United States
Environmental Protection
Jf lkAgency
EPA/600/R-23/061
March 2023
www.epa.gov/isa
Integrated Science
Assessment for Lead
Appendix 4: Cardiovascular Effects
External Review Draft
March 2023
Health and Environmental Effects Assessment Division
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Exernal Review Draft
DRAFT: Do not cite or quote
-------
DISCLAIMER
1 This document is an external review draft for peer review purposes only. This information is
2 distributed solely for the purpose of predissemination peer review under applicable information quality
3 guidelines. It has not been formally disseminated by the Environmental Protection Agency. It does not
4 represent and should not be construed to represent any agency determination or policy. Mention of trade
5 names or commercial products does not constitute endorsement or recommendation for use.
6
Exernal Review Draft
4-ii
DRAFT: Do not cite or quote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
DOCUMENT GUIDE
This Document Guide is intended to orient readers to the organization of the Lead (Pb) Integrated Science
Assessment (ISA) in its entirety and to the sub-section of the ISA at hand (indicated in bold). The ISA consists of
the Front Matter (list of authors, contributors, reviewers, and acronyms), Executive Summary, Integrated Synthesis,
and 12 appendices, which can all be found at https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=357282.
Front Matter
Executive Summary
Integrative Synthesis
Appendix 1. Lead Source to Concentration
Appendix 2. Exposure, Toxicokinetics, and Biomarkers
Appendix 3. Nervous System Effects
Appendix 4. Cardiovascular Effects
Appendix 5. Renal Effects
Appendix 6. Immune System Effects
Appendix 7. Hematological Effects
Appendix 8. Reproductive and Developmental Effects
Appendix 9. Effects on Other Organ Systems and Mortality
Appendix 10. Cancer
Appendix 11. Effects of Lead in Terrestrial and Aquatic Ecosystems
Appendix 12. Process for Developing the Pb Integrated Science Assessment
Exernal Review Draft
4-iii
DRAFT: Do not cite or quote
-------
CONTENTS
LIST OF TABLES 4-v
LIST OF FIGURES 4-vi
ACRONYMS AND ABBREVIATION 4-viii
APPENDIX 4 CARDIOVASCULAR EFFECTS 4-1
4.1 Introduction and Summary of the 2013 Integrated Science Assessment 4-1
4.1.1 Hypertension and Increased Blood Pressure 4-2
4.1.2 Subclinical Atherosclerosis 4-4
4.1.3 Coronary Heart Disease 4-5
4.1.4 Cerebrovascular Disease 4-6
4.2 Scope 4-6
4.3 Blood Pressure and Hypertension 4-8
4.3.1 Epidemiologic Studies of Blood Pressure and Hypertension 4-8
4.3.2 Toxicological Studies of Blood Pressure and Hypertension 4-39
4.3.3 Integrated Summary of Blood Pressure and Hypertension 4-41
4.4 Ischemic Heart Disease and Associated Cardiovascular Effects 4-43
4.4.1 Epidemiologic Studies of Ischemic Heart Disease 4-43
4.4.2 Summary of Ischemic Heart Disease 4-47
4.5 Heart Failure and Impaired Cardiac Function 4-48
4.5.1 Epidemiologic Studies of Impaired Cardiac Function 4-48
4.5.2 Toxicological Studies of Impaired Cardiac Function 4-49
4.5.3 Integrated Summary of Impaired Cardiac Function 4-50
4.6 Endothelial Dysfunction 4-51
4.6.1 Toxicological Studies of Endothelial Dysfunction 4-51
4.6.2 Summary of Endothelial Dysfunction 4-52
4.7 Cardiac Electrophysiology and Arrythmia 4-52
4.7.1 Cardiac Depolarization, Repolarization, and Arrythmia 4-52
4.7.2 Heart Rate and Heart Rate Variability 4-54
4.7.3 Integrated Summary of Cardiac Electrophysiology and Arrythmia 4-56
4.8 Atherosclerosis and Peripheral Artery Disease 4-57
4.8.1 Epidemiologic Studies of Atherosclerosis and Peripheral Artery Disease 4-57
4.8.2 Toxicological Studies of Atherosclerosis 4-60
4.8.3 Integrated Summary of Atherosclerosis 4-61
4.9 Cerebrovascular Disease 4-61
4.9.1 Epidemiologic Studies of Cerebrovascular Disease 4-61
4.9.2 Summary of Cerebrovascular Disease 4-62
4.10 Cardiovascular Mortality 4-62
4.10.1 Epidemiologic Studies of Cardiovascular Mortality 4-62
4.10.2 Summary of Cardiovascular Mortality 4-70
4.11 Biological Plausibility 4-71
4.12 Summary and Causality Determination 4-75
4.13 Evidence Inventories - Data Tables to Summarize Study Details 4-83
4.14 References 4-154
Exernal Review Draft
4-iv
DRAFT: Do not cite or quote
-------
LIST OF TABLES
Table 4-1 Summary of causality determinations from the 2013 Pb Integrated Science Assessment. 4-2
Table 4-2 Summary of evidence indicating a causal relationship between Pb exposure and
cardiovascular effects. 4-80
Table 4-3 Epidemiologic studies of Pb exposure and blood pressure. 4-83
Table 4-4 Epidemiologic studies of Pb exposure and hypertension. 4-100
Table 4-5 Epidemiologic studies of Pb exposure and blood pressure and hypertension among
children. 4-110
Table 4-6 Animal toxicological studies of Pb exposure and blood pressure/hypertension. 4-117
Table 4-7 Epidemiologic studies of Pb exposure and coronary and ischemic heart disease. 4-121
Table 4-8 Epidemiologic studies of Pb exposure and cardiac function. 4-128
Table 4-9 Animal toxicological studies of cardiac function. 4-131
Table 4-10 Animal toxicological studies of Pb exposure and endothelial dysfunction. 4-133
Table 4-11 Epidemiologic studies of Pb exposure cardio electrophysiology and arrythmia. 4-134
Table 4-12 Animal toxicological studies of Pb exposure and cardiac electrophysiology. 4-138
Table 4-13 Epidemiologic studies of Pb exposure and atherosclerosis and peripheral artery disease. 4-140
Table 4-14 Animal toxicological studies of Pb exposure and atherosclerosis. 4-143
Table 4-15 Epidemiologic studies of Pb exposure and cerebrovascular disease. 4-144
Table 4-16 Epidemiologic studies of Pb exposure and cardiovascular mortality. 4-146
Exernal Review Draft
4-v
DRAFT: Do not cite or quote
-------
LIST OF FIGURES
Figure 4-1 Association between biomarkers of Pb exposure and blood pressure. 4-10
Figure 4-2 Association between blood Pb level quartiles and systolic blood pressure, diastolic blood
pressure, and hypertension, polluted region of rural southwest China. 4-13
Figure 4-3 Association between blood Pb level quartiles and systolic blood pressure, diastolic blood
pressure, and hypertension, unpolluted region of rural southwest China. 4-13
Figure 4-4 Blood pressure (systolic and diastolic) and a doubling of blood Pb levels stratified by sex
and race National Health and Nutrition Examination Survey (2003-2010). 4-17
Figure 4-5 Blood pressure (systolic, diastolic, and pulse pressure) and a doubling of blood Pb level
stratified by sex and race National Health and Nutrition Examination Survey (1999-2006). 4-18
Figure 4-6 Blood pressure (systolic, diastolic, and pulse pressure) and a doubling of blood Pb levels
stratified by sex and race, National Health and Nutrition Examination Survey (2001-2008). 4-19
Figure 4-7 Interactive effect between blood Pb levels, race, and education level, National Health and
Nutrition Examination Survey (2001-2008). 4-20
Figure 4-8 Interactive effect between blood Pb levels, race, and poverty level, National Health and
Nutrition Examination Survey (2001-2008). 4-21
Figure 4-9 Association (odds ratio) between blood Pb levels and systolic blood pressure >140 mmHg
or diastolic blood pressure >90 mmHg by allostatic load, National Health and Nutrition
Examination Survey (1999-2008). 4-23
Figure 4-10 Relationship between tibia Pb levels and systolic blood pressure by perceived stress,
Normative Aging Study cohort. 4-24
Figure 4-11 Relationship between blood Pb levels and systolic blood pressure by sex and age,
Canadian Health Measures Survey. 4-26
Figure 4-12 Relationship between blood Pb levels and diastolic blood pressure by sex and age,
Canadian Health Measures Survey. 4-27
Figure 4-13 Pulse pressure and tibial Pb levels, modified by vitamin D receptor variant, Normative
Aging Study cohort. 4-28
Figure 4-14 Associations between biomarkers of Pb exposure and hypertension. 4-30
Figure 4-15 Dose-response curve between tibia Pb levels and resistant hypertension, Normative Aging
Study cohort. 4-33
Figure 4-16 Dose-response curve between blood Pb and any hypertension or uncontrolled
hypertension, restricted cubic splines, National Health and Nutrition Examination Survey
(1999-2006). 4-34
Figure 4-17 Odds ratios for the association between quartiles of blood Pb and prevalent hypertension,
stratified by sex. 4-35
Figure 4-18 Odds ratios for the association between quartiles of blood Pb levels and prevalent
hypertension, stratified by body mass index, China National Human Biomonitoring cohort. 4-37
Exernal Review Draft 4-vi DRAFT: Do not cite or quote
-------
Figure 4-19 Relationship between blood Pb levels and common carotid artery plaques, common
carotid artery diameter, and cardiovascular disease among diabetic patients. 4-46
Figure 4-20 Meta-analysis of the association between biomarkers of Pb exposure and coronary heart
disease. 4-47
Figure 4-21 Association between aortic pulse wave velocity with blood Pb levels and age. 4-59
Figure 4-22 Stratified associations between abdominal aortic calcification score and blood Pb levels. 4-60
Figure 4-23 Associations between blood Pb level and cardiovascular mortality. 4-64
Figure 4-24 Dose-response relationship between blood Pb levels and cardiovascular and ischemic
heart disease mortality. 4-67
Figure 4-25 Cumulative incidence function of cardiovascular mortality by blood Pb level, National
Health and Nutrition Examination Survey III (1988-1994). 4-68
Figure 4-26 Potential biological pathways for cardiovascular effects following exposure to Pb. 4-72
Exernal Review Draft
4-vii
DRAFT: Do not cite or quote
-------
ACRONYMS AND ABBREVIATION
A peak late diastolic velocity
AAC abdominal aortic calcification
AAS atomic absorption spectrometry
AAS atomic absorption spectrometry
ABLES Adult Blood Lead Epidemiology and
Surveillance
ACE angiotensin-converting enzyme
ACh acetylcholine
ADMA asymmetric dimethylarginine
AF atrial fibrillation
AGT angiotensinogen
AHA American Heart Association
AL allostatic load
ALAD S-aminolevulinic acid dehydratase
ANS autonomic nervous system
APOE apolipoprotein E
AQCD Air Quality Criteria Document
ARCA Automobile Racing Club of America
ASCVD atherosclerotic cardiovascular disease
ASV anodic stripping voltammetry
BEST Bangladesh Vitamin E and Selenium
Trial
BLL blood lead level
BMI body mass index
BP blood pressure
BRHS British Regional Heart Study
BW body weight
Ca2+ calcium ion
C282Y HFE mutant of the HFE wildtype
CAD coronary artery disease
CAS coronary artery stenosis
CCA common carotid artery
Cd cadmium
CHD coronary heart disease
CHF congestive heart failure
CI confidence interval
CIF cumulative incidence function
CRP C-reactive protein
CT computerized tomography
CVD cardiovascular disease
d day(s)
DBP diastolic blood pressure
E peak early diastolic velocity
e' peak early diastolic mitral annular
velocity
EAF electric arc furnace
ECG electrocardiography
eGFR estimated glomerular filtration rate
EMM effect measure modification
ETAAS Electrothermal Atomic Absorption
Spectrometry
Eyr erythrocyte
FABP4 adipocyte fatty acid-binding protein
FBG fasting blood glucose
Fe iron
FRS Framingham risk score
GFAAS graphite furnace atomic absorption
spectrometry
GFR glomerular filtration rate
GLS global longitudinal strain
GM geometric mean
GRS genetic risk score
GSD geometric standard deviation
GSE geometric standard error
GuLF Gulf Long-T erm F ollow-up
GW gestational week
FlbAlc hemoglobin Ale
H63D HFE mutant of the HFE
HDL high-density lipoprotein
HDL-C high-density lipoprotein cholesterol
HF high frequency
HFE hemochromatosis gene
HMOX1 heme oxygenase-1
HOME Health Outcomes and Measures of the
Environment Study
hr hour(s)
HR hazard ratio
HRV heart rate variability
HTN hypertension
ICD International Classification of Diseases
ICP-MS inductively coupled plasma mass
spectrometry
IHD ischemic heart disease
IMT intimal medial thickening
IQR interquartile ratio
ISA Integrated Science Assessment
IVS interventricular septum
KNHANES Korea National Health and Nutrition
Examination Survey
K-XRF K-Shell X-ray fluorescence
LCL lower confidence limit
Exernal Review Draft
4-viii
DRAFT: Do not cite or quote
-------
LDL low-density lipoprotein
LDL-C low-density lipoprotein
LF low frequency
LV left ventricular
LVDP left ventricular diastolic pressure
LVMI left ventricular mass index
LVPW left ventricular posterior wall
LVSP left ventricular systolic pressure
MAP mean arterial pressure
MDCS-CC cardiovascular cohort of the Malmo
Cancer and Diet Study
METAL Environmental Pollutant Exposure and
Metabolic Diseases in Shanghai
METS Modeling the Epidemiologic Transition
Study
MI myocardial infarction
mo month(s)
NA not available
NAS Normative Aging Study
NASCAR National Association for Stock Car
Auto Racing
NH non-Hispanic
NHANES National Health and Nutrition
Examination Survey
NN normal-to-normal
NO nitric oxygen
NR not reported
nu normalized units
OR odds ratio
PAD peripheral artery disease
Pb lead
PECOS Population, Exposure, Comparison,
Outcome, and Study
PHQ Patient Health Questionnaire
PIR poverty-to-income ratio
PIVUS Prospective Investigation of the
Vasculature in Uppsala Seniors
PND postnatal day
PP pulse pressure
PR prevalence ratio
PROGRESS Programming Research in Obesity,
Growth Environment and Social Stress
PVD peripheral vascular disease
PWV pulse wave velocity
Q quartile
QRS QRS complex in ECG
QRSc corrected QRS duration
QT QT interval in ECG
QTc corrected QT interval
RAAS renin-angiotensin-aldosterone system
RLS regional longitudinal strain
rMSSD root-mean-square of successive
differences
ROS reactive oxygen species
RR relative risk
RRS regional radial strain
RV right ventricular
RVDP right ventricular diastolic pressure
RVSP right ventricular systolic pressure
RWT relative wall thickness
SBP systolic blood pressure
SD standard deviation
SE standard error
SES socioeconomic status
SNP single nucleotide polymorphisms
SOD superoxide dismutase
SOF Study of Osteoporotic Fractures
ST segment measured from the J point to
the end of the T wave in an ECG
T tertile
TC total cholesterol
TPR Total peripheral resistance
TRI Toxic Release Inventory
UCL upper confidence limit
VA-NAS Veterans Affairs Normative Aging
Study
VDR vitamin D receptor
yr year(s)
Exernal Review Draft
4-viii
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
APPENDIX 4 CARDIOVASCULAR EFFECTS
Causality Determination for Lead (Pb) Exposure and
Cardiovascular Effects
This appendix characterizes the scientific evidence that supports causality determinations for
lead (Pb) exposure and cardiovascular effects. The types of studies evaluated within this appendix are
consistent with the overall scope of the ISA as detailed in the Process Appendix (see Section 12.4). In
assessing the overall evidence, the strengths and limitations of individual studies were evaluated based
on scientific considerations detailed in Table 12-5 of the Process Appendix (see Section 12.6.1). More
details on the causal framework used to reach these conclusions are included in the Preamble to the
ISA (U.S. EPA. 2015). The evidence presented throughout this appendix supports the following
causality conclusions:
Outcome
Causality Determination
Cardiovascular Effects
Causal
The Executive Summary, Integrated Synthesis, and all other appendices of this Pb ISA can be found at
https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=357282.
4.1 Introduction and Summary of the 2013 Integrated Science
Assessment
The 2013 Pb ISA (U.S. EPA. 2013a) made four causality determinations with respect to
cardiovascular disease, using the U.S. Surgeon General's Report on Smoking as a guideline to group
evidence into health outcome categories (CDC. 2004). The categories included hypertension, subclinical
atherosclerosis, coronary heart disease (CHD), and cerebrovascular disease. Evidence was sufficient to
conclude causal relationships between Pb exposure and hypertension and CHD. The causal determination
for hypertension was not only informed by evidence of hypertension and increases in blood pressure (BP),
but also cardiovascular-related mortality. The 2013 Pb ISA indicated a coherence between epidemiologic
and toxicological studies, and animal toxicological studies provided strong evidence to support biologic
plausibility. Specifically, oxidative stress from Pb exposure can result in an inactivation of nitrous oxide,
which can lead to increased vasoconstriction and therefore increased BP. The causal determination for
CHD was informed by epidemiologic evidence for heart rate variability (HRV); myocardial infarction
(MI); ischemic heart disease (IHD); mortality from MI, IHD, and CHD; and increased thrombosis,
coagulation, and arrhythmia in animals. The biological plausibility and mode of action for these
cardiovascular effects was provided by evidence for oxidative stress, inflammation, and vascular cell
activation or dysfunction. Specifically, coherence between epidemiologic and toxicologic evidence
demonstrated that Pb exposure may promote a procoagulant state that can potentially contribute to
thrombus formulation and therefore reduced blood supply to the heart. Causality determinations for each
Exernal Review Draft
4-1
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
24
of the four categories are summarized in Table 4-1 and some of the evidence supporting these
determinations is discussed in Sections 4.1.1 to 4.1.4.
Table 4-1 Summary of causality determinations from the 2013 Pb Integrated
Science Assessment.
Outcome Group
Causality Determination
Hypertension and Increased Blood Pressure
Causal
Subclinical Atherosclerosis
Suggestive
Coronary Heart Disease
Causal
Cerebrovascular Disease
Inadequate
The current ISA is consistent with more recent ISAs (e.g., 2019 Particulate Matter and 2020
Ozone ISAs) (U.S. EPA. 2020. 2019) in that it makes a single causality determination for cardiovascular
effects. This approach recognizes that many cardiovascular endpoints are inter-related (e.g., both
atherosclerosis and endothelial dysfunction can contribute to increases in BP), and therefore not easily
discussed in isolation. The remainder of this section summarizes the evidence for Pb exposures and
cardiovascular effects assessed in the 2013 Pb ISA, including the evidence for hypertension and increased
BP (Section 4.1.1), atherosclerosis (Section 4.1.2), CHD (Section 4.1.3), and cerebrovascular disease
(Section 4.1.4). Subsequent sections of this appendix provide an overview of study inclusion criteria for
the cardiovascular effects evidence in the current ISA (Section 4.2), summaries and evaluations of recent
health effects evidence (Sections 4.3 to 4.10), a discussion of biological plausibility (Section 4.1.1), and a
discussion of how all the individual lines of cardiovascular evidence were considered and integrated to
inform the causality determination for Pb exposure and cardiovascular effects (Section 4.1.2). Study-
specific details, including information on study design; exposure metrics, concentrations, and durations;
and select results are presented in summary tables in Section 4.1.3.
4.1.1 Hypertension and Increased Blood Pressure
The 2013 Pb ISA (U.S. EPA. 2013a') indicated that the combined evidence from epidemiologic
and animal toxicological studies was sufficient to conclude that there is a causal relationship between Pb
exposure and hypertension. This conclusion was informed by the coherence of effects observed between
epidemiologic and toxicological studies with respect to hypertension and its related endpoints. A number
of prospective epidemiologic studies clearly supported the relationship between biomarkers of Pb
exposure and hypertension incidence and changes in BP (U.S. EPA. 2013a). The prospective evidence
was supported by meta-analyses that underscored the consistency and reproducibility of Pb-associated
increases in BP and hypertension across diverse populations and different study designs. Importantly,
Exernal Review Draft
4-2
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
many epidemiologic studies evaluated in the 2013 Pb ISA (U.S. EPA. 2013a) adjusted for a wide range of
potential confounders to reduce uncertainty due to potential unmeasured confounding. Although the
adjustment for specific factors varied by study, the collective body of evidence included adjustments for
multiple potential key confounding factors, including age, diet, sex, body mass index (BMI), BP-lowering
medication use, socioeconomic status (SES), race/ethnicity, alcohol consumption, cholesterol, smoking,
preexisting disease (e.g., diabetes), measures of renal function, and co-pollutant exposures (e.g., cadmium
[Cd]), while still maintaining positive associations between biomarkers of Pb exposure and changes in
BP/hypertension.
Results from animal toxicological studies examining BP-related endpoints were coherent with the
epidemiologic findings. In the previous review, all the animal toxicological studies providing blood Pb
level (BLL) and BP measurements reported increases in BP with increasing BLLs. Most of these studies
examined Pb exposures that resulted in mean BLLs >10 j^ig/dL: however, a single animal toxicological
study conducted in rats after drinking water exposure found a continuous increase in BP in animals with
mean BLLs ranging from 0.05 to 29 (ig/dL with no evidence of a threshold (U.S. EPA. 2013a).
Animal toxicological studies also provided strong support for the biological plausibility of the Pb-
associated increases in BP and hypertension observed in epidemiologic studies. Studies evaluated in the
2013 Pb ISA (U.S. EPA. 2013a) demonstrated that oxidative stress following Pb exposure inactivates
vasodilator nitric oxide, which may lead to increased vasoconstriction and increased BP. If increases in
BP persist, the result is hypertension (i.e., chronically elevated BP). Oxidative stress can also damage the
endothelium, further disrupting endothelium-dependent vascular relaxation and increasing the contractile
response. Studies also suggested Pb exposure disrupts normal contractile processes by altering the
sympathetic nervous system, the renin-angiotensin-aldosterone system, and the balance between
production of vasodilators and vasoconstrictors (U.S. EPA. 2013a).
Although the relationship between exposure to Pb and increases in BP in adults was well
established at the time of the last review, some uncertainties were identified in the evidence for BP
changes, specifically among children. The 2013 Pb ISA noted that some of the BP results (and other
cardiovascular effects) observed in children may be antecedent to later-in-life effects. Therefore, there is
at least some uncertainty in the level, timing, frequency, and duration of Pb exposure contributing to the
reported cardiovascular effects in adults. That is, although there is a clear relationship between exposure
to Pb and changes in BP in adults, it is possible that childhood Pb exposures could contribute to adult
BLLs through processes such as bone remodeling that occurs during aging and/or pregnancy. Thus, Pb-
associated changes in BP reported in adults may be appreciably influenced by past Pb exposures, perhaps
as early as childhood.
Overall, epidemiologic and toxicological evidence from the previous review consistently
demonstrated that Pb exposures are associated with increased BP and hypertension in adults. The
epidemiologic studies have been replicated by different researchers in different cohorts and associations
reported in these studies have largely remained positive after adjusting for numerous potential
Exernal Review Draft
4-3
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
24
25
26
27
28
29
30
31
32
33
confounding factors. These studies are also coherent with numerous animal toxicological studies
demonstrating increases in BP following Pb exposure. In addition, toxicological studies provided
biological plausibility and a potential mode of action for the results observed in epidemiologic studies.
Thus, in the 2013 Pb ISA, the combined evidence from epidemiologic and animal toxicological studies
was sufficient to conclude that there is a causal relationship between Pb exposure and hypertension.
Since the last review, the evidence relating Pb exposure to increases in BP and hypertension have
expanded greatly (see Section 4.3), further reinforcing an already strong evidence base established in the
last ISA. As a result, evidence from epidemiologic and animal toxicological studies related to BP and
hypertension are a key driver for the current ISA's conclusion of a causal relationship between Pb
exposure and cardiovascular effects. A discussion of how the epidemiologic and animal toxicological
evidence of hypertension and increased BP contributes to the determination of a causal relationship
between exposure to Pb and cardiovascular effects in this ISA can be found in Section 4.1.2.
4.1.2 Subclinical Atherosclerosis
The 2013 Pb ISA (U.S. EPA. 2013a') concluded that the evidence was suggestive of, but not
sufficient to infer, a causal relationship between exposure to Pb and subclinical atherosclerosis. Studies
considered in the last review included an analysis from the 2006 Pb Air Quality Criteria Document
(AQCD) indicating that exposure to Pb was associated with peripheral artery disease (PAD) in the
National Health and Nutrition Examination Survey (NHANES) population, and that co-exposure with Cd
did not confound the association (U.S. EPA. 2006a). Additional epidemiologic findings presented in the
2013 Pb ISA were limited to cross-sectional analyses. One such analysis reported an increasing trend in
the odds of PAD across concurrent BLL groups in adult NHANES participants. Furthermore, in an
epidemiologic study conducted in a Pb-exposed population, positive associations were reported between
BLLs and increases in intima-media thickness and atherosclerotic plaque presentation (U.S. EPA. 2013a).
However, the 2013 ISA noted that most of the available epidemiologic analyses were cross-sectional in
nature, contributing to uncertainty in the level, timing, frequency, and duration of the Pb exposures that
contributed to the observed associations.
In addition to the epidemiologic evidence, toxicological studies provided limited evidence to
suggest Pb exposure may initiate atherosclerotic vessel disease. The 2013 Pb ISA (U.S. EPA. 2013a)
noted that in vitro Pb exposure resulted in a concentration-dependent increase in arterial intimal thickness
in human radial and internal mammary arteries. Moreover, exposure to Pb in rats increased aortic medial
thickness. Following Pb exposure, toxicological studies also demonstrated evidence of oxidative stress
and systemic inflammation, processes which are important to the development of atherosclerosis. Finally,
toxicological studies indicated a relationship between Pb exposure and elevation of cholesterol (U.S.
EPA. 2013a). Taken together with the epidemiologic evidence and its associated uncertainties, the 2013
Exernal Review Draft
4-4
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
24
25
26
27
28
29
30
31
32
33
34
Pb ISA concluded that the evidence was suggestive of a causal relationship between Pb exposure and
subclinical atherosclerosis.
Since the last review, there is additional evidence supporting the potential contribution of Pb
exposures to atherosclerosis. This evidence includes recent epidemiologic studies reporting positive
associations with markers of atherosclerosis and a recent toxicological study in rats demonstrating
morphological changes in the aorta consistent with the potential for atherosclerosis (see Section 4.3). A
discussion of how the epidemiologic and animal toxicological evidence of atherosclerosis contributes to
the determination of a causal relationship between exposure to Pb and cardiovascular effects in this ISA
can be found in Section 4.1.2.
4.1.3 Coronary Heart Disease
The 2013 Pb ISA (U.S. EPA. 2013a) concluded that the evidence supports a causal relationship
between exposure to Pb and CHD. This conclusion was primarily based on the results of epidemiologic
studies examining the incidence of MI, IHD, and HRV, and on studies examining mortality from CHD,
MI, or IHD. The rationale for this determination is summarized below.
The 2013 Pb ISA (U.S. EPA. 2013a') described longitudinal studies in cohorts in different
locations with follow-up periods of up to 12 years. These studies consistently reported that biomarkers of
Pb exposure are associated with risk of mortality from MI, IHD, or CHD. The strongest associations were
observed with MI mortality. Despite the differences in design and methods used across epidemiologic
studies, associations between higher levels of tissue Pb (blood, bone) and higher risk of CHD-related
mortality were generally observed 2013 Pb ISA (U.S. EPA. 2013a). The body of evidence demonstrating
associations with mortality from CHD was substantiated by several studies indicating associations
between biomarkers of Pb exposure and incidence of CHD-related outcomes. For example, a prospective
analysis examined the incidence of IHD (physician-confirmed MI, angina pectoris) and reported that
blood and bone Pb levels contributed independently to IHD incidence. The 2013 Pb ISA further noted
that coherence for the associations in humans was provided by an animal toxicological study suggesting
that Pb exposure promoted a procoagulant state that could contribute to thrombus formation, and thus,
potentially reduce the blood supply to the heart (U.S. EPA. 2013a).
Previous research has indicated that decreased HRV is associated with higher mortality from MI
and can be used as a predictor of the physiological processes underlying CHD. The 2013 Pb ISA
described several cohort studies demonstrating associations between Pb exposure and decreases in HRV
(U.S. EPA. 2013a). Additionally, a prospective analysis reported that higher tibia Pb levels were
associated with increases in certain ECG measurements, including the corrected QT interval (QTc) and
corrected QRS duration (QRSc), which can be indicative of impaired cardiac electrophysiology. As CHD
is the result of vascular blockage, the previous Pb ISA also noted that these epidemiologic associations
were supported, at least in part, by the limited evidence for subclinical atherosclerosis (Section 4.1.2). The
Exernal Review Draft
4-5
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
24
25
26
27
28
29
30
31
2013 Pb ISA (U.S. EPA. 2013a') additionally noted that the strong and consistent evidence for Pb-induced
hypertension supported the biological plausibility of the Pb-induced increase in CHD risk.
In summary, in the 2013 Pb ISA (U.S. EPA. 2013a). several studies examining CHD morbidity
and mortality and contributing cardiovascular effects reported consistent associations between Pb
exposure and CHD. In addition, both animal toxicological and epidemiologic studies describe a
biologically plausible potential mode of action (e.g., hypertension, atherosclerosis, potentially adverse
changes in cardiac electrophysiology). Taken together, the 2013 Pb ISA (U.S. EPA. 2013a) concluded
that epidemiologic evidence, supported by toxicological evidence, was sufficient to conclude a causal
relationship exists between Pb exposure and CHD.
Since the last review, the epidemiologic evidence describing the relationship between Pb
exposure and endpoints related to CHD has expanded (see Section 4.4) and further strengthens the
evidence base established in the last ISA. In particular, there are new epidemiologic studies reporting
associations with IHD and MI mortality. Results of animal toxicological studies of HRV and cardiac
electrophysiology published since the last review have been largely mixed. A discussion of how this
epidemiologic and animal toxicological evidence contributed to the determination of a causal relationship
between exposure to Pb and cardiovascular effects in this current review can be found in Section 4.1.2.
4.1.4 Cerebrovascular Disease
The 2013 Pb ISA (U.S. EPA. 2013a) concluded there was insufficient evidence to inform the
relationship between cerebrovascular disease and Pb exposure. Despite strong evidence indicating effects
of Pb exposure on hypertension and CHD, very few studies evaluated in the 2013 Pb ISA examined the
effects of Pb exposure on cerebrovascular disease. Furthermore, the studies that were available reported
relatively imprecise associations between BLLs and stroke-related mortality. With respect to animal
toxicological studies, there was some evidence for processes that could lead to cerebrovascular disease,
such as an increase in markers of oxidative stress, inflammation, and coagulation that could potentially
aid in clot formation. When considered as a whole, however, this limited evidence was insufficient to
inform the relationship between Pb exposure and cerebrovascular disease. In the current review, studies
examining the potential relationship between Pb exposure and cerebrovascular disease remain quite
limited (see Section 4.9). Consideration of this evidence in the causality determination for Pb exposures
and cardiovascular effects is presented in Section 4.1.2.
4.2 Scope
The scope of this appendix is defined by Population, Exposure, Comparison, Outcome, and Study
Design (PECOS) statements. The PECOS statement defines the objectives of the review and establishes
study inclusion criteria and thereby facilitates identification of the most relevant literature to inform the
Exernal Review Draft
4-6
DRAFT: Do not cite or quote
-------
1 Pb ISA.1 To identify the most relevant literature, the body of evidence from the 2013 Pb ISA was
2 considered in the development of the PECOS statements for this appendix. Specifically, well-established
3 areas of research; gaps in the literature; and inherent uncertainties in specific populations, exposure
4 metrics, comparison groups, and study designs identified in the 2013 Pb ISA inform the scope of this
5 appendix. The 2013 Pb ISA used different inclusion criteria than the current ISA, and many of the studies
6 referenced therein do not meet the current PECOS criteria (e.g., due to higher or unreported biomarker
7 levels). Many of those studies are discussed in this appendix to establish the state of the evidence prior to
8 this assessment. With the exception of supporting evidence used to demonstrate the biological plausibility
9 of Pb-associated cardiovascular effects, recent studies were only included if they satisfied all of
10 components of the following discipline-specific PECOS statements:
11 Epidemiologic Studies:
12 Population: Any human population, including specific populations or lifestages that might be at
13 increased risk of a health effect;
14 Exposure: Exposure to Pb2 as indicated by biological measurements of Pb in the body—with a
15 specific focus on Pb in blood, bone, and teeth; validated environmental indicators of Pb
16 exposure,3 or intervention groups in randomized trials and quasi-experimental studies;
17 Comparison: Populations, population subgroups, or individuals with relatively higher versus
18 lower levels of the exposure metric (e.g., per unit or log unit increase in the exposure metric,
19 or categorical comparisons between different exposure metric quantiles);
20 Outcome: Cardiovascular effects including but not limited to CHD, hypertension and increased
21 BP, and cardiovascular-related mortality; and
22 Study design: Epidemiologic studies consisting of longitudinal and retrospective cohort studies,
23 case-control studies, cross-sectional studies with appropriate timing of exposure for the health
24 endpoint of interest, randomized trials and quasi-experimental studies examining
25 interventions to reduce exposures.
26 Experimental Studies:
27 Population: Laboratory nonhuman mammalian animal species (e.g., mouse, rat, guinea pig,
28 minipig, rabbit, cat, dog) of any lifestage (including preconception, in utero, lactation,
29 peripubertal, and adult stages);
1 The following types of publications are generally considered to fall outside the scope and are not included in the
ISA: review articles (which typically present summaries or interpretations of existing studies rather than bringing
forward new information in the form of original research or new analyses), Pb poisoning studies or clinical reports
(e.g., involving accidental exposures to very high amounts of Pb described in clinical reports that may be extremely
unlikely to be experienced under ambient air exposure conditions), and risk or benefits analyses (e.g., that apply
concentration-response functions or effect estimates to exposure estimates for differing cases).
2 Recent studies of occupational exposure to Pb were considered insofar as they addressed a topic area that was of
particular relevance to the National Ambient Air Quality Standards review (e.g., longitudinal studies designed to
examine recent versus historical Pb exposure).
3 Studies that estimate Pb exposure by measuring Pb concentrations in PMio and PM2 5 ambient air samples are only
considered for inclusion if they also include a relevant biomarker of exposure. Given that size distribution data for
Pb-PM are fairly limited, it is difficult to assess the representativeness of these concentrations to population
exposure [Section 2.5.3 (U.S. EPA. 2013aYI. Moreover, data illustrating the relationships of Pb-PMio and Pb-PM2 5
with BLLs are lacking.
Exernal Review Draft
4-7
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
24
25
26
27
28
Exposure: Oral, inhalation, or intravenous routes administered to a whole animal (in vivo) that
results in a BLL of 30 (ig/dL or below;1'2
Comparators: A concurrent control group exposed to vehicle-only treatment or untreated control
Outcome: Cardiovascular effects; and
Study design: Controlled exposure studies of animals in vivo.
4.3 Blood Pressure and Hypertension
High BP typically is defined as a systolic BP (SBP) above 130 mmHg or a diastolic blood
pressure (DBP) above 80 mmHg. SBP represents the pressure in the arteries as the heart contracts, while
DBP represents the pressure in the arteries as the heart is relaxed and is filling with blood. Prolonged high
BP is known as hypertension and can lead to a thickening of the ventricular wall resulting in diminished
filling during diastole. Hypertension can contribute ultimately to the development of arrythmia and heart
failure. Pulse pressure (PP), or the difference between SBP and DBP, as well as mean arterial pressure
(MAP)— which is a function of cardiac output, systemic vascular resistance, and central venous
pressure—are additional metrics used in studies of air pollution's effects on BP. Moreover, hypertension
is one of several conditions, including high blood sugar, excess body fat around the waist, and abnormal
triglyceride levels, that comprise metabolic syndrome (see Appendix 9), which is a risk factor for heart
disease, stroke, and diabetes.
4.3.1 Epidemiologic Studies of Blood Pressure and Hypertension
Several epidemiologic studies evaluated in the 2013 Pb ISA (U.S. EPA. 2013a) and previous
AQCD documents (U.S. EPA. 2006b. 1990) indicate an association between biomarkers of Pb exposure
and changes in BP and hypertension risk. Although previous studies evaluated in the 2006 Pb AQCD
(U.S. EPA. 2006b) and a supplement to the 1986 Pb AQCD (U.S. EPA. 1990) most likely represented
populations historically exposed to higher levels of air Pb (measured during the 1970s and 1980s)
compared with populations today, they indicated there was no apparent threshold below which blood Pb
was not significantly associated with changes in BP, for mean BLLs ranging from 7 (ig/dL to 34 (ig/dL.
The 2013 Pb ISA (U.S. EPA. 2013a') further demonstrated an association between Pb biomarkers and
increased BP and hypertension risk at BLLs <2 (ig/dL. The majority of the evidence for this association
was derived from the Normative Aging Study (NAS) cohort of mostly older white men (Zhang et al..
2010; Perlstein et al.. 2007; Elmarsafawv et al.. 2006) and a Korean study composed of workers with high
BLLs (mean BLLs -20-35 (ig/dL), due to occupational Pb exposure (Weaver et al.. 2008; Glenn et al..
1 Pb mixture studies are included if they employ an experimental arm that involves exposure to Pb alone.
2 This level represents an order of magnitude above the upper end of the distribution of U.S. young children's BLL.
The 95th percentile of the 2011-2016 NHANES distribution of BLL in children (1-5 years; n = 2,321) is 2.66 (ig/dL
(Eganetal.. 20211 and the proportion of individuals with BLL that exceed this concentration varies depending on
factors including (but not limited to) housing age, geographic region, and a child's age, sex, and nutritional status.
Exernal Review Draft
4-8
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
24
25
26
27
28
2006). The 2013 Pb ISA (U.S. EPA. 2013a') also highlighted specific groups that may be at higher risk of
an adverse BP outcome with increased Pb biomarkers, including those with high stress, certain genetic
variants, and minority populations.
Recent studies continue to provide consistent evidence that exposure to Pb is associated with
increased BP and hypertension risk. The majority of recent studies evaluating Pb biomarkers and changes
in BP or hypertension are cross-sectional, which can be useful for assessing concurrent associations
between blood Pb and increased BP or hypertension risk. A smaller number of studies implemented a
longitudinal study design, useful for evaluating long-term effects of elevated Pb biomarkers. Generally,
the evidence continues to indicate that changes in BP are most strongly associated with concurrent BLLs,
whereas increased risk of hypertension is more likely to be associated with cumulative Pb measures (such
as bone Pb levels).
4.3.1.1 Blood Pressure
Several recent studies specifically evaluated SBP and DBP, while other studies examined changes
in PP and MAP. Study-specific details, including blood/bone Pb levels, study population characteristics,
confounders, and selected results from these studies, are highlighted in Table 4-3. Studies in Figure 4-1
are standardized to be interpreted as changes in BP associated with a 1 (ig/dL increase in BLL or a
10 jj.g/g increase in bone Pb level. Study details in Table 4-3 include standardized results as well as results
that could not be standardized on the basis of information provided in each paper. Many of these studies
evaluated this association cross-sectionally. Specifically, most used population-level cross-sectional study
designs using NHANES (Huang. 2022; Everson et al.. 2021; Teve et al.. 2020; Obcng-Gvasi. 2019;
Obeng-Gvasi et al.. 2018; Hara et al.. 2015; Hicken et al.. 2013; Zota et al.. 2013; Hicken et al.. 2012;
Scinicariello et al.. 2011). Korea National Health and Nutrition Examination Survey (KNHANES) (Lee et
al.. 2016a'). a Canadian population-level survey (Canadian Health Measures Survey) (Bushnik et al..
2014). or a Chinese longitudinal survey (China National Human Biomonitoring) (Qu et al.. 2022). These
types of population-level cross-sectional studies have the advantage of assessing relatively low average
blood Pb (<5 (ig/dL) levels with concurrent BP measurements among a large sample size of participants.
A single study (Scinicariello et al.. 2010) used this type of data in the 2013 Pb ISA to evaluate BLLs and
changes in BP measurements. Additionally, two recent studies longitudinally evaluated the association
between biomarkers of Pb exposure and BP changes (Yu et al.. 2020; Bulka et al.. 2019).
Exernal Review Draft
4-9
DRAFT: Do not cite or quote
-------
Reference Population
SBP:
-fAlmeida Lopes et al, 2017 Adults >40 Cambe, Brazil
tHuang et al, 2022
NHANES
All
Men
Mexican American
Other Hispanic
Non-Hispanic White
Non-Hispanic Black
Other Race
Women
Mexican American
Other Hispanic
Non-Hispanic White
Non-Hispanic Black
Other Race
Pb distribution Pb biomarker
Geometric mean: 1.97 (95%CI: 1.90-2.04; Blood
Mean (SD): 1.73(1.71) Blood
tTeye et al, 2020
tEverson et al, 2021
Glenn et al, 2006
Weaver et al, 2008
Scinicariello et al, 2010
NHANES
NH White
NH Black
Hispanic
Other
NHANES
Korean Pb Workers
Short term: Longitudinal blood Pb
Short term: Concurrent blood Pb
Long term: Longitudinal blood Pb
Long term: Concurrent blood Pb
Korean Pb Workers
Median (IQR)
Men: 1.50 (0.99,2.29)
Women: 1.06(0.69,1.60)
Men: 1.60(1.00,2.60)
Women: 1.11 (0.71,1.77)
Men: 1.58(0.99,2.43)
Women: 0.95 (0.62,1.51)
Men: 1.54(1.05,2.39)
Women: 1.16 (0.75,1.79)
Median: 1.5
Mean (SD): 30.0(16.7)
Mean (SD): 75.1 (101.1)
Mean (SE)
Overall: 2.99 (0.99)
Non-Hispanic White: 2.87 (0.09)
Non-Hispanic Black 3.59 (0.20)
Mexican American 3.33 (0.11)
Blood
Blood
Blood
Patella
AL = allostatic load; BP = blood pressure; HFE C282Y = mutant of the HFE wildtype; CI = confidence interval; DBP = diastolic blood pressure; GSE = geometric standard error; HFE
H63D = mutant of the HFE wildtype; HFE = hemochromatosis gene; IQR = interquartile range; NAS = Normative Aging Study; NH = non-Hispanic; NHANES = National Health and
Nutrition Examination Survey; OR = odds ratio; Pb = lead; Q# = quartile number; RR = relative risk; SBP = systolic blood pressure; SD = standard error; SE = standard error.
Note: Effect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb. If the Pb biomarker is log-transformed, effect estimates are standardized to
the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated interval. Categorical effect
estimates are not standardized.
Figure 4-1 Association between biomarkers of Pb exposure and blood pressure.
Exernal Review Draft
4-10
DRAFT: Do not cite or quote
-------
Figure 4-1 (Continued) Association between biomarkers of Pb exposure and blood pressure.
DBP:
tAlmeida Lopes et al, 2017 Adults, >40 Cambe, Brazil
tHuang et al, 2022
tTeye et al, 2020
tEverson et al, 2021
Scinicariello et al, 2010
PP:
Zhang etal, 2010
NHANES
All
Men
Mexican American
Other Hispanic
Non-Hispanic White
Non-Hispanic Black
Other Race
Women
Mexican American
Other Hispanic
Non-Hispanic White
Non-Hispanic Black
Other Race
Geometric mean: 1.97 (95%CI:1.90-2.04; Blood
Mean (SD): 1.73(1.71)
Blood
NHANES
Median (IQR)
Blood
NH White
Men: 1 50 (0.99,2.29)
Women. 1.06 (0.69,1.60)
NH Black
Men: 1.60 (1.00,2.60)
Women: 1.11 (0.71,1.77)
Hispanic
Men: 1.58 (0.99,2.43)
Women: 0.95 (0.62,1.51)
Other
Men: 1.54(1.05,2.39)
Women: 1.16(0.75,1.79)
NHANES
Median: 1.5
Blood
NHANES III
Mean (SE)
Blood
Overall: 2.99 (0.99)
Non-Hispanic White: 2.87 (0.09)
Non-Hispanic Black 3.59 (0.20)
Mexican American 3.33 (0.11)
NAS men
HFE WikJtype
18(12-27)
Tibia
HFE H63D
19(14-26)
Tibia
HFE C282Y
20(14-27)
Tibia
Any HFE variant
19(14-27)
Tibia
HFE Wildtype
26(17-34)
Patella
HFE H63D
27(19-37)
Patella
HFE C282Y
25(17-37)
Patella
Any HFE variant
26(18-37)
Patella
I I I I I
0.00 0.50 1.00 1.50 2.00
Change in BP (mmHg 95% CI) per 1 ng/dl_ increase in blood Pb or 10 ng/g increase in bone Pb
2.50
Exernal Review Draft
4-11 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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Figure 4-1 (Continued) Association between biomarkers of Pb exposure and
blood pressure.
Many nationally representative cross-sectional studies evaluated association between increases in
BLLs and changes in either SBP or DBP using continuous Pb biomarkers. Generally, increases in BLLs
were concurrently associated with increases in both SBP and DBP (Huang. 2022; Qu et al.. 2022; Teve et
al.. 2020; Lee et al.. 2016a; Hara et al.. 2015; Hicken et al.. 2013; Scinicariello et al.. 2011). However,
some nationally representative studies noted null associations for SBP, but positive associations for DBP
(Obcng-Gvasi et al.. 2018; Bushnik et al.. 2014; Zota et al.. 2013). while others noted positive
associations for SBP and null associations for DBP (Everson et al.. 2021). Studies containing the
necessary information to standardize effect estimates to a 1 (ig/dL increase in blood Pb or a 10 |ig/g
increase in bone represent similar trends (Figure 4-1) and conclusions as studies that did not contain the
necessary information needed for standardization (Figure 4-1 and Table 4-3). Specifically, in a
KNHANES (2008-2013) analysis, Lee et al. (2016a) reported a 0.71 mmHg increase in DBP with each
doubling of blood Pb (95% CI: 0.29, 1.13 mmHg) and a similar association with SBP (0.73 mmHg [95%
CI: 0.09, 1.36 mmHg]). In an NHANES (1999-2006) analysis Scinicariello et al. (2011) indicated a
1.07 mmHg increase in SBP (95% CI: 0.384, 1.76 mmHg) and a 0.71 mmHg increase in DBP (95% CI:
0.18, 1.24 mmHg) for a twofold increase in BLL. In contrast, in an analysis of more recent NHANES
cycles (2007-2010), Obeng-Gvasi et al. (2018) noted a 0.268 mmHg increase in DBP (95% CI: 0.079,
0.458 mmHg), but reported a null association between ln-Pb and SBP (0.052 mmHg [95% CI: -0.233,
0.458 mmHg]) (Table 4-3).
Several smaller cross-sectional studies have also examined the relationship between Pb
biomarkers and BP (Yanetal.. 2022; Xu et al.. 2021; Chung et al.. 2020; Wang et al.. 2020; Guo et al..
2019; Lopes et al.. 2017; Gambelunghe et al.. 2016; Ettinger et al.. 2014). These studies tended to support
the larger nationally representative studies. Several studies indicated positive linear increases in both SBP
and DBP (Yan et al.. 2022; Chung et al.. 2020; Wang et al.. 2020; Gambelunghe et al.. 2016). For
example, a moderately sized study in Taiwan noted an increase in SBP (1.34 mmHg [95% CI: 0.34,
2.52 mmHg]) and DBP (0.69 mmHg [95% CI: 0.01, 1.37 mmHg]) per 1 (ig/dL increase in blood Pb
(Chung et al.. 2020). Additionally, a large cohort (with a cross-sectional component) in Malmo, Sweden
(n = 4,452) assessed BP and BLLs in the early nineties (1991-1994). This population was likely exposed
to historically high levels of Pb in the environment. The fully adjusted model indicated positive increases
in both SBP (1.8 mmHg [95% CI: 0.52, 3.08 mmHg]) and DBP (1.4 mmHg [95% CI 0.57, 2.54 mmHg])
when comparing the highest quartile of BLLs (mean 4.7 (ig/dL) with the lower three quartiles (mean 1.5-
2.8 (ig/dL) (Gambelunghe et al.. 2016). Yan et al. (2022) cross-sectionally evaluated a Haitian population
with relatively higher BLLs (geometric mean [GM]: 4.73 (ig/dL). This study had a high limit of detection
(3.3 (ig/dL), however, and -30% of the study population had BLLs below the limit of detection. Yet, this
study indicated positive associations between blood Pb and SBP (2.42 mmHg [95% CI: 0.36, 4.49]) and
DBP (1.96 mmHg [95% CI: 0.56, 3.37]) when comparing the highest quartile (6.5-58.2 (ig/dL) with the
lowest (<3.3 (ig/dL).
Exernal Review Draft
4-12
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
In addition, a study among older adults (aged 40-75) living in rural southwest China compared
the associations between BLL quartiles and BP measurements among those subsisting off rice and
vegetables grown in a polluted region (Cd) concentration >0.2 mg/kg) with the associations among those
in an unpolluted region (Cd <0.05 mg/kg) (Wang et al.. 2020). In the polluted region, this study reported
positive associations with both SBP and DBP when the highest BLL quartile (>4.6 (ig/dL) was compared
with the lowest BLL quartile (<2.1 (ig/dL) (Figure 4-2). In contrast, there was no relationship observed in
the unpolluted region (Figure 4-3). The authors of this study hypothesize that this discrepancy in
association between polluted and unpolluted regions may be due in part to differences in mean BLLs in
the polluted (3.5 (ig/dL) and unpolluted (2.6 j^ig/dL) areas, or that more pollution may modify the
association between blood Pb and BP, in addition to the small sample size in the unpolluted area
(n = 214) compared with the polluted area (n = 602).
