantitative score, memory, executive function, memory span, verbal
28 memory, and working memory.
29 Wang et al. (2009a) evaluated whether traffic-related air pollution exposure affected
30 neurobehavioral function on 861 children aged 8-10 years old in Quanzhou, China. Their
31 observations were consistent with those of other studies (communities with higher air
32 pollution concentrations related to traffic exhaust was associated with a decrease in
33 scores/NAI). However, a major limitation of the study was that it did not conduct a direct
34 analysis of only NO2 effects. Rather, the independent variable of the study was the
35 location of the schools, not NO2 exposure.
36 Morales et al. (2009) conducted a longitudinal study observing associations of indoor
37 NO2 from household gas appliances (exposure during first 3 months of life) with
November 2013 5-109 DRAFT: Do Not Cite or Quote
-------
1 cognitive functioning in 398 four-year-old children from Menorca, Spain. Morales et al.
2 (2009) found that as the number of gas appliances increased, NO2 concentrations
3 increased (mean [SE]: 0 appliances: 6.10 [0.5] ppb; 1 appliance: 16.7 [1.0] ppb; 2
4 appliances: 25.7 [2.1] ppb). General cognitive, verbal, memory, and executive function
5 scores were negatively associated with the number of gas appliances at home, and indoor
6 NO2 exposures were also associated with general cognitive, verbal, and executive
7 function scores through a negative dose-response. Among children with any GSTP1
8 Val-105 allele, gas stove and fire exposure and indoor NO2 was associated with a
9 decrease in general cognitive score and general cognitive, verbal, and executive function
10 scores, respectively, but children with the lie/lie genotype did not show any cognitive
11 decrements. These results provide coherence with effects on other neurodevelopmental
12 outcomes after exposure to ambient NO2.
Infants
13 The Bayley Scales of Infant Development is a widely used test for infant mental
14 development as it is a reliable indicator of current development and cognitive functioning
15 in infants since it tests for markers such as motor function, memory, and early language
16 skills. However, the Bayley Scales of Infant Development is not necessarily correlated
17 with development of children at older ages, and the 1 year Bailey test does not have many
18 outcomes tested that are analogous to those tested at 2 and 3 years old.
19 Guxens et al. (2012) investigated whether or not residential air pollution exposure during
20 pregnancy affected mental development in 1,889 children (mean age 14.8 months) and
21 whether or not antioxidant and detoxification factors reduced the effect. Women from
22 four regions of Spain (Valencia, Sabadell, Asturias, Gipuzkoa; part of the Infancia y
23 Medio Ambiente [INMA] project) were recruited during their first trimester of
24 pregnancy, and children's mental development was tested around 14 months of age
25 (using the Bayley Scales of Infant Development. An overall inverse, but imprecise,
26 association with mental development was observed for NO2 exposure in adjusted models
27 (combined region (3 [95% CI]: -0.95 [-3.90, 1.89] per NO2 doubling), as well as for
28 benzene exposure. The strongest association was observed in the Gipuzkoa region
29 ((3 [95% CI]: -5.15 [-8.04, -2.27] and -5.49 [-9.21, -1.76] perNO2 doubling,
30 respectively). When stratified by antioxidant and detoxification variables, Guxens et al.
31 (2012) concluded that a higher consumption of fruits and vegetables (>405 g/day) during
32 the first trimester of pregnancy may have reduced the effect of air pollutants on infant
33 mental development, while higher maternal circulating Vitamin D levels and longer
34 breast feeding duration had less of an effect, and parental education level and social class
35 had no effect on infant mental development.
November 2013 5-110 DRAFT: Do Not Cite or Quote
-------
Attention-related Behaviors
1 In their study of children 9-11 years old, van Kempen et al. (2012) found that none of the
2 NES test outcomes related to attention (e.g., Simple Reaction Time Test [SRTT],
3 Switching Attention Test [SAT]) were associated with concurrent school NO2.
4 Additionally, no associations were found with home NO2, with or without adjustment for
5 road traffic and aircraft noise. The combined exposure to NO2 and road traffic noise was
6 negatively associated with the reaction times measured during the SAT 'block' condition
7 in the school setting and the reaction times measured during the SRTT and the SAT
8 'arrow' condition in the home setting. At both school and home, high NO2 concentrations
9 had more of an effect on the reaction time in high road traffic noise than in low traffic
10 noise. There was not strong evidence of associations of other attention tests with school
11 or home air pollution, school or home road traffic noise, or school or home air pollution
12 and road traffic noise combined exposure.
13 Morales et al. (2009) observed the effects of indoor NO2 exposure during the first three
14 months of life on ADHD-related symptoms tested by psychologists in 365 children (at
15 4-years old). Exposure to gas appliances and higher indoor NO2 increased the risk of
16 ADHD symptoms and inattention in children, but hyperactivity risk was not increased in
17 association with either exposure. Children with any GSTP1 Val-105 allele showed a
18 higher risk of ADHD symptoms and inattention in association with exposure to gas
19 appliances and indoor NO2, but children with the lie/lie genotype did not show any
20 negative ADHD associations with either exposure. Morales et al. (2009) concluded that
21 early-life exposure to gas appliances in the home and higher concentrations of indoor
22 NO2 had a higher risk of developing ADHD symptoms in 4-year-olds, and that those
23 with the GSTP1 Val-105 allele had a higher risk of ADHD symptoms. These findings are
24 coherent with associations found between other neurodevelopmental outcomes in
25 association with exposure to ambient NO 2.
Motor function
26 van Kempen et al. (2012) used NES tests in 9-11 year old children to determine if there
27 was a relationship between concurrent air pollution and noise exposure and motor
28 function in home and school settings. Based on the Hand Eye Coordination Test (HECT),
29 there was not strong evidence of association between locomotion and NO2 exposure in
30 either the school or home setting, with or without adjustment for road traffic and aircraft
31 noise. There were also no associations found between locomotion and road traffic noise
32 or aircraft noise after adjusting for NO2, in the school or home settings, or when a
33 combination of NO2 and noise was considered.
November 2013 5-111 DRAFT: Do Not Cite or Quote
-------
1 Freire etal. (2010) used a standardized Spanish adaptation of the McCarthy Scales of
2 Children's Abilities (MSCA) to test motor abilities of four-year-old children to determine
3 if there was an association between concurrent NO2 exposure and motor function. The
4 predicted NO2 levels were higher in urban settings (15.8 ppb) than non-urban settings
5 (4.9 ppb). A negative association was found with both the lower and higher predicted
6 NO2 exposures in fully adjusted motor function and gross motor function, but fine motor
7 skills showed a positive association (P [95% CI]: lower NO2: 3.28 [-6.83, 13.40]; higher
8 NO2: 0.91 [-10.22, 12.05]). In contrast, Morales et al. (2009) found no association
9 between indoor NO2 exposure during the first three months of life and motor function in
10 398 four year old children.
11 Tabacova et al. (1985) examined postnatal development of pups from dams that were
12 exposed to 50, 500, or 5,300 ppb NO2 (5h/day, gestation day 0-21). Neuromotor deficits
13 in the righting reflex and postural gait were also seen in pups with 50, 500 and 5,300 ppb
14 NO2. Comparison of the animal toxicology studies and the epidemiologic findings points
15 to multiple sensitive windows of NO2 exposure with potential effects on motor function.
Psychological distress
16 Clark etal. (2012) measured psychological distress in 9-10 year old children using the
17 parental version of the Strengths and Difficulties Questionnaire, and measured aircraft
18 noise, road traffic noise, and NO2 at school. Clark et al. (2012) found that psychological
19 distress showed no associations with noise exposures before or after adjustment for NO2
20 and no substantial associations with NO2 before or after adjustment for noise.
21 In an animal toxicological study, Pi Giovanni et al. (1994) reported developmental
22 neurobehavioral decrements, i.e., decreased pup vocalization, in males removed from the
23 nest at PND5, PND10, or PND15 (3,000 ppb continuous dam NO2 exposure,
24 GDO-GD21).
Autism
25 Autism is a neurodevelopmental disorder characterized by impaired social interaction,
26 verbal and non-verbal communication deficits, and repetitive or stereotypic behavior.
27 Although the causes of autism are not fully understood, many potential factors have been
28 implicated.
29 Becerraetal. (2013) and Volketal. (2013) evaluated whether prenatal and first year of
30 life exposure to NO2 increased the risk of developing autism in children. Becerra et al.
31 (2013) observed 7,603 autistic disorder (AD) and 75,782 control children born in Los
November 2013 5-112 DRAFT: Do Not Cite or Quote
-------
1 Angeles County, California at 36 to 71 months old. AD children (diagnosis based on the
2 Diagnostic and Statistical Manual of Mental Disorders and reported on the Client
3 Development Evaluation Report) were selected from seven regional centers contracted by
4 the California Department of Developmental Services. Control children were selected at
5 random without replacement from birth certificates and had no documentation of autism.
6 NO2 estimates were extracted from LUR model surfaces at each residential location to
7 classify traffic-related exposure (LUR based), and ambient NO2 was measured from the
8 nearest monitoring stations to the residence (monitor-based). Relative increases in risk of
9 AD diagnosis were seen in unseasonalized LUR (U-LUR) and seasonalized LUR (S-
10 LUR) for NO2 and for monitor-based NO2 during entire pregnancy exposure (adjusted
11 OR [95%CI]: U-LUR: 1.13 [1.06, 1.23] per lOppb; S-LUR: 1.05 [0.98, 1.12] per 10
12 ppb; Monitor-based: 1.04 [0.98, 1.10] per 10 ppb), but consistent patterns were not
13 observed across trimester-specific effect estimates. The least educated mothers had the
14 strongest association between AD and LUR-based estimates compared to mothers with
15 the highest education. Adjusted two-pollutant models with U-LUR-NO2 and O3, NO,
16 CO, PMio, and PM2 5 showed an increased risk of 13-17% per 10 ppb for AD in entire
17 pregnancy. Volketal. (2013) studied 524 children (279 with autism, 245 control, 24-60
18 months old; part of the Childhood Autism Risks from Genetics and the Environment
19 [CHARGE] study) in California. Autistic children were evaluated using the Autism
20 Diagnostic Observation Schedules (parents were given the Autism Diagnostic Interview-
21 Revised), while those with a developmental delay and control children were given the
22 Social Communication Questionnaire to screen for autistic features. Motor skills,
23 language, socialization, and daily living skills were assessed using the Mullen Scales of
24 Early Learning and the Vineland Adaptive Behavior Scales, and regional NO2 data was
25 collected from the U.S. EPA AQS. A 10-ppb increase in regional NO2 exposure was
26 associated with an increased risk of autism during the first year of life (adjusted OR [95%
27 CI]: 1.67 [1.25, 2.23]), the entire pregnancy (adjusted OR [95% CI]: 1.52 [1.16, 2.00]),
28 and each of the 3 trimesters (adjusted OR [95% CI]: 1st trimester: 1.30 [1.04, 1.14];
29 2nd trimester: 1.40 [1.10, 1.78]; 3rd trimester: 1.42 [1.12, 1.80]).
Nervous system histopathology: Animal toxicology evidence
30 A recent study examined nervous system effects in adult rats which were exposed over a
31 short-term period to NOX. The evidence is included in this chapter (long-term) because
32 there are no other studies of nevous system effects examined in relation to short-term
33 NO2 exposure (Chapter 4). Adult male Wistar rats were exposed for seven days (6 hours
34 per day) to 2,500-10,000 ppb NO2 (Li et al. 2012aV Brain tissue was collected 18 hours
35 following the last exposure. NO2 exposure had no effect on body weight, however
36 concentration-dependent reductions in brain to body weight ratios were observed, with
November 2013 5-113 DRAFT: Do Not Cite or Quote
-------
1 statistical significance reached at 5,000 and 10,000 ppb NO2. Histopathologic analysis of
2 cerebral cortex demonstrated a concentration-dependent increase in swollen or shrunken
3 nuclei and a concentration-dependent, statistically significant increase in apoptotic cell
4 number in all NO2-exposed rats. Oxidative stress biomarkers were also measured, as well
5 as gene expression and protein levels of oncogenes and apoptosis-related genes.
6 Statistically significant changes in antioxidant enzyme activities (Cu/Zn SOD, MnSOD
7 and glutathione peroxidase), protein carbonyls and malondialdehyde were observed in
8 response to 5,000 and 10,000 ppb NO2. While rats exposed to 2,500 ppb NO2
9 demonstrated a statistically significant increase in the protein level of p53, rats exposed to
10 the higher concentrations exhibited statistically significant increases in mRNA and
11 protein levels of c-fos, c-jun, p-53, and bax. These results are consistent with oxidative
12 stress especially at higher concentrations of NO2.
Summary of studies of Neurodevelopment
13 In summary, several studies found an association between long term exposure to NO2
14 and cognitive function in schoolchildren and infants, but the results were inconsistent.
15 Clark etal. (2012) found that air pollution had no effect on cognitive outcomes in
16 children 9-10 years old and saw little evidence that air pollution moderated the
17 association of noise exposure on cognition, but consuming more fruits and vegetables
18 during pregnancy, higher maternal circulating Vitamin D levels, and longer breast
19 feeding duration may reduce the effect of air pollution on infant mental development
20 (Guxens et al. 2012). While some studies considered birth outcomes and noise exposure,
21 none of the studies considered other pollutants associated with neurodevelopmental
22 decrements, such as lead (Pb) and PM (U.S. EPA. 2013a. 2009b). Generally inconsistent
23 results were found for several attention-related behaviors and motor function with
24 exposure to indoor NO2 or ambient NO2 alone and in combination with noise exposure
25 from road or air traffic. Psychological distress was found to have no substantial
26 associations with NO2 before or after adjustment for noise. Autism risk increased with
27 exposure to NO2 during pregnancy and the first year of life, with more pronounced risk
28 found for late gestation and early life exposure. Homes that had children with autism had
29 higher regional NO2. Consistent patterns were not seen in trimester-specific exposure
30 effects, and children with mothers with the least education showed the strongest
31 association between NO2 and autism.
November 2013 5-114 DRAFT: Do Not Cite or Quote
-------
5.4.4.2 Developmental Respiratory Effects
1 Several high-quality longitudinal epidemiologic studies have consistently found
2 associations with long-term NO2 exposure and decrements in lung function growth and
3 are described in detail in Section 5.2.3.1. Briefly, studies included children from the U.S.,
4 Mexico, and Europe followed from birth or age 8 years for periods of 8 to 11 years. NO2
5 exposures were assessed from central site measurements, dispersion modeling, and land
6 use regression modeling. Results consistently demonstrated associations of NO2 exposure
7 in the first year of life or change in the annual average with decrements in lung function
8 and partially irreversible decrements in lung function growth in children (Figure 5-5)
9 (Molter etal.. 2013: Schultzetal.. 2012: Breton etal.. 2011: Oftedal et al.. 2008: Roias-
10 Martinez et al., 2007a: Gauderman et al., 2004). Some studies found an NO2
11 concentration-dependent decrement in lung function and lung function growth (Rojas-
12 Martinez et al., 2007a: Gauderman et al.. 2004). A number of studies have demonstrated
13 that long-term exposure to NO2 alters lung morphology in experimental animals, though
14 these changes do not appear to contribute to altered lung function (Havashi etal.. 1987:
15 KubotaetaL 1987). Additionally, these effects were not clearly demonstrated in juvenile
16 animals (Chang et al.. 1986: Furiosi etal.. 1973). Thus, the animal toxicological evidence
17 does not provide clear biological plausibility for epidemiologic observations.
5.4.4.3 Early Life Mortality
18 During the neonatal and post-neonatal periods, the developing lung is highly sensitive to
19 environmental toxicants. The lung is not well developed at birth, with 80% of alveoli
20 being formed postnatally. An important question regarding the association between NO2
21 and infant mortality is the critical window of exposure during development for which
22 infants are at risk. Several age intervals have been explored: neonatal (<1 month);
23 postneonatal (1 month to 1 year); and an overall interval for infants that includes both the
24 neonatal and postneonatal periods (<1 year). The studies reflect a variety of study
25 designs, exposure periods, regions, and adjustment for potential confounders. As
26 discussed below, a handful of studies have examined the effect of ambient air pollution
27 on neonatal and postneonatal mortality, with the former the least studied. These studies
28 varied somewhat with regard to the outcomes and exposure periods examined and study
29 designs employed.
30 Overall, the evidence for an association between exposure to NO2 and infant mortality is
31 inconsistent. In an animal toxicological study, Tabacova et al. (1985) examined postnatal
32 development of pups from dams that were exposed to 50, 500, or 5,300 ppb NO2 (5h/day,
33 gestation day GDO-GD21). Significantly decreased pup viability was seen at PND21 with
November 2013 5-115 DRAFT: Do Not Cite or Quote
-------
1 5,300 ppb NO2. Recent epidemiologic studies have examined the association between
2 long-term exposure to NO2 and stillbirths, with one study (Faiz et al.. 2012) observing an
3 association and another (Hwang et al., 2011) observing associations near the null value.
4 One study investigated the association between short-term exposure to NO2 and mortality
5 during the neonatal period (Lin et al., 2004a). and did not observe a positive association.
6 More studies have examined the association between exposure to NO2 and mortality
7 during the postneonatal period. Son et al. (2008). Tsai et al. (2006). and Yang et al.
8 (2006) examined the association between short-term exposure to NO2 and postneonatal
9 mortality, while Ritz et al. (2006) investigated the association between long-term
10 exposure to NO2 and post-neonatal mortality; none observed a consistent, positive
11 association. Finally, two studies examined the association between NO2 and sudden
12 infant death syndrome (SIDS). Dales et al. (2004) and Ritz et al. (2006) observed positive
13 associations with short-term and long-term exposure to NO2, respectively. Supplemental
14 Table S5-8 (U.S. EPA. 2013m) provides a brief overview of the epidemiologic studies of
15 infant mortality.
November 2013 5-116 DRAFT: Do Not Cite or Quote
-------
Table 5-13 Key reproductive and developmental epidemiologic studies for NO2.
Study
Location
(Sample Size)
Mean NO2
(ppb)
Exposure
assessment
Selected Effect Estimates3 (95% Cl)
Fertility, Reproduction and Pregnancy
Pereira et al. (2013)
Wu et al. (2009)
Malmqvist et al.
(2013)
Leqroetal. (2010)
Perth, Australia
/ OO A COX
(n - 23,452)
Southern
California
(n = 81,186)
Sweden
(n-81,110)
Northeastern
U.S.
(n = 7,403)
NR
NOX
Entire
Pregnancy: 7.23
T1:7.45
T2: 7.29
T3:7.14
NOX
7.5
NO2
19
LUR model
CALINE4
dispersion model
to estimate local
traffic-generated
pollution
Modeled NOX
with data from
emission
database and
AERMOD with a
spatial resolution
of 500 x 500
meters
Spatially
interpolated
concentrations
from kriging
based on
monitoring data
Pre-eclampsia
-T- A . A r\A i r\ r\ A -i -i o\
T1. 1.04 (0.94, 1.16)
T2: 1.02(0.91, 1.15)
T3: 1.17(1.04, 1.32)
Entire Pregnancy: 1.22 (1.02, 1.49)
Pre-eclampsia
Entire pregnancy: 1.44 (1.23, 1.68)
Pre-eclampsia
Third trimester
Q1: ref
Q2: 1.28(1.13, 1.46)
Q3: 1.33(1.17, 1.52)
Q4: 1.51 (1.32, 1.73)
Gestational Diabetes
Third trimester
Q1: Ref
Q2: 1.19(0.99, 1.44)
Q3: 1.52(1.28, 1.82)
Q4: 1.69(1.41,2.03)
Odds of Live Birth Following IVF
Medication start to oocyte retrieval:
0.80(0.71, 0.91)
Oocyte retrieval to embryo transfer:
0.87(0.79, 0.96)
Embryo transfer to pregnancy test (14
days): 0.76 (0.66, 0.86)
Embryo transfer to live birth: 0.76
(0.56, 1.02)
November 2013
5-117
DRAFT: Do Not Cite or Quote
-------
Table 5-13 (Continued): Key Reproductive and Developmental Epidemiologic Studies for NO2.
Birth Outcomes
Aauileraetal (2010) Catalonia, Spain 16.9-17.2
(n = 562)
Land-use Fetal Length
Regression T1: -2.04 (-7.01, 2.95)
T2:-1.69 (-7.05, 3.69)
T3: 0.33 (-4.06, 4.72)
Head Circumference
T1: 0.25 (-5.42, 5.91)
T2: 1.70 (-3.69, 7.07)
T3: 0.23 (-4.32, 4.77)
Abdominal Circumference
T1: -2.82 (-8.24, 2.59)
T2:-0.13 (-5.64, 5.38)
T3: 0.74 (-3.926, 5.40)
Biparietal Diameter
T1: 3.87 (-2.04, 9.75)
T2: 4.90 (-0.34, 10.11)
T3: 1.48 (-3.41, 6.35)
Estimated Fetal Weight
T1:-2.22 (-7.39, 2.98)
T2: 0.46 (-5.82, 6.72)
T3: 0.91 (-3.65, 5.45)
Head Circumference
Entire pregnancy: -0.11 (-0.25, 0.03)
Birth Length
Ballester et al. (2010)
Valencia, Spain
(n = 785)
19.1-20.2
Land-use
Regression
SGA - weight
Entire pregnancy: 1.59 (0.89, 2.84)
SGA - length
Entire pregnancy: 1.48 (0.628, 3.49)
Estarlich et al. (2011) Asturias,
uipuzKoa,
Sabadell,
Valencia, Spain
(n = 2,337)
Hansen et al. (2007) Brisbane,
Australia
(n = 26,617)
Overall: 15.5
Urban: 15.9
Rural: 8.7
Median: 7.8
75th: 11.4
Max: 24.2
Land-Use Birth Length
Regression Entire Pregnancy:
-1.69 cm (-0.34, -0.02)
Head Circumference
Entire Pregnancy:
-0.01 cm (0.13, 0.11)
City-wide avg Head Circumference (cm)
T1: 0.05 (-0.05, 0.17)
T2: 0.08 (-0.02, 0.19)
T3: 0.00 (-0.10, 0.10)
Crown-Heel Length (cm)
T1: 0.24(0.05, 0.42)
T2: 0.07 (-0.10, 0.24)
T3: -0.15 (-0.25, -0.05)
November 2013
5-118
DRAFT: Do Not Cite or Quote
-------
Table 5-13 (Continued): Key Reproductive and Developmental Epidemiologic Studies for NO2.
Hansen et al. (2008)
Brisbane,
Australia
(n = 15,623)
9.8
Closest monitor
(Within 2-14 km
of one of 17
monitors)
Head Circumference
M1: 0.54 (-1.88, 2.94)
M2:-0.16 (-2.54, 2.20)
M3:-0.60 (-3.18, 2.00)
M4:-0.30 (-2.30, 1.68)
Biparietal Diameter
M1: 0.14 (-0.62, 0.88)
M2: -0.20 (-0.88, 0.50)
M3:-0.12 (-0.82, 0.58)
M4:-0.16 (-0.74, 0.42)
Abdominal Circumference
M1: 0.48 (-1.98, 2.94)
M2: 0.98 (-1.40, 3.34)
M3: 0.20 (-2.12, 2.52)
M4: 0.30 (-1.80, 2.40)
Femur Length
M1: 0.06 (-0.50, 0.62)
M2:-0.18(-0.78, 0.44)
M3: 0.02 (-0.52, 0.56)
M4:-0.26 (-0.80, 0.26)
Ifiiquezetal. (2012)
Valencia, Spain Median: 20.2 Land-use Fetal Length
(n = 818) Regression 11:0.97(0.92,1.02)
12:0.96(0.92, 1.00)
13:0.97(0.92, 1.02)
Abdominal Circumference
11:0.96(0.92, 0.99)
12:0.98(0.94, 1.02)
13:0.98(0.94, 1.03)
Biparietal Diameter
11:0.96(0.92, 1.00)
12:0.97(0.92, 1.01)
13:0.98(94, 1.02)
Estimated Fetal Weight
11:0.96(0.92, 1.00)
12:0.98(0.94, 1.02)
13:0.97(0.93, 1.02)
van den Hooven et al.
(2012b)
Rotterdam, the
Netherlands
(n = 7,772)
Mean: 21. 2
Median: 21.1
75th: 22.4
Max: 30.3
Combination of
continuous
monitoring data
and CIS-based
dispersion
modeling
techniques
Head Circumference (mm)
T3:
Q1: Ref
Q2: -0.40 (-1.00, 0.20)
Q3:-0.81 (-1.42, -0.20)
Q3: -1.28 (-1.96, -0.61)
Length (mm)
T3:
Q1: Ref
Q2:-0.02 (-0.17, 0.13)
Q3: -0.09 (-0.24, 0.06)
Q4:-0.33 (-0.50,-0.16)
SGA
Entire Pregnancy:
Q1: ref
Q2: 0.93(0.66, 1.31)
Q3: 1.25(0.90, 1.73)
Q4: 1.35(0.94, 1.94)
Bell et al. (2007)
Connecticut and
Massachusetts
(n = 358,504)
17.4
County-level avg Birth Weight (g)
Entire Pregnancy: -18.54 (-22.50,
-14.58)
Black mothers: -26.46 (-37.50, -15.63)
White mothers: -17.29 (-21.67 -13.13)
LEW
1.06(1.00, 1.11)
November 2013
5-119
DRAFT: Do Not Cite or Quote
-------
Table 5-13 (Continued): Key Reproductive and Developmental Epidemiologic Studies for NO2.
Darrowet al. (2011 b)
Atlanta, GA
(N = 406,627)
23.6
Population- Birth Weight (g)
weighted spatial Entire pregnancy: -18.40 (-28.00,
average -9.00)
First 28 days: 0.8 (-3.60, 5.20)
T3:-9.00 (-17.00,-1.20)
Non-Hispanic white: -9.20 (-18.60,
0.20)
Non-Hispanic black: -7.8 (-17.40, 1.60)
Hispanic:-11.60 (-24.80, 1.40)
Postnatal Development
Freire et al. (2010)
Spain
(n = 210)
11.1
Land-use High exposure (>13.2 ppb)
regression General cognitive model: -4.19 (-14.02,
5.64)
Perceptual-performance: -2.17 (-12.76,
8.41)
Verbal:-3.09 (-13.31, 7.13)
Quantitative: -6.71 (-17.91, 4.49)
Memory: -5.52 (-16.18, 5.13)
Motor function: -5.30 (-15.96, 5.36)
Executive function: -4.93 (-14.90, 5.05)
Memory span: -3.46 (-13.93, 7.01)
Verbal memory: -2.71 (-14.02, 8.59)
Working memory: -7.37 (-18.96, 1.74)
Gross motor function: -8.61 (-18.96,
1.74)
Fine motor skills: 0.91 (-10.22, 12.05)
November 2013
5-120
DRAFT: Do Not Cite or Quote
-------
Table 5-13 (Continued): Key Reproductive and Developmental Epidemiologic Studies for NO2.
van Kempen et al.
(2012)
the Netherlands School: 16.5
(n = 485) School: 16.9
Home: 16.4
Modeled data
linked to home
and school
address
School - (3 (95% Cl)
A/C>2 (adjusted for noise)
Memory:
-0.30 (-0.55, 0.04)
SRTT, reaction time:
-2.23 (-22.13, 17.66)
SAT, block, no. of errors:
-0.02 (-0.42, 0.38)
SAT, block, reaction time:
13.92 (-16.70, 43.92)
SAT, arrow, no. of errors:
-0.30 (-0.92, 0.30)
SAT, arrow, reaction time:
21.55 (-19.72, 62.81)
SAT, switch, no. of errors:
-1.19 (-3.62, 1.26)
SAT, switch, reaction time:
21.5 (-45.17, 88.19)
Locomotion:
0.08 (-0.08, 0.25)
Perceptual coding:
0.04 (-0.21, 0.30)
Home-(3 (95% Cl)
NO2 (adjusted for noise)
Memory: 0.17 (-0.08, 0.42)
SRTT, reaction time:
-2.11 (-20.96, 16.72)
SAT, block, no. of errors:
-0.04 (-0.40, 0.32)
SAT, block, reaction time:
15.85 (-11.28, 42.96)
SAT, arrow, no. of errors:
-0.34 (-0.94, 0.26)
SAT, arrow, reaction time:
-3.32 (-42.96, 36.34)
SAT, switch, no. of errors:
-1.23 (-3.32, 0.87)
SAT, switch, reaction time:
-20.21 (-74.92, 34.51)
Locomotion: 0.06 (-0.08, 0.21)
Perceptual coding: -0.06 (-0.26, 0.15)
Clark etal. (2012)
U.K.
(n = 719)
22.7
Modeled data
linked to home
and school
address
NO2 (adjusted for traffic noise)
Reading comprehension:
1.078(0.844, 1.404)
Recognition memory:
1.254(0.673,2.294)
Information recall:
1.327(0.537, 3.221)
Conceptual recall:
1.004(0.797, 1.278)
Working memory:
1.058(0.004,292.728)
Physiological distress:
1.603(0.537,4.788)
November 2013
5-121
DRAFT: Do Not Cite or Quote
-------
Table 5-13 (Continued): Key Reproductive and Developmental Epidemiologic Studies for NO2.
Guxens et al. (2012)
Volketal. (2013)
Becerra et al. (2013)
"Relative risk per 10 ppb
Spain Overall: 15.4 Passive
(n = 1,889) Valencia: 19.6 Samplers; Land-
Sabadell: 17.1 use Regression
Asturias: 12.3
Gipuzkoa: 10.7
California NR CALINE4 line-
(n - 524) source air quality
dispersion model
(within 5 km of
child's home)
Monitoring
stations (within
50 km of home)
Inverse distance-
squared
weighting
California 30.8 Land-use
(n - 83,385) regression
Nearest
monitoring
station to birth
residence
Mental Development - (3 [95% Clf
Location
All regions: -0.95 (-3.90, 1.89)
Gipuzkoa: -5. 15 (-8.04, -2.27)
Asturias: 0.17 (-2.71, 3.04)
Sabadell: 1.98 (-1.69, 5.66)
Valencia: -0.43 (-2.86, 2.01)
Maternal fruit and vegetable intake
<405g/day: -4. 13 (-7. 06, -1.21)
>405g/day: 0.25 (-3.63, 4.12)
Breast feeding duration
None: -3.47 (-7.82, 0.98)
<6 mo: -0.71 (-4.06, 2.65)
> 6 mo: -0.61 (-2.97, 1.75)
Maternal Vitamin D circulation
Low: -2.49 (-6.87, 1.89)
Medium: -0.55 (-3.48, 2.39)
High: -0.11 (-2.72,2.49)
Autism
Traffic-related exposure
First year
Q2: 0.91 (0.56, 1.47)
Q3: 1.00(0.62, 1.62)
Q4: 3.10(1.76, 5.57)
All pregnancy
Q2: 1.26(0.77,2.06)
Q3: 1.09(0.67, 1.79)
Q4: 1.98(1.20, 3.31)
Autism
Entire pregnancy: 1.05 (0.98, 1.12)
T1: 1.03(0.98, 1.08)
T2: 1.03(0.98, 1.08)
T3: 1.04(0.98, 1.09)
Monitor-based
Entire pregnancy: 1.04 (0.98, 1.10)
T1: 1.04(0.99, 1.08)
T2: 1.01 (0.97, 1.06)
T3: 1.02(0.97, 1.07)
change in NO2, or 20 ppb change in NOX, unless otherwise noted.
bPer NO2 doubling
T1 = First Trimester, T2 = Second Trimester, T3 = Third Trimester
M1 = Month 1, M2 = Month 2, M3 = Month 3, M4 = Month 4
NR: No quantitative results reported
November 2013
5-122
DRAFT: Do Not Cite or Quote
-------
Table 5-14 Reproductive and developmental toxicological studies for NO2.
Reference
Tabacova et
al. (1985)
Shalamberidze
and Tsereteli
(1971a).
Shalamberidze
and Tsereteli
(1971b)
Di Giovanni et
al. (1994)
Concentration
N02 (ppb)
25, 50, 500, or
5,300 ppb
(0, 50, 100,
1,000, or
10,000|jg/m3)
1,300 ppb
1,500 or
3,000 ppb
Strain,
Age, Sex
(n)
Wistar
rat.Adult,
F(20)
Albino rat,
NR, Adult,
F(7)
Wistar rat,
M pups (7)
Exposure conditions
Developmental exposure with
postnatal neurotoxicity testing. 0,
25, 50, 500, or 5,300 ppb (0, 50,
100, 1,000, or 1 0,000 |jg/m3),
5 h/day during gestational days 0
through 21 ; progeny followed for
uptoPNDBO
1,300 ppb for 12 h/day for 3 mo
(further specifics unavailable)
Direct exposure of dams at 0,
1,500, or 3,000 ppb continuously
Endpoints Examined
Pup viability, developmental
endpoints (eye opening,
incisor eruption); neuromotor
(righting reflex, postural gait,
geotaxis); hepatic lipid
peroxidation; hepatic drug-
metabolizing enzyme activity.
Litter size, birth weight,
postnatal weight gain (body
weight).
Neurobehavior (ultrasonic
vocalization)
Kripke and
Sherwin
(1984)
1,000 ppb
LEW/f mai
rat, young
adults, M
(6)
throughout gestational days 0-21,
male offspring tested for
vocalizations on PND5, PND10,
andPND15.
0 or 1,000 ppb for 7 h/day,
5 days/week for 21 days
Spermatogenesis, germinal
cells histology, and testicular
interstitial cell histology.
5.4.5 Summary and Causal Determination
1 Overall, the evidence is suggestive of a causal relationship between exposure to NO2 and
2 all three of the reproductive and developmental outcomes: (1) fertility, reproduction and
3 pregnancy, (2) birth outcomes, and (3) postnatal development. Separate conclusions are
4 made for these three smaller groups of outcomes because they are likely to have different
5 etiologies and exposure patterns over different lifestages. In past reviews, a limited
6 number of epidemiologic studies had assessed the relationship between exposure to NO2
7 and reproductive and developmental effects. The 2008 ISA for Oxides of Nitrogen
8 concluded that there was not consistent evidence for an association between NO2 and
9 birth outcomes and concluded that evidence was inadequate to infer the presence or
10 absence of a causal relationship with reproductive and developmental effects overall. All
11 available evidence, including more than 100 recent studies, examining the relationship
12 between exposure to NO2 and reproductive and developmental effects were evaluated
November 2013
5-123
DRAFT: Do Not Cite or Quote
-------
1 using the framework described in Table II of the Preamble. The key evidence as it relates
2 to the causal framework is summarized in Table 5-15.
Fertility, Reproduction, and Pregnancy
3 A number of studies examined the association between exposure to measured
4 concentration of NO2 or modeled predictions of NOX concentration and effects on
5 fertility, reproduction, and pregnancy. These types of health endpoints and their
6 relationship with air pollution have only recently begun to be evaluated, and thus the
7 number of studies for any one endpoint is limited. There is generally no evidence for an
8 association between NO2 concentrations and sperm quality. A single study (Legro et al..
9 2010) observed a decreased odds of live birth associated with higher NO2 concentrations
10 during ovulation induction and the period after embryo transfer. There is emerging
11 evidence for an association between modeled predictions of NOX concentrations during
12 the third trimester and pre-eclampsia; however, evidence for pregnancy induced
13 hypertension, gestational diabetes, and reduced placental growth and function is limited
14 and inconsistent. Collectively, the limited and inconsistent evidence is suggestive of a
15 causal relationship between exposure to NO2 and effects on fertility, reproduction and
16 pregnancy.
Birth Outcomes
17 While the collective evidence for many of the birth outcomes examined is generally
18 inconsistent, there are several well-designed, well-conducted studies that indicate an
19 association between NO2 and adverse birth outcomes. For example, the Spanish cohort
20 that utilized anthropometric fetal measurements throughout pregnancy (Iniguez et al..
21 2012; Estarlich et al.. 2011: Aguilera etal.. 2010: Ballester etal. 2010) observed small,
22 yet consistent associations with impaired fetal growth and NO2 concentrations. Similarly,
23 several high-quality studies observed associations between decreases in birth weight and
24 NO2 concentrations (Darrow et al.. 20lib: Bell et al.. 2007). Studies that examined PTB
25 and birth defects (using both measured NO2 concentrations and modeled predictions of
26 NOX concentrations) generally found inconsistent results, with some studies observing
27 positive associations, while others observed negative associations, regardless of whether
28 NO2 or NOX were used to estimate exposure. Many of the studies examining PTB
29 observed associations very close to the null value. Collectively, the limited and
30 inconsistent evidence is suggestive of a causal relationship between exposure to NO2 and
31 effects on birth outcomes.
November 2013 5-124 DRAFT: Do Not Cite or Quote
-------
Postnatal Development Effects
1 There is limited though positive evidence from both epidemiologic and animal
2 toxicological studies for an association between prenatal and early life NO2 exposure and
3 postnatal development effects, including decrements in cognitive function in humans and
4 neurobehavioral development of pup vocalization and decreases in lung function growth
5 in children. Evidence does not indicate that prenatal or early life NO2 exposures are
6 associated with infant mortality. Collectively, the limited and inconsistent evidence is
7 suggestive of a causal relationship between exposure to NO2 and effects on postnatal
8 development.
November 2013 5-125 DRAFT: Do Not Cite or Quote
-------
Table 5-15 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and reproductive and
developmental effects.
Rationale for
Causal
Determination3
Key Evidence13
Key References'3
NC>2 Concentrations
Associated with
Effects0
Fertility, Reproduction, and Pregnancy -Suggestive
At least one high-
quality epidemiologic
study shows an
association with pre-
eclampsia
Consistent associations, especially
with exposure during the third
trimester, between NO2 concentration
and pre-eclampsia, after adjustment
for common potential confounders. .
Limited evidence for a linear
concentration-response relationship.
Associations not evaluated in
copollutants models. Generally high
correlations between exposures to
NOxand PM2.5 (R2 > 0.7)
Wu et al. (2009)
Pereiraetal. (2013)
Malmqvist et al. (2013)
Mean: 7.2 ppb
Mean: 23 ppb
Mean: 7.5 ppb
Limited and
inconsistent
evidence for other
pregnancy-related
health effects
Limited and inconsistent
epidemiologic evidence for
associations with pregnancy-induced
hypertension, gestational diabetes,
and placental growth and function
Limited toxicological evidence for
deficits in maternal weight gain during
pregnancy, rats.
Means: 8.7-28.6 ppb
Hampeletal. (2011).
Leeetal. (2012b).
Mobasheret al. (2013).
Malmqvist et al. (2013),
van den Hooven et al.
(2012c)
Tabacova et al. (1984) 5,300 ppb
At least one high-
quality epidemiologic
study shows an
association in vitro
fertilization failure
Decreased odds of live birth
associated with higher NO2
concentrations during ovulation
induction and the period after embryo
transfer
Leqroetal. (2010)
19 ppb
Lack of evidence
from available
toxicological and
epidemiologic
studies to support an
association of NO2
exposure with
detrimental effects
on sperm quality
Overall, a limited number of
toxicological and epidemiologic
studies provide no evidence for an
association between exposure to
ambient NO2 concentrations and
effects on sperm.
Rats:Kripke and
Sherwin (1984)
Humans: Rubes et al.
(2010). Sokoletal.
(2006)
1,000 ppb 7h/day,
5 days/week
Mean exposure
averaged over 90-
days:16.8 ppb
Mean daily
concentration: 30.1 ppb
November 2013
5-126
DRAFT: Do Not Cite or Quote
-------
Table 5-15 (Continued): Summary of evidence supporting a suggestive of a causal
relationship between long-term NOi exposure and reproductive
and developmental effects.
Rationale for
Causal
Determination3
Key Evidence13
Key References'3
NO2 Concentrations
Associated with
Effects0
Birth Outcomes - Suggestive
At least one high-
quality epidemiologic
study shows an
association with fetal
growth restriction
and decreased birth
weight
Strong evidence from a well-
conducted Spanish cohort studies that
observes associations with NC>2
concentrations and fetal growth
restriction.
Supported by consistent evidence,
SGAand IUGR.
Outcomes assessed with
anthropometric fetal measurements,
Well-conducted studies show
association with lower birth weight,
though the body of evidence is
generally inconsistent.
Section 5.4.3.3
Darrowet al. (2011 b).
Bell et al. (2007)
Mean exposure
averaged over
trimesters: 7.8-36.1 ppb
Mean exposure
averaged over 3rd
trimester of pregnancy:
23.8 ppb
Mean exposure
averaged over entire
pregnancy: 17.4 ppb
Evidence for other
birth outcomes
generally
inconsistent
Some studies observe an association Section 5.4.3.2
between NO2 exposure and PTB or Section 5434
birth detects while other studies
observe no consistent pattern of
association
Mean exposure
averaged over
trimesters: 8.8-37.6 ppb
Mean exposure
averaged over early
pregnancy (e.g., weeks
3-8): 8.2-28.0 ppb
Limited toxicological
evidence with
relevant NC>2
exposures
Evidence for decreased litter size and
late embryonic lethality in rats.
Shalamberidze and
Tsereteli(1971a),
Shalamberidze and
Tsereteli(1971b)
1,300-5,300 ppb
Limited evidence for
key events informing
mode of action
Inflammation
Increase in CRP concentration in
human umbilical cord blood
associated with NO2 concentration
van den Hooven et al.
(2012a)
Mean exposure
averaged over week
before delivery:
21.4 ppb
Postnatal Development - Suggestive
At least one high-
quality animal
toxicological study
shows an association
with postnatal
development
Impairments in postnatal development
including pup viability, postnatal eye
opening, and incisor eruption
Tabacova et al. (1985) 50, 500, 5,300 ppb
Consistent
epidemiologic
evidence for
developmental
respiratory effects
Consistent associations in children
with decrements in lung function
growth.
Lack of analogous toxicological
evidence
Section 5.2.3.1
Mean annual avg:
14-21 ppb
Mean 6-mo avg:
34 ppb
November 2013
5-127
DRAFT: Do Not Cite or Quote
-------
Table 5-15 (Continued): Summary of evidence supporting a suggestive of a causal
relationship between long-term NOi exposure and reproductive
and developmental effects.
Rationale for
Causal
Determination3
Limited but
supporting evidence
for
neurodevelopmental
effects
Results of
epidemiologic
studies for an effect
on infant mortality
generally
inconsistent
Key Evidence13
Some epidemiologic studies showed
cognitive function decrements in
infants and schoolchildren in
association with NO2 exposure
Some studies did not indicate
associations with cognitive function
More limited and inconsistent
epidemiologic evidence for attention-
related behaviors, motor function,
psychological distress.
Impaired vocalization and neuromotor
function (righting reflex and postural
gait) in pups after in utero NO2
exposure.
Prenatal NO2 exposure associated
with autism in first year of life or at
ages 3-6 years in California
Evidence for a positive association
between exposure to NO2 and infant
mortality is inconsistent across
studies and across post-natal periods
Key References'3
van Kempen et al.
(2012), Morales et al.
(2009). Guxens et al.
(2012)
Clark etal. (2012),
Freireetal. (2010).
Di Giovanni etal. (1994)
Volketal. (2013).
Becerra etal. (2013)
Section 5.4.4.3
NO2 Concentrations
Associated with
Effects0
Mean concurrent:
16.5, 16.9 ppb
Mean prenatal:
15.7 ppb
3,000 ppb
Mean: 30.8 ppb
Mean daily
concentrations:
20.3-50.3 ppb
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
""Describes the key evidence and references contributing most heavily to causal determination and where applicable to uncertainties
and inconsistencies. References to earlier sections indicate where full body of evidence is described.
°Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
5.5 Mortality
i
2
3
4
5
6
7
In past reviews, a limited number of epidemiologic studies had assessed the relationship
between long-term exposure to NO2 and mortality in adults, including cause-specific and
total mortality. The 2008 ISA for Oxides of Nitrogen concluded that the amount of
evidence was "inadequate to infer the presence or absence of a causal relationship" (U.S.
EPA. 2008c). In the current ISA, findings for cause-specific mortality (i.e., respiratory,
cardiovascular) are used to assess the continuum of effects and inform the causality
determinations for respiratory and cardiovascular effects. The causality determination for
total mortality contained herein (Section 5.5) is based primarily on the evidence for non-
accidental mortality but also is informed by the extent to which evidence for the spectrum
November 2013
5-128
DRAFT: Do Not Cite or Quote
-------
1 of cardiovascular and respiratory effects provides biological plausibility for NO2-related
2 total mortality.
5.5.1 Review of Mortality Evidence from 2008 ISA for Oxides of Nitrogen
3 Two seminal studies of long-term exposure to air pollution and mortality among adults
4 have been conducted in the United States; the American Cancer Society (ACS) and the
5 Harvard Six Cities (HSC) cohorts have undergone extensive independent re-analyses and
6 have reported extended results including additional years of follow-up. The initial reports
7 from the ACS (Popeetal.. 1995) and the HCS (Dockerv et al. 1993) did not include
8 results for gaseous pollutants. However, as reported in the 2008 ISA for Oxides of
9 Nitrogen (U.S. EPA. 2008c). in re-analyses of these studies, Krewski et al. (2000)
10 examined the association between gaseous pollutants, including NO2, and mortality.
11 Krewski et al. (2000) observed a positive association between long-term exposure to NO2
12 and mortality in the HSC cohort, with effect estimates1 similar in magnitude to those
13 observed with PM2 5. The effect estimates were positive for different causes of mortality,
14 but were the strongest for cardiopulmonary and total mortality. In their reanalyses of the
15 ACS cohort data, Krewski et al. (2000) long-term exposure to NO2 was not associated
16 with mortality. An extended study of the ACS cohort (Pope et al.. 2002) doubled the
17 follow-up time and tripled the number of deaths compared to the original study, but still
18 observed no association between long-term exposure to NO2 and mortality.
19 A series of studies (Lipfert et al.. 2006a: Lipfert et al.. 2006b: Lipfert et al.. 2003. 2000)
20 characterized a national cohort of over 70,000 male U.S. military veterans who were
21 diagnosed as having hypertension in the mid 1970s and were followed up through 2001.
22 In the earlier studies, the authors reported increased risk of mortality associated with both
23 concurrent and delayed exposure to NO2; these excess risks were in the range of 5-9%
24 (Lipfert et al.. 2003, 2000). In the later studies, the authors focused on traffic density in
25 this cohort. Lipfert et al. (2006a): Lipfert et al. (2006b) reported that traffic density was a
26 better predictor of mortality than ambient air pollution variables, though they still
27 observed a positive and statistically significant association between mortality and NO2
28 exposure. The results from the series of studies characterizing the Veterans cohort are
29 indicative of a traffic-related air pollution effect on mortality, but the study population
30 (lower SES, males with hypertension and a very high smoking rate) was not
31 representative of the general U.S. population.
1 Quantitative effect estimates from studies reviewed in the 2008 NOX ISA (U.S. EPA. 20080) can be found
alongside effect estimates from more recent studies in Figure 5-9. Figure 5-10. and Figure 5-11 and the
corresponding Tables (Table 5-16. Table 5-17. and Table 5-18. respectively).
November 2013 5-129 DRAFT: Do Not Cite or Quote
-------
1 In another cohort conducted in the U.S. (the California Seventh-day Adventist cohort
2 [AHSMOG]), Abbey etal. (1999) enrolled young adult, non-smoking Seventh-day
3 Adventists throughout California. Generally, NO2 was not associated with all-cause,
4 cardiopulmonary, or respiratory mortality in either men or women. The authors observed
5 large risk estimates for lung cancer mortality for most of the air pollutants examined,
6 including NO2, but the number of lung cancer deaths in this cohort was very small (12
7 for females and 18 for males); therefore, it is difficult to interpret these results.
8 Several studies conducted in European countries have examined the relationship between
9 long-term exposure to traffic-related pollutants (including NO2 and NOX) and mortality
10 among adults. Hoek et al. (2002) observed an association between NO2 and mortality in
11 the Netherlands Cohort Study on Diet and Cancer (NLCS), though the association with
12 living near a major road was stronger in magnitude. On the other hand, Gehring et al.
13 (2006) observed that NO2 was generally more strongly associated with mortality than an
14 indicator for living near a major road in a cohort of women from Germany. Results from
15 the PAARC survey (Air Pollution and Chronic Respiratory Diseases) conducted in
16 France, demonstrated increased risk between long-term exposure to NO2 and total,
17 cardiopulmonary, and lung cancer mortality (Filleul et al.. 2005). Similarly, Nafstad et al.
18 (2004) observed an association between NOX and total mortality, as wells as deaths due
19 to respiratory causes, lung cancer, and ischemic heart disease in a cohort of Norwegian
20 men. Nyberg et al. (2000) observed similar results for lung cancer mortality in a case-
21 control study of men in Stockholm, Sweden. (Naess et al.. 2007) investigated the
22 concentration-response relationships between NO2 and cause-specific mortality among a
23 cohort from Oslo, Norway aged 51-90 years. Total mortality, as well as death due to
24 cardiovascular causes, lung cancer, and COPD were associated with NO2 for both men
25 and women in two different age groups, 51-70 and 71-90 years. Naess et al. (2007)
26 reported that the effects appeared to increase at NO2 levels higher than 21 ppb in the
27 younger age group (with little evidence of an association below 21 ppb), while a linear
28 effect was observed between 10 and 31 ppb in the older age group.
29 The results from these studies led to the conclusion that the evidence was inadequate to
30 infer the presence or absence of a causal relationship in the 2008 ISA for Oxides of
31 Nitrogen (U.S. EPA. 2008c). The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c)
32 noted that potential confounding by copollutants was an important uncertainty when
33 interpreting the evidence for the association between long-term exposure to NO2 and
34 mortality. Collinearity among criteria pollutants is another uncertainty that needs to be
35 considered; several studies reported high correlations between NO2 and PM indices. The
36 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) also acknowledged that NO2 could
37 be a surrogate or marker for traffic-related pollution. These uncertainties do not preclude
November 2013 5-130 DRAFT: Do Not Cite or Quote
-------
1 the possibility of an independent effect of NO2, or of NO2 playing a role in interactions
2 among traffic-related pollutants.
5.5.2 Recent Evidence for Mortality from Long-term Exposure to Oxides of
Nitrogen
3 Several recent studies provide extended analyses of existing cohort studies of adult
4 populations. In a reanalysis that extended the follow-up period for the ACS cohort to 18
5 years (1982-2000), Krewski et al. (2009) reported generally null associations between
6 long-term exposure to NO2 and total and cause-specific mortality, similar to what was
7 reported in the initial reanalysis of this cohort (Krewski et al., 2000). In an update to the
8 Veterans cohort study, Lipfert et al. (2009) looked at markers for specific emission
9 sources, including NOX as a marker of traffic, and their relationship with mortality,
10 utilizing a 26-year follow-up period now available for this cohort. The authors observed
11 an association between long-term exposure to NOX and mortality, and noted that this
12 association was stronger among men living in areas with high traffic density compared to
13 men living in areas with lower traffic density. The authors also demonstrate that traffic-
14 related air pollutants (including NOX) are better predictors of mortality than a measure of
15 traffic density in this cohort. Updated results have also been reported for the NCLS
16 cohort (the same effect estimates are reported by both Beelen et al. (2008b) and
17 Brunekreef et al. (2009)). Consistent with previous results from this cohort, the authors
18 observe an association with total mortality. In the updated results, the authors observe the
19 strongest effect between long-term exposure to NO2 and respiratory mortality; this
20 association is stronger than any observed with the traffic variables and total or cause-
21 specific mortality.
22 In an update to a cohort of women in Germany (Gehring et al.. 2006). Heinrich et al.
23 (2013) includes five additional years of follow-up and twice as many fatalities compared
24 to the original analysis. In the updated analyses, the authors observed positive
25 associations between NO2 and all-cause and cardiopulmonary mortality. The effect
26 estimates for lung cancer or respiratory mortality were positive, though less precise and
27 not statistically significant. The effect estimates were highest for women living within
28 50 meters of a road with median daily traffic volume of 5,000 cars or greater. The effect
29 estimates for the associations between all-cause and cardiopulmonary mortality and NO2
30 were generally lower for the follow-up period compared to the original analysis.
31 In a recent U.S. cohort study, Hartetal. (2010) examined the association between
32 residential exposure to NO2 and mortality among men in the U.S. trucking industry. The
33 authors observed an increase in cardiovascular disease mortality and a decrease in COPD
November 2013 5-131 DRAFT: Do Not Cite or Quote
-------
1 mortality associated with NO2 exposure. The association between NO2 exposure and all-
2 cause mortality was robust to the inclusion of PMi0 or SO2 in copollutant models. This
3 association was stronger when the cohort was restricted to truck drivers that maintained
4 local routes, and long haul drivers were excluded. COPD mortality was positively
5 associated with NO2 exposure in the sensitivity analysis excluding long haul drivers. The
6 associations for other causes of death (i.e., lung cancer, IHD, respiratory disease) were
7 generally positive, but were not statistically significant. Another recent U.S. cohort study,
8 The California Teachers Study (Lipsett et al.. 2011) examined the association between
9 long-term exposure to NOX and NO2 and mortality among current and former female
10 public school teachers. The authors observed the strongest associations between IHD
11 mortality and exposure to NOX and NO2; the associations for other causes of death (i.e.,
12 CVD, cerebrovascular, respiratory, lung cancer and all-cause) were less consistent and
13 generally close to the null value.
14 A number of recent studies have examined the association between long-term exposure to
15 NO2 and mortality in Canadian cities. Chen et al. (2013) conducted a cohort study of air
16 pollution and cardiovascular mortality in three cities in Ontario. They used land-use
17 regression models to assign exposure to NO2, and observed that long-term exposure to
18 NO2 was associated with an increased risk of cardiovascular mortality. The association
19 was stronger when mortality from IHD was evaluated separately. In a single-city study
20 conducted in Toronto, Ontario, Jerrett et al. (2009) examined the association between
21 long-term exposure to NO2 and all-cause mortality among subjects from a respiratory
22 clinic. The authors observed positive associations with all-cause and circulatory
23 mortality; the associations with respiratory and lung cancer mortality were also positive,
24 though less precise. In a model that included both NO2 and proximity to traffic, the effect
25 estimate for NO2 remained robust and the effect attributable to traffic was attenuated. In a
26 single-city study conducted in Vancouver, British Columbia, Gan et al. (2011) conducted
27 a population-based cohort study to evaluate the association between traffic-related
28 pollutants and risk of mortality due to CHD. Land-use regression models were used to
29 estimate exposure over a 5 year period (1994-1998) and the cohort was followed up for 4
30 years (1999-2002). The authors observed the strongest associations (i.e., highest
31 magnitude) for exposures to NO2 and CHD mortality; however these associations were
32 greatly attenuated when PM2 5 and BC were included in the model.
33 In a large cohort study in Rome, Italy, Cesaroni et al. (2013) observed positive
34 associations between long-term exposure to NO2 and total, cardiovascular, IHD,
35 respiratory and lung cancer mortality among the adult population. These associations
36 were robust to the inclusion of PM2 5 in the model. Tonne and Wilkinson (2013)
37 evaluated the association between long-term exposure to NO2 and NOX among survivors
38 of hospital admissions for acute coronary system in England and Wales and observed
November 2013 5-132 DRAFT: Do Not Cite or Quote
-------
1 evidence of a null association after adjustment for PM2 5. Rosenlund et al. (2008b)
2 conducted a cohort study in Rome, Italy to investigate the effects of long-term exposure
3 to NO 2 and cardiovascular deaths, including mortality among previous myocardial
4 infarction (MI) survivors. The authors observed a positive association between long-term
5 exposure to NO2 and fatal coronary events, though they did not observe an association
6 with mortality among survivors of a first coronary event. In Brisbane, Australia, Wang et
7 al. (2009b) examined the association between long-term exposure to NO2 and cardio-
8 respiratory mortality. The relative risk for NO2 and cardio-respiratory mortality was near
9 the null value and not statistically significant, indicating no association.
10 A number of studies were conducted in Asian countries to evaluate the association
11 between long-term-exposure to NO2 and mortality. Dong et al. (2012) observed a strong,
12 positive association between long-term exposure to NO2 and respiratory mortality in a
13 cohort study conducted in Shenyang, China. In Shizuoka, Japan, Yorifuji et al. (2010)
14 observed positive associations between NO2 and all-cause, cardiopulmonary, IHD, and
15 respiratory disease mortality, with the strongest effects observed for IHD mortality.
16 When the analysis was restricted to non-smokers, a positive association was observed
17 with lung cancer mortality. Similar observations were reported for lung cancer by
18 Katanoda et al. (2011) among a cohort in Tokyo, Japan and Liu et al. (2008) for a study
19 of women living in Taiwan. In a related study, Liu et al. (2009a) also observed a positive
20 association between long-term exposure to NO2 and bladder cancer.
21 A supplemental Table S5-9 (U.S. EPA. 2013n) provides an overview of the
22 epidemiologic studies of long-term exposure to NOX and mortality. These studies are
23 also characterized in Figure 5-9. Figure 5-10. and Figure 5-11; and Table 5-16. Table
24 5-17. and Table 5-18.
25
November 2013 5-133 DRAFT: Do Not Cite or Quote
-------
STUDY COHORT CONCENTRATION
Abbeyetal. 1999 AHSMOG 36.7
Abbeyetal. 2000 AHSMOG 36.7
Lipsettet al. 2011 CA Teachers 33.6
Lipsettet al. 2011 CA Teachers 33.6
Krewskietal. 2000 ACS 27.9
Filleuletal. 2005 PAARC 24.5
Filleuletal. 2005 PAARC 24.5
Cesaroniet al. 2013 Rome 23.4
Tonne and Wilkinson 2013 22.64; 14.2
Lipfertetal. 2000 Veterans 21.5-27.8
Popeetal. 2002 ACS-Extended 21.4-27.9
Lipfertetal. 2006 Veterans 19.8-27.2
Brunekreefetal. 2009 NLCS-AIR 20.7
Brunekreefetal. 2009 NLCS-AIR 20.7
Gehringet al. 2006 German Women's Health 20.7
Gehringet al. 2006 German Women's Health 20.7
Heinrichetal. 2013 German Women's Health 20.7
Hoeketal. 2002 NLCS 19.5
Jerrettetal. 2009 Toronto 19.5
Lipfertetal. 2006 Veterans 16.3
Hart etal. 2010 TrIPS 14.2
Hart etal. 2010 TrIPS 14.2
Yorifuji etal. 2010 Shizuoka Elderly 13.3
Nafstad etal. 2004 Norwegian Men 10.6
Tonne and Wilkinson 2013 10; 6.5
Krewskietal. 2000 Six Cities Study-Reanalysis 6.1-21.9
Lipfertetal. 2009 Veterans 5.58
NOTES
Men —
Women — 1
Women •
Women
«
24 areas -•
18 areas
H
Full Cohort
Full Cohort
Excluding Long Haul Drivers
•
t
-•
»-
-•^-
-•
^
•
0.75 1 1.25 1.5 1.7
Hazard Ratio (95% Cl)
Note: Red = Recent Studies; Black = Studies reviewed in the 2008 ISA for Oxides of Nitrogen. Hazard ratios are standardized to a
10-ppb increase in NO2 or NOX concentration. Studies are presented in descending order, with the largest mean concentration
(ppb) at the top and the smallest at the bottom of the figure. Circles = NO2; Diamonds = NOX.
Figure 5-9 Results of studies of long-term exposure to NO2 or NOX and all-
cause mortality.
November 2013
5-134
DRAFT: Do Not Cite or Quote
-------
Table 5-16 Corresponding risk estimates for Figure 5-9.
Study
Location
Notes
Relative Risk3 (95% Cl)
Abbey et al. (1999)
Abbey et al. (1999)
Lipsett et al. (2011)
Lipsett et al. (2011)
Krewski et al. (2000)
Filleul et al. (2005)
Filleul et al. (2005)
Cesaroni et al. (2013)
Tonne and Wilkinson (2013)
Lipfert et al. (2000)
Pope et al. (2002)
Lipfert et al. (2006b)
Brunekreef et al. (2009)
Brunekreef et al. (2009)
Gehrinq et al. (2006)
Gehrinq et al. (2006)
Heinrich et al. (2013)
Hoek et al. (2002)
Jerrett et al. (2009)
Lipfert et al. (2006a)
Hartetal. (2010)
Hartetal. (2010)
Yorifuji et al. (2010)
Nafstad et al. (2004)
Tonne and Wilkinson (2013)
Krewski et al. (2000)
Lipfert et al. (2009)
U.S.
U.S.
California
California
U.S.
France
France
Italy
England and Wales
U.S.
U.S.
U.S.
the Netherlands
the Netherlands
Germany
Germany
Germany
the Netherlands
Canada
U.S.
U.S.
U.S.
Japan
Norway
England and Wales
U.S.
U.S.
Men
Women
Women, NC>2
Women, NOx
24 areas
18 areas
NOX
Full cohort
Case cohort
1 -yr avg
5-yr avg
Full Cohort
Excluding Long Haul Drivers
NOX
1.02(0.95,
0.99(0.94,
0.97(0.94,
1.01 (0.99,
0.99(0.99,
0.98(0.96,
1.22(1.10,
1.06(1.04,
1.04(1.01,
1.07(1.04,
1.00(0.98,
1.03(0.98,
1.05(1.00,
0.92(0.79,
1.20(1.02,
1.23(1.02,
1.21 (1.08,
1.15(0.89,
1.48(1.00,
1.04(0.97,
1.10(1.06,
1.19(1.13,
1.04(0.93,
1.16(1.12,
1.12(1.06,
1.15(1.04,
1.04(1.03,
1.08)
1.05)
1.05)
1.03)
1.00)
1.00)
1.34)
1.06)
1.06)
1.10)
1.02)
1.02)
1.10)
1.06)
1.41)
1.47)
1.36)
1.49)
2.16)
1.13)
1.15)
1.26)
1.32)
1.22)
1.20)
1.27)
1.05)
Note: Studies correspond to studies presented in Figure 5-9.
a Effect estimates are standardized to a 10-ppb increase in NO2 or NOX concentration.
November 2013
5-135
DRAFT: Do Not Cite or Quote
-------
STUDY
NaGSSGtal. 2007
LipsGttGt al. 2011
CGsaroniGtal. 2013
BrunGkrGGfGtal. 2009
Hartetal. 2010
Chenetal. 2013
Abbey etal. 1999
KrcwskiGtal. 2000
Filleuletal. 2005
GchringGt al. 2006
II- • L. -inl -1
HGinrich Gt al. 2013
WangGtal. 2009
HoGkGtal. 2002
Yorifuji Gt al. 2010
KrcwskiGtal. 2000
RosGnlundGtal. 2008
GanGtal. 2011
LipsGttGt al. 2011
KrcwskiGtal. 2000
CGsaroniGt al. 2013
Hartetal. 2010
.
Nafstad Gt ai. 2004
Chenetal. 2013
Jerrettetal. 2009
Yorifuji Gt al. 2010
LipsGttGt al. 2011
CGsaroniGtal. 2013
Nafstad Gt ai. 2004
Chenetal. 2013
COHORT MEAN CONCENTRATION
Oslo
CATGachGrs
RomG
NLCS-AIR
TrIPS
Ontario Cohort
AHSMOG Cohort
ACS Cohort
PAARC SurvGy Cohort
GGrman WomGn's HGalth
1 , 1 1 ,
GGrman WomGn s HGalth
BrisbanG Cohort
NLCS
Shizuoka EldGrly Cohort
SixCitiGS Study-RGanalysis
RomG Cohort
VancouvGr Cohort
CATGachGrs
ACS Cohort
RomG
TrIPS
.
NorwGgian MGn Cohort
Ontario Cohort
Toronto Cohort
Shizuoka EldGrly Cohort
CATGachGrs
RomG
NorwGgian MGn Cohort
Ontario Cohort
42.1
33.6
23.4
20.7
14.2
12.1-21.7
36.7
27.9
24.5
20.7
20.7
19.8
19.5
13.3
6.1-21.9
~23.9
17.01
33.6
27.9
23.4
14.2
13 3
me
12.1-21.7
19.5
13.3
33.6
23.4
13 3
icf.6
12.1-21.7
NOTES
Women
Women — •
Full Cohort
Case Cohort •—
Full Cohort
Excluding Long Haul Drivers
Men —
Women -
24 areas -<
18 areas ~~
1-yravg
5-yr avg
_
Outof hospital
In hospital —
Following non-fatal event — • —
Women —
Full Cohort — '
Excluding Long Haul Drivers -
Women •
Women -"
— •-!
+
—
•
•—
— • —
_^_
\—
>—
t
Cardiovascular
Cardiopulmonary
_
_
^ *
•
— • —
— •
•
-•
I
-%-
1 —
-•
+
— • —
^
>—
t-
^ *
CHD
IHD
Circulatory
* *
Cerbrovascular
0.6 1 1.4 1.8 2.2 2.6
Hazard Ratio (95% Cl)
Note: Red = Recent Studies; Black = Studies reviewed in the 2008 ISA for Oxides of Nitrogen. Hazard ratios are standardized to a
10-ppb increase in NO2, NOX or NO concentration. Studies are presented in descending order, with the largest mean concentration
(ppb) at the top and the smallest at the bottom of the figure. Circles = NO2; Diamonds = NOX; Triangles = NO.
Figure 5-10 Results of studies of long-term exposure to NO2, NO, or NOX and
cardiovascular mortality.
November 2013
5-136
DRAFT: Do Not Cite or Quote
-------
Table 5-17 Corresponding risk estimates for Figure 5-10.
Study
Location
Notes
Relative Risk3
(95% Cl)
Cardiovascular Disease
Naess et al. (2007)
Norway
Women
1.06(1.00, 1.12)
Lipsettetal. (2011)
Lipsettetal. (2011)
Cesaroni et al. (2013)
Brunekreef et al. (2009)
Brunekreef et al. (2009)
Hartetal. (2010)
Hartetal. (2010)
Chenetal. (2013)
California
California
Italy
the Netherlands
the Netherlands
U.S.
U.S.
Canada
Women, NC>2
Women, NOx
Full cohort
Case cohort
Full Cohort
Excluding Long Haul Drivers
0.98(0.88,
1.03(1.00,
1.06(1.04,
1.04(0.96,
0.92(0.77,
1.09(1.01,
1.14(1.03,
1.17(1.10,
1.09)
1.06)
1.08)
1.13)
1.10)
1.17)
1.25)
1.23)
Cardiopulmonary Disease
Abbey et al. (1999)
Abbey et al. (1999)
Krewski et al. (2000)
Filleul et al. (2005)
Filleul et al. (2005)
Gehrinq et al. (2006)
Gehrinq et al. (2006)
Heinrich et al. (2013)
Wanq et al. (2009b)
Hoek et al. (2002)
Yorifuii et al. (2010)
Krewski et al. (2000)
U.S.
U.S.
U.S.
France
France
Germany
Germany
Germany
Australia
the Netherlands
Japan
U.S.
Men
Women
24 areas
18 areas
1 -yr avg
5-yr avg
1.01 (0.93, 1.09)
1.02(0.95, 1.09)
1.01 (1.00, 1.02)
1.00(0.96, 1.04)
1.16(0.93, 1.45)
1.70(1.28,2.26)
1.92(1.35,2.71)
1.67(1.36,2.05)
0.95(0.74, 1.22)
1.45(0.99,2.13)
1.32(1.12, 1.54)
1.17(1.03, 1.34)
Coronary Heart Disease (CHD)
Rosenlund et al. (2008b)
Rosenlund et al. (2008b)
Rosenlund et al. (2008b)
Ganetal. (2011)
Ganetal. (2011)
Italy
Italy
Italy
Canada
Canada
Out of hospital
In hospital
Following non-fatal coronary event
NO2
NO
1.16(1.04, 1.26)
1.10(0.94, 1.30)
0.91 (0.80, 1.04)
1.09(1.02, 1.19)
1.09(1.03, 1.15)
Ischemic Heart Disease (IHD)
Lipsettetal. (2011)
Lipsettetal. (2011)
Krewski et al. (2000)
Cesaroni et al. (2013)
Hartetal. (2010)
Hartetal. (2010)
Yorifuii et al. (2010)
California
California
U.S.
Italy
U.S.
U.S.
Japan
Women, NO2
Women, NOX
Full Cohort
Excluding Long Haul Drivers
1.07(0.92, 1.24)
1.05(1.00, 1.09)
1.02(1.00, 1.03)
1.10(1.06, 1.14)
1.01 (0.92, 1.11)
1.07(0.95, 1.21)
1.57(1.04,2.36)
November 2013
5-137
DRAFT: Do Not Cite or Quote
-------
Table 5-17 (Continued): Corresponding risk estimates for Figure 5-10.
Study
Nafstad et al. (2004)
Chenetal. (2013)
Location Notes
Norway NOX
Canada
Relative Risk3
(95% Cl)
1.16(1.06,
1.19(1.08,
1.24)
1.30)
Circulatory Disease
Jerrett et al. (2009)
Yorifuii et al. (2010)
Canada
Japan
2.53(1.27,
1.30(1.06,
5.11)
1.59)
Cerebrovascular Disease
Lipsett et al. (2011)
Lipsettetal. (2011)
Cesaroni et al. (2013)
Yorifuii et al. (2010)
Nafstad et al. (2004)
Chenetal. (2013)
California Women, NO2
California Women, NOx
Italy
Japan
Norway NOx
Canada
0.86(0.71,
1.01 (0.95,
1.02(0.98,
1.18(0.89,
1.08(0.89,
0.92(0.81,
1.06)
1.07)
1.06)
1.57)
1.30)
1.10)
Note: Studies correspond to studies presented in Figure 5-10.
aEffect estimates are standardized to a 10-ppb increase in NO2, NOX or NO concentration.
November 2013 5-138 DRAFT: Do Not Cite or Quote
-------
STUDY
Abbey etal. 1999
Lipsettet al.2011
Dong etal. 2012
Cesaronietal. 2013
Brunekreefetal. 2009
Heinrichetal. 2013
Jerrettetal. 2009
Hart etal. 2010
Yorifuji et al. 2010
Nafstad etal. 2004
Katanoda etal. 2011
Naessetal. 2007
Hart etal. 2010
Yorifuji et al. 2010
Abbey etal. 1999
Lipsettet al.2011
Naessetal. 2007
Krewskietal. 2000
Filleuletal. 2005
Cesaronietal. 2013
Brunekreefetal. 2009
Heinrichetal. 2013
Hart etal. 2010
Nybergetal. 2000
Yorifuji et al. 2010
Nafstad etal. 2004
Krewskietal. 2000
Katanoda etal. 2011
COHORT COI\
AHSMOG Cohort
CA Teachers
Shenyang Cohort
Rome
NLCS-AIR
German Women
Toronto Cohort
TrIPS
Shizuoka Elderly
Norwegian Men
Three-prefecture
Oslo Cohort
TrIPS
Shizuoka Elderly
AHSMOG Cohort
CA Teachers
Oslo Cohort
ACS Cohort
PAARCSurvey
Rome
NLCS-AIR
German Women
TrIPS
Stockhom
Shizuoka Elderly
Norwegian Men
Six Cities -Reanalysis
Three-prefecture
MEAN
CENTRATION
36.7
33.6
24.5
23.4
20.7
20.7
19.5
14.2
13.3
10.6
1.2-33.7
27.5-44.9
14.2
13.3
36.7
33.6
27.5-44.9
27.9
24.5
23.4
20.7
20.7
14.2
13.3
13.3
10.6
1.2-33.7
NOTES
Men •—
Women 1
Women -^
Full Cohort
Ca_c Cohort
,- M ,. i
ExcludingLong Haul Drivers
All
Male
Female
Men
Women —
Women
Women —4
Men
Women
4
24 areas -•-
Full Cohort •-
C3jC COhOI t •
Full Cohort
ExcludingLong Haul Drivers —
jU ycai cj\pOjUi c
All
Men
Women
Respiratory
6.1(5.2^7.2)
-•-
*
*
__~^ COPD
-•
Lung Cancer
-• —
-•
— •
Hazard Ratio (95% Cl)
Note: Red = Recent Studies; Black = Studies reviewed in the 2008 ISA for Oxides of Nitrogen. Hazard ratios are standardized to a
10-ppb increase in NO2 or NOX concentration. Studies are presented in descending order, with the largest mean concentration
(ppb) at the top and the smallest at the bottom of the figure. Circles = NO2; Diamonds = NOX.
Figure 5-11 Results of studies of long-term exposure to NO2 or NOX and
respiratory mortality.
November 2013
5-139
DRAFT: Do Not Cite or Quote
-------
Table 5-18 Corresponding risk estimates for Figure 5-11.
Study
Location
Notes
Hazard Ratio3 (95% Cl)
Respiratory
Abbey et al. (1999)
Abbey et al. (1999)
Lipsett et al. (2011)
Lipsettetal. (2011)
Donqetal. (2012)
Cesaroni et al. (2013)
Brunekreef et al. (2009)
Brunekreef et al. (2009)
Heinrich et al. (2013)
Jerrett et al. (2009)
Hartetal. (2010)
Hartetal. (2010)
Yorifuii et al. (2010)
Nafstad et al. (2004)
Katanoda et al. (2011)
Katanoda et al. (2011)
Katanoda et al. (2011)
U.S.
U.S.
California
California
China
Italy
the Netherlands
the Netherlands
Germany
Canada
U.S.
U.S.
Japan
Norway
Japan
Japan
Japan
Men
Women
Women, NC>2
Women, NOx
Full cohort
Case cohort
Full Cohort
Excluding Long Haul Drivers
NOX
All
Men
Women
0.93(0.82, 1.07)
0.98(0.87, 1.11)
0.93(0.76, 1.15)
0.97(0.91, 1.03)
6.1 (5.2, 7.2)
1.06(1.00, 1.12)
1.22(1.00, 1.48)
1.16(0.83, 1.62)
1.15(0.67,2.00)
1.16(0.37,2.71)
1.07(0.91, 1.27)
1.26(1.01, 1.56)
1.39(1.04, 1.83)
1.32(1.12, 1.54)
1.16(1.12, 1.21)
1.11 (1.05, 1.18)
1.25(1.18, 1.33)
COPD
Naess et al. (2007)
Naess et al. (2007)
Hartetal. (2010)
Hartetal. (2010)
Yorifuii et al. (2010)
Norway
Norway
U.S.
U.S.
Japan
Men
Women
Full Cohort
Excluding Long Haul Drivers
1.18(1.04, 1.33)
1.05(0.93, 1.18)
0.97(0.79, 1.19)
1.01 (0.82, 1.44)
1.22(0.63,2.31)
Lung Cancer
Abbey et al. (1999)
Abbey et al. (1999)
Lipsettetal. (2011)
Lipsettetal. (2011)
Naess et al. (2007)
Naess et al. (2007)
Krewski et al. (2000)
Filleul et al. (2005)
U.S.
U.S.
California
California
Norway
Norway
U.S.
France
Men
Women
Women, NC>2
Women, NOx
Men
Women
24 areas
1.35(0.96, 1.90)
1.69(1.07,2.65)
1.00(0.76, 1.32)
0.98(0.90, 1.07)
1.06(0.97, 1.15)
1.20(1.09, 1.32)
0.99(0.97, 1.01)
0.94(0.89, 1.02)
November 2013
5-140
DRAFT: Do Not Cite or Quote
-------
Table 5-18 (Continued): Corresponding risk estimates for Figure 5-11.
Study
Filleuletal. (2005)
Cesaroni et al. (2013)
Brunekreef et al. (2009)
Brunekreef et al. (2009)
Heinrich et al. (2013)
Location
France
Italy
the Netherlands
the Netherlands
Germany
Notes
18 areas
Full cohort
Case cohort
Hazard Ratio3 (95% Cl)
1.12(0.77, 1.61)
1.08(1.04, 1.14)
0.94(0.81, 1.09)
0.87(0.66, 1.14)
1.56(0.91,2.69)
Hartetal. (2010)
Hartetal. (2010)
Nvberq et al. (2000)
Nvberq et al. (2000)
Yorifuii et al. (2010)
Nafstad et al. (2004)
Krewski et al. (2000)
Katanoda et al. (2011)
Katanoda et al. (2011)
Katanoda et al. (2011)
U.S.
U.S.
Sweden
Sweden
Japan
Norway
U.S.
Japan
Japan
Japan
Full Cohort
Excluding Long Haul Drivers
30-year Exposure
10-year Exposure
NOX
All
Men
Women
1.07(0.96,
1.09(0.95,
1.10(0.87,
1.20(0.94,
0.91 (0.63,
1.22(1.06,
1.09(0.76,
1.17(1.10,
1.18(1.11,
1.13(1.01,
1.19)
1.25)
1.36)
1.48)
1.34)
1.39)
1.57)
1.26)
1.26)
1.27)
Note: Studies correspond to studies presented in Figure 5-11.
a Effect estimates are standardized to a 10-ppb increase in NO2 or NOX concentration.
5.5.3 Summary and Causal Determination
1 Collectively, the evidence is suggestive of a causal relationship between long-term
2 exposure to NO2 and mortality among adults. The strongest evidence comes from cohort
3 studies conducted in the U.S. and Europe, which show consistent, positive associations
4 with total mortality, as well as deaths due to respiratory and cardiovascular disease
5 (Lipsett et al.. 2011; Hart etal.. 2010; Brunekreef etal.. 2009; Beelen et al.. 2008b:
6 Krewski et al.. 2000). The results from these studies are coherent with studies that have
7 observed associations between long-term exposure to NO2 and respiratory hospital
8 admissions (Andersen et al.. 2012; Andersen et al.. 2011) and cardiovascular effects
9 (Lipsett et al.. 2011; Hartetal.. 2010). Additionally, the evidence for short- and long-
10 term respiratory and cardiovascular morbidity provides some biological plausibility for
11 mortality.
12 In past reviews, a limited number of epidemiologic studies had assessed the relationship
13 between long-term exposure to NO2 and mortality in adults. The 2008 ISA for Oxides of
14 Nitrogen concluded that the scarce amount of evidence was "inadequate to infer the
15 presence or absence of a causal relationship" (U.S. EPA. 2008c). Recent studies provide
November 2013
5-141
DRAFT: Do Not Cite or Quote
-------
1 evidence for an association between long-term exposure to NOX and mortality from
2 extended analyses of existing cohorts as well as original results from new cohorts in the
3 U.S., Europe and Asia. While the results were generally consistent across studies, there
4 were several well-designed, well-conducted studies that did not observe an association
5 between long-term exposure to NO2 and mortality (Krewski et al., 2009; Pope et al.,
6 2002; Abbey et al.. 1999). All available evidence for mortality due to long-term exposure
7 to NOX was evaluated using the framework described in Table II of the Preamble. The
8 key evidence as it relates to the causal framework is summarized in Table 5-19. The
9 overall evidence is suggestive of a causal relationship between long-term exposure to
10 NO2 and mortality among adults.
November 2013 5-142 DRAFT: Do Not Cite or Quote
-------
Table 5-19 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and total mortality.
Rationale for
Causal
Determination3
At least one high-
Key Evidence13
Positive association between long-term
Key References'3
Krewski et al. (2000)
NOX or NO2
Concentrations
Associated with
Effects0
Mean concentrations
quality exposure to NO2 and mortality in the
epidemiologic Harvard Six Cities (HSC) Cohort, with
study shows an effect estimates similar in magnitude to
association those observed with PM25, even after
adjustment for common potential
confounders. Associations generally not
evaluated in copollutants models.
Updated results from the Netherlands
Cohort Study (NLCS) report a positive
association with total mortality, effects for
respiratory mortality stronger than any
observed with traffic variables and total
or other cause-specific mortality
across cities (1980):
6.1-21.9 ppb
Beelen et al. (2008b),
Brunekreef et al. (2009)
Mean (1987-1996):
20.7 ppb
Max: 35.5 ppb
Some studies
show no
association
Recent cohort studies in the U.S.
observe increases in mortality due to
cardiovascular disease in separate
cohorts of men and women
No association in several re-analyses of
the American Cancer Society (ACS)
cohort
Hartetal. (2010)
Lipsettetal. (2011)
Krewski et al. (2000)
Pope et al. (2002)
Krewski et al. (2009)
Mean (1985-2000):
14.2 ppb
Mean (1996-2005):
33.6; Max: 67.2 ppb
Mean (1982-1 998):
21. 4-27.9 ppb
Mean (1982-1 998)
No association with total,
cardiopulmonary or respiratory mortality
in the California Seventh-day Adventists
cohort (AHSMOG)
Abbey et al. (1999)
27.9; Max 51.1 ppb
Mean (1973-1992):
36.8 ppb
Limited coherence Limited evidence for respiratory
with evidence for
respiratory and
cardiovascular
morbidity
hospitalizations in adults coherent with
evidence for respiratory mortality
Andersen et al. (2011)
Andersen et al. (2012)
35-yr mean: 9.0 ppb
25-yr mean: 9.5 ppb
Some inconsistencies reported for
cardiovascular morbidity. Evidence for Ml
and heart failure coherent with evidence
for cardiovascular mortality
Lipsettetal. (2011)
Mean: 33.6;
Max: 67.2 ppb
Atkinson et al. (2013)
Mean: 12.0;
Max: 32.3 ppb
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble.
bDescribes the key evidence and references contributing most heavily to causal determination and, where applicable, to
uncertainties or inconsistencies. References to earlier sections indicate where full body of evidence is described.
""Describes the NOX or NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
November 2013
5-143
DRAFT: Do Not Cite or Quote
-------
5.6 Cancer
1 The 1993 Oxides of Nitrogen AQCD and the 2008 ISA for Oxides of Nitrogen reported
2 that there was no clear evidence that NO2 or oxides of nitrogen act as a complete
3 carcinogen. The U.S. Department of Health and Human Services, the International
4 Agency for Research on Cancer, and the U.S. EPA have not classified nitrogen oxides for
5 potential carcinogenicity. The American Conference of Industrial Hygienists has
6 classified NO2 as A4 (Not classifiable for humans or animals). The 2008 Oxides of
7 Nitrogen ISA (U.S. EPA. 2008c) included a few epidemiologic studies of oxides of
8 nitrogen and cancer, both examining lung cancer incidence and reporting positive
9 associations. Since the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c). additional
10 studies have been published exploring this relationship. In addition, studies have been
11 performed examining the relationship between NO2 and leukemia, bladder cancer, breast
12 cancer, and prostate cancer. These are all described in more detail in supplementary
13 Table S5-10 (U.S. EPA. 2013o).
5.6.1 Lung Cancer Incidence
14 Two previous studies included in the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
15 2008c) reported positive associations between oxides of nitrogen and lung cancer
16 incidence (Nafstadetal.. 2003: Nvberg et al.. 2000). Nvberg et al. (2000) reported an
17 association between NO2 and lung cancer at the highest 10-year average concentrations
18 of NO2 with a 20-year lag. This association was robust to inclusion of SO2, which was
19 not observed to be associated with lung cancer (Pearson correlation coefficient between
20 SO2 and NO2 ranged from 0.5 to 0.7). Nafstad et al. (2003) performed a study with 24
21 years of follow-up and reported a positive association between NOX concentrations and
22 lung cancer incidence during the early years of the study, but the authors report more
23 recent years had weaker associations (results were not provided). The Pearson correlation
24 coefficient between NOX and SO2 was 0.63 and no association was observed between
25 SO2 concentration and cancer.
26 An HEI Research Report examined the association between NO2 concentration and lung
27 cancer incidence within the NLCS using over 11 years of follow-up (Brunekreef et al..
28 2009: Beelen et al.. 2008a). The researchers observed no association in unadjusted and
29 adjusted analyses using case-cohort and full cohort approaches. The associations between
30 lung cancer and SO2 (correlation coefficient with NO2>0.6) and PM2 5 (correlation
31 coefficient with NO2>0.8) were also examined and found to be null.
November 2013 5-144 DRAFT: Do Not Cite or Quote
-------
1 A Danish study combined three cohorts and reported an association between increased
2 NOX concentrations and lung cancer incidence (Raaschou-Nielsen et al.. 2010a). This
3 increased incidence with NOX exposure persisted in some models of specific cancer
4 types, such as squamous cell carcinomas. When examining the associations by sex,
5 length of education, and smoking status, the precision was decreased and no differences
6 were observed between the groups. One of these cohorts was used in another study where
7 the follow-up period was extended five years to include more cases (Raaschou-Nielsen et
8 al.. 2011). This study detected an increased incidence rate of lung cancer in the highest
9 quartile of NOX concentrations. Further analyses evaluated interactions with sex,
10 smoking status, length of school attendance, and daily fruit intake. An increased
11 association between NOX concentration and lung cancer incidence was observed among
12 individuals with at least 8 years of schooling but no association was apparent among
13 those with less schooling.