Q1(<20 80)
02(20 80-34 10)
Pb Q3(34 10-46 40)
Q4(>46 40)
Hypertension
jMrend SBP
p- trend
DBP
p-trend
i
0.438
0.00$
<0.001
2 3
Odds Ratios
0 S
Coefficients
10
0 5
Coefficients
10
DBP = diastolic blood pressure; Pb = lead; Q = quartile; SBP = systolic blood pressure.
Source: Adapted from Wang et al. (2020).
Figure 4-2 Association between blood Pb level quartiles and systolic blood
pressure, diastolic blood pressure, and hypertension, polluted
region of rural southwest China.
Q1(<16 00)
02(16 00-2375)
Pb 03(23 75-37 98)
Q4(>37 98)
Hypertension
ptrend SBF
p-trcnd DBP
0.567
0.769
3 6 9
Odds Ratios
-20 -10 0 10 20
Coefficients
p i rend
0.524
-5 0 S 10
Coefficients
DBP = diastolic blood pressure; Pb = lead; Q = quartile; SBP = systolic blood pressure.
Source: Adapted from Wang et al. (2020).
Figure 4-3 Association between blood Pb level quartiles and systolic blood
pressure, diastolic blood pressure, and hypertension, unpolluted
region of rural southwest China.
In contrast, a study in Cambe, Brazil (n = 948) indicated a null association between BLLs and
SBP. However, when comparing the 90th percentile (6.03 j^ig/dL) with the 10th percentile (0.74 (ig/dL),
Exernal Review Draft
4-13
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
there was a small but precise increase in DBP (0.005 mmHg [95% CI: 0.002, 0.008 mmHg]) per 1 (ig/dL
increase in blood Pb (Lopes et al.. 2017). Additionally, a recent study cross-sectionally evaluated the
association between blood Pb and BP among participants in the Gulf Long-Term Follow-up (GuLF) study
(Xu et al.. 2021). The GuLF study is a longitudinal cohort of individuals involved in the 2010 Deepwater
Horizon oil spill. Baseline blood Pb and BP measurements were obtained between 2011 and 2013. BLLs
within this study were low overall (quartile 1: 0.06 (ig/dL, quartile 4: 0.27 (ig/dL). This study indicated
null associations between the highest quartile and the lowest quartile for SBP (-0.96 [95% CI: -4.13,
2.22]) and DBP (-0.01 mmHg [95% CI: -2.21, 2.10]).
Additionally, a smaller number of cross-sectional studies evaluated either SBP or DBP
categorically. Typically, these studies dichotomized either SBP or DBP at a particular clinically relevant
threshold prior to conducting categorical statistical analyses. The results of these studies were more mixed
compared with the results presented using continuous BLLs, presented above. For example, Ettinger et al.
(2014) evaluated the association between BLLs and high SBP (>130 mmHg) and high DBP (>85 mmHg)
among young adults (aged 25-45) of African descent. This study yielded null results for both high SBP
(>130 mmHg) (OR: 1.69 [95% CI: 0.55, 5.15]) and high DBP (>85 mmHg) 2.20 [95% CI: 0.59, 8.16])
when comparing blood Pb values above and below the median (1.66 (ig/dL). In contrast, a different study
among young adults (aged 18-44) indicated positive increases in the odds of SBP >120 mmHg (OR: 1.21
[95% CI: 1.07, 1.38]) and DBP >80 mmHg (OR: 1.32 [95% CI: 1.10, 1.58]) when comparing BLLs
above and below 5 (ig/dL . These cross-sectional results were similar to the results generated from studies
evaluating concurrent BLLs and hypertension (see Section 4.3.1.2).
While most cross-sectional studies evaluated the association between concurrent BLLs and SBP
and DBP, some studies assessed in the 2013 Pb ISA also considered concurrent bone Pb measurements
and BP. Bone Pb tends to represent cumulative or long-term exposure to Pb, whereas BLLs are
representative of recent exposure. Several analyses previously presented in the 2013 Pb ISA indicated
mixed results for the association between bone Pb levels and SBP and DBP, although associations were
generally positive. For example, (Elmarsafawv et al.. 2006) evaluated whether calcium intake affects the
relationship between SBP and bone Pb levels in a cross-sectional analysis of the NAS cohort.
(Elmarsafawv et al.. 2006) reported increases in SBP for each 10 jj.g/g increase in bone Pb level in both
high (>800 mg/day) and low calcium (<800 mg/day) groups. However, the low calcium group had a
substantially larger increase (4.00 mmHg [95% CI: 1.05, 6.95]), compared with the high calcium group
1.90 mmHg [95% CI: 0.10, 3.70]). Although these results are not presented in Figure 4-1, they are
comparatively larger than results presented for SBP and BLLs. No recent studies evaluated the
association between bone Pb levels and BP.
In addition to the numerous cross-sectional studies previously mentioned, several recent studies
longitudinally evaluated the relationship between biomarkers of Pb exposure and BP measurements.
Bulka et al. (2019) evaluated a small Bangladeshi cohort with baseline BLLs (median: 8.5 (ig/dL)
measurements between April 2006 and August 2009, from an arsenic-endemic area. Residents in this area
Exernal Review Draft
4-14
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
24
25
26
27
28
29
30
31
32
33
34
are chronically exposed to high levels of Pb in the air, water, and other industrial sources. BP was
assessed biennially for a total of 6 years. This study indicated that SBP was higher in the highest quartile
of baseline blood Pb compared with the lowest quartile, corresponding to a 1.16 mmHg annual increase
(95% CI: 0.21, 2.11 mmHg). Results for DBP (0.53 mmHg [95% CI: -0.10, 1.16 mmHg]) and PP
(0.63 mmHg [95% CI: -0.08, 1.34]) were smaller in magnitude, compared with SBP when comparing the
highest quartile of BLLs to the lowest quartiles. All analyses considered several appropriate confounders,
in addition to urinary arsenic (creatinine standardized). While there was an annual increase in the
association between blood Pb and SBP, BP measurements remained stable across visits, but
antihypertensive medication use increased from 7.5% at baseline to 15.3% at the last visit, which was
controlled for as a confounder in all statistical models. A longitudinal study in Belgium evaluated the
association between baseline BLLs (collected between 1985 and 2005) and BP measured an average of
9.4 years following blood Pb measurement. For each doubling of BLLs there were null associations
between peripheral SBP (2.41 mmHg [95% CI: -0.38, 5.20 mmHg]), DBP (0.50 mmHg [95% CI: -1.07,
2.07 mmHg]), and PP (1.91 mmHg [95% CI: -0.32, 4.14 mmHg]). Similarly, the association between a
doubling of BLLs and central SBP, DBP, and PP were also null. Overall, associations from these studies
remained stable even after controlling for Cd at baseline and considering the high endemic levels of
arsenic in the Bangladeshi cohort.
Several studies also assessed PP in addition to SBP and DBP. To reiterate, PP is the force the
heart requires to contract and is calculated by subtracting DBP from SBP. Overall, there was a null
relationship between BLLs and PP in both cross-sectional (Hara et al.. 2015; Scinicariello et al.. 2010
Perlstein. 2007. 194019) and cohort analyses (Yu et al.. 2020; Bulka etal.. 2019). However, the
relationship between bone Pb levels and PP was positive in cross-sectional analyses (Zhang et al.. 2010;
Perlstein et al.. 2007). Hara et al. (2015) additionally evaluated MAP, which is the average BP during a
single cardiac cycle. This study indicated an increase in MAP associated with BLLs.
The 2013 Pb ISA included two different meta-analyses focused on the relationship between Pb
exposure biomarkers and BP changes or hypertension status. Nawrot et al. (2002) included over 30 cross-
sectional and prospective studies on BLLs and BP, including >58,000 adults. This meta-analysis
concluded that each doubling of concurrent BLLs was associated with a 1 mmHg increase in systolic BP
and a 0.6 mmHg increase in diastolic BP. Furthermore, Navas-Acien et al. (2008) conducted a similar
meta-analysis based on bone Pb measurements (three prospective, five cross-sectional). The pooled
estimate from the cross-sectional studies indicated an increase in SBP of 0.26 mmHg (95% CI: 0.02,
0.50) per 10 jxg/g tibia Pb. When considering hypertension, pooled results indicated increased odds of
hypertension (OR: 1.04 [95% CI: 1.01, 1.07]) per 10 jxg/g increase in tibia Pb and 1.04 (95% CI: 0.96,
1.12) per 10 jxg/g increase in patella Pb.
Exernal Review Draft
4-15
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
24
25
26
27
28
29
30
31
32
33
34
35
4.3.1.1.1 Effect Measure Modification
Several recent studies went beyond only evaluating the direct association between BLL and BP,
but also evaluated effect measure modification (EMM) by a variety of outcomes, including race (Huang.
2022; Teve et al.. 2020; Hara et al.. 2015; Hicken et al.. 2013; Hicken et al.. 2012; Scinicariello et al..
2011). sex (Gambelunghe et al.. 2016; Hara et al.. 2015; Bushnik et al.. 2014; Hicken et al.. 2013; Hicken
et al.. 2012; Scinicariello et al.. 2011). age (Huang. 2022; Obeng-Gvasi. 2019; Gambelunghe et al.. 2016;
Bushnik et al.. 2014). stress/depression (Hicken et al.. 2013; Zota et al.. 2013). genetic variations (Jhun et
al.. 2015). and smoking (Gambelunghe et al.. 2016). These analyses can help to further highlight specific
subgroups of the population that may have an increased risk of elevated BP associated with Pb
biomarkers of exposure.
Race was a common measure to evaluate differential effects of biomarkers of Pb exposure and
BP. An NHANES (1999-2016) study indicated statistically significant increases in SBP and DBP among
non-Hispanic White individuals (SBP: 0.34 mmHg [95% CI: 0.11, 0.57 mmHg], DBP: 0.38 mmHg [95%
CI: 0.19, 0.57 mmHg]) and non-Hispanic Black individuals (SBP: 0.67 mmHg [95% CI: 0.29,
1.05 mmHg], DBP: 0.36 mmHg [95% CI: 0.06, 0.66 mmHg]), with each 1 (ig/dL increase in BLLs . As
described in the 2013 Pb ISA, Scinicariello et al. (2010) used NHANES III (1988-1994) to evaluate BP
and BLLs by race/ethnicity. This study indicated a greater increase in both SBP (1.615 mmHg [95% CI:
1.007, 2.223 mmHg]) and DBP (1.261 mmHg [95% CI: 0.716, 1.805 mmHg]) among non-Hispanic Black
individuals per 1 jj.g/dL of blood Pb, compared with non-Hispanic White and Mexican American
individuals (Figure 4-1, Table 4-3).
Stratification by sex was less common, and the results were less consistent than for race. The
Malmo Diet and Cancer Study evaluated the relationship between BLLs and changes in BP stratified by
sex (Gambelunghe et al.. 2016). In this study, sex did not modify the positive association between BLLs
and SBP or DBP increases. Additionally, an NHANES (2003-2010) analysis reported increased SBP and
DBP for each doubling of BLLs among both sexes (Hara et al.. 2015) (Figure 4-4, Table 4-3).
Several studies evaluated both sex and racial differences together in the association between
BLLs and changes in BP. In a cross-sectional study using NHANES (2003-2010), Hara et al. (2015)
evaluated SBP and DBP stratified by both sex and race (Figure 4-4). This study indicated that
qualitatively, compared with White females, Black females experienced a greater increase in magnitude
in SBP with each doubling of BLLs. White females had a greater increase in DBP, compared with Black
females, however. Compared with White males, Black and Hispanic males had greater increases in SBP
with each doubling of BLLs. White men had a similar increase in DBP, however, compared with Black
men (Figure 4-4, Table 4-3). Scinicariello et al. (2011) used NHANES (1999-2006) to evaluate sex and
racial disparities for changes in BP and BLLs (Figure 4-5). This study reported the largest SBP increases
among Black males (2.40 mmHg [95% CI: 0.91, 3.69 mmHg]) and Black females (2.40 mmHg [95% CI:
0.17, 4.63 mmHg]) associated with a doubling of BLLs, compared with other races. Conversely, Mexican
Exernal Review Draft
4-16
DRAFT: Do not cite or quote
-------
1 American males had decreases in DBP associated with a doubling of BLLs (-1.34 mmHg [95% CI:
2 -2.63, -0.05 mmHg]). Huang (2022) also conducted an analysis using NHANES (1999-2006). This
3 study generally noted similar positive associations with increased SBP and DBP among non-Hispanic
4 White males and females and non-Hispanic Black males and females; however, there were null
5 associations for Mexican American and other Hispanic males and females for both SBP and DBP.
Population
SBP:
All
Women
Black
Hispanic
White
Men
Black
Hispanic
White
DBP:
All
Women
Black
Hispanic
White
Men
Black
Hispanic
White
PP:
All
Women
Black
Hispanic
White
Men
Black
Hispanic
White
MAP:
All
Women
Black
Hispanic
White
Men
Black
Hispanic
White
Pb distribution
1.37(0.88-2.10)
1.21 (0.80-1.78)
1.22(0.80-1.86)
1.86(1.20-2.85
1.94 (1.25-2.83)
1.73 (1.16-2.57)
1.37(0.88-2.10) *
1.21 (0.80-1.78)
1.22 (0.80-1.86)
1.86(1.20-2.85
1.94 (1.25-2.83)
1.73 (1.16-2.57)
1.37 (0.88-2.10)
1.21 (0.80-1.78)
1.22 (0.80-1.86)
1.86(1.20-2.85
1.94 (1.25-2.83)
1.73 (1.16-2.57)
1.37 (0.88-2.10)
1.21 (0.80-1.78)
1.22 (0.80-1.86)
1.86(1.20-2.85
1.94 (1.25-2.83)
1.73 (1.16-2.57)
—I 1—
0.00 0.50
mmHg (per doubling of blood Pb)
EE
LCL
UCL
0.76
0.38
1.13
0.58
0.01
1.17
¦* 1.18
-0.19
2.55
- 0.56
-0.57
1.69
0.61
-0.18
1.40
0.79
0.30
1.27
1.61
0.45
2.76
+ 0.95
0.05
1.84
0.65
-0.03
1.32
0.43
0.18
0.68
0.43
0.07
0.80
0.52
-0.34
1.37
-0.13
-0.81
0.55
0.73
0.23
1.24
0.47
0.13
0.81
0.81
-0.04
1.66
-0.03
-0.64
0.58
0.70
0.24
1.17
0.33
-0.02
0.67
0.15
-0.37
0.67
+ 0.64
-0.56
1.84
0.69
-0.29
1.67
-0.11
-0.84
0.61
0.32
-0.13
0.77
+ 0.81
-0.27
1.88
+ 0.98
0.14
1.82
-0.06
-0.68
0.57
0.54
0.29
0.79
0.48
0.10
0.86
0.73
-0.16
1.62
0.10
-0.62
0.82
0.69
0.18
1.21
0.57
0.24
0.91
+ 1.08
0.26
1.90
0.30
-0.30
0.89
0.68
0.23
1.14
1.00
1.50
DBP = diastolic blood pressure; EE = effect estimate; MAP = mean arterial pressure; Pb = lead; PP = pulse pressure; SBP = systolic
blood pressure; LCL = lower confidence limit; UCL = upper confidence limit.
Note: Pb distribution presented as geometric mean (IQR).
Source: Hara et al. (2015).
Figure 4-4 Blood pressure (systolic and diastolic) and a doubling of blood
Pb levels stratified by sex and race National Health and Nutrition
Examination Survey (2003-2010).
Exernal Review Draft
4-17
DRAFT: Do not cite or quote
-------
Population
SBP:
All
White:
Men
Woman
Black:
Men
Women
Mexican-American:
Pb distribution
2.2 (0.03)
1.55 (0.02)
2.44 (0.05)
1.81 (0.06)
Men
Women
DBP;
All
White:
Hen
Women
Black:
Men
Women
2.47 (0.06)
1.56 (0.04)
2.2 (0.03)
1.55 (0.02)
2 44 fO 05)
1 81 i0 06!
MeMcan-^mencan
Men
Women
PP:
All
White
Men
Women
Black
Men
Women
2.47 (0.06)
1.56 (0.04)
2.2 (0.03)
1.55 (0.02)
2,44 (0.05)
1.81 (0.06)
Mexican-American:
Men 2.47 (0.06)
Women 1.56(0.04)
EE LCL DCL
1.07 0.38 1.76
0.87 -0.17 1.91
0.89 -0.19 1.97
~2 30 0.91 3.69
~2 40 0.17 4.63
0.10 -1.27 1.47
-0.03 -1.28 1.22
0.71 0.18 1.24
0.S0 0.02 1.78
0.95 0.21 1,69
~2.75 1.14 436
0.30 -1.29 1.89
-1.34 -2.83 -0.05
-0.74 -1.60 0.12
0.37 -0.30 1.04
-0.02 -1.10 1.06
-0.03 -1.25 1.19
-0.42 -2.24 1.40
-2,21 -0.08 4.50
1.42 0.05 2.79
0.70 -0,53 1.93
—I 1 1 1 I I 1 1
-1.00 0.00 0.50 1.00 1.50 2.00 2.50 3.00
mmHg (per doubling of Dlood Pb)
DBP = diastolic blood pressure; EE = effect estimate; PP = pulse pressure; SBP = systolic blood pressure; LCL = lower confidence
limit; UCL = upper confidence limit.
Note: Pb distribution presented as mean (SE).
Source: Scinicariello et al. (2011).
Figure 4-5 Blood pressure (systolic, diastolic, and pulse pressure) and a
doubling of blood Pb level stratified by sex and race National
Health and Nutrition Examination Survey (1999-2006).
1 Additionally, using NHANES (2001-2008) Hicken et al. (2012) indicated differences in SBP,
2 DBP, and PP when comparing White and Black males and females (Figure 4-6). The associations
3 between BLLs and SBP, DBP, and PP were consistently higher among Black females compared with
4 White females. Furthermore, this discrepancy by race and sex was also altered by educational attainment
5 (Figure 4-7) and family poverty (Figure 4-8).
Exernal Review Draft
4-18
DRAFT: Do not cite or quote
-------
Population
SBP:
White Men
< High School
Ł High School
Poor
Non-Poor
White Women
< High School
a High School
Poor
Non-Poor
Black Men
< High School
a High School
Poor
Non-Poor
Black Women
< High School
2 High School
Poor
Non-Poor
DBP:
White:
Men
Women
Black:
Men
Women
PP:
White:
Men
Women
Black:
Men
Women
Blood Pb
1.7(1.7)
1.2 (1.2)
1.9 (1.8)
1.4(1.3)
EE LCL UCL
1.7(1.7)
1.2 (1.2)
1.9 (18)
1.4(1.3)
1.7 (1.7)
1.2 (1.2)
1.9 (1.8)
1.4(1.3)
0.3
0.3
0.3
0.6
0.3
0.8
1.0
-0.88
-2.25
-2.25
-1.16
-1.07
-0.96
-2.14
0.3 -1.86
0.7 -1.06
1.1
2.8
4.1
2.2
4.8
0.4
4.0
1 6.7
2.9
" 4.3
•3.7
-0.47
1.43
1.55
0.24
4.21
-1.36
1.84
2.58
-0.24
1.36
-0.02
1.3
2.1
1.48
2.85
2.85
2.36
1.67
2.56
4.14
2.46
2.46
2.67
4.17
6.65
4.16
5.39
2.16
6.16
10.82
6.04
7.24
7.42
0.6 -0.18 1.38
1.3 0.32 2.28
1.5 -0.07 3.07
1.9 0.53 3.27
-0.3 -1.28 0.68
-0.5 -1.87 0.87
-0.27
0.34
2.87
3.86
1 1 1 1—
0.00 1.00 2.00 3.00
mmHg (per doubling of blood Pb)
4.00
5.00
6.00
7.00
DBP = diastolic blood pressure; EE = effect estimate; LCL = lower confidence limit; Pb = lead; PP = pulse pressure; SBP = systolic
blood pressure; UCL = upper confidence limit.
Note: Pb distribution presented as mean (median).
Source: Hicken et al. (2012).
Figure 4-6 Blood pressure (systolic, diastolic, and pulse pressure) and a
doubling of blood Pb levels stratified by sex and race, National
Health and Nutrition Examination Survey (2001-2008).
Exernal Review Draft
4-19
DRAFT: Do not cite or quote
-------
a
White, high school
— Black, < high school Black, Ł high school
b
-0.53 -0.33 -0.12 0.00 0.18 0.34 0.48 0.70 0.97
Blood Lead, jig/dL
SBP = systolic blood pressure.
Note: Association between SBP and log-transformed BLL by educational attainment for men (a) and women (b).
Source: Hicken et al. (2012).
Figure 4-7 Interactive effect between blood Pb levels, race, and education
level, National Health and Nutrition Examination Survey (2001—
2008),
Exernal Review Draft
4-20
DRAFT: Do not cite or quote
-------
118-1 T , , , T , , ,
-0.19 0.05 0.25 0.40 0.53 0.60 0.83 1.03 1.30
Blood Lead, |ag/dL
White, PIR < 1.85 (poor) White, PIR >1.85 (nonpoor)
Black, PIR < 1.85 (poor) ¦¦¦¦« Black, PIR >1.85 (nonpoor)
Blood Lead, |.ig/dL
SBP = systolic blood pressure, PIR= poverty income ratio.
Note: Association between SBP and log-transformed BLL by poverty level for men (a) and women (b).
Source: Hicken et al. (2012).
Figure 4-8 Interactive effect between blood Pb levels, race, and poverty level,
National Health and Nutrition Examination Survey (2001-2008).
Exernal Review Draft
4-21
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Another NHANES (2005-2008) analysis evaluated racial differences in the effects of a doubling
of BLLs on changes in BP, stratified by depressive symptoms (Hickcn et al.. 2013). First, the association
between increased SBP and a doubling of BLLs was larger among Black participants (3.2 mmHg [95%
CI: 1.5, 5.0 mmHg]) than White participants (1.0 mmHg [95% CI: -0.3, 2.4 mmHg]). However, increases
in DBP were similar when comparing Black and White participants. This study further stratified the data
by considering depressive symptoms, defined using the Patient Health Questionnaire (PHQ-9), which
may indicate psychosocial stress. The PHQ-9 score was parsed into low (score <3) and high (score >3).
The association between BLLs and BP (both SBP and DBP) was greater among those with high PHQ-9
scores. High psychosocial stress (PHQ-9 score >3) particularly modified the association between blood
Pb and BP among Black individuals, compared with White individuals (Table 4-3). Specifically, a
doubling of BLLs was associated with a 5.6 mmHg (95% CI: 2.0, 9.2 mmHg) increase in SBP among
Black individuals with high levels of depression (PHQ-9 score >3), compared with only a 1.2 mmHg
(95% CI: -0.5, 2.9 mmHg) increase in SBP among White individuals with high levels of depression
(PHQ-9 score >3). Similarly, another NHANES (1999-2008) analysis evaluated the relationship between
BP and BLLs by allostatic load (AL), or a measure of the combined biologic burden due to chronic stress
(Zotaetal.. 2013). Overall, this study indicated there were greater odds of an association between BLLs
and having SBP >140 mmHg or DBP >90 mmHg among those with high AL (Figure 4-9).
Exernal Review Draft
4-22
DRAFT: Do not cite or quote
-------
3
%/>
k_
CL
"O
O
o
jO
(A)
1 (B)
O o
¦O
Si
ro
>
CL>
O
#
kn
o>
cr
O
"D
"O
<
1 r
2 3 4
Quintle of lead exposure
3
¦o
<
"i 1 r
2 3 4
Quindle of lead exposure
T3
<
t 1 1 r
2 3 4 5
Quintile of lead exposure
1 1 1 r
2 3 4 5
Quintile of lead exposure
In Figures A and B, high allostatic load (AL) is defined as the top 50th percentile whereas in Figures C and D, high AL is defined as
the top 20th percentile.
AL = allostatic load; CI = confidence interval; OR = odds ratio.
Source: Zota et al. (2013).
Figure 4-9 Association (odds ratio) between blood Pb levels and systolic
blood pressure >140 mmHg or diastolic blood pressure
>90 mmHg by allostatic load, National Health and Nutrition
Examination Survey (1999-2008).
Exernal Review Draft
4-23
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
In a previous NAS study included in the 2013 Pb ISA, Peters et al. (2007) assessed self-reported
(perceived) stress using a questionnaire to evaluate if perceived stress modifies the association between
bone Pb levels (tibia and patella) and changes in BP. In this study, there was an interaction between SBP
and tibia Pb levels, indicating that increases in tibia Pb levels were associated with increased SBP among
those with high stress (Figure 4-10). However, there was no statistically significant interaction with high
stress for the association between increased SBP and patella Pb levels or for DBP and either bone Pb
measurement.
140-
135-
3 130-
E
E 125-
| 120-
115-
110
-10 0 10 30 50 70 90
Tibia lead (ug/g)
SBP = systolic blood pressure.
Source: Peters et al. (2007).
Figure 4-10 Relationship between tibia Pb levels and systolic blood pressure
by perceived stress, Normative Aging Study cohort.
Several studies also evaluated the relationship between biomarkers of Pb exposure and changes in
BP stratified by age. Specifically, an NHANES (2009-2016) analysis evaluated the odds of the
associations between BLLs and increased SBP (>120 mmHg) or DBP (>80 mmHg) for middle-aged (46-
65 years) and young (aged 18-44 years) adults (Obcng-Gvasi. 2019). This study demonstrated similar
odds of increased SBP for middle-aged adults (OR: 1.32 [95% CI: 1.14, 1.52]) as with young adults (OR:
1.21 [95% CI: 1.07, 1.38]) when comparing BLLs above and below 5 (ig/dL. The association between
BLLs and increased DBP was also similar in middle-aged (OR: 1.16 [95% CI: 0.98, 1.38]) and young
(OR: 1.32 [95% CI: 1.10, 1.58]) adults. The young adults included in this analysis were likely not
exposed to air emissions associated with leaded gasoline in the past, and therefore can help disentangle
the effects of past high Pb exposures on CVD health endpoints. Additionally, the Malmo Diet and Cancer
o High perceived stress
• Low perceived stress
- - Trend (high stress)
Trend (law stress!
T
T
Exernal Review Draft
4-24
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
Study also considered stratification by both sex and age when evaluating associations between BLLs and
changes in BP (Gambclunghc et al.. 2016). This study noted marginally increased associations between
BLLs and increased SBP among adults aged >57 years (2.4 mmHg [95% CI: 1.20, 3.60 mmHg])
compared with adults <57 years (1.3 mmHg [95% CI: -0.55, 3.15]), when comparing the highest blood
Pb quartile (4.7 j^ig/dL) with the lowest three quartiles (range 1.5-2.8 (ig/dL). There were no differences,
however, in the association between BLLs and increased DBP by age. However, this cohort was likely
exposed to air emissions associated with leaded gasoline in the past. In addition, a cross-sectional study,
using the Canadian Health Measures Survey (2007-2011), stratified by both age and sex (Bushnik et al..
2014) demonstrated a steep increase in SBP and DBP associated with BLLs, up to 3 (ig/dL, especially
among middle-aged adults (40-54 years) and men. Specifically, this study indicated that a 1 (ig/dL
increase in BLL would correspond with a 1-2 mmHg increase in SBP and a 2-3 mmHg increase in DBP
(Figure 4-11, Figure 4-12).
Exernal Review Draft
4-25
DRAFT: Do not cite or quote
-------
Systolic blood pressure (mm Hg)
3 4
Blood lead level (pg/dL)
¦ Model 1 (40 to 79 years)
¦ Model 3 (55 to 79 years)
Model 5 (Women, 40 to 79 years)
- Model 2 (40 to 54 years)*
¦¦ Model 4 (Men, 40 to 79 years)
* significant association between blood Pb level and systolic blood pressure (p < 0.05)
BMI = body mass index; HDL = high-density lipoprotein.
Source: Bushnik et al. (2014).
Figure 4-11 Relationship between blood Pb levels and systolic blood pressure
by sex and age, Canadian Health Measures Survey.
Exernal Review Draft
4-26
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
Diastolic blood pressure (mm Hg)
80 n
79-
78-
77-
76-
75-
74-
73-
72-
71-
70-
1 2 3 4 5 8 7
Blood lead level (pg>dL)
Model 1 *40 tc 79 years)* Model 2 {40to 54 years)*
— — — Model 3 (55 to 79 year.3) Model 4 (Men, 40 to 79 years)*
— — Model 5 {Women, 40 to 79 years)
* significant association between blood Pb level and diastolic blood pressure (p < 0.05)
BMI = body mass index; HDL = high-density lipoprotein.
Source: Bushnik et al. (2014).
Figure 4-12 Relationship between blood Pb levels and diastolic blood
pressure by sex and age, Canadian Health Measures Survey.
Certain genetic polymorphisms can be important in assessing the risk of increased BP as a result
of elevated levels of biomarkers of Pb exposure. In a longitudinal analysis of the NAS cohort, Jhun et al.
(2015) evaluated potential EMM by vitamin D receptors (VDR) between PP and bone level and BLL.
Genetic variations in VDR genes can potentially influence the accumulation, absorption, and retention of
Pb in the body. After the initial baseline bone Pb, blood Pb, and BP measurements, PP was reassessed
every 3-5 years. At the initial visit, an IQR increase in either tibia or patella Pb level was associated with
an increased PP among those with the variant (opposed to ancestral) genotype (single nucleotide
polymorphisms [SNPs] in Bsml, Taql, Apal, or Fokl). Although there was an association with PP and
tibia Pb levels by VDR genotype at baseline, this relationship appeared to diminish with time (Figure 4-
13). However, the three-way interaction terms between bone Pb levels, VDR receptor type, and time since
baseline was almost zero, indicating that VDR consistently modifies the association between bone Pb
I Aernnl Review Draft
4-27
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
levels and PP. In addition to genetic polymorphisms in VDR, the 2013 Pb ISA also evaluated studies that
assessed other genetic factors that may increase susceptibility to Pb. Specifically, ScinicarieHo et al
(2010) used NHANES III (1988-1994), to stratify by 5-aminolevulinic acid dehydratase (ALAD) status.
A critical mechanism of Pb toxicity is its ability to interact and inhibit key enzymes, such as A LAD, in
the heme biosynthetic pathway. This study indicated that non-Hispanic White carriers of the ALAD2
polymorphism had a greater increase in SBP and DBP associated with BLLs. However, in a South
Korean occupational study, BLLs were associated with increased SBP only, and there was no evidence of
EMM by either VDR or ALAD (Weaver et al.. 2008). In another evaluation of the NAS, Zhang et al.
(2010) examined changes in the hemochromatosis gene (HFE), which can promote excessive iron
absorption and is thought to also alter Pb biomarker concentrations. Two mutations to HFE (C282Y and
H63D) were examined within this older population. This study suggested that those with the H63D
mutation were more likely to have an increase in PP with a 10 j.ig/g increase in tibia (2.54 mml lg [95%
CI: 0.12, 4.96 mml Ig|) and patella (2.23 mmHg [95% CI: 0.23, 4.23 mmHgj) Pb levels. Taken together,
certain genetic polymorphisms appear capable of predisposing some groups to greater effects on BP
related to biomarkers of Pb exposure.
At baseline
_ 55.00
DO
I 54.00
Ł 53.00
S! 52.00
ZJ
| 51-00
50.00
-5 49.00
Q.
52.37
50.77
50.84
49.77
13 28
Tibia lead levels (fig/g)
-Ancestral -¦»- Variant
tfj
I
E
E
55.00
54.00
53.00
52.00
51.00
50.00
49.00
After 10 years since baseline
54.18 53.81
53.18
52.28
13 28
Tibia lead levels (ng/g)
-Ancestral -<•« Variant
BMI = body mass index; VDR = vitamin D receptor.
Source! Jhun et al. (2015).
Figure 4-13 Pulse pressure and tibial Pb levels, modified by vitamin D
receptor variant, Normative Aging Study cohort.
The Malmo Diet and Cancer Study further evaluated the association between BLLs and changes
in BP through stratification by smoking status (Gambelunghe et al.. 2016). The cross-sectional
component of the study indicated that smokers (ever-smokers) had a 3.9 mmHg (95% CI: 1.59,
6.21 mmHg) increase in SBP, compared with only a 0.6 mmHg (-1.46, 2.66 mmHg) increase among
never-smokers when comparing the highest quartile of BLLs (mean 4.7 ug/dL) with the lower three
quartiles (mean 1.5-2.8 (ig/dL). Similarly, smokers had a 1.6 mmHg (95% CI: 0.65, 2.54 mmHg) increase
in DBP, compared with a 1.1 mmHg (95% CI: -0.05, 2.25 mmHg) increase among never-smokers.
Exernal Review Draft
4-28
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
4.3.1.2
Hypertension
Fewer recent studies evaluated the relationship between biomarkers of Pb exposure and
hypertension. Study-specific details, including blood and bone Pb levels, study population characteristics,
potential confounders, and select results from these studies are highlighted in Table 4-4 and Figure 4-14.
Studies in Figure 4-4 are standardized to represent the risk of prevalent or incident hypertension
associated with a 1 (ig/dL increase in BLL or a 10 jj.g/g increase in bone Pb level. Study details shown in
Table 4-4 include standardized results as well as results that could not be standardized based on the
information provided in each paper. Generally, hypertension refers to chronic BP readings of
>140 mmHg for SBP and >90 mmHg for DBP, while prehypertension, typically thought of as a precursor
to chronic hypertension, is usually defined as SBP 120-139 mmHg or DBP 80-89 mmHg. However,
some studies may choose to define hypertension, or prehypertension, differently. Some cross-sectional
studies evaluated associations between biomarkers of Pb exposure and prevalent or preexisting
hypertension (Huang. 2022; Ou et al.. 2022; Xu et al.. 2021; Teve et al.. 2020; Wang et al.. 2020; Choi et
al.. 2018; Lopes et al.. 2017; Hara et al.. 2015; Bushnik et al.. 2014). However, other studies specifically
evaluated prehypertension, or SBP or DBP values that approach a predefined clinical definition of
hypertension (Qu et al.. 2022; Lee etal.. 2016b; Lee etal.. 2016a). Additionally, longitudinal studies
examined associations between baseline Pb biomarkers and incident, or newly developed hypertension
(Gambclunghc et al.. 2016) whereas other studies evaluated associations with hypertension that may not
respond to medication (resistant) or completely untreated (uncontrolled) hypertension using NHANES
(Miao et al.. 2020) or the NAS cohort (Zheutlin et al.. 2018).
Exernal Review Draft
4-29
DRAFT: Do not cite or quote
-------
fAlmeida Lopes et al, 2017 Cross-sectional Adults >40 Cambe, Brazil
Pb distribution Pb biomarker
Geometric mean: 1.97 (95%CI:1.90-2.04) Blood
TChoi et al, 2018
Cross-sectional
KNHANES
Curry Intake
Non-Curry Intake
Mean (SE): 2.01 (0.025)
fMiao etal, 2020
Cross-sectional NHANES
Mean (SE):
Male: 1.50 (0.02)
Female: 1.07 (1.01)
Uncontrolled Hyptertension
vs Non-hypertension
Male
Female
Uncontrolled Hypertension
vs Controlled Hypertension
Male
Female
Uncontrolled Hypertension vs
Controlled and Non-Controlled Hypertension
Male
Female
Cross-sectional
Overall
NH White
NH Black
Hispanic
Other
Median (IQR)
Men: 1.50(0.99,2.29)
Women: 1.06 (0.69,1.60)
Men: 1.60 (1.00,2.60)
Women: 1.11 (0.71,1.77)
Men: 1.58 (0.99,2.43)
Women: 0.95 (0.62,1.51)
Men: 1.54 (1.05,2.39)
Women: 1.16(0.75,1.79)
tHuang etal,2022
Cross sectional
NHANES
All
Women
Men
Mexican American
Other Hispanic
Non-Hispanic White
Non-Hispanic Black
Other Race
Mean (SD): 1.73 (1.71)
Elmarsafawy et al, 2006 Cross-sectional NAS Men
Low Calcium
Mean (SD)
21.6(12.0)
31.7 (18.3)
6.6 (4.3)
Tibia
Patella
Blood
21.6 (12.0)
31.7 (18.3)
6.6 (4.3)
Tibia
Patella
Blood
TZheutlin etal,2018
NAS Men, Resistant hypertension
Median IQR
20.0 (13.0-28.5)
27.0 (18.0-40.0)
5 (3.4-5.0)
Tibia
Patella
Blood
Peters et al, 2007
NAS Men
High vs Low Stress
High vs Low Stress
Mean (SD)
21.5 (13.4)
31.5(19.3)
Tibia
Patella
I 1 1
0.75 1.00 1.25
Effect Estimate (95% CI) per 1 ng/dL increase in blood Pb or 10 ng/g increase in bone Pb
BP = blood pressure; CI = confidence interval; IQR = interquartile range; KNHANES = Korean National Health and Nutrition
Examination Survey; NAS = Normative Aging Study; NH = non-Hispanic; NHANES = National Health and Nutrition Examination
Survey; OR = odds ratio; Pb = lead; RR = relative risk; SD = standard error; SE = standard error.
Figure 4-14 Associations between biomarkers of Pb exposure and
hypertension.
1 Cross-sectional studies identified positive associations between BLLs and prevalent hypertension
2 but were not statistically significant. Wang et al. (2020) indicated no association between BLLs and
3 hypertension prevalence among an older adult Chinese population (Figure 4-2 and Figure 4-3). Similarly,
4 Bushnik et al. (2014). also indicated no association between BLLs and hypertension prevalence among
5 participants of the Canadian Health Measures Survey. Studies evaluating NHANES (1999-2016) (Teve et
Exernal Review Draft
4-30
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
al.. 2020). NHANES (1999-2018) (Huang. 2022). and NHANES (2003-2010) (Hara et al.. 2015) did not
identify associations between BLLs and prevalent hypertension. Additionally, the GuLF study, which
cross-sectionally evaluated concurrent BLLs and prevalent hypertension among those involved in the
2010 Deepwater Horizon oil spill, indicated no associations (Xu et al.. 2021). However, this study had
low (quartile 1: 0.06 (ig/dL, quartile 4: 0.27 (ig/dL) mean BLLs. These results are consistent with the
2013 Pb ISA, which generally summarized studies reporting null associations between concurrent BLLs
and prevalent hypertension. For example, a study of South Korean Pb workers indicated no association
between BLLs and prevalent hypertension, despite this population having relatively high BLLs (mean:
31.9 (ig/dL) (Weaver et al.. 2008).
Although most cross-sectional studies did not observe associations between BLLs and prevalent
hypertension, some of these studies did report positive associations. A cross-sectional Brazilian study
evaluated BLLs among adults >40 years and indicated an association between BLLs and prevalent
hypertension (Lopes et al.. 2017). There were increased odds of hypertension prevalence noted (OR: 1.08
[95% CI: 1.03, 1.14]), for each l(ig/dL increase in BLLs. Additionally, a KNHANES (2008-2013) study
also indicated a marginal association for each doubling of BLLs for prevalent hypertension (OR: 1.09
[95% CI: 0.98, 1.22]) (Lee et al.. 2016a). Another more recent analysis using the China National Human
Biomonitoring longitudinal survey evaluated the relationship between concurrent BLLs and several
different definitions of hypertensive status (Ou et al.. 2022). When hypertension was defined according to
the 2010 Chinese Hypertension Guidelines (SBP >140 mmHg, DBP >90 mmHg), there were increased
odds of hypertension associated with BLLs (OR: 2.33 [95% CI: 1.67, 3.24]), when comparing the largest
quartile (>3.2 (ig/dL) with the lowest (<1.5 (ig/dL). Another recent NHANES (1999-2016) analysis (Tsoi
et al.. 2021) indicated increased odds of prevalent hypertension for each doubling of BLLs (OR: 1.09
[95% CI: 1.04, 1.14]) and when comparing the highest quartile (>2.10 (ig/dL) with the lowest quartile
(<0.89 (ig/dL) (OR: 1.21 [95% CI 1.07, 1.36]).
Several studies also evaluated the association between BLLs and prehypertension, a common
precursor to chronic hypertension. Ou et al. (2022) considered several prehypertension definitions. Using
the 2010 Chinese Hypertension Guidelines for prehypertension (SBP 120-139 mmHg, DBP 80-
89 mmHg), there were increased odds of prehypertension comparing the highest with the lowest quartile
(OR: 1.56 [95% CI: 1.22, 1.99]). This study also considered the 2017 American College of Cardiologists
(ACC)/American Heart Association (AHA) guidelines for elevated BP (SBP 120-129 mmHg, DBP <80)
and stage 1 hypertension (SBP 130-139 mmHg, DBP 80-89). Using these definitions, there was a null
association between BLLs and elevated BP (OR: 1.18 [95% CI: 0.88, 1.57]), but a positive association
with stage 1 hypertension (OR: 1.75 [95% CI: 1.31, 2.33]). Lee et al. (2016a) also evaluated
prehypertension, which was defined as DBP >80 mmHg or SBP >120 mmHg in a KNHANES (2008-
2013) analysis. This study indicated that for each doubling of BLLs there was an increased association
with prehypertension (OR: 1.09 [95% CI: 0.99, 1.21]). In another KNHANES (2007-2013) analysis, Lee
et al. (2016b) also specifically evaluated prehypertension, which was defined as DBP between 80-
89 mmHg or SBP between 120-139 mmHg and the absence of any current treatment or diagnosis of
Exernal Review Draft
4-31
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
24
25
hypertension. When comparing the highest quartile (2.717 to 24.532 (ig/dL) to the lowest quartile (0.206
to 1.539 (ig/dL) there was an association between BLLs and prehypertension (OR: 1.30 [95% CI: 1.07,
1.60]).
A recent longitudinal study (Malmo Diet and Cancer Study), within a cohort with high historical
Pb exposure, explored BLLs as they relate to incident hypertension (Gambclunghc et al.. 2016). This
study defined hypertension status as SBP >140 mmHg or DBP >90 mmHg or the use of antihypertensive
medication. At baseline (time = 0) there was a cross-sectional relationship between hypertension and the
highest quartile of BLLs, compared with the lowest three quartiles (OR: 1.3 [95% CI: 1.1, 1.5]).
Participants in this study were followed for approximately 16 years. When analyzed at the follow-up,
there was no association between baseline BLLs and the use of antihypertensive medication (OR: 1.0
[95% CI: 0.8, 1.2]) or high BP at follow-up (OR: 1.0 [0.7, 1.3]).
Another longitudinal analysis evaluated resistant hypertension and both blood and bone Pb levels
among participants of the NAS cohort (Zhcutlin et al.. 2018). Resistant hypertension was defined as
having either uncontrolled hypertension (SBP >140 or DBP >90 while taking >3 antihypertensive
medications), or controlled hypertension (SBP <140 and DBP <90 while taking >4 antihypertensive
medications). Overall, a 10 jj.g/g increase in tibia Pb level was associated with resistant hypertension (RR:
1.12 [95% CI: 1.01, 1.25]) but the association with same increase in patella Pb levels was smaller in
magnitude (RR: 1.04 [95% CI: 0.96, 1.13]). The dose-response relationship between tibia Pb levels and
resistant hypertension risk is relatively linear, with the steepest slope noted in the lower part of the
distribution of tibia Pb concentrations (0 to 20 jj.g/g) (Figure 4-15); a flattening of the slope between 20
and 80 jj.g/g; and a steepening of the slope for the highest bone Pb concentrations (>80 jj.g/g). This dose-
response relationship supports previous findings of a supralinear association between Pb exposures and
Pb-related health outcomes (U.S. EPA. 2013a). However, among the same study participants. (Zheutlin et
al.. 2018) for a 1 (ig/dL increase in BLLs, the association between BLL and resistant hypertension was
smaller in magnitude (RR: 1.02 [95% CI: 0.97, 1.08]).
Exernal Review Draft
4-32
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Tibia Lead (|jgvgJ
HTN = hypertension; RR = relative risk.
Source: Zheutlin et al. (2018).
Figure 4-15 Dose-response curve between tibia Pb levels and resistant
hypertension, Normative Aging Study cohort.
4.3.1.2.1 Effect Measure Modification
Several recent studies also evaluated EMM by a variety of factors when assessing the relationship
between biomarkers of Pb exposure and hypertension outcomes. In a recent NHANES (1999-2006)
analysis, Miao et al. (2020) evaluated BLLs and any hypertension status as well as uncontrolled
hypertension, stratified by sex (Figure 4-14, Figure 4-16, Table 4-4). Any hypertension was defined as
SBP >130 mmHg or DBP >80 mmHg or the use of antihypertension medication, while uncontrolled
hypertension was defined as an average SBP >130 mmHg or DBP >80 mmHg, regardless of
antihypertension medication use. When considering continuous BLLs, for each 1 (ig/dL increase in blood
Pb there were increased odds of any hypertension among males (OR: 1.037 [95% CI: 1.015, 1.060]), but
less so among females (OR: 1.020 [95% CI: 0.970, 1.074]). However, for each 1 (ig/dL increase in BLLs,
there were increased odds of uncontrolled hypertension for both hypertensive males (OR: 1.157 [95% CI:
1.080, 1.239]) and females (OR: 1.109 [95% CI: 1.020, 1.205]) and a smaller increase in the odds of
uncontrolled hypertension among all males (OR: 1.062 [95% CI: 1.036, 1.088]) and females (OR: 1.056
[95% CI: 1.011, 1.102]). The dose-response relationship, when considering a restricted cubic spline for
BLLs, indicated a steeper slope up to around 2 (ig/dL, like has been observed for other Pb exposure and
hypertension outcomes (Figure 4-16, See Section 4.3.1.2). This relationship appears to be more
pronounced in males than in females, especially when comparing uncontrolled hypertension to those with
controlled hypertension.
Exernal Review Draft
4-33
DRAFT: Do not cite or quote
-------
Adjusted ORs of any hypertension in men
BLL4.43 ISBi percenMJfi In adults) as ral
3.0-
S
S 0.8-
0 1 2 a 4
BlOflfl lend revel. \tjnlL
Adjusted ORs of any hypertension in women
BLL=G.13 <5tti pareennsa in adufts) as rsf
3,0-
3
8 0.5-
0.0 H
0 1 2 I J"
teatf level. ji^WL
Adjusted ORs of unconlrolled HTN in hypertensive men
BLL-0.43 i5th percenfeteiil aduHi) mm ref
Adjusted ORs of uncontrolled HTN in hypertensive women
a.o-
|«
i 4-
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
KNHANES (2008-2013) and observed positive associations between BLLs and hypertension (OR: 1.29
[95% CI: 1.10, 1.51]) and prehypertension (OR: 1.21 [95% CI: 1.06, 1.38]) in females only for each
doubling of BLLs. Similarly, in the cross-sectional analysis of the Malmo Diet and Cancer Study,
Gambclunghc et al. (2016) indicated an elevated effect of prevalent hypertension among females (OR: 1.4
[95% CI: 1.1, 1.7]), compared with males (OR: 1.2 [(0.6, 1.51). However. Ou et al. (2022) indicated
associations larger in magnitude, but with less precision, among males compared with females (Figure 4-
17) in the China National Human Biomonitoring cohort. In contrast, the cross-sectional analysis of the
Canadian Health Measures Survey, indicated no association between BLL and prevalent hypertension and
was not further modified by sex (Bushnik et al.. 2014).
02 O* 04 OI Oi O* 04
Elevated Stage 1 HTN
Ol 02 03 04 Oi Oi O^ 04 Oi 02 O* 04
Stage 2 HTN Pre-HTN HTN
The 2017 ACC/AHA guidclino
2010 Chinos© HTN guideline
Men
++144W-
01 02 03 0-» 01 02 03 01 01 02 OJ 01 01 02 03 Oi
Elevated Stagel HTN Stage 2 HTN Pre-HTN
01 02 02 01
HTN
The 2017 ACC,1 AHA guidelin
2010 Chinese HTN guidelin
Women
ACC = American College of Cardiologists; AHA = American Heart Association; HTN = hypertension; Q = quartile.
Source: Adapted from Qu et al. (2022).
Figure 4-17 Odds ratios for the association between quartiles of blood Pb and
prevalent hypertension, stratified by sex.
Stratification by race/ethnicity and other socioeconomic factors was less common in studies of
hypertension, compared with studies examining BP alone. Scinicariello et al. (2011) examined the
relationship between blood Pb and prevalent hypertension, stratified by both race/ethnicity and sex using
NHANES (1999-2006). Although the overall relationship between hypertension and blood Pb was null,
an association was reported among Black males (OR: 2.69 [95% CI: 1.08, 6.72]) when comparing those
with BLLs at the 10th percentile (<0.6 (ig/dL) to those at the 90th percentile (3.5-10 (ig/dL). Hara et al.
(2015) indicated an overall null association between blood Pb and hypertension, however, an outcome
that persisted even when stratifying by race/ethnicity in NHANES (2003-2010). In addition, the GuLF
study (Xu et al.. 2021) indicated a null association between concurrent blood Pb and hypertension in the
full sample and in analyses stratified by race (Table 4-4).
Exernal Review Draft
4-35
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
Diet has also been considered as an EMM of this association. A recent KNHANES (2013)
analysis evaluated the association between BLLs and hypertension by curry intake (Choi et al.. 2018).