14 A study using the GEN-AIR case-control data reported on non-smokers and lung cancer
15 incidence (Papathomas et al.. 2011). This study used multiple statistical analysis
16 techniques to evaluate the associations between air pollutants and lung cancer incidence.
17 Although profile regression analyses reported higher NO2 exposures for the higher risk
18 grouping, logistic regression analyses did not find an association between NO2 and lung
19 cancer incidence. The same was true of PMi0. In another statistical model by the authors,
20 NO2 was not chosen as a predictor, whereas PMi0 concentration was.
21 In summary, multiple studies have examined the associations between concentrations of
22 oxides of nitrogen and lung cancer incidence. Positive associations were reported in
23 multiple studies, but other studies reported no associations. The inconsistency observed
24 between studies does not appear to be due to the inclusion of other pollutants in the
25 models nor does it appear to relate to the length of the exposure or follow-up period.
26 Potentially important confounders, such as smoking status, were included in the analyses.
27 Variables examined as effect measure modifiers, such as education, may be important in
28 further understanding the association.
5.6.2 Lung Cancer Mortality
29 Two HEI Research Reports have investigated the association between NO2 concentration
30 and lung cancer mortality using large cohorts with follow-ups of at least 10 years.
31 Brunekreef etal. (2009) (see also. Beelen et al.. 2008b) reported no association between
32 NO2 and lung cancer mortality using the Netherlands Cohort Study (NLCS) and results
33 were not changed with the inclusion of a traffic-intensity variable. No association was
34 observed between lung cancer mortality and other pollutants (SO2, correlation coefficient
November 2013 5-145 DRAFT: Do Not Cite or Quote
-------
1 with NO2 >0.6, or PM25, correlation coefficient with NO2 >0.8). Krewski et al. (2009)
2 utilized an extended follow-up of the American Cancer Society Study and reported no
3 associations between NO2 and lung cancer mortality. An association with lung cancer
4 mortality for PM2 5 was noted in this report, but not for CO, O3, or SO2.
5 Inconsistent findings between NO2 and lung cancer mortality have been reported in
6 studies conducted across Europe. A positive association was observed between NO2 and
7 lung cancer mortality in a large study conducted in Rome, Italy (Cesaroni et al.. 2013).
8 The association demonstrated a linear relationship. No effect measure modification was
9 apparent by age, sex, educational level, area-based socioeconomic position, or moving
10 history. NO2 was highly correlated with PM2 5, which was also associated with lung
11 cancer mortality. A study in France reported a positive association between NO2 and lung
12 cancer mortality only after exclusion of areas with air monitoring sites reporting a high
13 ratio of NO to NO2 (which implied a strong influence of heavy traffic near the monitor
14 that may not represent the air pollution concentrations in the entire area) (Filleul et al..
15 2005). Correlations between NO2 and other air pollutants ranged from -0.22 to 0.86. No
16 other air pollutants examined in the study (SO2, total suspended particles, black smoke,
17 and NO) were associated with lung cancer mortality. A study in Norway examined four
18 years of air pollution and mortality data (Naess et al.. 2007). Positive associations
19 between NO2 and lung cancer mortality were observed among women aged 51-70 years
20 and 71-90 years but not among men in these age groups (although a positive association
21 was reported in the crude HR for 71-90 year-old men). Correlations between the
22 pollutants examined (NO2, PMi0, and PM25) were not reported individually but ranged
23 from 0.88 to 0.95. Associations between lung cancer and the other pollutants were similar
24 to those observed for NO2. In a non-parametric smooth analysis that combined the sexes,
25 the increase in log odds for lung cancer appears to begin around 21.3 ppb for 51-70
26 year-olds while the increase appears to be at lower concentrations among those aged
27 71-90 years. A large study of women from Germany followed up women who were
28 originally enrolled in cross-sectional studies in the 1980s and 1990s (Heinrich et al..
29 2013). Using NO2 concentration from their address at the baseline examination, the
30 authors reported no association between NO2 concentration and lung cancer mortality.
31 The Spearman's correlation coefficient for PM10, which was observed to be associated
32 with lung cancer, and NO2 was 0.5. A large cohort of men employed by the U.S. trucking
33 industry in 1985 were matched to records in the National Death Index through 2000 (Hart
34 et al.. 2011). Using NO2 concentrations at their residential address, the association with
35 lung cancer mortality was examined. No association was detected and this persisted when
36 long-haul drivers who are away from the home at least one night per week were excluded
37 from the analyses. Similar results were observed for PM10 and SO2.
November 2013 5-146 DRAFT: Do Not Cite or Quote
-------
1 Multiple studies of NO2 and lung cancer mortality have been conducted in Asia. A study
2 in Japan followed individuals aged 65-84 years at enrollment for about 6 years (Yorifuji
3 et al., 2010). No overall association was reported between NO2 concentration and lung
4 cancer mortality. In stratified analyses, the association between NO2 concentration and
5 lung cancer mortality was higher among non-smokers compared to former/current
6 smokers but the findings were imprecise and the 95% confidence intervals overlapped.
7 No difference in the association was observed among other stratification variables.
8 Another study in Japan followed individuals for 10 years and observed a positive
9 association between NO2 concentration and lung cancer mortality (Katanoda et al., 2011).
10 An association was also observed for suspended PM (Pearson correlation coefficient with
11 NO2 = 0.26) but not for SO2. When the association between NO2 concentration and lung
12 cancer mortality was examined by region, the association appears to persist only in the
13 areas of study with the highest NO2 values (data on association by region only presented
14 in figures; numerical estimates not provided). A national study of urban areas in China
15 had a follow-up of less than 10 years and reported no association between NOX and lung
16 cancer mortality (Cao et al., 2011). This lack of association was robust to inclusion of
17 TSP or SO2, of which SO2 concentrations were found to be associated with lung cancer
18 mortality. A study performed in Taiwan used a case-control approach, comparing women
19 who died of lung cancer or other non-respiratory related causes (Liu et al.. 2008). The
20 highest tertile of NO2 concentration was positively associated with lung cancer mortality.
21 Associations between pollution concentrations and lung cancer mortality were also
22 observed for CO, but not SO2, PMi0, or O3. A combined exposure category was created
23 examining those women with estimated exposure concentrations of CO and NO2 in the
24 highest tertiles compared to those in the lowest tertiles. The results were similar to those
25 of the single-pollutant estimates.
26 Overall, there are inconsistent findings among studies of NO2 and lung cancer mortality.
27 The inconsistency appears unrelated to exposure assessment or length of follow-up
28 periods. Most of these studies controlled for confounders, such as smoking.
5.6.3 Leukemia Incidence and Mortality
29 A study of acute leukemia incidence identified cases from the French National Registry
30 of Childhood Blood Malignancies. Controls were randomly selected from the population
31 at a distribution of age and sex that matched that of the cases (Amigou et al.. 2011). NO2
32 concentration over 6.5 ppb were positively associated with the odds of leukemia. This
33 was also observed for specific types of leukemia. There was no difference in results
34 based on urban or rural residence. The authors stated that results were strengthened when
35 including only children who had been in the residences utilized in the study for at least 2
November 2013 5-147 DRAFT: Do Not Cite or Quote
-------
1 years (data not included in the paper). A study in Taiwan matched children with a cause
2 of death related to leukemia to children with a cause of death unrelated to neoplasms or
3 respiratory problems based on sex, year of birth, and year of death (Weng et al., 2008).
4 NO 2 concentrations were positively associated with the odds of death related to leukemia.
5.6.4 Bladder Cancer Mortality
5 A study performed in Taiwan examined mortality records, comparing individuals
6 (matched on sex, year of birth, and year of death) with and without mortality due to
7 bladder cancer (Liu et al.. 2009a). Increased odds of bladder cancer mortality was
8 associated with increased NO2 concentrations. This trend was also observed for SO2. The
9 highest tertile of PMi0 concentration was also associated with bladder cancer mortality
10 but no association was observed for CO or O3 concentrations. Liu et al. (2009a) further
11 examined a three-level variable, with the lowest level being individuals in the lowest
12 tertile of SO2 and NO2 concentrations (< 4.32 ppb and < 20.99 ppb, respectively), the
13 highest level being individuals in the highest tertile of SO2 and NO2 concentrations
14 (>6.49 ppb and >27.33 ppb, respectively), and all others being categorized in the middle.
15 The resulting ORs, adjusted for urbanization of residential area and marital status, were
16 1.37 (95% CI: 1.03, 1.82) for the middle level and 1.98 (95% CI: 1.36, 2.88) for the
17 highest level. Although the point estimates for NO2 and SO2 combined are higher than
18 those observed forNO2 or SO2 alone [see Supplemental Table S5-10, (U.S. EPA.
19 20J_3o)]), the 95% confidence intervals overlap. Therefore, the conclusion that NO2 and
20 SO2 combined contribute to higher odds of mortality than either alone cannot be drawn.
5.6.5 Breast Cancer Incidence
21 A Canadian study of post-menopausal breast cancer incidence using a hospital-based
22 case-control study design estimated NO2 concentrations using two methods:
23 extrapolating data from fixed-site monitoring stations or extrapolating data from
24 predicted concentrations determined with land-use regression using a dense network of
25 air samplers (Grouse etal. 2010). Although point estimates were elevated for some of the
26 associations between NO2 concentrations and post-menopausal breast cancer incidence,
27 most of the associations were not statistically significant. In sensitivity analyses limited
28 to subjects who were residents of the same address for at least 10 years prior to the study,
29 the point estimates were slightly higher but precision was reduced. This study suggests a
30 possible association between post-menopausal breast cancer incidence and NO2
31 concentration.
November 2013 5-148 DRAFT: Do Not Cite or Quote
-------
1 An ecologic study was performed by Wei et al. (2012) using data from the Surveillance,
2 Epidemiology, and End Results (SEER) program to determine the breast cancer incidence
3 rate of various states and metropolitan areas and data from the US EPA's Geographic
4 Area AirDatato determine NOX emissions. Results of Pearson's correlations
5 demonstrated a relationship between NOX emissions and breast cancer incidence. The
6 state with the highest NOX emissions also had the highest rate of breast cancer incidence
7 and the state with the lowest emissions had the lowest breast cancer incidence rate.
8 However, this study is limited by its ecologic nature and the lack of individual level data.
9 There is no control for potential confounders or examination of factors other than air
10 pollutants (of which CO, SO2, and VOCs, but not PMi0, also had positive correlations)
11 that could be associated with breast cancer incidence rates.
5.6.6 Prostate Cancer Incidence
12 Men enrolled in the Prostate Cancer and Environment Study (PROtEuS) were included in
13 an investigation of NO2 concentration and prostate cancer incidence (Parent et al.. 2013).
14 Cases were men diagnosed with prostate cancer and recruited through pathology
15 departments. Population-based controls were indentified through electoral lists and
16 frequency matched by five-year age groups. A positive association was observed between
17 recent NO2 concentration and odds of prostate cancer. The association was also observed
18 using back-extrapolated estimates of NO2 ten years prior. Multiple sensitivity analyses
19 were performed, including back-extrapolation of NO2 estimates for 20 years, addition of
20 smoking and alcohol consumption as confounders, exclusion of proxy subjects, exclusion
21 of subjects without a prostate cancer screening in the past 5 years, exclusion of subjects at
22 their residence for less than 10 years, and comparisons of subjects with geo-coding to
23 their exact address or to a centroid of their postal code. The results, while not always
24 statistically significant (in some parts due to decreases in sample size and precision),
25 were similar to the overall results reported.
5.6.7 Animal and In Vitro Carcinogenicity and Genotoxicity Studies
26 Animal toxicology studies characterizing the carcinogenicity and genotoxicity of NO2
27 follow. NO2 has been reported to act as a tumor promoter at the site of contact, i.e., in the
28 respiratory tract after inhalation exposure. This is consistent with mechanistic evidence of
29 observed hyperplasia of the lung epithelium with NO2 exposure (see Section 5.2.10).
30 Ex vivo exposure of human nasal epithelial mucosa cells cultured at the air-liquid
31 interface to 10 ppb NO2 (Koehler et al.. 2013: Koehler etal. 2010) or 100 ppb NO2
32 (Koehler et al., 2011) produced increased DNA fragmentation measured with the
November 2013 5-149 DRAFT: Do Not Cite or Quote
-------
1 COMET assay as early as 30 minutes after exposure and micronuclei formation after
2 3-hour exposure to 100 ppb NO2 (Koehler et al.. 2011). Percent of DNA content in the
3 tail as detected with the COMET assay decreased with increasing exposure duration (0.5,
4 1,2, and 3-hour exposure) (Koehler et al.. 2013). Of the in vivo assays reported in the
5 previous ISA [see U.S. EPA (2008c). Annex Table 4-12, Table 4-13, and Table 4-11, on
6 pages 4-36 and 4-37 of the 2008 Annex], results were mixed with positive findings of
7 genotoxicity seen in two studies that employed rat lung cells (mutations and chromosome
8 abnormalities, 50,000-560,000 ppb NO2 >12 days; 27,000 ppb NO2, 3 h) and negative
9 findings of genotoxicity seen in tests employing Drosophila recessive lethals
10 (500,000-7,000,000 ppb NO2, 1 h), Drosophila wing spot test (50,000-280,000 ppb NO2,
11 2 days), mouse bone marrow micronuclei (20,000 ppb, 23 h), and mouse spermatocyte
12 and lymphocyte chromosomal aberrations (100-10,000 ppb NO2, 6 h). In vitro exposures
13 to NO2 yielded positive findings in a majority of the tests in rodent (2,000-3,000 ppb
14 NO2, 10 minutes) and human cell lines, bacteria (5,000-90,000 ppb NO2, 30 minutes)
15 and plants (5,000 ppb NO2, 24 h).
5.6.8 Animal Toxicology Studies of Co-exposure with Known Carcinogens
16 The 1993 AQCD for Oxides of Nitrogen and the 2008 ISA for Oxides of Nitrogen
17 detailed NO2 co-exposure with known carcinogens. Rats injected with the carcinogen
18 N-bis (2-hydroxy-propyl) nitrosamine (BHPN) and continuously exposed to 40, 400 or
19 4.000 ppb NO2 for 17 months developed a non-statistically significant five-fold increase
20 in incidence of adenomas or adenocarcinomas versus control animals (4,000 ppb NO2)
21 (Ichinose et al.. 1991). Another study by the same lab (Ichinose and Sagai. 1992) showed
22 statistically significant increases in BHPN-induced lung tumors with combined NO2 + O3
23 exposure, a multipollutant effect absent with exposure to either single pollutant (BHPN
24 injection followed the next day by either clean air 0% NO2, 500 ppb NO2, 50 ppb
25 NO2 + 400 ppb O3, or 400 ppb O3 + 1 mg/m3 H2SO4 for 13 months, and then recovery
26 with clean air for another 11 months; continuous NO2 exposure, 11 h/day H2SO4 or O3
27 exposure).
28 Another study with co-exposure of F344 male rats to diesel exhaust particle extract-
29 coated carbon black particles (DEPcCBP) and NO2 and/or SO2 found significantly
30 increased incidences of lung tumors (alveolar adenomas) for the animals co-exposed to
31 DEcCBP and NO2 and/or SO2 but not in those with DEcCBP exposure alone (Ohyama et
32 al.. 1999). The National Toxicology Program's Report on Carcinogens has stated DEP is
33 reasonably anticipated to be a human carcinogen (NTP. 2011). Exposed rats received IT
34 installation of DEPcCBP once per week for 4 weeks, and 6,000 ppb NO2, 4,000 ppb SO2
November 2013 5-150 DRAFT: Do Not Cite or Quote
-------
1 or 6,000 ppb NO 2 + 4,000 ppb SO2 was administered 16 h/day for 8 months, and
2 followed by 8 months of clean air exposure.
5.6.9 Studies in Animals with Spontaneous High Tumor Rates
3 The previous ISA and AQCDs described studies in animals with spontaneously high
4 tumor rates including strain A/J mice, AKR/cum mice, and CAFl/Jax mice. Strain A/J
5 mice exposed to 10,000 ppb NO2 for 6 h/day, 5 days/week for 6 months (Adkins et al..
6 1986) had a small but statistically significant increase in pulmonary adenomas (increased
7 tumor multiplicity) with NO2 exposure (1,000 and 5,000 ppb NO2 had no effect). In
8 another study, increased survival rates of NO2-exposed animals were reported in a model
9 of spontaneous T cell lymphoma, i.e., AKR/cum mice that were exposed intermittently (7
10 hours/day, 5days/week) to 250 ppb NO2 for up to 26 weeks (Richters and Damii. 1990).
11 Another study using CAFl/Jax mice (Wagner et al.. 1965) showed that continuous
12 exposure to 5,000 ppb NO2 produced significant increases in the number of year-old
13 animals with pulmonary tumors when compared to control; this finding was no longer
14 significant at 14 or 16 months exposure.
5.6.10 Facilitation of Metastases
15 The previous ISA and AQCDs summarized a group of experiments by one lab that
16 focused on the role of NO2 in metastases facilitation. Richters and Kuraitis (1981).
17 Richters and Kuraitis (1983). and Richters et al. (1985) exposed mice to multiple
18 concentrations and durations of NO2, and after exposure the mice were injected I.V. with
19 the B16 melanoma cell line. Lung tumors were then counted with results of some of the
20 experiments showing significantly increased numbers of tumors.
5.6.11 Production of N-Nitroso Compounds and other Nitro Derivatives
21 Daily chemical transformations involving UV, NO2 and hydrocarbons, products of
22 automobile exhaust, and oxygen/ozone can generate peroxyacetyl nitrate (PAN) in the
23 gas fraction as part of photochemical smog. Mutagenicity assays demonstrated that PAN
24 is weakly mutagenic in the lungs of the highly susceptible big Blue (R) mice and in
25 Salmonella and that PAN produces a unique signature mutation (Demarini et al.. 2000).
26 N-nitroso compounds can be generated endogenously in the human body from NO2 via
27 processes that generate nitrite (NO2~) or nitrate. Further, NO2" is known to react with
November 2013 5-151 DRAFT: Do Not Cite or Quote
-------
1 amines to produce nitrosamines, known animal carcinogens. The possibility that NO2
2 could produce cancer via nitrosamine formation has been investigated and was reported
3 in the previous NOX ISA U.S. EPA (2008c).
November 2013 5-152 DRAFT: Do Not Cite or Quote
-------
Table 5-20
Reference
Koehler et al.
(2013)
Koehler et al.
(2010)
Koehler et al.
(2011)
Ohvama et al.
(1999)
Adkins et al.
(1986)
Richters and
Damii (1990)
Wagner et al.
(1965)
Richters and
Kuraitis
Animal toxicological studies of carcinogenicity and genotoxicity
Concentration
NO2
10 ppb
100, 1,000 or
10,000 ppb
100 ppb
1,000, 5,000,
6,000 ppb
10,000 ppb
250 ppb
5,000 ppb
400 or 800 ppb
Strain, Age,
Sex (n)
Human
Human
Human
F344 rats, male,
Strain A/J mice,
AKR/cum mice
CAF1/Jax mice
Swiss Webster
mice, males (24);
Exposure conditions
Ex vivo cell culture at the air liquid interface, primary
human nasal epithelia cells from n = 10 donors, NC>2
exposure for 0, 0.5, 1 , 2 and 3 h
Ex vivo cell culture at the air liquid interface, primary
human nasal epithelia cells from n - 10 donors, NC>2
exposure for 0 or 0.5 h
Ex vivo cell culture at the air liquid interface, primary
human nasal epithelia cells from n - 10 donors, NC>2
exposure for 0, 0.5, 1 , 2 and 3h
Exposure to diesel exhaust particle extract-coated
carbon black particles (DEPcCBP) and NO2 IT
installation of DEPcCBP 1x/week for 4 weeks. 6,000
ppb NC>2 was administered 16h/day for 8 mo, and
followed by 8 mo of clean air exposure.
Exposure of mice with spontaneous high tumor rates
to NC>2 for 6h/day, 5 days/week for 6 mo
Exposure of mice intermittently (7 h/day,
5days/week) to NC>2 for up to 26 weeks
Continuous exposure to 5,000 ppb NC>2
NO2 exposure 8h/day, 5days/week for 10weeks
(Swiss mice) or 12 weeks (C57BL/6J mice); Then all
with exposure to NO2.
Endpoints Examined
COMET assay, Micronucleus formation,
proliferation assay, apoptosis, necrosis,
cytotoxicity
COMET assay, Micronucleus formation,
proliferation assay, cytotoxicity
COMET assay, Micronucleus formation,
proliferation assay, apoptosis, necrosis,
cytotoxicity
Lung tumor incidence (alveolar adenomas)
Lung tumor multiplicity (pulmonary adenomas)
Rodent survival rate.
Lung tumor multiplicity at 12, 14 and 16 mo.
Facilitation of lung tumor metastasis (incidence of
lung tumors)
(1981)
C57BL/6J males
(90)
animals were infused i.v. with B16 melanoma cells
that are known to metastasize to the lung. 3 weeks
post-infusion, animals were sacrificed and lungs
scored for tumor incidence.
November 2013
5-153
DRAFT: Do Not Cite or Quote
-------
Table 5-20 (Continued): Animal toxicological studies of carcinogenicity and genotoxicity with exposure to NO2.
Concentration Strain, Age,
Reference NO2 Sex (n) Exposure conditions
Richters and 300, 400 or
Kuraitis 500 ppb
(1983)
C57BL/6J mice NC>2 exposure 7 h/day, 5 days/week for 10 weeks.
(25, 51, 23) Then all animals were infused i.v. with B16
melanoma cells that are known to metastasize to the
lung. 3 weeks post-infusion, animals were sacrificed
and lungs scored for tumor incidence.
Endpoints Examined
Facilitation of lung tumor metastasis (incidence of
lung tumors)
Richters et al. 400 ppb
C57BL/6J mice 12 weeks of continuous exposure to NO2 . Then all
animals were infused i.v. with B16 melanoma cells. 3
weeks post-infusion, animals were sacrificed and
lungs scored for tumor incidence.
Facilitation of lung tumor metastasis (incidence of
lung tumors)
Ichinose et al.
40, 400 or
4,000 ppb
Adult Rats
Co-exposure with carcinogen NHPN and NC>2.
exposure for 17 mo.
NC>2 Incidence of NHPN-induced lung tumors
(adenoma or adenocarcinomas).
Ichinose and
Saqai(1992)
500 ppb NO2;
50 ppb NO2 +
400 ppb O3
Adult rats
Carcinogen exposure plus air pollutant mixture
exposure (O3 + NO2). 500 ppb NO2, 50 ppb NO2 +
400 ppb Os, for 13 mo, and then recovery with clean
air for another 11 mo; continuous NO2 exposure, 11
h/day Os exposure
Incidence of NHPN-induced lung tumors
(adenoma or adenocarcinomas).
November 2013
5-154
DRAFT: Do Not Cite or Quote
-------
5.6.12 Summary and Causal Determination
1 The overall evidence for long-term NO2 exposure and cancer is suggestive of a causal
2 relationship. This conclusion is based on evidence from some prospective epidemiologic
3 studies reporting associations between NO2 or NOX exposure and cancer incidence and
4 mortality. Animal toxicology studies employing NO2 exposure with other known
5 carcinogens provide further supporting evidence, showing that inhaled NO2 can increase
6 tumor load in laboratory rodents. Nonetheless, toxicological data provide no clear
7 evidence of NO2 acting as a complete carcinogen and not all epidemiologic studies report
8 positive associations.
9 In past reviews, a limited number of epidemiologic studies had assessed the relationship
10 between long-term NO2 or NOX exposure and cancer incidence and mortality. The 2008
11 ISA for Oxides of Nitrogen concluded that the evidence was "inadequate to infer the
12 presence or absence of a causal relationship" (U.S. EPA. 2008c). Recent studies include
13 evidence on lung cancer as well as new types of cancer, evaluating both incidence and
14 mortality. All available evidence for cancer due to long-term NO2 or NOX exposure was
15 evaluated using the framework described in Table II of the Preamble. The key evidence
16 as it relates to the causal framework is summarized in Table 5-21.
17 Epidemiologic studies of NO2 or NOX and lung cancer incidence have had mixed results,
18 with some studies reporting no associations while other studies report positive
19 associations. Most of these studies included large sample sizes, similar NOX or NO2
20 concentrations, and control for many potential confounders, including smoking
21 exposures. Most studies of NO2 or NOX and lung cancer mortality reported no
22 association, but there are some studies reporting positive associations. Recent studies of
23 leukemia have reported associations with NO2 concentration. Similarly, a study of
24 bladder cancer mortality reported an association with NO2. Breast cancer incidence was
25 positively correlated with NOX concentration in an ecologic analysis but a study of post-
26 menopausal women observed no increase in odds with higher NO2 concentrations. A
27 positive association was observed between NO2 concentration and prostate cancer
28 incidence. Toxicological data provide no clear evidence of NO2 acting as a complete
29 carcinogen and agencies that classify carcinogens including the Department of Health
30 and Human Services, the International Agency for Research on Cancer, and the US EPA
31 have not classified oxides of nitrogen for potential carcinogenicity. The American
32 Conference of Industrial Hygienists has classified NO2 as A4 (Not classifiable for
33 humans or animals). However, in some animal toxicological models NO2 may act as a
34 tumor promoter at the site of contact, possibly due to its ability to produce cellular
35 damage, induce respiratory epithelial hyperplasia (Section 5.2.10). or promote
November 2013 5-155 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
5
6
regenerative cell proliferation. Genotoxic and mutagenic studies with NO2 have mixed
results. Some studies with co-exposure to other known carcinogens demonstrated that
inhaled NO2 can increase tumor burden in rodents. Collectively, while some studies
observed no associations, the evidence from several, high-quality toxicological and
epidemiologic studies is suggestive of a causal relationship between long-term exposure
to NO2 and cancer incidence and mortality.
Table 5-21 Summary of evidence supporting a suggestive of a causal
relationship between long-term NO2 exposure and cancer.
Rationale for
Causal
Determination3
Key Evidence
Key References
NO2 or NOX
Concentrations
Associated with
Effects0
Cancer - Suggestive
Some high-quality,
epidemiologic
studies provide
evidence of
associations.
Positive associations were
observed between overall cancer
incidence and mortality in
multiple studies conducted in
Europe and Asia
Positive associations were also
observed in studies of NC>2
concentrations and leukemia,
bladder cancer, and prostate
cancer.
Nafstad et al. (2003).
Nvberq et al. (2000).
Raaschou-Nielsen et al.
(201 Oa).
Raaschou-Nielsen et al.
(2011),
Cesaroni et al. (2013),
Filleuletal. (2005),
Katanoda et al. (2011).
Liu et al. (2008).
Naess et al. (2007)
Sections 5.6.1-5.6.6
Amiqou et al. (2011).
Wenq et al. (2008).
Liu et al. (2009a).
Parent et al. (2013)
Sections 5.6.1-5.6.6
Means varied with some
studies including areas
estimating
concentrations of NO2
or NOx as low as 1.2
ppb to studies with
areas estimated at 32.4
ppb.
Associations observed
at levels as low as
6.5-8.6 ppb for leukemia
Some high-quality,
epidemiologic
studies demonstrate
no associations.
No associations were observed
between overall cancer incidence
and mortality in multiple studies
conducted in the United States,
Europe, and Asia.
Brunekreef et al. (2009).
Beelen et al. (2008a).
Papathomas et al. (2011),
Brunekreef et al. (2009),
Beelen et al. (2008a),
Caoetal. (2011),
Hartetal. (2011),
Heinrichetal. (2013).
Krewski et al. (2009).
Yorifuii et al. (2010)
Sections 5.6.1-5.6.6
Means varied with
estimated
concentrations of NC>2
or NOX ranging from
13.3 to 27.9 ppb.
November 2013
5-156
DRAFT: Do Not Cite or Quote
-------
Table 5-21 (Continued): Summary of evidence supporting a suggestive of a causal
relationship between long-term NOi exposure and cancer.
Rationale for
Causal
Determination3 Key Evidence13
NO2 or NOX
Concentrations
Associated with
Key References" Effects0
Limited evidence
from high-quality,
toxicological studies
Studies of facilitation of
metastasis and co-exposures
with known carcinogens show
NOX related effects. Studies of
NOx as a direct carcinogen are
lacking.
Adkinsetal. (1986).
Richters and Damii (1990).
Wagner et al. (1965),
Richters and Kuraitis (1981).
Richters and Kuraitis (1983),
Richters et al. (1985),
Ichinose et al. (1991).
Ichinose and Saqai (1992)
Sections 5.6.8 and 5.6.10
10,000 ppb
250 ppb
5,000 ppb
400, 800 ppb
300, 400, 500 ppb
400 ppb
4,000 ppb
500 ppb
Limited evidence for
key events to inform
mode of action
Finding of mutagenicity and
micronucleus formation in ex vivo
culture of primary human nasal
epithelial cells exposed to NO2.
Mixed findings of mutagenicity
and carcinogenicity in various
models of NO2 exposure in older
studies, mainly in non-human
species.
Koehler et al. (2013),
Koehler et al. (2011),
Koehler et al. (2010)
Section 5.6.7
fU.S. EPA (2008c), Annex
Table AX4-11,
Table AX 4-12, and
Table AX 4-13]
100, 1,000, 10,000 ppb
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table I
Preamble.
""Describes the key evidence and references contributing most heavily to causal determination and, where applicable, to
uncertainties or inconsistencies. References to earlier sections indicate where full body of evidence is described.
°Describes the NO2 concentrations with which the evidence is substantiated (for experimental studies, below 5,000 ppb).
ofthe
November 2013
5-157
DRAFT: Do Not Cite or Quote
-------
References for Chapter 5
Abbey. DE; Nishino. N; McDonnell. WF; Burchette. RJ: Knutsen. SF; Beeson. WL; Yang. JX. (1999).
Long-term inhalable particles and other air pollutants related to mortality in nonsmokers. Am J Respir
Crit Care Med 159: 373-382. http://dx.doi.org/10.1164/airccm. 159.2.9806020
Ackermann-Liebrich. U; Leuenberger. P; Schwartz. J; Schindler. C: Monn. C: Bolognini. G: Bongard. JP;
Brandli. O: Domenighetti. G: Elsasser. S: Grize. L; Karrer. W: Keller. R; Keller-Wossidlo. H; Kiinzli.
N; Martin. BW: Medici. TC: Perruchoud. AP; Schoni. MH; Tschopp. JM; Villiger. B; Wiithrich. B;
Zellweger. JP; Zemp. E. (1997). Lung function and long term exposure to air pollutants in Switzerland.
Am J Respir Crit Care Med 155: 122-129. http://dx.doi.Org/10.1164/airccm.155.l.9001300
Adkins. B. Jr; Van Stee. EW: Simmons. JE; Eustis. SL. (1986). Oncogenic response of strain A/J mice to
inhaled chemicals. J Toxicol Environ Health 17: 311-322.
http://dx.doi.org/10.1080/15287398609530825
Aguilera. I; Garcia-Esteban. R; Iniguez. C: Nieuwenhuijsen. MJ; Rodriguez. A; Paez. M; Ballester. F;
Sunyer. J. (2010). Prenatal exposure to traffic-related air pollution and ultrasound measures of fetal
growth in the INMA Sabadell cohort. Environ Health Perspect 118: 705-711.
http://dx.doi.org/10.1289/ehp.0901228
Aguilera. I; Guxens. M; Garcia-Esteban. R; Corbella. T; Nieuwenhuijsen. MJ: Foradada. CM; Sunyer. J.
(2009). Association between GIS-based exposure to urban air pollution during pregnancy and birth
weight in the INMA Sabadell cohort. Environ Health Perspect 117: 1322-1327.
Aguilera. I; Pedersen. M; Garcia-Esteban. R; Ballester. F; Basterrechea. M; Esplugues. A; Fernandez-
Somoano. A; Lertxundi. A; Tardon. A; Sunyer. J. (2013). Early-life exposure to outdoor air pollution
and respiratory health, ear infections, and eczema in infants from the INMA study. Environ Health
Perspect 121: 387-392. http://dx.doi.org/10.1289/ehp.1205281
Akinbami. LJ; Lynch. CD; Parker. JD; Woodruff. TJ. (2010). The association between childhood asthma
prevalence and monitored air pollutants in metropolitan areas, United States, 2001-2004. Environ Res
110: 294-301. http://dx.doi.Org/10.1016/j.envres.2010.01.001
Amigou. A; Sermage-Faure. C: Orsi. L; Leverger. G: Baruchel. A; Bertrand. Y; Nelken. B; Robert. A;
Michel. G: Margueritte. G: Perel. Y; Mechinaud. F; Bordigoni. P; Hemon. D; Clavel. J. (2011). Road
traffic and childhood leukemia: The ESCALE Study (SFCE). Environ Health Perspect 119: 566-572.
http://dx.doi.org/10.1289/ehp. 1002429
Andersen. ZJ: B0nnelykke. K; Hvidberg. M; Jensen. SS: Ketzel. M; Loft. S: S0rensen. M; Tj0nneland. A;
Overvad. K; Raaschou-Nielsen. O. (2012). Long-term exposure to air pollution and asthma
hospitalisations in older adults: A cohort study. Thorax 67: 6-11. http://dx.doi.org/10.1136/thoraxjnl-
2011-200711
Andersen. ZJ: Hvidberg. M; Jensen. SS: Ketzel. M; Loft. S: S0rensen. M; Tj0nneland. A; Overvad. K;
Raaschou-Nielsen. O. (2011). Chronic obstructive pulmonary disease and long-term exposure to traffic-
related air pollution: A cohort study. Am J Respir Crit Care Med 183: 455-461.
http://dx.doi.org/10.1164/rccm.201006-0937OC
Anderson. HR; Favarato. G: Atkinson. RW. (2013). Long-term exposure to air pollution and the incidence
of asthma: Meta-analysis of cohort studies. Air Qual Atmos Health 6: 47-56.
http://dx.doi.org/10.1007/sll869-011-0144-5
Annesi-Maesano. I; Hulin. M; Lavaud. F; Raherison. C: Kopferschmitt. C: de Slav. F; Charpin. DA; Denis.
Cl (2012a). Poor air quality in classrooms related to asthma and rhinitis in primary schoolchildren of the
French 6 Cities Study. Thorax 67: 682-688. http://dx.doi.org/10.1136/thoraxinl-2011-200391
November 2013 5-158 DRAFT: Do Not Cite or Quote
-------
Annesi-Maesano. I; Moreau. D; Caillaud. D; Lavaud. F; Le Moullec. Y; Tavtard. A; Pauli. G: Charpin. D.
(2007). Residential proximity fine particles related to allergic sensitisation and asthma in primary school
children. RespirMed 101: 1721-1729. http://dx.doi.Org/10.1016/j.rmed.2007.02.022
Aranyi. C: Fenters. J: Erhlich. R; Gardner. D. (1976). Scanning electron microscopy of alveolar
macrophages after exposure to oxygen, nitrogen dioxide, and ozone [Abstract]. Environ Health Perspect
16: 180.
Arner. EC: Rhoades. RA. (1973). Long-term nitrogen dioxide exposure: Effects on lung lipids and
mechanical properties. Arch Environ Health 26: 156-160.
http://dx.doi.org/10.1080/00039896.1973.10666244
Atkinson. RW: Carey. IM; Kent. AJ; van Staa. TP; Anderson. HR; Cook. DG. (2013). Long-term exposure
to outdoor air pollution and incidence of cardiovascular diseases. Epidemiology 24: 44-53.
http://dx.doi.org/10.1097/EDE.Ob013e318276ccb8
Ayaz. KL; Csallany. AS. (1978). Long-term NO2 exposure of mice in the presence and absence of vitamin
E. II. Effect of glutathione peroxidase. Arch Environ Occup Health 33: 292-296.
Ballester. F; Estarlich. M; Iniguez. C: Llop. S: Ramon. R; Esplugues. A; Lacasafia. M; Rebagliato. M.
(2010). Air pollution exposure during pregnancy and reduced birth size: A prospective birth cohort
study in Valencia, Spain. Environ Health 9: 6. http://dx.doi.org/10.1186/1476-069X-9-6
Basu. R: Malig. B; Ostro. B. (2010). High ambient temperature and the risk of preterm delivery. Am J
Epidemiol 172: 1108-1117. http://dx.doi.org/l0.1093/aie/kwa 170
Becerra. TA; Wilhelm. M; Olsen. J: Cockburn. M; Ritz. B. (2013). Ambient air pollution and autism in Los
Angeles county, California. Environ Health Perspect 121: 380-386.
http://dx.doi.org/10.1289/ehp.1205827
Beelen. R; Hoek. G: van den Brandt. PA; Goldbohm. RA; Fischer. P; Schouten. LJ; Armstrong. B;
Brunekreef. B. (2008a). Long-term exposure to traffic-related air pollution and lung cancer risk.
Epidemiology 19: 702-710. http://dx.doi.org/10.1097/EDE.Ob013e318181b3ca
Beelen. R; Hoek. G: van den Brandt. PA; Goldbohm. RA: Fischer. P; Schouten. LJ: Jerrett. M; Hughes. E;
Armstrong. B; Brunekreef. B. (2008b). Long-term effects of traffic-related air pollution on mortality in
a Dutch cohort (NLCS-AIR study). Environ Health Perspect 116: 196-202.
http://dx.doi.org/10.1289/ehp.10767
Belanger. K; Holford. TR; Gent. JF; Hill. ME; Kezik. JM; Leaderer. BP (2013). Household levels of
nitrogen dioxide and pediatric asthma severity. Epidemiology 24: 320-330.
http://dx.doi.org/10.1097/EDE.Ob013e318280e2ac
Bell. ML; Ebisu. K; Belanger. K. (2007). Ambient air pollution and low birth weight in Connecticut and
Massachusetts. Environ Health Perspect 115: 1118-1124. http://dx.doi.org/10.1289/ehp.9759
Bentaveb. M; Helmer. C: Raherison. C: Dartigues. JF; Tessier. JF; Annesi-Maesano. I. (2010). Bronchitis-
like symptoms and proximity air pollution in French elderly. Respir Med 104: 880-888.
http://dx.doi.0rg/10.1016/i.rmed.2010.01.004
Bernstein. L; Allen. M; Anton-Culver. H; Deapen. D; Horn-Ross. PL; Peel. D: Finder. R; Reynolds. P:
Sullivan-Hallev. J: West. D: Wright W: Ziogas. A; Ross. RK. (2002). High breast cancer incidence rates
among California teachers: Results from the California Teachers Study (United States). Cancer Causes
Control 13: 625-635.
Blair. WH; Henry. MC: Ehrlich. R. (1969). Chronic toxicity of nitrogen dioxide: II. Effect on
histopathology of lung tissue. Arch Environ Health 18: 186-192.
http://dx.doi.org/10.1080/00039896.1969.10665391
Bobak. M. (2000). Outdoor air pollution, low birth weight, and prematurity. Environ Health Perspect 108:
173-176.
November 2013 5-159 DRAFT: Do Not Cite or Quote
-------
Boggia. B; Carbone. U: Farinaro. E; Zarrilli. S: Lombardi. G: Colao. A; De Rosa. N; De Rosa. M. (2009).
Effects of working posture and exposure to traffic pollutants on sperm quality. J Endocrinol Invest 32:
430-434.
Boogaard. H; Fischer. PH; Janssen. NA; Kos. GP; Weijers. EP; Cassee. FR; van der Zee. SC: de Hartog. JJ:
Meliefste. K; Wang. M; Brunekreef. B; Hoek. G. (2013). Respiratory effects of a reduction in outdoor
air pollution concentrations. Epidemiology 24: 753-761.
http://dx.doi.org/10.1097/EDE.Ob013e31829el639
Braback. L; Forsberg. B. (2009). Does traffic exhaust contribute to the development of asthma and allergic
sensitization in children: findings from recent cohort studies [Review]. Environ Health Global Access
Sci Source 8:17. http://dx.doi.org/10.1186/1476-069X-8-17
Brauer. M; Hoek. G: Smit HA; de Jongste. JC: Gerritsen. J; Postma. PS: Kerkhof. M; Brunekreef. B.
(2007). Air pollution and development of asthma, allergy and infections in a birth cohort. Eur Respir J
29: 879-888. http://dx.doi.org/10.1183/09031936.00083406
Brauer. M; Hoek. G: Van Vliet. P; Meliefste. K; Fischer. PH; Wijga. A; Koopman. LP; Neijens. HJ;
Gerritsen. J: Kerkhof. M; Heinrich. J: Bellander. T; Brunekreef. B. (2002). Air pollution from traffic and
the development of respiratory infections and asthmatic and allergic symptoms in children. Am J Respir
Crit Care Med 166: 1092-1098. http://dx.doi.org/10.1164/rccm.200108-007OC
Brauer. M; Lencar. C: Tamburic. L; Koehoorn. M; Demers. P; Karr. C. (2008). A cohort study of traffic-
related air pollution impacts on birth outcomes. Environ Health Perspect 116: 680-686.
http://dx.doi.org/10.1289/ehp.10952
Breton. CV: Salam. MT; Vora. H; Gauderman. WJ: Gilliland. FD. (2011). Genetic variation in the
glutathione synthesis pathway, air pollution, and children's lung function growth. Am J Respir Crit Care
Med 183: 243-248. http://dx.doi.org/10.1164/rccm.201006-0849OC
Brunekreef. B; Beelen. R; Hoek. G: Schouten. L; Bausch-Goldbohm. S: Fischer. P; Armstrong. B; Hughes.
E; Jerrett. M; van den Brandt. P. (2009). Effects of long-term exposure to traffic-related air pollution on
respiratory and cardiovascular mortality in the Netherlands: The NLCS-AIR Study. (139). Boston, MA:
Health Effects Institute.
Cao. J: Yang. C: Li. J: Chen. R; Chen. B; Gu. B; Kan. H. (2011). Association between long-term exposure
to outdoor air pollution and mortality in China: A cohort study. J Hazard Mater 186: 1594-1600.
http://dx.doi.0rg/10.1016/i.jhazmat.2010.12.036
Carlsten. C: Brauer. M; Dimich-Ward. H; Dybuncio. A; Becker. AB; Chan-Yeung. M. (2011 a). Combined
exposure to dog and indoor pollution: incident asthma in a high-risk birth cohort. Eur Respir J 37: 324-
330. http://dx.doi.org/10.1183/09031936.00187609
Carlsten. C: Dybuncio. A; Becker. A; Chan-Yeung. M; Brauer. M. (2011b). GSTP1 polymorphism modifies
risk for incident asthma associated with nitrogen dioxide in a high-risk birth cohort [Letter]. Occup
Environ Med In Press: In Press, http://dx.doi.org/10.1136/oem.2010.063560
Carlsten. C: Dybuncio. A; Becker. A; Chan-Yeung. M; Brauer. M. (2011c). Traffic-related air pollution and
incident asthma in a high-risk birth cohort. Occup Environ Med 68: 291-295.
http://dx.doi.org/10.1136/oem.2010.055152
Castro-Giner. F; Kiinzli. N; Jacquemin. B; Forsberg. B; de Cid. R: Sunyer. J: Jarvis. D; Briggs. D;
Vienneau. D; Norback. D; Gonzalez. JR; Guerra. S: Janson. C: Anto. JM; Wist. M; Heinrich. J: Estivill.
X; Kogevinas. M. (2009). Traffic-related air pollution, oxidative stress genes, and asthma (ECHRS).
Environ Health Perspect 117: 1919-1924.
Cesaroni. G: Badaloni. C: Gariazzo. C: Stafoggia. M; Sozzi. R; Davoli. M; Forastiere. F. (2013). Long-term
exposure to urban air pollution and mortality in a cohort of more than a million adults in rome. Environ
Health Perspect 121: 324-331. http://dx.doi.org/10.1289/ehp.1205862
Chang. LY; Graham. JA; Miller. FJ; Ospital. JJ: Crapo. JD. (1986). Effects of subchronic inhalation of low
concentrations of nitrogen dioxide: I. The proximal alveolar region of juvenile and adult rats. Toxicol
Appl Pharmacol 83: 46-61. http://dx.doi.org/10.1016/0041-008X(86)90321-2
November 2013 5-160 DRAFT: Do Not Cite or Quote
-------
Channell. MM; Paffett ML; Devlin. RB; Madden. MC: Campen. MJ. (2012). Circulating factors induce
coronary endothelial cell activation following exposure to inhaled diesel exhaust and nitrogen dioxide in
humans: Evidence from a novel translational in vitro model. Toxicol Sci 127: 179-186.
http://dx.doi.org/10.1093/toxsci/kfs084
Chen. H; Goldberg. MS: Burnett. RT; Jerrett. M; Wheeler. AJ; Villeneuve. PJ. (2013). Long-term exposure
to traffic-related air pollution and cardiovascular mortality. Epidemiology 24: 35-43.
http://dx.doi.org/10.1097/EDE.Ob013e318276c005
Chen. L; Bell. EM; Caton. AR; Druschel. CM; Lin. S. (2010a). Residential mobility during pregnancy and
the potential for ambient air pollution exposure misclassification. Environ Res 110: 162-168.
http://dx.doi.0rg/10.1016/i.envres.2009.ll.001
Cheng. MF; Tsai. SS: Wu. TN; Chen. PS: Yang. CY. (2007). Air pollution and hospital admissions for
pneumonia in a tropical city: Kaohsiung, Taiwan. J Toxicol Environ Health A 70: 2021-2026.
http://dx.doi.org/10.1080/15287390701601020
Chiu. HF; Cheng. MH; Yang. CY. (2009). Air pollution and hospital admissions for pneumonia in a
subtropical city: Taipei, Taiwan. Inhal Toxicol 21: 32-37. http://dx.doi.org/10.1080/08958370802441198
Clark. C: Crombie. R; Head. J: van Kamp. I; van Kempen. E; Stansfeld. SA. (2012). Does traffic-related air
pollution explain associations of aircraft and road traffic noise exposure on children's health and
cognition? A secondary analysis of the United Kingdom sample from the RANCH project. Am J
Epidemiol 176: 327-337. http://dx.doi.org/10.1093/aie/kws012
Clark. NA; Demers. PA; Karr. CJ: Koehoorn. M; Lencar. C: Tamburic. L; Brauer. M. (2010). Effect of early
life exposure to air pollution on development of childhood asthma. Environ Health Perspect 118: 284-
290. http://dx.doi.org/10.1289/ehp.0900916
Cloughertv. JE; Lew. JI; Kubzansky. LD; Ryan. PB; Suglia. SF; Canner. MJ: Wright. RJ. (2007).
Synergistic effects of traffic-related air pollution and exposure to violence on urban asthma etiology.
Environ Health Perspect 115: 1140-1146. http://dx.doi.org/10.1289/ehp.9863
Crapo. JD; Barry. BE; Chang. LY; Mercer. RR. (1984). Alterations in lung structure caused by inhalation of
oxidants. J Toxicol Environ Health 13: 301-321. http://dx.doi.org/10.1080/15287398409530500
Grouse. PL; Goldberg. MS: Ross. NA; Chen. H; Labreche. F. (2010). Postmenopausal breast cancer is
associated with exposure to traffic-related air pollution in Montreal, Canada: A casecontrol study.
Environ Health Perspect 118: 1578-1583. http://dx.doi.org/10.1289/ehp.1002221
Dales. R: Burnett. RT; Smith-Doiron. M; Stieb. DM; Brook. JR. (2004). Air pollution and sudden infant
death syndrome. Pediatrics 113: 628-631.
Dales. R: Wheeler. A; Mahmud. M; Frescura. AM; Smith-Doiron. M; Nethery. E; Liu. L. (2008). The
influence of living near roadways on spirometry and exhaled nitric oxide in elementary schoolchildren.
Environ Health Perspect 116: 1423-1427. http://dx.doi.org/10.1289/ehp.10943
Darrow. LA; Klein. M; Flanders. WD; Waller. LA; Correa. A; Marcus. M; Mulholland. JA; Russell. AG:
Tolbert. PE. (2009). Ambient air pollution and preterm birth: A time-series analysis. Epidemiology 20:
689-698. http://dx.doi.org/10.1097/EDE.Ob013e3181a7128f
Darrow. LA; Klein. M; Strickland. MJ: Mulholland. JA; Tolbert. PE. (20 lib). Ambient air pollution and
birth weight in full-term infants in Atlanta, 1994-2004. Environ Health Perspect 119: 731-737.
http://dx.doi.org/10.1289/ehp.1002785
de Burbure. CY; Heilier. JF; Neve. J: Becker. A; Albrecht. C: Borm. PJA; Gromadzinska. J: Wasowicz. W:
Rydzynski. K; Bernard. AM. (2007). Lung permeability, antioxidant status, and NO2 inhalation: A
selenium supplementation study in rats. J Toxicol Environ Health A 70: 284-294.
http://dx.doi.org/10.1080/15287390600884875
Delamater. PL; Finlev. AO; Banerjee. S. (2012). An analysis of asthma hospitalizations, air pollution, and
weather conditions in Los Angeles County, California. Sci Total Environ 425: 110-118.
http://dx.doi.0rg/10.1016/i.scitotenv.2012.02.015
November 2013 5-161 DRAFT: Do Not Cite or Quote
-------
Demarini. DM; Shelton. ML; Kohan. MJ; Hudgens. EE; Kleindienst. TE; Ball. LM; Walsh. D; de Boer. JG:
Lewis-Sevan. L; Rabinowitz. J. R.; Claxton. LD; Lewtas. J. (2000). Mutagenicity in lung of Big
Blue(R) mice and induction of tandem-base substitutions in Salmonella by the air pollutant
peroxyacetyl nitrate (PAN): Predicted formation of intrastrand cross-links. Mutat Res-Fundam Mol
Mech Mutagen 457: 41-55. http://dx.doi.org/10.1016/80027-5107(00)00121-4
Di Giovanni. V; Cagiano. R; Carratu. MR; De Salvia. MA; Giustino. A; Cuomo. V. (1994). Alterations in
the ontogeny of rat pup ultrasonic vocalization produced by prenatal exposure to nitrogen dioxide.
Psychopharmacology 116: 423-427.
Dockerv. DW: Pope. CA. Ill: Xu. X: Spengler. JD: Ware. JH: Fav. ME: Ferris. BG. Jr: Speizer. FE. (1993).
An association between air pollution and mortality in six US cities. N Engl J Med 329: 1753-1759.
http://dx.doi.org/10.1056/NEJM199312093292401
Dong. GH; Chen. T; Liu. MM; Wang. D; Ma. YN; Ren. WH; Lee. YL; Zhao. YD; He. QC. (2011). Gender
differences and effect of air pollution on asthma in children with and without allergic predisposition:
Northeast Chinese children health study. PLoS ONE 6: e22470.
http://dx.doi.org/10.1371/iournal.pone.0022470
Dong. GH: Zhang. P: Sun. B: Zhang. L: Chen. X: Ma. N: Yu. F: Guo. H: Huang. H: Lee. YL: Tang. N:
Chen. J. (2012). Long-term exposure to ambient air pollution and respiratory disease mortality in
Shenyang, China: A 12-year population-based retrospective cohort study. Respiration 84: 360-368.
http://dx.doi.org/10.1159/000332930
Douglas. GJ: Price. JF; Page. CP (1995). The effect of prolonged exposure to NO2 from birth on airways
responsiveness in rabbits sensitized at birth. Eur Respir J 8: 246-252.
http://dx.doi.org/10.1183/09031936.95.08020246
Ebisu. K; Bell. ML. (2012). Airborne PM2.5 chemical components and low birth weight in the
Northeastern and Mid-Atlantic regions of the United States. Environ Health Perspect 120: 1746-1752.
http://dx.doi.org/10.1289/ehp.1104763
Ehrlich. R. (1980). Interaction between environmental pollutants and respiratory infections. Environ Health
Perspect 35: 89-99.
Ehrlich. R; Henry. MC. (1968). Chronic toxicity of nitrogen dioxide: I: Effect on resistance to bacterial
pneumonia. Arch Environ Health 17: 860-865. http://dx.doi.org/10.1080/00039896.1968.10665342
Estarlich. M; Ballester. F; Aguilera. I; Fernandez-Somoano. A; Lertxundi. A; Llop. S: Freire. C: Tardon. A;
Basterrechea. M; Sunyer. J: Iniguez. C. (2011). Residential exposure to outdoor air pollution during
pregnancy and anthropometric measures at birth in a multicenter cohort in Spain. Environ Health
Perspect 119: 1333-1338. http://dx.doi.org/10.1289/ehp.1002918
Faiz. AS: Rhoads. GG: Demissie. K; Kruse. L; Lin. Y; Rich. DQ. (2012). Ambient air pollution and the risk
of stillbirth. Am J Epidemiol 176: 308-316. http://dx.doi.org/10.1093/aje/kws029
Feichtinger. W: Papalambrou. K; Poehl. M; Krischker. U; Neumann. K. (1997). Smoking and in vitro
fertilization: Ameta-analysis. J Assist Reprod Genet 14: 596-599.
http://dx.doi.Org/10.1023/A:1022584802711
Felber Dietrich. D; Gemperli. A; Gaspoz. JM; Schindler. C: Liu. LJ; Gold. PR; Schwartz. J: Rochat. T;
Barthelemy. JC: Pons. M; Roche. F; Probst Hensch. NM; Bridevaux. PO; Gerbase. MW: Neu. U:
Ackermann-Liebrich. U. (2008). Differences in heart rate variability associated with long-term exposure
to NO2. Environ Health Perspect 116: 1357-1361. http://dx.doi.org/10.1289/ehp.11377
Fenters. JD: Findlav. JC: Port. CD; Ehrlich. R; Coffin. PL. (1973). Chronic exposure to nitrogen dioxide:
Immunologic, physiologic, and pathologic effects in virus-challenged squirrel monkeys. Arch Environ
Health 27: 85-89. http://dx.doi.org/10.1080/00039896.1973.10666324
Filleul. L; Rondeau. V: Vandentorren. S: Le Moual. N; Cantagrel. A; Annesi-Maesano. I; Charpin. D;
Declercq. C: Neukirch. F; Paris. C: Vervloet. D: Brochard. P; Tessier. JF; Kauffmann. F; Baldi. I.
(2005). Twenty five year mortality and air pollution: results from the French PAARC survey. Occup
Environ Med 62: 453-460. http://dx.doi.org/10.1136/oem.2004.014746
November 2013 5-162 DRAFT: Do Not Cite or Quote
-------
Forbes. LJ; Patel. MD; Rudnicka. AR; Cook. DG: Bush. T; Stedman. JR; Whincup. PH; Strachan. DP;
Anderson. HR. (2009a). Chronic exposure to outdoor air pollution and diagnosed cardiovascular
disease: meta-analysis of three large cross-sectional surveys. Environ Health Global Access Sci Source
8: 30. http://dx.doi.org/10.1186/1476-069X-8-30
Forbes. LJL; Kapetanakis. V: Rudnicka. AR; Cook. DG: Bush. T; Stedman. JR; Whincup. PH; Strachan.
DP; Anderson. HR. (2009b). Chronic exposure to outdoor air pollution and lung function in adults.
Thorax 64: 657-663. http://dx.doi.org/10.1136/thx.2008.109389
Freire. C: Ramos. R; Puertas. R; Lopez-Espinosa. MJ; Julvez. J; Aguilera. I; Cruz. F; Fernandez. MF;
Sunyer. J; Olea. N. (2010). Association of traffic-related air pollution with cognitive development in
children. J Epidemiol Community Health 64: 223-228. http://dx.doi.org/10.1136/iech.2008.084574
Fu. A; Leaderer. BP; Gent. JF; Leaderer. D; Zhu. Y. (2012a). An environmental epigenetic study of ADRB2
5'-UTR methylation and childhood asthma severity. Clin Exp Allergy 42: 1575-1581.
http://dx.doi.0rg/10.llll/i.1365-2222.2012.04055.x
Fujimaki. H; Nohara. O. (1994). Changes in the response of lung mast cells isolated from rats and guinea
pigs exposed to nitrogen dioxide. Inhal Toxicol 6: 515-520.
http://dx.doi.org/10.3109/08958379409040508
Furiosi. NJ; Crane. SC: Freeman. G. (1973). Mixed sodium chloride aerosol and nitrogen dioxide in air:
Biological effects on monkeys and rats. Arch Environ Occup Health 27: 405-408.
Gan. W: Koehoorn. M; Davies. H; Demers. P; Tamburic. L; Brauer. M. (2011). Long-term exposure to
traffic-related air pollution and the risk of coronary heart disease hospitalization and mortality. Environ
Health Perspect 119: 501-507. http://dx.doi.org/10.1289/ehp. 1002511
Gan. WQ: Fitzgerald. JM; Carlsten. C: Sadatsafavi. M; Brauer. M. (2013). Associations of ambient air
pollution with chronic obstructive pulmonary disease hospitalization and mortality. Am J Respir Crit
Care Med 187: Ill-Ill. http://dx.doi.org/10.1164/rccm.201211-2004OC
Gao. Y; Chan. EY; Li. LP; He. QQ: Wong. TW. (2013). Chronic effects of ambient air pollution on lung
function among Chinese children. Arch Dis Child 98: 128-135. http://dx.doi.org/10.1136/archdischild-
2011-301541
Gasana. J: Dillikar. D; Mendv. A; Forno. E; Ramos Vieira. E. (2012). Motor vehicle air pollution and
asthma in children: A meta-analysis. Environ Res 117: 36-45.
http://dx.doi.0rg/10.1016/i.envres.2012.05.001
Gauderman. WJ: Avol. E; Gilliland. F; Vora. H; Thomas. D: Berhane. K; Mcconnell. R: Kuenzli. N;
Lurmann. F; Rappaport. E; Margolis. H; Bates. D: Peters. J. (2004). The effect of air pollution on lung
development from lOto 18 years of age. NEngl J Med 351: 1057-1067.
http://dx.doi.org/10.1056/NEJMoa040610
Gauderman. WJ: Avol. E; Lurmann. F; Kuenzli. N; Gilliland. F; Peters. J: McConnell. R. (2005).
Childhood asthma and exposure to traffic and nitrogen dioxide. Epidemiology 16: 737-743.
http://dx.doi.org/10.1097/01.ede.0000181308.51440.75
Gauderman. WJ: Vora. H; McConnell. R: Berhane. K; Gilliland. F; Thomas. D: Lurmann. F; Avol. E;
Kunzli. N; Jerrett. M; Peters. J. (2007). Effect of exposure to traffic on lung development from 10 to 18
years of age: A cohort study. Lancet 369: 571-577. http://dx.doi.org/10.1016/S0140-6736(07)60037-3
Geer. LA; Weedon. J: Bell. ML. (2012). Ambient air pollution and term birth weight in Texas from 1998 to
2004. J Air Waste Manag Assoc 62: 1285-1295. http://dx.doi.org/10.1080/10962247.2012.707632
Gehring. U; Cyrys. J: Sedlmeir. G: Brunekreef. B; Bellander. T; Fischer. P; Bauer. CP; Reinhardt. D:
Wichmann. HE; Heinrich. J. (2002). Traffic-related air pollution and respiratory health during the first 2
yrs of life. Eur Respir J 19: 690-698. http://dx.doi.org/10.1183/09031936.02.01182001
November 2013 5-163 DRAFT: Do Not Cite or Quote
-------
Gehring. U; Heinrich. J; Kramer. U; Grote. V: Hochadel. M; Sugiri. D; Kraft. M; Rauchfuss. K; Eberwein.
HG: Wichmann. HE. (2006). Long-term exposure to ambient air pollution and cardiopulmonary
mortality in women. Epidemiology 17: 545-551.
http://dx.doi.org/10.1097/01.ede.0000224541.38258.87
Gehring. U: van Eijsden. M; Dijkema. MB; van der Wai. MF; Fischer. P; Brunekreef. B. (2011 a). Traffic-
related air pollution and pregnancy outcomes in the Dutch ABCD birth cohort study. Occup Environ
Med 68: 36-43. http://dx.doi.org/10.1136/oem.2009.053132
Gehring. U: Wijga. AH; Brauer. M; Fischer. P; de Jongste. JC: Kerkhof. M; Oldenwening. M; Smit. HA;
Brunekreef. B. (2010). Traffic-related air pollution and the development of asthma and allergies during
the first 8 years of life. Am J Respir Crit Care Med 181: 596-603.
http://dx.doi.org/10.1164/rccm.200906-0858OC
Gehring. U: Wijga. AH; Fischer. P; de Jongste. JC: Kerkhof. M; Koppelman. GH; Smit. HA; Brunekreef.
B^ (20lib). Traffic-related air pollution, preterm birth and term birth weight in the PIAMA birth cohort
study. Environ Res 111: 125-135. http://dx.doi.Org/10.1016/i.envres.2010.10.004
Ghosh. JKC: Wilhelm. M; Su. J: Goldberg. D; Cockburn. M; Jerrett. M; Ritz. B. (2012a). Assessing the
influence of traffic-related air pollution on risk of term low birth weight on the basis of land-use-based
regression models and measures of air toxics. Am J Epidemiol 175: 1262-1274.
http://dx.doi.org/10.1093/aie/kwr469
Ghosh. R; Joad. J: Benes. I; Postal. M; Sram. RJ; Hertz-Picciotto. I. (2012b). Ambient nitrogen oxides
exposure and early childhood respiratory illnesses. Environ Int 39: 96-102.
http://dx.doi.0rg/10.1016/i.envint.2011.10.001
Gotschi. T; Heinrich. J: Sunyer. J: Kunzli. N. (2008). Long-term effects of ambient air pollution on lung
function: A review [Review]. Epidemiology 19: 690-701.
http://dx.doi.org/10.1097/EDE.Ob013e318181650f
Gotschi. T; Sunyer. J: Chinn. S: de Marco. R; Forsberg. B; Gauderman. JW: Garcia-Esteban. R; Heinrich.
J; Jacquemin. B; Jarvis. D: Ponzio. M; Villani. S: Kunzli. N. (2008). Air pollution and lung function in
the European Community Respiratory Health Survey. Int J Epidemiol 37: 1349-1358.
http://dx.doi.org/10.1093/ije/dvnl36
Gouveia. N; Bremner. SA; Novaes. HMD. (2004). Association between ambient air pollution and birth
weight in Sao Paulo, Brazil. J Epidemiol Community Health 58: 11-17.
Powers. AM; Cullinan. P; Ayres. JG: Anderson. HR; Strachan. DP; Holgate. ST; Mills. 1C: Maynard. RL.
(2012). Does outdoor air pollution induce new cases of asthma? Biological plausibility and evidence; a
review. Respirology 17: 887-898. http://dx.doi.Org/10.llll/i.1440-1843.2012.02195.x
Greene. ND; Schneider. SL. (1978). Effects of NO2 on the response of baboon alveolar macrophages to
migration inhibitory factor. J Toxicol Environ Health 4: 869-880.
http://dx.doi.org/10.1080/15287397809529708
Gregory. RE; Pickrell. JA; Halm. FF; Hobbs. CH. (1983). Pulmonary effects of intermittent subacute
exposure to low-level nitrogen dioxide. J Toxicol Environ Health 11: 405-414.
http://dx.doi.org/10.1080/15287398309530354
Gruzieva. O: Bellander. T; Eneroth. K; Kull. I; Melen. E; Nordling. E; van Hage. M; Wickman. M;
Moskalenko. V; Hulchiy. O; Pershagen. G. (2012). Traffic-related air pollution and development of
allergic sensitization in children during the first 8 years of life. J Allergy Clin Immunol 129: 240-246.
http://dx.doi.0rg/10.1016/i.jaci.2011.ll.001
Gruzieva. O: Bergstrom. A; Hulchiy. O: Kull. I; Lind. T; Melen. E; Moskalenko. V; Pershagen. G:
Bellander. T. (2013). Exposure to air pollution from traffic and childhood asthma until 12 years of age.
Epidemiology 24: 54-61. http://dx.doi.org/10.1097/EDE.Ob013e318276clea
November 2013 5-164 DRAFT: Do Not Cite or Quote
-------
Guxens. M; Aguilera. I; Ballester. F; Estarlich. M; Fernandez-Somoano. A; Lertxundi. A; Lertxundi. N;
Mendez. MA; Tardon. A; Vrijheid. M; Sunyer. J. (2012). Prenatal exposure to residential air pollution
and infant mental development: Modulation by antioxidants and detoxification factors. Environ Health
Perspect 120: 144-149. http://dx.doi.org/10.1289/ehp.1103469
Hack. M; Fanaroff. AA. (1999). Outcomes of children of extremely low birth weight and gestational age in
the 1990s [Review]. Early Hum Dev 53: 193-218. http://dx.doi.org/10.1016/S0378-3782(98)00052-8
Hampel. R; Lepeule. J: Schneider. A; Bottagisi. S: Charles. MA; Ducimetiere. P; Peters. A; Slama. R.
(2011). Short-term impact of ambient air pollution and air temperature on blood pressure among
pregnant women. Epidemiology 22: 671-679. http://dx.doi.org/10.1097/EDE.Ob013e318226e8d6
Hansel. NN; Brevsse. PN; McCormack. MC: Matsui. EC: Curtin-Brosnan. J: Williams. DAL; Moore. JL;
Cuhran. JL; Diette. GB. (2008). A longitudinal study of indoor nitrogen dioxide levels and respiratory
symptoms in inner-city children with asthma. Environ Health Perspect 116: 1428-1432.
http://dx.doi.org/10.1289/ehp.11349
Hansen. C: Neller. A; Williams. G: Simpson. R. (2006). Maternal exposure to low levels of ambient air
pollution and preterm birth in Brisbane, Australia. BJOG 113: 935-941.
http://dx.doi.0rg/10.llll/i.1471-0528.2006.01010.x
Hansen. C: Neller. A; Williams. G: Simpson. R. (2007). Low levels of ambient air pollution during
pregnancy and fetal growth among term neonates in Brisbane, Australia. Environ Res 103: 383-389.
http://dx.doi.0rg/10.1016/i.envres.2006.06.010
Hansen. CA; Barnett. AG: Pritchard. G. (2008). The effect of ambient air pollution during early pregnancy
on fetal ultrasonic measurements during mid-pregnancy. Environ Health Perspect 116: 362-369.
http://dx.doi.org/10.1289/ehp.10720
Hart. JE; Garshick. E; Dockery. DW: Smith. TJ; Ryan. L; Laden. F. (2010). Long-term ambient multi-
pollutant exposures and mortality. Am J Respir Crit Care Med 183: 73-78.
http://dx.doi.org/10.1164/rccm.200912-1903OC
Hart. JE; Garshick. E; Dockery. DW: Smith. TJ: Ryan. L; Laden. F. (2011). Long-term ambient
multipollutant exposures and mortality. Am J Respir Crit Care Med 183: 73-78.
http://dx.doi.org/10.1164/rccm.200912-1903OC
Havashi. Y; Kohno. T; Ohwada. H. (1987). Morphological effects of nitrogen dioxide on the rat lung.
Environ Health Perspect 73: 135-145.
HEI (Health Effects Institute). (2010). Traffic-related air pollution: A critical review of the literature on
emissions, exposure, and health effects [HEI]. (Special Report 17). Boston, MA.
http://pubs.healtheffects.org/view.php?id=334
Heinrich. J: Thiering. E; Rzehak. P; Kramer. U; Hochadel. M; Rauchfuss. KM; Gehring. U; Wichmann.
HE. (2013). Long-term exposure to NO2 and PM10 and all-cause and cause-specific mortality in a
prospective cohort of women. Occup Environ Med 70: 179-186. http://dx.doi.org/10.1136/oemed-2012-
100876
Henry. MC: Findlav. J: Spangler. J: Ehrlich. R. (1970). Chronic toxicity of NO2 in squirrel monkeys: III.
Effect on resistance to bacterial and viral infection. Arch Environ Health 20: 566-570.
http://dx.doi.org/10.1080/00039896.1970.10665663
Hoek. G: Brunekreef. B; Goldbohm. S: Fischer. P; Van den Brandt. PA. (2002). Association between
mortality and indicators of traffic-related air pollution in the Netherlands: A cohort study. Lancet 360:
1203-1209. http://dx.doi.org/10.1016/s0140-6736(02)11280-3
Hutcheon. JA; Platt. RW. (2008). The missing data problem in birth weight percentiles and thresholds for
"small-for-gestational-age". Am J Epidemiol 167: 786-792. http://dx.doi.org/10.1093/aje/kwm327
Hwang. BF; Jaakkola. JJK; Lee. YL; Lin. YC: Y-LL. G. (2006). Relation between air pollution and allergic
rhinitis in Taiwanese schoolchildren. Respir Res 7: 23. http://dx.doi.org/10.1186/1465-9921-7-23
November 2013 5-165 DRAFT: Do Not Cite or Quote
-------
Hwang. BF; Lee. YL. (2010). Air pollution and prevalence of bronchitic symptoms among children in
Taiwan. Chest 138: 956-964. http://dx.doi.org/10.1378/chest.09-2600
Hwang. BF; Lee. YL; Jaakkola. JJ. (2011). Air pollution and stillbirth: A population-based case-control
study in Taiwan. Environ Health Perspect 119: 1345-1349. http://dx.doi.org/10.1289/ehp.1003056
Hwang. BF: Lee. YL; Lin. YC: Jaakkola. JJK; Quo. YL. (2005). Traffic related air pollution as a
determinant of asthma among Taiwanese school children. Thorax 60: 467-473.
http://dx.doi.org/10.1136/thx.2004.033977
Ichinose. T; Fujii. K; Sagai. M. (1991). Experimental studies on tumor promotion by nitrogen dioxide.
Toxicology 67: 211-225. http://dx.doi.org/10.1016/0300-483X(91)90144-P
Ichinose. T; Sagai. M. (1992). Combined exposure to NO2, O3 and H2SO4-aerosol and lung tumor
formation in rats. Toxicology 74: 173-184. http://dx.doi.org/10.1016/0300-483X(92)90137-4
Ichinose. T; Sagai. M; Kubota. K. (1983). Changes of lipid peroxidation and antioxidative protective
systems in lungs or rats exposed acutely, subacutely and chronically to nitrogen dioxide. J Jpn Soc Air
Pollut 18: 132-146.
Ifiiguez. C: Ballester. F; Estarlich. M; Esplugues. A; Murcia. M; Llop. S: Plana. A; Amoros. R; Rebagliato.
M. (2012). Prenatal exposure to traffic-related air pollution and fetal growth in a cohort of pregnant
women. Occup Environ Med 69: 736-744. http://dx.doi.org/10.1136/oemed-2011-100550
Islam. T; Gauderman. WJ: Berhane. K; Mcconnell. R; Avol. E; Peters. JM; Gilliland. FD. (2007). The
relationship between air pollution, lung function and asthma in adolescents. Thorax 62: 957-963.
http://dx.doi.org/10.1136/thx.2007.078964
Jacquemin. B; Sunyer. J; Forsberg. B; Aguilera. I; Bouso. L; Briggs. D; de Marco. R: Garcia-Esteban. R;
Heinrich. J; Jarvis. D; Maldonado. JA; Pavo. F; Rage. E; Vienneau. D; Kiinzli. N. (2009a). Association
between modelled traffic-related air pollution and asthma score in the ECRHS. Eur Respir J 34: 834-
842. http://dx.doi.org/10.1183/09031936.00138208
Jacquemin. B; Sunyer. J: Forsberg. B; Aguilera. I; Briggs. D: Garcia-Esteban. R; Gotschi. T; Heinrich. J:
Jarvholm. B; Jarvis. D: Vienneau. D: Kiinzli. N. (2009b). Home outdoor NO2 and new onset of self-
reported asthma in adults. Epidemiology 20: 119-126.
http://dx.doi.org/10.1097/EDE.Ob013e3181886e76
Jalaludin. B; Mamies. T; Morgan. G: Lincoln. D; Sheppeard. V: Corbett. S. (2007). Impact of ambient air
pollution on gestational age is modified by season in Sydney, Australia. Environ Health 6: 16.
http://dx.doi.org/10.1186/1476-069X-6-16
Jerrett M; Finkelstein. MM; Brook. JR; Arain. MA; Kanaroglou. P; Stieb. DM; Gilbert. NL; Verma. D:
Finkelstein. N; Chapman. KR; Sears. MR. (2009). A cohort study of traffic-related air pollution and
mortality in Toronto, Ontario, Canada. Environ Health Perspect 117: 772-777.
http://dx.doi.org/10.1289/ehp.11533
Jerrett. M; Shankardass. K; Berhane. K; Gauderman. WJ: Kiinzli. N; Avol. E; Gilliland. F; Lurmann. F;
Molitor. JN; Molitor. JT; Thomas. DC: Peters. J: McConnell. R. (2008). Traffic-related air pollution and
asthma onset in children: A prospective cohort study with individual exposure measurement. Environ
Health Perspect 116: 1433-1438. http://dx.doi.org/10.1289/ehp.10968
Jiang. LL; Zhang. YH; Song. GX; Chen. GH; Chen. BH; Zhao. NQ; Kan. HP. (2007). A time series
analysis of outdoor air pollution and preterm birth in Shanghai, China. Biomed Environ Sci 20: 426-
431.
Kannan. S: Misra. DP; Dvonch. JT; Krishnakumar. A. (2006). Exposures to airborne paniculate matter and
adverse perinatal outcomes: A biologically plausible mechanistic framework for exploring potential
effect modification by nutrition. Environ Health Perspect 114: 1636-1642.
Kashima. S: Naruse. H; Yorifuji. T; Ohki. S: Murakoshi. T; Takao. S: Tsuda. T; Doi. H. (2011). Residential
proximity to heavy traffic and birth weight in Shizuoka, Japan. Environ Res 111: 377-387.
http://dx.doi.0rg/10.1016/i.envres.2011.02.005
November 2013 5-166 DRAFT: Do Not Cite or Quote
-------
Katanoda. K; Sobue. T; Satoh. H; Taiima. K; Suzuki. T; Nakatsuka. H; Takezaki. T; Nakavama. T; Nitta. H;
Tanabe. K; Tominaga. S. (2011). An association between long-term exposure to ambient air pollution
and mortality from lung cancer and respiratory diseases in Japan. J Epidemiol 21: 132-143.
http://dx.doi.org/10.2188/iea.JE20100098
Kim. JJ; Huen. K; Adams. S: Smorodinsky. S: Hoats. A; Malig. B; Lipsett. M; Ostro. B. (2008). Residential
traffic and children's respiratory health. Environ Health Perspect 116: 1274-1279.
Kitabatake. M; Yamamoto. H; Yuan. PF; Manjurul. H; Murase. S: Yamauchi. T. (1995). Effects of exposure
to NO2 or SO2 on bronchopulmonary reaction induced by Candida albicans in guinea pigs. J Toxicol
Environ Health 45: 75-82. http://dx.doi.org/10.1080/15287399509531981
Kobavashi. T; Miura. T. (1995). Concentration- and time-dependent increase in specific airway resistance
after induction of airway hyperresponsiveness by subchronic exposure of guinea pigs to nitrogen
dioxide. Fundam Appl Toxicol 25: 154-158. http://dx.doi.org/10.1006/faat. 1995.1049
Koehler. C: Ginzkev. C: Friehs. G: Hackenberg. S: Froelich. K; Scherzed. A; Burghartz. M; Kessler. M;
Kleinsasser. N. (2010). Aspects of nitrogen dioxide toxicity in environmental urban concentrations in
human nasal epithelium. Toxicol Appl Pharmacol 245: 219-225.
http://dx.doi.0rg/10.1016/i.taap.2010.03.003
Koehler. C: Ginzkev. C: Friehs. G: Hackenberg. S: Froelich. K; Scherzed. A; Burghartz. M; Kessler. M;
Kleinsasser. N. (2011). Ex vivo toxicity of nitrogen dioxide in human nasal epithelium at the WHO
defined 1-h limit value. Toxicol Lett 207: 89-95. http://dx.doi.Org/10.1016/i.toxlet.2011.08.004
Koehler. C: Thielen. S: Ginzkev. C: Hackenberg. S: Scherzed. A; Burghartz. M; Paulus. M; Hagen. R;
Kleinsasser. NH. (2013). Nitrogen dioxide is genotoxic in urban concentrations. Inhal Toxicol 25: 341-
347. http://dx.doi.org/10.3109/08958378.2013.788104
Krewski. D; Burnett. RT; Goldberg. MS: Hoover. K; Siemiatvcki. J: Jerrett. M; Abrahamowicz. M; White.
WH. (2000). Reanalysis of the Harvard Six Cities study and the American Cancer Society study of
paniculate air pollution and mortality. Cambridge, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id=6
Krewski. D; Jerrett. M; Burnett. RT; Ma. R; Hughes. E; Shi. Y; Turner. MC: Pope AC. Ill; Thurston. G:
Calle. EE; Thun. MJ; Beckerman. B; DeLuca. P; Finkelstein. N; Ito. K; Moore. DK; Newbold. KB;
Ramsay. T; Ross. Z; Shin. H; Tempalski. B. (2009). Extended follow-up and spatial analysis of the
American Cancer Society study linking particulate air pollution and mortality. Res Rep Health Eff Inst
0: 5-114; discussion 115-136.
Kripke. BJ; Sherwin. RP. (1984). Nitrogen dioxide exposure—influence on rat testes. Anesth Analg 63: 526-
528.
Kubota. K; Murakami. M; Takenaka. S: Kawai. K; Kyono. H. (1987). Effects of long-term nitrogen dioxide
exposure on rat lung: Morphological observations. Environ Health Perspect 73: 157-169.
Kumae. T; Arakawa. H. (2006). Comparison of effects of in vivo nitrogen dioxide exposure starting from
different periods on alveolar macrophage activity, assessed by a chemiluminescence technique in
Brown-Norway rats. Luminescence 21: 226-232. http://dx.doi.org/10.1002/bio.911
Lafuma. C: Harf. A; Lange. F; Bozzi. L; Poncv. JL; Bignon. J. (1987). Effect of low-level NO2 chronic
exposure on elastase-induced emphysema. Environ Res 43: 75-84. http://dx.doi.org/10.1016/S0013-
9351(87)80059-2
Lampl. M; Jeantv. P. (2003). Timing is everything: A reconsideration of fetal growth velocity patterns
identifies the importance of individual and sex differences. Am J Hum Biol 15: 667-680.
http://dx.doi.org/10.1002/aihb. 10204
Lee. BE; Ha. EH; Park. HS: Kim. YJ; Hong. YC: Kim. H; Lee. JT. (2003). Exposure to air pollution during
different gestational phases contributes to risks of low birth weight. Hum Reprod 18: 638-643.
http://dx.doi.org/10.1093/humrep/degl02
November 2013 5-167 DRAFT: Do Not Cite or Quote
-------
Lee. PC: Talbott. EO; Roberts. JM; Catov. JM; Bilonick. RA; Stone. RA; Sharma. RK: Ritz. B. (2012b).
Ambient air pollution exposure and blood pressure changes during pregnancy. Environ Res 117: 46-53.
http://dx.doi.0rg/10.1016/i.envres.2012.05.011
Lee. PC: Talbott. EO: Roberts. JM: Catov. JM: Sharma. RK: Ritz. B. (20lie). Particulate air pollution
exposure and C-reactive protein during early pregnancy. Epidemiology 22: 524-531.
http://dx.doi.org/10.1097/EDE.Ob013e31821c6c58
Lee. YL; Hwang. BF; Chen. YA; Chen. JM: Wu. YF. (2012c). Pulmonary function and incident bronchitis
and asthma in children: a community-based prospective cohort study. PLoS ONE 7: e32477.
http://dx.doi.org/10.1371/iournal.pone.0032477
Lee. YL; Mcconnell. R; Berhane. K; Gilliland. FD. (2009). Ambient ozone modifies the effect of tumor
necrosis factor G-308A on bronchitic symptoms among children with asthma. Allergy 64: 1342-1348.
http://dx.doi.0rg/10.llll/i.1398-9995.2009.02014.x
Lee. YL: Wang. WH; Lu. CW: Lin. YH; Hwang. BF. (20lid). Effects of ambient air pollution on
pulmonary function among schoolchildren. Int J Hyg Environ Health 214: 369-375.
http://dx.doi.0rg/10.1016/i.iiheh.2011.05.004
Leem. JH; Kaplan. BM; Shim. YK; Pohl. HR; Gotwav. CA; Bullard. SM; Rogers. JF; Smith. MM; Tylenda.
CA. (2006). Exposures to air pollutants during pregnancy and preterm delivery. Environ Health Perspect
114: 905-910. http://dx.doi.org/10.1289/ehp.8733
Legro. RS: Sauer. MV; Mottla. GL; Richter. KS: Li. X; Dodson. WC: Liao. D. (2010). Effect of air quality
on assisted human reproduction. Hum Reprod 25: 1317-1324. http://dx.doi.org/10.1093/humrep/deq021
Lenters. V: Uiterwaal. CS: Beelen. R; Bots. ML; Fischer. P; Brunekreef. B; Hoek. G. (2010). Long-term
exposure to air pollution and vascular damage in young adults. Epidemiology 21: 512-520.
http://dx.doi.org/10.1097/EDE.Ob013e3181dec3a7
Lepeule. J: Cai'ni. F; Bottagisi. S: Galineau. J: Hulin. A; Marquis. N; Bohet A; Siroux. V; Kaminski. M;
Charles. MA; Slama. R. (2010). Maternal exposure to nitrogen dioxide during pregnancy and offspring
birth weight: Comparison of two exposure models. Environ Health Perspect 118: 1483-1489.
http://dx.doi.org/10.1289/ehp.0901509
Li. H; Chen. L; Guo. Z; Sang. N; Li. G. (2012a). In vivo screening to determine neurological hazards of
nitrogen dioxide (NO2) using Wistar rats. J Hazard Mater 225-226: 46-53.
http://dx.doi.0rg/10.1016/i.jhazmat.2012.04.063
Li. H; Han. M; Guo. L; Li. G: Sang. N. (2011 a). Oxidative stress, endothelial dysfunction and
inflammatory response in rat heart to NO inhalation exposure. Chemosphere 82: 1589-1596.
http://dx.doi.0rg/10.1016/i.chemosphere.2010.ll.055
Lin. CA; Pereira. LAA; Nishioka. DC: Conceicao. GMS: Graga. ALF; Saldiva. PHN. (2004a). Air
pollution and neonatal deaths in Sao Paulo, Brazil. Braz J Med Biol Res 37: 765-770.
http://dx.doi.org/10.1590/S0100-879X2004000500019
Linn. WS: Rappaport. EB; Berhane. KT; Bastain. TM; Avol. EL; Gilliland. FD. (2009). Exhaled nitric
oxide in a population-based study of southern California schoolchildren. Respir Res 10: 28.
http://dx.doi.org/10.1186/1465-9921-10-28
Lipfert. F; Wyzga. R: Batv. J: Miller. J. (2009). Air pollution and survival within the Washington
University-EPRI veterans cohort: Risks based on modeled estimates of ambient levels of hazardous and
criteria air pollutants. J Air Waste Manag Assoc 59: 473-489.