Curcumin, a major component of curry, is known to have anti-inflammatory properties and can act as a
chelating agent for heavy metals, such as Pb. This study defined hypertension as SBP >140 mmHg, DBP
>90 mmHg, or current use of antihypertensive medication. This study indicated a null association
between prevalent hypertension and blood Pb among those who regularly consumed curry (consumed at
least one curry dish/month in the past year) (OR: 1.108 [95% CI: 0.827, 1.485]),fora 1 (ig/dL increase in
BLLs; however, an association was reported in those who did not regularly consume curry (OR: 1.399
[95% CI: 1.054, 1.857]). A previous analysis of the NAS cohort (Elmarsafawv et al.. 2006) evaluated
whether calcium intake affects the relationship between hypertensive status and bone Pb levels. High
calcium intake has been associated with lower BP measurements and it has been hypothesized that
calcium and Pb may interact with one another biologically. Using detailed dietary information to estimate
calcium intake indicated there were moderate associations between either concurrent blood or bone Pb
measurements and prevalent hypertension, but this association did not differ among those with low
calcium intake (<800 mg/d) compared with those with high calcium intake (>800 mg/d).
A small number of studies also considered BMI as an EMM. In the examination of the China
National Human Biomonitoring cohort, Qu et al. (2022) indicated a marginally stronger effect between
concurrent BLLs and hypertension among those with elevated (>24 kg/m2) BMI (Figure 4-18). However,
results from the GuLF study indicated no differences in the association between concurrent BLLs and
prevalent hypertension when stratified by BMI. It is important to note that this study did not observe an
increase in association between concurrent BLLs and hypertension status, even when unstratified.
Exernal Review Draft
4-36
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Ql 1J3 Q4 Ql Ql Q.5 Q4
Elevated Stage 1 HTN
(Ł1 Q2
Stage 2 HTN
The 2017 ACC/AHA guideline
01 Q3 y.". Q4 yi Q2 Q:t
Pre-HTN HTN
2010 Chinese HTN guideline
BMI < 24 Kg/m
02 03 0-1 01 02 03 04 01 02 03 04
Elevated Stage 1 HTN Stage 2 HTN
The 2017 ACC/AHA guideline
01 02 03 0'11 01 02 03 01
Prc-HTN HTN
2010 Chinese HTN guideline
BMI > 24 Kg/m
ACC = American College of Cardiologists; AHA = American Heart Association; BMI = body mass index; HTN = hypertension;
Q = quartile.
Source: Adapted from Qu et al. (2022).
Figure 4-18 Odds ratios for the association between quartiles of blood Pb
levels and prevalent hypertension, stratified by body mass index,
China National Human Biomonitoring cohort.
4.3.1.3 Blood Pressure and Hypertension in Children
The 2013 Pb ISA (U.S. EPA. 2013a) indicated that the small body of evidence presented
suggested a relationship between biomarkers of Pb exposure and BP and hypertensive effects in children,
adding to the few studies presented in the 2006 Pb AQCD (U.S. EPA. 2006b). Although BP effects are
often more prevalent in adult populations compared with child populations, evidence from earlier studies
suggested BP increases related to Pb biomarkers levels in children and adolescents. In the 2013 Pb ISA
(U.S. EPA. 2013a). the strongest evidence of a relationship between Pb biomarkers and increased
childhood BP came from longitudinal studies (Zhang et al.. 2012; Gump et al.. 2005) and cross-sectional
studies (Gump et al.. 2011; Factor-Litvak et al.. 1999). More recent data supports the previous findings.
Study-specific details, including Pb biomarker levels, study population characteristics, potential
confounders, and select results from these studies are highlighted in Table 4-5. These details include
standardized results as well as those that could not be standardized based on the information provided in
each paper.
Several recent longitudinal studies highlight associations between increased BP associated with
increased levels of biomarkers of Pb exposure in children. A longitudinal study in Mexico City evaluated
cord BLLs (GM 4.67 (ig/dL) and maternal bone Pb levels (patella [median: 11.6 j^ig/g | and tibia [median
Exernal Review Draft
4-37
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
9.3 jJ-g/g]) 1-month postpartum and subsequently assessed BP in their offspring (aged 9-15) (Zhang ct al..
2012). The associations between any Pb biomarker and changes in BP were null, but when stratified by
sex, a 10 jj.g/g increase in maternal tibia Pb levels was associated with increased SBP (1.62 mmHg [95%
CI: 0.53, 2.71 mmHg]) and DBP (1.24 mmHg [95% CI: 0.23, 2.25]) in female children, but not in male
children. There was no such association for patella Pb or cord BLLs. Cortical (tibia) bone is reflective of
cumulative exposure, whereas trabecular (patella) bone has a shorter half-life and a higher turnover rate of
Pb. Additionally, cord blood is mostly representative of the BLLs at birth, and not necessarily the BLLs
the fetus was exposed to throughout pregnancy. In addition, a small prospective study among 9.5-year-old
children, described in the 2013 Pb ISA, observed an increase in SBP (12.16 mmHg [95% CI: 2.44,
21.88 mmHg]), but only suggested an increase in DBP (8.54 mmHg [-0.45, 17.35 mmHg])
corresponding with a 1 (ig/dL increase in cord BLLs (GM: 2.56 j^ig/dL) (Gump et al.. 2005); even so, null
associations were observed between concurrent blood levels and BP measurements.
In contrast, several recent longitudinal studies including mother-child pairs have been
implemented and generally have yielded null results. Kupsco et al. (2019) assessed blood levels for
several metals, including Pb (mean: 3.7 j^ig/dL). during the second trimester of pregnancy and specific
cardiac and metabolic endpoints were evaluated among children. The associations between the natural log
of maternal BLLs and children's SBP or DBP were null. Another study using maternal BLLs evaluated
the association between the erythrocyte fraction (Ery-Pb) in maternal blood and BP among children
(-4.5 years) (Skroder et al.. 2016). The Ery-Pb was assessed at both 14 weeks (median: 73 j^ig/kg) and
30 weeks (median: 86 j^ig/kg) gestation. Linear regression analyses identified no associations with SBP or
DBP among young children. In addition, another recent study Zhang et al. (2021) of mother-child pairs,
evaluated BLLs in mothers 24-72 hours after delivery, BP was then subsequently assessed in children
(aged 3-15). Among this cohort, there were null associations between mother's BLL and children's BP
measurements.
Several recent cross-sectional studies that evaluated the association between concurrent BLLs and
BP indicated positive associations. Gump et al. (2011) evaluated BP change as a response to acute stress.
Children aged 9-11 were subjected to a variety of experimental tasks to stimulate the stress response.
Children with higher quartiles of concurrent blood Pb (1.21 to 3.76 (ig/dL) exhibited a greater change in
SBP (7.23 mmHg; 95% CI not reported) compared with children with lower blood Pb (0.14 to
0.68 (ig/dL; 5.3 mmHg) (Table 4-5). An earlier study, included in the 2013 Pb ISA, evaluated children
with higher Pb blood levels (4.1 to 76.4 (ig/dL) from two different towns in Kosovo, when it was part of
Yugoslavia, with high (mean: 37.3 (ig/dL) and low (mean: 8.7 (ig/dL) BLLs. This study identified a
modest association between a 1 jj.g/dL increase in concurrent BLLs and SBP (0.05 mmHg [-0.02, 0.13])
(Factor-Litvak et al.. 1996). Additionally, a recent study from China, evaluated childhood BP and
concurrent child blood Pb (Lu et al.. 2018). Children in this study were recruited from two regions of
similar SES in China, corresponding to an e-waste (high environmental Pb, mean: 7.14 (ig/dL) exposed
area (Guiyu) and a reference (low environmental Pb, mean: 3.91 (ig/dL) area (Haojiang); no association
was noted between log-transformed BLLs and either SBP or DBP among these children.
Exernal Review Draft
4-38
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
24
25
26
27
28
29
30
31
32
33
34
35
Several recent studies assessing BP and BLLs in children have relied on cross-sectional
nationally representative data sets (NHANES, KNHANES). A large study evaluated seven 2-year
NHANES cycles (1999-2012) among adolescents aged 12-19, with an average BLL of 1.17 j^ig/dL (Xuet
al.. 2017). In this cohort, there was no association between BLLs and BP. Another NHANES (2009-
2016) analysis also indicated a null association between BP changes and blood Pb among children aged
8-17 (Desai et al.. 2021). Similarly, a smaller study included three KNHANES cycles (2010-2016)
among adolescents 10-18 years of age with aGM BLL of 1.19 (ig/dL (Ahn et al.. 2018). In this study,
there was no association reported between a doubling of BLLs (log-transformed) and BP or
prehypertension (SBP 120-140 mmHg, DBP 80-90 mmHg). Another NHANES analysis, used five 2-
year NHANES cycles (2007-2016) among children and adolescents aged 8-17. This cohort had a GM of
BLLs ranging between 0.98 (ig/dL and 0.60 (ig/dL from the first (2007-2008) to last (2015-2016)
NHANES cycle evaluated. Similarly, there were no associations between BLLs and BP. However, when
stratified by race/ethnicity, a twofold increase in BLLs was associated with decreased DBP among Black
children (-1.59 mmHg [95% CI: -3.04, -0.16 mmHg]), and increased DBP among White children
(1.38 mmHg [95% CI: 0.40, 2.36 mmHg]) (Yao et al.. 2020).
Several studies also evaluated total peripheral resistance (TPR) and its relationship with
biomarkers of Pb in children. In general, TPR measures the total amount of force circulating blood
imposes on the vasculature in the body and is represented by the ratio between MAP and cardiac output.
Gump etal. (2011) evaluated cardiovascular responses, including sympathetic and parasympathetic
activation, in response to acute stress in children. Children aged 9-11 were subjected to a variety of
experimental tasks to stimulate the stress response. Overall, increased BLL quartiles corresponded to an
increase in TPR. These results support a previous study by Gump et al. (2005). which reported higher Pb
exposures during early childhood. In this study, Gump et al. (2005) indicated that an increase in early
childhood (average age 2.6 years) BLLs was associated with a greater TPR response to acute stress years
later (at 9.5 years of age). Overall, in this cohort, TPR increased with increasing quartiles of BLLs.
Furthermore, BLL was identified as a mediator within this cohort between the relationship between SES
and TPR reactivity. Specifically, Gump et al. (2007) indicated that BLLs may also mediate the association
between SES and the cortical responses to acute stress. Furthermore, when controlling for childhood
BLLs, family income (a measure of SES) was no longer predictive of Cortisol levels.
4.3.2 Toxicological Studies of Blood Pressure and Hypertension
In the 2013 ISA for Pb and previous Pb AQCDs, animal toxicological studies have consistently
demonstrated a relationship between exposure to Pb and increases in BP. Nearly all animal toxicological
studies provided evidence that long-term Pb exposure (>4 weeks), resulting in BLLs less than 10 (ig/dL,
could result in the onset of hypertension (after a latency period) in experimental animals that persists long
after the cessation of Pb exposure (U.S. EPA. 2006a). For example, Tsao et al. (2000) presented evidence
for increased systolic and diastolic BP in rats with BLLs somewhat similar to the current U.S. adult
Exernal Review Draft
4-39
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
24
25
26
27
28
29
30
31
32
33
34
35
36
population (mean 2.15 (ig/dL blood Pb), compared with untreated controls. In addition, there was a
statistically significant, positive trend for increasing BP with increasing BLLs up to 56 (ig/dL, with the
effect leveling off at higher BLLs. There were a number of other studies from previous reviews
demonstrating increases in measures of BP following exposure to Pb (Mohammad et al.. 2010; Zhang et
al.. 2009; Badavi et al.. 2008); Grizzo and Cordellini (2008): (Rcza et al.. 2008; Bravo et al.. 2007; Robles
et al.. 2007); Hevdari et al. (2006); (Bagchi and Preuss. 2005; Nakhoul etal.. 1992). More information on
these studies can be found in Section 4.4.2.2 of the 2013 Pb ISA (U.S. EPA. 2013a).
Since the publication of the 2013 Pb ISA, animal toxicological studies with mean blood Pb values
of <30 (ig/dl have further demonstrated a relationship between exposure to Pb and increases in measures
of BP. More specifically, rats with amean BLL of 13.6 jj.g/dl following a 30-day drinking water exposure
had statistically significantly higher SBP (p < 0.05) at 1, 2, 3, and 4 weeks of exposure when compared
with control animals (Fioresi et al.. 2014). At the end of the 30-day exposure, these authors also reported
statistically significant increases in SBP, DBP, and MAP (Fioresi et al.. 2014). Similarly, Nunes et al.
(2015) reported that rats with an 8.4 jj.g/dl mean BLL had statistically significantly higher SBP from 7 to
28 days following a 30-day exposure, relative to control animals. In another multi-day measurement
study, Xu etal. (2015) reported a statistically significant increase (p < 0.05) in SBP and DBP between the
6th and 17th day of a 40-day Pb exposure, but no difference from days 19 to 40. Pb levels in this study
were 19.3 jj.g/dl at day 12 and 24.6 jj.g/dl on day 40 Xu et al. (2015).
In agreement with the studies that measured BP on multiple occasions throughout exposure, Silva
et al. (2015) reported that rats with a 12.3 jj.g/dl BLL had statistically significantly (p < 0.05) higher SBP
following a 30-day exposure relative to control animals. In an additional study, Shvachiv et al. (2018)
exposed rats first through lactation. After weaning, rats were then exposed by drinking water either
continuously until 28 weeks or were given 8 weeks of Pb abstinence and then exposed until 28 weeks.
This study reported a statistically significant increase in DBP and MAP in rats continuously or
intermittently exposed to Pb, as well as a significant increase in SBP in rats continuously exposed to Pb
relative controls. In addition, for both exposure groups, the authors reported a statistically significant
(p < 0.05) decrease in BP regulation as measured by differences in baroflex gain. Notably, a decreased
baroflex response can impair BP recovery (i.e., lowering of BP) following stimulation of chemoreceptors
that increase BP. BLLs in this study were -24 jj.g/dl for the constant exposure group and -19 jj.g/dl for the
intermittent exposed group (Shvachiv et al.. 2018). Similarly, in a study of rats exposed to Pb through
lactation and weaning, statistically significant increases (p < 0.05) in SBP attimepoints ranging from
PND 22 to PND 100 were reported relative to control animals. Mean BLLs ranged from -11 jj.g/dl to
20 (ig/dl in this study (Gaspar and Cordellini. 2014). In agreement with these studies, a pair of analyses
demonstrated a statistically significant increase (p < 0.05) in SBP (but not DBP) relative to control
animals following maternal exposure and then an additional a 1-year drinking water exposure that
resulted in a BLL of <30 jj.g/dl (Zhu etal.. 2019; Zhu et al.. 2018).
Exernal Review Draft
4-40
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
24
25
26
27
28
29
30
31
While the above studies all reported some statistically significant increases in BP at one or
multiple timepoints, Wildemann et al. (2015) reported no change relative to control animals for SBP,
DBP, or PP for rats with a 1.7 jj.g/dl or 8.6 jj.g/dl BLL after 4 weeks of exposure. Moreover, combined
exposure of Pb, mercury, and methylmercury resulted in no change in any of these BP measures relative
to control .
When considered as a whole, the animal toxicological evidence presented above continues to
demonstrate a clear relationship between Pb exposure and increases in measures of BP. All but one
animal toxicological study evaluated above reported at least some measure of increased BP following Pb
exposure. Additional details on these studies and their designs can be found in Table 4-6.
4.3.2.1 Renin-Angiotensin-Aldosterone System
The renin-angiotensin-aldosterone system (RAAS) plays an important role in the regulation of
BP. For example, angiotensin II (Ang II) stimulates arteriolar vasoconstriction leading to increases in BP.
Angiotensin-converting enzyme (ACE) is involved in the activation of Ang II. In the 2013 Pb ISA, most
studies demonstrated an effect of Pb on RAAS consistent with increases in BP. For example, following
Pb exposure, vascular reactivity to Ang II was found to increase (Robles et al.. 2007). Moreover,
exposure to Pb also resulted in increases in kidney and/or serum ACE activity and renal angiotensin II
positive cells (Rodrigucz-lturbc et al.. 2005; Sharifi et al.. 2004; Carmignani et al.. 1999). In addition, Pb
exposure increased activity and levels of the a-1 subunit protein of Na+/K+ATPase, which plays a major
role in Na+ reabsorption and is regulated by the RAAS (Fiorim, 2011, 786401; Simoes, 2011, 711507).
Other studies demonstrating effects on RAAS can be found in Section 4.4.2.3 of the 2013 Pb ISA (U.S.
EPA. 2013a).
Since the 2013 Pb ISA, Fioresi et al. (2014) reported a statistically significant increase in NA+ K+
ATPase (p < 0.05) but no change in ACE activity in plasma and cardiac tissue. Thus, there is limited
additional evidence for changes in RAAS following Pb exposure resulting in BLLs <30 jj.g/dl. Additional
details for this toxicological study can be found in Table 4-6 of this ISA.
4.3.3 Integrated Summary of Blood Pressure and Hypertension
Several studies presented in the 2013 Pb ISA demonstrated positive associations between BP
measurements and biomarkers of Pb exposure. The current literature continues to support these findings.
Since the 2013 Pb ISA, several nationally representative cross-sectional studies (e.g., NHANES,
KNHANES) have been conducted to specifically evaluate the association between concurrent blood Pb
values and BP measurements or hypertension status. These studies can contribute substantially to the
current evidence base, especially since there were fewer of nationally representative studies available at
the time of the 2013 Pb ISA. Typically, cross-sectional studies can provide information on the association
Exernal Review Draft
4-41
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
24
25
26
27
28
29
30
31
32
33
34
35
36
between concurrent blood Pb values and BP measurements or on hypertension status taken at the time of
the interview. However, longitudinal studies typically can provide information on the relationship
between historic Pb biomarker information and the change in BP since baseline or the development of
hypertension. Both study types are valuable in discerning the associations between biomarkers of Pb
exposure, BP, and hypertension.
Overall, recent cross-sectional studies provided consistent evidence that concurrent BLLs are
associated with increased SBP and DBP within adult populations. Evidence for increases in PP and MAP
was less consistent but these endpoints were examined in fewer studies. Associations between concurrent
BLLs and BP among children were inconsistent, and mostly suggested a null association. Yet, a series of
studies evaluating TPR among children indicated an association with increasing blood Pb values. In
addition, studies evaluating a concurrent BLL and BP at a particular threshold (i.e., SBP >130 mmHg),
mostly indicated null results.
Longitudinal studies less commonly evaluated changes in BP measurements (in mmHg) but were
more likely to evaluate the development of clinical hypertension or prehypertension over a prolonged
period of time. Most longitudinal studies evaluating incident hypertension or prehypertension and a
marker of cumulative Pb exposure (measured in bone) indicated positive associations. In contrast,
associations with incident hypertension or prehypertension were mostly null when using blood Pb
measurements. Of the few longitudinal studies that evaluated BP changes, the results were mostly null,
with a few indicating associations between baseline blood Pb measurements and changes in BP
measurements overtime.
Animal toxicological studies continue to support the epidemiologic evidence. Recent animal
toxicological studies reaffirm the clear association between Pb exposure in animals and increases in BP,
presented in the 2013 Pb ISA. Current studies specifically were restricted to only include lower BLLs
(<30 (ig/dL), and the majority of relevant studies indicated a persistent relationship between BLLs,
whether it be related to continuous or intermittent exposures, and increases in BP. The evidence
supporting changes in RAAS following Pb exposure is less consistent.
Several recent epidemiologic studies also evaluated EMM by various factors including
race/ethnicity, sex, age, diet, stress, and genetic polymorphisms. Taken together the evidence suggests
that in addition to having higher blood Pb measurements, associations between blood Pb and BP are
larger among non-Hispanic Black populations when compared with Hispanic or non-Hispanic White
populations. When considered alone, there were mixed conclusions as to whether there were any
differences in the association between Pb biomarkers and BP or hypertension by sex. However, when
combined with race, Black males clearly demonstrated increased risk of Pb-associated BP changes, when
compared with other sex/race groups. These results were consistent across several analyses. In addition,
those with high depressive symptoms and increased stress (perceived stress and AL) were identified as
having larger risk of Pb-associated increases in BP and hypertension. Taken together, the most recent
Exernal Review Draft
4-42
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
24
25
26
27
28
29
30
evidence supports the conclusions of the previous ISA, indicating an association between biomarkers of
Pb exposure and changes in either BP or hypertension status.
4.4 Ischemic Heart Disease and Associated Cardiovascular
Effects
IHD, also known as CHD or CAD, is a chronic condition characterized by atherosclerosis and
reduced blood flow to the heart. The majority of IHD is caused by atherosclerosis (Section 4.8), which
can lead to the blockage of the coronary arteries and restriction of blood flow to the heart muscle. An MI
or heart attack is an acute event that occurs when heart tissue death occurs secondary to prolonged
ischemia. Several studies within this section evaluate IHD as a composite measure mostly defined as the
presence of MI, angina pectoris, or CHD death, whereas other studies evaluate composite IHD- risk
scores using cross-sectional data. There were no animal toxicological studies examining indicators of IHD
at BLLs <30 (ig/dL published since the 2013 Pb ISA.
4.4.1 Epidemiologic Studies of Ischemic Heart Disease
The 2006 Pb AQCD (U.S. EPA. 2006b) indicated an association between Pb biomarker levels
and MI (Gustavsson et al.. 2001). The 2013 Pb ISA (U.S. EPA. 2013a') further contributed to this small
amount of evidence with the inclusion of a study among the NAS cohort. This longitudinal study among
(mostly white) men indicated an increased incidence of IHD associated with bone (both tibia and patella)
Pb levels (Jain et al.. 2007).
Several recent studies have been published since the 2013 Pb ISA that specifically evaluate the
association between biomarkers of Pb exposure and measures of IHD, CHD, or CAD. Study-specific
details, including biomarker Pb levels, study population characteristics, confounders, and select results
from these studies are highlighted in Table 4-7. These details include standardized results as well as those
that could not be standardized based on the information provided in each paper.
A recent study evaluated whether the relationship between CHD and bone Pb levels is modified
by certain genetic polymorphisms (Ding et al.. 2016). It is thought that certain genetic factors may
predispose an individual to increased Pb toxicity. Using the NAS cohort, several genes and encoding
proteins including, S-aminolevulinic acid dehydratase (ALAD), HFE, heme oxygenase-1 (HMOX1),
VDR, apolipoprotein E (APOE), glutathione S-transferases, and the RAAS, were evaluated as effect
measure modifiers of the relationship between bone Pb measurements and incident CHD. All these
different genes and encoding proteins appear to play a role in influencing Pb uptake and or retention or
may alter Pb toxicity. Overall, 22 different SNPs corresponding to these Pb-related genes were studied
separately and in combination in a genetic risk score (GRS). Two GRSs were constructed; the first (GRS
1) summed all 22 SNPs, whereas the second (GRS 2) only included the nine SNPs found to significantly
Exernal Review Draft
4-43
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
modify the association between patella Pb levels and incident CHD within this study. Overall, without
considering any genetic polymorphisms, the association between a twofold increase in patella Pb levels
and CHD incidence was positive (HR: 1.36 [95% CI: 1.15, 1.61]). Several genetic polymorphisms
appeared to further modify this relationship. Specifically, positive associations were observed for
individuals with at least one minor allele in VDR (rsl544410 (Bsml) HR: 1.65 [95% CI: 1.31,2.08]);
rs731236 (Taql) HR: 1.61 [95% CI: 1.29, 2.02]); rsl073581 (i^oŁ7)HR: 1.47 [95% CI: 1.17, 1.83]);
rs757343 (Tru91) HR: 1.48 [95% CI: 1.18, 1.85]) and HMOX1 (rs2071749) HR: 1.51 [95% CI: 1.22,
1.86]), whereas individuals without any minor alleles had null associations. However, positive
associations were observed among individuals without any minor alleles in HMOX1 (rs2071746 HR:
1.51 [95% CI: 1.07, 2.13]; rs5995098 HR: 1.62 [95% CI: 1.23, 2.14]), APOE (rs429358 HR: 1.43 (95%
CI: 1.17, 1.75]) and angiotensinogen (AGT; rs699 HR: 2.17 [95% CI: 1.51, 3.12]; rs5046 HR: 1.57 [95%
CI: 1.27, 1.94]). When considered in combination, both GRS values identified significant EMM for the
association between a twofold increase in patella Pb and risk of incident CHD (GRS 1 HR: 2.27 [95% CI:
1.50, 3.42] and GRS 2 HR: 2.77 [(95% CI: 1.78, 4.31]).
Another study of the NAS cohort measured if incident CAD and bone Pb measurements were
modified by diet (Ding et al.. 2019). Evidence suggests that a diet deficient in essential metals (zinc,
calcium, selenium, iron) can augment Pb absorption and retention in the body, while certain vitamins (C,
E, and Be) may function as antioxidants against Pb toxicity. Specifically, vitamins E and C can act by
inhibiting lipid peroxidation by neutralizing Pb-related reactive oxygen species (ROSs) by rapid electron
transfer, while vitamin Be can act by reducing Pb-related increases in homocysteine. Additionally,
vitamins Bi and Be are composed of ring structures containing nitrogen, which may mediate interactions
with Pb. This study collected detailed dietary information from each NAS member and classified diets
high in fruit, legumes, whole grains, tomatoes, seafood, poultry, cruciferous vegetables, dark-yellow
vegetables, leafy vegetables, and other vegetables as "prudent" diets. Alternatively, diets with a high
intake of processed meat, red meat, refined grains, butter, high-fat dairy products, eggs, and fries, was
considered a "Western" diet. The diet types were considered separately and were not mutually exclusive.
For example, a low prudent diet was not equivalent to a high Western diet, and there could be some
overlap in diet type between participants. Overall, results indicated that for each doubling of bone Pb
levels there was a higher risk of CAD associated with both tibia (HR: 1.25 [95% CI: 1.06, 1.48]) and
patella (HR: 1.30 [95% CI: 1.09, 1.56]). However, low prudent diet modified this association with patella
Pb levels. Those with a low prudent diet (HR: 1.64 [95% CI: 1.27, 2.11]) had a higher association
between patella Pb levels and CAD risk compared with those with a high prudent diet (HR: 1.07 [95% CI:
0.86, 1.34]). A Western diet did not appear to modify the results.
In a Canadian prospective cohort of patients on hemodialysis, incident cardiovascular events
during the 2-year follow-up period were evaluated (Tonclli et al.. 2018). Cardiovascular events were
defined as acute MI, percutaneous coronary angioplasty, coronary artery bypass grafting, heart failure,
and stroke or transient ischemic attack. Patients in this cohort (n = 1,278) had relatively low BLLs (1st
Exernal Review Draft
4-44
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
decile: 0.06 (ig/dL, 10th decile 1.74 (ig/dL), and there was no observed relationship between BLLs and
cardiovascular events when comparing the highest with the lowest decile (results not shown).
Several recent cross-sectional analyses have assessed 10-year CHD risks in association with
biomarkers of Pb exposure (Nguyen et al.. 2021; Park and Han. 2021; Choi et al.. 2020; Cho et al.. 2016).
Cho et al. (2016) calculated the Framingham risk score (FRS) to predict the 10-year risk of CHD in
asymptomatic patients associated with BLLs among Korean men and women taking part in KNHANES
IV and V (2008-2010). The FRS incorporates various CHD risk factors including age, gender, SBP, total
cholesterol, and high-density lipoprotein cholesterol (HDL-C). This study indicated that for each
increasing BLL quartile, there were statistically significant increased odds of an elevated FRS, compared
with the lowest quartile among men. Specifically, there was a positive effect (OR: 3.13 [95% CI: 2.09,
4.69]) for the highest quartile of BLLs (3.519-26.507 (ig/dL) compared with the lowest quartile of BLLs
(0.711-2.129 (ig/dL) among males. This effect was not observed among females (OR: 0.88 [95% CI:
0.26, 2.97]). Park and Han (2021) also calculated a CVD risk score based of the FRS from 2008. Again,
using data obtained from KNHANES, this study calculated the effect of a log increase in BLLs associated
with a 10-20% increase in FRS. Park and Han (2021) indicated that a one-unit increase in log-
transformed blood Pb was associated with an odds ratio of 2.4 (95% CI: 1.89, 3.18) of having an FRS
increase between 10-20% in men. However, this association was not observed among females (OR: 1.05
[95% CI: 0.68, 1.63]). Similarly, a >20% increase in the FRS score was associated with an odds ratio of
2.85 (95% CI: 2.02, 4.01) among males, but not among females (OR: 0.71 [0.19, 2.66]). Another
assessment of KNHANES indicated that a doubling of BLLs was associated with an 0.10% (0.02,
0.21%]) increase in 10-year CVD risk (Nguyen et al.. 2021) (Table 4-7)
(Choi et al.. 2020) used KNHANES to evaluate associations between BLLs and the 10-year
atherosclerotic cardiovascular disease (ASCVD) risk score. The ASCVD risk score was calculated first
based of the ACC and AHA guideline on the assessment of CVD risk. This formula incorporates factors
such as age, total cholesterol, HDL-C, hypertension treatment, smoking status, and diabetes. For this
analysis, the risk score was scaled to be more relevant to the Korean population, as the risk score was
created based on mostly non-Hispanic White and non-Hispanic Black populations in the United States.
This study also noted an increase in the ASCVD risk of 0.117 (95% CI: 0.005, 0.229) among men when
comparing the highest with the lowest quartiles (distribution information not reported), but this increase
was not noted among women (0.072, [95% CI: -0.004, 0.148]). When stratified by urban versus rural
locations, there was an increase in the ASCVD risk score effect estimate among men living in urban areas
(0.133 [95% CI: 0.011, 0.254]) but also an increase among women living in rural communities (0.212
[95% CI: 0.045,0.379]).
A recent cross-sectional study evaluated older, diabetic patients in China (Wan et al.. 2021). This
study evaluated BLLs and prevalent CVD. In this context, CVD was defined as a composite measure
including a history of CHD, MI, or stroke. When comparing the highest quartile of BLLs (>3.7 (ig/dL)
with the lowest quartile of BLLs (<1.8 (ig/dL), there were increased odds (OR: 1.44 [95% CI: 1.17, 1.76])
Exernal Review Draft
4-45
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
of CVD within this population at higher BLLs (Figure 4-19). Additionally, another recent study evaluated
a collection of emerging predictive CVD biomarkers including asymmetric dimethylarginine (ADMA),
adipocyte fatty acid-binding protein (FABP4, also known as aP2 and AFABP), adiponectin, and chemerin
(Ochoa-Martinez et al.. 2018). When comparing the highest tertile (T3: >9.1 (ig/dL) with the lowest tertile
(Tl: <3.5 (ig/dL), there was a positive association with ADMA (0.75 (imol/L [95% CI: 0.15,
1.85 |_imol/L|) and FABP4 (27.5 ng/mL [95% CI: 10.0, 34.5 ng/mL]). Other biomarkers evaluated had
null associations with BLLs.
CCA plaque
CVD
P for trend < 0.001
P for trend c 0.001
One In BLL SD increment
BLL Quartile 4
BLL Quartile 3
BLL Quartile 2
BLL Quartile 1
0.5 1,0 1.5
Odds ratio (95%CI)
2.0
One In BLL SD Increment
BLL Quartile 4
BLL Quartile 3
BLL Quartile 2
BLL Quartile 1
0.5 1.0 1.5
Odds ratio (95%CI)
2.0
Left CCA diameter
Pfor trend < 0.384
Ore In BLL SD increment
BLL Quartile 4
BLL Quartile 3
BLL Quartile 2
BLL Quartile 1
*0.2 -0,1 0 0.1 0.2
regression coefficients (95%CI)
Right CCA diameter
Pfor trend = 0.777
One In BLL SD increment
BLL Quartile 4
BLL Quartile 3 h-
BLL Quartile 2
BLL Quartile 1
r~
-0.2 -0.1 0 0.1 0.2
regression coefficients (95%CI)
BLL = blood lead level; CCA = common carotid artery; CI = confidence interval; CVD = cardiovascular disease; SD standard
deviation.
Source: Wan et al. (2021).
Figure 4-19 Relationship between blood Pb levels and common carotid artery
plaques, common carotid artery diameter, and cardiovascular
disease among diabetic patients.
A recent meta-analysis evaluating blood metals (including blood Pb) evaluated the aggregate
association between BLLs and CHD risk (Chowdhurv et al.. 2018). For this study, CHD was defined as
non-fatal MI, angina, coronary revascularization (i.e., percutaneous transluminal coronary angioplasty or
coronary artery bypass surgery) or CHD mortality. It included studies with cohort, case-control, or
nested-case-control study designs. In this analysis, a total of eight studies were identified including those
Exernal Review Draft
4-46
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
that evaluated CHD mortality, those that were previously included in the 2006 Pb AQCD (U.S. EPA.
2006b). and occupational studies. Even though none of the studies presented in this meta-analysis of CHD
and BLLs were included in this section (cardiovascular mortality discussed in Section 4.10), the overall
results further support an association between biomarkers of Pb exposure and IHD (Figure 4-20).
Lead
SOF
Glost
Zutph
BRI
Measurement No of No oS
soutce participants events
p Populatic
BIc
Bit
Bit
Bit
Relative risk
(95% CI)
Relative tisk
(95% CI)
4.9S
10
IS (1.00
S4 (1,21
McEtvenny (2015}
hh-\M: I..
Subtotal: P=0.001, i2=6?,6% 1.43(1,1610 1.76)
ABLES = Adult Blood Lead Epidemiology and Surveillance; BRHS = British Regional Heart Study; CI = confidence interval;
NHANES = National Health and Nutrition Examination Survey; SOF = Study of Osteoporotic Fractures; VA-NAS = Veterans Affairs
Normative Aging Study.
Source: Adapted from Chowdhurv et al. (2018).
Figure 4-20 Meta-analysis of the association between biomarkers of Pb
exposure and coronary heart disease.
4.4.2 Summary of Ischemic Heart Disease
Limited evidence was presented in the 2013 Pb ISA (U.S. EPA. 2013a) indicating an association
between biomarkers of Pb exposure and incident IHD. Although this effect was strong across both blood
and bone (patella) Pb measurements, there were not enough published studies at the time to fully evaluate
the association.
Several recent epidemiologic studies have been published further supporting this association.
Studies using the NAS cohort of elderly (mostly White) men indicated a positive association between
patella Pb levels and incident IHD (Ding et al.. 2019; Ding et al.. 2016). These studies had extensive
follow-up periods (-20 years), with patella Pb levels ranging between 29.2 and 32.2 jj.g/g. Additionally, a
series of 10-year CVD risk evaluations (Nguyen et al.. 2021; Park and Han. 2021; Choi et al.. 2020; Cho
et al.. 2016) were conducted using KNHANES data. These studies used cross-sectional data to create a
score that could be predictive of future CVD risk, and all indicated increased 10-year CVD risk with
increasing BLLs. BLLs in these studies generally averaged <3 (ig/dL.
While many of these studies evaluated the overall associations between biomarkers of Pb
exposure and IHD or other similar outcomes, many further stratified by sex, diet, and other distinguishing
Exernal Review Draft
4-47
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
24
25
26
27
28
29
30
31
characteristics such as genetic polymorphisms. Overall, males (Park and Han. 2021; Choi et al.. 2020;
Cho et al.. 2016) tended to have larger Pb-associated IHD risks than females and certain genetic
polymorphisms (Ding et al.. 2016) modified the relationship between bone Pb levels and incident IHD
Furthermore, diets low in fruit, whole grains, and vegetables were associated with a larger association
between bone Pb levels and incident IHD (Ding et al.. 2019).
4.5 Heart Failure and Impaired Cardiac Function
Heart failure refers to a set of conditions in which the heart's pumping action is weakened. With
congestive heart failure (CHF), the flow of blood from the heart slows and fails to meet the oxygen
demands of the body, and the returning blood can back up and cause swelling or edema in the lungs or
other tissues (typically in the legs and ankles). Right-sided heart failure is typically a consequence of left-
sided heart failure but can also result from damage to the pulmonary vasculature, which can result in
increased right ventricular (RV) mass, reduced flow to the left ventricle, and reduced left ventricular (LV)
mass. In chronic heart failure, the heart typically enlarges and develops more muscle mass. The 2006 Pb
AQCD (U.S. EPA. 2006b) presented limited epidemiologic evidence on the association between
biomarkers of Pb exposure and cardiac function. Little evidence was added in the 2013 Pb ISA. Since
then, the evidence has expanded modestly, with recent epidemiologic and toxicological studies providing
support for an effect between biomarkers of Pb exposure and cardiac function.
4.5.1 Epidemiologic Studies of Impaired Cardiac Function
The 2006 Pb AQCD presented a cross-sectional study indicating an association between Pb
biomarker levels and LV hypertrophy (Schwartz. 1991). More recent studies indicate an association
between Pb biomarkers and cardiac function. Study-specific details, including biomarker Pb levels, study
population characteristics, potential confounders, and select results from these studies are highlighted in
Table 4-8. These details include standardized results as well as those that could not be standardized based
on the information provided in each paper.
A recent small prospective study (Yang et al.. 2017) of a Flemish population evaluated potential
toxic effects of Pb on the myocardium by assessing the association between blood Pb and LV function.
Doppler imaging of transmural blood flow was used to assess systolic and diastolic LV function. In this
study, there was evidence of decreased LV systolic function for each doubling of blood Pb. Specifically,
there were decreases in global longitudinal strain (GLS) by 0.497% (95% CI: -0.957, -0.038%), regional
longitudinal strain (RLS) by 0.784% (-1.482, -0.087%), regional radial strain (RRS) by 2.316% (-4.748,
-0.115%), and regional longitudinal strain rate by 0.071s 1 (95% CI: -0.124, -0.019s '). There was no
association between BLLs and diastolic LV function. A cross-sectional study among a Swedish
population evaluated LV measurements using two-dimensional echocardiography measuring septal
Exernal Review Draft
4-48
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
24
25
26
27
28
29
30
thickness, posterior wall thickness, LV diameter in end diastole, and LV diameter in end systole (Lind et
al.. 2012). For natural log-transformed increases in serum Pb, there was a decrease in LV mass index
(LVMI) ((3: -0.73 [95% CI: -2.20, 0.74]) and an increase in relative wall thickness (RWT) ((3: 0.011
[95% CI: -0.001, 0.022]), but neither were statistically significant.
4.5.1.1 Impaired Cardiac Function in Children
The 2013 Pb ISA indicated that the small body of available evidence suggested a relationship
between biomarkers of Pb exposure and cardiac function in children, adding to the few studies presented
in the 2006 Pb AQCD . Specifically, Gump et al. (2011) evaluated cardiovascular responses, including
sympathetic and parasympathetic activation, to acute stress in children. Children aged 9-11 were
subjected to a variety of experimental tasks to stimulate the stress response. Cardiovascular
measurements, including cardiac output and stroke volume were assessed at baseline and following each
task. In general, increasing quartiles (Ql: 0.14-0.68 (ig/dL, Q4: 1.21-3.76 (ig/dL) of BLLs corresponded
to decreases in stroke volume and cardiac output, compared with baseline. These results support a
previous study by Gump et al. (2005). which had higher Pb exposures during early childhood.
A recent cross—sectional study provided further evidence of an association between more
sensitive cardiac outcomes (Chen et al.. 2021). This study evaluated Pb's potential effect on structural
function and inflammation related to the LV function in children. Children were recruited from two
different primary areas, including an e-waste exposed area (Guiyu) and a reference area (Haojiang).
Several different LV measurements were obtained. A 1-unit increase in BLLs was associated with smaller
(natural log-transformed) interventricular septum (IVS) measurements ([3: -0.004 (95% CI: -0.007,
-0.001). Other natural log-transformed echocardiogram measurements indicated null associations (LV
posterior wall [3: -0.001 [95% CI: -0.003, 0.001]); ejection fraction [3: -0.001 (95% CI: -0.002, 0.001)
with a unit increase in BLLs.
4.5.2 Toxicological Studies of Impaired Cardiac Function
The previous ISA did not include any animal toxicological studies examining impaired cardiac
function. However, animal toxicological studies published since the last review have looked at the
potential for Pb exposure to alter cardiac function. Wildemann et al. (2015) reported no evidence for an
effect of Pb exposure on stroke volume or cardiac output in rats. Moreover, combined exposure to Pb,
mercury, and methylmercury resulted in no change in these measures relative to controls. In contrast,
Fioresi et al. (2014) reported a statistically significant increase in some measures of cardiac contractility
in rats. More specifically, they found a statistically significant increase in left ventricular systolic pressure
(LVSP) and LV dP/dt, (change in pressure/change in time; p < 0.05), but not right ventricular systolic
Exernal Review Draft
4-49
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
24
25
26
27
28
29
30
31
32
33
34
pressure (RVSP) or RV dP/dt following a 30-day exposure to Pb (13.6 jj.g/dl mean BLL). There were also
no changes reported in left or right ventricular diastolic pressure (LVDP, RVDP) (Fioresi et al.. 2014).
Additional studies examining the potential for impaired cardiac function were done in isolated
LV papillary muscle. Silva et al. (2015) reported no significant difference in force generation between
muscle isolated from control or Pb-treated (15-day exposure, 12.3 jj.g/dl BLL) rats following pulse
stimulation. However, the time to peak tension and 90% relaxation was statistically significantly
(p < 0.05) shorter in LV papillary muscle derived from Pb-treated animals relative to muscle from control
animals. Moreover, inotropic contractile force was statistically significantly decreased in muscle from Pb-
treated animals following treatment with calcium chloride, but not isoproterenol (Silva et al.. 2015) and
Pb exposure significantly lowered tetanic (sustained) peak and plateau force. In a similar analysis in LV
papillary muscle, Fioresi et al. (2014) reported that following a 30-day exposure to Pb resulting in a mean
13.6 (ig/dl BLL, there were not significant differences in isometric contraction force, time to peak
contraction or relaxation rates. However, in contrast to Silva et al. (2015). following rest and calcium
treatment, there was a statistically significant increase in contractile force in muscle from Pb-treated
animals (Fioresi et al.. 2014). When considered as a whole, the animal toxicological evidence for changes
in cardiac function is limited and, in some cases, results across studies are conflicting. Additional details
for the toxicological studies discussed in this section can be found in Table 4-9 of this ISA.
4.5.3 Integrated Summary of Impaired Cardiac Function
Limited evidence was presented in the 2013 Pb ISA (U.S. EPA. 2013a) indicating an association
between biomarkers of Pb exposure and indicators of cardiac function. The recent epidemiologic evidence
suggests that the potential effect of Pb exposure on cardiac function may be more likely among children
and the elderly. An analysis of participants >70 years of age indicated positive associations between
markers of LV function and blood Pb, with relatively low mean BLLs (<2 ug/dL) (Lind et al.. 2012).
Associations with these same outcomes were null in a cohort of middle-aged participants (mean age
-39 years), although there was evidence of an association with markers of LV structure within this cohort
(Yang et al.. 2017). A study among children indicated a relationship between smaller IVS measurements
with increased BLLs (Chen et al.. 2021). Results of available animal studies examining cardiac function
have been inconsistent, and conflicting results were reported in studies examining contractile force in
isolated papillary muscle following calcium treatment (Silva et al.. 2015; Fioresi et al.. 2014).
A small number of studies presented associations between decreased stroke volume with
increasing BLLs (Gump etal.. 2011; Gump et al.. 2005). but results were less consistent when
considering cardiac output. In animal toxicological studies there was no evidence of an effect of Pb
exposure on stroke volume or cardiac output, but limited evidence for an effect on measures of cardiac
contractility. Taken together, there is limited evidence to support a relationship between biomarkers of Pb
exposure and cardiac function.
Exernal Review Draft
4-50
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
24
25
26
27
28
29
30
31
32
33
4.6 Endothelial Dysfunction
Endothelial dysfunction is the physiological impairment of the inner lining of blood vessels that is
characterized by an imbalance between vasodilators such as nitric oxide and vasoconstrictors such as
endothelin-1 (ET-1). High BP is often the result of an imbalance of these factors that leads to greater
vasoconstriction.
4.6.1 Toxicological Studies of Endothelial Dysfunction
In the 2013 Pb ISA, animal toxicological studies provided mixed evidence for Pb exposure
having an effect on vascular relaxation and constriction. For example, although Pb exposure decreased
acetylcholine (ACh)-induced vasodilation in isolated rat tail arteries (Silvcira ct al.. 2010; Zhang ct al..
2007). Skoczvnska and Stoiek (2005) reported that Pb exposure enhanced vasodilation by ACh in rat
mesenteric arteries. Moreover, in aortic rings of perinatally exposed rats, there was no change observed in
the relaxation response to ACh (Tiorim et al.. 2011; Rizzi et al.. 2009; Grizzo and Cordellini. 2008V More
information on these and other studies examining vascular reactivity from previous reviews can be found
in Section 4.4.2.3 of the 2013 Pb ISA (U.S. EPA. 2013b)
Since the publication of the 2013 Pb ISA, additional toxicological studies of vascular function
have been published in animals with BLLs <30 (ig/dl. In a study of young rats exposed to Pb through
lactation, Pb exposure (BLL of ~11 jj.g/dl to 20 jj.g/dl) resulted in a statistically significant (p < 0.05)
increase in the maximum contractile response to the vasoconstrictor noradrenaline in intact rat aortas at
days 52, 70, and 100 (but not at day 23) relative to control animals (Gaspar and Cordellini. 2014). In
denuded aortas (i.e., aortas with no endothelium), the contractile response to noradrenaline increased
comparably from both control and Pb-treated animals, thereby suggesting that the difference in the
contractile response in intact aortas was the result of Pb's effect on the endothelium (Gaspar and
Cordellini. 2014). However, in an additional study using adult rats, there was no difference in intact rat
aortic segments from control or Pb-exposed rats when treated with the vasodilators ACh or sodium
nitroprusside, and a statistically significant (p < 0.05) decrease in the contractile response following
exposure to the vasoconstrictor phenylephrine, but not potassium chloride (Nunes et al.. 2015). The BLL
in this study was 8.4 jj.g/dl and when the endothelium was mechanically removed, phenylephrine-induced
contractility increased in both groups but to a greater extent in aortic segments from Pb-treated rats. Using
a number of chemical inhibitors, the authors suggest that the decrease in contractility in response to
phenylephrine (in intact aortic segments) was not due to a Pb effect on the vasodilator NO, but rather to
increasing levels of hydrogen peroxide, which can also have vasodilatory effects. That is, incubation with
catalase increased the constriction response to phenylephrine in aortic segments from Pb-treated rats but
not control rats. The authors go on to show that differences in hydrogen peroxide activity between aortic
segments from Pb-treated and control rats is potentially due to Pb increasing the levels of the hydrogen
peroxide generating enzyme superoxide dismutase (SOD) (Nunes et al.. 2015).
Exernal Review Draft
4-51
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
24
25
26
27
28
29
4.6.2
Summary of Endothelial Dysfunction
Taken together, the limited toxicological evidence presented above suggests that Pb exposure
may result in changes in endothelial function. However, the direction of this response varies in that Pb
exposure can either increase or decrease the response to vasodilators/vasoconstrictors. These studies also
suggest that Pb's effects on the endothelium are complicated and differ depending on age, treatment (e.g.,
vasodilators testing endothelium-dependent versus endothelium-independent mechanisms), and/or type of
endothelial cells tested. Additional details for the toxicological studies discussed in this section can be
found in Table 4-10 of this ISA.
4.7 Cardiac Electrophysiology and Arrythmia
Electrical activity in the heart is crucial for regulating the heartbeat and is typically measured
using surface electrocardiography (ECG). ECGs measure electrical activity in the heart that is due to
depolarization and repolarization of the atria and ventricles. Changes in electrical activity can lead to
changes in cardiac depolarization, repolarization, and development of arrythmia (Section 4.7.1) and
changes in heart rate and HRV (Section 4.7.2)
4.7.1 Cardiac Depolarization, Repolarization, and Arrythmia
Experimental and epidemiologic studies typically use surface ECGs to measure electrical activity
in the heart resulting from depolarization and repolarization of the atria and ventricles. The P-wave of the
ECG corresponds to atrial depolarization, the QRS complex represents ventricular depolarization, and the
T-wave represents ventricular repolarization. The ventricles account for the largest proportion of heart
mass overall and thus are the primary determinants of the electrical activity recorded in the ECG.
Therefore, ECG changes indicating abnormal electrical activity in the ventricles are of greatest concern.
Endpoints denoting ventricular electrical activity include QTc interval, transmural dispersion duration,
and T-wave shape. Changes in QT and ST, as well as changes in T-wave shape, duration, or amplitude,
may indicate abnormal impulse propagation in the ventricles.
Cardiac arrhythmias can vary in severity from the benign to the potentially lethal, such as in
cardiac arrest when an electrical disturbance disrupts the heart's pumping action causing loss of heart
function. Atrial fibrillation (AF) is the most common type of arrhythmia. Clinical and subclinical forms of
AF are associated with reduced functional status and quality of life, as well as downstream consequences
such as ischemic stroke (Prvstowskv et al.. 1996; Anonymous. 1994) and CHF (Roy et al.. 2009).
contributing to both cardiovascular disease and all-cause mortality (Kannel et al.. 1983). Ventricular
fibrillation is a well-known cause of sudden cardiac death and is commonly associated with MI, heart
failure, cardiomyopathy, and other forms of structural (e.g., valvular) heart disease. Pathophysiologic
Exernal Review Draft
4-52
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
24
25
26
27
28
29
30
mechanisms underlying arrhythmia include electrolyte abnormalities, modulation of the autonomic
nervous system (ANS), membrane channels, gap junctions, oxidant stress, myocardial stretch, and
ischemia. Ventricular conduction and repolarization abnormalities such as QRS complex and QT interval
prolongation, as well as LV hypertrophy and clinical antecedents including hypertension, are also
associated with cardiac arrest (Rautahariu et al.. 1994).
4.7.1.1 Epidemiologic Studies of Cardiac Depolarization, Repolarization, and
Arrythmia
Numerous epidemiologic studies evaluated in the 2013 Pb ISA (U.S. EPA. 2013a) strengthened
the evidence presented in the 2006 Pb AQCD (U.S. EPA. 2006b) that described an association between
biomarkers of Pb exposure and changes in ECG measures. Current studies continue to support prior
analyses. Study-specific details, including blood and bone Pb levels, study population characteristics,
potential confounders, and select results from these studies, are highlighted in Table 4-11. These study
details include standardized results as well as results that could not be standardized based on the
information provided in each paper.
Previous ISAs described analyses evaluating the association between Pb biomarkers and
electrophysiologic outcomes using the NAS cohort, of mostly white men. For example, Cheng et al.
(1998) described an association between bone Pb and corrected QT interval (QTc) among men >65 years,
and Eum etal. (2011) prospectively evaluated ECG findings and bone Pb levels within the NAS cohort.
Eum etal. (2011) reported an association between tibia Pb levels and increases in QTc interval (7.94
msec [95% CI: 1.42, 14.45]) and QRSc duration (5.94 msec [95% CI: 1.66, 10.22]) when comparing the
highest tertile of bone Pb levels with the lowest. Additionally, a cross-sectional analysis of elderly NAS
men provided evidence of EMM of certain genetic polymorphisms in genes affecting iron (Fe)
metabolism (HFE C282Y and HMOX1 L variants) on the relationship between biomarkers of Pb exposure
and prolonged QT interval (Park et al.. 2009).
A recent study supports these previous findings in a more diverse population (NHANES),
compared with the NAS cohort of mostly white men. Jing et al. (2019) used NHANES III (1988-1994) to
evaluate the relationship between log-transformed BLLs and the QRS-T angle. The QRS-T angle can
quantify the relationship between ventricular depolarization (QRS-axis) and repolarization (T-axis) and is
a predictor of ventricular arrythmia. The QRS-T angle was measured using a standard 12-lead ECG, and
sex-specific tertiles of QRS-T angle were created. This study indicated that higher BLLs were associated
with a greater QRS-T angle (third tertile versus first tertile) among men (OR: 1.35 [95% CI: 1.05, 1.74]),
but not among women (OR: 1.05 [95% CI: 0.82, 1.36]).
Exernal Review Draft
4-53
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
24
25
26
27
28
29
4.7.1.2 Toxicological Studies of Cardiac Depolarization, Repolarization, and
Arrythmia
The 2013 Pb ISA evaluated an animal toxicological study demonstrating that exposure to Pb
resulted in increased incidence of arrhythmia and atrioventricular conduction block (i.e., disruption of
electrical signals from the atria to the ventricles) after 12 weeks of Pb exposure . This study also reported
a prolonged ST interval, without alteration in QRS duration. Since the last review, Wildemann et al.
(2015) reported no change relative to control animals for PR, QRS, or QT for rats with a 1.7 jj.g/dl or
8.6 (ig/dl BLL. A combined exposure of Pb, mercury, and methylmercury resulted, however, in
significant increases in the QRS and QT intervals. Taken together, there is little animal evidence for an
effect of Pb exposure alone on cardiac depolarization and/or repolarization.
4.7.2 Heart Rate and Heart Rate Variability
Heart rate is a key indicator of autonomic function. It is modulated at the sinoatrial node of the
heart by both parasympathetic and sympathetic branches of the ANS and represents the number of times
the heart beats in a given time frame (e.g., per minute). In general, increased sympathetic activation
increases heart rate, while enhanced activation of parasympathetic, vagal tone decreases heart rate (Lahiri
et al.. 2008). Heart rate variability (HRV) represents the degree of difference in the inter-beat intervals of
successive heartbeats. Given that both arms of the ANS contribute, changes in HRV are an indicator of
the relative balance of sympathetic and parasympathetic tone to the heart and their interaction (Rowan et
al.. 2007). Low HRV is associated with an increased risk of cardiac arrhythmia and an increased risk of
mortality in patients with CHF awaiting a heart or lung transplant (Fauchier et al.. 2004; Bigger et al..
1992). Low HRV has also been shown to be predictive of CAD (Kotccha et al.. 2012). Notably, increases
in HRV have also been associated with increases in mortality (Carll et al.. 2018). In general, the two most
common ways to measure HRV are time-domain measures of variability and frequency-domain analysis
of the power spectrum. With respect to time-domain measures, the standard deviation of normal-to-
normal (NN) intervals (i.e., the interval between consecutive normal beats) reflects overall HRV, and
root-mean-square of successive differences (rMSSD) in NN intervals reflects parasympathetic influence
on the heart. In terms of frequency domain, high-frequency (HF) domain is widely thought to reflect
cardiac parasympathetic activity while the low-frequency (LF) domain has been posited as an indicator of
the interaction of the sympathetic and parasympathetic nervous systems (Billman. 2013). although its
linkage with sympathetic tone is controversial and uncertain (Notarius etal.. 1999).
4.7.2.1 Epidemiologic Studies of Heart Rate and Heart Rate Variability
A small number of studies examining the relationship between Pb biomarkers and heart rate or
HRV were evaluated in the 2013 Pb ISA . However, the studies characterized in the 2013 Pb ISA related
Exernal Review Draft
4-54
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
24
25
26
27
28
29
30
31
32
33
to heart rate and HRV were all within the NAS cohort. Specifically, Park etal. (2006) presented evidence
of a relationship between patella Pb levels and decreased HRV among those with three or more metabolic
abnormalities (waist circumference >102 cm, hypertriglyceridemia >150 mg/dL, HDL-C <40 mg/dL, BP
>130/85 mmHg, fasting glucose >110 mg/dL). The results of this study supported previous research
presented in the 2006 Pb AQCD. Study-specific details, including blood and bone Pb levels, study
population characteristics, potential confounders, and select results from these studies are presented in
Table 4-11. These study details include standardized results as well as results that could not be
standardized based on the information provided in each paper.
In a more recent study, Gump et al. (2017) evaluated the association between BLLs and HRV
among children (aged 9-11) as part of the Environmental Exposures and Child Health Outcomes study.
However, associations between BLLs (range: 0.19-3.25 j^ig/dL) and HRV were null within this group.
Another recent analysis evaluated the effect of BLLs and HRV among children (age 12) . This study
obtained blood Pb measurements at two time points (aged 3-5 and 12), while HRV was measured only at
age 12. Children in this study were given a standardized stressful stimulus known as the Public Speaking
Stress task. In this task, children were asked to first plan a speech to deliver (planning phase) and then
present that speech to the research assistant (speaking phase) while being continuously monitored for
HRV. For the planning phase, there was a null association between a HRV frequency measure (LF/HF)
and BLLs at ages 3-5 (0.03 [-0.02, 0.09]) and at age 12 (-0.04 [-0.16, 0.07]). For the speaking phase,
there was a positive association between HRV frequency and BLLs at ages 3-5 (0.06 [0.01, 0.12]), but
not with BLLs at age 12 (0.05 [-0.18, 0.08]). An increase in the LF/HF ratio is associated with a shift to
sympathetic dominance and an overall decrease in HRV, which is suggestive of a dysregulated stress
response.
4.7.2.2 Toxicological Studies of Heart Rate and Heart Rate Variability
The 2013 Pb ISA discussed a limited number of animal toxicological studies demonstrating that
exposure to Pb increased heart rate (SimSes etal.. 2011; Badavi et al.. 2008; Lai et al.. 2002). There were
no studies that examined changes in HRV in response to Pb exposure.
Since the publication of the 2013 Pb ISA, there have been additional toxicological studies
published with respect to exposure to Pb and HR. Fioresi et al. (2014) reported statistically significantly
higher heart rate (p < 0.05) in rats with a BLL of 13.6 jj.g/dl, relative to control animals. However,
Wildemann et al. (2015) reported no change relative to control animals for heart rate in rats with a
1.7 (ig/dl or 8.6 jj.g/dl BLL or following combined exposure with mercury or methylmercury. Other
studies were similarly mixed, with some reporting statistically significant increases in heart rate following
Pb exposure (Zhu et al.. 2019; Zhu et al.. 2018). while another study using two different exposure
scenarios did not (Shvachiv et al.. 2018). Thus, overall, there is mixed evidence from animal toxicological
Exernal Review Draft
4-55
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
24
25
26
27
28
29
30
31
32
33
34
studies for an increase in heart rate following Pb exposure. Additional details for the toxicological studies
discussed above can be found in Table 4-12 of this ISA.
Since the 2013 Pb ISA, there have also been animal toxicological studies published with BLLs
<30 (ig/dl examining the relationship between Pb exposure and changes in HRV. Shvachiv et al. (2018)
reported a statistically significant increase in LF in rats continuously, but not intermittently exposed to Pb
relative to control animals. However, no changes in HF, or the LF/HF ratio were reported in either group
relative to controls. BLLs in this study were approximately 24 jj.g/dl for the constant exposure group and
approximately 19 jj.g/dl for the intermittent exposed group (Shvachiv et al.. 20 IS). Additionally, in a pair
of analyses by the same laboratory, there was a statistically significant increase (p < 0.05) in the LF/HF
ratio and a statistically significant decrease in LF and HF. BLLs in this study were <30 jj.g/dl (Zhu et al..
2019; Zhu et al.. 2018). Thus, there is only limited evidence from animal toxicological studies for an
effect of Pb exposure on measures of HRV at BLLs <30 jj.g/dl. Additional details for the toxicological
studies discussed above can be found in Table 4-12 of this ISA.
4.7.3 Integrated Summary of Cardiac Electrophysiology and Arrythmia
Exposure to Pb has been shown to affect contractility in animals and to be associated with cardiac
contractility in epidemiologic studies. The epidemiologic evidence supports an association between
altered ECG measures and biomarkers of Pb exposure. Specifically, a series of studies using the NAS
cohort presented in the 2013 Pb ISA indicated an association between bone Pb levels and a prolonged QT
interval (Eum etal.. 2011; Park et al.. 2009; Cheng et al.. 1998). There is evidence suggesting that a
lengthening of the QT interval increases risk of future abnormal heart rhythm or sudden cardiac arrest.
However, these NAS studies were small and evaluated mostly white, elderly men. A recent study
evaluated an earlier cohort of NHANES participants (NHANES III 1988-1994) (Jing et al.. 2019). This
study included a much larger sample size and a more diverse group of subjects. In this cross-sectional
analysis, there was evidence of an increased QRS-T angle associated with BLLs. The effect was most
prominent in males compared with females. Despite the relatively consistent evidence observed within the
epidemiologic literature, the toxicological literature is sparce and more mixed. There was a single study in
the last review demonstrating increased incidence of arrhythmia, atrioventricular block, and a prolonged
ST segment interval (Reza etal.. 2008). Since the last review, an additional study reported no change in
the PR, QRS, or QT segments in rats (Wildemann et al.. 2015)
The epidemiologic evidence for an association between biomarkers of Pb exposure and either
heart rate or HRV are less compelling. Few studies evaluate this outcome. An earlier analysis of the NAS
cohort indicated an association between bone Pb measurements and a decrease in HRV among elderly
white men (Park et al.. 2006). This supported evidence from occupational studies presented in the 2013
Pb ISA (U.S. EPA. 2013a). Results from recent studies in children are not consistent. An analysis among
a small group of children yielded no association between BLLs and HRV Gump et al. (2017). whereas a
Exernal Review Draft
4-56
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
24
25
26
27
28
29
30
31
separate analysis indicated a slight decrease in HRV among 12-year-old children with their BLLs when
they were between the ages of 3-5 (Halabickv et al.. 2022). The toxicological evidence for exposure to Pb
and changes in heart rate was largely mixed. Some animal toxicological studies reported increases in heart
rate following Pb exposures (Zhu et al.. 2019; Zhu et al.. 2018; Fioresi etal.. 2014). whereas other animal
studies reported no change (Shvachiv et al.. 2018; Wildemann et al.. 2015). With respect to HRV, there
was a limited number of animal toxicological studies, but they reported changes in some measures of
HRV following Pb exposure (Zhu et al.. 2019; Shvachiv et al.. 2018; Zhu et al.. 2018). That said, there
were differences among these studies with respect to which measures of HRV changed, or the direction of
change for a given measure. Taken together, the relatively small body of evidence from epidemiologic
and toxicological studies examining Pb exposure and changes in cardiac electrophysiology and arrythmia
have reported mixed results.
4.8 Atherosclerosis and Peripheral Artery Disease
Atherosclerosis is the process of plaque buildup into lesions on the walls of the coronary arteries
that can lead to vessel narrowing, reduced blood flow to the heart, and IHD. The development of
atherosclerosis is dependent on the interplay between plasma lipoproteins, inflammation, endothelial
activation, and neutrophil attraction to the endothelium, extravasation, and lipid uptake. Risk factors for
atherosclerosis include high low-density lipoprotein (LDL)/low HDL cholesterol, high BP, diabetes,
obesity, smoking, and increasing age. Measures of subclinical atherosclerosis provide the opportunity to
assess the pathogenesis of vascular disease at an earlier stage. PAD is an indicator of atherosclerosis and
is measured by the ankle brachial index, which is the ratio of BP between the posterior tibia artery and the
brachial artery. An ankle brachial index of less than 0.9 is typically indicative of the presence of PAD.
Prior toxicological studies have reported that Pb can increase atheromatous plaque formation in pigeons,
increase arterial pressure, decrease heart rate and blood flow, and alter cardiac energy metabolism and
conduction (Prentice and Kopp. 1985; Revis et al.. 1981).
4.8.1 Epidemiologic Studies of Atherosclerosis and Peripheral Artery
Disease
A limited number of studies have evaluated the effects of biomarkers of Pb exposure and
atherosclerosis. The 2013 Pb ISA (U.S. EPA. 2013a) described an association between BLLs and both
intimal medial thickening (IMT) and atherosclerotic plaque presentation in an occupational study, among
those with high concentrations of blood Pb (-25 (ig/dL) (Poreba et al.. 2011). Recent studies further
expand the knowledge base for the relationship between Pb biomarkers and atherosclerosis and PAD.
Study-specific details, including BLLs, study population characteristics, potential confounders, and select
results from these studies are highlighted in Table 4-13. These details include standardized results as well
as those that could not be standardized based on the information provided in each paper.
Exernal Review Draft
4-57
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
A study published since the 2013 Pb ISA evaluated diabetic patients in China (Wan et al.. 2021).
This study evaluated the association between common carotid artery (CCA) plaques and BLLs. When
comparing the highest quartile of BLLs (>3.7 (ig/dL) with the lowest quartile of BLLs (<1.8 (ig/dL), there
were increased odds (OR: 1.53 [95% CI: 1.29, 1.82]) of CCA plaque. The diameter of the CCA did not
appear related to BLLs (Figure 4-19).
Another recent analysis described the association between hemodynamic measures (peripheral
BP, central BP, and time-dependent hemodynamics), which assess arterial stiffness, and BLLs among a
Flemish population (Yu et al.. 2020). Blood Pb was collected at least once during the study period (1985
to 2005), and participants were followed for a median of 9.4 years. BLLs within this population were
relatively low (GM: 2.93 (ig/dL, IQR: 1.8-4.7). At the final follow-up, trained personnel assessed
measures of arterial stiffness. Overall, measures of peripheral BP or central BP were not associated with
BLLs. However, for every doubling of BLLs, several measures of time-dependent hemodynamics were
elevated, including augmentation ratio (1.74% [95% CI: 0.95, 2.53%]), augmentation index (3.03% [95%
CI: 1.56, 4.50]), forward pulse peak time (6.62% [95% CI: 2.21, 11.0%]), backward PP amplitude
(1.02 mmHg [95% CI: 0.02, 2.02 mmHg]), and reflection index (3.98% [95% CI: 1.71, 6.24%]).
However, the association with aortic pulse wave velocity (aPWV) was null (0.14 ms [95% CI: -0.08,
0.35 ms]), and age appeared to be a major component of increases in aPWV (Figure 4-21). The sum of
these results from this study indicates an association between relatively low BLLs and evidence of
atherosclerosis (Table 4-13).
Exernal Review Draft
4-58
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
Panel A: pulse wave velocity was only standardized to a heart rate of 60 beats per minute; Panel B: the associations were fully
adjusted.
Source: Yu et al. (2020).
Figure 4-21 Association between aortic pulse wave velocity with blood Pb
levels and age.
A large Korean study evaluated the association between BLLs and coronary artery stenosis
(CAS), which is the blockage or narrowing of the arteries that supply blood to the heart (Kim et al..
2021). This study performed a coronary computerized tomography (CT) angiography and classified
participants with CAS if they had >25% stenosis. Overall, each 1 (ig/dL increase in BLLs was associated
with increased odds of CAS (OR: 1.14 [95% CI: 1.02, 1.26]). Many studies of atherosclerosis focus on
calcification or blockage of the coronary artery, but a recent NHANES analysis focused on abdominal
aortic calcification (AAC) (Qin et al.. 2021). AAC is a marker of subclinical atherosclerosis and a
predictor of future CVD events. Lateral lumbar spine images, using the Kauppila score system were used
to score the AAC severity on a scale from 0 to 24, and a total AAC score >6 was considered to be
substantial calcification of the abdominal aorta. Overall, each one-unit increase in BLLs corresponded to
a 0.15-unit increase (95% CI: 0.02, 0.27) in total AAC score and an 11% increase in severe AAC (OR:
1.11 [95% CI: 1.00, 1.22]). There were no differences in association when stratified by race, sex, age,
BMI, hypertension, or diabetic status (Figure 4-22).
Exernal Review Draft
4-59
DRAFT: Do not cite or quote
-------
AAC Score
P (95% CI)
P for trend
P for interaction
Race
Mexican American
-0.02 (-0.28.0.24)
0.9
0.49
Other Hispanic
0.37 (-0.05. 0.78)
0.084
Non-Hispanic White
0.34 (0 09. 0 59)
0.0075
Non-Hispanic Black
-0.07 (-0.22. 0.09)
0.4
¦—•—>
Other Races
0.01 (-0.33. 0.34)
098
Gender
Male
0.16(0.02, 0.30)
0.025
•—•—i
0.22
Female
0.27 (0.17.0.37)
0.045
-¦-
Age (years)
Age < 60
0.11 (0.00.0.22)
0.031
0.34
Age 2 60
0.23 (0.16. 0.29)
0.042
BMI
Normal weight
0.05 (-0.14, 0.38)
0.36
0.61
Overweight
0.09 (0.01,0.22)
0.0019
>-•—
Obese
0.32(0.12, 0.52)
0.0023
Hypertension
Yes
0 17(0 09, 0 4?)
n n?i
i-m ¦
0 64
No
0.13(0.00, 0.25)
0 044
Diabetes
Yes
0.05 (0.01.0.12)
0.02
¦ <
0.33
NO
0.16(0.04. 0,28)
0.01
1 1
1
AAC = abdominal aortic calcification; BLL = blood lead level; BMI = body mass index; CI = confidence interval; P = p-value.
Source: Qin et aL (2021).
Figure 4-22 Stratified associations between abdominal aortic calcification
score and blood Pb levels.
1 The 2013 Pb ISA described epidemiologic studies assessing the relationship between biomarkers
2 of Pb exposure and prevalent PAD. An NHANES (1999-2002) analysis observed an increasing trend in
3 the odds of PAD with increasing concurrent BLLs (Muntner et al.. 2005). Another NHANES (1999-
4 2000) analysis also indicated a trend of increasing odds of PAD with increasing quartiles of concurrent
5 BLLs, among adults >40 years (Navas-Acien et al.. 2004). However, these results were not statistically
6 significant for any quartile of Pb exposure. To date, no recent studies evaluating biomarkers of Pb
7 exposure and PAD have been conducted for inclusion in the current review.
4.8.2 Toxicological Studies of Atherosclerosis
8 In the 2013 Pb ISA, a study in rats demonstrated that Pb exposure increased the aortic media
9 thickness, media-lumen ratio, and medial collagen content (Zhang et al.. 2009). Since the publication of
10 that document, Xu et al. (2015) reported a statistically significant increase in proliferating cell nuclear
11 antigen in cardiac tissue (p < 0.05) in rats exposed to Pb up to 12 or 40 days. This result potentially
Exernal Review Draft
4-60
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
24
25
26
27
28
indicates increased cellular division and/or DNA repair following exposure to Pb, which is relevant given
that increased cellular proliferation plays a role in atherosclerotic plaque growth. Moreover, in the 40-day
exposure group, these authors reported a statistically significant increase in the diameter of the cells of the
aorta, as well as changes in the shape (i.e., loss of curvature) of the aortic internal elastic lumen relative to
control animals. The blood Pb concentrations in this study were 19.3 jj.g/dl on day 12 and 24.6 jj.g/dl on
day 40 (Xu et al.. 2015). Additional details for the animal toxicological studies discussed in this section
can be found in Table 4-14 of this ISA.
4.8.3 Integrated Summary of Atherosclerosis
At the time of the 2013 Pb ISA, there were few studies examining the relationship between
biomarkers of Pb exposure and measures of atherosclerosis and PAD. Overall, these studies were mixed,
with some indicating an association between BLLs and IMT or an increase in the odds of PAD
prevalence, while others did not indicate a relationship between BLLs and prevalent PAD. Recent
epidemiologic evidence indicates a consistent positive association between markers of atherosclerosis and
Pb exposure. Atherosclerotic evidence is measured differently between the included studies, but further
supports the notion of an association between BLLs and plaque formation. While there is strong evidence
that markers of atherosclerosis increase with age, BLLs also appear to play a substantial role. The
toxicological evidence from the previous and current ISA is limited but supports epidemiologic studies
demonstrating a positive association between Pb exposure and markers of atherosclerosis. More
specifically, there is animal toxicological evidence of morphological changes in the aorta consistent with
the potential for atherosclerosis. Taken together, there is evidence from both epidemiologic and
toxicological studies to support an association between biomarkers of Pb exposure and makers of
atherosclerosis development.
4.9 Cerebrovascular Disease
Cerebrovascular disease describes a group of conditions involving the cerebral blood vessels that
result in transient or permanent disruption of blood flow to the brain. These conditions include stroke,
transient ischemic attack, and subarachnoid hemorrhage. Both hypertension and atherosclerosis are risk
factors for cerebrovascular disease and the mechanisms for these outcomes also apply to cerebrovascular
disease. Very few studies have examined the effects of Pb exposure on cerebrovascular disease.
4.9.1 Epidemiologic Studies of Cerebrovascular Disease
The 2013 Pb ISA described a limited number of epidemiologic studies that examined
associations between Pb exposure and cerebrovascular disease. Two previous prospective epidemiologic
Exernal Review Draft
4-61
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
24
25
26
27
28
studies evaluated mortality from stroke . In an NHANES analysis, Menke et al. (2006) indicated that
increases in BLLs were associated with an increase in stroke mortality, although the association was
imprecise. In contrast, Khalil et al. (2009) reported a null association between BLLs and stroke mortality.
In a cross-sectional study in Taiwan, Lee et al. (2009) reported an association between increased
intracranial and extracranial stenosis (>50%) and urine Pb concentrations but not blood Pb
concentrations.
In a recent small case-control analysis, Mousavi-Mirzaei et al. (2020) evaluated acute ischemic
stroke in relation to BLLs among patients in Iran. Cases (n = 44) of acute ischemic stroke were matched
to controls (n = 44) based on age, sex, occupation, opium addiction, and sampling time. Participants in
this study had relatively high BLLs (median: 6.38 (ig/dL, IQR: 1.75-34.87). There was an association
between increased BLLs and increased risk of acute ischemic stroke (OR: 1.04 [95% CI: 1.02, 1.07] for a
1 (ig/dL increase in blood Pb). Study-specific details, including BLLs, study population characteristics,
potential confounders, and select results are highlighted in Table 4-15. Study details in Table 4-15 include
standardized results as well as results that could not be standardized based on the information provided in
each paper There were no animal toxicological studies at BLLs <30 (ig/dL that examined the relationship
between Pb exposure and cerebrovascular disease.
4.9.2 Summary of Cerebrovascular Disease
Few studies have examined the relationship between biomarkers of Pb exposure and
cerebrovascular disease. A small amount of evidence was presented in the 2013 Pb ISA suggesting an
association between Pb exposure and stroke mortality or stenosis in the intracarotid system. Since the
publication of these prior documents, however, very little additional epidemiologic information can be
added to the current evidence base. Moreover, there were no relevant animal toxicological studies at
BLLs <30 (ig/dL. Thus, the evidence to suggest an association between Pb exposure and cerebrovascular
disease is limited.
4.10 Cardiovascular Mortality
4.10.1 Epidemiologic Studies of Cardiovascular Mortality
Studies that examine the association between biomarkers of Pb exposure and cause-specific
mortality outcomes, such as cardiovascular mortality, provide additional evidence for Pb-related
cardiovascular effects, specifically whether there is evidence of an overall continuum of effects. Several
epidemiologic studies evaluated in the 2013 Pb ISA (U.S. EPA. 2013a) strengthened the evidence
presented in the 2006 Pb AQCD (U.S. EPA. 2006b) indicating an association between Pb biomarkers of
Exernal Review Draft
4-62
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
exposure and cardiovascular mortality. The strongest evidence came from multiple prospective cohort
studies that observed consistent positive associations with CVD mortality across different populations,
while also using different model specifications and approaches to control for a wide range of potential
confounders. The majority of cohort studies evaluated in the 2013 Pb ISA utilized blood Pb data from
NHANES II and III, which was then linked prospectively to mortality data, with between 8-16 years of
follow-up (Mcnkc et al.. 2006; Schober et al.. 2006; Lustberg and Silbcrgcld. 2002). Additional
prospective cohort studies, specifically among older adults, reported that CVD mortality was associated
with Pb measured in blood (Khalil et al.. 2009) and in the tibia and the patella (Wcisskopf et al.. 2009).
Notably, adult BLLs may be representative of contributions from both recent Pb exposures and
mobilization of legacy Pb from bone, therefore it remains unclear as to what extent either recent, past, or
cumulative Pb exposures contribute to the observed associations with cardiovascular mortality. Because
of the rapid decline in ambient air Pb concentrations and population BLLs that corresponded with the
phase out of leaded gasoline, participants ofNHANES II (1976-1980) and NHANES III (1988-1994)
likely had higher past Pb exposures compared with exposure at the time of blood collection—further
complicating the determination of BLLs that might contribute to the observed associations. Recent studies
continue to provide evidence of consistent positive associations between exposure to Pb and CVD
mortality (Figure 4-23). Study-specific details, including biomarker Pb levels, study population
characteristics, confounders, and select results from these studies, are highlighted in Table 4-16.
Exernal Review Draft
4-63
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
„ . ... „ . .• „ .. . ..... Pb measurement -Years of ,, .
Reference Study Population Pb distribution Mortality
year follow-up
Cardiovascular Mortality:
Menkeetaletal,2006 NHANESIII Adults *20 Mean:2.59 1988-1994 12 CVD
TLanphear et al et al, 2018 NHANES III Adults Ł20 ?Mm0SC 1988-1994 19 CVD
Geometric SE: 1.32
Median
TVanBemmel etal etal, 2011 NHANES III Adults *40 <5ug/dL2.6 1988-1994 7.5-7.8 CVD
* 5 ug/dL 7.5
TDuan etal etal,2020* NHANES Adults220
1.49(0.93,2.31)
Cause-specific Cardiovascular Mortality:
CVD ALAD GG
CVD ALAD CG/GG
Menke etal etal,2006
Menke etal et al, 2006
NHANES II
NHANES II
Adults 2 20
Adults s 20
Mean: 2.60
Mean: 2.61
1988-1994
1988-1994
12
12
Ml
Stroke
TLanphear etal etal, 2018 NHANES III Adults s 20
Geometric Mean: 2.71
Geometric SE: 1.33
0.90 1.00 1.10 1.20
Effect Estimate (95% CI) per 1 ug/dL increase in blood Pb
ALAD = 6-aminolevulinic acid dehydratase; ALAD GG and ALAD CG/GG = variants of 6-aminolevulinic acid dehydratase;
CI = confidence interval; CVD = cardiovascular disease; IHD = Ischemic heart disease; IQR = Interquartile range Ml = myocardial
infarction; NHANES = National Health and Nutrition Examination Survey; Pb = lead; T = fertile.
Note: fRed text: Studies published since the 2013 Pb ISA, Black text: Studies included in the 2013 Pb ISA.
Effect estimates are standardized to a 1 |jg/dL increase in blood Pb. If the Pb biomarker is log-transformed, effect estimates are
standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed
to be linear within the evaluated interval.
*Study estimated relative risk.
Figure 4-23 Associations between blood Pb level and cardiovascular
mortality.
In an analysis of the NHANES III cohort, Lanphear et al. (2018) reported that a 1 j^ig/dL increase
in BLLs was associated with hazard ratios (HRs) of 1.10 (95% CI: 1.05, 1.15]) for CVD mortality and
1.14 [95% CI: 1.08, 1.20]) for IHD mortality. Lanphear et al. (2018) extended the average follow-up time
of the Menke et al. (2006) analysis of the same NHANES III cohort by over 7 years (from ~12 to
-19 years), resulting in a substantial increase in observed cardiovascular deaths (766 versus 1,801).
Several other recent studies that analyzed NHANES cycles reported associations of similar magnitude for
CVD mortality (Duan et al.. 2020: Ruiz-Hernandez et al.. 2017: Aoki et al.. 2016: van Bemmel et al..
2011). Specifically, van Bemmel et al. (2011). Ruiz-Hernandez et al. (2017). and Cook et al. (2022)
assessed cohorts using NHANES III data with similar results. Aoki et al. (2016) evaluated the relationship
between CVD mortality and BLLs with additional control for either hemoglobin or hematocrit values
using NHANES (1999-2010) data with mortality follow-up through 2011. A 10-fold increase in BLLs
was associated with an RR of 1.26 (95% CI: 0.91, 1.78). However, when BLLs were hemoglobin-
corrected (see details below), there was a greater increase in magnitude and precision in predicting CVD
mortality, compared with the association with whole blood Pb alone (RR: 1.46 [95% CI: 1.06, 2.01]).
Similar results were obtained when evaluating hematocrit-corrected whole BLLs and CVD mortality (RR:
1.44 [95% CI: 1.05, 1.98]).
Exernal Review Draft
4-64
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Duan etal. (2020) evaluated the relationship between CVD mortality and BLLs using NHANES
(1999-2014) with mortality follow-up through 2015. Here, a 1 (ig/dL increase in BLLs was associated
with an RR of 1.39 (95% CI: 1.28, 1.51). Both Aoki et al. (2016) and Obeng-Gvasi et al. (2021) also
relied on more recent NHANES blood Pb data, 1999-2010 and 1999-2008, respectively. Because of the
phaseout of leaded paint and gas, these more recent NHANES studies will capture populations potentially
less affected by the earlier period of elevated Pb exposure. . It is expected, however, that these
populations would still have had a substantial period of elevated BLLs in early life due to the gradual
decline in BLLs over time.
A number of studies have additionally evaluated either hemoglobin- or hematocrit-corrected
BLLs and mortality. Aoki et al. (2016) used six 2-year NHANES cycles (1999-2010) linked with
mortality data through the end of 2011 (median follow-up time: 6.2 years) to evaluate the association
between whole BLLs, hematocrit-corrected and hemoglobin-corrected BLLs, and CVD mortality among
subjects >40 years of age at baseline. Hematocrit- or hemoglobin-corrected whole blood Pb was
calculated by dividing whole blood Pb by either hematocrit or hemoglobin, respectively. To make the
results more comparable with whole blood Pb, the values were multiplied by the weighted arithmetic
mean of either hematocrit or hemoglobin . In models assessing whole BLLs (not corrected for either
hematocrit or hemoglobin), every 10-fold increase in whole BLLs was associated with an RR of 1.26
(95% CI: 0.91, 1.78) for cardiovascular mortality. Results were similar when controlling for hematocrit
(RR: 1.35 [95% CI: 0.98, 1.86]) or hemoglobin (RR: 1.35 [95% CI: 0.98, 1.87]) as a covariate in the
model. However, the association was stronger in terms of both magnitude and precision when evaluating
hematocrit-corrected (RR: 1.44 [95% CI: 1.05, 1.98]) or hemoglobin-corrected (RR: 1.46 [95% CI: 1.06,
2.01]) whole BLLs. Another study , also examined BLLs corrected for hemoglobin. Lin et al. (2011)
examined Taiwanese adults with end-stage renal disease with relatively high (mean: 11.5 (ig/dL) BLLs.
To correct for hemoglobin, the authors used the following equations, for males: BLL x 14/hemoglobin
concentration; for females: BLL x 12/hemoglobin concentration. The hemoglobin-corrected blood Pb
results were similar in magnitude (HR: 7.35 [95% CI: 1.64, 33.33]) than the noncorrected blood Pb values
(HR: 9.71 [95% CI: 2.11, 23.26]) when comparing the highest tertile (>12.64 (ig/dL) with the lowest
tertile (<8.51 (ig/dL).
In a recent study, Hollingsworth and Rudik (2021) implemented a quasi-experimental design to
examine the effect of the phase out of leaded gasoline in automotive racing on mortality rates in older
adults. Comparing time periods prior to and after the phaseout of leaded gasoline in professional racing
series (i.e., the National Association for Stock Car Auto Racing [NASCAR] and the Automobile Racing
Club of America [ARCA]), the authors used a difference-in-differences technique to estimate county-
level changes in air Pb concentrations, elevated BLL prevalence among children, and mortality rates in
race counties and counties bordering race counties relative to control counties. A detailed discussion of
results for air Pb concentrations and BLLs is presented in Appendix 2. Section 2.4.1. In short, there were
substantial declines in both air Pb concentrations and the prevalence of children with elevated BLLs
associated with the phaseout of leaded gasoline. The authors also reported significant declines in
Exernal Review Draft
4-65
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
24
25
26
27
28
29
30
31
32
33
34
cardiovascular mortality rates over this same period. Specifically, in the period following de-leading of
gasoline, there was an estimated decline in annual age-standardized cardiovascular mortality rates of 37
deaths per 100,000 in race counties and 12 deaths per 100,000 in border counties. Additionally, there was
a similar decline for IHD-related deaths, with 53 deaths per 100,000 in race counties and 20 deaths per
100,000 in border counties. Similar to the exposure results, the mortality estimates appear to demonstrate
a distance gradient. The difference-in-difference approach controls for spatially varying confounders by
estimating the difference in mortality rates in adjoining years in the same county and controls for
temporally varying confounders by taking the difference of those differences between locations. The
authors additionally adjust for potential confounders that may vary spatially and temporally (e.g.,
unemployment rate and quantity of Toxic Release Inventory [TRI] lead emissions). Hollingsw orth and
Rudik (2021) did not adjust for potential copollutant exposures, but provides evidence that there is no
differential effect of leaded and unleaded races on other copollutant concentrations (i.e., CO, VOCs,
PMio, PM2.5, NO2, and O3) in the weeks leading up to and following the race. However, because the
mortality rates are an annual measure, there is remaining uncertainty regarding potential differential
trends in the long-term average of other pollutants that could be correlated with the phaseout of leaded
gasoline in NASCAR and ARCA.
Additionally, a recent meta-analysis (Chowdhurv et al.. 2018) evaluated several metal
biomarkers, including BLLs, and evaluated the overall association between BLLs and CHD, specifically
including studies of CHD mortality. It included studies with cohort, case-control, or nested-case-control
study designs. As described in Section 4.4, this analysis provided evidence of an increased risk of CHD
mortality associated with increasing BLLs (Figure 4-20).
4.10.1.1 Dose-Response Relationship
An examination of the dose-response relationship between biomarkers of Pb exposure and
cardiovascular mortality helps to further evaluate the continuum of effect between biomarkers of Pb
exposure and cardiovascular outcomes. Because of differences in exposure historically, it is expected that
adult BLLs would be influenced by historical Pb exposures. Therefore, studies examining a single blood
Pb measurement in adulthood may not fully capture the true effect of biomarkers of Pb exposure and
cardiovascular mortality.
Several recent studies, however, have summarized mortality outcomes over a range of blood Pb
values. Lanphear et al. (2018) extended the follow-up period of the Menke et al. (2006) study and
evaluated the dose-response relationship among the same population. Using a five-knot restricted cubic
spline analysis, this study generally indicated a supralinear dose-response relationship between BLLs and
CVD and IHD mortality. The authors also stated that this dose-response relationship was steeper (HRs
were larger in magnitude) at lower blood Pb concentrations. Overall, this study reported increased risk of
CVD or IHD mortality among those with BLLs <5 (ig/dL (Figure 4-24).
Exernal Review Draft
4-66
DRAFT: Do not cite or quote
-------
Concentration of lead in blood (ng/dL)
CI = confidence interval.
Note: Restricted cubic spline (5 knots) (red lines) and adjusted HRs (black lines) with 95% CIs (hatched lines) for (B) cardiovascular
disease mortality and (C) IHD mortality.
Source: Adapted from Lanphear et al. (2018).
Figure 4-24 Dose-response relationship between blood Pb levels and
cardiovascular and ischemic heart disease mortality.
1 These results are similar to other assessments of the dose-response relationship described in
2 previous assessments of Pb, including the 2006 Pb AQCD (U.S. EPA. 2006b) along with the 2013 Pb ISA
3 (U.S. EPA. 2013a). In the original evaluation of the NHANES III data and mortality, Menke et al. (2006)
4 noted a similar linear shape of the dose-response curve. Specifically, the dose-response relationship was
5 steeper at lower blood Pb concentrations.
6 A similar NHANES III (1988-1994) analysis evaluated total CVD, heart disease (CVD diagnosis
7 codes excluding stroke), and MI deaths through 2010 (Cook et al.. 2022). This study indicated a greater
8 risk of CVD mortality among those with the highest BLLs (>6.23 (ig/dl for men and >3.74 (ig/dl for
Exernal Review Draft
4-67
DRAFT: Do not cite or quote
-------
1 women) (Figure 4-25). Similar patterns were reported for heart disease and MI mortality. The results of
2 this analysis provide no evidence of a threshold below which an association between blood Pb and
3 mortality does not exist, at least within the blood Pb ranges within this study (10th percentile: 1.0 (ig/dl,
4 90th percentile: 6.7 (ig/dl). A sensitivity analysis evaluated BLLs continuously. In the model, a 1-unit
5 increase in log-transformed BLLs was associated with an 8% (HR: 1.08 [95% CI: 1.00, 1.16]) increase in
6 total CVD mortality risk and a 9% (HR: 1.09 [95% 1.02, 1.16]) increase in heart disease risk. However,
7 there was no reported increased risk of acute MI mortality associated with a 1-unit increase in log-
8 transformed BLLs (HR: 0.95 [95% CI: 0.84, 1.08]).
CIF = cumulative incidence function; CVD = cardiovascular disease; NHANES = National Health and Nutrition Examination Survey.
Source: Cook et al. (2022).
Figure 4-25 Cumulative incidence function of cardiovascular mortality by
blood Pb level, National Health and Nutrition Examination Survey
III (1988-1994).
4.10.1.2 At-Risk Populations
9 Several recent analyses of biomarkers of Pb exposure and cardiovascular mortality have
10 evaluated EMM or stratification of the relationship by specific parameters such as sex, genetic factors,
Exernal Review Draft
4-68
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
24
25
26
27
28
29
30
31
32
33
stress, and behavior factors like smoking, whereas other analyses have primarily focused on populations
or lifestages that may be particularly vulnerable to premature mortality associated with biomarkers of Pb
exposure. The differences observed in at-risk populations are described below.
4.10.1.2.1 Stratification
Ruiz-Hernandez et al. (2017) used NHANES III (1988-1994) and three 2-year NHANES cycles
(1999-2004) linked with mortality data—through the end of 1996 for the NHANES III cohort and
through the end of 2006 for the NHANES 1999-2004 cohort data—to assess the relationship between
BLLs and CVD and CHD mortality. This study showed that in fully adjusted models, there were
increases in both CVD mortality (relative risk [RR]: 1.19 [95% CI: 1.07, 1.31]) and CHD mortality (RR:
1.24 [95% CI: 1.10, 1.41]) for each doubling of BLLs. The RRs for both CVD and CHD mortality were
stronger among women compared with men and among never-smokers compared with ever-smokers.
Despite a higher mean BLL in the NHANES 111(1988-1994) cohort, the RR was 1.17 (95% CI: 1.06,
1.29) compared with 1.43 (95% CI: 1.16, 1.78) in 1999 to 2004.
van Bemmel et al. (2011) investigated a smaller subset of NHANES III (1988-1994) subjects.
This study specifically evaluated EMM of the relationship between BLLs and cardiovascular mortality by
polymorphisms in ALAD. A critical mechanism of Pb toxicity is its ability to interact and inhibit key
enzymes, such as ALAD, in the heme biosynthetic pathway. This analysis identified a null association
between elevated BLLs (>5|_ig/dL) and cardiovascular mortality. When further stratified by ALAD
variant, this study continued to observe null associations of elevated BLLs (>5|_ig/dL) among both
ALADGG variants for cardiovascular mortality (HR: 1.01 [95% CI: 0.92, 1.10]) and among ALADCG/CC
variants (HR: 1.13 [95% CI: 0.93, 1.36]).
In a more recent analysis, (Obcng-Gvasi et al.. 2021) evaluated whether the association between
BLLs and CVD mortality was modified by AL, a measure of cumulative stress. This study used
NHANES (1999-2008) data linked to mortality data through 2014. First, the study indicated that higher
BLLs were associated with a higher AL index. There was also an increased risk of CVD mortality among
those with BLLs >1.55 (ig/dL (median) (HR: 2.35 [95% CI: 1.77, 2.93]), when compared with those
below the median BLL. This study also indicated that the interaction between BLLs and AL was
significant (p = 0.014) but did not present stratified results.
Evidence of EMM is in direct contrast to stratified analyses presented in the previous Pb ISA.
Menke etal. (2006) demonstrated that there were no interactions between BLLs and other adjusted
variables, when comparing the 80th percentile (4.92 (ig/dL) with the 20th percentile (1.46 (ig/dL) of
BLLs. Specifically, the association between BLLs and cardiovascular mortality was positive but not
different when stratified by age, race, sex, urban/rural residence, smoking, BMI, and comorbid conditions
(hypertension, diabetes, low kidney function).
Exernal Review Draft
4-69
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
24
25
26
27
28
29
30
31
32
33
4.10.1.2.2 Specific Populations
Several recent analyses have focused on the analysis of biomarkers of Pb exposure and
cardiovascular mortality among certain populations with comorbid conditions or at specific lifestages. Lin
et al. (2011) evaluated the relationship between BLLs and mortality among patients on maintenance
hemodialysis in a relatively short (-18 months of follow-up) prospective cohort in Taiwan. Study subjects
had a relatively high average BLL (mean: 11.5 (ig/dL), which is higher than the general Taiwanese
population (mean: 7.7 (ig/dL). It is suspected that hemodialysis patients may experience higher BLLs
because their kidneys may no longer be able to excrete Pb from the body due to a total loss of renal
function (see Appendix 5: Renal Effects). There was an increased HRamong those in the third tertile of
BLLs (>12.64 (ig/dL) for cardiovascular mortality (HR: 9.71 [2.11, 23.26]), compared with the first tertile
of BLLs (<8.51 (ig/dL). Additionally, when considering hemoglobin-corrected blood Pb values, the
association between the highest and lowest tertile was smaller in magnitude (HR: 4.98 [95% CI: 1.86,
13.33]) compared with whole blood Pb measurements, but was still imprecise (both measurements had
large confidence intervals).
4.10.2 Summary of Cardiovascular Mortality
The CVD mortality results in this review supported and expanded on findings from both the 2006
Pb AQCD, which included NHANES mortality studies (Schober et al.. 2006: Lustbcrg and Silbcrgcld.
2002). and the 2013 Pb ISA, which included an NHANES mortality study (Menke et al.. 2006) and non-
NHANES cohort analyses (Khalil. 2010: Khalil et al.. 2009: Weisskopf et al.. 2009). Several of the most
recent NHANES analyses (Duan et al.. 2020: Lanphear et al.. 2018: Ruiz-Hernandez et al.. 2017: Aoki et
al.. 2016) further strengthen the evidence provided within the 2013 Pb ISA, by including a wide range of
potential confounders and further consideration of a dose-response relationship. Furthermore, the most
recent NHANES analyses provide evidence of an association between BLLs and mortality at lower mean
population blood Pb concentrations (mean or median blood Pb range between 1.49 and 3.2 (ig/dL) (Duan
et al.. 2020: Lanphear et al.. 2018: Ruiz-Hernandez et al.. 2017: Aoki et al.. 2016). Despite the differences
observed within the studies, associations between increased concentrations of Pb biomarkers and
mortality were generally observed (Figure 4-23, Table 4-16).
There still remains uncertainty regarding the relative contributions of recent, past, and cumulative
Pb exposure for the relationship between BLLs and cardiovascular mortality. The more recent NHANES
analyses evaluate cycles as recent as 2015 and continue to observe strong associations between
increasingly lower levels of blood Pb and cardiovascular mortality; however, these analyses still contain
populations greatly influenced by high historic Pb exposure. Additionally, further confounder control,
such as the inclusion of Cd concentrations in blood or urine, can also reduce the uncertainty noted in the
2006 Pb AQCD. van Bemmel et al. (2011) reported null associations between BLLs and cardiovascular
mortality. However, despite using data from NHANES III, the authors were not able to sufficiently
Exernal Review Draft
4-70
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
account for all confounders, and were limited to a smaller sample size, given their study hypothesis. The
cohort of Taiwanese hemodialysis patients provided evidence that there may be subsets of the population
at an increased risk of Pb-related cardiovascular mortality, compared with the general population . Taken
together, despite differences in the design, methods, and considerations across studies, associations
between elevated levels of Pb biomarkers and increased mortality risk were generally observed.
4.11 Biological Plausibility
Sections 4.1 to 4.10 of this appendix describe the cardiovascular health effects associated with
exposure to Pb from epidemiologic and animal toxicological studies. Informed largely by the animal
toxicological evidence presented in these sections, as well as in previous ISAs and AQCDs, this section
describes the biological pathways that potentially underlie the cardiovascular associations observed in
epidemiologic studies. Figure 4-26 graphically depicts these proposed pathways as a continuum of
pathophysiological responses—connected by arrows—that may ultimately lead to the apical
cardiovascular events associated with exposures to Pb at concentrations observed in epidemiologic studies
(e.g., IHD, MI). Note that the role of biological plausibility in contributing to the weight-of-evidence
causality determinations reached in the current Pb ISA are discussed in Section 4.12.
Exernal Review Draft
4-71
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
Pb Exposure |
Pro-
Atherosclerotic
Environment
(e.g., increased
inflammation,
clotting factors,
cholesterol)
Exacerbation
of Ischemic
Heart Disease/
Potential
Myocardial
infarction or
Stroke
Cardiovascular
¦H Related
Mortality
^
of
the
¦
1
Nervous jUWWUWWMS
System
HR = heart rate; HRV = heart rate variability; Pb = lead.
Note: The boxes above represent the effects for which there is experimental or epidemiologic evidence related to Pb exposure, and
the arrows indicate a proposed relationship between those effects. Shading around multiple boxes is used to denote a grouping of
these effects. Arrows may connect individual boxes, groupings of boxes, and individual boxes within groupings of boxes.
Progression of effects is generally depicted from left to right and color-coded (gray, exposure; green, initial effect; blue, intermediate
effect; orange, effect at the population level or a key clinical effect). Here, population-level effects generally reflect results of
epidemiologic studies. The structure of the biological plausibility sections and the role of biological plausibility in contributing to the
weight-of-evidence analysis used in this Pb ISA are discussed in Section 4.12.
Figure 4-26 Potential biological pathways for cardiovascular effects following
exposure to Pb.
Considering the available health evidence, Figure 4-26 shows plausible pathways connecting Pb
exposure to the apical events reported in epidemiologic studies. The first potential pathway begins with
oxidative stress leading to impaired vascular function, systemic inflammation, a pro-atherosclerotic
environment, and increases in BP. The second potential pathway involves Pb perturbation of the RAAS
leading to increases in BP and impaired vascular function. The third potential pathway involves
modulation of the ANS leading to increases in BP and exacerbation of conduction abnormalities and
arrythmia. Once these pathways are initiated, there is evidence from in vitro and in vivo toxicological
studies that exposure to Pb may result in a series of pathophysiological responses that could lead to
cardiovascular events such as IHD, MI, and stroke, and thus, possible cardiovascular mortality.
As noted above, one potential pathway for Pb exposure to result in the associations reported in
epidemiologic studies is through the induction of oxidative stress and inflammation. Exposure to Pb can
Exernal Review Draft
4-72
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
stimulate the production of ROSs in the blood, heart, and/or vasculature (SimSes et al.. 2015; Dewaniee et
al.. 2013; Farmand et al.. 2005; Ni et al.. 2004; Attri et al.. 2003; Courtois et al.. 2003; Vaziri et al.. 1999;
Gonicketal.. 1997) For example, Pb exposure in rats resulted in increased levels of superoxides and
hydrogen peroxide in human coronary endothelial cells. Similarly, Vaziri et al. (1999) demonstrated
increased plasma and cardiac levels of the oxidative stress marker 3-nitrotyrosine following Pb exposure
in rats.
Pb exposure resulting in the production of ROSs is important because several studies have
demonstrated a role for Pb-induced oxidative stress in impaired vascular function (SimSes et al.. 2015;
Dursun et al.. 2005; Attri et al.. 2003; Gonicketal.. 1997; Vaziri et al.. 1997; Khalil-Manesh et al.. 1994;
Khalil-Manesh et al.. 1993). Impaired vascular function is often the result of impaired functioning of the
endothelium, which maintains the normal balance of mediators that promote vasorelaxation (e.g., nitric
oxide) and vasoconstriction (e.g., endothelin-1). Animal toxicological studies demonstrate that exposure
to Pb results in altered vascular function (Nunes et al.. 2015; Gaspar and Cordellini. 2014; Silveira et al..