Lipfert. FW: Batv. JD; Miller. JP; Wyzga. RE. (2006a). PM2.5 constituents and related air quality variables
as predictors of survival in a cohort of U.S. military veterans. Inhal Toxicol 18: 645-657.
http://dx.doi.org/10.1080/08958370600742946
Lipfert. FW: Perry. HM. Jr; Miller. JP; Batv. JD: Wyzga. RE; Carmodv. SE. (2000). The Washington
University-EPRI veterans' cohort mortality study: Preliminary results. Inhal Toxicol 12: 41-73.
http://dx.doi.org/10.1080/713856640
November 2013 5-168 DRAFT: Do Not Cite or Quote
-------
Lipfert FW: Perry. HM. Jr; Miller. JP; Batv. JD; Wyzga. RE; Carmodv. SE. (2003). Air pollution, blood
pressure, and their long-term associations with mortality. Inhal Toxicol 15: 493-512.
http://dx.doi.org/10.1080/08958370304463
Lipfert. FW: Wyzga. RE; Batv. JD: Miller. JP. (2006b). Traffic density as a surrogate measure of
environmental exposures in studies of air pollution health effects: Long-term mortality in a cohort of US
veterans. Atmos Environ 40: 154-169. http://dx.doi.Org/10.1016/i.atmosenv.2005.09.027
Lipsett. MJ; Ostro. BD; Reynolds. P; Goldberg. D; Hertz. A; Jerrett. M; Smith. DF; Garcia. C: Chang. ET;
Bernstein. L. (2011). Long-term exposure to air pollution and cardiorespiratory disease in the California
teachers study cohort. Am J Respir Crit Care Med 184: 828-835.
http://dx.doi.org/10.1164/rccm.201012-2082OC
Liu. CC: Tsai. SS: Chiu. HF; Wu. TN; Chen. CC: Yang. CY. (2009a). Ambient exposure to criteria air
pollutants and risk of death from bladder cancer in Taiwan. Inhal Toxicol 21: 48-54.
http://dx.doi.org/10.1080/08958370802207326
Liu. CC: Tsai. SS: Chiu. HF; Wu. TN; Yang. CY. (2008). Ambient exposure to criteria air pollutants and
female lung cancer in Taiwan. Inhal Toxicol 20: 311-317.
http://dx.doi.org/10.1080/08958370701866107
Liu. S: Krewski. D; Shi. Y; Chen. Y; Burnett. R. (2007). Association between maternal exposure to ambient
air pollutants during pregnancy and fetal growth restriction. J Expo Sci Environ Epidemiol 17: 426-432.
http://dx.doi.org/10.1038/sj.jes.7500503
Liu. S: Krewski. D; Shi. Y; Chen. Y; Burnett. RT. (2003). Association between gaseous ambient air
pollutants and adverse pregnancy outcomes in Vancouver, Canada. Environ Health Perspect 111: 1773-
1778.
Llop. S: Ballester. F; Estarlich. M; Esplugues. A; Rebagliato. M; Iniguez. C. (2010). Preterm birth and
exposure to air pollutants during pregnancy. Environ Res 110: 778-785.
http://dx.doi.0rg/10.1016/i.envres.2010.09.009
Madsen. C: Gehring. U; Walker. SE; Brunekreef. B; Stigum. H; Naess. O: Nafstad. P. (2010). Ambient air
pollution exposure, residential mobility and term birth weight in Oslo, Norway. Environ Res 110: 363-
371. http://dx.doi.0rg/10.1016/j.envres.2010.02.005
Malmqvist. E; Jakobsson. K; Tinnerberg. H; Rignell-Hydbom. A; Rylander. L. (2013). Gestational diabetes
and preeclampsia in association with air pollution at levels below current air quality guidelines. Environ
Health Perspect 121: 488-493. http://dx.doi.org/10.1289/ehp.1205736
Malmqvist. E; Rignell-Hydbom. A; Tinnerberg. H; Bjork. J: Stroh. E; Jakobsson. K; Rittner. R; Rylander.
L_. (2011). Maternal exposure to air pollution and birth outcomes. Environ Health Perspect 119: 553558.
http://dx.doi.org/10.1289/ehp.1002564
Mamies. T; Jalaludin. B; Morgan. G: Lincoln. D; Sheppeard. V: Corbett. S. (2005). Impact of ambient air
pollution on birth weight in Sydney, Australia. Occup Environ Med 62: 524-530.
http://dx.doi.org/10.1136/oem.2004.014282
Maroziene. L; Grazuleviciene. R. (2002). Maternal exposure to low-level air pollution and pregnancy
outcomes: A population-based study. Environ Health Global Access Sci Source 1: 6.
http://dx.doi.org/10.1186/1476-069X-l-6
Martinez. FD; Wright. AL; Taussig. LM; Holberg. CJ: Halonen. M; Morgan. WJ. (1995). Asthma and
wheezing in the first six years of life. N Engl J Med 332: 133-138.
http://dx.doi.org/10.1056/NEJM199501193320301
McConnell. R: Berhane. K; Gilliland. F; Molitor. J: Thomas. D: Lurmann. F; Avol. E; Gauderman. WJ:
Peters. JM. (2003). Prospective study of air pollution and bronchitic symptoms in children with asthma.
Am J Respir Crit Care Med 168: 790-797. http://dx.doi.org/10.1164/rccm.200304-466OC
November 2013 5-169 DRAFT: Do Not Cite or Quote
-------
McConnell. R: Berhane. K: Molitor. J: Gilliland. F: Kunzli. N: Thome. PS: Thomas. D: Gauderman. WJ: Avol.
E: Lurmann. F: Rappaport. E: Jerrett. M: Peters. JM. (2006). Dog ownership enhances symptomatic
responses to air pollution in children with asthma. Environ Health Perspect 114: 1910-1915.
http://dx.doi.org/10.1289/ehp.8548
McConnell R: Islam. T: Shankardass. K: Jerrett M: Lurmann. F: Gilliland. F: Gauderman. J: Avol E:
Kuenzli. N: Yao. L: Peters. J: Berhane. K. (2010). Childhood incident asthma and traffic-related air
pollution at home and school. Environ Health Perspect 118: 1021-1026.
http://dx.doi.org/10.1289/ehp.0901232
McCormack. MC: Brevsse. PN; Hansel. NN; Matsui. EC: Tonorezos. ES: Curtin-Brosnan. J: Williams. PL;
Buckley. TJ; Eggleston. PA; Diette. GB. (2008). Common household activities are associated with
elevated paniculate matter concentrations in bedrooms of inner-city Baltimore pre-school children.
Environ Res 106: 148-155. http://dx.doi.Org/10.1016/j.envres.2007.08.012
Melen. E; Nyberg. F; Lindgren. CM; Berglind. N; Zucchelli. M; Nordling. E; Hallberg. J: Svartengren. M;
Morgenstern. R; Kere. J: Bellander. T; Wickman. M; Pershagen. G. (2008). Interactions between
glutathione S-transferase PI, tumor necrosis factor, and traffic-related air pollution for development of
childhood allergic disease. Environ Health Perspect 116: 1077-1084.
http://dx.doi.org/10.1289/ehp.11117
Meng. QY; Svendsgaard. D; Kotchmar. DJ; Pinto. JP (2012b). Associations between personal exposures
and ambient concentrations of nitrogen dioxide: A quantitative research synthesis. Atmos Environ 57:
322-329. http://dx.doi.0rg/10.1016/i.atmosenv.2012.04.035
Meng. YY; Rull. RP; Wilhelm. M; Lombardi. C: Balmes. J: Ritz. B. (2010). Outdoor air pollution and
uncontrolled asthma in the San Joaquin Valley, California. J Epidemiol Community Health 64: 142-147.
http://dx.doi.org/10.1136/iech.2008.083576
Mercer. RR; Costa. PL; Crapo. JD. (1995). Effects of prolonged exposure to low doses of nitric oxide or
nitrogen dioxide on the alveolar septa of the adult rat lung. Lab Invest 73: 20-28.
Mi. YH; Norback. D; Tao. J: Mi. YL; Perm. M. (2006). Current asthma and respiratory symptoms among
pupils in Shanghai, China: Influence of building ventilation, nitrogen dioxide, ozone, and formaldehyde
in classrooms. Indoor Air 16: 454-464. http://dx.doi.Org/10.llll/i.1600-0668.2006.00439.x
Miller. FJ; Graham. JA; Raub. JA; Illing. JW: Menache. MG: House. DE; Gardner. DE. (1987). Evaluating
the toxicity of urban patterns of oxidant gases. II. Effects in mice from chronic exposure to nitrogen
dioxide. J Toxicol Environ Health 21: 99-112. http://dx.doi.org/10.1080/15287398709531005
Miller. KA; Siscovick. PS: Sheppard. L; Shepherd. K; Sullivan. JH; Anderson. GL; Kaufman. JD. (2007).
Long-term exposure to air pollution and incidence of cardiovascular events in women. N Engl J Med
356: 447-458.
Mobasher. Z; Salam. MT; Goodwin. TM; Lurmann. F; Ingles. SA; Wilson. ML. (2013). Associations
between ambient air pollution and Hypertensive Disorders of Pregnancy. Environ Res 123: 9-16.
http://dx.doi.0rg/10.1016/i.envres.2013.01.006
Modig. L; Toren. K; Janson. C: Jarvholm. B; Forsberg. B. (2009). Vehicle exhaust outside the home and
onset of asthma among adults. Eur Respir J 33: 1261-1267.
http://dx.doi.org/10.1183/09031936.00101108
Molter. A; Agius. RM; de Vocht. F; Lindlev. S: Gerrard. W: Lowe. L; Belgrave. D: Custovic. A; Simpson.
A. (2013). Long-term exposure to PM10 and NO2 in association with lung volume and airway
resistance in the MAAS birth cohort. Environ Health Perspect 121: 1232-1238.
http://dx.doi.org/10.1289/ehp.1205961
Molter. A; Lindlev. S: de Vocht. F; Agius. R: Kerry. G: Johnson. K; Ashmore. M; Terry. A;
Dimitroulopoulou. S: Simpson. A. (2012). Performance of a microenviromental model for estimating
personal NO2 exposure in children. Atmos Environ 51: 225-233.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.01.030
November 2013 5-170 DRAFT: Do Not Cite or Quote
-------
Molter. A; Lindlev. S: de Vocht. F; Simpson. A; Agius. R. (2010a). Modelling air pollution for
epidemiologic research-Part I: A novel approach combining land use regression and air dispersion. Sci
Total Environ 408: 5862-5869. http://dx.doi.Org/10.1016/i.scitotenv.2010.08.027
Molter. A; Lindlev. S: de Vocht. F; Simpson. A; Agius. R. (2010b). Modelling air pollution for
epidemiologic research-Part II: Predicting temporal variation through land use regression. Sci Total
Environ 409: 211-217. http://dx.doi.Org/10.1016/i.scitotenv.2010.10.005
Morales. E; Julvez. J; Torrent. M; de Cid. R: Guxens. M; Bustamante. M; Kiinzli. N; Sunyer. J. (2009).
Association of early-life exposure to household gas appliances and indoor nitrogen dioxide with
cognition and attention behavior in preschoolers. Am J Epidemiol 169: 1327-1336.
http://dx.doi.org/10.1093/aie/kwp067
Morello-Frosch. R; Jesdale. BM; Sadd. JL; Pastor. M. (2010). Ambient air pollution exposure and full-term
birth weight in California. Environ Health 9: 44. http://dx.doi.org/10.1186/1476-069X-9-44
Morgenstern. V; Zutavern. A; Cyrys. J: Brockow. I; Gehring. U; Koletzko. S: Bauer. CP; Reinhardt. D;
Wichmann. HE; Heinrich. J. (2007). Respiratory health and individual estimated exposure to traffic-
related air pollutants in a cohort of young children. Occup Environ Med 64: 8-16.
http://dx.doi.org/10.1136/oem.2006.028241
Morgenstern. V; Zutavern. A; Cyrvs. J: Brockow. I; Koletzko. S: Kramer. U; Behrendt. H; Herbarth. O: von
Berg. A; Bauer. CP: Wichmann. HE; Heinrich. J. (2008). Atopic diseases, allergic sensitization, and
exposure to traffic-related air pollution in children. Am J Respir Crit Care Med 177: 1331-1337.
http://dx.doi.org/10.1164/rccm.200701-036OC
Moseler. M; Hendel-Kramer. A; Karmaus. W: Forster. J: Weiss. K; Urbanek. R; Kuehr. J. (1994). Effect of
moderate NO2 air pollution on the lung function of children with asthmatic symptoms. Environ Res 67:
109-124. http://dx.doi.org/10.1006/enrs.1994.1068
Naess. O: Nafstad. P; Aamodt. G: Claussen. B; Rosland. P. (2007). Relation between concentration of air
pollution and cause-specific mortality: four-year exposures to nitrogen dioxide and particulate matter
pollutants in 470 neighborhoods in Oslo, Norway. Am J Epidemiol 165: 435-443.
http://dx.doi.org/10.1093/aie/kwk016
Nafstad. P; Haheim. LL; Oftedal. B; Gram. F; Holme. I; Hjermann. I; Leren. P. (2003). Lung cancer and air
pollution: A 27 year follow up of 16 209 Norwegian men. Thorax 58: 1071-1076.
http://dx.doi.org/10.1136/thorax.58.12.1071
Nafstad. P: Haheim. LL; Wisloff. T; Gram. F; Oftedal. B; Holme. I; Hjermann. I; Leren. P. (2004). Urban
air pollution and mortality in a cohort of Norwegian men. Environ Health Perspect 112: 610-605.
Neupane. B; Jerrett. M; Burnett. RT; Marrie. T; Arain. A; Loeb. M. (2010). Long-term exposure to ambient
air pollution and risk of hospitalization with community-acquired pneumonia in older adults. Am J
Respir Crit Care Med 181: 47-53. http://dx.doi.org/10.1164/rccm.200901-0160OC
NHLBI (National Institutes of Health, National Heart Lung and Blood Institute). (2002). Global initiative
for asthma: Global strategy for asthma management and prevention. U.S. Department of Health and
Human Services. http://www.ginasthma.org/Guidelines/guidelines-2002-original%3a-workshop-
report%2c-global-strategv-for-asthm am anagement-and-prevention.html
Nicolai. T; Carr. D; Weiland. SK; Duhme. H; Von Ehrenstein. O: Wagner. C: Von Mutius. E. (2003). Urban
traffic and pollutant exposure related to respiratory outcomes and atopy in a large sample of children.
Eur Respir J 21: 956-963. http://dx.doi.org/10.1183/09031936.03.00041103
Nishimura. KK; Galanter. JM; Roth. LA; Oh. SS: Thakur. N; Nguyen. EA; Thyne. S: Farber. HJ;
Serebrisky. D; Kumar. R; Brigino-Buenaventura. E; Davis. A; LeNoir. MA; Meade. K; Rodriguez-
Cintron. W: Avila. PC: Borrell. LN; Bibbins-Domingo. K; Rodriguez-Santana. JR; Sen. S: Lurmann. F;
Balmes. JR; Burchard. EG. (2013a). Early-life air pollution and asthma risk in minority children: The
GALA II and SAGE II studies. Am J Respir Crit Care Med 188: 309-318.
http://dx.doi.org/10.1164/rccm.201302-0264OC
November 2013 5-171 DRAFT: Do Not Cite or Quote
-------
Nishimura. KK; Galanter. JM; Roth. LA; Oh. SS: Thakur. N; Nguyen. EA; Thyne. S: Farber. HJ;
Serebrisky. D; Kumar. R; Brigino-Buenaventura. E; Davis. A; Lenoir. MA; Meade. K; Rodriguez-
Cintron. W: Avila. PC: Borrell. LN; Bibbins-Domingo. K; Rodriguez-Santana. JR: Sen. S: Lurmann. F;
Balmes. JR: Burchard. EG. (2013b). Early-life air pollution and asthma risk in minority children: The
GALA II and SAGE II studies. Online supplemental material [Supplemental Data]. Am J Respir Crit
Care Med 188.
Nordling. E; Berglind. N; Melen. E; Emenius. G: Hallberg. J; Nyberg. F: Pershagen. G: Svartengren. M;
Wickman. M; Bellander. T. (2008). Traffic-related air pollution and childhood respiratory symptoms,
function and allergies. Epidemiology 19: 401-408. http://dx.doi.org/10.1097/EDE.Ob013e31816alce3
NTP (National Toxicology Program). (2011). Diesel exhaust particulates. In Report on Carcinogens,
Twelfth Edition (pp. 153-155). Research Triangle Park, NC: U.S. Department of Health and Human
Services, Public Health Service, National Toxicology Program.
Nyberg. F; Gustavsson. P: Jarup. L; Bellander. T; Berglind. N; Jakobsson. R; Pershagen. G. (2000). Urban
air pollution and lung cancer in Stockholm. Epidemiology 11: 487-495.
Oftedal. B; Brunekreef. B; Nystad. W: Madsen. C: Walker. SE; Nafstad. P. (2008). Residential outdoor air
pollution and lung function in schoolchildren. Epidemiology 19: 129-137.
http://dx.doi.org/10.1097/EDE.Ob013e31815c0827
Oftedal. B; Nystad. W: Brunekreef. B; Nafstad. P. (2009). Long-term traffic-related exposures and asthma
onset in schoolchildren in Oslo, Norway. Environ Health Perspect 117: 839-844.
http://dx.doi.org/10.1289/ehp.11491
Ohashi. Y; Nakai. Y; Sugiura. Y; Ohno. Y; Okamoto. H; Tanaka. A; Kakinoki. Y; Havashi. M. (1994).
Nitrogen dioxide-induced eosinophilia and mucosal injury in the nose of the guinea pig. Acta
Otolaryngol 114: 547-551.
Ohyama. K; Ito. T; Kanisawa. M. (1999). The roles of diesel exhaust particle extracts and the promotive
effects of NO2 and/or SO2 exposure on rat lung tumorigenesis. Cancer Lett 139: 189-197.
http://dx.doi.org/10.1016/S0304-3835(99)00040-3
Olsson. D; Ekstrom. M; Forsberg. B. (2012). Temporal Variation in Air Pollution Concentrations and
Preterm Birth-A Population Based Epidemiological Study. Int J Environ Res Public Health 9: 272-285.
http://dx.doi.org/10.3390/ijerph9010272
Padula. AM; Tager. IB; Carmichael. SL; Hammond. SK; Lurmann. F: Shaw. GM. (2013a). The association
of ambient air pollution and traffic exposures with selected congenital anomalies in the San Joaquin
Valley of California. Am J Epidemiol 177: 1074-1085. http://dx.doi.org/10.1093/aie/kws367
Padula. AM; Tager. IB; Carmichael. SL; Hammond. SK; Yang. W: Lurmann. F: Shaw. GM. (2013b).
Ambient air pollution and traffic exposures and congenital heart defects in the san joaquin valley of
California. Paediatric and Perinatal Epidemiology (Online) 27: 329-339.
http://dx.doi.org/10.llll/ppe.12055
Panasevich. S: Leander. K; Rosenlund. M; Ljungman. P: Bellander. T; de Faire. U: Pershagen. G: Nyberg.
K (2009). Associations of long- and short-term air pollution exposure with markers of inflammation and
coagulation in a population sample. Occup Environ Med 66: 747-753.
http://dx.doi.org/10.1136/oem.2008.043471
Papathomas. M; Molitor. J: Richardson. S: Riboli. E; Vineis. P. (2011). Examining the joint effect of
multiple risk factors using exposure risk profiles: Lung cancer in non-smokers. Environ Health Perspect
119: 84-91. http://dx.doi.org/10.1289/ehp.1002118
Parent. ME; Goldberg. MS: Grouse. PL; Ross. NA; Chen. H; Valois. MF; Liautaud. A. (2013). Traffic-
related air pollution and prostate cancer risk: a case-control study in Montreal, Canada. Occup Environ
Med 70: 511-518. http://dx.doi.org/10.1136/oemed-2012-101211
Parker. JD; Akinbami. LJ; Woodruff. TJ. (2009). Air pollution and childhood respiratory allergies in the
United States. Environ Health Perspect 117: 140-147.
November 2013 5-172 DRAFT: Do Not Cite or Quote
-------
Parker. JD; Rich. DQ; Glinianaia. SV: Leem. JH; Wartenberg. D; Bell. ML; Bonzini. M; Brauer. M;
Darrow. L; Gehring. U; Gouveia. N; Grillo. P; Ha. E; van den Hooven. EH; Jalaludin. B; Jesdale. BM;
Lepeule. J: Morello-Frosch. R; Morgan. GG: Slama. R; Pierik. FH; Pesatori. AC: Sathyanaravana. S:
Seo. J: Strickland. M; Tamburic. L; Woodruff. TJ. (2011). The international collaboration on air
pollution and pregnancy outcomes: Initial results. Environ Health Perspect 119: 1023-1028.
http://dx.doi.org/10.1289/ehp.1002725
Pathmanathan. S: Krishna. MT; Blomberg. A; Helledav. R: Kelly. FJ; Sandstrom. T; Holgate. ST; Wilson.
SJ: Frew. AJ. (2003). Repeated daily exposure to 2 ppm nitrogen dioxide upregulates the expression of
IL-5, IL-10, IL-13, and ICAM-1 in the bronchial epithelium of healthy human airways. Occup Environ
Med 60: 892-896. http://dx.doi.org/10.1136/oem.60.ll.892
Pattenden. S: Hoek. G: Braun-Fahrlander. C: Forastiere. F; Kosheleva. A; Neuberger. M; Fletcher. T.
(2006). NO2 and children's respiratory symptoms in the PATY study. Occup Environ Med 63: 828-835.
http://dx.doi.org/10.1136/oem.2006.025213
Pereira. G: Cook. AG: Haggar. F; Bower. C: Nassar. N. (2012). Locally derived traffic-related air pollution
and fetal growth restriction: A retrospective cohort study. Occup Environ Med 69: 815-822.
http://dx.doi.org/10.1136/oemed-2011-100509
Pereira. G: Haggar. F; Shand. AW: Bower. C: Cook. A; Nassar. N. (2013). Association between pre-
eclampsia and locally derived traffic-related air pollution: A retrospective cohort study. J Epidemiol
Community Health 67: 147-152. http://dx.doi.org/10.1136/iech-2011-200805
Pilotto. LS: Nitschke. M; Smith. BJ; Pisaniello. D; Ruffin. RE; McElroy. HJ; Martin. J: Hiller. JE. (2004).
Randomized controlled trial of unflued gas heater replacement on respiratory health of asthmatic
schoolchildren. Int J Epidemiol 33: 208-214.
Pope. CA. Ill; Burnett. RT; Thun. MJ; Calle. EE; Krewski. D; Ito. K; Thurston. GD. (2002). Lung cancer,
cardiopulmonary mortality, and long-term exposure to fine paniculate air pollution. JAMA 287: 1132-
1141.
Pope. CA. Ill: Thun. MJ: Namboodiri. MM: Dockerv. DW: Evans. JS: Speizer. FE: Heath. CW. Jr. (1995).
Paniculate air pollution as a predictor of mortality in a prospective study of U.S. adults. Am J Respir
Crit Care Med 151: 669-674. http://dx.doi.Org/10.1164/airccm/151.3 Pt 1.669
Pujades-Rodriguez. M; Mckeever. T; Lewis. S: Whyatt. D: Britton. J: Venn. A. (2009). Effect of traffic
pollution on respiratory and allergic disease in adults: Cross-sectional and longitudinal analyses. BMC
Pulm Med 9: 42. http://dx.doi.org/10.1186/1471-2466-9-42
Raaschou-Nielsen. O; Andersen. ZJ; Hvidberg. M; Jensen. SS: Ketzel. M; S0rensen. M; Loft. S: Overvad.
K; Tj0nneland. A. (2011). Lung cancer incidence and long-term exposure to air pollution from traffic.
Environ Health Perspect 119: 860-865. http://dx.doi.org/10.1289/ehp.1002353
Raaschou-Nielsen. O; Bak. H; S0rensen. M; Jensen. SS: Ketzel. M; Hvidberg. M; Schnohr. P; Tj0nneland.
A; Overvad. K; Loft. S. (2010a). Air pollution from traffic and risk for lung cancer in three Danish
cohorts. Cancer Epidemiol Biomarkers Prev 19: 1284-1291. http://dx.doi.org/10.1158/1055-9965.EPI-
10-0036
Raaschou-Nielsen. O: Hermansen. MN; Loland. L; Buchvald. F; Pipper. CB; S0rensen. M; Loft. S:
Bisgaard. H. (201 Ob). Long-term exposure to indoor air pollution and wheezing symptoms in infants.
Indoor Air 20: 159-167. http://dx.doi.Org/10.llll/i.1600-0668.2009.00635.x
Rage. E; Jacquemin. B; Nadif. R: Oryszczyn. MP; Siroux. V; Aguilera. I; Kauffmann. F; Kiinzli. N. (2009).
Total serum IgE levels are associated with ambient ozone concentration in asthmatic adults. Allergy 64:
40-46. http://dx.doi.0rg/10.llll/i.1398-9995.2008.01800.x
Rahmalia. A; Giorgis-Allemand. L; Lepeule. J: Philippat. C: Galineau. J: Hulin. A; Charles. MA; Slama. R.
(2012). Pregnancy exposure to atmospheric pollutants and placenta! weight: An approach relying on a
dispersion model. Environ Int 48: 47-55. http://dx.doi.Org/10.1016/i.envint.2012.06.013
November 2013 5-173 DRAFT: Do Not Cite or Quote
-------
Renzetti. G: Silvestre. G: D'Amario. C: Bottini. E; Gloria-Bottini. F; Bottini. N; Auais. A; Perez. MK;
Piedimonte. G. (2009). Less air pollution leads to rapid reduction of airway inflammation and improved
airway function in asthmatic children. Pediatrics 123: 1051-1058. http://dx.doi.org/10.1542/peds.2008-
1153
Rich. DQ; Demissie. K; Lu. SE; Kamat. L; Wartenberg. D; Rhoads. GG. (2009). Ambient air pollutant
concentrations during pregnancy and the risk of fetal growth restriction. J Epidemiol Community Health
63: 488-496. http://dx.doi.org/10.1136/iech.2008.082792
Richters. A; Damji. KS. (1990). The relationship between inhalation of nitrogen dioxide, the immune
system, and progression of a spontaneously occurring lymphoma in AKR mice. J Environ Pathol
Toxicol Oncol 10: 225-230.
Richters. A; Kuraitis. K. (1981). Inhalation of NO2 and blood borne cancer cell spread to the lungs. Arch
Environ Health 36: 36-39.
Puchters. A; Kuraitis. K. (1983). Air pollutants and the facilitation of cancer metastasis. Environ Health
Perspect52: 165-168.
Puchters. A; Richters. V: Alley. WP. (1985). The mortality rate from lung metastases in animals inhaling
nitrogen dioxide (NO2). J Surg Oncol 28: 63-66. http://dx.doi.org/10.1002/jso.2930280115
Ritz. B; Wilhelm. M. (2008). Ambient air pollution and adverse birth outcomes: Methodologic issues in an
emerging field [Review]. Basic Clin Pharmacol Toxicol 102: 182-190. http://dx.doi.org/10.llll/U742-
7843.2007.00161.x
Ritz. B; Wilhelm. M; Hoggatt. KJ; Ghosh. JK. (2007). Ambient air pollution and preterm birth in the
environment and pregnancy outcomes study at the University of California, Los Angeles. Am J
Epidemiol 166: 1045-1052. http://dx.doi.org/l0.1093/aje/kwm 181
Ritz. B; Wilhelm. M; Zhao. Y. (2006). Air pollution and infant death in southern California, 1989-2000.
Pediatrics 118: 493-502. http://dx.doi.org/10.1542/peds.2006-0027
Ritz. B; Yu. F; Chapa. G: Fruin. S. (2000). Effect of air pollution on preterm birth among children born in
Southern California between 1989 and 1993. Epidemiology 11: 502-511.
Rojas-Martinez. R; Perez-Padilla. R; Olaiz-Fernandez. G: Mendoza-Alvarado. L; Moreno-Macias. H;
Fortoul. T; McDonnell. W: Loomis. D; Romieu. I. (2007a). Lung function growth in children with long-
term exposure to air pollutants in Mexico City. Am J Respir Crit Care Med 176: 377-384.
http://dx.doi.org/10.1164/rccm.200510-1678OC
Rojas-Martinez. R; Perez-Padilla. R: Olaiz-Fernandez. G: Mendoza-Alvarado. L; Moreno-Macias. H;
Fortoul. T; McDonnell. W: Loomis. D: Romieu. I. (2007b). Lung function growth in children with long-
term exposure to air pollutants in Mexico City. Online data supplement [Supplemental Data]. Am J
Respir Crit Care Med [176].
Romero. R; Espinoza. J: Kusanovic. JP; Gotsch. F; Hassan. S: Erez. O; Chaiworapongsa. T; Mazor. M.
(2006). The preterm parturition syndrome [Review]. BJOG 113: 17-42. http://dx.doi.org/10.llll/U471-
0528.2006.01120.x
Rosenlund. M; Bellander. T; Nordquist. T; Alfredsson. L. (2009a). Traffic-generated air pollution and
myocardial infarction. Epidemiology 20: 265-271. http://dx.doi.org/10.1097/EDE.Ob013e318190ea68
Rosenlund. M; Forastiere. F; Porta. D: De Sario. M; Badaloni. C: Perucci. CA. (2009b). Traffic-related air
pollution in relation to respiratory symptoms, allergic sensitisation and lung function in schoolchildren.
Thorax 64: 573-580. http://dx.doi.org/10.1136/thx.20Q7.094953
Rosenlund. M; Picciotto. S: Forastiere. F; Stafoggia. M; Perucci. CA. (2008b). Traffic-related air pollution
in relation to incidence and prognosis of coronary heart disease. Epidemiology 19: 121-128.
http://dx.doi.org/10.1097/EDE.Ob013e31815cl921
Rubes. J: Rybar. R; Prinosilova. P; Veznik. Z; Chvatalova. I; Solansky. I; Sram. RJ. (2010). Genetic
polymorphisms influence the susceptibility of men to sperm DNA damage associated with exposure to
air pollution. Mutat Res 683: 9-15. http://dx.doi.Org/10.1016/j.mrfmmm.2009.09.010
November 2013 5-174 DRAFT: Do Not Cite or Quote
-------
Sagai. M; Ichinose. T; Kubota. K. (1984). Studies on the biochemical effects of nitrogen dioxide. IV.
Relation between the change of lipid peroxidation and the antioxidative protective system in rat lungs
upon life span exposure to low levels of NO2. Toxicol Appl Pharmacol 73: 444-456.
Sagai. M; Ichinose. T; Oda. H; Kubota. K. (1982). Studies on biochemical effects of nitrogen dioxide II
Changes of the protective systems in rat lungs and of lipid peroxidation by acute exposure. J Toxicol
Environ Health 9: 153-164. http://dx.doi.org/10.1080/15287398209530150
Salam. MT; Millstein. J; Li. YF; Lurmann. FW: Margolis. HG: Gilliland. FD. (2005). Birth outcomes and
prenatal exposure to ozone, carbon monoxide, and paniculate matter: Results from the Children's Health
Study. Environ Health Perspect 113: 1638-1644. http://dx.doi.org/10.1289/ehp.8111
Schikowski. T; Sugiri. D; Ranft. U: Gehring. U; Heinrich. J: Wichmann. HE; Kramer. U. (2005). Long-
term air pollution exposure and living close to busy roads are associated with COPD in women. Respir
Res 22: 152-161. http://dx.doi.org/10.1186/1465-9921-6-152
Schultz. ES: Gruzieva. O: Bellander. T; Bottai. M; Hallberg. J: Kull. I; Svartengren. M; Melen. E;
Pershagen. G. (2012). Traffic-related air pollution and lung function in children at 8 years of age: A
birth cohort study. Am J Respir Crit Care Med 186: 1286-1291. http://dx.doi.org/10.1164/rccm.201206-
1045OC
Seilkop. SK; Campen. MJ; Lund. AK; McDonald. JD; Mauderly. JL. (2012). Identification of chemical
components of combustion emissions that affect pro-atherosclerotic vascular responses in mice. Inhal
Toxicol 24: 270-287. http://dx.doi.org/10.3109/08958378.2012.667455
Sekine. K; Shima. M; Nitta. Y; Adachi. M. (2004). Long term effects of exposure to automobile exhaust on
the pulmonary function of female adults in Tokyo, Japan. Occup Environ Med 61: 350-357.
http://dx.doi.org/10.1136/oem.2002.005934
Shalamberidze. OP; Tsereteli. NT. (1971a). Effect of low concentrations of sulfur and nitrogen dioxides on
the estrual cycle and reproductive functions of experimental animals. Gig Sanit 36: 178-182.
Shalamberidze. OP; Tsereteli. NT. (1971b). [Effect of small concentrations of sulfur dioxide and nitrogen
peroxide on the estrus cycle and reproductive function in experiments on animals]. Gig Sanit 36: 13-17.
Sherwin. RP; Richters. V. (1982). Hyperplasia of type 2 pneumocytes following 0.34 ppm nitrogen dioxide
exposure: Quantitation by image analysis. Arch Environ Health 37: 306-315.
http://dx.doi.org/10.1080/00039896.1982.10667584
Shima. M; Nitta. Y; Ando. M; Adachi. M. (2002). Effects of air pollution on the prevalence and incidence
of asthma in children. Arch Environ Health 57: 529-535. http://dx.doi.org/10.1080/00039890209602084
Slama. R; Darrow. L; Parker. J: Woodruff. TJ; Strickland. M; Nieuwenhuijsen. M; Glinianaia. S: Hoggatt.
KJ; Kannan. S: Hurley. F; Kalinka. J: Sram. R; Brauer. M; Wilhelm. M; Heinrich. J: Ritz. B. (2008).
Meeting report: Atmospheric pollution and human reproduction. Environ Health Perspect 116: 791-798.
http://dx.doi.org/10.1289/ehp.11074
Slama. R: Morgenstern. V; Cyrys. J: Zutavern. A; Herbarth. O; Wichmann. HE; Heinrich. J: Lisa Group. S.
(2007). Traffic-related atmospheric pollutants levels during pregnancy and offspring's term birth weight:
A study relying on a land-use regression exposure model. Environ Health Perspect 115: 1283-1292.
http://dx.doi.org/10.1289/ehp.10047
S10rdal. LH; Walker. SE; Solberg. S. (2003). The urban air dispersion model EPISODE applied in
AirQUIS2003: Technical description. (TR 12/2003). Kjeller, Norway: Norwegian Institute for Air
Research.
Sokol. RZ; Kraft. P; Fowler. IM; Mamet. R: Kim. E; Berhane. KT. (2006). Exposure to environmental
ozone alters semen quality. Environ Health Perspect 114: 360-365. http://dx.doi.org/10.1289/ehp.8232
Son. JY; Cho. YS: Lee. JT. (2008). Effects of air pollution on postneonatal infant mortality among firstborn
infants in Seoul, Korea: Case-crossover and time-series analyses. Arch Environ Occup Health 63: 108-
113. http://dx.doi.Org/10.3200/AEOH.63.3.108-113
November 2013 5-175 DRAFT: Do Not Cite or Quote
-------
Sonnenschein-Van der Voort. AM; de Kluizenaar. Y; Jaddoe. VW: Gabriele. C: Raat H; Moll. HA;
Hofman. A; Pierik. FH; Miedema. HM; de Jongste. JC: Duijts. L. (2012). Air pollution, fetal and infant
tobacco smoke exposure, and wheezing in preschool children: a population-based prospective birth
cohort. Environ Health Global Access Sci Source 11: 91. http://dx.doi.org/10.1186/1476-069X-ll-91
Stevens. MA; Menache. MG: Crapo. JD; Miller. FJ; Graham. JA. (1988). Pulmonary function in juvenile
and young adult rats exposed to low-level NO2 with diurnal spikes. J Toxicol Environ Health 23: 229-
240. http://dx.doi.org/10.1080/15287398809531109
Sunyer. J; Jarvis. D; Gotschi. T; Garcia-Esteban. R; Jacquemin. B; Aguilera. I; Ackerman. U; de Marco. R;
Forsberg. B; Gislason. T; Heinrich. J; Norback. D; Villani. S: Kiinzli. N. (2006). Chronic bronchitis and
urban air pollution in an international study. Occup Environ Med 63: 836-843.
http://dx.doi.org/10.1136/oem.2006.027995
Sunyer. J: Pekkanen. J: Garcia-Esteban. R; Svanes. C: Kiinzli. N; Janson. C: de Marco. R: Anto. JM;
Burnev. P. (2007). Asthma score: Predictive ability and risk factors. Allergy 62: 142-148.
http://dx.doi.0rg/10.llll/i.1398-9995.2006.01184.x
Sunyer. J: Puig. C: Torrent. M; Garcia-Algar. O; Calico. I; Munoz-Ortiz. L; Barnes. M; Cullinan. P. (2004).
Nitrogen dioxide is not associated with respiratory infection during the first year of life. Int J Epidemiol
33: 116-120. http://dx.doi.org/10.1093/iie/dvh037
Suzuki. AK; Tsubone. H; Ichinose. T; Oda. H; Kubota. K. (1981). [Effects of subchronic nitrogen dioxide
exposure on arterial blood pHa, PaCO2 and PaO2 in rats]. Nippon Eiseigaku Zasshi 36: 816-823.
Svendsen. ER; Gonzales. M; Mukerjee. S: Smith. L; Ross. M; Walsh. D; Rhonev. S: Andrews. G:
Ozkaynak. H; Neas. LM. (2012). GIS-modeled indicators of traffic-related air pollutants and adverse
pulmonary health among children in El Paso, Texas. Am J Epidemiol 176 Suppl 7: S131-S141.
http://dx.doi.org/10.1093/aie/kws274
Tabacova. S: Balabaeva. L. (1988). Nitrogen dioxide embryotoxicity and lipid peroxidation [Abstract].
Teratology 38: 29A.
Tabacova. S: Balabaeva. L; Vardev. F. (1984). Nitrogen dioxide: maternal and fetal effects. 25th Congress
of the European Society of Toxicology, June 11-14, 1984, Budapest, Hungary.
Tabacova. S: Nikiforov. B; Balabaeva. L. (1985). Postnatal effects of maternal exposure to nitrogen
dioxide. Neurotoxicol Teratol 7: 785-789.
Tager. IB; Balmes. J: Lurmann. F; Ngo. L; Alcorn. S: Kiinzli. N. (2005). Chronic exposure to ambient
ozone and lung function in young adults. Epidemiology 16: 751-759.
http://dx.doi.org/10.1097/01.ede.0000183166.68809.bO
Takano. H; Yanagisawa. R: Inoue. KI; Shimada. A; Ichinose. T; Sadakane. K; Yoshino. S: Yamaki. K;
Morita. M; Yoshikawa. T. (2004). Nitrogen dioxide air pollution near ambient levels is an atherogenic
risk primarily in obese subjects: A brief communication. Exp Biol Med 229: 361-364.
Takenoue. Y; Kaneko. T; Miyamae. T; Mori. M; Yokota. S. (2012). The influence of outdoor NO(2)
exposure on asthma in childhood: Meta-analysis. Pediatrics International 54: 762-769.
http://dx.doi.Org/10.llll/i.1442-200X.2012.03674.x
Tepper. JS: Costa. PL; Winsett. DW: Stevens. MA; Doerfler. PL; Watkinson. WP. (1993). Near-lifetime
exposure of the rat to a simulated urban profile of nitrogen dioxide: Pulmonary function evaluation.
Fundam Appl Toxicol 20: 88-96. http://dx.doi.Org/10.1093/toxsci/20.l.88
Tonne. C: Wilkinson. P. (2013). Long-term exposure to air pollution is associated with survival following
acute coronary syndrome. European Heart Journal (Online) 34: 1306-1311.
http://dx.doi.org/10.1093/eurhearti/ehs480
Toren. K; Gislason. T; Omenaas. E; Jogi. R; Forsberg. B; Nystrom. L; Olin. AC: Svanes. C: Janson. C.
(2004). A prospective study of asthma incidence and its predictors: The RHINE study. Eur Respir J 24:
942-946. http://dx.doi.org/10.1183/09031936.04.00044804
November 2013 5-176 DRAFT: Do Not Cite or Quote
-------
Trasande. L; Wong. K; Roy. A; Savitz. DA; Thurston. G. (2013). Exploring prenatal outdoor air pollution,
birth outcomes and neonatal health care utilization in a nationally representative sample. J Expo Sci
Environ Epidemiol 23: 315-321. http://dx.doi.org/10.1038/jes.2012.124
Tsai. S. -S: Chen. C. -C: Hsieh. H. -J; Chang. C. -C: Yang. C. -Y. (2006). Air pollution and postneonatal
mortality in atropical city: Kaohsiung, Taiwan. Inhal Toxicol 18: 185-189.
http://dx.doi.org/10.1080/08958370500434214
Tsubone. H; Suzuki. AK. (1984). Reflex cardiopulmonary responses by stimulation to type J receptors in
rats exposed to NO2. J Toxicol Environ Health 13: 905-917.
htrp://dx.doi.org/10.1080/15287398409530550
Tung. KY; Tsai. CH; Lee. YL. (2011). Microsomal epoxide hydroxylase genotypes/diplotypes, traffic air
pollution, and childhood asthma. Chest 139: 839-848. http://dx.doi.org/10.1378/chest.10-2479
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park,NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2009b). Integrated science assessment for particulate
matter - Annexes with errata [EPA Report]. (EPA/600/R-08/139). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2013a). Integrated science assessment for lead [EPA
Report]. (EPA/600/R-10/075F). Research Triangle Park, NC.
http://ofmpub.epa.gov/eims/eimscomm.getfile7p download id=514513
U.S. EPA (U.S. Environmental Protection Agency). (2013e). Supplemental table Sl-1. Epidemiologic
studies of health effects not evaluated in the ISA for oxides for nitrogen [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013g). Supplemental table S5-2. Epidemiologic
studies of exposure to oxides of nitrogen and pregnancy effects [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013h). Supplemental table S5-3. Epidemiologic
studies of exposure to oxides of nitrogen and fetal growth [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013i). Supplemental table S5-4. Epidemiologic
studies exposure to oxides of nitrogen and preterm birth [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013J). Supplemental table S5-5. Epidemiologic
studies of exposure to oxides of nitrogen and birth weight [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013k). Supplemental table S5-6. Epidemiologic
studies of exposure to oxides of nitrogen and birth defects [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (20131). Supplemental table S5-7. Epidemiologic
studies of exposure to nitrogen dioxide and neurodevelopmental effects [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013m). Supplemental table S5-8. Epidemiologic
studies of exposure to nitrogen dioxide and infant mortality [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013n). Supplemental table S5-9. Epidemiologic
studies of long-term exposure to oxides of nitrogen and mortality [EPA Report].
U.S. EPA (U.S. Environmental Protection Agency). (2013o). Supplemental table S5-10. Epidemiologic
studies of exposure to oxides of nitrogen and cancer incidence and mortality [EPA Report].
November 2013 5-177 DRAFT: Do Not Cite or Quote
-------
van den Hooven. EH; de Kluizenaar. Y; Pierik. FH; Hofman. A; van Ratingen. SW: Zandveld. PY;
Lindemans. J: Russcher. H; Steegers. EA; Miedema. HM; Jaddoe. VW. (2012a). Chronic air pollution
exposure during pregnancy and maternal and fetal C-reactive protein levels: The generation R study.