2010; Zhang et al.. 2007; Skoczvnska and Stoiek. 2005). including impairment that would be consistent
with greater vasoconstriction (SimSes et al.. 2015; Gaspar and Cordellini. 2014) or decreased vasodilation
(Silveira etal.. 2010; Zhang et al.. 2007). Toxicological studies have also demonstrated that Pb-induced
oxidative stress results in impaired vascular function through the inactivation or downregulation of
vasodilators such as nitric oxide and/or soluble guanylate cyclase, thereby increasing the potential for
vasoconstriction (Goncalves-Rizzi et al.. 2016; Dursun et al.. 2005; Attri et al.. 2003; Gonicketal.. 1997;
Vaziri et al.. 1997; Khalil-Manesh et al.. 1994; Khalil-Manesh et al.. 1993). For example, Pb-induced
ROSs can inactivate or sequester the vasodilator nitric oxide (Malvezzi et al.. 2001; Vaziri et al.. 1999).
and inhibition of nicotinamide adenine dinucleotide phosphate oxidase was able to block Pb-enhanced
contraction of cultured rat aorta cells in response to the vasoconstrictor 5 hydroxytryptamine (Zhang et
al.. 2005).
Continuing along this potential pathway, impaired vascular function also promotes plaque
formation potentially leading to atherosclerosis. In addition to its hallmark feature of impaired
vasodilation, impaired vascular function is further characterized by decreased vascular integrity, increased
expression of adhesion molecules, and cytokine upregulation (Lind et al.. 2021). In total, this increases
the potential for atherosclerotic disease and formation of thrombi (i.e., blood clots). Following Pb
exposure, toxicological studies have demonstrated that Pb induces markers of systemic inflammation in
blood (Fernandez-Cabezudo et al.. 2007; Iavicoli et al.. 2006; Chen et al.. 2004; Dvatlov and Lawrence.
2002; Miller etal.. 1998; Heo et al.. 1997; Heo et al.. 1996). as well as increases in C-reactive protein in
cardiac tissue (Roshan et al.. 2011). In addition, Pb was also found to induce interleukin (IL)-8, which
mediated vessel intima hyperplasia in human endothelial cells (Zeller et al.. 2010). Similarly, Pb exposure
in rats increased aortic media thickness, media-lumen ratio, and medial collagen content (Zhang etal..
2009). Exposure to Pb also increased coagulation and promoted thrombus formation (Shin et al.. 2007). In
agreement with these toxicological studies demonstrating Pb-induced inflammation, an epidemiologic
study found that higher BLLs in children were correlated with higher serum levels of IL-4 (Lutz et al..
Exernal Review Draft
4-73
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
1999). which can stimulate the liver to produce additional coagulation factors and further the pro-
atherosclerotic environment. Moreover, as discussed in the metabolic effects appendix (Section 9.2),
exposure to Pb can also result in the upregulation of cholesterol, another key contributor to developing
atherosclerosis. Taken together, evidence of a pro-atherosclerotic environment is important given that
atherosclerosis can lead to plaque and thrombosis (i.e., clot) formation. If dislodged, those plaques could
obstruct blood flow to the heart or stimulate intravascular clotting (Karolv et al.. 2007). both of which
could result in acute myocardial ischemia. If the dislodged plaque obstructs blood flow to the brain, there
is potential for stroke.
Impaired vascular function (e.g., resulting from Pb-induced oxidative stress) can also lead to
increases in BP through vasoconstriction. Increases in BP may then exacerbate IHD or heart failure
through conduction abnormalities or arrythmias, and further impair vascular function. For example, in
patients with high BP, changes in arterial shear stress due to changes in blood flow (i.e., laminar versus
turbulent) are associated with impaired vascular function (Khderetal.. 1998). which, as noted above,
could lead to a worsening of IHD or heart failure. Importantly, there are numerous studies demonstrating
that Pb can increase measures of BP: (Nunes et al.. 2015; Xu et al.. 2015; Fioresi et al.. 2014; Mohammad
et al.. 2010; Zhang et al.. 2009; Badavi et al.. 2008; Grizzo and Cordellini. 2008: Reza et al.. 2008; Bravo
et al.. 2007; Robles et al.. 2007; Hevdari et al.. 2006; Bagchi and Preuss. 2005; Nakhoul et al.. 1992). In
addition, a toxicological study has demonstrated that Pb exposure can result in conduction abnormalities
and potential arrythmia (Reza et al.. 2008).
The second pathway by which Pb exposure may result in the cardiovascular associations reported
in epidemiologic studies is through activation of the RAAS, which is responsible for fluid homeostasis
and BP regulation. Exposure of experimental animals to Pb increases ACE activity; plasma kininase II;
kininase I; and kallikrein activities in plasma, aorta, heart, and kidney, as well as renal angiotensin II
positive cells (Rodrigucz-lturbc et al.. 2005; Sharifi et al.. 2004; Carmignani etal.. 1999). These changes
can result in increases in BP, which as noted above, could lead to worsening of IHD or heart failure
potentially through conduction abnormalities, arrythmia, and/or impaired vascular function. Additional
information on the effect of Pb on the RAAS system is discussed in the renal effects appendix. This
summary includes a discussion of Pb accumulation in the kidney resulting in cellular damage, thereby
increasing the potential for RAAS disfunction (see Appendix 5).
The third pathway by which Pb exposure may result in the cardiovascular associations reported in
epidemiologic studies is through modulation of the ANS. As noted in the nervous system appendix, Pb
can deposit in the brain where it causes cellular damage and altered neurological function (see
Appendix 3). Similarly, it has also been shown that exposure to Pb can modulate autonomic tone (e.g.,
increased sympathetic tone) to the heart and vasculature, possibly through stimulation of the P2X4 and
P2X7 receptors in satellite glial cells (Zhu et al.. 2019; Zhu et al.. 2018). Shifts toward increased
sympathetic nervous system tone may result in increases in heart rate and BP as well as decreases in
vascular function, which as mentioned above, could exacerbate IHD and/or heart failure. It is therefore
Exernal Review Draft
4-74
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
24
25
26
27
28
29
30
important to note evidence from animal toxicological studies for increases in heart rate (Simocs et al..
2011; Badavi et al.. 2008; Lai et al.. 2002) and changes in HRV consistent with a shift toward increased
sympathetic tone (Zhu et al.. 2019; Shvachiv et al.. 2018; Zhu et al.. 2018; Geraldes et al.. 2016)
following Pb exposure. Similarly, evidence from an animal toxicological study suggests that Pb exposure
can result in conduction abnormalities or arrhythmia (Rcza et al.. 2008). Conduction abnormalities or
arrhythmia could exacerbate IHD and/or HF. Taken together, there are potential pathways by which ANS
modulation may lead to worsening of IHD or HF, thereby increasing the risk for mortality.
When considering the available evidence presented throughout this appendix, there are plausible
pathways connecting Pb exposure to the cardiovascular associations reported in epidemiologic studies
(Figure 4-26). Thus, these proposed pathways provide biological plausibility for the associations reported
in epidemiologic studies between Pb and IHD, MI, stroke, and therefore, mortality.
4.12 Summary and Causality Determination
A large body of health evidence published since the 2013 Pb ISA continues to demonstrate a
causal relationship between exposure to Pb and cardiovascular health effects. The 2013 Pb ISA concluded
that the evidence supported a causal relationship between exposure to Pb and hypertension and increased
BP in adults, as well as between Pb exposure and CHD (based largely on epidemiologic studies of CVD-
related mortality). For other cardiovascular-related outcomes, the evidence was suggestive of but not
sufficient to infer a causal relationship for subclinical atherosclerosis and inadequate to infer the presence
or absence of a relationship with cerebrovascular disease (Table 4-2). More recent studies greatly expand
the evidence base discussed in the 2013 Pb ISA and serve to strengthen the support for relationships
between exposure to Pb and a number of cardiovascular-related health effects. In particular, there is
substantially more evidence of hypertension, increases in BP, and cardiovascular-related mortality
following exposure to Pb. Moreover, there is additional health evidence for effects such as changes in
cardiac electrophysiology (e.g., ECG measures of cardiac depolarization, repolarization, and HRV),
arrythmia, and markers of atherosclerosis. Thus, in the current ISA, the evidence supports a causal
relationship between exposure to Pb and cardiovascular effects} After a brief discussion of the health
evidence and key uncertainties found in the 2013 Pb ISA, the rest of this summary and causal
determination section discusses the health evidence and rationale for the causal determination reached in
this Pb ISA. This discussion will rely upon the framework for causality determinations described in the
preamble to the IS As (U.S. EPA. 2015). Key health evidence supporting this determination is also
summarized in Table 4-2.
1 The current ISA follows the approach of more recent IS As, including the 2019 Particulate Matter and 2021 Ozone
ISAs, in making a single causality determination for cardiovascular effects. Additional information regarding this
decision can be found in Section 4.1 of this appendix.
Exernal Review Draft
4-75
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
In the 2013 Pb ISA, the strongest evidence for an effect of Pb on cardiovascular outcomes was
on BP and CVD-related mortality. Prospective epidemiologic studies clearly supported the relationship
between biomarkers of Pb exposure and hypertension incidence and changes in BP (U.S. EPA. 2013a).
The prospective evidence was supported by meta-analyses that underscored the consistency and
reproducibility of Pb-associated increases in BP and hypertension and epidemiologic studies that adjusted
for a wide range of potential confounders to reduce uncertainty due to potential unmeasured confounding
(U.S. EPA. 2013a). With respect to cardiovascular-related mortality, the previous Pb ISA (U.S. EPA.
2013a) described longitudinal studies in adult cohorts in a number of locations reporting that biomarkers
of Pb exposure were associated with risk of mortality from MI, IHD, or CHD, with the strongest of these
associations being with MI mortality. In addition, epidemiologic studies reviewed in the 2013 Pb ISA
included some evidence of a positive association between exposure to Pb and changes in cardio
electrophysiology (e.g., changes in HRV and QT interval) and atherosclerotic plaque formation. Key
uncertainties noted with respect to the epidemiologic evidence from the last review included inconsistent
evidence for BP changes in children and uncertainty in the level, timing, frequency, and duration of Pb
exposure contributing to the reported cardiovascular effects in adults. That is, given the appreciable
history of exposure in decades past (see Appendix 2. Section 2.4.1), and that Pb accumulates in the body
over a lifetime, the extent to which past Pb exposures contribute to the BLLs and positive associations
reported in epidemiologic studies remains uncertain.
In the 2013 Pb ISA, animal toxicological studies provided supporting evidence and biological
plausibility for the associations observed in epidemiologic studies, particularly with respect to BLLs and
changes in BP and/or hypertension. Increases in BP following exposure to Pb were generally reported in
animal toxicological studies. The previous ISA further noted toxicological studies indicating the
production of oxidative stress species that could inactivate the vasodilator, nitric oxide, which could
potentially lead to increased vasoconstriction, and thus, increases in BP. Animal toxicological studies
discussed in the last review also provided at least some evidence that exposure to Pb may contribute to a
pro-atherosclerotic environment and result in changes in HRV.
More recent studies greatly expand the evidence base from the 2013 Pb ISA and serve to
strengthen support for the relationship between exposure to Pb and cardiovascular effects in adults. In
particular, the strongest evidence continues to be Pb's effect on increases in BP. Numerous additional
epidemiologic studies published since the last review report positive associations between measures of Pb
in the body and increases in BP. Nationally representative cross-sectional studies in countries including
the United States and Canada reported positive associations between increases in BLLs and changes in
SBP, DBP, or both (Huang. 2022; Qu et al.. 2022; Everson et al.. 2021; Teve et al.. 2020; Obeng-Gvasi et
al.. 2018; Lee et al.. 2016a; Hara et al.. 2015; Bushnik et al.. 2014; Hicken et al.. 2013; Zota et al.. 2013;
Scinicariello et al.. 2011). These nationally representative studies of adult cohorts (with most participants
born before 1970, some in the 1930s) generally reported positive increases in BP (mmHg) with mean
BLLs -1.5-3 (ig/dL. Consistent with these nationally representative studies, smaller cross-sectional
studies generally reported positive associations between measures of Pb burden and changes in BP (Yan
Exernal Review Draft
4-76
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
et al.. 2022; Chung et al.. 2020; Wang et al.. 2020; Guo et al.. 2019; Gambelunghe et al.. 2016; Ettinger et
al.. 2014) within a slightly larger range of mean BLLs within each study (~1.5- 8.5 (ig/dL). While not all
studies reported positive associations with SBP or DBP [e.g., (Yu et al.. 2020; Ettinger et al.. 2014)1. the
generally positive cross-sectional results were consistent with a longitudinal analysis in a small
Bangladeshi cohort. This study indicated there was an annual increase in SBP associated with the largest
quartile of baseline BLLs compared with the lowest quartile Bulka et al. (2019). The majority of recent
analyses consider a wide range of confounders including demographics, comorbid conditions,
antihypertensive medication use, and other co-exposures to metals such as Cd. In addition, an extensive
amount of literature also considered effect measure modifiers, including, sex, age, and race, among
others. Combined with epidemiologic results from the previous ISA and AQCDs, there is clear and
substantial evidence that increasing body Pb levels is associated with increases in measures of BP.
However, uncertainty remains regarding the role of extensive historical exposure (magnitude, duration,
timing) of these cohorts.
The epidemiologic associations summarized above are coherent with animal toxicological studies
published since the 2013 Pb ISA that examined BP. BLLs in these animal studies were <30 (ig/dL, and
most of these studies reported that in animals exposed to Pb, there were increases in BP when compared
with control treated animals (Zhu et al.. 2019; Shvachiv et al.. 2018; Zhu et al.. 2018; Nunes et al.. 2015;
Silvaetal.. 2015; Xu et al.. 2015; Fioresi et al.. 2014; Gaspar and Cordellini. 2014). It should be noted
that although these studies found some measure of BP at some time point to be increased following
exposure to Pb, there was variability among studies with respect to which measure of BP increased (e.g.,
SBP or DBP) and the timing of those increases. Moreover, there was a single animal toxicological study
in rats that reported no changes in measures of BP following a Pb drinking water exposure (Wildemann et
al.. 2015). Nonetheless, when considered in total, these animal toxicological studies provide clear
evidence for exposure to Pb resulting in increases in measures of BP. These animal toxicological studies
are coherent with, and provide support for, the mostly positive associations reported in epidemiologic
studies between body Pb levels and BP increases.
As noted above, a number of prospective cohort studies evaluated in the 2013 Pb ISA (U.S. EPA.
2013a) and in the 2006 Pb AQCD (U.S. EPA. 2006b) indicated positive associations between Pb
biomarkers of exposure and cardiovascular mortality. Moreover, the results of these previously reviewed
studies remained positive when controlling for a wide range of potential confounders. Since the
publication of the 2013 Pb ISA, additional evidence of cardiovascular-related mortality has been reported.
In an analysis of the NHANES III cohort, a 1 (ig/dL increase in BLLs was associated with HRs of 1.10
(95% CI: 1.05, 1.15]) for CVD mortality and 1.14 [95% CI: 1.08, 1.20]) for IHD mortality (Lanphear et
al.. 2018). Consistent with these results, additional studies analyzing NHANES cycles reported
associations of similar magnitudes between BLLs and CVD-related mortality (Duan et al.. 2020; Ruiz-
Hernandez et al.. 2017; Aoki et al.. 2016; van Bemmel et al.. 2011). These more recent studies also
reported that associations between BLLs and CVD-related mortality remained positive after accounting
for risk factors such as physical activity, serum cholesterol, (Lanphear et al.. 2018; Ruiz-Hernandez et al..
Exernal Review Draft
4-77
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
24
25
26
27
28
29
30
31
32
33
34
35
36
37
2017) and Cd levels in blood or urine (Aoki et al.. 2016) (Table 4-16). In addition. Duan et al. (2020)
specifically evaluated NHANES participants enrolled in cycles between 1999 and 2014 (mortality data
included through 2015), with ~7 years of mortality follow-up. Although some members of this population
may have had lower Pb exposures due to the phaseout of leaded gasoline, especially when compared with
studies assessing adults in NHANES II (1976-1980) and NHANES III (1988-1994), the vast majority of
the participants were born well before the phaseout.
Epidemiologic studies of mortality are consistent not only with the large amount of evidence for
changes in BP and hypertension described above, but also with evidence of associations between blood or
bone Pb levels and other cardiovascular outcomes. Studies using the NAS cohort of older adult men
indicated an association between patella Pb levels and incident IHD (Ding et al.. 2019; Ding et al.. 2016).
Additionally, a series of 10-year CVD risk evaluations using KNHANES data observed increased 10-year
CVD risk with increasing BLLs (Nguyen et al.. 2021; Park and Han. 2021; Choi et al.. 2020; Cho et al..
2016). These studies are also consistent with a series of NAS analyses presented in the 2013 Pb ISA
indicating an association between bone Pb levels and a prolonged QT interval (Eum et al.. 2011; Park et
al.. 2009; Cheng et al.. 1998). Recent studies among children are less consistent than those in adults.
However, there is some evidence to support clinically relevant changes in HRV (Halabickv et al.. 2022)
and increases in SBP and TPR (Gump etal.. 2011) following an acute stressor.
Toxicological studies evaluated in the 2013 ISA demonstrated increased incidence of arrhythmia,
atrioventricular block, and a prolonged ST segment interval in Pb-exposed animals (Reza et al.. 2008).
That said, an additional toxicological study published since the last review reported no change in the PR,
QRS, or QT segments in Pb-exposed rats (Wildemann et al.. 2015). Similarly, although more limited
and/or mixed, there is at least some epidemiologic and animal toxicological evidence for changes in heart
rate and HRV (Section 4.7) and potential indicators of atherosclerosis following exposure to Pb. Notably,
a toxicological study published since the 2013 Pb ISA reported a statistically significant increase in the
diameter of the cells of the aorta, as well as changes in the shape (i.e., loss of curvature) of the aortic
internal elastic lumen in Pb-exposed rats, relative to control rats. This study also reported a statistically
significant increase in proliferating cell nuclear antigen in rat cardiac tissue in Pb-exposed rats (p < 0.05),
potentially consistent with the type of cellular proliferation that is involved in atherosclerotic plaque
growth (Xu et al.. 2015). Moreover, these results are consistent with a study discussed in the 2013 Pb ISA
demonstrating increased aortic media thickness, media-lumen ratio, and medial collagen content
following exposure to Pb (Zhang et al.. 2009).
In support of epidemiologic studies reporting positive associations between BLLs and CVD-
related mortality, animal and in vitro toxicological evidence provides plausible pathways by which
exposure to Pb could lead to serious CVD-related outcomes such as IHD, MI, and/or stroke. In brief, one
such pathway posits that exposure to Pb resulting in oxidative stress and systemic inflammation could
potentially lead to impaired vascular function, a pro-atherosclerotic environment, and increases in BP.
Importantly, there is animal toxicological evidence demonstrating all these effects following exposure to
Exernal Review Draft
4-78
DRAFT: Do not cite or quote
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Pb (Section 4.8). In addition, these effects, in particular atherosclerosis and increases in BP, can set the
stage for an MI or stroke that could result in mortality. More information on this and other potential
pathways can be found in Section 4.11, in which each potential pathway is described in detail. Several
recent epidemiologic studies have been published further supporting this association.
Taken together, the recent evidence described throughout this appendix extends the consistency
and coherence of the evidence base reported in the 2013 Pb ISA. Direct evidence for Pb exposure-related
cardiovascular effects can be found in numerous animal toxicological studies. In coherence with these
results are epidemiologic studies reporting that Pb exposure is associated with some of the same
cardiovascular endpoints reported in animal toxicological studies (e.g., increased BP, changes in cardiac
electrophysiology). While some cardiovascular outcomes are examined in relatively few studies, and
results across studies are inconsistent (see Sections 4.7-4.9), the evidence overall demonstrates a
relationship between exposure to Pb and indicators of cardiovascular disease. Animal toxicological
studies following Pb exposure provide coherence and biological plausibility for the consistent
epidemiologic associations reported between body Pb concentrations and cardiovascular outcomes such
as increased BP, hypertension, and cardiovascular mortality. However, uncertainties still remain
regarding the timing, frequency, and duration of Pb exposure levels, which contribute to cardiovascular
health effects. Yet, the collective evidence is sufficient to conclude that there is a causal relationship
between Pb exposure and cardiovascular effects.
Exernal Review Draft
4-79
DRAFT: Do not cite or quote
-------
Table 4-2 Summary of evidence indicating a causal relationship between Pb
exposure and cardiovascular effects.
Rationale for
Causality
Determination3
Key Evidence13
References'3
Pb Biomarker Levels
Associated with Effects0
Generally consistent
evidence from
epidemiologic studies
of BP in adults
Epidemiologic studies
consistently demonstrating
increases in at least some
measure of BP and Pb
biomarkers
Hara et al. (2015)
Hicken etal. (2012)
Hicken etal. (2013)
Obenq-Gvasi et al. (2018)
Scinicariello et al. (2011)
Teve et al. (2020)
Zota etal. (2013)
Everson et al. (2021)
Huang (2022)
Obenq-Gvasi (2019)
Tsoi etal. (2021)
Lee etal. (2016a)
Bushniket al. (2014)
Qu et al. (2022)
Lopes etal. (2017)
Chung et al. (2020)
Yan et al. (2022)
Gambelunqhe et al. (2016)
Bulka et al. (2019)
Mean blood Pb: -1.0 to
3 [jg/dL
Generally consistent
evidence from
epidemiologic studies
of hypertension
Epidemiologic studies
consistently demonstrating
increases in incident
hypertension risk with Pb
biomarkers
Mostly positive associations
between prevalent
hypertension and Pb
biomarkers
Gambelunqhe et al. (2016)
Zheutlin et al. (2018)
Huang (2022)
Tsoi etal. (2021)
Miao etal. (2020)
Scinicariello et al. (2011)
Lee etal. (2016b)
Lee etal. (2016a)
Choi etal. (2018)
Mean blood Pb: ~2.5-5 [jg/dL
Mean bone Pb: -20 (tibia) -
27 (patella) |jg/g
Mean blood Pb: ~1.5-
3.5 [jg/dL
Qu et al. (2022)
Lopes etal. (2017)
Generally consistent
evidence from
epidemiologic studies
of cardiovascular
mortality
Epidemiologic studies
consistently demonstrating
increases in cardiovascular
mortality risk with Pb
biomarkers
Menke et al. (2006)
Lanphear et al. (2018)
van Bemmel et al. (2011)
Cook et al. (2022)
Ruiz-Hernandez et al. (2017)
Duan etal. (2020)
Aoki etal. (2016)
Obenq-Gvasi et al. (2021)
Lin etal. (2011)
Mean blood Pb: ~1.5-
3.2 [jg/dL
Exernal Review Draft
4-80
DRAFT: Do not cite or quote
-------
Rationale for
Causality
Determination3
Key Evidence13
References'3
Pb Biomarker Levels
Associated with Effects0
Generally consistent
evidence from
epidemiologic studies
of ischemic heart
disease
Mostly positive associations
between incident IHD and Pb
biomarkers
Mostly positive associations
between estimates of 10-yr
CHD risk and Pb biomarkers
Jain et al. (2007)
Ding etal. (2016)
Ding et al. (2019)
Cho et al. (2016)
Choi et al. (2020)
Park and Han (2021)
Nguyen et al. (2021)
Mean blood Pb: -6.5 [jg/dL
Mean bone (patella) Pb:
~30 |jg/g
Mean bone (tibia) Pb: -23 |jg/
Mean blood Pb: -3 [jg/dL
Generally consistent Mostly positive associations
evidence from between left ventricle
epidemiologic studies structure/function and Pb
of cardiac function biomarkers
Yang etal. (2017)
Lind et al. (2012)
Chen etal. (2021)
Mean blood Pb: -2-5 [jg/dL
Limited evidence
from epidemiologic
studies for changes
in HRV
A single study reported a
change in HRV following a
stress response in children
Halabickv et al. (2022)
Mean blood Pb: -3-6 [jg/dL
Limited evidence
from epidemiologic
studies for
atherosclerosis
A small number of studies
demonstrated development
of atherosclerosis within
different populations
Wan etal. (2021)
Kim etal. (2021)
Qin etal. (2021)
Mean blood Pb: 1.5-3 [jg/dL
Consistent evidence
from animal
toxicological studies
of BP
Animal toxicological studies
consistently demonstrating
increases in at least some
measure of BP
Fioresi et al. (2014)
Nunes etal. (2015)
Xu et al. (2015)
Silva etal. (2015)
Shvachiv et al. (2018)
Gaspar and Cordellini (2014)
Zhu etal. (2018)
Zhu etal. (2019)
Mean blood Pb: -8-30 [jg/dL
Limited evidence
from animal
toxicological studies
for changes in HRV
A small number of studies
demonstrated changes in at
least some measure of HRV
(e.g., LF)
Shvachiv et al. (2018)
Zhu etal. (2018)
Zhu etal. (2019)
Mean blood Pb:
28 [jg/dL
-24-
Exernal Review Draft
4-81
DRAFT: Do not cite or quote
-------
Rationale for
Causality Key Evidence* References* Associated'wfth^E2sc
Determination3 Associated witn tnects
Limited but
consistent evidence
from animal
toxicological studies
for structural
changes consistent
with the development
of atherosclerosis
A single animal toxicological
study reported an increase in
the aortic media thickness,
media-lumen ratio, and
medial collagen content
following Pb exposure
A single animal toxicological
study reporting an increase in
the diameter of the cells of
the aorta, changes in the
shape of the aortic internal
elastic lumen, and an
increase in proliferating cell
nuclear antigen in rat cardiac
tissue
Zhang et al. (2009)
Mean blood Pb: -28 [jg/dL
Xu et al. (2015)
Mean blood Pb:
25 [jg/dL
-20-
Biological Plausibility A few well-defined potential Section 4.11 NA
pathways by which exposure
to Pb could reasonably result
in the health outcomes
reported in epidemiologic
studies
BP = blood pressure; CHD = coronary heart disease; HRV = heart rate variability; IHD = ischemic heart disease; LF = low
frequency; NA = not available; Pb = lead; yr = year(s).
aBased on aspects considered in judgments of causality and weight of evidence in causal framework in Table I and Table II of the
Preamble to the ISAs (U.S. EPA. 2015).
bDescribes the key evidence and references, supporting or contradicting, contributing most heavily to causality determination and,
where applicable, to uncertainties or inconsistencies. References to earlier sections indicate where the full body of evidence is
described.
°Describes the Pb biomarker levels at which the evidence is substantiated.
1
Exernal Review Draft
4-82
DRAFT: Do not cite or quote
-------
4.13 Evidence Inventories - Data Tables to Summarize Study Details
Table 4-3 Epidemiologic studies of Pb exposure and blood pressure.
Reference and Study
Study Design Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Cross-Sectional Studies
Hara et al. (2015) NHANES
n = 12,725
>20 yr
United States
NHANES 2003-
2010
Cross-sectional
Blood Pb (ICP-MS) (|jg/dL) BP (SBP, DBP, Linear models adjusted for BP change (mmHg) per doubling of
Average
individual born
-1957
GM (IQR):
See Figure 4-4
Age at measurement:
Mean (SD)
Black Women: 48.31 (6.8)
Hispanic Women: 48.1 (16.8)
White Women: 53.0 (8.4)
Black Men: 47.7 (16.9)
Hispanic Men: 46.1 (6.8)
White Men: 53.1 (18.6)
PP, MAP) ethnicity, sex, age, BMI,
heart rate, hematocrit, serum
total calcium y-
glutamyltransferase, cotinine,
dietary sodium to potassium
intake ratio, college
education, antihypertensive
drug treatment
blood Pb
See Figure 4-4b
Hicken et al.
(2012)
United States
NHANES 2005-
2008
Cross-sectional
NHANES
n =10,971
>20 yr
Average
individual born
-1963
Blood Pb (ICP-MS) (|jg/dL)
Mean (Median)
See Figure 4-6
Age at measurement
Mean (SD)
White Men: 45.6 (15.8)
Black Men: 40.6 (14.4)
White Women: 47.3 (16.7)
Black Women: 42.4 (15.1)
BP (SBP, DBP,
PP)
Linear regression adjusted
forage, BMI, heavy alcohol
use, smoking status,
diabetes diagnosis,
antihypertensive medication
use, and dietary intake of
sodium, calcium, and
potassium
Change in BP (mmHg)
See Figure 4-6b
External Review Draft
4-83
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Hicken et al.
(2013)
United States
NHANES 2005-
2008
Cross-sectional
NHANES
n = 4,470
Nonpregnant
adults (>20 yr)
Average
individual born
-1962
Blood Pb (ICP-MS) (pg/dL)
Mean (SD)
lack: 1.9 (2.2)
White: 1.7 (0.9)
Age at measurement:
Mean (SD)
Black: 42.2 (16.4)
White: 47.1 (10.8)
BP (SBP, DBP)
Linear regression adjusted
for race/ethnicity, age, sex,
high school education, family
poverty, hematocrit, BMI,
heavy alcohol use, smoking
status, and diabetes
BP (mmHg) per doubling of blood Pbb
SBP
Black 3.2 (1.5, 5.0)
High Depression 5.6 (2.0, 9.2)
Low Depression 1.8 (0.2, 3.5)
White 1.0 (-0.3, 2.4)
High Depression 1.2 (-0.5, 2.9)
Low Depression 1.0 (-0.6, 2.6)
DBP
Black 1.8 (0.7, 2.8)
white 0.9 (0.1, 1.8)
Obenq-Gvasi et NHANES
al. (2018)
United States
2007-2010
Cross-sectional
n = 12,153
>20 yr
Average
individual born
-1958
Blood Pb (ICP-MS) (pg/dL) BP (SBP, DBP)
Mean (SD)
Q1 (0-2): 1.09 (0.01)
Q2 (2-5): 2.78 (0.02)
Q3 (5-10): 6.40 (0.10)
Q4 (>10): 16.11 (1.40)
Age at measurement
Mean (SD):
Linear regression adjusted
for age, sex, race/ethnicity,
BMI, antihypertensive
medication
BP (mmHg) and In-blood Pbbc
DBP 0.268 (0.079, 0.458)
SBP 0.052 (-0.233, 0.458)
Q1
Q2
Q3
Q4
44.25 (0.32)
56.05 (0.54)
54.77 (1.13)
47.56 (2.56)
External Review Draft
4-84
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Scinicariello et al.
(2010)
United States
NHANES III
1988-1994
Cross-sectional
NHANES III
n =6,016
>17 yr
Average
individual born
-1963, -1949,
and in or
before -1931
Blood Pb (GFAAS) (|jg/dL)
Mean (SE)
Overall, 2.99 (0.09)
NH white 2.87 (0.09)
NH black 3.59 (0.20)
Mexican American 3.33 (0.11)
Age at measurement:
17-3947%
40-59 30.9%
>60 22.1%
BP (SBP, DBP)
Multivariable linear
regression adjusted for age,
sex, education, smoking
status, alcohol intake, BMI,
serum creatinine levels,
serum calcium, glycosylated
hemoglobin, and hematocrit
BP (mmHg) and blood Pb
SBP
NH white 0.707 (0.216, 1.199)
NH black 1.615 (1.007, 2.223)
Mexican American 0.471 (0.062, 0.879)
DBP
NH white -0.094 (-0.741, 0.553)
NH black 1.261 (0.716, 1.805)
Mexican American 0.414 (-0.001, 0.83)
Significant interactions with blood Pb and
ALAD genotype observed in relation to
SBP for NH white and NH black
individuals
Scinicariello et al.
(2011)
United States
NHANES 1999-
2006
Cross-sectional
NHANES
n =16,222
>20 yr with
blood Pb
<10 |jg/dL
Average
individual born
-1959
Blood Pb (ICP-MS) (ug/dL)
Mean (SE)
See Figure 4-5
Age at measurement:
Mean (SE)
White men: 47.14 (0.37)
White women: 49.64 (0.36)
Black men: 42.86 (0.37)
Black women: 45.10 (0.42)
Mexican-American men: 37.64
(0.48)
Mexican-American women:
40.67 (0.65)
BP (SBP, DBP, Multivariable logistic and
PP) linear regression models
adjusted for age, BMI, self-
reported diabetes alcohol
ingestion, smoking status,
education, serum
creatinine, serum total
calcium, sodium, hematocrit,
and blood Cd
BP (mmHg) and twofold increase in blood
Pbb
See Figure 4-5
External Review Draft
4-85
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Teve et al. (2020) NHANES
n = 30,467
United States 20-79 yr
NHANES 1999-
2016
Cross-sectional
Average
individual born
-1965
Blood Pb (ICP-MS)d (|jg/dL)
Median (IQR)
NH White
Men: 1.50 (0.99, 2.29)
Women: 1.06 (0.69, 1.60)
NH Black
Men: 1.60 (1.00, 2.60)
Women: 1.11 (0.71, 1.77)
Hispanic
Men: 1.58 (0.99, 2.43)
Women: 0.95 (0.62, 1.51)
Other race
Men: 1.54 (1.05, 2.39)
Women: 1.16 (0.75, 1.79)
BP (SBP, DBP)
Linear regression adjusted
for age/ethnicity, age,
gender, education level, BMI,
and PIR
BP (mmHg)e
SBP
NH White: 0.34 (0.11, 0.57)
NH Black: 0.67 (0.29, 1.05)
Hispanic: 0.10 (-0.01, 0.21)
Other: 0.44 (-0.51, 1.39)
DBP
NH White: 0.38 (0.19, 0.57)
NH Black: 0.36 (0.06, 0.66)
Hispanic: -0.08 (-0.21, 0.05)
Other: 0.27 (-0.15, 0.69)
Age at measurement
Mean age
NH White men: 46.37
NH White women: 47.00
NH Black men: 43.09
NH Black women: 43.28
Hispanic men: 39.67
Hispanic women: 40.51
Other men: 42.92
Other women: 43.54
External Review Draft
4-86
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Zota etal. (2013) NHANES
n =8,194
United States 40-65 yr
1999-2008
Cross-sectional
Blood Pb (ICP-MS) (|jg/dL) BP (SBP, DBP) Logistic and linear regression OR(Q5vs. Q1)
GM 1.69
Geometric SE (GSE) 0.02
Quintiles GM (GSE)
Average
individual born
-1953
Q1
Q2
Q3
Q4
Q5
0.76 (0.01)
1.25 (0.00)
1.67 (0.00)
2.25 (0.01)
3.88 (0.03)
Age at measurement
Mean: 50.9
SE: 0.15
adjusted for age, educational
attainment, race/ethnicity,
smoking, alcohol
consumption, marital status,
and antihypertensive
medication use
Elevated SBP (>140 mmHg)
All participants: 1.23 (0.92, 1.65)
LowAL: 1.14 (0.79, 1.66)
High AL: 1.40 (0.99, 1.97)
Elevated DBP (>90 mmHg)
All participants: 1.77 (1.25, 2.50)
LowAL: 1.46 (0.80, 2.68)
High AL: 2.28 (1.33, 3.91)
BP change (mmHg, Q5 vs. Q1)
SBP
All participants: 0.36 (-1.07, 2.33)
LowAL: 0.67 (-1.24, 2.58)
High AL: 1.60 (-0.62, 3.82)
DBP
All participants: 1.76 (0.75, 2.78)
LowAL: 1.72 (0.62, 2.95)
High AL: 2.01 (0.24, 3.79)
Everson et al. NHANES Blood Pb (ICP-MS) (pg/dL) BP (SBP, DBP) Linear regression models BP (mmHg)
(2021) n =2,413 Median: 1.5 adjusted for age, age2, race, SBP 0.73 (0.03, 1.44)
sex, BMI, and smoking status „ „„ , „ „„
a DBP 0.41 (-0.10, 0.92)
United States 20-59 yr Age at measurement:
Range 20-59 yr
NHANES 1999- Average
2004 individual born
-1962
Cross-sectional
External Review Draft
4-87
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Huang (2022)
United States
NHANES
n = 32,289
>20 yr
NHANES 1999- Average
2018 individual born
-1958
Cross-sectional
Blood Pb (ICP-MS) (pg/dL)
Mean (SD) 1.73 (1.71)
Age at measurement
Mean (SD) 49.68 (18.04)
BP (SBP, DBP)
Linear regression models
adjusted for age, sex, race,
education, family income
poverty ratio, BMI, alcohol
use, and smoking
BP (mmHg)
SBP
All 0.30 (0.19, 0.42)
Men
Mexican American 0.01 (-0.13, 0.34)
Other Hispanic 0.07 (-0.31, 0.45)
NH White 0.44 (0.22, 0.66)
NH Black 0.37 (0.07, 0.67)
Other Race 0.49 (-0.04, 1.03)
Women
Mexican American 0.14 (-0.28, 0.57)
Other Hispanic 0.84 (-0.15, 1.83)
NH White 0.63 (0.22, 1.04)
NH Black 0.99 (0.48, 1.50)
Other Race 0.49 (-0.35, 1.34)
DBP
All 0.23 (0.14, 0.32)
Men
Mexican American 0.08 (-0.11, 0.26)
Other Hispanic -0.20 (-0.51, 0.11)
NH White 0.40 (0.22, 0.58)
NH Black 0.26 (0.00, 0.51)
Other Race 0.05 (-0.37, 0.48)
Women
Mexican American 0.08 (-0.25, 0.40)
Other Hispanic 0.42 (-0.30, 1.14)
NH White 0.74 (0.41, 1.07)
NH Black 0.80 (0.40, 1.20)
Other Race 0.16 (-0.47, 0.79)
External Review Draft
4-88
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Obena-Gvasi
(2019)
United States
NHANES 2009-
2016
Cross-sectional
NHANES
young adults
(18-44 yr)
(n = 7,730),
middle-aged
adults (45-
65 yr)
(n = 5,744)
Average
individual born
-1981 and
-1957
Blood Pb (ICR-MS) (ijg/dL)
mean (SE):
young adults:
1.03 (0.026)
Middle-aged adults:
1.62 (0.044)
BP (SBP, DBP)
Logistic regression adjusted
for sex, BMI, income,
ethnicity, alcohol
consumption, and smoking
OR (above/below 5 |jg/dL)b
SBP >120 mmHg
Young adults: 1.21 (1.07, 1.38)
Middle-aged adults: 1.32 (1.14, 1.52)
DBP >80mmHg
Young adults: 1.32 (1.10, 1.58)
Middle-aged adults: 1.16 (0.98, 1.38)
Tsoietal. (2021) NHANES
United States
NHANES 1999-
2016
Cross-sectional
N = 39,477
adults >20
Average
individual born
-1960
Blood Pb ICP-MS (|jg/dL)
Median 1.30
Q1 <0.89
Q2 0.89-1.30
Q3 1.30-2.10
Q4 >2.10
Age at measurement:
Mean (SE)
Hypertensive
54.08 (0.23) yr
Non-hypertensive
39.87 (0.19) yr
BP (SBP) Multivariable linear
regression adjusted for age,
sex ethnicity, waist
circumference, PIR,
education, ever cigarette
smoking, diabetes, and stage
3-5 chronic kidney diseases
SBP (mmHg)
For every doubling of blood Pbb
0.52 (0.19, 0.86)
External Review Draft
4-89
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Lee etal. (2016a)
South Korea
KNHANES IV
(2008-2009), V
(2010-2012), and
VI (2013)
Cross-sectional
Korean
NHANES
n = 11,797
>19 yr
Average
individual born
in or before
-1991
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
GM (95% CI)
Male: 2.396 (2.362, 2.430)
Female: 1.919 (1.889, 1.949)
Age at measurement >19 yr
BP (SBP, DBP)
Linear models adjusted for
sex, age, residence area,
education level, smoking,
drinking status, BMI, physical
activity, serum creatinine,
and hemoglobin
BP (mmHg) doubling of blood Pbb
SBP
All 0.73 (0.09, 1.36)
Male: 0.30 (-0.53, 1.14)
Female: 1.08 (0.26, 1.90)
DBP:
All 0.71 (0.29, 1.13)
Male: 0.59 (0.01, 1.17)
Female: 0.80 (0.28, 1.33)
Bushnik et al.
(2014)
Canada
2007-2011
Cross-sectional
Canadian
Health
Measures
Survey
n = 4,550
Nonpregnant
individuals
aged 40-79
Blood Pb (ICP-MS) (ijg/dL)
Mean 1.64 (1.58-1.71)
Age at measurement:
mean: 55.4
BP (SBP, DBP)
Linear regression adjusted
for age, sex, education,
smoking, alcohol, physical
activity, BMI, non-HDL
cholesterol, diabetes, chronic
kidney disease, family history
of high BP, antihypertension
medication use
BP (mmHg)
SBP
See Figure 4-11
DBP
See Figure 4-12
Average
individual born
-1954
External Review Draft
4-90
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Qu et al. (2022)
China
2017-2018
Cross-sectional
China National
Human
Biomonitoring
Study
n = 11,037
>18 yr
Average
individual born
-1969
Blood Pb (ICP-MS) (|jg/dL)f BP change
Quartiles (SBP, DBP)
Q1
Q2
Q3
Q4
<1.59
1.59-2.24
2.24-3.21
>3.21
Age at measurement
Range 18-79
Multiple linear regression
adjusted for sex, age, BMI,
regions, education, smoking
status, alcohol consumption,
family history of
hypertension, residence
area, rice consumption, red
meat consumptions,
vegetable consumptions,
FBG, TC, HDL-C, urinary
arsenic levels, and blood Cd
levels
BP Change (mmHg)b
SBP
Q2 vs. Q1 1.36 (0.25-2.47)
Q3 vs. Q1 1.38 (-0.25-3.00)
Q4 vs. Q1 4.72 (2.70-6.74)
DBP
Q2 vs.
Q1
1.06
(0.23-1.90)
Q3 vs.
Q1
1.94
(0.79-3.09)
Q4 vs.
Q1
4.42
(3.02-5.83)
Lopes et al.
(2017)
Cambe, Brazil
2011
Cross-sectional
n = 948
Adults 40 yr
and older,
randomly
sampled from
census tracts in
the region
Average
individual born
-1956
Blood Pb (ICP-MS) (|jg/dL)
GM (95% CI):
1.97 (1.90-2.04)
10th percentile: 0.74
90th percentile: 6.03
Age at measurement:
Mean: 54.5 yr
BP (DBP, SBP)
Multiple linear regression
adjusted for age, sex, race,
income, education,
antihypertensive medication,
total cholesterol,
triglycerides, glycemia,
smoking, alcohol
consumption, and BMI
Change in BP (mmHg) (10th vs. 90th
percentile)
SBP no association (all Cis ranged
between 0 and 0)
DBP 0.005 (0.002, 0.008)
External Review Draft
4-91
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Chung et al.
(2020)
Taiwan
recruited 2010-
2011 and 2015-
2016
n = 770 Blood Pb (ICP-MS) (pg/dL)
GM (IQR)
Community Distance from EAF
residents living
<500 m: 2.41 (1.22-6.19)
near an electric v '
arc furnace 500—1000m: 2.26 (1.16—4.83)
(EAF) 1000-1500 m: 2.12 (1.05—
4.67)
Average 1500—2000 m: 2.23 (0.98—
Cross-sectional individual born 4.31)
BP (SBP, DBP) General linear models Change in BP (mmHg)b
adjusting for age, sex,
ethnicity, living near the main ggp. -| 43 34 2 52)
road and smoking
DBP: 0.69 (0.01, 1.37)
-1953
>2000m: 2.03 (1.03-4.31)
Age at measurement:
Median 60
Wang et al.
(2020)
China
Cross-sectional
n = 816
Adults 40-75,
residing in area
for >15 yr, and
subsisting on
rice and
vegetables
grown in the
polluted (Cd
concentration
>0.2 mg/kg) or
unpolluted (Cd
concentration
<0.05 mg/kg)
area
Blood (ICP-MS) (|Jg/dL)f
Median (IQR)
Polluted 3.54 (2.42-4.89)
Unpolluted 2.61 (1.70-3.84)
Age at measurement:
mean (SD)
Polluted area
Hypertensive: 60.32 (8.08)
Normotensive: 55.61 (8.52)
Unpolluted area Hypertensive:
59.92 (9.19)
Normotensive: 56.86 (9.22)
BP (SBP, DBP)
Linear regression adjusted
forage, gender, smoking
status, and BMI
BP (mmHg) and Blood Pbb
See Figure 4-2 and Figure 4-3
External Review Draft
4-92
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Zhang et al.
(2010)
Boston, MA
August 1991 and
December 2001
Cross-sectional
NAS
n = 619
Elderly men
(mostly white)
Average
individual born
-1933
Bone Pb (K-XRF) (pg/g)
Median (IQR)
Wild type HFE
Tibia: 8 (12-27)
Patella: 26 (17-37)
C282Y HFE
Tibia: 20 (14-27)
Patella: 25 (17-37)
H63D HFE
Tibia: 19 (14-26)
Patella: 27 (19-37)
Age at measurement
Mean: 67
BP (PP) Linear mixed-effects
regression models with
repeated measurements
adjusted for age, education,
alcohol intake, smoking, daily
intakes of calcium, sodium,
and potassium, total calories,
family history of
hypertension, diabetes,
height, heart rate, HDL, total
cholesterol, HDL ratio, and
waist circumference
PP (mmHg)
Tibia Pb
Wild Type HFE: 0.29 (-0.46, 1.05)
H63D HFE: 2.54 (0.12, 4.96)
C282Y HFE: 0.68 (-1.33, 2.70)
Any HFE variant: 2.23 (0.23, 4.23)
Patella Pb
Wild Type HFE: 0.14 (-0.33, 0.61)
H63D HFE: 1.53 (-0.005, 3.11)
C282Y HFE: 0.29 (-0.15, 0.73)
Any HFE variant: 1.49 (0.16, 2.82)
Weaver et al. n = 652 Blood Pb (GFAAS with
(2008) current and Zeeman correction) (pg/dL)
former Pb Mean (SD): 30.9 (16.7)
South Korea workers
Bone (Patella) Pb (K-XRF)
1999-2001 Average (MQ/g)
individual born
-1957
Cross-sectional
Mean (SD): 75.1 (101.1)
Age at measurement:
Mean: 43.3, SD: 9.8
BP (SBP) Multivariable linear
regression adjusted for age,
sex, BMI, diabetes,
antihypertensive and
analgesic medication use, Pb
job duration, tobacco, and
alcohol use
SBP (mmHg)e
Blood Pb 0.1007 (0.02, 0.18)
Patella Pb 0.059 (-0.08, 0.20)
External Review Draft
4-93
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Elmarsafawv et
al. (2006)
Boston, MA
Participants' first
visit after 1991
Cross-sectional
NAS
n = 471
Elderly men
(mostly white)
Average
individual born
-1924
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Mean (SD): 6.6 (4.3)
Bone (K-XRF) (pg/g)
Mean (SD):
Tibia: 21.6 (12.0)
Patella: 31.7 (18.3)
Age at measurement
Mean: 67
BP (SBP) Multivariable linear
regression adjusted for age,
BMI, family history of
hypertension, smoking,
dietary sodium intake, and
cumulative alcohol ingestion
No control for potential
confounding SES factors
SBP (mmHg)
Tibia
Low Ca2+ group (<800 mg/d):
4.00 (1.05, 6.95)
High Ca2+ group (>800 mg/d):
1.90 (0.10, 3.70)
Yan et al. (2022) Haitian CVD
Cohort Study
Port-au-Prince,
Haiti
2019-2021
Cross-sectional
n = 2,504
General
population >18
living in Port-
au-Prince
Average
individual born
-1980
Blood Pb (LeadCare II Blood
Level Analyzer) (pg/dL)
(Pb measurement device had
high limit of detection
(3.3 pg/dL), and only 71% of
the population had
quantifiable blood Pb values)
GM: 4.73
Geometric SE: 1.62
Age at measurement
Median: 40 yr
BP (SBP, DBP)
Multivariable linear
regression models adjusted
forage, sex, BMI, smoking
status, alcohol use, physical
activity, income, and use of
antihypertensive medication
BP (mmHg)b
SBP
Q2 vs. Q1 0.62 (-1.46, 2.70)
Q3 vs. Q1 1.73 (-0.24, 3.70)
Q4 vs. Q1 2.42(0.36, 4.49)
>3.3 pg/dL vs. <3.3 pg/dL 1.65 (0.05,
3.24)
>5 pg/dL vs. <5 pg/dL 1.16 (-0.35, 2.68)
DBP
Q2 vs. Q1 0.19 (-1.26, 1.64)
Q3 vs. Q1 1.16 (-0.25, 2.57)
Q4 vs. Q1 1.96 (0.56, 3.37)
>3.3 pg/dL vs. <3.3 pg/dL 1.16 (0.04,
2.27)
>5 pg/dL vs. <5pg/dL: 0.96 (-0.1, 2.02)
External Review Draft
4-94
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Xu et al. (2021)
United States
2011-2013
Cross-sectional
GuLF Study
Cohort
n = 957
>21 yr
Average
individual born
in or before
-1991
Blood Pb (ICP-MS) (pg/dL)
Median: 0.09
75th percentile: 0.19
Maximum: 33.8
Age at measurement: >21 yr
BP (SBP, DBP)
Multiple linear regression
adjusted for age, sex, race,
educational attainment, and
household income
BP (mmHg)
SBP
Q2 vs. Q1 1.19 (-1.73, 4.11)
Q3 vs. Q1 0.54 (-2.48, 3.55)
Q4 vs. Q1 -0.96 (-4.13, 2.22)
DBP
Q2 vs. Q1 0.21 (-1.81, 2.24)
Q3 vs. Q1 0.26 (-1.84, 2.35)
Q4 vs. Q1 -0.01 (-2.21, 2.19)
Perlstein et al.
(2007)
Boston, MA
1991-1997
Cross-sectional
NAS
n = 593
Elderly men
(mostly white)
Average
individual born
-1927
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Mean (SD)
Overall: 6.12 (4.03)
Q1
Q2
Q3
Q4
Q5
2.3 (0.8)
3.9 (0.3)
5.4 (0.5)
7.4 (0.6)
12.4 (4.4)
BP (PP) Multiple linear regression
adjusted for age, height,
race, heart rate, waist
circumference, diabetes,
family history of
hypertension, education level
achieved, smoking, alcohol
intake, fasting plasma
glucose, and ratio of total
cholesterol to HDL
cholesterol
PP (mmHg)b
Tibia Pb (above/below median): 4.2 (1.9,
6.50)
Tibia Pb (mean difference)
Q5 vs. Q1: 2.58 (-1.15, 6.48)
Blood Pb (mean difference)
Q5 vs. Q1: -1.49 (-4.93, 1.94)
Tibia (K-XRF) (pg/g)
Median: 19
Mean (SD)
Q1: 7.4 (3.2)
Q2: 14.1 (1.4)
Q3 18.9 (1.4)
Q4 24.9 (2.2)
Q5 40.9 (14)
External Review Draft
4-95
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Ettinaer et al.
(2014)
Ghana, South
Africa,
Seychelles,
Jamaica, and the
United States
2010-2011
Cross-sectional
Modeling the
Epidemiologic
Transition
Study (METS)
n = 150 (30
randomly
selected from
each site)
Young adults
(25-45) of
African descent
Average
individual born
Blood Pb (ICP-MS) (pg/dL)
GM:
BP (SBP, DBP)
1.55 (95% CI:
Median
1.66 (95% CI:
75th: 2.6
Max: 31.82
1.30, 1.85);
1.34, 1.93)
Age at measurement (years):
Mean (SD):
Males: 34.7 (6)
Females: 35.2 (6.2)
Logistic regression adjusted
for age, sex, site location,
education, paid employment,
marital status, smoking,
alcohol use, fish intake
(percent body fat in models
not assessing models of
BMI)
OR above/below median (1.66 pg/dL)b
High SBP (>130 mmHg) 1.69 (0.55, 5.15)
High DBP (>85 mmHg) 2.20 (0.59, 8.16)
-1975
Guoetal. (2019) n = 145
China
2015
Cross-sectional
Males free of
cardiovascular
disease
(including
angina
pectoris, Ml,
and thrombus)
Average
individual born
-1976
Blood Pb (ICP-MS) (pg/dL)
mean (SD): 8.50 (3.77)
Median: 7.85
75th: 10.08
Max: 28.17
Age at measurement (years):
mean (SD): 39 (12)
BP (SBP, DBP)
Linear (log-transformed) and
logistic regression adjusting
for age
Linear regression (log-transformed)
(mmHg)b
SBP: 7.28 (-7.68, 22.24)
DBP: 4.34 (-7.12, 15.80)
OR above/below median (7.85 pg/dL)b
SBP >134 mmHg: 2.28 (1.01, 5.12)
DBP >84 mmHg: 1.55 (0.70, 3.40)
External Review Draft
4-96
DRAFT: Do not cite or quote
-------
Stilly'o^sfg!? Population Exposure Assessment Outcome Confounders Effect Estimates and 95% Clsa
Cohort Studies
Gambelunqhe et
al. (2016)
Malmo
Sweden
1991-1994, re-
examination
2007-2012
Cohort with
cross-sectional
component
Malmo Diet
and Cancer
Study (MDCS-
CC)
n = 4,452
Aged 46-67
living in Malmo
Sweden
Average
individual born
-1935
Blood Pb (ICP-MS) (pg/dL)
Mean: 2.8
Max: 25.8
Quartile Means
BP (SBP, DBP)
Linear regression adjusted
for sex, age, smoking,
alcohol, waist circumference,
education
BP differences (mmHg) (Q4 vs.
Q1+Q2+Q3)be
SBP: 1.8 (0.5, 3.1)
SBP: Smokers 3.9 (1.6, 6.2)
1.5, 2.7)
Q1: 1.5
SBP Never-smokers: 0.6
Q2: 2.2
Males: 2.1 (0.3, 3.9)
Q3: 2.8
Females: 1.5 (-0.4, 3.4)
Q4: 4.7
<57 yr: 2.4 (1.2, 3.6)
>57 yr: 1.3 (-0.6, 3.2)
Age at measurement (years):
Mean: 57
DBP: 1.4 (0.6, 2.2)
DBP Smokers: 1.6 (0.7, 2
DBP Never-smokers: 1.1
Males: 1.7 (0.9, 2.5)
Females: 1.1 (0.2, 2.0)
<57 yr: 1.4 (0.7, 2.1)
>57 yr: 1.5 (0.75, 2.5)
External Review Draft
4-97
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Bulka et al.
(2019)
Bangladesh
Participants were
randomized
between April
2006 and August
2009, BP
measurements
taken at baseline,
and every 2 yr for
a total of 6 yr
Cohort
Bangladesh
Vitamin E and
Selenium Trial
(BEST)
n = 255
Participants
from the trial
were from
randomized
sample of
those taking
part of the
placebo arm of
the BEST study
Blood Pb (whole) (ICP-MS)
(Hg/dL)
Median 8.5
Age at measurement
25-37 yr (88/255)
38-46 yr (82/255)
47-64 yr (85/255)
BP (SBP, DBP, Mixed-effects regression
PP) models adjusting for
manganese, selenium, age,
sex, site, smoking,
educational duration,
creatine-corrected urinary
arsenic, diabetes, BMI,
antihypertensive medication
use
Yearly change in BP (mmHg) (4th quartile
vs. 1st quartile)b
SBP: 1.16 (95% CI: 0.21, 2.11)
DBP: 0.53 (95% CI: -0.10, 1.16)
PP: 0.63 (95% CI: -0.08, 1.34)
Average
individual born
-1976, -1965,
-1952
Glenn et al.
(2006)
South Korea
1997-2001
Cohort
n = 575
Pb-exposed
workers
Average
individual born
-1956
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Females
Visit 1
Visit 2
Visit 3
Males
Visit 1
Visit 2
Visit 3
20.3 (9.6)
20.8 (10.8)
19.8 (10.7)
35.0 (13.5)
36.5 (14.2)
35.4 (15.9)
BP (SBP) Multivariable models using
generalized estimating
equations were used in
longitudinal analyses
adjusted for visit number,
baseline age, baseline age
squared, baseline lifetime
alcohol consumption,
baseline BMI, sex, baseline
BP-lowering medication use,
alcohol consumption
Age at measurement:
Range 18-67 yr
SBP (mmHg)
Model 1 (short-term)
Blood Pb (longitudinal): 0.009 (0.002,
0.02)
Blood Pb (concurrent): 0.008 (-0.001,
0.02)
Model 4 (short and longer-term, controls
for tibia Pb)
Blood Pb (longitudinal): 0.009 (0.002,
0.02)
Blood Pb (concurrent): 0.01 (0.001, 0.019)
External Review Draft
4-98
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Yu et al. (2020)
Belgium
Blood Pb
collected in
1985-2005,
arterial stiffness
measured a
median of 9.4 yr
later
Cohort
Cadmium in
Belgium study
n = 267
Average
individual born
-1958
Blood Pb (ETAAS) (|jg/dL)
GM (IQR):
2.93 (1.8-4.7)
Age at measurement
Mean: 37 yr
Central and
Peripheral BP
Linear multivariable models
adjusting for sex, enrollment
characteristics (age, BMI,
smoking, drinking, serum
total to HDL-C ration, plasma
glucose, eGFR (estimated
from serum creatinine),
SES), the time interval
between measurement of
exposure biomarkers and
hemodynamic assessment,
and antihypertensive drug
treatment at enrollment and
follow-up
Per doubling of Pb concentration15
Peripheral
Systolic Pressure: 2.41 (-0.38, 5.20)
Diastolic Pressure: 0.50 (-1.07, 2.07)
PP: 1.91 (-0.32, 4.14)
Central
Systolic Pressure: 2.65 (-0.17, 5.46)
Diastolic Pressure: 0.42 (-1.18, 2.02)
PP: 2.23 (-0.03, 4.48)
Similar results when controlling for
baseline urinary Cd
AL = allostatic load; ALAD = 6-aminolevulinic acid dehydratase; BEST = Biomonitoring of Environmental Status and Trends; BMI = body mass index; BP = blood pressure; C282Y
HFE = mutant of the HFE wildtype; Ca2+ = calcium ion; Cd= cadmium; CI = confidence interval; CVD = cardiovascular disease; DBP = diastolic blood pressure; EAF = electric arc
furnace; eGFR = estimated glomerular filtration rate; ETAAS = electrothermal atomic absorption spectrometry; FBG = fasting blood glucose; GFAAS = graphite furnace atomic
absorption spectrometry; GM = geometric mean; GSE = geometric standard error; GuLF = Gulf Long-Term Follow-up; HDL-C = high-density lipoprotein cholesterol; H63D
HFE = mutant of the HFE wildtype; HFE = hemochromatosis gene; ICP-MS = inductively coupled plasma mass spectrometry; IQR = interquartile ratio; KNHANES = Korea National
Health and Nutrition Examination Survey; K-XRF = K-Shell X-ray fluorescence; LF = low frequency; MAP = mean arterial pressure; MDCS-CC = cardiovascular cohort of the Malmo
Cancer and Diet Study; METS = Modeling the Epidemiologic Transition Study; Ml = myocardial infarction; mo = month(s); NAS = Normative Aging Study; NH = non-Hispanic;
NHANES = National Health and Nutrition Examination Survey; OR = odds ratio; Pb = lead; PIR = poverty-to-income ratio; PP = pulse pressure; Q = quartile; SBP = systolic blood
pressure; SD = standard deviation; SE = standard error; SES = socioeconomic status; TC = total cholesterol; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
°lncrement unclear.
dBlood Pb analysis method unclear, assumed based on data source.
Confidence intervals estimated based on reported p values.
'Original results reported in |jg/L.
External Review Draft
4-99
DRAFT: Do not cite or quote
-------
Table 4-4 Epidemiologic studies of Pb exposure and hypertension.
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome Confounders
Effect Estimates and 95% Clsa
Teve et al. (2020)
United States
NHANES 1999-2016
Cross-sectional
NHANES
n = 30,467
20-79 yr
Average
individual born
-1965
Blood Pb ICP-MSC (pg/dL)
NH White men: 1.89
NH White women: 1.30
NH Black men: 2.20
NH Black women: 1.49
Hispanic men: 2.18
Hispanic women: 1.30
Other men: 1.93
Other women: 1.42
Hypertension
(SBP
>140 mmHg,
DBP >90, use of
antihypertensive
medication)
Race/ethnicity, age,
gender, education
level, BMI, and PIR
OR (95% CI):
1.002 (0.983, 1.021)
Age at measurement:
NH White men: 46.37
NH White women: 47.00
NH Black men: 43.09
NH Black women: 43.28
Hispanic men: 39.67
Hispanic women: 40.51
Other men: 42.92
Other women: 43.54
External Review Draft
4-100
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
(Huang. 2022. pp. 2author- NHANES
year) n = 32,289
>20 yr
United States
NHANES 1999-2018
Cross-sectional
Average
individual born
-1958
Blood Pb (ICP-MS) (pg/dL)
Mean (SD): 1.73 (1.71)
Age at measurement (years):
Mean (SD): 49.68 (18.04)
Hypertension
(SBP
>130 mmHg,
DBP >80, use of
antihypertensive
medication, or
self-reported
hypertension)
Linear regression
models adjusted for
age, sex, race,
education, family
income poverty
ratio, BMI, alcohol
use, and smoking
Hypertension (OR)
All 1.01 (0.99, 1.03)
Women 1.03 (0.99, 1.07)
Men 1.01 (0.99, 1.03)
Mexican American 0.99 (0.96, 1.02)
Other Hispanic 1.01 (0.95, 1.06)
NH White 1.03 (1.00, 1.06)
NH Black 1.02 (0.98, 1.06)
Other race 1.04 (0.97, 1.11)
BMI>30 1.00 (0.97, 1.03)
BMI 25-30 1.00 (0.98, 1.03)
BMI <25 1.03 (1.00, 1.06)
External Review Draft
4-101
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Tsoietal. (2021)
United States
NHANES 1999-2016
Cross-sectional
NHANES
n = 39,477
Adults >20
Average
individual born
-1960
Blood Pb ICP-MS (pg/dL)
Median: 1.30
Q1
Q2
Q3
Q4
<0.89
0.89-1.30
1.30-2.10
>2.10
Age at measurement
Mean (SE)
Hypertensive:
54.08 (0.23) yr
Non-hypertensive:
39.87 (0.19) yr
Hypertension
(SBP
>130 mmHg,
DBP >80, use of
antihypertensive
medication, or
self-reported
hypertension)
Multivariable logistic
regression adjusted
for age, sex
ethnicity, waist
circumference, PIR,
education, ever
cigarette smoking,
diabetes, and stage
3-5 chronic kidney
diseases
ORb
For every doubling of blood Pb
All 1.09 (1.04, 1.14)
Male 1.10 (1.05, 1.16)
Female 1.04 (0.97, 1.11)
Mex. American 0.98 (0.89, 1.08)
Other Hispanic 1.07 (0.93, 1.23)
NH White 1.12 (1.05, 1.19)
NH Black 1.06 (0.99, 1.15)
Other ethnicity 1.10 (0.95, 1.28)
Quartiles (All)
Q2 vs. Q1 1.15(1.04, 1.26)
Q3 vs. Q1 1.17(1.05, 1.31)
Q4 vs. Q1 1.21 (1.07, 1.36)
Miao etal. (2020)
United States
NHANES 1999-2016
Cross-sectional
NHANES
n = 30,762
>20 yr
Average
individual born
-1978, -1958, or
before -1947
Whole Blood Pb (GFAAS with
Zeeman correction (1999-
2002 and ICP-MS (2003-
2016) (pg/dL)
mean (SE)
Male: 1.50 (0.02)
Female: 1.07 (0.01)
Age at measurement (years):
20-39 36.5%
40-59 38.9%
>60 24.6%
Uncontrolled
hypertension
(SBP
>130 mmHg or
DBP >80 mmHg
or
antihypertension
medication use)
and uncontrolled
hypertension
(SBP
>130 mmHg or
DBP >80 mmHg
regardless
antihypertension
medication use)
Logistic regression
adjusted for age,
sex, race/ethnicity,
ratio of family
income to poverty,
education, smoking
status, serum
cotinine, alcohol
intake, BMI, and
menopausal status
among females
OR (95% CI)
Uncontrolled hypertension vs. non-
hypertension:
Male: 1.037 (1.015, 1.060
Female: 1.020 (0.970, 1.074)
Uncontrolled Hypertension vs.
Controlled Hypertension
Male: 1.157 (1.080, 1.239)
Female: 1.109 (1.020, 1.205)
Uncontrolled hypertension vs.
Controlled and Non-hypertension:
Male: 1.062 (1.036, 1.088)
Female: 1.056 (1.011, 1.102)
External Review Draft
4-102
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Scinicariello et al. (2011)
United States
NHANES 1999-2006
Cross-sectional
NHANES
n = 16,222
Blood Pb
<10 |jg/dL
Average
individual born
-1959
Blood Pb (ICP-MS) (|jg/dL)
Percentile:
10th: <0.7
90th: 3.5-10
Age at measurement (years):
Mean (SE)
White males: 47.14 (0.37)
White females: 49.64 (0.36)
Black males: 42.86 (0.37)
Black females: 45.10 (0.42)
Mexican-Am males: 37.64
(0.48)
Mex-Am females: 40.67 (0.65)
Hypertension
prevalence (SBP
>140 mmHg,
DBP >90, use of
antihypertensive
medication)
Multivariable logistic
and linear
regression models.
Confounders: age,
BMI, self-reported
diabetes alcohol
ingestion, smoking
status, education,
serum creatinine,
serum total calcium,
sodium, hematocrit,
and blood Cd
Prevalence OR (90th vs. 10th
percentile)15
All: 1.26 (0.98, 1.61)
White males: 1.20 (0.74, 1.96)
White females: 1.07 (0.69, 1.66)
Black males: 2.69 (1.08, 6.72)
Black females: 1.04 (0.50, 2.16)
Mex-Am males: 1.03 (0.23, 4.59)
Mex-Am females: 0.67 (0.37, 1.20)
Hara et al. (2015)
United States
NHANES 2003-2010
Cross-sectional
NHANES
n = 12,725
>20 yr
Average
individual born
-1957
Blood Pb (ICP-MS) (|jg/dL)
GM (IQR):
Females
Black: 1.37 (0.88-2.10)
Hispanic: 1.21 (0.80-1.78)
White: 1.22 (0.80-1.86)
Males
Black: 1.86 (1.20-2.85)
Hispanic: 1.94 (1.25-2.83)
White: 1.73 (1.16-2.57)
Age at measurement:
mean (SD)
Black females: 48.31 (6.8)
Hispanic females: 48.1 (16.8)
White females: 53.0 (8.4)
Black males: 47.7 (16.9)
Hispanic males: 46.1 (6.8)
White males: 53.1 (18.6)
Hypertension
(SBP
>140 mmHg or
DBP >90 mmHg
or the use of
antihypertensive
medication)
Logistic models
adjusted for
ethnicity, sex, age,
BMI, heart rate,
hematocrit, serum
total calcium y-
glutamyltransferase,
cotinine, dietary
sodium to
potassium intake
ratio, college
education,
antihypertensive
drug treatment
OR (95% CI) for doubling blood Pbb
All: 0.95 (0.90, 1.01)
Females: 0.95 (0.87, 1.04)
Black: 0.82 (0.67, 0.99)
Hispanic: 0.86 (0.72, 1.04)
White: 1.06 (0.94, 1.21)
Males: 0.95 (0.87, 1.02)
Black: 1.00 (0.84, 1.20)
Hispanic: 0.84 (0.71, 0.99)
White: 0.99 (0.89, 1.10)
External Review Draft
4-103
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Lee etal. (2016b)
Korea
KNHANES
2007-2013
Cross-sectional
KNHANES
n = 8,493
>20 yr
Average
individual born
-1981, -1961, or
before -1950
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Quartiles
Q1
Q2
Q3
Q4
0.206-1.539
1.540-2.056
2.057-2.716
2.717-24.532
Age at measurement:
20-39 (58.8%)
40-59 (34.9%)
>60 (6.3%)
Prehypertension
prevalence (DBP
80-89 mmHg or
SBP 120-
139 mmHg and
the absence of
any current
treatment or
diagnosis of
hypertension)
Logistic regression
adjusted for age,
sex, education,
occupation, income
residence, smoking
alcohol
consumption,
exercise level,
serum creatinine
clearance, chronic
disease, and
antihypertensive
medication
ORb (95% CI):
Q2 vs. Q1 1.24
(1
.04,
1.48)
Q3 vs. Q1 1.27
(1
.06,
1.52)
Q4 vs. Q1 1.30
(1
.07,
1.60)
Lee etal. (2016a)
KNHANES
2008-2013
Cross-sectional
KNHANES
n = 11,797
>19 yr
Average
individual born in
or before -1991
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
GM (95% CI)
Male: 2.396 (2.362, 2.430)
Female: 1.919 (1.889, 1.949)
Hypertension
(SBP
>140 mmHg or
DBP >90 mmHg)
and
Prehypertension
(SBP
>120 mmHg or
DBP >80 mmHg)
Logistic models
adjusted for sex,
age, residence
area, education
level, smoking,
drinking status,
BMI, physical
activity, serum
creatinine, and
hemoglobin
OR for doubling blood Pbb
Hypertension
All: 1.09 (0.98, 1.22)
Male: 0.92 (0.80, 1.07)
Female: 1.29 (1.10, 1.51)
Prehypertension
All: 1.09 (0.99, 1.21)
Male: 0.98 (0.85, 1.12)
Female: 1.21 (1.06, 1.38)
Choi etal. (2018)
South Korea
KNHANES 2013
Cross-sectional
KNHANES
n = 1,350
19-64 yr
Average
individual born
-1989, -1978,
-1968, -1958,
-1951
Blood Pb (GFAAS with
Zeeman correction) (ug/dL)
Mean 2.01
SE 0.025
Age at measurement:
19-29 (23.1%)
30-39 (22.7%)
40-49 (21.9%)
50-59 (22.1%)
60-64 (10.1%)
Hypertension Logistic regression ORd (95% CI)
prevalence (SBP
>140 mmHg, or
DBP >90 mmHg,
or use of
antihypertensive
medication)
adjusting for age,
sex, smoking, and
BMI
Curry intake:
1.108 (0.827, 1.484)
Non-curry intake:
1.399 (1.054, 1.857)
External Review Draft
4-104
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Qu et al. (2022)
China
2017-2018
Cross-sectional
China National
Human
Biomonitoring
Study
n = 11,037
18-79 yr
Average
individual born
-1969
Blood Pb (ICP-MS) (|jg/dL)e
Quartiles
Q1
Q2
Q3
Q4
<1.59
1.59-2.24
2.24-3.21
>3.21
Prehypertension
(SBP 120-139,
DBP 80-89),
hypertension
(Chinese
guideline: SBP
>140, or DBP
>90), elevated
BP (2017
ACC/AHA: SBP
120-129, DBP
<80), stage 1
hypertension
(2017 ACC/AHA:
SBP 130-139,
DBP 80-89)
Logistic regression
adjusted for sex,
age, BMI, regions,
education, smoking
status, alcohol
consumption, family
history of
hypertension,
residence area, rice
consumption, red
meat consumptions,
vegetable
consumptions,
FBG, TC, HDL-C,
urinary arsenic
levels, and blood
Cd levels
ORb (95% CI)
Prehypertension
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
1.24 (1.04-1.47)
1.27 (1.02-1.59)
1.56 (1.22-1.99)
Hypertension
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
1.23 (0.96-1.56)
1.49 (1.12-1.96)
2.33 (1.67-3.24)
Elevated BP
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
1.00 (0.77-1.31)
1.10 (0.83-1.47)
1.18 (0.88-1.57)
Stage 1 Hypertension
Q2 vs. Q1: 1.35 (1.10-1.65)
Q3 vs. Q1: 1.35 (1.04-1.77)
Q4 vs. Q1: 1.75 (1.31-2.33
Bushnik et al. (2014)
Canada
2007-2011
Cross-sectional
Canadian Health Blood Pb (ICP-MS) (|jg/dL)
Measures Survey Mean: 1.64 (1.58-1.71)
n = 4,550
Nonpregnant
individuals aged
40-79
Average
individual born
-1954
Age at measurement:
Mean: 55.4
Hypertension:
SBP >140, or
DBP >90, or use
of
antihypertensive
medication, or
health care
provider
diagnosis of
hypertension
Logistic regression
adjusted for age,
sex, education,
smoking, alcohol,
physical activity,
BMI, non-HDL
cholesterol,
diabetes, chronic
kidney disease,
family history of
high BP,
antihypertension
medication use
OR
Age 40-79: 0.02 (0.00, 0.43)
Age 40-54: 0.01 (0.00, 0.20)
Age 55-79: 0.01 (0.00, 1.00)
Male: 0.00 (0.00, 0.91)
Female: 0.02 (0.00, 0.60)
*Results correspond to linear model.
Concentration response function for
splines not shown. Authors indicated
no relationship between hypertension
and BLL
External Review Draft
4-105
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Wang et al. (2020)
China
Cross-sectional
n = 816
Adults 40-75,
residing in area
for >15 yr, and
subsisting on rice
and vegetables
grown in the
polluted (Cd
concentration
>0.2 mg/kg) or
unpolluted (Cd
concentration
<0.05 mg/kg)
area
Blood Pb (ICP-MS) (|jg/dL)
Median (IQR)
Polluted: 3.54 (2.4-4.89)
Unpolluted: 2.61 (1.70-3.84)
Age at measurement:
Mean (SD)
Polluted area
Hypertensive: 60.32 (8.08)
Normotensive 55.61 (8.52)
Unpolluted area Hypertensive:
59.92 (9.19)
Normotensive 56.86 (9.22)
Hypertension
(SBP
>140 mmHg,
DBP >90 mmHg,
self-reported
physician
diagnosis, or
current use of
antihypertensive
medication)
Logistic regression
adjusted for age,
gender, smoking
status, and BMI
ORb (95% CI)
See Figure 4-2 and Figure 4-3
Lopes et al. (2017)
Cambe, Brazil
2011
Cross-sectional
n = 948
adults 40 yr and
older, randomly
sampled from
census tracts in
the region
Average
individual born
-1956
Blood Pb (ICP-MS) (pg/dL)
GM:
1.97 (95% Cl:1.90-2.04)
Age at measurement:
Mean: 54.5 yr
Hypertension
(SBP
>140 mmHg,
DBP >90 mmHg
or current
antihypertensive
medication)
Multiple linear
regression adjusted
for age, sex, race,
income, education,
antihypertensive
medication, total
cholesterol,
triglycerides,
glycemia, smoking,
alcohol
consumption, and
BMI
OR 1.079 (1.026, 1.136)
External Review Draft
4-106
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Xu et al. (2021)
United States
2011-2013
Cross-sectional
GuLF Study
Cohort
n = 957
Adults >21
Average
individual born in
or before -1991
Blood Pb (ICP-MS) (pg/dL)
Median: 0.09
75th percentile: 0.19
Maximum: 33.8
Age at measurement: >21 yr
Hypertension
(SBP
>140 mmHg,
DBP >90 mmHg
or current
antihypertensive
medication)
Multivariable
Poisson regression
adjusted for age,
sex, race,
educational
attainment, and
household income
Prevalence Ratio (PR)
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
BMI >30
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
0.96 (0.73,1.25)
0.91 (0.71, 1.17)
0.86 (0.66, 1.12)
1.05 (0.78, 1.42)
1.09 (0.82, 1.45)
1.14 (0.84, 1.55)
BMI <30
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
0.89 (0.52, 1.52)
0.81 (0.50, 1.31)
0.82 (0.50, 1.32)
Weaver et al. (2008)
South Korea
1999-2001
Cross-sectional
n = 652
current and
former Pb
workers
Average
individual born
-1957
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Mean (SD): 31.9 (14.8)
Patella Pb (K-XRF) (pg/g)
Mean (SD): 75.1 (101.1)
Age at measurement:
Mean: 43.3, SD: 9.8
Hypertension
(SBP
>140 mmHg,
DBP >90 mmHg;
and/or use of
antihypertensive
medications; or
physician
diagnosis)
Logistic regression
models adjusted for
age, sex, BMI,
diabetes,
antihypertensive
and analgesic
medication use, Pb
job duration, work
status, tobacco, and
alcohol use
Quantitative results not reported.
None ofthe examined Pb exposure
metrics (blood, patella, and
logarithmic (In)-transformed patella)
were significantly associated with
hypertension
External Review Draft
4-107
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Elmarsafawv et al. (2006)
Boston, MA
1991
Cross-sectional
NAS
n = 471
Elderly men
(mostly white)
Average
individual born
-1924
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Mean (SD): 6.6 (4.3)
Bone (K-XRF) (pg/g)
Mean (SD)
Tibia: 21.6 (12.0)
Patella: 31.7 (18.3)
Age at measurement
Mean: 67
Hypertension
(SBP
>160 mmHg,
DBP >95 mmHg;
and/or physician
diagnosis with
current use of
antihypertensive
medications)
Logistic regression
models adjusted for
age, BMI, family
history of
hypertension,
history of smoking,
dietary sodium
intake, and
cumulative alcohol
ingestion
OR (95% CI)
Low Ca2+ group (<800 mg/d):
Blood: 1.02 (1.00, 1.03)
Tibia: 1.02 (1.00, 1.03)
Patella: 1.01 (1.00, 1.01)
High Ca2+ group (>800 mg/d):
Blood Pb: 1.01 (0.99, 1.02)
Tibia Pb: 1.02 (1.00, 1.05)
Patella Pb: 1.01 (1.00, 1.02)
Gambelunahe et al. (2016) Malmo Diet and Blood Pb (ICP-MS) (pg/L)
Malmo, Sweden
1991-1994, re-examination
2007-2012
Cohort
Cancer Study
(MDCS-CC)
n = 4,452
aged 46-67
living in Malmo
Sweden
Average
individual born
-1935
All: 2.8
Max: 25.8
Age at measurement:
Mean: 57
Range: 46-67
Hypertension
(SBP
>140 mmHg, or
DBP >90 mmHg,
or
antihypertensive
medication use)
Logistic regression OR (Q4 vs. Q1+Q2+Q3)b
adjusted for sex,
age, smoking,
alcohol, waist
circumference,
education
All: 1.3 (1.1, 1.5)
Smokers: 1.5 (1.2, 1.8)
Never-smokers: 0.96 (0.7, 1.3)
<57 yr: 1.5(1.2, 1.9)
>5 yr: 1.3 (0.9, 1.4)
Male: 1.20 (0.60, 1.5)
Female1.4 (1.1, 1.7)
At Follow-up
Antihypertensive medication use: 1.0
(0.8, 1.2)
High BP: 1.0 (0.7, 1.3)
External Review Draft
4-108
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Zheutlin et al. (2018)
Boston, MA
1986-2013
Cohort
NAS
n = 475
Male volunteers
aged 21 to 80 yr
(bone Pb
measurement
started in 1991,
resistant
hypertension
assessed starting
at visit prior to
first bone
measurement)
Average
individual born
-1931
Blood Pb (GFAAS with
Zeeman correction) (pg/dL)
Median (IQR): 5.0 (3.4-8.0)
Bone (K-XRF) (ug/g)
Median (IQR)
Tibia: 20.0 (13.0-28.5) Patella:
27.0 (18.0-40.0)
Age at measurement:
67.9 (63.2-72.6)
Incident resistant
hypertension
(inadequate
control (SBP
>140 mmHg,
DBP >90 mmHg)
while taking >3
antihypertensive
medications, or
adequate control
(SBP
<140 mmHg,
DBP <90mmHg)
while taking >4
antihypertensive
medications
Poisson regression
with robust error
variation adjusting
for age,
race/ethnicity,
education
attainment, income
level, BMI, family
history of
hypertension, and
cigarette smoking
RR
Tibia: 1.12 (1.01, 1.25)
Patella: 1.04 (0.96, 1.13)
Blood: 1.02 (0.97, 1.08)
ACC = American College of Cardiologists; AHA = American Heart Association; BLL = blood lead level; BMI = body mass index; BP = blood pressure; Ca2+ = calcium ion;
Cd = cadmium; CI = confidence interval; DBP = diastolic blood pressure; FBG = fasting blood glucose; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric
mean; GuLF = Gulf Long-Term Follow-up; HDL-C = high-density lipoprotein cholesterol; ICP-MS = inductively coupled plasma mass spectrometry; IQR = interquartile ratio;
KNHANES = Korea National Health and Nutrition Examination Survey; LF = low frequency; MDCS-CC = cardiovascular cohort of the Malmo Cancer and Diet Study; Mex-
Am = Mexican-American; NAS = Normative Aging Study; NH = non-Hispanic; NHANES = National Health and Nutrition Examination Survey; OR = odds ratio; PIR = poverty-to-
income ratio; Pb = lead; Q = quartile; PR = prevalence ratio; RR = relative risk; SBP = systolic blood pressure; SD = standard deviation; SE = standard error; TC = total cholesterol;
K-XRF = K-Shell X-ray fluorescence; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
°Blood Pb analysis method unclear, assumed based on data source.
Confidence intervals estimated based on reported p values.
eOriginal results reported in |jg/L.
External Review Draft
4-109
DRAFT: Do not cite or quote
-------
Table 4-5
Epidemiologic studies of Pb exposure and blood pressure and hypertension among children.
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Zhang et al. (2012)
Mexico City, Mexico
1994-2003
Follow-up 2008-2010
Cohort
Early Life Exposures
in Mexico to
Environmental
Toxicants project
n = 457 mother-child
pairs
Average individual
born -1973
Cord Blood (AAS) (pg/dL)
Mean (SD): 5.51 (3.45)
Bone (K-shell X-ray) (pg/g)
Median (IQR)
Tibia: 9.3 (3.3-16.1)
Patella: 11.6 (4.5-19.9)
Age at measurement
Mean (SD): 25.6 (5.4)
BP (SBP, DBP) in
children
Age at outcome
Mean (SD): 10.7
(2.4)
Multiple
regression models
and generalized
estimating
equations (log
linear for cord
blood, linear for
concurrent blood
and maternal
bone) adjusted for
maternal
education, birth
weight, BMI, sex,
and child
concurrent age
Difference in BP (mmHg)
Cord Blood
SBP All: 0.23 (-0.14, 0.60)
DBP All: 0.23 (-0.03 0.49)
Tibia
SBP All: 0.74 (-0.10, 1.58)
SBP Female: 1.62 (0.54, 2.71)
SBP Male: -0.26 (-1.52, 1.00)
DBP All: 0.35 (-0.36, 1.07)
DBP Female: 1.24 (0.23, 2.25)
DBP Male: -0.64 (-1.57, 0.30)
Patella
SBP All: 0.28 (-0.45, 1.00)
DBP All: 0.14 (-0.59, 0.88)
Gump et al. (2005)
Oswego, NY (born at a
single hospital in New York
from 1991-94)
Cohort
Oswego Children's
Study
n = 122 children aged
9.5 yr
Average individual
born -1990
Cord blood (ETAAS)
(Hg/dL)
Mean (SD): 2.97 (1.75)
Child blood (ETAAS and
ASV) (pg/dL)
Mean (SD): 4.62 (2.51)
BP (SBP, DBP) and
TPR in children
Age at outcome 9.5
Linear regression
models examined
the adjusted for
HOME score, SES,
birth weight, child
BMI, and child sex
Baseline BP (mmHg) per 1 [jg/dL
increase in cord blood Pbb
SBP: 12.16 (2.44, 21.88)
DBP: 8.45 (-0.45, 17.35)
TPR, dyne-s/cm5
no association, results NR
Age of child blood Pb
measurement
Mean: 2.6
Relationship of blood Pb with
change in z-score for outcome
(post and prestress) per 1 [jg/dL
increase in childhood blood Pb
SBP: -0.009 (-0.074, 0.055)
DBP: 0.069 (-0.001, 0.138)
TPR, dyne-s/cm5
0.088 (0.024, 0.152)
External Review Draft
4-110
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Gump et al. (2007)
Oswego, NY (born at a
single hospital in New York
from 1991-94)
Cohort
Oswego Children's
Study
n = 122 children aged
9.5 yr
Average individual
born -1990
Cord blood (ETAAS)
(Hg/dL)
Mean (SD): 2.97 (1.75)
Child blood (ETAAS and
ASV) (pg/dL)
Mean (SD): 4.62 (2.51)
Age of child blood Pb
measurement (years):
Mean: 2.6
BP (SBP) and TPR
in children
Age at outcome 9.5
Linear regression
models adjusting for
the same covariates
as in Gump et al.
(2005). Separate
models testing
whether Pb is a
mediator of SES
associations (Sobel
test) and whether Pb
moderates SES
associations (Pb-
SES interaction)
Blood Pb was a mediator of the
SES-TPR relationship
SES alone: -0.62 dyne-s/cm5
(p < 0.05)
SES with Blood Pb: -0.40 dyne-
s/cm5 (p > 0.10), change in R2
attributable to SES: -55.3%
Blood Pb was a potential
moderator of the SES-TPR
relationship.
Blood Pb x SES interaction:
p = 0.07
Blood Pb was a moderator of
SES-SBP relationship
Pb x SES interaction: p = 0.007
Kupsco et al. (2019)
Mexico City, Mexico
2007-2011
Cohort
Research in Obesity, Maternal Blood (ICP-MS) BP (SBP, DBP)
Growth Environment
and Social Stress
(PROGRESS) birth
study
n: 548
Mother/child pairs
Maternal blood
tested for metals in
second trimester,
children assessed at
age 4-6
(Hg/dL)
Mean: 3.7, SD: 2.7
Range 0.75-18
Max: 18
Age at measurement
(years):
28 (5.6)
Age at Outcome:
Mean (SD): 4.8
(0.55)
Range: 4-6.8
Linear regression
adjusted for birth
weight, gestational
age, prepregnancy
BMI, education,
socioeconomic
status, parity,
environmental
tobacco smoke
BP (mmHg) per 1 In unit increase
in maternal blood Pbc
SBP: -0.05 (-0.09, 0.07)
DBP: 0 (-0.23, 0.54)
Average individual
born -1981
External Review Draft
4-111
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Skroder et al. (2016)
Bangladesh
(2002- 2004)
Cohort
Maternal and Infant
Nutrition
Interventions, Matlab
n = 1,511
(gestational week
[GW] 14); 713 (GW
30)
Mother-child pairs
Erythrocyte Pb (Eyr-
Pb) measured at
GW14 and GW30,
children assessed at
age 4.5
Maternal Blood (Eyr-Pb)
(ICP-MS) (pg/kg)
GW: 14
Median: 73
95th: 172
GW: 30
Media: 86
95th : 506
Age at measurement:
Mean (SD): 26 (6)
BP (SBP, DBP)
Age at Outcome
Mean: 4.5 yr
Linear regression
adjusted for sex,
birth weight, season
of birth, age at
outcome
measurements,
height for age z-
score, maternal BMI
at GW8, parity, SES,
and supplementation
group
BP (mmHg) per pg/kg Eyr-Pbc
GW14:
SBP: 0.042 (-0.058, 0.14)
DBP: -0.0058 (-0.090, 0.077)
GW30:
SBP: 0.042 (
DBP: 0.072 I
-0.090, 0.17)
-0.039, 0.18)
Average individual
born -1977
Gump et al. (2011)
Oswego, NY
Cross-sectional
n = 140 children
ages 9-11 yr
Blood (ICP-MS) (pg/dL)
GM: 1.01
BP (SBP), TPR
Q1
Q2
Q3
Q4
0.14-0.68
0.69-0.93
0.94-1.20
1.21-3.76
Linear regression
adjusted for sex,
SES, BMI, and age
Change in SBP (mmHg) across
quartiles in response to acute
stresscd
Q1: 5.30, Q2: 7.33, Q3: 7.07, Q4:
7.23, p for trend = 0.31
Change in TPR (%) across
quartiles in response to acute
stresscd
Q1: 2.91, Q2: 8.18, Q3: 9.55, Q4:
9.51, p for trend = 0.03
External Review Draft
4-112
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Factor-Litvak et al. (1996)
Kosovo (when part of
Yugoslavia)
January 1st, 1984-July 31st,
1986
Cross-sectional
Yugoslavia
Prospective Study
n = 281 from two
towns (Kovoska
Mitrovica [Exposed
town] and Pristina
[Unexposed town 25
miles south]
Average individual
born -1980
Blood (GFAAS) (pg/dL)
Exposed Town
Mean (SD): 37.3 (12.0)
Range: 9.5- 76.4
Unexposed Town
Mean (SD): 8.7 (2.8)
Range: 4.1-20.2
Blood Pressure
(SBP, DBP)
Linear regression
adjusted for ethnic
group, birth order
(and height, BMI,
and sex for SBP and
waist circumference
for DBP)
BP (mmHg) per 1 pg/dL blood Pb
SBP: 0.054 (-0.024, 0.13)
DBP: 0.042 (-0.010, 0.090)
Lu et al. (2018)
Guiyu (e-waste exposed),
Haojiang (reference)
China
2016
Cross-sectional
n = 590
children (aged 3-7)
residing in either
Guiyu or Haojiang
China
Average individual
born -2011
Blood (GFAAS) (|jg/dL)
Median
Exposed: 7.14
Unexposed: 3.91
Age at measurement
Mean (SD)
Exposed: 4.52 (0.86)
Unexposed 4.40 (1.04)
Blood pressure
(SBP, DBP)
Age at Outcome
Mean (SD)
Exposed: 4.52
(0.86)
Unexposed: 4.4
(1.04)
Linear regression
adjusted for outdoor
activities, family
member smoking,
parent education and
diet (including
cooking oil, picky
eating, sweetmeat
consumption, salted
products, vegetable
and fruit
consumption, dairy
products, bean
products, marine
products), age, sex,
BMI, and family
history of diseases
(hypertension,
diabetes, obesity)
Ln-transformed Blood Pb and BP
(mmHgf
SBP: -0.30 (-2.6, 2.00)
DBP: -2.11 (-4.38, 0.17)
External Review Draft
4-113
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Ahn etal. (2018)
Korea
KNHANES
2010-2016
Cross-sectional
KNHANES
n = 1,776
Adolescents (10-
18 yr)
Average individual
born -1999
Blood (GFAAS with
Zeeman background
correction (pg/dL)
GM (95% CI): 1.19 (1.17-
1.22)
Age at measurement:
Range: 10-18 yr
BP (SBP, DBP)
Prehypertension
(SBP 120-
140 mmHg, DBP
80-90 mmHg)
Linear and logistic
regression adjusted
for sex, age,
residence area,
smoking status,
drinking status, BMI,
year of
measurement,
physical activities,
hemoglobin, and
serum creatinine
Mean difference (mmHg) for
doubling blood Pbc
DBP: -0.680 (-1.561, 0.221)
SBP: -0.0999 (1.098, 0.898)
Prehypertension (OR [95% CI]) for
doubling blood Pbc
0.906 (0.629, 1.305)
Xu et al. (2017)
United States
NHANES 1999-2012
Cross-sectional
NHANES
n = 11,662
Adolescents 12-19
participating in
NHANES
Average individual
born -1990
Blood (ICP-MS)C (|jg/dL)
Mean (SD) 1.17 (1.20)
Q1: <0.6
Q2: 0.6-0.9
Q3: 0.0-1.34
Q4: >1.34
Age at measurement
Range: 12-19 yr
BP (SBP, DBP)
Linear models
adjusted for age,
sex, PIR, waist
circumference,
serum cotinine,
physical activity and
NHANES cycle
BP (mmHg) (Q4 vs. Q1)c
SBP: 0.001 (-0.001, 0.004)
DBP: 0.001 (-0.006, 0.008)
External Review Draft
4-114
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Yao et al. (2020)
United States
2007-2016
Cross-sectional
NHANES
n = 7,076
Children and
adolescents
Average individual
born -1999
Blood (ICP-MS) (ug/dL)
Mean: 0.67
Q1
Q2
Q3
Q4
<0.46
0.46-0.65
0.65-0.96
>0.96
Age at measurement
Mean (SD): 11.99 (2.88)
BP (SBP, DBP)
High BP (self (or
parent) reported
hypertension
diagnosis or
antihypertension
medication use for
those >16, or SBP
>120 mmHg or
DBP >80)
Linear or logistic
regression adjusted
for age, sex,
race/ethnicity, BMI,
cycle, serum cotinine
levels, hematocrit,
annual family
income, and intake
of calcium, sodium,
and potassium
BP (Regression coefficient) for
log-transformed blood Pbbc
SBP
All: -0.48 (-1.07, 0.11)
Male: -0.55 (-1.35, 0.25)
Female: -0.53 (-1.41, 0.35)
Mexican-American: -0.10 (-1.06,
0.86)
Other Hispanic: -1.77 (-3.46,
-0.08)
White: -0.27 (-1.21, 0.67)
Black: 0.17 (-0.15, 1.65)
DBP
All: 0.75 (-1.01, 1.49)
Male:1.16 (-0.13, 2.45)
Female: 0.24 (-1.01, 1.49)
Mexican-American: 1.12 (-0.49,
2.73)
Other Hispanic: -0.86 (-2.53,
0.81)
White: 1.99 (0.58, 3.40)
Black: -2.30 (-4.38, -0.22)
High BP (OR) Q4 vs. Q1
All: 0.89 (0.62-1.27)
"Change in BP associated with a
twofold increase in blood Pb was
calculated by dividing the
regression coefficient by log2(e)
External Review Draft
4-115
DRAFT: Do not cite or quote
-------
Reference and Study
Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Desai et al. (2021)
NHANES
2009-2016
Cross-sectional
NHANES
n = 1,642
participants aged 8-
17 yr
Average individual
born -2000
Blood Pb (ICP-MS) (pg/dL)
Median: 0.57
95th percentile: 1.6
Age at measurement
Median: 152 mo (-12.7 yr)
BP (SBP, DBP, PP)
Multivariate linear
regression models
adjusted for age,
sex, race, BMI, total
energy intake,
NHANES cycle,
education of
household head, and
income to poverty
ratio
BP (mmHg)
SBP: -0.351 (-1.391, 0.689)
DBP: -0.078 (-1.365, 1.209)
PP: -0.273 (-1.781, 1.235)
Zhang et al. (2021)
Boston, MA (United States)
Baseline 2002-2013, follow-
up through 2018
Cohort
Boston Birth Cohort
n = 1,194
Average individual
born -1980
Maternal red blood Pb
(measured 24-72 hr
postdelivery, ICP-MS)
(pg/dL)
Median: 2.42
75th percentile: 3.68
Max: 24.8
Age at measurement
(age at delivery)
Mean (SD): 27.7 (6.5)
BP (SBP) percentile
(based on child
age, sex, and
height according to
the 2017 American
Academy of
Pediatric Clinical
Practice Guideline)
Age at outcome
Median (IQR):
8.4 (6.2-10.6)
Multivariate linear
regression models
adjusting for
maternal age,
race/ethnicity,
educational level,
prepregnancy BMI,
and smoking history
Difference in child SBP percentile0
Quartiles
Q2 vs. Q1: 0.92 (-3:13, 4.97)
Q3 vs. Q1: 1.39 (-2:85, 5.64)
Q4 vs. Q1: -0.62 (-4.97, 3.74)
Per 1 pg/dL increase in blood Pb
0.142 (-0.673, 0.958)
AAS = atomic absorption spectrometry; ASV = anodic stripping voltammetry; BMI = body mass index; BP = blood pressure; CI = confidence interval; DBP = diastolic blood pressure;
ETAAS = Electrothermal Atomic Absorption Spectrometry with Zeeman background correction; Eyr = erythrocyte; GFAAS = graphite furnace atomic absorption spectrometry;
GM = geometric mean; GSD = geometric standard deviation; GW = gestational week; HOME = Health Outcomes and Measures of the Environment Study; ICP-MS = Inductively
Coupled Plasma Mass Spectrometry; IQR = interquartile ratio; KNHANES = Korean National Health and Nutrition Examination Survey; mo = month(s); Mex-Am = Mexican American;
NHANES = National Health and Nutrition Examination Survey; NR = not reported; OR = odds ratio; PIR = poverty-to-income ratio; PP = pulse pressure; PROGRESS = Programming
Research in Obesity, Growth Environment and Social Stress; Q = quartile; SBP = systolic blood pressure; SD = standard deviation; SES = socioeconomic status; TPR = total
peripheral resistance; Pb = lead; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bConfidence intervals estimated based on reported standard errors.
°Unable to be standardized.
Confidence intervals not provided and unable to calculate based on given information.
External Review Draft
4-116
DRAFT: Do not cite or quote
-------
Table 4-6 Animal toxicological studies of Pb exposure and blood pressure/hypertension.
Species Timinn of ExDosure Details
Study (Stock/Strain), Exposgure (Concentration, Duration) BLL As Rep°rted (Mg/dL) Endpoints Examined
Fioresi et al. Rat (Wistar) Age 2 mo to 3 mo 100 ppm Pb acetate in drinking <0.5 [jg/dL for control SBP, measured weekly from 0 to
(2014) Control (tap water for 30 d 4wk
water), M, 13.6 ± 1.07 [jg/dL for MAP, DBP, ACE activity measured
n = 9-12 100 ppm group post 30-d exposure
100 ppm
group, M,
n = 9-12
Gaspar and
Cordellini (2014)
Rat (Wistar)
Control (tap
water), M,
n = 15
8500 ppm
group, M,
n = 20
In utero to PND 22
Pregnancy day 0: females
divided into tap water and
500 ppm Pb acetate in drinking
water groups. Exposure lasted
through pregnancy. At birth,
pups were exposed to Pb (or
control) through nursing. Pups
were weaned at 22 d
<5 [jg/dL at all times for
tap water
19.98 ±6.31 - PND 52
13.15 ± 0.97 - PND 70
11.17 ± 2.11 -PND 100
SBP weekly measurements starting
at PND 23 to PND 100
Nunes et al. (2015) Rat (Wistar) 2 mo old rats 100 ppm Pb acetate in drinking NR for control SBP measured post 30 d exposure
Control exposed for 30 d water for 30 d for 28 d
(Distilled 8.4 [jg/dL for 100 ppm
water),
M, n = 5
100 ppm Pb
acetate group,
M, n = 5
External Review Draft
4-117
DRAFT: Do not cite or quote
-------
Study
Species
(Stock/Strain),
n, Sex
Timing of
Exposure
Exposure Details
(Concentration, Duration)
BLL As Reported (|jg/dL)
Endpoints Examined
Silva etal. (2015)
Rat (Wistar)
Control
(distilled water)
M, n = 6
3 mo old rats 100 ppm Pb acetate in drinking 12.3 ± 2 |jg/dL
exposed for 15 d water for 15 d
SBP measured weekly
100 ppm Pb
acetate group
M, n = 6
Wildemann et al.