Environ Health Perspect 120: 746-751. http://dx.doi.org/10.1289/ehp.1104345
van den Hooven. EH; de Kluizenaar. Y; Pierik. FH; Hofman. A; van Ratingen. SW: Zandveld. PY;
Mackenbach. JP; Steegers. EA; Miedema. HM; Jaddoe. VW. (2011). Air pollution, blood pressure, and
the risk of hypertensive complications during pregnancy: The generation R study. Hypertension 57:
406-412. http://dx.doi.org/10.1161/HYPERTENSIONAHA.110.164087
van den Hooven. EH; Pierik. FH: de Kluizenaa. Y; Willemsen. SP; Hofman. A; van Ratingen. SW:
Zandvel. PYJ; Mackenbach. JP; Steegers. EAP; Miedema. HME; Jaddoe. VWV. (2012b). Air pollution
exposure during pregnancy, ultrasound measures of fetal growth, and adverse birth outcomes: A
prospective cohort study. Environ Health Perspect 120: 150-156. http://dx.doi.org/10.1289/ehp.1003316
van den Hooven. EH; Pierik. FH: de Kluizenaar. Y; Hofman. A; van Ratingen. SW: Zandveld. PY;
Russcher. H; Lindemans. J: Miedema. HM; Steegers. EA; Jaddoe. VW. (2012c). Air pollution exposure
and markers of placenta! growth and function: The Generation R Study. Environ Health Perspect 120:
1753-1759. http://dx.doi.org/10.1289/ehp.1204918
van Kempen. E; Fischer. P; Janssen. N; Houthuijs. D; van Kamp. I; Stansfeld. S: Cassee. F. (2012).
Neurobehavioral effects of exposure to traffic-related air pollution and transportation noise in primary
schoolchildren. Environ Res 115: 18-25. http://dx.doi.Org/10.1016/j.envres.2012.03.002
Volk. HE; Lurmann. F; Penfold. B; Hertz-Picciotto. I; Mcconnell. R. (2013). Traffic-related air pollution,
paniculate matter, and autism. Arch Gen Psychiatry 70: 71-77.
http://dx.doi.org/10.1001/iamapsvchiatry.2013.266
Vrijheid. M; Martinez. D; Manzanares. S: Dadvand. P; Schembari. A; Rankin. J: Nieuwenhuijsen. M.
(2011). Ambient air pollution and risk of congenital anomalies: A systematic review and meta-analysis
[Review]. Environ Health Perspect 119: 598-606. http://dx.doi.org/10.1289/ehp.1002946
Wagner. WD; Duncan. BR; Wright. PG: Stokinger. HE. (1965). Experimental study of threshold limit of
NO2. Arch Environ Health 10: 455-466. http://dx.doi.org/10.1080/00039896.1965.10664029
Wang. SO: Zhang. JL; Zeng. XD; Zeng. YM; Wang. SC: Chen. SY (2009a). Association of traffic-related
air pollution with children's neurobehavioral functions in Quanzhou, China. Environ Health Perspect
117: 1612-1618. http://dx.doi.org/10.1289/ehp.0800023
Wang. XY; Hu. W: Tong. S. (2009b). Long-term exposure to gaseous air pollutants and cardio-respiratory
mortality in Brisbane, Australia. Geospat Health 3: 257-263.
Wei. Y; Davis. J: Bina. WF. (2012). Ambient air pollution is associated with the increased incidence of
breast cancer in US. Int J Environ Health Res 22: 12-21.
http://dx.doi.org/10.1080/09603123.2011.588321
Weng. HH; Tsai. SS: Chen. CC: Chiu. HF; Wu. TN; Yang. CY (2008). Childhood leukemia development
and correlation with traffic air pollution in Taiwan using nitrogen dioxide as an air pollutant marker. J
Toxicol Environ Health A 71: 434-438. http://dx.doi.org/10.1080/15287390701839042
Wilhelm. M; Ghosh. JK; Su. J: Cockburn. M; Jerrett. M; Ritz. B. (2012). Traffic-related air toxics and term
low birth weight in Los Angeles County, California. Environ Health Perspect 120: 132-138.
http://dx.doi.org/10.1289/ehp.1103408
Wilhelm. M; Ritz. B. (2005). Local variations in CO and particulate air pollution and adverse birth
outcomes in Los Angeles County, California, USA. Environ Health Perspect 113: 1212-1221.
http://dx.doi.org/10.1289/ehp.7751
Wollmann. HA. (1998). Intrauterine growth restriction: Definition and etiology. Horm Res 49: 1-6.
http://dx.doi.org/10.1159/000053079
November 2013 5-178 DRAFT: Do Not Cite or Quote
-------
Wood. AM; Harrison. RM; Semple. S: Ayres. JG: Stocklev. RA. (2009). Outdoor air pollution is associated
with disease severity in al-antitrypsin deficiency. Eur Respir J 34: 346-353.
http://dx.doi.org/10.1183/09031936.00087908
Woodruff. TJ; Parker. JD; Adams. K; Bell. ML; Gehring. U; Glinianaia. S: Ha. EH; Jalaludin. B; Slama. R.
(2010). International Collaboration on Air Pollution and Pregnancy Outcomes (ICAPPO). Int J Environ
Res Public Health 7: 2638-2652. http://dx.doi.org/10.3390/iierph7062638
Woodruff. TJ; Parker. JD; Darrow. LA; Slama. R; Bell. ML; Choi. H; Glinianaia. S: Hoggatt. KJ; Karr. CJ:
Lobdell. DT; Wilhelm. M. (2009). Methodological issues in studies of air pollution and reproductive
health. Environ Res 109: 311-320. http://dx.doi.Org/10.1016/j.envres.2008.12.012
Wu. J: Ren. C: Delfino. RJ; Chung. J: Wilhelm. M; Ritz. B. (2009). Association between local traffic-
generated air pollution and preeclampsia and preterm delivery in the south coast air basin of California.
Environ Health Perspect 117: 17731779.
Wu. J: Wilhelm. M; Chung. J: Ritz. B. (2011 a). Comparing exposure assessment methods for traffic-related
air pollution in an adverse pregnancy outcome study. Environ Res 111: 685-692.
http://dx.doi.0rg/10.1016/i.envres.2011.03.008
Yamamoto. I; Takahashi. M. (1984). Ultrastructural observations of rat lung exposed to nitrogen dioxide for
7 months. Kitasato Archives of Experimental Medicine 57: 57-65.
Yang. CY; Hsieh. HJ; Tsai. SS: Wu. TN; Chiu. HF. (2006). Correlation between air pollution and
postneonatal mortality in a subtropical city: Taipei, Taiwan. J Toxicol Environ Health A 69: 2033-2040.
http://dx.doi.org/10.1080/15287390600746181
Yorifuji. T; Kashima. S: Tsuda. T; Takao. S: Suzuki. E; Doi. H; Sugiyama. M; Ishikawa-Takata. K; Ohta. T.
(2010). Long-term exposure to traffic-related air pollution and mortality in Shizuoka, Japan. Occup
Environ Med 67: 111-117. http://dx.doi.org/10.1136/oem.2008.045542
Zemp. E; Elsasser. S: Schindler. C: Kiinzli. N; Perruchoud. AP; Domenighetti. G: Medici. T; Ackermann-
Liebrich. U; Leuenberger. P; Monn. C: Bolognini. G: Bongard. JP; Brandli. O: Karrer. W: Keller. R;
Schoni. MH; Tschopp. JM; Villiger. B; Zellweger. JP (1999). Long-term ambient air pollution and
respiratory symptoms in adults (SAPALDIA study). Am J Respir Crit Care Med 159: 1257-1266.
http://dx.doi.0rg/10.1164/airccm.159.4.9807052
November 2013 5-179 DRAFT: Do Not Cite or Quote
-------
CHAPTER 6 POPULATIONS POTENTIALLY AT
INCREASED RISK FOR HEALTH EFFECTS
RELATED TO EXPOSURE TO OXIDES OF
NITROGEN
6.1 Introduction
1 Interindividual variation in human responses to air pollution exposure can result in some
2 groups being at increased risk for detrimental effects in response to ambient exposure to
3 an air pollutant. The NAAQS are intended to provide an adequate margin of safety for
4 both the population as a whole and those potentially at increased risk for health effects in
5 response to ambient air pollution exposure (see Preface to this ISA). To facilitate the
6 identification of populations and lifestages at greater risk for air pollutant related health
7 effects, this chapter evaluates studies that examine factors that may contribute to the
8 susceptibility and/or vulnerability of an individual to air pollutants. The definitions of
9 susceptibility and vulnerability have been found to vary across studies, but in most
10 instances "susceptibility" refers to biological or intrinsic factors (e.g., lifestage, sex,
11 pre-existing disease/conditions) while "vulnerability" refers to non-biological or extrinsic
12 factors (e.g., socioeconomic status) (Sacks etal.. 2011; U.S. EPA. 2010b. 2009a). In
13 some cases, the terms "at-risk" and "sensitive" populations have been used to encompass
14 these concepts more generally. The main goal of this evaluation of evidence in this
15 chapter is to identify and understand those factors that may result in a population or
16 lifestage being at increased, or in some cases decreased, risk of health effects related to
17 exposure to oxides of nitrogen, and not to categorize the factors by definition.
18 Individuals, and ultimately populations, could experience increased, or in some instances
19 decreased risk, for air pollutant-induced health effects via multiple avenues. As discussed
20 in the Preamble, risk may be modified by intrinsic or extrinsic factors, differences in
21 dose/exposure, or differences in exposure to air pollutant concentrations. It is important
22 to note that the emphasis of this chapter is to identify and understand the factors that
23 potentially increase or decrease the risk of health effects related to exposure to oxides of
24 nitrogen, regardless of whether the change in risk is due to intrinsic factors, extrinsic
25 factors, increased dose/exposure, or a combination. The following sections examine
26 factors that potentially lead to increased or decreased risk of health effects related to
27 exposure to oxides of nitrogen and characterize the overall weight of evidence and the
28 magnitude of effect, when possible, for each factor.
November 2013 6-1 DRAFT: Do Not Cite or Quote
-------
Approach to Classifying Potential At-Risk Factors
1 The systematic approach used to classify potential at-risk factors is described in more
2 detail in the Preamble. The evidence evaluated includes recent studies discussed in
3 Chapter 4 and Chapter 5 of this ISA building on the evidence presented in the 2008 ISA
4 for Oxides of Nitrogen (U.S. EPA. 2008c) and the 1993 Air Quality Criteria for Oxides
5 of Nitrogen (U.S. EPA. 1993) and using the current framework to systematically classify
6 at-risk populations or lifestages that has been presented in past ISAs (U.S. EPA. 2013a.
7 b). In general, the current approach builds on the causal framework used throughout the
8 ISA; conclusions made regarding the strength of evidence are based on evaluation and
9 synthesis across scientific disciplines for each factor that may contribute to increased or
10 decreased risk of a health effect related to exposure to oxides of nitrogen. Important
11 considerations in the evaluation of stratified results include a priori versus post-hoc
12 analyses, multiple comparisons, and small sample sizes in individual strata. These factors
13 can increase the probability of finding associations by chance or reduce power to detect
14 associations in subgroup analyses. Thus, coherence and biological plausibility from other
15 lines of evidence are important to inform these potential uncertainties in epidemiologic
16 results. As discussed in the Preamble, this evaluation focuses on epidemiologic studies
17 that conducted stratified analyses to compare populations or lifestages exposed to similar
18 air pollutant concentrations within the same study design in addition to controlled human
19 exposure and toxicological studies in animals examining effects various at-risk factors
20 (e.g., genetic background or pre-existing disease) on response to exposure to oxides of
21 nitrogen. More detailed discussions of these individual studies are presented in Chapter 4
22 and Chapter 5 as the objective of this chapter is to evaluate and categorize the evidence
23 for each factor as adequate, suggestive, inadequate, or no effect. These categories are
24 described in more detail in Table 6-1. and a summary of the classification of evidence for
25 the factors considered for increased risk of health effects related to exposure to oxides of
26 nitrogen is presented in Section 6.6.
November 2013 6-2 DRAFT: Do Not Cite or Quote
-------
Table 6-1
Classification
Adequate
evidence
Suggestive
evidence
Inadequate
evidence
Evidence of
no effect
Classification of Evidence for Potential At-Risk Factors.
Health Effects
There is substantial, consistent evidence within a discipline to conclude that a factor results in a
population or lifestage being at increased or decreased risk of air pollutant-related health effect(s)
relative to some reference population or lifestage. Where applicable this includes coherence
across disciplines. Evidence includes multiple high-quality studies.
The collective evidence suggests that a factor results in a population or lifestage being at
increased or decreased risk of an air pollutant-related health effect relative to some reference
population or lifestage, but the evidence is limited due to some inconsistency within a discipline or,
where applicable, a lack of coherence across disciplines.
The collective evidence is inadequate to determine if a factor results in a population or lifestage
being at increased or decreased risk of an air pollutant-related health effect relative to some
reference population or lifestage. The available studies are of insufficient quantity, quality,
consistency, and/or statistical power to permit a conclusion to be drawn.
There is substantial, consistent evidence within a discipline to conclude that a factor does not
result in a population or lifestage being at increased or decreased risk of air pollutant-related
health effect(s) relative to some reference population or lifestage. Where applicable this includes
coherence across disciplines. Evidence includes multiple high-quality studies.
6.2 Genetic Factors
1 Genetic variation in the human population is known to contribute to numerous diseases
2 and differential physiologic responses. Furthermore, genetic background has been
3 considered as a response modifying factor in studies examining air pollution-related
4 health effects, including NO2. Studies included in this ISA that evaluate genetic factors
5 have used a targeted approach, focusing on specific genes that are suggested to have a
6 role in signaling pathways involved in biological responses to air pollutants. In particular,
7 most studies examined variants for genes encoding antioxidant enzymes (glutathione S-
8 transferases [GSTM1 and GSTP1], glutathione synthetase [GSS], glutathione reductase
9 [GSR], and NADPH reductase quinone 1[NQO1]) and mediators of immune response
10 (tumor necrosis factor [TNF] and toll-like receptor 4 [TLR4]). Modification by gene
11 variants has been examined primarily for NO2-associated respiratory outcomes, although
12 a few studies examined other health outcomes (i.e., cognitive function, heart rate
13 variability). It is important to note that the functional or biological consequence of some
14 of the gene variants examined in the literature is unknown; however, when available, the
15 variant effect is described (Table 6-2).
16 Oxidative stress has been described as a key process underlying the respiratory effects of
17 NO2 exposure (Section 3.3.2.1): however, studies did not find NO2-associated respiratory
18 effects to be modified consistently by variants in GSTM1 or GSTP1, which encode
19 enzymes with altered oxidative metabolizing activity. Romieu et al. (2006) found
November 2013
6-3
DRAFT: Do Not Cite or Quote
-------
1 associations of short-term NO2 exposure with respiratory symptoms and asthma
2 medication to be larger for children with the GSTM1 positive genotype compared to
3 children who were GSTM1 null. GSTM1 positive is associated with normal antioxidant
4 activity and characterized 62% of the study population. In contrast, Castro-Giner et al.
5 (2009) did not observe the association between long-term NO2 exposure and asthma to
6 differ by GSTM1 genotype. For the GSTP1 variant at codon 105, short-term NO2
7 exposure was associated with larger risk of respiratory symptoms and medication use
8 among children with asthma with the lie/lie or Ile/Val genotype (Romieu et al.. 2006).
9 which is not associated with reduced antioxidant activity. However, studies of long-term
10 exposure found no difference between the group having the lie/lie genotype and the
11 group with Ile/Val or Val/Val genotypes for associations of NO2 with risk of asthma or
12 wheeze (Castro-Giner et al.. 2009; Melen et al.. 2008). Melen et al. (2008). however, did
13 find evidence for increased NO 2-related risk of asthma in children with the GSTP 114Val
14 genotype compared to children having the GSTP Alal 14Ala genotype.
15 Variant genotypes in other glutathione metabolism pathway genes (GSS, GSR, GCLM,
16 and GCLC) were evaluated for potential effect modification of impaired lung function
17 growth in children and exposure to oxides of nitrogen (Breton et al.. 2011). Among the
18 multiple comparisons made, only variation in the GSS haplotype containing a
19 polymorphism of unknown function (rs!801310) was associated with differences in lung
20 function growth (FEVi and MMEF) attributable to NO2 exposure. Bajaetal. (2010) was
21 the only study to report on differences in non-respiratory effects (heart rate-corrected QT
22 interval) of NO2 exposure across glutathione genotypes. A strength of this study was
23 analysis of genetic variants as a composite rather than performing multiple comparisons
24 of individual variants. A genetic susceptibility score was determined for each subject
25 based on genotypes for 10 different genes involved in oxidative stress responses; subjects
26 with a high genetic susceptibility score had mostly unfavorable genotypes while subjects
27 with a low genetic susceptibility score had mostly genotypes associated with protection
28 against oxidative stress. The association between heart-rate-corrected QT interval and
29 NO2 was greater among subjects with a high genetic susceptibility score (Baja et al..
30 2010).
31 The enzyme NADPH dehydrogenase (quinone-1) (NQO1) is also associated with
32 oxidative metabolism, and adults homozygous for the major allele of NQO1 rs2917666
33 were found to have higher NO2-associated prevalence of asthma and bronchial
34 hyperresponsiveness (Castro-Giner et al.. 2009). though the functional consequence of
35 this polymorphism is unknown. Associations between NO2 and asthma prevalence did
36 not vary for other genotypes of NQO1 (rs!800566 and rslOS 17). Further, this study
37 examined several genetic variants, and the prevalence of any given variant was 15% or
38 less.
November 2013 6-4 DRAFT: Do Not Cite or Quote
-------
1 Mediators of the immune response including TNF and TLR4 are known to have a role in
2 oxidant-induced inflammation and asthma pathogenesis and as such they have been
3 examined as potential factors that may increase the risk of NO2-related health effects, but
4 evidence does not clearly demonstrate effect modification by TNF variants. Castro-Giner
5 et al. (2009) evaluated associations between NO2 and asthma prevalence across three
6 polymorphisms in TNF, comparing homozygotes for the major allele to heterozygotes
7 and homozygotes for the minor allele. Subjects homozygous for the major allele of
8 TNFA rs2844484 had higher odds of NO 2-associated asthma prevalence compared to
9 other genotypes; however, no differences were observed for other polymorphisms,
10 including the common TNF 308 variant (rs 1800629). Melen et al. (2008) found that this
11 TNF 308 variant genotype in combination with GSTP Val/Val increased risk for NO2-
12 associated sensitization to allergens in children relative to other diplotypes. This analysis
13 was based on small numbers, and the association in the group with both variants was
14 estimated with large imprecision. Risk in individuals with polymorphisms in TLR4 was
15 also evaluated, but no genotype differences were observed for NO 2-associated asthma
16 (Castro-Giner et al.. 2009).
17 The beta-2-adrenergic receptor (ADRB2) is an encoded G protein-coupled receptor that
18 plays an important role in regulation of airway smooth muscle tone and is the
19 pharmacological target of beta-agonist asthma medications (Hizawa. 2011). NO2
20 exposure has been shown to induce AHR in adults (Section 4.2.2.1). providing a
21 plausible role for variants in this gene in increasing the risk of NO2-associated respiratory
22 effects. However, evidence for effect modification is inconsistent. Castro-Giner et al.
23 (2009) found that variant genotypes for several ADRB2 polymorphisms did not modify
24 odds of NO 2-associated asthma prevalence. In contrast, Fu etal. (2012a) demonstrated
25 that the association between indoor NO2 exposure and severe childhood asthma was
26 stronger among children with higher methylation of the ADRB2 promoter, which is
27 associated with reduced expression of the receptor. Coherent with the mixed evidence for
28 effect modification by ADRB2 variants, there is mixed evidence for bronchodilator use
29 modifying NO2-associated respiratory effects (Section 4.2.2.2).
30 Antioxidant and immune modulation have been described as key events to inform the
31 mode of action underlying the health effects associated with NO2 exposure (Sections
32 3.3.2.1 and 3.3.2.6). The epidemiologic evidence demonstrates that some antioxidant and
33 immune-related gene variants can modify response to NO2 exposure, but there are
34 inconsistent results for the modification by any particular gene variant of associations
35 with respiratory outcomes across studies. Several results are based on post-hoc analyses
36 comprising small proportions of study populations and multiple comparisons. Further,
37 there are no controlled human exposure or toxicological studies comparing effects of
38 NO2 across different genotypes. However, a role for variants in enzymes involved in
November 2013 6-5 DRAFT: Do Not Cite or Quote
-------
1
2
3
4
oxidative metabolism is supported by findings in animals that dietary antioxidant vitamin
levels influence NO2-related oxidative stress (Section 6.5.1). Overall, the collective
evidence suggests that genetic factors modify risk for NO2-related asthma outcomes,
based on evidence from variant genotypes in glutathione metabolism.
Table 6-2
Gene variant
GSTM1 null
GSTP1
VaMOSVal
(rsID 947894)
GSTP
Ala114Valor
Val114Val
(rs1799811)
GSS haplotype
(0100000;
rs1801310)
GSR
Various SNPs
GCLM
Various SNPs
GCLC
Various SNPs
NQO1 CC
(rs2917666)
TNF 308
308 GA/AA
Summary of epidemiologic studies evaluating effect modification by
genetic variants.
Referent
genotype
GSTM1 positive
Ile105lle
GSTP
Ala114Ala
Other
haplotypes
Other
haplotypes
Other
haplotypes
Other
haplotypes
GC or GG
GG
Variant effect
Null oxidant
metabolizing
capacity
Reduced oxidant
metabolizing
capacity
(Val/Val)
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Altered TNF
expression
Direction of
effect
modification by Health outcome/
variant Population
I Respiratory
symptoms and
medication use in
asthmatic children
<-> Asthma prevalence in
adults
| Respiratory
symptoms and
medication use in
asthmatic children
<-> Respiratory
symptoms and
asthma prevalence in
children
<-> Asthma prevalence in
adults
f Asthma prevalence in
children
f Lung function growth
in children
<-> Lung function growth
in children
<-> Lung function growth
in children
<-> Lung function growth
in children
f Asthma prevalence in
adults
<-> Asthma prevalence in
adults
Reference
Romieu et al. (2006)
Castro-Giner et al.
(2009)
Romieu et al. (2006)
Melen et al. (2008)
Castro-Giner et al.
(2009)
Melen et al. (2008)
Breton et al. (2011)
Breton et al. (2011)
Breton et al. (2011)
Breton et al. (2011)
Castro-Giner et al.
(2009)
Castro-Giner et al.
(2009)
November 2013
6-6
DRAFT: Do Not Cite or Quote
-------
Table 6-2 (Continued): Summary of epidemiologic studies evaluating effect modification by
genetic variants.
Direction of
effect
Gene variant
GSTP1 105
lle/ValorVal/Val
+ TNF 308
GA/AA
ADRB2
Intermediate or
high levels of
methylation
ADRB2
rs 104271 3 GG
rs 104271 4 C/C
rs 104271 8 C/C
rs 104271 9 G/G
Referent
genotype
Any other
diplotype
Low levels of
methylation
G/A or AA
C/G or G/G
C/A or A/A
G/C or C/C
Variant effect
Lower oxidant
metabolizing
capacity and
altered TNF
expression
Reduced
expression of
ADRB2
Unknown
modification by Health outcome/
variant Population
t Sensitization and lung
function in children
t Asthma severity in
children
<-> Asthma prevalence in
adults
Reference
Melen et al. (2008)
Fuetal. (2012a)
Castro-Giner et al.
(2009)
6.3 Pre-existing Disease/Conditions
1 Individuals with pre-existing disease are often considered as a subpopulation at greater risk for air
2 pollution-related health effects because they may be at a compromised biological state depending
3 on the disease and severity. The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) concluded
4 that those with pre-existing pulmonary conditions were likely to be at greater risk for NO2-
5 associated health effects, especially individuals with asthma. The majority of recent studies
6 examining effect modification by pre-existing disease continued to focus on asthma, though some
7 studies provide evidence for COPD, cardiovascular disease, and diabetes. Sections 6.3.1, 6.3.2.
8 6.3.3. and 6.3.4 discuss these studies and draw conclusions regarding risk related to each
9 pre-existing disease. Table 6-3 presents the prevalence of these diseases according to the CDC's
10 National Center for Health Statistics (Schiller et al.. 2012). including the proportion of adults with
11 a current diagnosis categorized by age and geographic region. Data for children, when available,
12 are discussed within the relevant sections.
November 2013
6-7
DRAFT: Do Not Cite or Quote
-------
Table 6-3 Prevalence of respiratory diseases, cardiovascular diseases, and
diabetes among adults by age and region in the U.S. in 2010.
Adults (18+) Age (%)a Region(%)b
Chronic Disease/ North
Condition N (in thousands) 18-44 45-64 65-74 75+ east Midwest South West
All (N, in thousands)
229,505
110,615
80,198
21,291
17,401
40,577
53,316
81,721
53,891
Selected Respiratory Diseases
Asthmac
18,
734
8.
1
8.4
8.7
7.4
8.7
8.2
7.7
8.4
COPD - ----- . . .
Chronic Bronchitis
Emphysema
Selected Cardiovascular
All Heart Disease
Coronary Heart Disease
Hypertension
Stroke
Diabetes
9,
4,
883
314
3.
0.
,0
,3
5.3
2.1
6.0
5.4
6.3
6.3
3.8
1.7
4.7
2.3
4.7
1.9
3.1
1.2
Diseases
27,
15,
59,
6.
20,
066
262
259
226
974
4.
1.
9.
0.
2.
,4
,4
,3
,6
8
13.2
7.3
34.4
3.0
12.3
24.3
16.5
54.2
6.1
22.0
37.1
25.8
57.3
10.7
21.7
10.7
6.1
24.0
2.0
7.1
12.2
6.6
24.7
2.9
8.9
12.3
7.2
27.1
2.9
10.1
10.1
5.4
21.7
2.5
8.3
aPercent of individual adults within each age group with disease, based on N (at the top of each age column).
bPercent of individual adults (18+) within each geographic region with disease, based on N (at the top of each region column).
°Asthma prevalence is reported for "still has asthma."
Source: Schiller et al. (2012): National Center for Health Statistics: Data from Tables 1 and 2; Tables 3 and 4; and Tables 7 and 8 of
the CDC report.
6.3.1 Asthma
1 Approximately 8.2% of adults and 9.5% of children in the United States currently have
2 asthma (Schiller et al.. 2012; Bloom etal. 2011). and it is the leading chronic illness of
3 children. A variety of factors affecting the health of individuals with asthma have been
4 identified including ambient air pollution. The 2008 ISA for Oxides of Nitrogen (U.S.
5 EPA. 2008c) concluded that individuals with pre-existing pulmonary conditions are
6 likely at greater risk for ambient NO2-related health effects, with the strongest evidence
7 for individuals with asthma. This conclusion was based on evidence for NO2-related
8 increases in asthma-related hospital admissions and emergency department (ED) visits
9 and AHR and respiratory symptoms in people with asthma. A number of studies
10 discussed in this ISA have evaluated the potential for increased risk of NO2-related
11 respiratory effects among individuals with asthma through comparisons of individuals
November 2013 6-8 DRAFT: Do Not Cite or Quote
-------
1 with or without asthma, comparisons among groups varying in asthma severity, and
2 comparisons among groups varying in asthma medication use.
3 Among epidemiologic studies comparing children with and without asthma, a few studies
4 that defined comparisons a priori found larger NO2-related increases in respiratory
5 symptoms (Patel et al.. 2010). or decrements in lung function (Timonen and Pekkanen.
6 1997) in children with asthma. However, other studies reported similar (Lin et al.. 2011;
7 Gauderman et al., 2004) or larger NO2-related respiratory effects in children without
8 asthma (Berhane etal. 2011; Barraza-Villarreal et al.. 2008) or null associations in either
9 group (Flamant-Hulin et al., 2010; Holguin et al.. 2007). In most studies, asthma was
10 ascertained as self or parental report of physician-diagnosed asthma; however, children
11 with asthma were not at increased risk of NO2-related respiratory effects even with
12 asthma assessed by a pediatric allergist (Barraza-Villarreal et al.. 2008). Barraza-
13 Villarreal et al. (2008) found that associations of interleukin 8 and forced vital capacity
14 with NO2 were actually stronger in children without asthma, though the majority of those
15 children had positive atopy as defined by having a positive skin prick test to an allergen.
16 Berhane etal. (2011) found that associations were stronger in children with a history of
17 respiratory allergies compared to those without. These results suggest increased risk
18 associated with allergy, which is supported by evidence for NO 2 -induced increases in
19 allergic inflammation in controlled human exposure and animal toxicological studies
20 (Sections 3.3.2.6.2. and 4.2.4.3). However, comparisons of children with and without
21 atopic asthma were inconsistent, with Mann etal. (2010) finding larger NO2 -related
22 increases in wheeze among children with asthma sensitized to cat or fungal allergens but
23 other studies finding no difference in associations of NO2 with eNO or lung function
24 decrements in children with and without atopic asthma (Sarnat et al.. 2012; Ranzi et al..
25 2004).
26 Among studies, there was heterogeneity in asthma severity and asthma medication use,
27 and there was some evidence for these factors contributing to heterogeneity in NO2-
28 related respiratory effects. Mann etal. (2010) examined asthma severity and found larger
29 NO 2 -related increases in wheeze among boys with mild, intermittent asthma. There were
30 several studies examining medication use, and overall, results for effect modification
31 were mixed. Some studies found larger NO2-related increases in pulmonary inflammation
32 or oxidative stress among ICS users (Oian et al.. 2009a; Delfino et al.. 2006) while others
33 reported stronger associations with pulmonary inflammation among ICS nonusers (Sarnat
34 etal.. 2012; Hernandez-Cadena et al.. 2009; Liu et al.. 2009b). Similarly, larger NO2-
35 related lung function decrements were found among bronchodilator users in one study
36 (Qian et al.. 2009b) and among those not using bronchodilators in another (Delfino et al..
37 2008a). It is difficult to interpret these findings without a consistent definition of asthma
November 2013 6-9 DRAFT: Do Not Cite or Quote
-------
1 and medication use across studies. For example, ICS use could possibly represent
2 different subgroups of asthma (i.e., severe or well-controlled). Furthermore, most
3 comparisons for medication use were not specified a priori, so these observations could
4 be attributable to a higher probability of finding differences by chance or a lower
5 probability of finding differences because of insufficient statistical power in subgroup
6 analyses.
7 Consistent with epidemiologic studies, controlled human exposure studies provide some
8 evidence that asthma status may be related to health outcomes associated with NO2
9 exposure, particularly for airway hyperresponsiveness (AHR), but there are
10 inconsistencies in results across studies and outcomes. Across studies, short-term
11 exposure to oxides of nitrogen was found to induce AHR in adults with and without
12 asthma. Although no individual study compared adults with and without asthma,
13 increased risk of adults with asthma was indicated by lower concentrations of NO2
14 exposure inducing nonspecific AHR in adults with asthma (200-300 ppb for 30 minutes,
15 100 ppb for 1 h, Section 4.2.2.2) than healthy adults without asthma (1,500-2,000 ppb for
16 1-3 hours, Section 4.2.2.1). A few controlled human exposure studies did compare
17 subjects with asthma and healthy controls with respect to other respiratory outcomes, and
18 NO2-induced decreases in FEVi were found in subjects with asthma but not healthy
19 subjects (Torres etal.. 1995). However, other studies did not find NO2-related effects on
20 lung function or inflammation among subjects with asthma or healthy subjects
21 (Vagaggini et al.. 1996); Linn et al. (1985b). Most studies that examined only subjects
22 with asthma did not find NO 2 -induced decrements in lung function (Jenkins etal.. 1997;
23 Torres and Magnussen. 1991; Kleinman et al.. 1983). In contrast, Bauer etal. (1986)
24 reported that NO2 exposure with exercise in subjects with asthma yielded a significant
25 reduction, approximately 10%, in forced expiratory volume in 1 second (FEVi) and
26 partial expiratory flow rates at 60% of total lung capacity. Studies of allergic responses in
27 adults with asthma also were mixed in reporting NO2-related effects on Th2 cytokines,
28 eosinophil activation, and pulmonary neutrophilia (Riedl etal.. 2012; Witten et al.. 2005;
29 Barck et al.. 2002).
30 As described in Section 4.2.9. the strongest evidence for a causal relationship for
31 respiratory effects of short- and long-term NO2 exposure is that for asthma morbidity.
32 Compelling evidence indicating increased risk of individuals with asthma is provided by
33 controlled human exposure studies showing increased sensitivity of adults with asthma to
34 NO2-induced AHR. While some epidemiologic studies with a priori-defined comparisons
35 of children with and without asthma indicated larger NO2-associated respiratory effects
36 in children with asthma, other studies did not indicate increased risk of children with
37 asthma. Some studies showed effect modification by atopy, asthma severity, or asthma
November 2013 6-10 DRAFT: Do Not Cite or Quote
-------
1 medication use, whereas others did not. Controlled human exposure studies in adults with
2 asthma indicated greater NO2-related effects for some outcomes but not others, but the
3 lack of comparison to healthy adults limits interpretation. Despite some inconsistencies
4 across disciplines, the collective evidence suggests that asthma may increase risk for
5 NO 2-related health effects, particularly AHR, as demonstrated in controlled human
6 exposure studies and some epidemiologic studies.
6.3.2 Chronic Obstructive Pulmonary Disease (COPD)
7 Chronic lower respiratory disease, including COPD, was ranked as the third leading
8 cause of death in the United States in 2011 (Hoyert and Xu. 2012). COPD comprises
9 chronic bronchitis and emphysema which affect approximately 4.3% and 1.9% of the
10 U.S. adult population (Schiller etaL 2012). Given that people with COPD have
11 compromised respiratory function, they may be at increased risk of NO 2 -related health
12 effects.
13 Of the epidemiologic studies evaluated in this ISA, only Suh and Zanobetti (201 Ob)
14 conducted stratified analyses to examine the potential of differential risk in people with
15 and without COPD, albeit in relation to cardiovascular rather than respiratory-related
16 health effects. Further, associations were compared between people with COPD and
17 people with previous MI. NO2-associated decreases in PNN50, a measure of heart rate
18 variability were larger among people with COPD than those with MI; however,
19 associations with other measures of heart rate variability were similar between groups or
20 larger among people with MI. In contrast, a previous study found with larger NO2-related
21 cardiovascular-related ED visits among people with COPD than people without COPD
22 (Peel et al.. 2007).
23
24 Unlike the epidemiologic study discussed above, controlled human exposure studies that
25 examined NO2-related health effects in people with COPD focused on measuring
26 respiratory endpoints (Gong et al.. 2005; Vagaggini etal., 1996; Morrow etal.. 1992;
27 LinnetaL 1985a). In comparisons of older adults with and without COPD, Morrow et al.
28 (1992) reported larger NO 2 -induced decrements in lung function in adults with COPD
29 than never-smoker elderly subjects: 8.2% decrease versus 0.22% decrease in FVC and a
30 4.82% decrease versus a 1.25% increase in FEVi. Similarly, Vagaggini et al. (1996)
31 reported decreased (approximately 10%) FEVi in subjects with COPD following NO2
32 exposure compared to air control exposures, while decrements were not observed in
33 healthy adult subjects. However, other studies that examined only adults with COPD did
November 2013 6-11 DRAFT: Do Not Cite or Quote
-------
1 not report any changes in lung function or pulmonary inflammation following NO2
2 exposure (Gong et al.. 2005; Linn et al.. 1985a).
3 In conclusion, controlled human exposure studies provide some evidence indicating that
4 NO2 exposure can result in larger pulmonary function decrements in individuals with
5 COPD, relative to healthy controls. Epidemiologic evidence points to increased risk in
6 adults with COPD, but in relation to cardiovascular effects, limiting the ability to assess
7 coherence between disciplines. However, the cardiovascular and respiratory systems are
8 linked in that inflammation and poor gas exchange associated with COPD can lead to
9 cardiac tissue damage. Overall, the collective evidence, particularly from controlled
10 human exposure studies, suggests that people with COPD are at increased risk of NO2-
11 related health effects relative to individuals without COPD.
6.3.3 Cardiovascular Disease (CVD)
12 Cardiovascular disease is the primary cause of death in the United States, and it is
13 estimated that approximately 12% of adults report a diagnosis of heart disease. In
14 addition, hypertension has been diagnosed in roughly 25% of the adult U.S. population
15 (Schiller et al.. 2012). Many studies investigating health effects associated with NO2 have
16 included individuals with pre-existing CVD, allowing for evaluation of whether
17 pre-existing CVD modifies risk for NO2-related health effects.
18 Associations between short-term increases in ambient NO2 concentrations and
19 cardiovascular hospital admissions or ED visits were not consistently greater among
20 individuals with pre-existing cardiovascular disease. Individuals with hypertension were
21 found to have larger NO2-related risks of ED visits for arrhythmia (Peel et al., 2007) but
22 not ED visits for ischemic heart disease (IHD) or congestive heart failure (CHF) (Peel et
23 al., 2007) or hospital admissions for myocardial infarction (MI) (D'Ippoliti et al., 2003).
24 The association between ambient NO2 and hospital admissions for MI was larger among
25 individuals with conduction disorders but not individuals with cardiac arrhythmia or heart
26 failure (D'Ippoliti et al.. 2003). A larger NO2-associated increase in hospital admissions
27 for IHD was found among individuals with a secondary diagnosis of CHF; however, this
28 could have been attributable to the large percentage of IHD cases with a secondary CHF
29 diagnosis (Mann et al., 2002).
30 Other studies found that pre-existing cardiovascular disease modified risk of NO2-
31 associated mortality. Chiusolo et al. (2011) found the strongest evidence for increased
32 risk of NO2-related mortality among individuals with diseases of the cardiovascular
33 system, which was consistent with results from Berglind et al. (2009) that demonstrated
34 increased NO2-associated mortality among survivors of MI.
November 2013 6-12 DRAFT: Do Not Cite or Quote
-------
1 Some studies have found that individuals with pre-existing CVD are not at increased risk
2 for NO 2-associated effects on cardiac function; however, another study found no
3 difference in NO 2 -associated ventricular tachyarrhythmia among individuals with and
4 without ischemic heart disease or among categories of left ventricular ejection fraction
5 (Ljungman et al., 2008). The latter finding is supported by results from a controlled
6 human exposure study in which subjects with coronary heart disease or impaired left
7 ventricular systolic function experienced no changes in heart rate or HRV with a 1-hour
8 exposure to 400 ppb NO2 (Scaife etal.. 2012).
9 Toxicological studies using ApoE deficient (ApoE~/-) mice as a model of hyperlipidemia
10 and atherosclerosis have demonstrated effects following NO2 exposure. Campen et al.
11 (2010) and (Seilkop et al., 2012) reported changes in heme oxygenase-1, endothelin-1,
12 and tissue inhibitor of metalloproteinase-2 as well as lipid peroxidation in the aorta
13 following exposure to NO2. However the connection between expression of these genes
14 and cardiovascular outcomes is not well described, limiting the ability of these studies to
15 provide biological plausibility for the potential increased risk of NO2-related health
16 effects observed in epidemiologic studies.
17 The evidence evaluating risk for NO2-associated health effects in individuals with and
18 without pre-existing CVD is inconsistent within and across various study designs and
19 outcomes in epidemiologic and toxicological studies. The epidemiologic evidence
20 indicates that risk for NO2-associated mortality may be greater among individuals with
21 pre-existing CVD, but evidence is inconsistent for cardiovascular hospital admissions,
22 ED visits, and measures of cardiac function. Further, animal toxicological studies do not
23 provide biological plausibility for epidemiologic results. Because of the inconsistencies
24 across epidemiologic study designs and outcomes and lack of clear biological
25 plausibility, the evidence is inadequate to determine whether pre-existing CVD increases
26 risk for health effects associated with NO2 exposure.
6.3.4 Diabetes
27 Diabetes mellitus is a group of diseases characterized by high blood glucose levels that
28 result from defects in the body's ability to produce and/or use insulin. An estimated 20
29 million Americans had diagnosed diabetes mellitus in 2010, representing 9.1% of the
30 adult population (Schiller et al.. 2012). The causes of type 2 diabetes are not fully
31 understood but chronic inflammation has been suggested as an important factor in the
32 development of the disease. The available epidemiologic studies each examined a
33 different outcome and produced inconsistent evidence of associations with NO2
34 exposure. An Italian multeity study found stronger associations between NO2 and
November 2013 6-13 DRAFT: Do Not Cite or Quote
-------
1 mortality among individuals with pre-existing diseases of the cardiovascular system,
2 including diabetes (Chiusolo et al.. 2011). whereas no effect modification by diabetes
3 was observed in studies of respiratory hospital admissions or lung cancer mortality
4 (Faustini etal.. 2013; Yorifuji et al.. 2010). Results also were inconsistent for
5 cardiovascular effects. A stronger association between NO2 and heart rate-corrected QT
6 interval was found in individuals with diabetes (Bajaet al.. 2010). whereas a stronger
7 association between decreased HRV and ambient NO2 was found among individuals
8 without a history of diabetes (Huang et al.. 2012a). No effect modification was reported
9 for ventricular tachyarrhythmia (Ljungman et al., 2008). Overall, the epidemiologic
10 evidence across studies is inconsistent for cardiovascular effects as well as other
11 outcomes, the collective evidence is inadequate to determine if diabetes increases risk
12 NO2-associated health effects.
6.4 Sociodemographic Factors
6.4.1 Lifestage
13 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) indicated that there was
14 supporting evidence for age-related differences in health effects related to NO2 exposure,
15 particularly for children with asthma and older adults. Differential health effects of NO2
16 across age groups may be due to several factors:
1) The human respiratory system is not fully developed until 18-20 years of age,
and therefore, it is plausible to consider children to have intrinsic risk for
respiratory effects due to potential perturbations in normal lung development.
2) Older adults (typically considered those 65 years of age or greater) are
generally at greater risk for ill health for a variety of reasons, including
weakened immune function, impaired healing, decrements in pulmonary and
cardiovascular function, and greater prevalence of pre-existing disease (Table
3) Dose/exposure to oxides of nitrogen due to ventilation and time-activity
patterns may vary across age groups.