(2015)
Rat (Wistar)
Control (tap
water), M,
n =6
357 or
1607 |jg/kg
BW/d, M, n = 5
per group
Unknown start age
for 4 wk
Tap water with 0.2% nitic acid,
or 357 or 1607 |jg/kg BW/d Pb
acetate in drinking water for
4 wk
1.4 ± 1.2 |jg/L for tap water/
0.2% nitic acid
(0.14 ± 0.12 Mg/dL)
17 ± 7 |jg/L for 357 [jg/kg
BW/d Pb acetate
(1.7 ± 0.7 Mg/dL)
86 ±29 [jg/L for 1607 pg/kg
BW/d Pb acetate
(8.6 ±2.9 pg/dL)
DBP, PP, SBP all measured post 4-
wk exposure
Xu et al. (2015)
Rat (Sprague
6-7-wk-old rats
Distilled water for 40 d or 1%
Day 12:
DBP, SBP, measured intermittently
Dawley)
exposed for 12 or
Pb acetate in drinking water
193.3 pg/L (19.33 pg/dL)
from 0-40 d
Control 1, M/F,
40 d
for 12 or 40 d
n =6
Day 40: 245.9 pg/L
(24.59 pg/dL)
1% PB acetate
group 12 d,
M/F, n = 15
1% PB acetate
group 40 d,
M/F, n = 15
External Review Draft
4-118
DRAFT: Do not cite or quote
-------
Study
Species
(Stock/Strain),
n, Sex
Timing of
Exposure
Exposure Details
(Concentration, Duration)
BLL As Reported (|jg/dL)
Endpoints Examined
Shvachiv et al.
(2018)
Rat (Wistar) In utero to 28 wk Pregnant Wistar rats were given 18.8 ± 2.0 [jg/dL for
Control, M/F,
n = 8
0.2% Pb
acetate,
intermittent Pb
group, M/F,
n = 9
0.2% Pb acetate in drinking
water or tap water. After a 21 d
weaning period, pups were
either continuously exposed to
Pb acetate in drinking water
until 28 wk or were given 8 wk
of Pb abstinence and then
exposed until 28 wk with Pb
acetate
intermittent exposure
24.4 ± 4.9 [jg/dL for
continuously exposed
DBP, MAP, Baroreceptor Reflex,
Chemoreceptor Reflex, SBP,
all measured 2-hr post 28-wk
exposure
0.2% Pb
acetate,
permanent Pb
group, M/F,
n = 9
Zhu et al. (2018)
Rat (Sprague
Dawley)
Control
(distilled
water), M,
n = 10
In utero to 1 yr Female rats were given either 0
or 500 mg/L Pb acetate for 10 d
before mating. Male offspring
continued receiving 0 or
500 mg/L Pb acetate for 1 yr
0.28 ± 0.02 mg/L
(28 ± 2 [jg/dL)
SBP, DBP measured post 1 -yr
exposure
0.5 g/L Pb
acetate, M,
n = 10
External Review Draft
4-119
DRAFT: Do not cite or quote
-------
Study
Species
(Stock/Strain),
n, Sex
Timing of
Exposure
Exposure Details
(Concentration, Duration)
BLL As Reported (|jg/dL)
Endpoints Examined
Zhu etal. (2019)
Rat (Sprague
Dawley)
Control
(distilled
water), M,
n = 100
0.5 g/L Pb
acetate, M,
n = 10
In utero to 1 yr Female rats were given either 0
or 0.5 g/L Pb acetate for 10 d
before mating. Male offspring
continued receiving 0 or 0.5 g/L
Pb acetate for 1 yr
0.27 ± 0.02 mg/L
(27 ± 2 pg/dL)
SBP, DBP measured post 1 -yr
exposure
ACE = angiotensin-converting enzyme; BLL = blood lead level; BW = body weight; d = day(s); DBP = diastolic blood pressure; F = female; M = male; MAP = mean arterial pressure;
NR = not reported; Pb = lead; PND = postnatal day; PP = pulse pressure; SBP = systolic blood pressure; yr = year(s); wk = week(s).
External Review Draft
4-120
DRAFT: Do not cite or quote
-------
Table 4-7
Epidemiologic studies of Pb exposure and coronary and ischemic heart disease.
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Jain et al. (2007)
Boston, MA
1991-2001
Cohort
NAS
n = 837
elderly men
(mostly white)
Average
individual
born -1933
Blood (GFAAS with Zeeman
correction) (pg/dL)
Non-cases
Mean (SD): 6.2 (4.3)
Range: 0 to 35
Cases
Mean (SD): 7.0 (3.8)
Range: 1.0 to 20.0
Bone (K-XRF) (pg/g)
Patella
Non-cases
Mean (SD): 30.6 (19.7)
Range: -10 to 165
Cases
Mean (SD): 36.8 (20.8)
Range: 5.0 to 101
Tibia Pb
Non-Cases
Mean (SD): 21.4 (13.6)
Range: -3 to 126
Cases
Mean (SD): 24.2 (15.9)
Range: -5 to 75
IHD
(Ml or angina
pectoris)
Cox proportional hazards models
adjusted forage, BMI, education,
race, smoking status, pack-years
smoked, alcohol intake, history of
diabetes mellitus and hypertension,
family history of hypertension, DBP,
SBP, serum triglycerides, serum
HDL, and total serum cholesterol
BLL >5 pg/dLb
Per 1 SD increase in Pb biomarker
OR over 10-yr follow-up:
1.73 (1.05, 2.87)
Ln (blood Pb)
OR: 1.45 (1.01,
2.06)
Ln (patella Pb)
OR: 2.64 (1.09, 6.37)
Ln (tibia Pb)
OR: 1.84 (0.57, 5.90)
Age at measurement
Mean 67
External Review Draft
4-121
DRAFT: Do not cite or quote
-------
Study "oesfg"" Population Exposure Assessment Outcome
Confounders
Effect Estimates and 95% Clsa
Ding etal. (2016)
Boston
1991 through 2011
participants followed
up to 20 yr
Cohort
No minor allele in HFE
rs1799945 (H63D) 1.41 (1.15,
1.73)
rs1800562 (C282Y) 1.36 (1.13,
1.64)
No minor allele HMOX1
rs2071746 1.51 (1.07, 2.14)
rs5995098 1.63 (1.23, 2.14)
At least one minor allele in HMOX1
rs2071749^.b'\ (1.22, 1.86)
No minor allele in APOE
rs429358 1.43(1.17, 1.76)
rs449647 1.29(1.05, 1.60)
rs7412 1.34 (1.10, 1.64)
At least one minor allele in APOE
rs7412 1.53 (1.07, 2.19)
No minor allele in AGT
rs699 2.17 (1.50, 3.12)
rs5046 1.53 (1.27, 1.94)
rs5050 1.36 (1.09, 1.69)
At least one minor allele in AGT
NAS
n = 589
Elderly men
(mostly white)
Average
individual
born -1935
Bone (K-XRF) (pg/g) Mean
(SD)
No CHD
Patella: 29.2 (16.1)
Tibia 20.2 (12.5)
CHD
Patella: 32.1 (18.8)
Tibia: 22.6 (13.5)
Age at measurement:
Mean: 66
Range: 48-96
Incident
Coronary Heart
Disease (Ml,
angina pectoris
or CHD deaths
Cox regression adjusted for age,
smoking status, BMI, and the ratio
of total cholesterol to
HDL-C level
HR (twofold increase in blood Pb)b
Bone (Patella) 1.36 (1.15, 1.61)
At least one minor allele in VDR
rs1544410 (Bsm1) 1.65 (1.31,
2.08)
rs731236 (Taq1) 1.61 (1.29, 2.02)
rs7975232 (Apa1) 1.28 (1.04, 1.57)
rs1073581 (Fok1) 1.47 (1.17, 1.83)
rs757343 (Tru91) 1.48 (1.18, 1.85)
At least one minor allele in ALAD
rs1833435 1.11 (0.79, 1.55)
External Review Draft
4-122
DRAFT: Do not cite or quote
-------
Study "oesfg"" Population Exposure Assessment Outcome
Confounders
Effect Estimates and 95% Clsa
Ding etal. (2019)
Boston, MA
United States
August 1991- June
2011
Mean (SE) 8.52
(5.75) yr of follow-up
Cohort
rs50501A1 (1.03, 1.94)
No minor allele in angiotensin II
receptor type 1
rs12695908 1.43 (1.20, 1.74)
No minor allele in glutathione S-
transferase pi 1
rs1695 1.39 (1.10, 1.76)
GRS 1 2.27 (1.50, 3.42)
GRS 2 2.77 (1.78. 4.31)
NAS
n = 594
elderly men
(mostly white)
without CHD
at baseline
Average
individual
born -1935
Bone (K-XRF) (pg/g) Mean
(SD)
CHD
Patella: 32.2 (18.9)
Tibia: 22.6 (13.5)
Non-CHD
Patella: 29.4 (18.9)
Tibia: 20.9 (13.2)
Age at measurement
Mean (SD)
CHD: 65.5 (6.2)
Non-CHD: 66.5 (7.5)
Incident
Coronary Heart
Disease (Ml,
angina pectoris
or CHD deaths
Cox proportional hazards adjusting
for BMI, total energy intake,
smoking status, TC to HDL ratio,
education level, and occupation
HR (95 % CI) for twofold increase
in Bone Pbb
Patella 1.30 (1.09, 1.56)
Tibia 1.25 (1.06, 1.48)
Prudent Diet
Patella
Low: 1.64 (1.27, 2.11)
High: 1.07 (0.86, 1.34)
Tibia
Low: 1.24 (0.96, 1.59)
High: 1.26 (1.02, 1.55)
Western Diet:
Patella
Low: 1.35 (1.05, 1.72)
High 1.27 (0.96, 1.61)
Tibia
Low: 1.43 (1.14, 1.80)
High: 1.08 (0.86, 1.34)
External Review Draft
4-123
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Tonelli etal. (2018) n = 1,278
Plasma Pb (ICP-MS) (pg/dL)
Deciles
Canada
Patients on
incident
hemodialysis
Participant recruited
between March 2005
and November 2012 Average
Cohort (year of
follow-up)
individual
born -1946
0.06
0.19
0.28
0.35
0.44
0.55
0.68
0.83
1.08
10: 1.74
Cardiovascular Logistic regression adjusting for
event (acute Ml, age, sex, race/ethnicity,
percutaneous unemployment prior to dialysis,
coronary year dialysis initiated, dialysis
angioplasty, duration, predialysis care,
coronary artery arteriovenous access, comorbidities
bypass grafting, (AF, Ml, BMI, cancer,
heart failure, cerebrovascular disease, CHF, lung
and stroke or disease, diabetes, dementia,
transient hypertension, liver disease, PVD,
ischemic attack) psychiatric disease, substance
misuse), albumin, and creatinine.
*AII variables were considered
candidate variables and were
included based on stepwise
regression results
ORb
Cardiovascular events: NR
Authors indicate a null relationship
between blood Pb deciles and all-
cardiovascular events, results not
reported
Cho et al. (2016)
KNHANES
Blood (GFAAS with Zeeman
>10% increase Logistic regression adjusted for
ORb (95%
CI):
n = 5,361
correction) (pg/dL)
in 10-yrCHD BMI, triglycerides, and LDL-C
Males
South Korea
Participants in
KNHANES
Males
Mean (SE): 2.81 (0.32)
Risk (FRS)
based on age,
gender, SBP,
Q2 vs. Q1
Q3 vs. Q1
1.59 (1.03, 2.46)
2.31 (1.52, 3.50)
2008-2019
aged 20-70 y
Q1: 0.71-2.13
total cholesterol,
Q4 vs. Q1
3.13 (2.9, 4.69)
Q2: 2.13-2.70
and HDL-C)
Females
Cross-sectional
Average
Q3: 2.70-3.52
Q2 vs. Q1
1.84 (0.61, 5.55)
individual
Q4: 3.52-26.51
Q3 vs. Q1
1.43 (0.44, 4.59)
born -1973
Females
Q4 vs. Q1
0.88 (0.26, 2.97)
Mean (SE): 2.04 (0.02)
Q1
Q2
Q3
Q4
0.42-1.49
1.49-1.95
1.95-2.51
2.51-9.59
Age at measurement (years):
Mean (SE)
Males: 39.3 (0.30)
External Review Draft
4-124
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Females: 40.9 (0.30)
Choi etal. (2020)
KNHANES
Blood (GFAAS with Zeeman
10-yr
Multiple linear regression analysis
Linear increase in 10-yr ASCVD
n = 2,424
correction) (pg/dL)
atherosclerotic
adjusting for age, income, job type,
risk score (Q4 vs. Q1 )b c
Korea
cardiovascular
physical activity, location, and sleep
Participants in
Pb distribution NR
disease
Males: 0.117 (0.01, 0.23)
2016-2017
KNHANES
aged 40-80 yr
(ASCVD) risk
Urban: 0.133 (0.01, 0.25)
Rural: 0.079 (-0.15, 0.31)
Cross-sectional
Average
individual
born -1956
<7 hr sleep: 0.097 (-0.03, 0.23)
>7 hr sleep: 0.183 (0.01, 0.36)
Females: 0.072 (-0.00, 0.15)
Urban: 0.038 (-0.05, 0.12)
Rural: 0.212 (0.05, 0.38)
<7 hr sleep: 0.110 (0.016, 0.20)
>7 hr sleep: 0.021 (-0.09, 0.13)
Park and Han (2021)
KNHANES
Blood Pb (GFAAS with
10-20% and
Logistic regression adjusted for
OR (<10% increase in 10-yr CVD
n = 1,929
Zeeman correction (pg/dL)
>20% increase
SBP, HDL cholesterol, and total
risk as referent)bcd
South Korea
Distribution: NR
in 10-yr CVD
risk estimated
cholesterol
Males
>20 y
using FRS
10-20% 10 yr CVD risk (vs. <10%):
KNHANES VII-1
2.407 (1.885, 3.075)
(2017)
Average
>20% 10 yr CVD risk (vs. <10%):
2.847 (2.020, 4.011)
individual
Cross-sectional
born in or
before -1997
Females
10-20% 10 yr CVD risk (vs. <10%):
1.051 (0.676, 1.633)
>20% 10 yr CVD risk (vs. <10%):
0.706 (0.188, 2.659)
External Review Draft
4-125
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Nguyen et al. (2021)
South Korea
KNHANES IV (2009),
V (2010-2012), VI
(2013), and VII
(2016-2017)
Cross-sectional
KNHANES
n = 9,602
>20 y
Average
individual
born -1965
Serum Pb (GFAAS with
Zeeman correction (pg/dL)
GM (95% CI)
2.02 (2.00-2.03)
Age at measurement
Mean (SD)
Males: 47.76 (15.25)
Females: 46.87 (15.16)
10-yr CVD risk Multivariable models adjusted for
estimated using serum cotinine, age group, sex,
FRS high-risk drinking, physical activity,
BMI, family history of CVDs,
diabetes or dyslipidemia, and type
2 diabetes.
*Assume linear regression, but not
specified
Linear increase in 10-yr CVD risk
score (log2 transformed blood Pb)b
0.104 (0.016, 0.214)
Ochoa-Martinez et al. Women living Blood Pb (GFAAS) (pg/dL)
(2018)
San Luis Potosi
Mexico
2015-2016
Cross-sectional
in
communities
with a high-
risk of
environmental
Pb
contamination
Average
individual
born -1967
mean (SD) 11.5 (9.0)
Tertiles
T1
T2
T3
<3.5
3.6-9.0
>9.1
Age at measurement:
mean (SD): 48.5 (18.0)
Predictive CVD
biomarkers
[asymmetric
dimethylarginine
(ADMA),
FABP4,
adiponectin,
and chemerin]
Linear regression controlling for
age, weight, waist circumference,
hip circumference, SBP, DBP, BMI,
body fat %, visceral fat %, glucose,
triglycerides, total cholesterol, HDL-
C, LDL-C
Predictive CVD biomarkersb
ADMA (pmol/L):
T2: 0.51 (-0.25, 0.69)
T3: 0.75 (0.15, 1.85)
FABP4 (ng/mL):
T2: 11.0 (-15.0, 16.0)
T3: 27.5 (10.0, 34.5)
Adiponectin (|jg/mL):
T2: 9.50 (-17.0, 21.0)
T3: 12.5 (-7.5-, 18.0)
Chemerin (ng/mL):
T2: 195 (-75.0, 275)
T3: 220 (-25, 300)
External Review Draft
4-126
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Wan etal. (2021)
China
May-August 2018
Cross-sectional
Environmental Blood (AAS) (pg/dL)
Pollutant
Exposure and
Metabolic
Diseases in
Shanghai
(METAL
study)
n = 4,234
Median (IQR):
2.6 (1.8-3.6)
Age at measurement
Median (IQR):
67 (62-72) yr
Presence of
CVD (Self-
reported
diagnosis by a
physician,
including CHD,
Ml or stroke)
Linear or logistic regression
adjusting for age, sex, duration of
diabetes, education status, current
smoking, BMI, HbA1c,
dyslipidemia, hypertension
OR (95% CI) (4th vs. 1st quartile of
Blood Pb)b
1.44 (1.17, 1.76)
Average
individual
born -1951
AAS = atomic absorption spectrometry; ADMA = asymmetric dimethylarginine; AF = atrial fibrillation; AGT = angiotensinogen; ALAD = 6-aminolevulinic acid dehydratase;
APOE = apolipoprotein E; ASCVD = atherosclerotic cardiovascular disease; BLL = blood lead level; BMI = body mass index; C282Y HFE = mutant of the HFE wildtype;
CHD = congenital heart disease; CHF = congestive heart failure; CI = confidence interval; CVD = cardiovascular disease; DBP = diastolic blood pressure; FABP4 = adipocyte fatty
acid-binding protein 4; FRS = Framingham risk score; GFAAS = graphite furnace atomic absorption spectrometry; GRS = genetic risk score; HbA1c = hemoglobin A1C; HDL-
C = high-density lipoprotein cholesterol; HFE = hemochromatosis gene; HMOX1 = heme oxygenase-1; HR = hazard ratio; ICP-MS = inductively coupled plasma mass spectrometry;
IHD = ischemic heart disease; IQR = interquartile ratio; KNHANES = Korean National Health and Nutrition Examination Survey; K-XRF = K-Shell X-ray fluorescence; LDL-C = low-
density lipoprotein cholesterol; METAL = Environmental Pollutant Exposure and Metabolic Diseases in Shanghai ; Ml = myocardial infarction; NR = not reported; OR = odds ratio;
Pb = lead; PVD = peripheral vascular disease; Q = quartile; SBP = systolic blood pressure; SD = standard deviation; SE = standard error; T = fertile; TC = total cholesterol;
VDR = vitamin D receptor; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
Confidence intervals estimated from standard error.
increment unclear.
External Review Draft
4-127
DRAFT: Do not cite or quote
-------
Table 4-8
Epidemiologic studies of Pb exposure and cardiac function.
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Schwartz (1991)
NHANES II
Blood- Method NR
Left ventricle
Logistic regression adjusting
OR
n = 9,932
hypertrophy
for age, sex, and race
1.028 (1.009, 1.048)
United States
Distribution NR
(based on
body habitus,
Average
BMI, and tricep
1976-1980
individual born
skinfold)
-1931
Cross-sectional
Yang etal. (2017)
Belgium
baseline 1985-
2000; follow-up:
2005-2010
Median follow-up
11.9 yr
Cohort
Cadmium in
Belgium study
n = 179
Average
individual born
-1953
Blood (ETAAS with Left ventricle
Zeeman correction) (pg/dL) structure and
GM 4.14 function
Age at measurement
Mean: 39.1
Linear regression adjusting
for measures at the time of
the echocardiography
including age, sex, MAP,
heart rate, BMI, fasting
plasma glucose, total to HDL
cholesterol ratio, serum
creatinine, y-
glutamyltransferase,
smoking, and
antihypertensive medication
class
For each doubling of blood Pbbc
LV Structure
LVMI, g/m2-1.399 (-4.504,1.707)
End diastolic diameter, cm -0.064 (-0.134,
-0.006)
RWT 0.0065 (-0.0031, 0.0162)
Systolic LV Function
Ejection fraction, % 0.190 (-1.293, 1.675)
GLS, % -0.497 (-0.957, -0.038)
RLS, % -0.784 (-1.482, -0.087)
RLS rate, (s-1) -0.071 (-0.124, -0.019)
RRS, % -2.316 (-4.748, -0.115)
RRS rate, (s-1) -0.135 (-0.292, 0.022)
Diastolic LV Function
E peak, cm/s 1.308 (-1.120, 3.736)
E/A ratio -0.036 (-0.085, 0.014)
e' peak, cm/s -0.188 (-0.494, 0.118 (E/e'
ratio 0.172 (-0.133, 0.477)
External Review Draft
4-128
DRAFT: Do not cite or quote
-------
Study DCesfgnnd Population Exposure Assessment Outcome
Confounders
Effect Estimates and 95% Clsa
Lindetal. (2012) Prospective Blood (ICP-MS) (pg/dL)
Investigation of Median (IQR): 1.72 (1.22,
Uppsa,a, Sweden Z28)
Seniors (PIVUS)
Cross-sectional study
Left ventricle Linear regression adjusting
structure for sex, blood pressure,
antihypertensive medication,
diabetes, and BMI
Per In-transformed unit increase in serum
Pbb
LVMI, g/m2 -0.73 -(2.20, 0.74)
RWT 0.011 (-0.001, 0.022)
n = 993
elderly (70 yr)
individuals
Chen et al. (2021) n = 486
Guangdong
province
China
2018
Cross-sectional
Preschool
children (aged
2-6) from two
towns with
similar SES but
different Pb
exposure
Average
individual born
-2013
Blood (GFAAS) (pg/dL)
Median (IQR):
Exposed: 4.51 (3.70-5.67)
Reference: 3.98 (3.25-
4.84)
Age at measurement
Mean (SD):
Exposed: 4.74 (0.84)
Reference: 4.75 (1.01)
Left ventricle Linear regression adjusted
structure and for gender, age, BMI, e-
function waste contamination w/ in
50 m of residence, residence
as workplace, distance of
residence from road, family
member daily smoking,
monthly household income,
maternal work associated
with e-waste, duration of
outdoor play, child contact
with e-waste, washing hands
before eating, nail biting
habit, chewing pencil habit,
yearly canned food
consumption, yearly
fruit/vegetable consumption,
yearly iron rich food
consumption, yearly marine
product consumption, and
yearly salted food
consumption
Ln-transformed parameters per one-unit
increase in blood Pbb
IVS, cm -0.004 (-0.007, 0.001)
LV posterior wall, mm-0.001 (-0.003,
0.001)
Ejection fraction, % 0.001 (-0.002, 0.001)
External Review Draft
4-129
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Gump et al. (2005) Oswego
Oswego, NY (born
at a single hospital
in New York from
1991-94)
Cohort
Children's Study
n = 122 children
aged 9.5 yr
Average
individual born
-1990
Cord blood (ETAAS)
(Hg/dL)
Mean (SD): 2.97 (1.75)
Child blood (ETAAS and
ASV) (pg/dL)
Mean (SD): 4.62 (2.51)
Age of child blood Pb
measurement
Mean 2.6
Stroke volume, Multivariate linear regression Cord BLL
cardiac output
adjusted for HOME score,
SES, birth weight, child BMI,
child sex
No association, results not reported
Childhood BLL
Stroke volume, mL -0.069 (-0.124, -0.015)
Cardiac output, L/min -0.056 (-0.113,
0.001)
Gump et al. (2011) n = 140 children Blood (ICP-MS) (pg/dL)
ages 9-11 yr
Oswego, NY
Cross-sectional
GM
1.01
Q1:
0.14-0.68
Q2:
0.69-0.93
Q3:
0.94-1.20
Q4:
1.21-3.76
Stroke volume, Linear regression models
cardiac output adjusted for sex, SES, BMI,
and age
Change in Stroke Volume (%) across
quartiles in response to acute stressbd
Q1: 2.23, Q2: 0.91, Q3: -3.47, Q4: -0.89,
p for trend = 0.04
Change in Cardiac Output (%) across
quartiles in response to acute stressbd
Q1: 3.26, Q2: 1.19, Q3: -2.31, Q4: -0.20,
p for trend = 0.05
A = peak late diastolic velocity; ASV = anodic stripping voltammetry; BLL = blood lead level; BMI = body mass index; E = peak early diastolic velocity; e' = peak early diastolic mitral
annular velocity; ETAAS = electrothermal atomic absorption spectrometry; GFAAS = graphite furnace atomic absorption spectrometry; GLS = global longitudinal strain;
GM = geometric mean; GSD = geometric standard deviation; HDL = high-density lipoprotein; HOME = Health Outcomes and Measures of the Environment Study; ICP-
MS = inductively coupled plasma mass spectrometry; IQR = interquartile ratio; IVS = interventricular septum; LV = left ventricular; LVMI = left ventricular mass index; MAP = mean
arterial pressure; NHANES = National Health and Nutrition Examination Survey; NR = not reported; OR = odds ratio; Pb = lead; PIVUS = Prospective Investigation of the Vasculature
in Uppsala Seniors; RLS = regional longitudinal strain; RRS = regional radial strain; RWT = relative wall thickness; SD = standard deviation; SES = socioeconomic status;
Q = quartile.
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
°Corrected for regression dilution bias using quintile method.
Confidence intervals not provided and unable to calculate based on given information.
External Review Draft
4-130
DRAFT: Do not cite or quote
-------
Table 4-9 Animal toxicological studies of cardiac function.
Study Species (Stock/Strain),
n, Sex
Timing of
Exposure
Exposure Details
(Concentration,
Duration)
BLL As Reported (|jg/dL)c
Endpoints
Examined
Wildemann et al. (2015) Rat (Wistar)
Control (tap water), M,
n = 6
Unknown start age
for 4 wk
Tap water with 0.2% nitic
acid, or 357 or 1607 pg/kg
BW/d Pb acetate in
drinking water for 4 wk
1.4 ± 1.2 pg/L for tap water/
0.2% nitic acid
(0.14 ± 0.12 pg/dL)
Stroke volume and
cardiac output post
4-wk exposure
357 or 1607 pg/kg BW/d,
M, n = 5 per group
17 ± 7 pg/L for 357 pg/kg
BW/d Pb acetate
(1.7 ± 0.7 pg/dL)
86 ±29 pg/L for 1607 pg/kg
BW/d Pb acetate
(8.6 ±2.9 pg/dL)
Silvaetal. (2015) Rat (Wistar)
Control (distilled water)
M, n = 6
3 mo old rats
exposed for 15 d
100 ppm Pb acetate in
drinking water for 15 d
12.3 ± 2 pg/dL
Force generation in
LV papillary muscle
following pulse
stimulation post 15-d
exposure
100 ppm Pb acetate group
M, n = 6
Time to peak tension
and 90% relaxation
post 15-d exposure
Inotropic force
following calcium or
isoproterenol
stimulation post 15-d
exposure
External Review Draft
4-131
DRAFT: Do not cite or quote
-------
Study
Species (Stock/Strain), Timing of
n, Sex Exposure
Exposure Details
(Concentration,
Duration)
BLL As Reported (|jg/dL)c
Endpoints
Examined
Fioresi et al. (2014)
Rat (Wistar)
Control (tap water), M,
n = 9-12
100 ppm group, M, n = 9-
12
Age 2 mo to 3 mo
100 ppm Pb acetate in
drinking water for 30 d
<0.5 [jg/dL for control
13.6 ± 1.07 [jg/dL for
100 ppm group
LVSP and RVSP, left
and right diastolic
pressure all
measured post 30 d
exposure
Isometric contraction
force, time to peak
contraction, and
relaxation rates in LV
papillary muscle post
30 d exposure
Contractile force
following calcium
treatment in LV
papillary muscle 30 d
post exposure
BW = body weight; d = day(s); LV = left ventricular; LVSP = right ventricular systolic pressure; M = male; mo = month(s); Pb = lead; RVSP = right ventricular systolic pressure;
wk = week(s).
External Review Draft
4-132
DRAFT: Do not cite or quote
-------
Table 4-10 Animal toxicological studies of Pb exposure and endothelial dysfunction.
st . Species (Stock/Strain), n, Timing of Exposure Details BLL As Reported Endpoints
y Sex Exposure (Concentration, Duration) (Mg'dL) Examined
Gaspar and Cordellini
(2014)
Rat (Wistar)
Rat (Wistar)
Control (tap water), M,
n = 8
500 ppm group, n = M, 6
In utero to PND 22
Pregnancy day 0: females
divided into tap water and
500 ppm Pb acetate in drinking
water groups. Exposure lasted
through pregnancy. At birth,
pups were exposed to Pb (or
control) through nursing. Pups
were weaned at 22 d.
<5 [jg/dL at all time
points for tap water
19.98 ± 6.31 -
PND 52
13.15 ± 0.97-
PND 70
Vascular reactivity
in aortic rings post
exposure at
PND 23, 52, 70,
and 100
11.17 ± 2.11
PND 100
Nunes et al. (2015)
Rat (Wistar)
Control (distilled water), M,
n = 16
2-mo-old rats
exposed for 30 d
100 ppm Pb acetate in drinking
water for 30 d.
NR for control
8.4 |jg/dL for
100 ppm
Vascular reactivity
in aortic rings
measured post 30-
d exposure
100 ppm Pb acetate
treatment, M, n = 16
5-8 rats from control or
100 ppm group for other
treatments (e.g.,
phenylephrine in control or
Pb-treated mice)
BLL = blood lead level; d = day(s); M = male; Pb = lead; PND = postnatal day; NR = not reported.
External Review Draft
4-133
DRAFT: Do not cite or quote
-------
Table 4-11
Epidemiologic studies of Pb exposure cardio electrophysiology and arrythmia.
Reference and
Study Design
Study Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Cheng et al.
(1998)
Boston, MA
NAS
n = 750
Elderly men (mostly
white)
NAS August 1991 Average individual
and December born -1925
1995
Cross-sectional
Blood Pb (GFAAS with
Zeeman correction)
(Hg/dL)
Mean (SD): 5.79 (3.44)
Bone (K-XRF) (pg/g)
Mean (SD):
Patella: 30.82 (19.19)
Tibia: 22.9 (13.36)
Age at measurement
Mean (SD): 67.81 (7.27)
ECG Linear regression adjusted for
conduction age and DBP. Model forQTc
(QTc, QRSc) additionally adjusted for
alcohol intake and BMI. Model
for QRSc additionally adjusted
for fasting glucose level.
QTc (msec)
Bone (Patella) Pb
<65 yr: 3.00 (0.16, 5.84)
>65 yr: 0.39 (-1.05, 1.83)
Tibia Pb
<65 yr: 5.03 (0.83, 9.22)
>65 yr: 1.41 (-0.67, 3.49
QRSc (msec)
Patella Pb
<65 yr: 2.23 (0.10, 4.36)
>65 yr: -0.11 (-1.07, 0.85)
Tibia Pb
<65 yr: 4.83 (1.83, 7.83)
>65 yr: -0.83 (-2.21, 0.56)
No association with blood Pb
Eum et al. (2011)
Boston, MA
NAS 1989 and
1996
Cohort
NAS
n = 600
Elderly men (mostly
white)
Average individual
born -1925
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Mean (SD): 5.8 (3.6)
Bone (K-XRF) (pg/g)
Mean (SD):
Patella: 30.3 (17.7)
Tibia: 21.6 (12.0)
Tibia Quartiles:
ECG Linear regression adjusted for
conduction age, education, smoking, BMI,
(QTc, QRSc) albumin-adjusted serum Ca2+,
and diabetes status at
baseline, years between ECG
tests, and QT-prolongation
drugs at the time of ECG
measurement.
Bone (Tibia) Pbb
Adjusted 8-yr change
QTc(Q1 reference)
Q1: 7.49 (1.22, 13.75) msec
Q3: 7.94 (1.42, 14.45) msec
p for trend = 0.03
QRSc (Q1 reference)
Q2: 0.52 (-3.60, 4.65) msec
Q1:
<16
p for trend = 0.005
Q2:
16.0-23
Q3:
CO
CM
A
External Review Draft
4-134
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Age at measurement:
Mean 67
No associations with patella
or blood Pb
Park et al. (2009)
Boston, MA
NAS August 1991
and December
1995
Cross-sectional
NAS
n = 613
Elderly men (mostly
white)
Average individual
born -1925
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Median (IQR): 5 (4-7)
Bone (K-XRF) (pg/g)
Median (IQR)
Patella: 26 (18-37)
Tibia: 19 (14-27)
QTc interval Linear regression models
adjusted forage, BMI, smoking
status, serum Ca2+, and
diabetes. No SES indicator
was considered.
QTc interval (msec)
0.433 (-0.253, 1.12)
Patella Pb:
1.389 (0.068, 2.711)
Tibia Pb:
2.192 (0.227, 4.158)
Age at measurement
Mean 67
Jinq et al. (2019)
United States
NHANES III
1988-1994
Cross-sectional
NHANES
n = 7,179
Participants without
self-reported history of
Ml (or ECG results
indicating Ml), without
a history of CHF
Average individual
born -1934, -1936,
-1932, -1935
Blood (GFAAS) (pg/dL)
GM
Men: 4.10
Women: 2.93
Age at measurement:
Mean (SD)
Men
T1
T2
T3
57.12 (0.51)
55.42 (0.53)
57.00 (0.65)
Women
T1
T2
T3
59.01 (0.64)
55.88 (0.62)
59.45 (0.98)
Ventricular
arrhythmia
(QRS-T angle)
Spatial QRS-T
angle
estimated
using a 12-Pb
ECG.
Multivariate weighted logistic
regression adjusting for
impaired fasting glucose,
hypertension, poverty index,
age, race, and smoking status.
One-unit increase in log of
blood Pbb
OR (95% CI) (3rd vs. 1st
fertile)
Men: 1.35 (1.05, 1.74)
Women: 1.05 (0.82, 1.36)
External Review Draft
4-135
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Park et al. (2006) NAS
n = 413
Boston, MA
2000-2004
Cross-sectional
Elderly men (mostly
white)
Average individual
born -1935
Bone (K-XRF) (pg/g)
Median (IQR)
Tibia: 19 (11-28)
Patella (measured within
6 mo of HRV):
23 (15-34)
Estimated Patella
(accounting for time
difference):
16.3 (10.4-25.8)
Age at measurement:
Mean: 67
HRV
Linear regression models
adjusted for age, cigarette
smoking, alcohol consumption,
room temperature, season
(model 2) BMI, fasting blood
glucose, HDL-C, triglyceride,
use of p-blockers, Ca2+
channel blockers, and/or ACE
inhibitors. No SES indicator
was considered.
HRV
Tibia
HF: -0.529 (-2.265, 1.206)
nu
LF: 0.529 (-1.206, 2.265) nu
Log LF/HF:
1.941 (-6.941, 10.824) %
Corrected Patella
HF: -0.39 (-2.013, 1.234) nu
LF: 0.39 (-1.234, 2.013) nu
Log LF/HF:
1.948 (-6.136, 10.032)
Effect estimates were more
pronounced among those
with greater# metabolic
abnormalities.
Gump et al.
(2011)
Oswego, NY
Cross-sectional
n = 140
Children aged 9-11 yr
Blood (ICP-MS) (pg/dL)
GM: 1.01
Heart rate Linear regression adjusted for
sex, SES, BMI, and age.
Q1
Q2
Q3
Q4
0.14-0.68
0.69-0.93
0.94-1.20
1.21-3.76
Change in heart rate
(beats/min) across quartiles
in response to acute stressbc
Q1: 0.91, Q2: 0.19, Q3: 0.86,
Q4: 0.58, p for trend = 0.85
No association between Pb
levels and baseline
cardiovascular levels
Gump et al.
(2017)
New York
Environmental
Exposures and Child
Health Outcomes
study
n = 203
Children aged 9-11
Blood (ICP-MS) (pg/dL) Heart rate
Mean (SD): 0.98 (0.61) variability
Range: 0.19-3.25
Linear regression adjusted for
sex, race, age, and SES.
No association between BLLs
and HRV. Results not shown
(p > 0.25)
External Review Draft
4-136
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study Population
Exposure Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Environmental
Exposures and
Child Health
Outcomes
Halabickv et al.
(2022)
Jintan, China
Children aged 3-5
in 2004-2005
wave, aged 11-12
in 2011-2013
wave
Cohort
China Jintan Child
Cohort Study
n = 408
Average individual
born -2000
Whole blood (GFAAS)
(Hg/dL)
Median (IQR)
3-5 yr: 6.4 (4.9-8.0)
11-13 yr: 2.9(2.3-3.6)
Age at measurement:
3-5 and 11-13 yr
HRV in
children during
a stress test
Age at
outcome -12
General linear models
adjusted for parental
occupation, child sex, serum
Fe, crowded neighborhood
*No baseline HRV
measurement and no
adjustment for BMI or physical
activity. Serum Fe less precise
measure of iron status
opposed to ferritin or
transferrin.
Change in planning phase
HRV per log-transformed
increase in blood Pbb
3-5 yr Pb 0.03 (-0.02, 0.09)
12 yr Pb -0.04 (-0.16, 0.07)
Change in speaking phase
HRV per log-transformed
increase in blood Pbb
3-5 yr Pb: 0.06 (0.01, 0.12)
12 yr Pb: -0.05 (-0.18, 0.08)
ACE = angiotensin-converting enzyme; BLL = blood lead level; BMI = body mass index; Ca2+ = calcium ion; CHF = congestive heart failure; CI = confidence interval;
DBP = diastolic blood pressure; ECG = electrocardiogram; GFAAS = graphite furnace atomic absorption spectrometry; GM = geometric mean; HDL = high-density lipoprotein;
HF = high-frequency power in normalized units; HRV = heart rate variability; ICP-MS = inductively coupled plasma mass spectrometry; IQR = interquartile ratio; K-XRF = K-Shell
X-ray fluorescence; LF = low-frequency; Ml = myocardial infarction; mo = month(s); NAS = Normative Aging Study; NHANES = National Health and Nutrition Examination Survey;
nu = normalized units; OR = odds ratio; Pb = lead; Q = quartile; QRSc = corrected QRS duration; QTc = corrected QT interval; SD = standard deviation; SES = socioeconomic
status; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
Confidence intervals not provided and unable to calculate based on given information
External Review Draft
4-137
DRAFT: Do not cite or quote
-------
Table 4-12 Animal toxicological studies of Pb exposure and cardiac electrophysiology.
Species
Study (Stock/Strain), n,
Sex
Timing of
Exposure
Exposure Details
(Concentration, Duration)
BLL 0JSg/dL)Orted EndP°ints Examined
Fioresi et al. (2014)
Rat (Wistar) Age 2 mo to
Control (tap water), 3 mo
M, n =9-12
100 ppm group, M,
n = 9-12
100 ppm Pb acetate in drinking water for 30 d
<0.5 [jg/dL for
control
13.6 ± 1.07 [jg/dL
for 100 ppm group
Heart rate post 30-d
exposure
Wildemann et al.
(2015)
Rat (Wistar)
Control (tap water),
M, n = 6
357 or 1607 [jg/kg
BW/d, M, n = 5 per
group
Unknown start Tap water with 0.2% nitic acid, or 357 or
age for 4 wk 1607 [jg/kg BW/d Pb acetate in drinking water for
4 wk
1.4 ± 1.2 |jg/L for
tap water/ 0.2%
nitic acid
(0.14 ± 0.12 Mg/dL)
17 ± 7 |jg/L for
357 [jg/kg BW/d
Pb acetate
(1.7 ± 0.7 Mg/dL)
Heart rate measured
baseline and post 4-wk
exposure
PR and QRS interval
post 4-wk exposure
86 ± 29 |jg/L for
1607 [jg/kg BW/d
Pb acetate
(8.6 ± 2.9 [jg/dL)
External Review Draft
4-138
DRAFT: Do not cite or quote
-------
Species
Study (Stock/Strain), n,
Sex
Timing of
Exposure
Exposure Details
(Concentration, Duration)
BLL As Reported
(pg/dL)
Endpoints Examined
Shvachiv et al. (2018)
Rat (Wistar)
Control, M/F, n = f
0.2% Pb acetate,
intermittent Pb
group, M/F, n = 9
In utero to Pregnant Wistar rats were given 0.2% Pb acetate
28 wk in drinking water or tap water. After a 21-d
weaning period, pups were either continuously
exposed to Pb acetate in drinking water until 28 wk
or were given 8 wk of Pb abstinence and then
exposed until 28 wk with Pb acetate
18.8 ±2.0 [jg/dL
for intermittent
exposure
24.4 ± 4.9 [jg/dL
for continuously
exposed
Heart rate, HF, LF,
LF/HF measured 24-hr
post 28-wk exposure
0.2% Pb acetate,
permanent Pb
group, M/F, n = 9
Zhu et al. (2018) Rat (Sprague In utero to 1 yr Female rats were given either 0 or 500 mg/L Pb 0.28 ± .02 mg/L Heart rate, HF, LF,
Dawley) acetate for 10 d before mating. Male offspring (28 ± 2 |jg/dL) LF/HF measured post 1-
Control (distilled continued receiving 0 or 500 mg/L Pb acetate for yr exposure
water), M, n = 100 1 V
0.5 g/L Pb acetate,
M, n = 10
M,n = 100
0.5 g/L Pb acetate,
M, n = 10
Zhu et al. (2019) Rat (Sprague In utero to 1 yr Female rats were given either 0 or 0.5 g/L Pb 0.27 ± 0.02 mg/L HF, LF, LF/HF, heart
Dawley) acetate for 10 d before mating. Male offspring (27 ± 2 |jg/dL) measured post 1-yr
Control (distilled continued receiving 0 or 0.5 g/L Pb acetate for exposure
water), M, n = 100 1 V
0.5 g/L Pb acetate,
M, n = 10
BLL = blood lead level; BW = body weight; d = day(s); F = female; HF = high-frequency; LF = low-frequency; M = male; mo = month(s); Pb = lead; PR = prevalence ratio;
wk = week(s); yr = year(s).
External Review Draft
4-139
DRAFT: Do not cite or quote
-------
Table 4-13
Epidemiologic studies of Pb exposure and atherosclerosis and peripheral artery disease.
Reference and
Study Design
Study
Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Wan etal. (2021)
China
May-August 2018
Cross-sectional
METAL study
n = 4,234
Average
individual born
-1951
Blood (AAS) (|jg/dL)
Median (IQR)
2.6 (1.8-3.6)
Age at measurement
Median (IQR):
67 (62-72) yr
CCA plaques and
diameter
Linear or logistic
regression adjusting for
age, sex, duration of
diabetes, education
status, current smoking,
BMI, HbA1c,
dyslipidemia,
hypertension
OR (4th vs. 1st quartile of Blood Pb)b
1.53 (1.29, 1.82)
Yu et al. (2020)
Belgium
Blood Pb collected in
1985-2005, arterial
stiffness measured a
median of 9 yr later
Cohort
Cadmium in
Belgium
n = 267
Average
individual born
-1958
Blood (ETAAS) (pg/dL)
GM (IQR):
2.93 (1.8-4.7)
Age at measurement:
Mean 37 yr
Arterial stiffness
Hemodynamic
measures
Linear multivariable
models adjusting for sex,
enrollment characteristics
(age, BMI, smoking,
drinking, serum total to
HDL-C ration, plasma
glucose, eGFR
(estimated from serum
creatinine), SES), the
time interval between
measurement of
exposure biomarkers and
hemodynamic
assessment, and
antihypertensive drug
treatment at enrollment
and follow-up
Time-dependent hemodynamics per doubling of
Pb concentration15
Augmentation ratio, % 1.74 (0.95, 2.53)
Augmentation index, % 3.03 (1.56, 4.50)
Pressure amplification -0.06 (-0.08, -0.04)
Pulse wave velocity, m/s 0.14 (-0.08, 0.35)
Forward pulse peak time, ms 6.62 (2.21, 11.0)
Backward pulse peak time, ms 1.02 (-1.31, 3.35)
Forward PP amplitude, mmHg
-0.43 (-1.92, 1.06)
Backward PP amplitude, mmHg
1.02 (0.02, 2.02)
Reflection index, %: 3.98 (1.71, 6.24)
External Review Draft
4-140
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Kimetal. (2021)
South Korea
2011-2018
Cross-sectional
n =2,193
Adults >2 yr of
age who
completed
voluntary
medical
examinations at
the Chonnam
National
University
Hwasun
Hospital
Average
individual born
-1961
Whole blood Pb
(GFAAS) (|jg/dL)
Mean (Median)
All participants:
2.71 (2.53)
Men: 2.98 (2.78)
Women: 2.18 (2.03)
Age at measurement:
Mean (SD): 53.5 (8.3)
Range: 23-81
Moderate to
severe CAS
(>25% stenosis)
Logistic regression
adjusted for age, sex,
hypertension, diabetes
mellitus, dyslipidemia,
BMI, regular exercise,
smoking, and alcohol
drinking
OR
All participants: 1.14 (1.02, 1.26)
Men: 1.13 (1.01, 1.27)
Women: 1.10 (0.86, 1.41)
Qin etal. (2021)
United States
NHANES 2013-2014
Cross-sectional
NHANES
n = 1,503
>40 yr
Average
individual born
-1960
Blood (ICP-MS)
(|jg/dL)c
Mean (SD): 1.45 (1.31)
Q1: 0.16-0.80
Q2: 0.81-1.21
Q3: 1.22-1.84
Q4: 1.85-24.6
Age at measurement
Mean: 52.7
AAC score (0-24 Linear or logistic
for total score),
and severe AAC
(AAC score >6)
regression adjusted for
sex, age, race,
education, BMI, blood
pressure, creatinine, A1c,
uric acid, serum calcium,
serum phosphorus, total
cholesterol, cotinine,
hemoglobin, hydroxy-
vitamin D, hypertension,
diabetes
Change in AAC score
Per [jg/dL increase in blood Pb 0.15 (0.02, 0.27)
Q2 vs. Q1: 0.58 (0.15, 1.02)
Q3 vs. Q1: 0.60 (0.14, 1.07)
Q4 vs. Q1: 0.99 (0.50, 1.48)
OR
Per [jg/dL increase in blood Pb 1.11 (1.00, 1.22)
Q2 vs. Q1
Q3 vs. Q1
Q4 vs. Q1
1.68 (0.86, 3.25)
2.15 (1.10, 4.19)
3.72 (1.94, 7.12)
External Review Draft
4-141
DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study
Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Muntner et al. (2005)
United States
NHANES 1999-2002
Cross-sectional
NHANES
n = 9,961
Average
individual born
-1954
Or
Average
individual born
in or before
-1983
Blood (GFAAS)
(Hg/dL)
Mean (95% CI): 1.64
(1.59-1.68)
PAD
Q1
Q2
Q3
Q4
<1.06
1.06-1.63
1.63-2.47
>2.47
Age at measurement
Mean NR
Logistic regression
models adjusted for age,
race/ethnicity, sex,
diabetes mellitus, BMI,
cigarette smoking,
alcohol consumption,
high school education,
health insurance status
OR (95% CI) (vs. 1st quartile)b
Q2
Q3
Q4
1.00 (0.45, 2.22)
1.21 (0.66, 2.23)
1.92 (1.02, 3.61)
Navas-Acien et al.
(2004)
United States
1999-2000
Cross-sectional
NHANES
n =2,125
participants,
age >40 yr,
Average
individual born
in or before
-1960
Blood (GFAAS)
(Hg/dL)
Q1
Q2
Q3
Q4
<1.45
1.45-2.07
2.07-2.90
>2.90
PAD
Logistic regression
adjusted for age, sex,
race, education, BMI,
alcohol intake,
hypertension, diabetes,
hypercholesterolemia,
eGFR, C-reactive
protein, self-reported
smoking status, serum
cotinine and Cd
OR (95% CI) (vs. 1st quartile)b
Q2
Q3
Q4
1.63 (0.50-5.27)
1.77 (0.55-5.63)
2.52 (0.75-8.51)
AAC = abdominal aortic calcification; AAS = atomic absorption spectrometry; BMI = body mass index; CAS = coronary artery stenosis; CCA = common carotid artery; Cd = cadmium;
CI = confidence interval; eGFR = estimated glomerular filtration rate; ETAAS = electrothermal atomic absorption spectrometry; GFAAS = graphite furnace atomic absorption
spectrometry; GM = geometric mean; HbA1c = hemoglobin A1C; HDL-C = high-density lipoprotein cholesterol; ICP-MS = inductively coupled plasma mass spectrometry;
IQR = interquartile ratio; METAL = Environmental Pollutant Exposure and Metabolic Diseases in Shanghai ; mo = month(s); NHANES = National Health and Nutrition Examination
Survey; NR = not reported; OR = odds ratio; PAD = peripheral artery disease; Pb = lead; PP = pulse pressure; Q = quartile; Q = quartile; SD = standard deviation;
SES = socioeconomic status; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
°Results reported as ng/dL but assumed to be ug/dL based on data source (NHANES).
External Review Draft
4-142
DRAFT: Do not cite or quote
-------
Table 4-14
Animal toxicological studies of Pb exposure and atherosclerosis.
Study
Species (Stock/Strain),
n, Sex
Timing of Exposure
Exposure Details
(Concentration,
Duration)
BLL As Reported
(pg/dL)
Endpoints
Examined
Xu et al. (2015)
Rat (Sprague Dawley)
Control 1, M/F, n = 6
1% PB acetate group
12 d, M/F, n = 15
6-7-wk old rats exposed for
12 or 40 d
Two control groups given
distilled water for 12 or
40 d. Two 1% Pb acetate
groups exposed for 12 or
40 d
Day 12: 193.3 pg/L
(19.33 pg/dl)
Day 40: 245.9 pg/L
(24.59 pg/L)
Cardiovascular
histology measured at
12 and 40 d
1% PB acetate group
40 d, M/F, n = 15
BLL = blood lead level; d = day(s); F = female; M = male; Pb = lead; wk = week(s).
External Review Draft
4-143
DRAFT: Do not cite or quote
-------
Table 4-15
Epidemiologic studies of Pb exposure and cerebrovascular disease.