17 More specifically, studies of exposure to oxides of nitrogen have identified time-activity
18 patterns and spatial variability in NO2 concentrations to be determinants of inter-
19 individual variability in exposure [(Molter et al., 2012; Kousaet al., 2001). and Section
20 2.6] and time-activity patterns have been shown to differ between children and adults. In
21 comparisons of children (mostly less than 8 years of age), parents of young children
22 (mostly under age 55), and older adults (mostly older than 54 years of age), children were
23 more likely than adults or older adults to take part in vigorous activity or aerobic exercise
November 2013 6-14 DRAFT: Do Not Cite or Quote
-------
1 (indoors and outdoors). Children were also more likely to spend over 30 minutes
2 performing vigorous outdoor physical activity compared with either group of adults (Wu
3 et al., 20lib). Additionally, this study demonstrated small differences among age groups
4 in time spent in various microenvironments. Differences in time-activity patterns across
5 age groups could potentially result in different NO2 exposure and may inform differential
6 health effects associated with NO2 in children and older adults. However, Meng et al.
7 (2012a) suggested a weaker association between personal NO2 exposure of children and
8 ambient NO2 concentrations in a meta-analysis.
6.4.1.1 Children
9 According to the 2010 census, 24% of the U.S. population is less than 18 years of age,
10 with 6.5% less than age 6 (Howden and Meyer. 2011). Furthermore, it is generally
11 recognized that children spend more time outdoors than adults in addition to having
12 higher ventilation rates relative to lung volume, and thus, may have greater exposure to
13 air pollutants, including NO 2.
14 The 2008 ISA for Oxides of Nitrogen reviewed several studies demonstrating larger
15 increases in asthma-related hospital admissions in association with ambient NO2 in
16 children compared to adults and reported that children may be at greater risk for NO2-
17 associated health effects (U.S. EPA. 2008c). Consistent with these findings, several
18 recent studies reported larger NO2-related asthma hospital admissions and ED visits in
19 children living in diverse locations including the U.S., Canada, Greece, and Hong Kong
20 (Son etal.. 2013; Ko et al.. 2007b; Villeneuve et al.. 2007). Son etal. (2013) found that
21 children had a 3.4 fold increase in asthma hospitalizations relative to adults (15-64 years),
22 similar to a study by Ko et al. (2007b) reporting a 2.2 fold increase in hospitalizations due
23 to acute asthma exacerbation in children (0-14 years) compared to adults (15-65). In
24 comparisons of children in different age groups, studies found larger NO2-related risks of
25 asthma hospital admissions or ED visits among younger children (e.g., age 0-4 years, 2-4
26 years) than children ages 5-14 years (Samoli etal.. 2011; Villeneuve et al., 2007). These
27 latter results may have weaker implications since diagnosis in children below the age of 5
28 years is less reliable. Evidence for other asthma outcomes did not clearly indicate
29 increased risk for children. Sinclair et al. (2010) found increased risk for children for
30 NO2-related asthma outpatient visits, whereas Burra et al. (2009) did not find differences
31 for asthma physician visits by age. NO2-associated asthma medication sales also did not
32 vary in children or adults (Laurent et al.. 2009).
33 Because there are differences in lung development over the course of childhood, risk may
34 vary among children according to the time window of exposure. Information on critical
November 2013 6-15 DRAFT: Do Not Cite or Quote
-------
1 time windows of exposure has been provided by studies of long-term NO2 exposure and
2 respiratory effects (Section 5.2.2.1). which found associations of asthma, lung function
3 decrements, and allergic sensitization or conditions with NO2 concentrations at birth or
4 the first year of life and with exposure in the year of diagnosis, other periods in
5 childhood, or lifetime exposure. Among studies that compared time periods of exposure,
6 several found larger risks of respiratory effects in individuals ages 4-21 years (mostly
7 children) associated with NO2 exposure in the first year of life compared with NO2
8 exposure in the first three years of life, year before diagnosis, or over a lifetime (Gruzieva
9 etal., 2013; Nishimuraetal., 2013a; Gruzieva etal.. 2012; SchultzetaL 2012). These
10 results suggest that higher NO2 exposure of children early rather than later in life may
11 increase their risk of developing respiratory effects.
12 Toxicological studies have evaluated differential effects of NO2 among juvenile and
13 mature animals on indicators of lung injury, inflammation, and lung host defense. These
14 studies found effects of NO2 exposure in mature rats and guinea pigs (exposures
15 beginning at 5 or 8 weeks) and not juvenile animals (exposures beginning after birth or
16 5 days) with respect to increases in pulmonary inflammation and impaired alveolar
17 macrophage function (Kumae and Arakawa. 2006) and lung damage (Azoulay-Dupuis et
18 al.. 1983). Although these results are in contrast to epidemiologic observations, the
19 endpoints examined in experimental animals do not have direct coherence with asthma
20 morbidity endpoints examined in the majority of epidemiologic studies.
21 In conclusion, recent epidemiologic studies generally demonstrate that NO2-associated
22 respiratory and asthma hospital admissions and ED visits are greater in children
23 compared to adults, which is consistent with previous conclusions from the 2008 ISA for
24 Oxides of Nitrogen (U.S. EPA, 2008c). There are some inconsistencies in the evidence as
25 some studies did not find children to be at greater risk and age groups examined across
26 studies varied. In children, results indicate that NO2 exposure early in life may increase
27 risk of respiratory effects. Limited toxicological evidence suggests greater NO2-related
28 respiratory effects in mature animals than juveniles, though the endpoints examined do
29 not have direct coherence with the asthma morbidity found in the epidemiologic evidence
30 and is not considered to be in conflict. In addition, time-activity patterns and ventilation
31 rates have been shown to differ between children and adults, such that children are more
32 likely to partake in vigorous outdoor activities and have higher ventilation. While these
33 factors can influence exposure and/or dose, such information is not available for oxides
34 of nitrogen. Overall, there is substantial, consistent evidence from epidemiologic studies
35 to conclude that children are at greater risk of NO2-associated health effects, particularly
36 respiratory hospital admissions and ED visits.
November 2013 6-16 DRAFT: Do Not Cite or Quote
-------
6.4.1.2 Older Adults
1 Older adults may be at greater risk for NO 2-associated health effects due to a number of
2 intrinsic factors as well as differences in time-activity patterns relative to other lifestages.
3 In addition, time-activity patterns and ventilation rates are different among older adults
4 than children and young to middle-aged adults, which likely contribute to different NO2
5 exposure for this lifestage. According to the 2008 National Population Projections issued
6 by the U.S. Census Bureau, approximately 12.9% of the U.S. population is 65 years or
7 older and it is estimated that by 2030, this fraction will grow to 20% (Vincent and
8 Velkoff 2010). Thus, this lifestage represents a substantial proportion of the U.S.
9 population that is potentially at increased risk for health effects associated with NO2
10 exposure.
11 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) indicated that older adults were
12 at increased risk for NO 2-associated morality and respiratory hospital admissions but not
13 cardiovascular effects. Recent studies strengthen the evidence for hospital admissions and
14 indicate elevated risk of older adults for admissions for asthma, respiratory disease,
15 COPD, and lower respiratory tract infections (LRTI). Ko et al. (2007b) and Villeneuve et
16 al. (2007) demonstrated larger associations between NO2 and asthma hospital admissions
17 in adults older than 65 years compared to individuals 15-64 years old or 45-64 years old,
18 respectively. Adults aged 65 years and older were found to have <1 or 2.6 fold greater
19 risk than younger adults in these studies. Villeneuve et al. (2007) also found the largest
20 risk among adults ages 75 years and older, though larger associations were not observed
21 for adults 45-64 years compared to individuals 15-44 years. Wong et al. (2009) found
22 NO2-related respiratory hospital admissions (i.e., respiratory disease, COPD, and LRTI
23 in individuals with COPD) to be greater among older adults (>65 years) relative to all
24 ages. Sonetal. (2013) also evaluated risk for NO2-related respiratory hospital
25 admissions, but only found older adults (65-74 years and > 75 years) to have larger
26 associations for allergic disease relative to other adults (15-64 years). Arbex et al. (2009)
27 specifically examined ED visit for COPD association with ambient NO2 and found
28 greater risk among those older than 65 compared to adults 40-64. Evidence for biological
29 plausibility of differential effects of NO2 by age is limited. Controlled human exposure
30 studies of older adults found no significant or small decrements in lung function with
31 NO2 exposure (Gong et al.. 2005; Morrow et al.. 1992) and no changes in SpO2 or
32 sputum cell counts (Gong et al.. 2005).
33 Increased risk in older adults is substantiated by evidence for mortality, particularly
34 recent multicity studies finding NO2-associated mortality to be greatest among
35 individuals older than 65 years, and even greater among those older than 75 years (Chen
36 et al.. 2012b: Cakmak et al.. 201 Ib). Chiusolo et al. (2011) also observed larger
November 2013 6-17 DRAFT: Do Not Cite or Quote
-------
1 associations between NO2 and mortality among those greater than 85 years old compared
2 to other adults; however, this study did not demonstrate increasing risk with increasing
3 age as individuals between 35-64 years of age had larger associations compared to those
4 65-74 or 75-84 years of age. A few studies did not find elevated risk of NO2-associated
5 mortality among older adults. Cesaroni et al. (2013) reported that adults less than age 60
6 years old have higher NO2-associated all-cause, cardiovascular, and ischemic heart
7 disease mortality. For NO2-associated lung cancer mortality, risk was increased among
8 adults age 60 years and older (Cesaroni et al.. 2013) but similar between those less than
9 or older than age 75 years (Yorifuji et al.. 2010). Although age comparisons differed
10 among studies, adults ages 65-74 years were found to have 2 or 6 fold greater risk of
11 NO2-related mortality (Cakmaket al.. 20lib: Chiusolo et al.. 2011)
12 Consistent with previous studies, recent studies did not consistently show older adults to
13 be at increased risk of NO2-related cardiovascular effects. A larger NO2-associated
14 decrease in the LF component of HRV was found among adults older than age 50 years
15 compared with younger adults or children (Min et al.. 2008). whereas no association was
16 found between NO2 and ventricular tachyarrhythmias in any age group (Ljungman et al..
17 2008). Controlled human exposure studies have not evaluated effects of NO2 on
18 cardiovascular function in older adults.
19 There is substantial and consistent evidence demonstrating older adults are at increased
20 risk for NO2-associated respiratory hospitalizations and mortality. Furthermore, this
21 recent evidence is consistent with that from the 2008 ISA for Oxides of Nitrogen (U.S.
22 EPA. 2008c) and strengthens previous conclusions. There is inconsistent evidence of
23 effect modification by age for cardiovascular outcomes associated with NO2 exposure.
24 Additionally, the limited evidence from controlled human exposure studies in older,
25 healthy adults is not coherent with the evidence from epidemiologic studies. Overall,
26 there is consistent epidemiologic evidence for respiratory effects and mortality across
27 various age ranges to conclude that there is adequate evidence indicating that older adults
28 are at increased risk for NO? -related health effects.
6.4.2 Socioeconomic Status
29 Socioeconomic status (SES) is a measure of an individual's income, education, and
30 occupation that can indicate inequities in access to resources such as healthcare.
31 According to U.S. Census data, 15.9% (approximately 48.5 million) of Americans were
32 of poverty status in 2011 according to household income, which is one metric used to
33 define SES (Bishaw. 2012). A number of other indicators are also used including:
34 education level, employment status, insurance status, social deprivation, and access to
November 2013 6-18 DRAFT: Do Not Cite or Quote
-------
1 health care. Exposure to air pollution may vary according to poverty status and SES, and
2 studies have further demonstrated that SES indicators can modify the association between
3 air pollution and health effects, though the influence of SES on this relationship is not
4 uniform. Deguen and Zmirou-Navier (2010) concluded that individuals having lower
5 SES generally are exposed to higher levels of ambient air NO2. However, O'Neill et al.
6 (2003) noted that several factors might alter this relationship, including changing
7 development, migration, and transportation patterns that could result in individuals of
8 higher socioeconomic status having higher exposures. Yu and Stuart (2013) and Chaix et
9 al. (2006) estimated the relationship between NOX concentrations and income brackets in
10 Tampa, FL and between NO2 concentrations and income brackets in Malmo, Sweden,
11 and observed that the ambient exposures declined with increasing income. Kruize et al.
12 (2007) also estimated the relationship between residential NO2 and income; at the top
13 50% of the distribution, little difference was observed among the income brackets, but
14 across the lower 30% of the distribution, exposures were higher for the lower income
15 groups. Several studies have shown that NO2 exposures increase with factors reflecting
16 lower SES, such as unemployment, overcrowding, lack of car ownership, lack of home
17 ownership, low educational attainment, and not residing in country of birth (Mitchell.
18 2005; Strohetal. 2005; Rotko etal.. 2001). However, Mitchell (2005) did observe that
19 discrepancy in NO2 exposures for a unit increase in a deprivation index declined over the
20 period 1993-2005. These studies support the possibility that individuals of low SES may
21 be at increased risk for exposure to NO2.
22 Studies presented in this ISA provide evidence for SES as a potential risk factor for NO2
23 associated health effects, although it is important to note that some studies were
24 conducted outside of the U.S., and definitions of SES can vary across countries based on
25 population demographics, bureaucracy, and the local economy which can contribute to
26 varying degrees of deprivation or inequities. It can be challenging to make comparisons
27 among these studies given this variation, but there are a number of studies available
28 across disciplines to evaluate SES as a risk factor for NO2-related health effects.
29 Across respiratory outcomes, positive associations were generally reported with ambient
30 NO2 exposure (Section 4.2). but studies stratifying results by various SES indicators did
31 not consistently show differences. Grineski et al. (2010) conducted a study in Phoenix,
32 AZ and found that children without insurance were at greater risk of NO2-associated
33 asthma hospital admissions compared to children with private insurance or Medicaid.
34 However, a study in Toronto, Canada that compared quintiles of household income did
35 not find any evidence of differences in the risk of asthma physician visits by income
36 (Burra et al.. 2009). Studies that examined community-level SES indicators, which may
37 not correspond with individual-level SES, produced inconsistent results. Larger NO2-
November 2013 6-19 DRAFT: Do Not Cite or Quote
-------
1 related increases in asthma were found in children with higher maternal- or self-reported
2 community violence (Cloughertv et al.. 2007); however, associations of asthma ED visits
3 and asthma medication sales did not vary by income-based health insurance premiums in
4 Korean communities (Kim et al.. 2007). across census blocks or areas with varying
5 average household income in Vancouver (Lin et al.. 2004b). or across a composite index
6 of SES indicators in Strasbourg, France (income, job category, education, and housing
7 characteristics) (Laurent et al.. 2009).
8 Results for effect modification by SES were more consistent for mortality. A number of
9 studies that reported associations between short-term NO2 exposure and mortality
10 (Section 4.4.5). also demonstrated effect modification by SES. In a study of 10 Italian
11 cities, Chiusolo et al. (2011) found inconsistent results when examining socioeconomic
12 position and income by dividing the average of the census tract for both indicators into
13 low (<20th percentile), middle (20th to 80th percentile) and high (>80th percentile)
14 categories. Increased risk was observed for the low and high socioeconomic position
15 groups while higher risk was identified for low and middle income groups. In contrast, a
16 study of 7 Chilean cities using several indicators of SES (education attainment,
17 community income level, and employment category) to measure effect modification
18 found evidence of increased risk of NO 2 -related mortality for individuals of low SES
19 (i.e., were not white-collar workers, had a lower income and education level) (Cakmak et
20 al.. 20 lib). These results are consistent with those of Wong et al. (2008b) and (Chen et
21 al.. 2012b): Wong et al. (2008a) found evidence that the most socially deprived areas of
22 Hong Kong, as measured by a composite metric of socioeconomic status, had higher
23 mortality risks, and Chen et al. (2012b) found that education level modified the
24 relationship between NO2 exposure and mortality in a study of 17 Chinese studies. More
25 specifically, examining education as low, illiterate or completion of primary school; high,
26 middle school education and above, Chen et al. (2012b) found those with less education
27 to have increased risk of NO2-related mortality. Overall, the trend across these studies
28 indicates that those of low SES are at increased risk for mortality associated with short-
29 term NO2 exposure.
30 In addition to short-term exposure studies, a long-term study of mortality found evidence
31 for low SES to increase risk of NO 2-associated mortality. In a cohort study conducted in
32 Rome, Italy, Cesaroni et al. (2013) found positive associations between ambient NO2 and
33 mortality for all SES categories and education levels, but associations were more precise
34 and generally larger for low SES and education. Other studies of long-term NO2
35 exposure did not find SES to consistently modify the association between NO2 and lung
36 cancer incidence and mortality. Yorifuji et al. (2010) reported that financial capability did
37 not modify the association between NO2 and lung cancer mortality, while studies that
November 2013 6-20 DRAFT: Do Not Cite or Quote
-------
1 examined long-term NO2 exposure and lung cancer incidence reported inconsistent
2 findings when stratifying results by educational attainment (<8 years, 8-10 years,
3 >8 years) (Raaschou-Nielsen et al., 2011; Raaschou-Nielsen et al., 2010a).
4 The potential modification of NO2 associations by SES was also examined in studies of
5 other health effects including birth and developmental outcomes. Results from these
6 studies were generally null or inconsistent (Section 5.4). and stratified analyses did not
7 clearly identify a particular level of SES within the population for which there is evidence
8 of association (Becerra et al.. 2013; Guxens et al.. 2012; Pereira et al.. 2012; Morello-
9 Froschetal.. 2010). While evidence overall demonstrates no association between low
10 birth weight and long-term NO2 exposure (Section 5.4.3.3), one study demonstrated that
11 decreases in birth weight associated with NO2 were greatest for mothers living in
12 communities with high rates of neighborhood-level poverty (Morello-Frosch et al.. 2010).
13 In contrast, a study in Australia found inconsistent results for modification by SES of the
14 association between NO2 and fetal growth measurements; those with low and high SES
15 (community-level index) had positive associations for small for gestational age with NO2
16 exposure in the 2nd trimester while those of middle SES had positive associations in the
17 3rd trimester (Pereira et al.. 2012). Effect modification by SES was also inconsistent for
18 associations between prenatal NO2 exposure and neurodevelopmental effects (Becerra et
19 al.. 2013; Guxens etal. 2012) (Section 5.4.4.1). Guxens etal. (2012) did not find
20 parental social class or education level to modify the relationship between NO2 and
21 mental development in infants, whereas Becerra etal. (2013) reported the strongest
22 associations between NO2 and autistic disorder in children born to mothers with the
23 lowest education level.
24 Across epidemiologic studies, there is consistent evidence demonstrating risk for NO2-
25 related mortality among individuals with low SES, while there is limited and inconsistent
26 evidence regarding cancer incidence and reproductive and developmental outcomes.
27 Interpreting this body of evidence is challenging given the wide variety of SES indicators
28 used across studies in addition to the breadth of countries where studies have been
29 conducted. Educational attainment was the most commonly used indicator of SES across
30 studies, and generally, lower levels of education were associated with NO2-related health
31 effects. Overall, the collective epidemiologic evidence suggests that low SES may
32 increase risk for NO2-related health effects.
November 2013 6-21 DRAFT: Do Not Cite or Quote
-------
6.4.3 Race/Ethnicity
1 Based on the 2010 U.S. Census, 63.7% of the U.S. population identified as non-Hispanic
2 whites; 12.6% reported their race as non-Hispanic black; and 16.3% reported being
3 Hispanic (Humes et al.. 2011). Race and ethnicity are complex risk factors that are often
4 closely related to other risk factors including genetics, diet, and socioeconomic status,
5 and both intrinsic and extrinsic mechanisms are likely to be involved in risk attributed to
6 race and ethnicity. While it can be difficult to understand the complexities of these
7 relationships, race/ethnicity is routinely examined as a risk factor for health outcomes and
8 has been studied in the context of NO2-related birth outcomes. Although NO2 exposure
9 during pregnancy was not consistently associated with birth weight across a multitude of
10 studies (Section 5.4.3.3), some studies that identified associations examined effect
11 modification by race (Darrow et al.. 201 Ib: Madsen et al.. 2010; Morello-Frosch et al..
12 2010; Bell et al., 2007). The results were varied as one study found NO2-associated
13 decreases in birth weight to be greatest for black mothers compared to white (Bell et al..
14 2007) while another study found NO2-associated decreases in birth weight to be greatest
15 for white (non-Hispanic) mothers, moderate for black (non-Hispanic), Asian (non-
16 Hispanic), and Pacific Islander mothers, and smallest for Hispanic mothers (Morello-
17 Frosch et al.. 2010). Additionally, Darrow et al. (201 Ib) found decreases in birth weight
18 associated with NO2 exposure in the 3rd trimester to be similar among white (non-
19 Hispanic), black (non-Hispanic), and Hispanic mothers. Rich et al. (2009) did find effect
20 modification by race for associations between NO2 and very small for gestational age;
21 however, risk estimates were greatest for Hispanic mothers compared to those for white
22 (non-Hispanic) and black (non-Hispanic). Further evidence is available from a study
23 conducted outside of the U.S. that did not find differences in associations of NO2 with
24 birth weight, low birth weight, or small for gestational age (SGA) among mothers of non-
25 Western or Western descent living in Oslo, Norway (Madsen et al.. 2010).
26 Beyond birth outcomes, only one study examined effect modification by race. Grineski et
27 al. (2010) examined whether race modified the association between NO2 exposure and
28 asthma hospital admission in children <14 years of age. Black children were at increased
29 risk of NO 2 -related asthma ED visits compared to Hispanic children, but no difference
30 was observed between black and white children.
31 The results from some individual studies demonstrate effect modification by race, but the
32 evidence across these studies is not consistent for any particular race/ethnicity
33 Furthermore, there are studies that demonstrate no effect modification for race. Overall,
34 this evidence is inadequate to determine whether race increases risk of NO2-related
35 health effects.
November 2013 6-22 DRAFT: Do Not Cite or Quote
-------
6.4.4 Sex
1 According to the 2010 U.S. Census, the distribution of males and females is
2 approximately equal: 49.2% male and 50.8% female (Howden and Meyer. 2011). This
3 distribution does vary by age with a greater prevalence of females above 65 years of age
4 compared to males. Additionally, sex is a potential risk factor for a vast number of health
5 conditions and diseases, including air pollutant-associated health effects. It is likely that
6 there are complex biological phenomena that underlie these differences in addition to
7 sex-based differences in exposure. A few studies have examined this more closely and
8 report mixed results. Kan et al. (2007). Kan et al. (2008). and Sunyer et al. (2006)
9 observed no difference in NO2 exposure between men and women in the multi-country
10 EXPOLIS study. Studies in children showed mixed results with respect to sex-related
11 differences in both proximity to roads and NO2 exposure, with some studies in Canada,
12 Europe, and Mexico showing larger effects in girls (Rosenlund et al.. 2009b; Oftedal et
13 al.. 2008; Rojas-Martinez et al.. 2007a; Luginaah et al.. 2005) and studies in Germany
14 and Los Angeles showing larger effects in boys (Gehring et al.. 2002; Peters etal.. 1999).
15 The inconsistency in NO2 exposure differences among men and women is mirrored in
16 studies examining sex-specific respiratory effects associated with NO2 exposure. Some
17 studies did not clearly indicate differences in associations between males and females
18 (Sarnat et al.. 2012; Liu et al.. 2009b; Lin et al.. 2005). Stronger associations of short-
19 term NO2 exposure with outcomes such as wheeze and asthma hospital admissions were
20 observed in boys with intermittent asthma or boys of low SES relative to girls (Mann et
21 al.. 2010; Lin et al.. 2004b) and stronger associations for outcomes such as respiratory
22 hospital admissions or ED visits for otitis media were observed in females (Zemek et al..
23 2010; Luginaah et al.. 2005). Inconsistency was also reported for respiratory effects of
24 long-term NO2 exposure. The association with asthma diagnosis was greater in females
25 (Kim et al.. 2004). although no differences in lung function growth were reported
26 between males and females (Rojas-Martinez et al.. 2007a).
27 Other long-term and short-term exposure studies examined associations between NO2
28 and mortality, and in general, indicate that females may be at greater risk. Some studies
29 (Cesaroni et al.. 2013; Katanoda et al.. 2011; Raaschou-Nielsen et al.. 2011; Raaschou-
30 Nielsen et al.. 2010a; Yorifuji et al.. 2010) did not observe differences between males and
31 females for NO2 -associated lung cancer incidence or mortality, but other studies
32 demonstrated that associations were significantly greater in females (Naess et al.. 2007;
33 Abbey etal.. 1999). Associations between long-term NO2 and cardiovascular or
34 respiratory mortality were not different among men or women in a few studies (Cesaroni
35 etal.. 2013; Naess et al.. 2007; Abbey etal.. 1999). while Katanoda etal. (2011) reported
36 respiratory mortality to be greater for females and Naess et al. (2007) reported COPD
November 2013 6-23 DRAFT: Do Not Cite or Quote
-------
1 mortality to be greater for males. In short-term studies, Chiusolo etal. (2011) found
2 males to be at slightly greater risk for NO2-related mortality, but more studies found
3 associations to be greater for females (Chen et al., 2012b; Cakmak et al., 201 Ib; Kan et
4 al.. 2008).
5 For other health outcomes, results did not clearly indicate greater NO 2 -associated risk in
6 males or females. The association with small for gestation age was greater in baby girls
7 than boys. The association between short-term NO2 exposure and the HRV measure
8 SDNN was stronger among females (Huang et al., 2012a).
9 Taken together, the results vary across studies, with recent evidence for increased risk for
10 NO2-related health effects present for males in some studies and females in others. The
11 majority of studies examining a sex being at increased or decreased risk for NO2-related
12 health effects focused on respiratory outcomes and did not clearly indicate increased risk
13 for either sex; however, a considerable number of epidemiologic studies examining NO2-
14 associated mortality demonstrate that females are at greater risk compared to males.
15 Overall, the collective evidence suggests that women may be at increased risk for NO2-
16 related health effects.
6.5 Behavioral and Other Factors
6.5.1 Diet
17 Diet is a plausible risk factor for health effects of air pollutants as it is an important
18 contributor to health status, but few epidemiologic studies have evaluated NO2-associated
19 health outcomes in the context of diet and the 2008 ISA for Oxides of Nitrogen (U.S.
20 EPA. 2008c) did not discuss diet as a risk factor; however, toxicological evidence is
21 available from the 2008 ISA and previous AQCDs that has primarily focused on
22 respiratory endpoints in animals deficient in or supplemented with Vitamin C and
23 Vitamin E.
24 Vitamin C, or ascorbic acid, can act as an antioxidant in the airways and neutralize
25 reactive oxygen species, and while epidemiologic studies are not available, a wealth of
26 toxicological studies have demonstrated that a diet rich is Vitamin C can mitigate NO2-
27 related oxidant injury. Guinea pigs with a Vitamin C-deficient diet had increased BALF
28 protein following NO2 exposure relative to air controls or guinea pigs with a normal diet
29 (Hatch etal.. 1986; Selgrade et al.. 1981). Rats with diets deficient in Vitamin E had
30 increases in lipid peroxidation and protein content in lung homogenates following NO2
31 exposure (Elsaved and Mustafa. 1982; Sevanian et al.. 1982b). Additional support for an
November 2013 6-24 DRAFT: Do Not Cite or Quote
-------
1 influence of Vitamin E is provided by observations that NO2-induced increases in BALF
2 protein or decreases in glutathione peroxidase activity were attenuated in animals on
3 Vitamin E-supplemented diets, relative to animals not supplemented with Vitamin E
4 (Guth and Mavis. 1986; Ayaz and Csallany. 1978). These studies demonstrate that
5 Vitamin C and E can modify the effects of NO2 on pulmonary injury in animals.
6 Limited, recent epidemiologic evidence stratifying results according to dietary
7 components includes two studies discussed in this ISA (Guxens et al.. 2012; Raaschou-
8 Nielsen et al.. 2011). Guxens et al. (2012) found the association between prenatal NO2
9 exposure during pregnancy and mental development in children during their second year
10 of life was greater among children with low maternal fruit and vegetable intake during
11 the first trimester of pregnancy. There was little evidence of an association among
12 mothers with higher fruit and vegetable intake. In addition, children of mothers with low
13 levels of circulating Vitamin D were also at greater risk for NO2-associated decrements
14 in mental development. Raaschou-Nielsen et al. (2011) found an association between the
15 highest quartile of NOX concentrations and lung cancer that was larger in individuals
16 with the lowest fruit intake compared with those in the highest quartile of fruit intake.
17 Although the limited epidemiologic evidence examined different outcomes, both
18 demonstrated response modification by dietary fruit or vegetable intake. The
19 toxicological evidence consistently demonstrated that Vitamin C or E dietary levels
20 affected pulmonary injury following NO2 exposure, but no epidemiologic or controlled
21 human exposure studies have examined these effects in humans. Coherence for the
22 animal toxicological evidence is provided by the well-characterized effect of NO2 in
23 inducing ROS/RNS (Section 3.3.2.1). Overall, the toxicological evidence suggests that
24 diet may modify response to NO2 and contribute to increased or decreased risk for health
25 effects associated with NO2 exposure.
6.5.2 Obesity
26 Obesity is defined as having a body mass index (BMI) of 30 kg/m2 or greater. It is an
27 issue of increasing importance as obesity rates in adults have continually increased over
28 several decades in the U.S., reaching an estimated 27% in 2010 (Schiller et al.. 2012).
29 Obesity, or BMI, may increase risk of NO2-related health effects through multiple
30 mechanisms including persistent, low-grade inflammation, increasing likelihood of
31 comorbidities, and poor diet.
32 The 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c) did not evaluate obesity as a
33 risk factor, but recent studies of NO2-related health effects allow for evaluation of effect
34 modification by obesity status. The majority of these studies have examined effect
November 2013 6-25 DRAFT: Do Not Cite or Quote
-------
1 modification of cardiovascular outcomes. There is biological plausibility for effect
2 modification given that weight and cardiovascular disease are related; however, the
3 evidence is inconsistent. In panel studies of individuals with pre-existing cardiovascular
4 disease, NO2 was associated with decreases in HRV but not ventricular
5 tachyarrhythmias; however, BMI did not modify either association (Huang etal. 2012a;
6 Ljungman et al.. 2008). In contrast, Baiaet al. (2010) found that the association between
7 NO2 and heart rate-corrected QT interval was greater among obese individuals compared
8 to non-obese individuals.
9 Animal evidence regarding effect modification by obesity is limited. Long-term exposure
10 to 160 ppb NO2 resulted in increased triglycerides and decreased in HDL in obese rats
11 compared to non-obese rats, though differences between strains were not observed at
12 higher NO2 exposures. This study suggests that obesity may contribute to NO2-induced
13 dyslipidemia, which is a known risk factor for cardiovascular disease.
14 Effect modification by obesity status has also been examined in the context of lung
15 cancer (Yorifuji et al.. 2010) and hypertensive disorders in pregnancy (Mobasher et al..
16 2013); associations between NO2 and these outcomes were not modified by obesity
17 status.
18 The epidemiologic and toxicological studies evaluating obesity as a risk factor are limited
19 in quantity, and results are inconsistent across studies. Overall, the evidence is inadequate
20 to determine whether obese individuals are at increased risk for NO2-associated health
21 effects.
6.5.3 Smoking
22 Smoking is a common behavior within the U.S. adult population as approximately 19.2%
23 of individuals report being current smokers and 21.5% report being a former smoker
24 (Schiller etal.. 2012). Smoking is a well-documented risk factor for a variety of diseases,
25 but it is unclear if smoking exacerbates health effects associated with air pollutant
26 exposures, including NO2. Only a few studies included in this ISA examined effect
27 modification by smoking status, and the limited amount of evidence is inconsistent.
28 In studies of lung cancer incidence, associations with ambient NO2 were not stronger
29 among smokers or nonsmokers (Raaschou-Nielsen et al.. 2011; Raaschou-Nielsen et al..
30 2010a). Studies evaluating lung cancer mortality reported contrasting results as Yorifuji
31 et al. (2010) found an increased hazard ratio among non-smokers and Katanoda et al.
32 (2011) found greater associations among male smokers and female nonsmokers.
November 2013 6-26 DRAFT: Do Not Cite or Quote
-------
1 Although many controlled human exposure studies report smoking status, comparisons
2 between smokers and non-smokers are infrequent due to small sample size. Despite this
3 limitation, Morrow et al. (1992) did analyze smoking subgroups and demonstrated that
4 among seven smoking subjects, the mean FEVi during a 4-hour exposure to 300 ppb
5 NO2 was significantly lower than the mean from 13 non-smoking subjects.
6 Taken together, evidence from both epidemiologic and controlled human exposure
7 studies is limited and inconsistent. Further, the epidemiologic evidence is provided by
8 studies of lung cancer, for which overall results are inconsistent. The limited and
9 inconsistent evidence is inadequate to determine whether smoking is a risk modifying
10 factor for NO2-related health effects.
6.5.4 Residential Location
11 A few epidemiologic studies examined health effects related to exposure to oxides of
12 nitrogen relative to residential location, specifically urban versus rural. Differences in
13 NO2 concentrations by residential location and building characteristics suggest that these
14 factors may contribute to variation in NO2 exposure. For example, Rotko et al. (2001)
15 found higher NO2 concentrations associated with downtown versus suburban location
16 (14.9 ppb versus 11.7 ppb), high-rise building versus single family home (13.1 ppb
17 versus 11.3 ppb), and older versus newer construction (before 1970: 14.5 ppb versus
18 during or after 1970: 11.5 ppb). Additionally, location of residence with respect to a busy
19 roadway can also affect exposure. Depending on atmospheric stability, NO2
20 concentrations can dilute with distance from the road or extend beyond 1 km of the
21 roadway (Section 2.5.3). The influence of roadway proximity on personal NO2 exposure
22 was demonstrated in a study finding the lowest correlations among personal, indoor, and
23 outdoor NO2 andNOx concentrations within a 500 meter buffer (r = 0.1-0.3) of a road
24 compared to 100 meter (r = 0.2-0.3) and 250 meter (r = 0.3-0.4) buffers (Schembari et al.,
25 2013). However, closest proximity to roadways was not the only determinant influencing
26 personal exposure as correlations were strongest for NO2 and the 250 meter buffer (r =
27 0.3-0.4), suggesting that NO2 is higher after some of the NO reacts photochemically to
28 become NO2. The topography of urban communities also may contribute to higher NO2
29 exposure as the presence of street canyons enhances mixing at elevations closer to the
30 street canyon-urban boundary layer interface, resulting in higher NO2 concentrations at
31 lower elevations (Section 2.5.3). This may have implications for higher exposures to
32 oxides of nitrogen for pedestrians, outdoor workers, and those living on lower levels.
33 Although the potential for higher exposure to NO2 in urban areas is well characterized,
34 epidemiologic comparisons of NO2-related health effects between urban and nonurban
November 2013 6-27 DRAFT: Do Not Cite or Quote
-------
1 residence are mixed. Further, these studies encompassed diverse outcomes: lung function
2 and respiratory symptoms, leukemia incidence, and autism. For respiratory effects and
3 leukemia incidence in children, no associations were found with NO2 in the entire study
4 populations or in analyses stratified by urban or rural residence in Italy (Ranzi et al..
5 2004) or results stratified by urban, semi-urban or rural residence in France (Amigou et
6 al.. 2011). In a study of autism in children in California, in the highest quartile of NO2
7 exposure in infancy, the association with autism was slightly lower among children living
8 in an urban area relative to children residing in rural areas (Volket al.. 2013).
9 While it has been demonstrated that urban residence can result in higher exposure to
10 oxides of nitrogen, the limited epidemiologic evidence does not consistently indicate
11 residential location to be a risk modifying factor. Because there are so few studies,
12 consistency within or across outcomes cannot be determined. Thus, the evidence is
13 inadequate to determine whether residential location modifies NO 2 -associated health
14 risks.
6.6 Summary
15 In this section, epidemiologic, exposure assessment, controlled human exposure, and
16 toxicological studies have been evaluated and indicate that various factors may lead to
17 increased risk of NO2-related health effects (Table 6-4)
18 The evaluation of evidence in this section demonstrates that there is adequate evidence to
19 conclude that different lifestages may be at increased risk for health effects related to
20 exposure to oxides of nitrogen. Moreover, both children and older adults are potentially
21 at greater risk for NO2-related health effects. Recent evidence demonstrates that children
22 are at increased risk for NO2-associated asthma hospitalizations and ED visits, which is
23 consistent with conclusions from the 2008 ISA for Oxides of Nitrogen (U.S. EPA.
24 2008c). There is limited toxicological evidence to provide biological plausibility, but
25 these observations are plausible based on studies demonstrating differences in NO2
26 exposure among children in addition to higher ventilation rates. There is also consistent
27 and substantial evidence showing larger associations in older adults between NO2
28 exposure and hospitalization and mortality, though limited evidence of respiratory
29 outcomes from controlled human exposures is not coherent with these observations.
November 2013 6-28 DRAFT: Do Not Cite or Quote
-------
Table 6-4 Summary of evidence for potential increased risk of health effects
related to exposure to oxides of nitrogen.
Evidence Classification Potential At Risk Factor
Adequate evidence Lifestage (Children [Section 6.4.1.11 and Older adults [Section 6.4.1.21
Suggestive evidence Genetic background (Section 6.2)
Asthma (Section 6.3.1)
COPD (Section 6.3.2)
SES (Section 6.4.2)
Sex (Section 6.4.4)
Diet (Section 6.5.1)
Inadequate evidence Cardiovascular disease (Section 6.3.3)
Diabetes (Section 6.3.4)
Race/Ethnicity (Section 6.4.3)
Obesity(Section 6.5.2)
Smoking (Section 6.5.3)
Residential location (Section 6.5.4)
Evidence of no effect
1 For several other factors, including genetic background, pre-existing asthma or COPD,
2 SES, sex, and diet, it was determined that there is suggestive evidence of increased risk
3 for respective populations as presented in Table 6-4. Studies have indicated increased risk
4 of exposure to oxides of nitrogen with declining measures of SES. For all of these
5 factors, there is a body of evidence for an outcome or scientific discipline indicating
6 differential response to NO2; however, there are inconsistencies within or across
7 outcomes or scientific disciplines that present uncertainties in the interpretation. In
8 contrast to conclusions from the 2008 ISA for Oxides of Nitrogen (U.S. EPA. 2008c).
9 recent epidemiologic evidence provides inconsistent evidence that pre-existing asthma
10 increases risk for hospitalizations or ED visits related to ambient NO2; however, meta-
11 analyses of controlled human exposure studies of AHR demonstrate that individuals with
12 asthma are more sensitive to NO2. Consistent with previous observations, individuals of
13 low SES may be at greater risk for NO 2-related health effects, but limitations exist due to
14 variations in definitions of SES and indicators of SES within and across different
15 countries. Furthermore, epidemiologic evidence demonstrates that females appear to be at
16 greater risk for NO2-related mortality, though there are inconsistencies for other
17 outcomes relative to NO2. Animal studies provide a wealth of evidence that diets rich in
18 Vitamin C or E may protect against respiratory effects of NO2, while those deficient in
19 these vitamins may be at greater risk for NO2-induced pulmonary injury and
20 inflammation.
November 2013 6-29 DRAFT: Do Not Cite or Quote
-------
1 In addition, there was inadequate evidence to determine increased risk for pre-existing
2 cardiovascular disease, diabetes, race/ethnicity, obesity, smoking, and residential location
3 (Table 6-4). For the majority of these factors, there was limited evidence to allow for
4 evaluation of consistency or coherence.
5 Overall, the available evidence evaluated in this ISA in combination with previous
6 evidence from the 2008 ISA for Oxides of Nitrogen demonstrates that lifestage is an
7 important contributor to increased risk of NO 2 -related health effects.
November 2013 6-30 DRAFT: Do Not Cite or Quote
-------
References for Chapter 6
Abbey. DE; Nishino. N; McDonnell. WF; Burchette. RJ: Knutsen. SF; Beeson. WL; Yang. JX. (1999).
Long-term inhalable particles and other air pollutants related to mortality in nonsmokers. Am J Respir
Crit Care Med 159: 373-382. http://dx.doi.org/10.1164/airccm. 159.2.9806020
Amigou. A; Sermage-Faure. C: Orsi. L; Leverger. G: Baruchel. A; Bertrand. Y; Nelken. B; Robert. A;
Michel. G: Margueritte. G: Perel. Y; Mechinaud. F; Bordigoni. P; Hemon. D; Clavel. J. (2011). Road
traffic and childhood leukemia: The ESCALE Study (SFCE). Environ Health Perspect 119: 566-572.
http://dx.doi.org/10.1289/ehp. 1002429
Arbex. MA; de Souza Conceicao. GM; Cendon. SP; Arbex. FF; Lopes. AC: Moyses. EP; Santiago. SL;
Saldiva. PHN; Pereira. LAA; Braga. ALF. (2009). Urban air pollution and chronic obstructive
pulmonary disease-related emergency department visits. J Epidemiol Community Health 63: 777-783.
http://dx.doi.org/10.1136/iech.2008.078360
Ayaz. KL; Csallany. AS. (1978). Long-term NO2 exposure of mice in the presence and absence of vitamin
E. II. Effect of glutathione peroxidase. Arch Environ Occup Health 33: 292-296.
Azoulav-Dupuis. E; Torres. M; Soler. P; Moreau. J. (1983). Pulmonary NO2 toxicity in neonate and adult
guinea pigs and rats. Environ Res 30: 322-339.
Baja. ES: Schwartz. JD; Wellenius. GA; Coull. BA; Zanobetti. A; Vokonas. PS: Suh. HH. (2010). Traffic-
related air pollution and QT interval: Modification by diabetes, obesity, and oxidative stress gene
polymorphisms in the Normative Aging Study. Environ Health Perspect 118: 840-846.
http://dx.doi.org/10.1289/ehp.0901396
Barck. C: Sandstrom. T; Lundahl. J: Hallden. G: Svartengren. M; Strand. V: Rak. S: Bylin. G. (2002).
Ambient level of NO2 augments the inflammatory response to inhaled allergen in asthmatics. Respir
Med 96: 907-917.
Barraza-Villarreal. A; Sunyer. J: Hernandez-Cadena. L; Escamilla-Nunez. MC: Sienra-Monge. JJ: Ramirez-
Aguilar. M; Cortez-Lugo. M; Holguin. F; Diaz-Sanchez. D; Olin. AC: Romieu. I. (2008). Air pollution,
airway inflammation, and lung function in a cohort study of Mexico City schoolchildren. Environ
Health Perspect 116: 832-838. http://dx.doi.org/10.1289/ehp.10926
Bauer. MA; Utell. MJ; Morrow. PE; Speers. DM; Gibb. FR. (1986). Inhalation of 0.30 ppm nitrogen
dioxide potentiates exercise-induced bronchospasm in asthmatics. Am Rev Respir Dis 134: 1203-1208.
Becerra. TA; Wilhelm. M; Olsen. J: Cockburn. M; Ritz. B. (2013). Ambient air pollution and autism in Los
Angeles county, California. Environ Health Perspect 121: 380-386.
http://dx.doi.org/10.1289/ehp. 1205827
Bell. ML; Ebisu. K; Belanger. K. (2007). Ambient air pollution and low birth weight in Connecticut and
Massachusetts. Environ Health Perspect 115: 1118-1124. http://dx.doi.org/10.1289/ehp.9759
Berglind. N; Bellander. T; Forastiere. F; von Klot. S: Aalto. P: Elosua. R; Kulmala. M; Lanki. T; Lowel. H;
Peters. A; Picciotto. S: Salomaa. V: Stafoggia. M; Sunyer. J: Nyberg. F. (2009). Ambient air pollution
and daily mortality among survivors of myocardial infarction. Epidemiology 20: 110-118.
http://dx.doi.org/10.1097/EDE.Ob013e3181878b50
Berhane. K; Zhang. Y; Linn. WS: Rappaport. EB; Bastain. TM; Salam. MT; Islam. T; Lurmann. F;
Gilliland. FD. (2011). The effect of ambient air pollution on exhaled nitric oxide in the Children's
Health Study. Eur Respir J 37: 1029-1036. http://dx.doi.org/10.1183/09031936.00081410
Bishaw. A. (2012). Poverty: 2010 and 2011. Bishaw, A. http://www.census.gov/prod/2012pubs/acsbr 11 -
Ol.pdf
November 2013 6-31 DRAFT: Do Not Cite or Quote
-------
Bloom. B; Cohen. RA; Freeman. G. (2011). Summary health statistics for U.S. children: National Health
Interview Survey, 2010. Vital Health Stat 10 0: 1-80.
Breton. CV: Salam. MT; Vora. H; Gauderman. WJ: Gilliland. FD. (2011). Genetic variation in the
glutathione synthesis pathway, air pollution, and children's lung function growth. Am J Respir Crit Care
Med 183: 243-248. http://dx.doi.org/10.1164/rccm.201006-0849OC
Burra. TA; Moineddin. R; Agha. MM; Glazier. RH. (2009). Social disadvantage, air pollution, and asthma
physician visits in Toronto, Canada. Environ Res 109: 567-574.
http://dx.doi.0rg/10.1016/i.envres.2009.03.004
Cakmak. S: Dales. RE; Angelica Rubio. M; Blanco Vidal. C. (20 lib). The risk of dying on days of higher
air pollution among the socially disadvantaged elderly. Environ Res 111: 388-393.
http://dx.doi.0rg/10.1016/i.envres.2011.01.003
Campen. MJ; Lund. AK; Doyle-Eisele. ML; McDonald. JD; Knuckles. TL; Rohr. AC: Knipping. EM;
Mauderly. JL. (2010). A comparison of vascular effects from complex and individual air pollutants
indicates a role for monoxide gases and volatile hydrocarbons. Environ Health Perspect 118: 921-927.
httrj://dx.doi.org/10.1289/ehp.09012Q7
Castro-Giner. F; Kiinzli. N; Jacquemin. B; Forsberg. B; de Cid. R; Sunyer. J; Jarvis. D; Briggs. D;
Vienneau. D; Norback. D; Gonzalez. JR; Guerra. S: Janson. C: Anto. JM; Wist. M; Heinrich. J; Estivill.
X; Kogevinas. M. (2009). Traffic-related air pollution, oxidative stress genes, and asthma (ECHRS).
Environ Health Perspect 117: 1919-1924.
Cesaroni. G: Badaloni. C: Gariazzo. C: Stafoggia. M; Sozzi. R; Davoli. M; Forastiere. F. (2013). Long-term
exposure to urban air pollution and mortality in a cohort of more than a million adults in rome. Environ
Health Perspect 121: 324-331. http://dx.doi.org/10.1289/ehp.1205862
Chaix. B; Gustafsson. S: Jerrett. M; Kristersson. H; Lithman. T; Boalt. A; Merlo. J. (2006). Children's
exposure to nitrogen dioxide in Sweden: Investigating environmental injustice in an egalitarian country.
J Epidemiol Community Health 60: 234-241. http://dx.doi.org/10.1136/iech.2005.038190
Chen. R; Samoli. E; Wong. CM; Huang. W: Wang. Z; Chen. B; Kan. H. (2012b). Associations between
short-term exposure to nitrogen dioxide and mortality in 17 Chinese cities: The China Air Pollution and
Health Effects Study (CAPES). Environ Int 45: 32-38. http://dx.doi.Org/10.1016/i.envint.2012.04.008
Chiusolo. M; Cadum. E; Stafoggia. M; Galassi. C: Berti. G: Faustini. A; Bisanti. L; Vigotti. MA; Dessi.
MP; Cernigliaro. A; Mallone. S: Pacelli. B; Minerba. S: Simonato. L; Forastiere. F. (2011). Short term
effects of nitrogen dioxide on mortality and susceptibility factors in 10 Italian cities: The EpiAir Study.
Environ Health Perspect 119: 1233-1238. http://dx.doi.org/10.1289/ehp.1002904
Cloughertv. JE; Lew. JI; Kubzansky. LD; Ryan. PB; Suglia. SF; Canner. MJ: Wright. RJ. (2007).
Synergistic effects of traffic-related air pollution and exposure to violence on urban asthma etiology.
Environ Health Perspect 115: 1140-1146. http://dx.doi.org/10.1289/ehp.9863
D'Ippoliti. D: Forastiere. F; Ancona. C: Agabiti. N; Fusco. D: Michelozzi. P; Perucci. CA. (2003). Air
pollution and myocardial infarction in Rome: a case-crossover analysis. Epidemiology 14: 528-535.
http://dx.doi.org/10.1097/01.ede.0000082046.22919.72
Darrow. LA; Klein. M; Strickland. MJ: Mulholland. JA; Tolbert. PE. (20 lib). Ambient air pollution and
birth weight in full-term infants in Atlanta, 1994-2004. Environ Health Perspect 119: 731-737.
http://dx.doi.org/10.1289/ehp.1002785
Deguen. S: Zmirou-Navier. D. (2010). Social inequalities resulting from health risks related to ambient air
quality—A European review [Review]. Eur J Public Health 20: 27-35.
http://dx.doi.org/10.1093/eurpub/ckp220
Delfino. RJ: Staimer. N: Gillen. D; Tjoa. T; Sioutas. C: Fung. K; George. SC: Kleinman. MT. (2006).
Personal and ambient air pollution is associated with increased exhaled nitric oxide in children with
asthma. Environ Health Perspect 114: 1736-1743.
November 2013 6-32 DRAFT: Do Not Cite or Quote
-------
Delfino. RJ; Staimer. N; Tjoa. T; Gillen. D; Kleinman. MT; Sioutas. C: Cooper. D. (2008a). Personal and
ambient air pollution exposures and lung function decrements in children with asthma. Environ Health
Perspect 116: 550-558. http://dx.doi.org/10.1289/ehp.10911
Elsaved. NM; Mustafa. MG. (1982). Dietary antioxidants and the biochemical response to oxidant
inhalation: I: Influence of dietary vitamin E on the biochemical effects of nitrogen dioxide exposure in
rat lung. Toxicol Appl Pharmacol 66: 319-328. http://dx.doi.org/10.1016/0041-008X(82)90298-8
Faustini. A; Stafoggia. M; Colais. P; Berti. G: Bisanti. L; Cadum. E; Cernigliaro. A; Mallone. S: Scarnato.
C: Forastiere. F. (2013). Air pollution and multiple acute respiratory outcomes. Eur Respir J 42: 304-
313. http://dx.doi.org/10.1183/09031936.00128712
Flamant-Hulin. M; Caillaud. D: Sacco. P; Penard-Morand. C: Annesi-Maesano. I. (2010). Air pollution and
increased levels of fractional exhaled nitric oxide in children with no history of airway damage. J
Toxicol Environ Health A 73: 272-283. http://dx.doi.org/10.1080/15287390903249206
Fu. A; Leaderer. BP; Gent. JF; Leaderer. D; Zhu. Y. (2012a). An environmental epigenetic study of ADRB2
5'-UTR methylation and childhood asthma severity. Clin Exp Allergy 42: 1575-1581.
http://dx.doi.0rg/10.llll/i.1365-2222.2012.04055.x
Gauderman. WJ: Avol. E; Gilliland. F; Vora. H; Thomas. D: Berhane. K; Mcconnell. R; Kuenzli. N;
Lurmann. F; Rappaport. E; Margolis. H; Bates. D: Peters. J. (2004). The effect of air pollution on lung
development from lOto 18 years of age. NEngl JMed 351: 1057-1067.
http://dx.doi.org/10.1056/NEJMoa040610
Gehring. U; Cyrys. J: Sedlmeir. G: Brunekreef. B; Bellander. T; Fischer. P; Bauer. CP; Reinhardt. D:
Wichmann. HE; Heinrich. J. (2002). Traffic-related air pollution and respiratory health during the first 2
yrs of life. Eur Respir J 19: 690-698. http://dx.doi.org/10.1183/09031936.02.01182001
Gong. H; Linn. WS: Clark. KW: Anderson. KR; Geller. MD; Sioutas. C. (2005). Respiratory responses to
exposures with fine particulates and nitrogen dioxide in the elderly with and without COPD. Inhal
Toxicol 17: 123-132. http://dx.doi.org/10.1080/08958370590904481
Grineski. SE; Staniswalis. JG: Peng. Y; Atkinson-Palombo. C. (2010). Children's asthma hospitalizations
and relative risk due to nitrogen dioxide (NO2): Effect modification by race, ethnicity, and insurance
status. Environ Res 110: 178-188. http://dx.doi.Org/10.1016/i.envres.2009.10.012
Gruzieva. O: Bellander. T; Eneroth. K; Kull. I; Melen. E; Nordling. E; van Hage. M; Wickman. M;
Moskalenko. V; Hulchiy. O; Pershagen. G. (2012). Traffic-related air pollution and development of
allergic sensitization in children during the first 8 years of life. J Allergy Clin Immunol 129: 240-246.
http://dx.doi.0rg/10.1016/i.jaci.2011.ll.001
Gruzieva. O: Bergstrom. A; Hulchiy. O: Kull. I; Lind. T; Melen. E; Moskalenko. V; Pershagen. G:
Bellander. T. (2013). Exposure to air pollution from traffic and childhood asthma until 12 years of age.
Epidemiology 24: 54-61. http://dx.doi.org/10.1097/EDE.Ob013e318276clea
Guth. DJ; Mavis. RD. (1986). The effect of lung alpha-tocopherol content on the acute toxicity of nitrogen
dioxide. Toxicol Appl Pharmacol 84: 304-314. http://dx.doi.org/10.1016/0041-008X(86)90138-9
Guxens. M; Aguilera. I; Ballester. F; Estarlich. M; Fernandez-Somoano. A; Lertxundi. A; Lertxundi. N;
Mendez. MA; Tardon. A; Vrijheid. M; Sunyer. J. (2012). Prenatal exposure to residential air pollution
and infant mental development: Modulation by antioxidants and detoxification factors. Environ Health
Perspect 120: 144-149. http://dx.doi.org/10.1289/ehp.1103469
Hatch. GE; Slade. R; Selgrade. MK; Stead. AG. (1986). Nitrogen dioxide exposure and lung antioxidants in
ascorbic acid-deficient guinea pigs. Toxicol Appl Pharmacol 82: 351-359.
http://dx.doi.org/10.1016/0041-008X(86)90212-7
Hernandez-Cadena. L; Holguin. F; Barraza-Villarreal. A; Del Rio-Navarro. BE; Sienra-Monge. JJ: Romieu.
L (2009). Increased levels of outdoor air pollutants are associated with reduced bronchodilation in
children with asthma. Chest 136: 1529-1536. http://dx.doi.org/10.1378/chest.08-1463
November 2013 6-33 DRAFT: Do Not Cite or Quote
-------
Hizawa. N. (2011). Pharmacogenetics of p2-agonists [Review]. Allergol Int 60: 239-246.
http://dx.doi.org/10.2332/allergolint.ll-RAI-0317
Holguin. F; Flores. S: Ross. Z; Cortez. M; Molina. M; Molina. L; Rincon. C: Jerrett. M; Berhane. K;
Granados. A; Romieu. I. (2007). Traffic-related exposures, airway function, inflammation, and
respiratory symptoms in children. Am J Respir Crit Care Med 176: 1236-1242.
http://dx.doi.org/10.1164/rccm.200611-1616OC
Howden. LM; Meyer. JA. (2011). Age and sex composition: 2010. (2010 Census Briefs, C2010BR-03).
Washington, DC: U.S. Department of Commerce, Economics and Statistics Administration, U.S. Census
Bureau, http://www.census.gov/prod/cen2010/briefs/c201 Obr-03.pdf
Hovert. PL; Xu. J. (2012). Deaths: Preliminary data for 2011. US Department of Health and Human
Services National Center for Health Statistics National Vital Statistics Reports 61: 1-51.
Huang. W: Zhu. T; Pan. X; Hu. M; Lu. SB; Lin. Y; Wang. T; Zhang. Y; Tang. X. (2012a). Air pollution and
autonomic and vascular dysfunction in patients with cardiovascular disease: Interactions of systemic
inflammation, overweight, and gender. Am J Epidemiol 176: 117-126.
http://dx.doi.org/10.1093/aie/kwr511
Humes. KR; Jones. NA; Ramirez. RR. (2011). Overview of race and Hispanic origin: 2010. (2010 Census
Briefs, C2010BR-02). Washington, DC: U.S. Department of Commerce, Economics and Statistics
Administration, U.S. Census Bureau, http://www.census.gov/prod/cen2010/briefs/c201 Obr-02.pdf
Jenkins. HS: Devalia. JL; Davies. RJ. (1997). The effect of ozone, nitrogen dioxide or their combination on
the airway response of mild asthmatics to inhaled allergen. Eur Respir J 10: 274S.
Jorres. R; Magnussen. H. (1991). Effect of 0.25 ppm nitrogen dioxide on the airway response to
methacholine in asymptomatic asthmatic patients. Lung 169: 77-85.
Jorres. R; Nowak. D; Grimminger. F; Seeger. W: Oldigs. M; Magnussen. H. (1995). The effect of 1 ppm
nitrogen dioxide on bronchoalveolar lavage cells and inflammatory mediators in normal and asthmatic
subjects. Eur Respir J 8: 416-424. http://dx.doi.org/10.1183/09031936.95.08030416
Kan. H; Heiss. G: Rose. KM; Whitsel. E; Lurmann. F; London. SJ. (2007). Traffic exposure and lung
function in adults: The Atherosclerosis Risk in Communities study. Thorax 62: 873-879.
http://dx.doi.org/10.1136/thx.2006.073015
Kan. H; London. SJ: Chen. G: Zhang. Y; Song. G: Zhao. N; Jiang. L; Chen. B. (2008). Season, sex, age,
and education as modifiers of the effects of outdoor air pollution on daily mortality in Shanghai, China:
The Public Health and Air Pollution in Asia (PAPA) Study. Environ Health Perspect 116: 1183-1188.
http://dx.doi.org/10.1289/ehp.10851
Katanoda. K; Sobue. T; Satoh. H; Taiima. K; Suzuki. T; Nakatsuka. H; Takezaki. T; Nakavama. T; Nitta. H;
Tanabe. K; Tominaga. S. (2011). An association between long-term exposure to ambient air pollution
and mortality from lung cancer and respiratory diseases in Japan. J Epidemiol 21: 132-143.
http://dx.doi.org/10.2188/iea.JE20100098
Kim. JJ; Smorodinsky. S: Lipsett. M; Singer. BC: Hodgson. AT; Ostro. B. (2004). Traffic-related air
pollution near busy roads: the East Bay children's Respiratory Health Study. Am J Respir Crit Care Med
170: 520-526. http://dx.doi.org/10.1164/rccm.200403-281OC
Kim. SY; O'Neill. MS: Lee. JT; Cho. Y; Kim. J: Kim. H. (2007). Air pollution, socioeconomic position,
and emergency hospital visits for asthma in Seoul, Korea. Int Arch Occup Environ Health 80: 701-710.
http://dx.doi.org/10.1007/s00420-007-0182-3
Kleinman. MT; Bailey. RM; Linn. WS: Anderson. KR; Whynot. JD; Shamoo. DA; Hackney. JD. (1983).
Effects of 0.2 ppm nitrogen dioxide on pulmonary function and response to bronchoprovocation in
asthmatics. J Toxicol Environ Health 12: 815-826. http://dx.doi.org/10.1080/15287398309530472
Ko. FWS: Tarn. W: Wong. TW: Lai. CKW (2007b). Effects of air pollution on asthma hospitalization rates
in different age groups in Hong Kong. Clin Exp Allergy 37: 1312-1319.
http://dx.doi.0rg/10.llll/i.1365-2222.2007.02791.x
November 2013 6-34 DRAFT: Do Not Cite or Quote
-------
Kousa. A; Monn. C: Rotko. T; Aim. S: Oblesbv. L; Jantunen. MJ. (2001). Personal exposures to NO2 in the
EXPOLIS-study: Relation to residential indoor, outdoor and workplace concentrations in Basel,
Helsinki and Prague. Atmos Environ 35: 3405-3412.
Kruize. H; Driessen. PP; Glasbergen. P; van Egmond. KN. (2007). Environmental equity and the role of
public policy: Experiences in the Rijnmond region. Environ Manage 40: 578-595.
http://dx.doi.org/10.1007/s00267-005-0378-9
Kumae. T; Arakawa. H. (2006). Comparison of effects of in vivo nitrogen dioxide exposure starting from
different periods on alveolar macrophage activity, assessed by a chemiluminescence technique in
Brown-Norway rats. Luminescence 21: 226-232. http://dx.doi.org/10.1002/bio.911
Laurent. O: Pedrono. G: Filleul. L; Segala. C: Lefranc. A; Schillinger. C: Riviere. E; D. B. (2009).
Influence of socioeconomic deprivation on the relation between air pollution and beta-agonist sales for
asthma. Chest 135: 717-723. http://dx.doi.org/10.1378/chest.08-1604
Lin. M; Chen. Y; Villeneuve. PJ: Burnett. RT; Lemyre. L; Hertzman. C: Mcgrail. KM; Krewski. D.
(2004b). Gaseous air pollutants and asthma hospitalization of children with low household income in
Vancouver, British Columbia, Canada. Am J Epidemiol 159: 294-303.
http://dx.doi.org/10.1093/aie/kwh043
Lin. M; Stieb. DM; Chen. Y. (2005). Coarse paniculate matter and hospitalization for respiratory infections
in children younger than 15 years in Toronto: A case-crossover analysis. Pediatrics 116: 235-240.
http://dx.doi.org/10.1542/rjeds.2004-2012
Lin. W: Huang. W: Zhu. T; Hu. M; Brunekreef. B; Zhang. Y; Liu. X; Cheng. H; Gehring. U; Li. C: Tang.
2L (2011). Acute respiratory inflammation in children and black carbon in ambient air before and during
the 2008 Beijing Olympics. Environ Health Perspect 119: 1507-1512.
http://dx.doi.org/10.1289/ehp.1103461
Linn. WS: Shamoo. DA; Spier. CE; Valencia. LM; Anzar. UT; Venet. TG: Avol. EL; Hackney. JD. (1985a).
Controlled exposure of volunteers with chronic obstructive pulmonary disease to nitrogen dioxide. Arch
Environ Health 40: 313-317. http://dx.doi.org/10.1080/00039896.1985.10545939
Linn. WS: Solomon. JC: Trim. SC: Spier. CE; Shamoo. DA; Venet. TG: Avol. EL; Hackney. JD. (1985b).
Effects of exposure to 4 ppm nitrogen dioxide in healthy and asthmatic volunteers. Arch Environ Occup
Health 40: 234-239.
Liu. L; Poon. R; Chen. L; Frescura. AM; Montuschi. P; Ciabattoni. G: Wheeler. A; Dales. R. (2009b).
Acute effects of air pollution on pulmonary function, airway inflammation, and oxidative stress in
asthmatic children. Environ Health Perspect 117: 668-674. http://dx.doi.org/10.1289/ehpll813
Ljungman. PLS: Berglind. N; Holmgren. C: Gadler. F; Edvardsson. N; Pershagen. G: Rosenqvist. M;
Sjogren. B; Bellander. T. (2008). Rapid effects of air pollution on ventricular arrhythmias. Eur Heart J
29: 2894-2901. http://dx.doi.org/!0.1093/eurheartj/ehn463
Luginaah. IN; Fung. KY; Gorev. KM; Webster. G: Wills. C. (2005). Association of ambient air pollution
with respiratory hospitalization in a government-designated "area of concern": The case of Windsor,
Ontario. Environ Health Perspect 113: 290-296. http://dx.doi.org/10.1289/ehp.7300
Madsen. C: Gehring. U; Walker. SE; Brunekreef. B; Stigum. H; Naess. O: Nafstad. P. (2010). Ambient air
pollution exposure, residential mobility and term birth weight in Oslo, Norway. Environ Res 110: 363-
371. http://dx.doi.0rg/10.1016/i.envres.2010.02.005
Mann. JK; Balmes. JR; Bruckner. TA; Mortimer. KM; Margolis. HG: Pratt. B; Hammond. SK; Lurmann. F;
Tager. IB. (2010). Short-term effects of air pollution on wheeze in asthmatic children in Fresno,
California. Environ Health Perspect 118: 1497-1502. http://dx.doi.org/10.1289/ehp.0901292
Mann. JK; Tager. IB; Lurmann. F; Segal. M; Quesenberrv. CP. Jr; Lugg. MM; Shan. J: Van den Eeden. SK.
(2002). Air pollution and hospital admissions for ischemic heart disease in persons with congestive
heart failure or arrhythmia. Environ Health Perspect 110: 1247-1252.
November 2013 6-35 DRAFT: Do Not Cite or Quote
-------
Melen. E; Nyberg. F; Lindgren. CM; Berglind. N; Zucchelli. M; Nordling. E; Hallberg. J; Svartengren. M;
Morgenstern. R; Kere. J: Bellander. T; Wickman. M; Pershagen. G. (2008). Interactions between
glutathione S-transferase PI, tumor necrosis factor, and traffic-related air pollution for development of
childhood allergic disease. Environ Health Perspect 116: 1077-1084.
http://dx.doi.org/10.1289/ehp.11117
Meng. Q; Williams. R; Pinto. JP (2012a). Determinants of the associations between ambient concentrations
and personal exposures to ambient PM2.5, NO2, and O3 during DEARS. Atmos Environ 63: 109-116.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.09.019
Min. KB; Min. JY; Cho. SI; Paek. D. (2008). The relationship between air pollutants and heart-rate
variability among community residents in Korea. Inhal Toxicol 20: 435-444.
http://dx.doi.org/10.1080/08958370801903834
Mitchell. G. (2005). Forecasting environmental equity: Air quality responses to road user charging in
Leeds, UK. J Environ Manage 77: 212-226. http://dx.doi.Org/10.1016/j.jenvman.2005.04.013
Mobasher. Z; Salam. MT; Goodwin. TM; Lurmann. F; Ingles. SA; Wilson. ML. (2013). Associations
between ambient air pollution and Hypertensive Disorders of Pregnancy. Environ Res 123: 9-16.
http://dx.doi.0rg/10.1016/i.envres.2013.01.006
Molter. A; Lindlev. S: de Vocht. F; Agius. R: Kerry. G: Johnson. K; Ashmore. M; Terry. A;
Dimitroulopoulou. S: Simpson. A. (2012). Performance of a microenviromental model for estimating
personal NO2 exposure in children. Atmos Environ 51: 225-233.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.01.030
Morello-Frosch. R; Jesdale. BM; Sadd. JL; Pastor. M. (2010). Ambient air pollution exposure and full-term
birth weight in California. Environ Health 9: 44. http://dx.doi.org/10.1186/1476-069X-9-44
Morrow. PE; Utell. MJ; Bauer. MA; Smeglin. AM; Frampton. MW: Cox. C: Speers. DM; Gibb. FR. (1992).
Pulmonary performance of elderly normal subjects and subjects with chronic obstructive pulmonary
disease exposed to 0.3 ppm nitrogen dioxide. Am J Respir Crit Care Med 145: 291-300.
http://dx.doi.0rg/10.1164/airccm/145.2 Pt 1.291
Naess. O: Nafstad. P; Aamodt. G: Claussen. B; Rosland. P. (2007). Relation between concentration of air
pollution and cause-specific mortality: four-year exposures to nitrogen dioxide and paniculate matter
pollutants in 470 neighborhoods in Oslo, Norway. Am J Epidemiol 165: 435-443.
http://dx.doi.org/10.1093/aie/kwk016
Nishimura. KK; Galanter. JM; Roth. LA; Oh. SS: Thakur. N; Nguyen. EA; Thyne. S: Farber. HJ;
Serebrisky. D; Kumar. R; Brigino-Buenaventura. E; Davis. A; LeNoir. MA; Meade. K; Rodriguez-
Cintron. W: Avila. PC: Borrell. LN; Bibbins-Domingo. K; Rodriguez-Santana. JR; Sen. S: Lurmann. F;
Balmes. JR; Burchard. EG. (2013a). Early-life air pollution and asthma risk in minority children: The
GALA II and SAGE II studies. Am J Respir Crit Care Med 188: 309-318.
http://dx.doi.org/10.1164/rccm.201302-0264OC
O'Neill. MS: Jerrett. M; Kawachi. I; Lew. JL Cohen. AJ; Gouveia. N; Wilkinson. P; Fletcher. T; Cifuentes.
L; Schwartz. J. (2003). Health, wealth, and air pollution: Advancing theory and methods [Review].
Environ Health Perspect 111: 1861-1870.
Oftedal. B; Brunekreef. B; Nystad. W: Madsen. C: Walker. SE; Nafstad. P. (2008). Residential outdoor air
pollution and lung function in schoolchildren. Epidemiology 19: 129-137.
http://dx.doi.org/10.1097/EDE.Ob013e31815c0827
Patel. MM; Chillrud. SN; Correa. JC: Hazi. Y; Feinberg. M; Kc. D: Prakash. S: Ross. JM; Lew. D: Kinnev.
PL. (2010). Traffic-related paniculate matter and acute respiratory symptoms among New York City
area adolescents. Environ Health Perspect 118: 1338-1343. http://dx.doi.org/10.1289/ehp.0901499
Peel. JL: Metzger. KB; Klein. M; Flanders. WD; Mulholland. JA; Tolbert. PE. (2007). Ambient air
pollution and cardiovascular emergency department visits in potentially sensitive groups. Am J
Epidemiol 165: 625-633. http://dx.doi.org/10.1093/aie/kwk051
November 2013 6-36 DRAFT: Do Not Cite or Quote
-------
Pereira. G: Cook. AG: Haggar. F; Bower. C: Nassar. N. (2012). Locally derived traffic-related air pollution
and fetal growth restriction: A retrospective cohort study. Occup Environ Med 69: 815-822.
http://dx.doi.org/10.1136/oemed-2011-100509
Peters. JM; Avol. E; Gauderman. WJ: Linn. WS: Navidi. W: London. SJ: Margolis. H; Rappaport. E; Vora.
H; Gong. H. Jr; Thomas. DC. (1999). A study of twelve southern California communities with differing
levels and types of air pollution: II. Effects on pulmonary function. Am J Respir Crit Care Med 159:
768-775. http://dx.doi.0rg/10.1164/airccm.159.3.9804144
Qian. Z; Lin. HM; Chinchilli. VM; Lehman. EB; Duan. Y; Craig. TJ; Wilson. WE; Liao. D; Lazarus. SC:
Bascom. R. (2009a). Interaction of ambient air pollution with asthma medication on exhaled nitric oxide
among asthmatics. Arch Environ Occup Health 64: 168-176.
http://dx.doi.org/10.1080/19338240903240616
Qian. Z; Lin. HM; Chinchilli. VM; Lehman. EB; Stewart. WF; Shah. N; Duan. Y; Craig. TJ: Wilson. WE;
Liao. D; Lazarus. SC: Bascom. R. (2009b). Associations between air pollution and peak expiratory flow
among patients with persistent asthma. J Toxicol Environ Health A 72: 39-46.
http://dx.doi.org/10.1080/15287390802445517
Raaschou-Nielsen. O; Andersen. ZJ; Hvidberg. M; Jensen. SS: Ketzel. M; S0rensen. M; Loft. S: Overvad.
K; Tj0nneland. A. (2011). Lung cancer incidence and long-term exposure to air pollution from traffic.
Environ Health Perspect 119: 860-865. http://dx.doi.org/10.1289/ehp.1002353
Raaschou-Nielsen. O; Bak. H; S0rensen. M; Jensen. SS: Ketzel. M; Hvidberg. M; Schnohr. P; Tj0nneland.
A; Overvad. K; Loft. S. (2010a). Air pollution from traffic and risk for lung cancer in three Danish
cohorts. Cancer Epidemiol Biomarkers Prev 19: 1284-1291. http://dx.doi.org/10.1158/1055-9965.EPI-
10-0036
Ranzi. A; Gambini. M; Spattini. A; Galassi. C: Sesti. D: Bedeschi. M; Messori. A; Baroni. A; Cavagni. G:
Lauriola. P. (2004). Air pollution and respiratory status in asthmatic children: Hints for a locally based
preventive strategy AIRE study. Eur J Epidemiol 19: 567-576.
Rich. DQ; Demissie. K; Lu. SE; Kamat. L; Wartenberg. D; Rhoads. GG. (2009). Ambient air pollutant
concentrations during pregnancy and the risk of fetal growth restriction. J Epidemiol Community Health
63: 488-496. http://dx.doi.org/10.1136/iech.2008.082792
Riedl. MA; Diaz-Sanchez. D; Linn. WS: Gong. H. Jr; Clark. KW: Effros. RM; Miller. JW: Cocker. PR;
Berhane. KT. (2012). Allergic inflammation in the human lower respiratory tract affected by exposure to
diesel exhaust [HEI] (pp. 5-43; discussion 45-64). (ISSN 1041-5505 Research Report 165). Boston,
MA: Health Effects Institute. http://pubs.healtheffects.org/view.php?id=373
Rojas-Martinez. R; Perez-Padilla. R: Olaiz-Fernandez. G: Mendoza-Alvarado. L; Moreno-Macias. H;
Fortoul. T; McDonnell. W: Loomis. D: Romieu. I. (2007a). Lung function growth in children with long-
term exposure to air pollutants in Mexico City. Am J Respir Crit Care Med 176: 377-384.
http://dx.doi.org/10.1164/rccm.200510-1678OC
Romieu. I; Ramirez-Aguilar. M; Sienra-Monge. JJ: Moreno-Macias. H; Del Rio-Navarro. BE; David. G:
Marzec. J: Hernandez-Avila. M; London. S. (2006). GSTM1 and GSTP1 and respiratory health in
asthmatic children exposed to ozone. Eur Respir J 28: 953-959.
http://dx.doi.org/10.1183/09031936.06.00114905
Rosenlund. M; Forastiere. F; Porta. D: De Sario. M; Badaloni. C: Perucci. CA. (2009b). Traffic-related air
pollution in relation to respiratory symptoms, allergic sensitisation and lung function in schoolchildren.
Thorax 64: 573-580. http://dx.doi.org/10.1136/thx.2007.094953
Rotko. T; Kousa. A; Aim. S: Jantunen. M. (2001). Exposures to nitrogen dioxide in EXPOLIS-Helsinki:
Microenvironment, behavioral and sociodemographic factors. J Expo Sci Environ Epidemiol 11:216-
223. http://dx.doi.org/10.1038/sj.jea.7500162
Sacks. JD: Stanek. LW: Luben. TJ: Johns. DO: Buckley. BJ; Brown. JS: Ross. M. (2011). Particulate-matter
induced health effects: Who is susceptible? [Review]. Environ Health Perspect 119: 446-454.
http://dx.doi.org/10.1289/ehp.1002255
November 2013 6-37 DRAFT: Do Not Cite or Quote
-------
Samoli. E; Nastos. PT; Paliatsos. AG: Katsouvanni. K; Priftis. KN. (2011). Acute effects of air pollution on
pediatric asthma exacerbation: Evidence of association and effect modification. Environ Res 111: 418-
424. http://dx.doi.0rg/10.1016/j.envres.2011.01.014
Sarnat. SE; Ravsoni. AU; Li. WW: Holguin. F; Johnson. BA; Flores Luevano. S: Garcia. JH; Sarnat. JA.
(2012). Air pollution and acute respiratory response in a panel of asthmatic children along the U.S.-
Mexico border. Environ Health Perspect 120: 437444. http://dx.doi.org/10.1289/ehp.1003169
Scaife. A; Barclay. J: Hillis. GS: Srinivasan. J; Macdonald. DW: Ross. JA; Ayres. JG. (2012). Lack of
effect of nitrogen dioxide exposure on heart rate variability in patients with stable coronary heart disease
and impaired left ventricular systolic function. Occup Environ Med 69: 587-591.
http://dx.doi.org/10.1136/oemed-2011-100126
Schembari. A; Triguero-Mas. M; De Nazelle. A; Dadvand. P; Vrijheid. M; Cirach. M; Martinez. D;
Figueras. F; Querol. X; Basagafia. X; Eeftens. M; Meliefste. K; Nieuwenhuijsen. MJ. (2013). Personal,
indoor and outdoor air pollution levels among pregnant women. Atmos Environ 64: 287-295.
http://dx.doi.0rg/10.1016/i.atmosenv.2012.09.053
Schiller. JS: Lucas. JW: Ward. BW: Peregov. JA. (2012). Summary health statistics for U.S. adults:
National Health Interview Survey, 2010. (DHHS Publication No. 20121580). Hyattsville, MD: National
Center for Health Statistics, http://www.cdc.gov/nchs/data/series/sr 10/srlO 252.pdf
Schultz. ES: Gruzieva. O: Bellander. T; Bottai. M; Hallberg. J: Kull. I; Svartengren. M; Melen. E;
Pershagen. G. (2012). Traffic-related air pollution and lung function in children at 8 years of age: A
birth cohort study. Am J Respir Crit Care Med 186: 1286-1291. http://dx.doi.org/10.1164/rccm.201206-
1045OC
Seilkop. SK; Campen. MJ: Lund. AK; McDonald. JD; Mauderly. JL. (2012). Identification of chemical
components of combustion emissions that affect pro-atherosclerotic vascular responses in mice. Inhal
Toxicol 24: 270-287. http://dx.doi.org/10.3109/08958378.2012.667455
Selgrade. MK: Mole. ML: Miller. FJ: Hatch. GE: Gardner. DE: Hu. PC. (1981). Effect of NO2 inhalation
and vitamin C deficiency on protein and lipid accumulation in the lung. Environ Res 26: 422-437.
Sevanian. A; Hacker. AD; Elsaved. N. (1982b). Influence of vitamin E and nitrogen dioxide on lipid
peroxidation in rat lung and liver microsomes. Lipids 17: 269-277.
Sinclair. AH; Edgerton. ES: Wyzga. R; Tolsma. D. (2010). A two-time-period comparison of the effects of
ambient air pollution on outpatient visits for acute respiratory illnesses. J Air Waste Manag Assoc 60:
163-175. http://dx.doi.0rg/10.3155/1047-3289.60.2.163
Son. JY; Lee. JT; Park. YH; Bell. ML. (2013). Short-term effects of air pollution on hospital admissions in
Korea. Epidemiology 24: 545-554. http://dx.doi.org/10.1097/EDE.Ob013e3182953244
Stroh. E; Oudin. A; Gustafsson. S: Pilesjo. P; Harrie. L; Stromberg. U; Jakobsson. K. (2005). Are
associations between socio-economic characteristics and exposure to air pollution a question of study
area size? An example from Scania, Sweden. Int J Health Geogr 4: 30-42.
Suh. HH; Zanobetti. A. (201 Ob). Exposure error masks the relationship between traffic-related air pollution
and heart rate variability. J Occup Environ Med 52: 685-692.
http://dx.doi.org/10.1097/JOM.Ob013e3181e8071f
Sunyer. J: Jarvis. D: Gotschi. T; Garcia-Esteban. R; Jacquemin. B; Aguilera. I; Ackerman. U; de Marco. R;
Forsberg. B; Gislason. T; Heinrich. J: Norback. D; Villani. S: Kiinzli. N. (2006). Chronic bronchitis and
urban air pollution in an international study. Occup Environ Med 63: 836-843.
http://dx.doi.org/10.1136/oem.2006.027995
Timonen. KL; Pekkanen. J. (1997). Air pollution and respiratory health among children with asthmatic or
cough symptoms. Am J Respir Crit Care Med 156: 546-552.
http://dx.doi.0rg/10.1164/airccm.156.2.9608044
November 2013 6-38 DRAFT: Do Not Cite or Quote
-------
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of nitrogen, vol.
1-3 [EPA Report]. (EPA/600/8-9!/049aF-cF). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (2008c). Integrated science assessment for oxides of
nitrogen Health criteria [EPA Report]. (EPA/600/R-08/071). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=194645
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2010b). Integrated science assessment for carbon
monoxide [EPA Report]. (EPA/600/R-09/019F). Research Triangle Park,NC.
http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid=218686
U.S. EPA (U.S. Environmental Protection Agency). (2013a). Integrated science assessment for lead [EPA
Report]. (EPA/600/R-10/075F). Research Triangle Park, NC.
http://ofmpub.epa.gov/eims/eimscomm.getfile7p download id=514513
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for ozone and related
photochemical oxidants (final report) [EPA Report]. (EPA/600/R-10/076F). Washington, DC.
http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=247492
Vagaggini. B; Paggiaro. PL; Giannini. D; Franco. AD; Cianchetti. S: Carnevali. S: Taccola. M; Bacci. E;
Bancalari. L; Dente. FL; Giuntini. C. (1996). Effect of short-term NO2 exposure on induced sputum in
normal, asthmatic and COPD subjects. Eur Respir J 9: 1852-1857.
Villeneuve. PJ: Chen. L; Rowe. BH; Coates. F (2007). Outdoor air pollution and emergency department
visits for asthma among children and adults: A case-crossover study in northern Alberta, Canada.
Environ Health Global Access Sci Source 6: 40. http://dx.doi.org/10.1186/1476-069X-6-40
Vincent. GK: Velkoff. VA. (2010). The next four decades: The older population in the United States: 2010 to
2050. (P25-1138). Washington, DC: U.S. Department of Commerce, Economics and Statistics
Administration, U.S. Census Bureau, http://www.census.gov/prod/2010pubs/p25-1138.pdf
Volk. HE; Lurmann. F; Penfold. B; Hertz-Picciotto. I; Mcconnell. R. (2013). Traffic-related air pollution,
paniculate matter, and autism. Arch Gen Psychiatry 70: 71-77.
http://dx.doi.org/10.1001/iamapsvchiatry.2013.266
Witten. A; Solomon. C: Abbritti. E; Arjomandi. M; Zhai. W: Kleinman. M; Balmes. J. (2005). Effects of
nitrogen dioxide on allergic airway responses in subjects with asthma. J Occup Environ Med 47: 1250-
1259.
Wong. CM; Ou. CO: Chan. KP; Chau. YK; Thach. TQ; Yang. L; Chung. RY; Thomas. GN; Peiris. JS:
Wong. TW: Hedlev. AJ; Lam. TH. (2008a). The effects of air pollution on mortality in socially deprived
urban areas in Hong Kong, China. Environ Health Perspect 116: 1189-1194.
http://dx.doi.org/10.1289/ehp.10850
Wong. CM; Vichit-Vadakan. N; Kan. H; Qian. Z. (2008b). Public Health and Air Pollution in Asia (PAPA):
Amulticity study of short-term effects of air pollution on mortality. Environ Health Perspect 116: 1195-
1202. http://dx.doi.org/10.1289/ehp. 11257
Wong. CM; Yang. L; Thach. TQ: Chau. PY; Chan. KP; Thomas. GN; Lam. TH; Wong. TW: Hedlev. AJ:
Peiris. JS. (2009). Modification by influenza on health effects of air pollution in Hong Kong. Environ
Health Perspect 117: 248-253. http://dx.doi.org/10.1289/ehp.11605
Wu. X; Bennett. DH; Lee. K; Cassadv. PL; Ritz. B; Hertz-Picciotto. I. (20lib). Longitudinal variability of
time-location/activity patterns of population at different ages: a longitudinal study in California. Environ
Health Global Access Sci Source 10: 80. http://dx.doi.org/10.1186/1476-069X-10-80
November 2013 6-39 DRAFT: Do Not Cite or Quote
-------
Yorifuji. T; Kashima. S: Tsuda. T; Takao. S: Suzuki. E; Doi. H; Sugiyama. M; Ishikawa-Takata. K; Ohta. T.
(2010). Long-term exposure to traffic-related air pollution and mortality in Shizuoka, Japan. Occup
Environ Med 67: 111-117. http://dx.doi.org/10.1136/oem.2008.045542
Yu. H; Stuart. AL. (2013). Spatiotemporal distributions of ambient oxides of nitrogen, with implications for
exposure inequality and urban design. J Air Waste Manag Assoc 63: 943-955.
http://dx.doi.org/10.1080/10962247.2013.800168
Zemek. R; Szyszkowicz. M; Rowe. BH. (2010). Air pollution and emergency department visits for otitis
media: A case-crossover study in Edmonton, Canada. Environ Health Perspect 118: 1631-1636.
http://dx.doi.org/10.1289/ehp.0901675
November 2013 6-40 DRAFT: Do Not Cite or Quote
------- |