Reference and
Study Design
Study Population Exposure Assessment
Outcome
Confounders
Effect Estimates and
95% Clsa
Menke et al. (2006) NHANES III
n = 13,946
United States
Average individual
NHANES III 1988- born-1946
1994, mortality
follow-up in 2001
Cohort
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Mean: 2.58
Tertiles:
T1
T2
T3
<1.93
1.94-3.62
>3.63
Age at measurement
Mean: 44.4 yr
Stroke mortality Cox proportional hazard regression
analysis adjusted age, race/ethnicity, sex,
urban residence, cigarette smoking,
alcohol consumption, education, physical
activity, household income, menopausal
status, BMI, CRP, total cholesterol,
diabetes mellitus, hypertension, GFR
category
HR: 1.15 (1.02, 1.28)
Khaliletal. (2009)
Baltimore, MD and
Monongahela Valley,
PA
Study of Osteoporotic
Fractures
n = 533
women
87 yr
ages 65-
Blood (GFAAS with
Zeeman correction)
(|jg/dL)
Mean (SD): 5.3 (2.3)
Range: 1-21
Stroke mortality Cox proportional hazards regression
analysis adjusted forage, clinic, BMI,
education, smoking, alcohol intake,
estrogen use, hypertension, total hip bone
mineral density, walking for exercise, and
diabetes
HR (95% CI) (>8 [jg/dL
blood Pb)b:
1.13 (0.34, 3.81)
Blood Pb measured Average individual
1990-1991, mortality born -1921
follow-up for -12 y
Age at measurement
(Mean): 70
Cohort
External Review Draft 4-144 DRAFT: Do not cite or quote
-------
Reference and
Study Design
Study Population Exposure Assessment Outcome
Confounders
Effect Estimates and
95% Clsa
Mousavi-Mirzaei et n = 88
Blood (GFAAS) (pg/dL)
Median (IQR):
6.38 (1.75-34.87)
Acute ischemic
stroke
Logistic regression controlling for lipid
profile and fasting blood sugar
OR (95% CI):
1.04 (1.02, 1.07)
al. (2020)
Birjand, Iran
(44 cases, 44 controls
matched on age and
sex, occupation,
Age at measurement
Mean (SD): 71.95
(11.37)
2016-2017
opium addiction, and
sampling time)
Case-control
Average individual
born -1944
BMI = body mass index; CI = confidence interval; CRP = C-reactive protein; GFAAS = graphite furnace atomic absorption spectrometry; GFR = glomerular filtration rate; HR = hazard
ratio; IQR = interquartile ratio; NHANES = National Health and Nutrition Examination Survey; OR = odds ratio; Pb = lead; SD = standard deviation; T = fertile; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect
estimates are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated
interval. Categorical effect estimates are not standardized.
bUnable to be standardized.
External Review Draft 4-145 DRAFT: Do not cite or quote
-------
Table 4-16
Epidemiologic studies of Pb exposure and cardiovascular mortality.
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Lustberq and Silberqeld
(2002)
United States
NHANES II 1976-1980,
mortality follow-up in
1992
NHANES
Blood (GFAAS with
n = 4 190 aged 30- Zeeman correction)"
74 (MQ/dL)
Mean (SD): 14.0 (5.1)
Median: 13
Tertiles
Circulatory Cox proportional hazard
mortality regression analysis adjusted for
age, sex, location, education,
race, income, smoking, BMI,
exercise
Average individual
born -1924
HRC
T2 vs. T1 1.27 (0.97,
T3 vs. T1 1.74 (1.25,
1.57)
2.40)
T1
T2
T3
<10
10-19
20-29
Cohort
Age at measurement:
Mean (SD) 54.1 (13.2)
Schober et al. (2006)
United States
NHANES III 1988-
1994, mortality follow-
up in 2006
-8.5 yr of follow-up
NHANES III
n = 9,686, >40 yr
Average individual
born in or before
-1951
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
CVD mortality Cox proportional hazard HR (95% CI)
regression analysis adjusted for q^d
sex, age, race/ethnicity,
T1
T2
T3
<5 (median 2.6)
5-9 (median 6.3)
>10 (median 11.8)
smoking, education level.
Did not evaluate BMI nor
comorbidities
T3 vs. T1 1.55 (1.16, 2.07)
Age at measurement:
>40 y
Cohort
External Review Draft
4-146
DRAFT: Do Not Cite or Quote
-------
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Menke et al. (2006)
United States
NHANES III 1988-
1994, mortality follow-
up in 2001
-12 yr of follow-up
Cohort
NHANES III
n = 13,946, >20 yr.
Average individual
born -1946
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Mean: 2.58
Tertiles:
T1
T2
T3
<1.93
1.94-3.62
>3.63
Age at measurement
Mean: 44.4 yr
CVD, Ml, and
stroke mortality
Cox proportional hazard
regression analysis adjusted
age, race/ethnicity, sex, urban
residence, cigarette smoking,
alcohol consumption,
education, physical activity,
household income,
menopausal status, BMI, CRP,
total cholesterol, diabetes
mellitus, hypertension, GFR
category
HR (95% CI):
CVD: 1.13 (1.06, 1.22)
Ml: 1.19 (1.05, 1.34)
Stroke: 1.15 (1.02, 1.28)
Lanphear et al. (2018)
United States
1988-1994 mortality
follow-up in 2011
-19 yr of follow-up (IQR
17.6-21.0 yr)
Cohort
NHANES III
n = 14,289 >20 yr.
Average individual
born -1947
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
GM: 2.71
Geometric SE: 1.31
10th percentile: 1.0
90th percentile: 6.7
Age at measurement:
Mean: 44.1 yr
CVD, and IHD
mortality
Cox proportional hazards
regression analysis adjusting
forage, sex, household
income, ethnic origin, BMI,
smoking status, alcohol
consumption, physical activity,
concentration of Cd in urine,
blood pressure, healthy eating
index tertiles, HbA1c, and
serum cholesterol
HR (95% CI)
CVD: 1.10 (1.05,
1.15)
IHD: 1.14 (1.08, 1.20)
External Review Draft
4-147
DRAFT: Do Not Cite or Quote
-------
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
van Bemmel et al.
(2011)
United States
1988-1994, follow-up
through 2007
-7 yr of follow-up for
those with low blood Pb
~7 yr of follow-up for
those with high blood
Pb
NHANES III
n = 3,349 adult age
>40 yr
Average individual
born -1932
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Median:
<5 [jg/dL: 2.6
>5 [jg/dL: 7.5
Age at measurement
Mean
<5 [jg/dL: 57
>5 [jg/dL: 61
CVD mortality
Cox proportional hazards
adjusting forage, education,
sex, smoking status, and
race/ethnicity
HR (95% CI):
CVD
All: 1.02 (0.93, 1.13)
ALADGG: 1.01 (0.92, 1.10)
ALADCG/GG: 1.13 (0.93, 1.37)
Cohort
Cook et al. (2022)
United States
Baseline 1988-1994,
mortality follow-up
through 2010 (2 yrof
follow-up)
Cohort
NHANES III
n = 15,036
adults >19 y
Average individual
born in or before
-1972
Blood Pb (GFAAS)
(Hg/dL)
Quartiles
Men:
Q1: <2.63
Q2 and Q3: 2.63-6.23
Q4: >6.23
Women:
Q1: <1.38
Q2 and Q3: 1.38-3.74
Q4: >3.74
Age at measurement:
>1 yr
Heart disease
mortality (ICD-10
100-113 I20-I22,
I24, I25-I28,
125.1-125.9, 130-
131, 133, I34-I38,
I40, 142-151, I70-
I78, Ml mortality
(121-122), and
CVD mortality
(heart disease
mortality plus
160-169
(cerebrovascular
disease)
Multivariate Cox model
adjusted for age, gender,
race/ethnicity, family income,
alcohol drinking, cigarette
smoking, BMI, physical activity,
and self-reported health status
at baseline
HR (95% Cl)c:
CVD mortality
Q2 & Q3 vs. Q1: 1.10 (0.84, 1.43)
Q4 vs. Q1: 1.35 (1.03, 1.77)
Per 1 [jg/dL increase in log-
transformed Pb: 1.08 (1.00, 1.16)
Heart disease mortality
Q2 & Q3 vs. Q1: 1.37 (1.04, 1.81)
Q4 vs. Q1: 1.60 (1.21, 2.13)
Per 1 [jg/dL increase in log-
transformed Pb: 1.09 (1.02, 1.16)
Ml mortality
Q2 & Q3 vs. Q1: 1.73 (1.08, 2.79)
Q4 vs. Q1: 1.45 (0.90, 2.32)
Per 1 [jg/dL increase in log-
transformed Pb: 0.95 (0.84, 1.08)
External Review Draft
4-148
DRAFT: Do Not Cite or Quote
-------
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Ruiz-Hernandez et al.
(2017)
United States
Baseline 1988-1994
and 1999-2004,
mortality follow-up
through 1996 and 2006
Cohort
NHANES III and
NHANES
n = 15,421
age >40yr
Average individual
born in or before
-1951
Blood (|jg/dL)
1988-1994 (GFAAS
with Zeeman
correction)
Mean: 3.2
1999-2004 (ICP-MS)
Mean: 1.9
Age at measurement:
>40 yr
CVD and CHD Poisson regression adjusting
mortality for age, sex, race, smoking
status, physical inactivity,
obesity, hypertension, diabetes,
high total cholesterol, low HDL
cholesterol, lipid lowering
medication, survey period, and
log-transformed urinary Cd
concentrations
RR (95% CI) for twofold increase in
blood Pbc
CVD:
CHD:
1.19 (1.07,
1.24 (1.10,
1.31)
1.41)
(Duan et al.. 2020)
United States
1999-2014, follow-up
through end of 2015
~7 yr of follow-up
NHANES
n = 18,602
Age >20 yr
Average individual
born -1960
Blood (ICP-MS)
(pg/dL)d
Median (IQR)
1.49 (0.93, 2.31)
Age at measurement:
Mean (SD): 45.9 (0.3)
CVD mortality Poisson regression analyses
adjusted for: sex, age, ethnicity,
education, PIR, cotinine
category, BMI, physical activity,
hypertension, and diabetes
RR (95% CI)
CVD: 1.35 (1.15, 1.59)
Cohort
External Review Draft
4-149
DRAFT: Do Not Cite or Quote
-------
Reference and Study Exposure
Design Study Population Assessment Outcome Confounders Effect Estimates and 95% Clsa
Aoki etal. (2016)
United States
NHANES
n = 18,602
age >40 yr
Blood (ICP-MS)
(Hg/dL)
Mean (SE): 1.73 (0.02)
1999-2010, follow-up Average individual Age a* measurement
through end of 2011 born-1947 Mean (SE): 57.5 (0.2)
~6 yr of follow-up
Cohort
CVD mortality Cox proportional hazards
(using age during follow-up as
the time scale) adjusting for
race, Hispanic origin, sex,
alcohol consumption, blood Cd,
serum iron, C-Reactive Protein
(CRP), and serum calcium
RR (95% CI) for 10-fold increase in
blood Pbc
Overall: 1.26 (0.91, 1.78
Control for hematocrit:
1.35 (0.98, 1.86
Control for hemoglobin:
1.35 (0.98, 1.87)
Hematocrit-corrected:
1.44 (1.05, 1.98)
Hemoglobin-corrected:
1.46 (1.06, 2.01)
Obenq-Gvasi et al.
(2021)
United States
Baseline 1999-2008,
mortality follow-up
through 2014
NHANES
n = 28,852
adults >20 yr
Average individual
born in or before
-1983
Blood Pb (ICP-MS)
(Hg/dL)
Median: 1.55
Age at measurement:
>20 yr
CVD mortality Multivariate Cox model
adjusting for sex, BMI,
smoking, alcohol consumption,
country of birth, and income
*No adjustment for age
HR (>1.55 [jg/dL Blood Pb)c:
2.35 (1.77, 2.93)
Cohort
External Review Draft
4-150
DRAFT: Do Not Cite or Quote
-------
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Lin etal. (2011)
Taiwan
Years not reported
Cohort (18 mo of follow-
up)
n = 927
Taiwanese adult
patients with end-
stage renal disease
on hemodialysis for
>6 mo, age >18
Baseline blood Pb
(ETAAS) (pg/dL)
Mean: 11.5
Median: 10.4
T1
T2
T3
CVD mortality
<8.51 pg/dL
8.51-12.64 pg/dL
>12.64 pg/dL
Age at measurement
Mean (SD): 55.2 (13.5)
Multivariate Cox model
adjusting forage, previous
cardiovascular diseases
(stroke, Ml, PVD, CHF),
education level, hemodialysis
vintage, using fistula,
normalized protein catabolic
rate, hemoglobin, serum
albumin, creatinine,
cardiothoracic ratio, and
logarithmic transformation of
high-sensitivity CRP
HR (95% CI) (T1: Referent)0
CVD
T2: 3.70 (2.06, 6.48)
T3: 9.71 (2.11, 23.26)
Hemoglobin-corrected:
CVD
T2: 3.52 (0.51, 6.33)
T3: 7.35 (1.64, 33.33)
Weisskopf et al. (2009)
United States
Baseline biomarkers
collected an average of
8 yr prior to death
Cohort
NAS
n = 868
Mostly white elderly
men
Average individual
born in or before
-1940
Blood Pb (GFAAS)
(pg/dL)
Mean (SD): 5.6 (3.4)
Bone Pb (K-XRF)
(pg/g)
Patella Pb
Mean (SD): 31.2 (19.4)
Tertiles:
T1: <22
T2: 22-35
T3: >35
Tibia Pb
Mean (SD): 21.8 (13.6)
CVD and IHD
mortality
Multivariate Cox model
adjusting forage, smoking,
education, alcohol intake,
physical activity, BMI, total
cholesterol, serum HDL,
diabetes, race, and
hypertension
HR (95% CI) (3rd vs. 1st fertile)0
Patella Pb:
CVD: 2.45 (1.07, 5.60)
IHD: 8.37 (1.29, 54.4)
Age at measurement:
>55 yr
External Review Draft
4-151
DRAFT: Do Not Cite or Quote
-------
Reference and Study
Design
Study Population
Exposure
Assessment
Outcome
Confounders
Effect Estimates and 95% Clsa
Khaliletal. (2009)
Baltimore, MD and
Monongahela Valley,
PA
Blood Pb measured
1990-1991, mortality
follow-up for -12 yr
Study of
Osteoporotic
Fractures
n = 533
Women, ages 65-
87 yr
Average individual
born -1921
Blood (GFAAS with
Zeeman correction)
(Hg/dL)
Mean (SD): 5.3 (2.3)
Range: 1-21
Age at measurement
(Mean): 70
CVD, and CHD
mortality
Cox proportional hazards
regression analysis adjusted for
age, clinic, BMI, education,
smoking, alcohol intake,
estrogen use, hypertension,
total hip bone mineral density,
walking for exercise, and
diabetes
HR (95% CI) (>8 [jg/dL blood Pb)c
CVD: 1.78 (0.92, 3.45)
CHD: 3.08 (1.23, 7.70)
Cohort
Hollinqsworth and
Rudik (2021) United
States
1999-2016
Quasi-experimental
design
Elderly population
(>65 yr)
Assessed the
change in deaths
(National Vital
Statistics System)
occuring among
this age group
before and after the
phaseout of leaded
gasoline in
professional racing
(NASCAR, ARCA).
County-level blood Pb
measurements in
children
CVD and IHD Difference-in-difference
mortality approach controlling for SES at
the county level (median
income, unemployment rates,
percent minority population),
TRI Pb emissions data
Decline in age-standardized mortality
rate per 100,000 population
CVD:
Race counties: 37
Border counties: 12
IHD:
Race counties: 53
Border counties: 20
Compared mortality
rates in race-
counties to
bordering counties
Average individual
born in or before
-1942
External Review Draft
4-152
DRAFT: Do Not Cite or Quote
-------
Reference and Study Exposure
Design Study Population Assessment Outcome Confounders Effect Estimates and 95% Clsa
ARCA = Automobile Racing Club of America; BMI = body mass index; CHD = coronary heart disease; CHF = congestive heart failure; Cd = cadmium; CI = confidence interval;
CRP = C-reactive protein; CVD = cardiovascular disease; ETAAS = electrothermal atomic absorption spectrometry; GFAAS = graphite furnace atomic absorption spectrometry;
GFR = glomerular filtration rate; GM = geometric mean; HDL = high-density lipoprotein; HR = hazard ratio; ICD = International Classification of Diseases; ICP-MS = inductively coupled
plasma mass spectrometry; IHD = ischemic heart disease; IQR = interquartile ratio; K-XRF = K-Shell X-ray fluorescence; Ml = myocardial infarction; mo = month(s); NAS = Normative
Aging Study; NASCAR = National Association for Stock Car Auto Racing; NHANES = National Health and Nutrition Examination Survey; Pb = lead; PIR = poverty-to-income ratio;
PVD = peripheral vascular disease; Q = quartile; RR = relative risk; SD = standard deviation; SE = standard error; T = fertile; yr = year(s).
aEffect estimates are standardized to a 1 |jg/dL increase in blood Pb or a 10 |jg/g increase in bone Pb, unless otherwise noted. If the Pb biomarker is log-transformed, effect estimates
are standardized to the specified unit increase for the 10th—90th percentile interval of the biomarker level. Effect estimates are assumed to be linear within the evaluated interval.
Categorical effect estimates are not standardized.
bPb analysis method assumed based on data source, not reported in paper.
°Unable to be standardized.
dUnits assumed to be |jg/dL (written as |jg/L in the paper).
External Review Draft
4-153
DRAFT: Do Not Cite or Quote
-------
4.14 References
Ahn. J: Kim. NS: Lee. BK: Park. J: Kim. Y. (2018). Association of blood pressure with blood lead and cadmium
levels in Korean adolescents: Analysis of data from the 2010-2016 Korean National Health and Nutrition
Examination Survey. J Korean Med Sci 33: e278.
Anonymous. (19941. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation: Analysis of
pooled data from five randomized controlled trials. Arch Intern Med 154: 1449-1457.
Aoki. Y: Brodv. DJ: Flegal. K. (20171. Blood lead and other metal biomarkers as risk factors for cardiovascular
disease mortality: Erratum [Erratum], Medicine (Baltimore) 96: e7997.
Aoki. Y: Brodv. DJ: Flegal. KM: Fakhouri. THI: Axelrad. DA: Parker. JD. (2016). Blood lead and other metal
biomarkers as risk factors for cardiovascular disease mortality. Medicine (Baltimore) 95: e2223.
Attri. J: Dhawan. V: Mahmood. S: Pandhi. P: Parwana. HK: Nath. R. (2003). Effect of vitamin C supplementation
on oxidative DNA damage in an experimental model of lead-induced hypertension. Ann Nutr Metab 47:
294-301.
Badavi. M: Mehreerdi. FZ: Sarkaki. A: Naseri. MKG: Dianat. M. (2008). Effect of grape seed extract on lead
induced hypertension and heart rate in rat. Pak J Biol Sci 11: 882-887.
Bagchi. D: Preuss. HG. (2005). Effects of acute and chronic oral exposure of lead on blood pressure and bone
mineral density in rats. J Inorg Biochem 99: 1155-1164.
Bigger. JT. Jr: Fleiss. JL: Steinman. RC: Rolnitzkv. LM: Kleiger. RE: Rottman. JN. (1992). Frequency domain
measures of heart period variability and mortality after myocardial infarction. Circulation 85: 164-171.
Billman. GE. (2013). The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance [Editorial].
Front Physiol 4: 26.
Bravo. Y: Ouiroz. Y: Ferrebuz. A: Vaziri. ND: Rodriguez-Iturbe. B. (2007). Mycophenolate mofetil administration
reduces renal inflammation, oxidative stress, and arterial pressure in rats with lead-induced hypertension.
Am J Physiol Renal Physiol 293: F616-F623.
Bulka. CM: Brvan. MS: Perskv. VW: Daviglus. ML: Durazo-Arvizu. RA: Parvez. F: Slavkovich. V: Graziano. JH:
Islam. T: Baron. JA: Ahsan. H: Argos. M. (2019). Changes in blood pressure associated with lead,
manganese, and selenium in a Bangladeshi cohort. Environ Pollut 248: 28-35.
Bushnik. T: Levallois. P: D'Amour. M: Anderson. TJ: McAlister. FA. (2014). Association between blood lead and
blood pressure: Results from the Canadian Health Measures Survey (2007 to 2011). Health Rep 25: 12-22.
Carll. AP: Farrai. AK: Roberts. AM. (2018). The role of the autonomic nervous system in cardiovascular toxicity. In
C McQueen (Ed.), Comprehensive toxicology (3rd ed., pp. 61-114). Amsterdam, Netherlands: Elsevier.
Carmignani. M: Boscolo. P: Poma. A: Volpe. AR. (1999). Kininergic system and arterial hypertension following
chronic exposure to inorganic lead. Immunopharmacology 44: 105-110.
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.
Chen. S: Golemboski. KA: Piepenbrink. M: Dietert. RR. (2004). Developmental immunotoxicity of lead in the rat:
Influence of maternal diet. J Toxicol Environ Health A 67: 495-511.
Chen. Z: Huo. X: Zhang. S: Cheng. Z: Huang. Y: Xu. X. (2021). Relations of blood lead levels to echocardiographic
left ventricular structure and function in preschool children. Chemosphere 268: 128793.
Cheng. Y: Schwartz. J: Vokonas. PS: Weiss. ST: Aro. A: Hu. H. (1998). Electrocardiographic conduction
disturbances in association with low-level lead exposure (the Normative Aging Study). Am J Cardiol 82:
594-599.
External Review Draft
4-154
DRAFT: Do Not Cite or Quote
-------
Cho. HM; Cho. DY: Kim. MY: Yang. SW: Seo. YS: Kim. KN. (2016). Combined effect of blood cadmium and lead
levels on coronary heart disease prediction risk in Korean men. Angiology 67: 582-586.
Choi. JW: Oh. C: Shim. SY: Jeong. S: Kim. HS: Kim. MS. (2018). Reduction in prevalence of hypertension and
blood heavy metals among curry-consumed Korean. Tohoku J Exp Med 244: 219-229.
Choi. S: Kwon. J: Kwon. P: Lee. C: Jang. SI. (2020). Association between blood heavy metal levels and predicted
10-year risk for a first atherosclerosis cardiovascular disease in the general Korean population. Int J
Environ Res Public Health 17: 2134.
Chowdhurv. R: Ramond. A: O'Keeffe. LM: Shahzad. S: Kunutsor. SK: Muka. T: Gregson. J: Willeit. P: Warnakula.
S: Khan. H: Chowdhurv. S: Gobin. R: Franco. OH: Di Angelantonio. E. (2018). Environmental toxic metal
contaminants and risk of cardiovascular disease: systematic review and meta-analysis. BMJ 362: k3310.
Chung. CJ: Wu. CD: Hwang. BF: Wu. CC: Huang. PH: Ho. CT: Hsu. HT. (2020). Effects of ambient PM 2.5 and
particle -bound metals on the healthy residents living near an electric arc furnace: A community- based
study. Sci Total Environ 728: 138799.
Cook. MK: Zhang. J: Wei. Y. (2022). Blood lead levels and risk of deaths from cardiovascular disease. Am J
Cardiol 173: 132-138.
Courtois. E: Marques. M: Barrientos. A: Casado. S: Lopez-Farre. A. (2003). Lead-induced downregulation of
soluble guanylage cyclase in isolated rat aortic segments mediated by reactive oxygen species and
cyclooxygenase-2. J Am Soc Nephrol 14: 1464-1470.
Desai. G: Niu. Z: Luo. W: Frndak. S: Shaver. AL: Kordas. K. (2021). Low-level exposure to lead, mercury, arsenic,
and cadmium, and blood pressure among 8-17-year-old participants of the 2009-2016 National Health and
Nutrition Examination Survey. EnvironRes 197: 111086.
Dewaniee. S: Sahu. R: Karmakar. S: Gangopadhvav. M. (2013). Toxic effects of lead exposure in Wistar rats:
Involvement of oxidative stress and the beneficial role of edible jute (Corchorus olitorius) leaves. Food
Chem Toxicol 55: 78-91.
Ding. N: Wang. X. in: Weisskopf. MG: Sparrow. D: Schwartz. J: Hu. H: Park. SK. (2016). Lead-related genetic loci,
cumulative lead exposure and incident coronary heart disease: The normative aging study. PLoS ONE 11:
e0161472.
Ding. N: Wang. X: Tucker. KL: Weisskopf. MG: Sparrow. D: Hu. H: Park. SK. (2019). Dietary patterns, bone lead
and incident coronary heart disease among middle-aged to elderly men. Environ Res 168: 222-229.
Duan. W: Xu. C: Liu. O: Xu. J: Weng. Z: Zhang. X: Basnet. TB: Dahal. M: Gu. A. (2020). Levels of a mixture of
heavy metals in blood and urine and all-cause, cardiovascular disease and cancer mortality: A population-
based cohort study. Environ Pollut 263: 114630.
Dursun. N: Arifoglu. C: Stier. C: Keskinol. L. (2005). Blood pressure relationship to nitric oxide, lipid peroxidation,
renal function, and renal blood flow in rats exposed to low lead levels. Biol Trace Elem Res 104: 141-149.
Dvatlov. VA: Lawrence. DA. (2002). Neonatal lead exposure potentiates sickness behavior induced by Listeria
monocytogenes infection of mice. Brain Behav Immun 16: 477-492.
Egan. KB: Cornwell. CR: Courtney. JG: Ettinger. AS. (2021). Blood lead levels in U.S. children ages 1-11 years,
1976-2016. Environ Health Perspect 129: 37003.
Elmarsafawv. SF: Jain. NB: Schwartz. J: Sparrow. D: Nie. HL: Hu. H. (2006). Dietary calcium as a potential
modifier of the relationship of lead burden to blood pressure. Epidemiology 17: 531-537.
Ettinger. AS: Bovet. P: Plange-Rhule. J: Forrester. TE: Lambert. EV: Lupoli. N: Shine. J: Dugas. LR: Shoham. D:
Durazo-Arvizu. RA: Cooper. RS: Luke. A. (2014). Distribution of metals exposure and associations with
cardiometabolic risk factors in the "Modeling the Epidemiologic Transition Study". Environ Health 13: 90.
Eum. KD: Nie. LH: Schwartz. J: Vokonas. PS: Sparrow. D: Hu. H: Weisskopf. MG. (2011). Prospective cohort
study of lead exposure and electrocardiographic conduction disturbances in the Department of Veterans
Affairs Normative Aging Study. Environ Health Perspect 119: 490-494.
External Review Draft
4-155
DRAFT: Do Not Cite or Quote
-------
Everson. TM: Niedzwiecki. MM: Toth. D: Tellez-Plaza. M: Liu. H: Barr. DB: Gribble. MO. (2021). Metal
biomarker mixtures and blood pressure in the United States: Cross-sectional findings from the 1999-2006
National Health and Nutrition Examination Survey (NHANES). Environ Health 20: 15.
Factor-Litvak. P: Kline. JK: Popovac. D: Hadzialievic. S: Lekic. V: Preteni-Rexhepi. E: Capuni-Paracka. S:
Slavkovich. V: Graziano. J. (1996). Blood lead and blood pressure in young children. Epidemiology 7:
633-637.
Factor-Litvak. P: Wasserman. G: Kline. JK: Graziano. J. (1999). The Yugoslavia prospective study of
environmental lead exposure. Environ Health Perspect 107: 9-15.
Farmand. F: Ehdaie. A: Roberts. CK: Sindhu. RK. (2005). Lead-induced dysregulation of superoxide dismutases,
catalase, glutathione peroxidase, and guanylate cyclase. Environ Res 98: 33-39.
Fauchier. L: Babutv. D: Melia A: Bonnet. P: Cosnav. P: Fauchier. JP. (2004). Heart rate variability in severe right
or left heart failure: the role of pulmonary hypertension and resistances. Eur J Heart Fail 6: 181-185.
Fernandez-Cabezudo. MJ: Ali. SAE: Ullah. A: Hasan. MY: Kosanovic. M: Fahim. MA: Adem. A: al-Ramadi. BK.
(2007). Pronounced susceptibility to infection by Salmonella enterica serovar Typhimurium in mice
chronically exposed to lead correlates with a shift to Th2-type immune responses. Toxicol Appl Pharmacol
218: 215-226.
Fioresi. M: Simoes. MR: Furieri. LB: Broseghini-Filho. GB: Vescovi. MVA: Stefanon. I: Vassallo. DV. (2014).
Chronic lead exposure increases blood pressure and myocardial contractility in rats. PLoS ONE 9: e96900.
Fiorim. J: Ribeiro Junior. RF: Silveira. EA: Padilha. AS: Vescovi. MVA: de Jesus. HC: Stefanon. I: Salaices. M:
Vassallo. DV. (2011). Low-level lead exposure increases systolic arterial pressure and endothelium-derived
vasodilator factors in rat aortas. PLoS ONE 6: el7117.
Gambelunghe. A: Sallsten. G: Borne. Y: Forsgard. N: Hedblad. B: Nilsson. P: Fagerberg. B: Engstrom. G:
Barregard. L. (2016). Low-level exposure to lead, blood pressure, and hypertension in a population-based
cohort. EnvironRes 149: 157-163.
Gaspar. AF: Cordellini. S. (2014). Combination therapy for the cardiovascular effects of perinatal lead exposure in
young and adult rats. Arq Bras Cardiol 103: 219-229.
Geraldes. V: Carvalho. M: Goncalves-Rosa. N: Tavares. C: Laranio. S: Rocha. I. (2016). Lead toxicity promotes
autonomic dysfunction with increased chemoreceptor sensitivity. Neurotoxicology 54: 170-177.
Glenn. BS: Bandeen-Roche. K: Lee. BK: Weaver. VM: Todd. AC: Schwartz. BS. (2006). Changes in systolic blood
pressure associated with lead in blood and bone. Epidemiology 17: 538-544.
Goncalves-Rizzi. VH: Nascimento. RA: Possomato-Vieira. JS: Dias-Junior. CA. (2016). Sodium nitrite prevents
both reductions in circulating nitric oxide and hypertension in 7-day lead-treated rats. Basic Clin Pharmacol
Toxicol 118: 225-230.
Gonick. HC: Ding. Y: Bondv. SC: Ni. Z: Vaziri. ND. (1997). Lead-induced hypertension: Interplay of nitric oxide
and reactive oxygen species. Hypertension 30: 1487-1492.
Grizzo. LT: Cordellini. S. (2008). Perinatal lead exposure affects nitric oxide and cyclooxygenase pathways in aorta
of weaned rats. Toxicol Sci 103: 207-214.
Gump. BB: Dvkas. MJ: MacKenzie. JA: Dumas. AK: Hruska. B: Ewart. CK: Parsons. PJ: Palmer. CD: Bendinskas.
K. (2017). Background lead and mercury exposures: Psychological and behavioral problems in children.
EnvironRes 158: 576-582.
Gump. BB: MacKenzie. JA: Bendinskas. K: Morgan. R: Dumas. AK: Palmer. CD: Parsons. PJ. (2011). Low-level
Pb and cardiovascular responses to acute stress in children: The role of cardiac autonomic regulation.
Neurotoxicol Teratol 33: 212-219.
Gump. BB: Reihman. J: Stewart. P: Lonkv. E: Darvill. T: Matthews. KA. (2007). Blood lead (Pb) levels: A potential
environmental mechanism explaining the relation between socioeconomic status and cardiovascular
reactivity in children. Health Psychol 26: 296-304.
External Review Draft
4-156
DRAFT: Do Not Cite or Quote
-------
Gump. BB: Stewart. P: Reihman. J: Lonkv. E: Darvill. T: Matthews. KA: Parsons. PJ. (2005). Prenatal and early
childhood blood lead levels and cardiovascular functioning in 9 1/2 year old children. Neurotoxicol Teratol
27: 655-665.
Guo. X: Yang. O: Zhang. W: Chen. Y: Ren. J: Gao. A. (2019). Associations of blood levels of trace elements and
heavy metals with metabolic syndrome in Chinese male adults with microRNA as mediators involved.
EnvironPollut 248: 66-73.
Gustavsson. P: Plato. N: Hallqvist. J: Hogstedt. C: Lewne. M: Reuterwall. C: Scheele. P. (2001). A population-
based case-referent study of myocardial infarction and occupational exposure to motor exhaust, other
combustion products, organic solvents, lead, and dynamite. Epidemiology 12: 222-228.
Halabickv. OM: Pinto-Martin. JA: Compton. P: Liu. J. (2022). Early childhood lead exposure and adolescent heart
rate variability: A longitudinal cohort study. EnvironRes 205: 112551.
Hara. A: Thiis. L: Asavama. K: Gu. YM: Jacobs. L: Zhang. ZY: Liu. YP: Nawrot. TS: Staessen. JA. (2015). Blood
pressure in relation to environmental lead exposure in the National Health and Nutrition Examination
Survey 2003 to 2010. Hypertension 65: 62-69.
Heo. Y: Lee. WT: Lawrence. DA. (19971. In vivo the environmental pollutants lead and mercury induce oligoclonal
T cell responses skewed toward type-2 reactivities. Cell Immunol 179: 185-195.
Heo. Y: Parsons. PJ: Lawrence. DA. (1996). Lead differentially modifies cytokine production in vitro and in vivo.
Toxicol Appl Pharmacol 138: 149-157.
Hevdari. A: Norouzzadeh. A: Khoshbaten. A: Asgari. A: Ghasemi. A: Naiafi. S: Badalzadeh. R. (2006). Effects of
short-term and subchronic lead poisoning on nitric oxide metabolites and vascular responsiveness in rat.
Toxicol Lett 166: 88-94.
Hicken. MT: Gee. GC: Connell. C: Snow. RC: Morenoff. J: Hu. H. (2013). Black-white blood pressure disparities:
Depressive symptoms and differential vulnerability to blood lead. Environ Health Perspect 121: 205-209.
Hicken. MT: Gee. GC: Morenoff. J: Connell. CM: Snow. RC: Hu. H. (2012). A novel look at racial health
disparities: The interaction between social disadvantage and environmental health. Am J Public Health 102:
2344-2351.
Hollingsworth. A: Rudik. I. (20211. The effect of leaded gasoline on elderly mortality: Evidence from regulatory
exemptions. Am Econ J EconPolicy 13: 345-373.
Huang. Z. (2022). Association between blood lead level with high blood pressure in US (NHANES 1999-2018).
Front Public Health 10: 836357.
Iavicoli. I: Carelli. G: Stanek. EJ. Ill: Castellino. N: Calabrese. EJ. (2006). Below background levels of blood lead
impact cytokine levels in male and female mice. Toxicol Appl Pharmacol 210: 94-99.
Jain. NB: Potula. V: Schwartz. J: Vokonas. PS: Sparrow. D: Wright. RO: Nie. H: Hu. H. (2007). Lead levels and
ischemic heart disease in a prospective study of middle-aged and elderly men: The VA Normative Aging
Study. Environ Health Perspect 115: 871-875.
Jhun. MA: Hu. H: Schwartz. J: Weisskopf. MG: Nie. LH: Sparrow. D: Vokonas. PS: Park. SK. (2015). Effect
modification by vitamin D receptor genetic polymorphisms in the association between cumulative lead
exposure and pulse pressure: A longitudinal study. Environ Health 14: 5.
Jing. J: Thapa. S: Delhev. L: Abouelenein. S: Morad. W: Delongchamp. R: Faramawi. MF. (2019). The relation of
blood lead and QRS-T angle in American adults. Arch Environ Occup Health 74: 287-291.
Kannel. WB: Abbott. RD: Savage. DP: McNamara. PM. (1983). Coronary heart disease and atrial fibrillation: The
Framingham Study. Am Heart J 106: 389-396.
Karolv. ED: Li. Z: Dailev. LA: Hvseni. X: Huang. YCT. (2007). Up-regulation of tissue factor in human pulmonary
artery endothelial cells after ultrafine particle exposure. Environ Health Perspect 115: 535-540.
External Review Draft
4-157
DRAFT: Do Not Cite or Quote
-------
Khalil-Manesh. F: Gonick. HC: Weiler. EWJ: Prins. B: Weber. MA: Purdv. R: Ren. O. (1994). Effect of chelation
treatment with dimercaptosuccinic acid (DMSA) on lead-related blood pressure changes. Environ Res 65:
86-99.
Khalil-Manesh. F: Gonick. HC: Weiler. EWJ: Prins. B: Weber. MA: Purdv. RE. (1993). Lead-induced hypertension:
Possible role of endothelial factors. Am J Hypertens 6: 723-729.
Khalil. N. (2010). Re: Blood lead mortality study: Khalil et al., 2009. Integrated science assessment for lead:
Supporting documents: Supplemental data from Dr. Khalil. Khalil, N.
Khalil. N: Wilson. JW: Talbott. EO: Morrow. LA: Hochberg. MC: Hillier. TA: Muldoon. SB: Cummings. SR:
Caulev. JA. (2009). Association of blood lead concentrations with mortality in older women: a prospective
cohort study. Environ Health 8: 15.
Khder. Y: Briancoa S: Petermann. R: Ouilliot. D: Stoltz. JF: Drouin. P: Zannad. F. (1998). Shear stress
abnormalities contribute to endothelial dysfunction in hypertension but not in type II diabetes. J Hypertens
16: 1619-1625.
Kim. S: Kang. W: Cho. S: Lim. DY: Yoo. Y: Park. RJ: Lee. BC: Moon. JD: Park. WJ. (2021). Associations between
blood lead levels and coronary artery stenosis measured using coronary computed tomography
angiography. Environ Health Perspect 129: 27006.
Kotecha. D: New. G: Flather. MP: Eccleston. D: Pepper. J: Krum. H. (2012). Five-minute heart rate variability can
predict obstructive angiographic coronary disease. Heart 98: 395-401.
Kupsco. A: Kioumourtzoglou. MA: Just. AC: Amarasiriwardena. C: Estrada-Gutierrez. G: Cantoral. A: Sanders.
AP: Braun. JM: Svensson. K: Brennan. KJM: Oken. E: Wright. RO: Baccarelli. AA: Tellez-Roio. MM.
(2019). Prenatal metal concentrations and childhood cardiometabolic risk using Bayesian kernel machine
regression to assess mixture and interaction effects. Epidemiology 30: 263-273.
Lahiri. MK: Kannankeril. PJ: Goldberger. JJ. (2008). Assessment of autonomic function in cardiovascular disease
[Review], J Am Coll Cardiol 51: 1725-1733.
Lai. CC: Lin. HH: Chen. CW: Chen. SH: Chiu. TH. (2002). Excitatory action of lead on rat sympathetic
preganglionic neurons in vitro and in vivo. Life Sci 71: 1035-1045.
Lanphear. BP: Rauch. S: Auinger. P: Allen. RW: Hornung. RW. (2018). Low-level lead exposure and mortality in
US adults: A population-based cohort study. The Lancet Public Health 3: el77-el84.
Lee. BK: Ahn. J: Kim. NS: Lee. CB: Park. J: Kim. Y. (2016a). Association of blood pressure with exposure to lead
and cadmium: Analysis of data from the 2008-2013 Korean National Health and Nutrition Examination
Survey. Biol Trace Elem Res 174: 40-51.
Lee. TH: Tseng. MC: Chen. CJ: Lin. JL. (2009). Association of high body lead store with severe intracranial carotid
atherosclerosis. Neurotoxicology 30: 876-880.
Lee. W: Yoon. JH: Roh. J: Lee. S: Seok. H: Lee. JH: Jung. PK: Rhie. J: Won. JU. (2016b). The association between
low blood lead levels and the prevalence of prehypertension among nonhypertensive adults in Korea. Am J
Hum Biol 28: 729-735.
Lin. JL: Lin-Tan. DT: Hsu. CW: Yea TH: Chen. KH: Hsu. HH: Ho. TC: Hsu. KH. (2011). Association of blood
lead levels with mortality in patients on maintenance hemodialysis. Am J Med 124: 350-358.
Lind. L: Arauio. JA: Barchowskv. A: Belcher. S: Berridge. BR: Chiamvimonvat. N: Chiu. WA: Cogliano. VJ:
Elmore. S: Farrai. AK: Gomes. AY: McHale. CM: Mever-Tamaki. KB: Posnack. NG: Vargas. HM: Yang.
X: Zeise. L: Zhou. C: Smith. MT. (2021). Key characteristics of cardiovascular toxicants. Environ Health
Perspect 129: 95001.
Lind. PM: Olsen. L: Lind. L. (2012). Elevated circulating levels of copper and nickel are found in elderly subjects
with left ventricular hypertrophy. Ecotoxicol Environ Saf 86: 66-72.
Lopes. ACB. D: Silbergeld. EK: Navas-Acien. A: Zamoiski. R: Martins. AD. Jr: Camargo. AEI: Urbano. MR:
Mesas. AE: Paoliello. MMB. (2017). Association between blood lead and blood pressure: a population-
based study in Brazilian adults. Environ Health 16: 27.
External Review Draft
4-158
DRAFT: Do Not Cite or Quote
-------
Lu. XL: Xu. XJ: Zhang. Y: Zhang. YL: Wang. CY: Huo. X. (2018). Elevated inflammatory Lp-PLA2 and IL-6 link
e-waste Pb toxicity to cardiovascular risk factors in preschool children. Environ Pollut 234: 601-609.
Lustberg. M: Silbergeld. E. (20021. Blood lead levels and mortality. Arch Intern Med 162: 2443-2449.
Lutz. PM: Wilson. TJ: Ireland. J: Jones. AL: Gorman. JS: Gale. NL: Johnson. JC: Hewett. JE. (1999). Elevated
immunoglobulin E (IgE) levels in children with exposure to environmental lead. Toxicology 134: 63-78.
Malvezzi. CK: Moreira. EG: Vassilieff. I: Vassilieff. VS: Cordellini. S. (2001). Effect of L-arginine,
dimercaptosuccinic acid (DMSA) and the association of L-arginine and DMSA on tissue lead mobilization
and blood pressure level in plumbism. Braz J Med Biol Res 34: 1341-1346.
Menke. A: Muntner. P: Batuman. V: Silbergeld. EK: Guallar. E. (2006). Blood lead below 0.48 (imol/1 (10 jxg/dl)
and mortality among US adults. Circulation 114: 1388-1394.
Miao. H: Liu. Y: Tsai. TC: Schwartz. J: Ji. JS. (2020). Association between blood lead level and uncontrolled
hypertension in the US population (NHANES 1999-2016). J Am Heart Assoc 9: e015533.
Miller. TE: Golemboski. KA: Ha. RS: Bunn. T: Sanders. FS: Dietert. RR. (1998). Developmental exposure to lead
causes persistent immunotoxicity in Fischer 344 rats. Toxicol Sci 42: 129-135.
Mohammad. A: Ali. N: Reza. B: Ali. K. (2010). Effect of ascorbic acid supplementation on nitric oxide metabolites
and systolic blood pressure in rats exposed to lead. Indian J Pharmacol 42: 78-81.
Mousavi-Mirzaei. SM: Khorasani. EY: Amirabadizadeh. A: Nakhaee. S: Baharshahi. A: Raiabpour-Sanati. A:
Talebi. A: Lamarine. RJ: Mehrpour. M: Mehrpour. O. (2020). Comparison of blood lead concentrations in
patients with acute ischemic stroke and healthy subjects. J Trace Elem Med Biol 61: 126532.
Muntner. P: Menke. A: DeSalvo. KB: Rabito. FA: Batuman. V. (2005). Continued decline in blood lead levels
among adults in the United States - The National Health and Nutrition Examination Surveys. Arch Intern
Med 165:2155-2161.
Nakhoul. F: Kavne. LH: Brautbar. N: Hu. MS: McDonough. A: Eggena. P: Golub. MS: Berger. M: Chang. CT:
Jamgotchian. N: Lee. DBN. (1992). Rapid hypertensinogenic effect of lead: Studies in the spontaneously
hypertensive rat. Toxicol Ind Health 8: 89-102.
Navas-Acien. A: Schwartz. BS: Rothenberg. SJ: Hu. H: Silbergeld. EK: Guallar. E. (2008). Bone lead levels and
blood pressure endpoints: A meta-analysis. Epidemiology 19: 496-504.
Navas-Acien. A: Selvin. E: Sharrett. AR: Calderon-Aranda. E: Silbergeld. E: Guallar. E. (2004). Lead, cadmium,
smoking, and increased risk of peripheral arterial disease. Circulation 109: 3196-3201.
Nawrot. TS: Thiis. L: DenHond. EM: Roels. HA: Staessen. JA. (2002). An epidemiological re-appraisal of the
association between blood pressure and blood lead: A meta-analysis. J Hum Hypertens 16: 123-131.
Nguyen. HP: Oh. H: Hoang. NHM: Kim. MS. (2021). Association between heavy metals, high-sensitivity C-
reaction protein and 10-year risk of cardiovascular diseases among adult Korean population. Sci Rep 11:
14664.
Ni. Z: Hou. S: Barton. CH: Vaziri. ND. (2004). Lead exposure raises superoxide and hydrogen peroxide in human
endothelial and vascular smooth muscle cells. Kidney Int 66: 2329-2336.
Notarius. CF: Butler. GC: Ando. S: Pollard. MJ: Senn. BL: Floras. JS. (1999). Dissociation between
microneurographic and heart rate variability estimates of sympathetic tone in normal subjects and patients
with heart failure. Clin Sci (Lond) 96: 557-565.
Nunes. KZ: Nunes. DO: Silveira. EA: Pereira. CAC: Filho. GBB: Vassallo. DV: Fioresi. M. (2015). Chronic lead
exposure decreases the vascular reactivity of rat aortas: The role of hydrogen peroxide. PLoS ONE 10:
e0120965.
Obeng-Gvasi. E. (2019). Lead exposure and cardiovascular disease among young and middle-aged adults. Med Sci
7: 103.
Obeng-Gvasi. E: Armiios. RX: Weigel. MM: Filippelli. GM: Savegh. MA. (2018). Cardiovascular-related outcomes
in U.S. adults exposed to lead. Int J Environ Res Public Health 15: 759.
External Review Draft
4-159
DRAFT: Do Not Cite or Quote
-------
Obeng-Gvasi. E: Ferguson. AC: Stamatakis. KA: Province. MA. (2021). Combined Effect of Lead Exposure and
Allostatic Load on Cardiovascular Disease Mortality-A Preliminary Study. Int J Environ Res Public Health
18.
Ochoa-Martinez. AC: Cardona-Lozano. ED: Carrizales-Yafiez. L: Perez-Maldonado. IN. (2018). Serum
concentrations of new predictive cardiovascular disease biomarkers in Mexican women exposed to lead.
Arch Environ Contam Toxicol 74: 248-258.
Park. SK: Hu. H: Wright. RO: Schwartz. J: Cheng. Y: Sparrow. D: Vokonas. PS: Weisskopf. MG. (2009). Iron
metabolism genes, low-level lead exposure, and QT interval. Environ Health Perspect 117: 80-85.
Park. SK: Schwartz. J: Weisskopf. M: Sparrow. D: Vokonas. PS: Wright. RO: Coull. B: Nie. H: Hu. H. (2006).
Low-level lead exposure, metabolic syndrome, and heart rate variability: The VA Normative Aging Study.
Environ Health Perspect 114: 1718-1724.
Park. Y: Han. J. (2021). Blood lead levels and cardiovascular disease risk: Results from the Korean National Health
and Nutrition Examination Survey. Int J Environ Res Public Health 18: 10315.
Perlstein. T: Weuve. J: Schwartz. J: Sparrow. D: Wright. R: Litoniua. A: Nie. H: Hu. H. (2007). Cumulative
community-level lead exposure and pulse pressure: The Normative Aging Study. Environ Health Perspect
115: 1696-1700.
Peters. JL: Kubzanskv. L: McNeelv. E: Schwartz. J: Spiro. A. Ill: Sparrow. D: Wright. RO: Nie. H: Hu. H. (2007).
Stress as a potential modifier of the impact of lead levels on blood pressure: The Normative Aging Study.
Environ Health Perspect 115: 1154-1159.
Poreba. R: Poroba. M: Gac. P: Andrzeiak. R. (2011). Ambulatory blood pressure monitoring and structural changes
in carotid arteries in normotensive workers occupationally exposed to lead. Hum Exp Toxicol 30: 1174-
1180.
Prentice. RC: Kopp. SJ. (1985). Cardiotoxicity of lead at various perfusate calcium concentrations: Functional and
metabolic responses of the perfused rat heart. Toxicol Appl Pharmacol 81: 491-501.
Prvstowskv. EN: Benson. DW. Jr: Fuster. V: Hart. RG: Kav. GN: Mverburg. RJ: Naccarelli. GV: Wvse. DG.
(1996). Management of patients with atrial fibrillation. A statement for healthcare professionals from the
Subcommittee on Electrocardiography and Electrophysiology, American Heart Association. Circulation 93:
1262-1277.
Oin. Z: Li. H: Xu. Y: Li. J: Su. B: Liao. R. (2021). Higher blood lead level is associated with increased likelihood of
abdominal aortic calcification. Front Cardiovasc Med 8: 747498.
Ou. Y: Lv. Y: Ji. S: Ding. L: Zhao. F: Zhu. Y: Zhang. W: Hu. X: Lu. Y: Li. Y: Zhang. X: Zhang. M: Yang. Y: Li. C:
Zhang. M: Li. Z: Chen. C: Zheng. L: Gu. H:... Shi, X. (2022). Effect of exposures to mixtures of lead and
various metals on hypertension, pre-hypertension, and blood pressure: A cross-sectional study from the
China National Human Biomonitoring. Environ Pollut 299: 118864.
Rautahariu. PM: Manolio. TA: Psatv. BM: Borhani. NO: Furberg. CD. (1994). Correlates of QT prolongation in
older adults (the Cardiovascular Health Study). Am J Cardiol 73: 999-1002.
Revis. NW: Zinsmeister. AR: Bull. R. (1981). Atherosclerosis and hypertension induction by lead and cadmium
ions: An effect prevented by calcium ion. Proc Natl Acad Sci USA 78: 6494-6498.
Reza. B: Ali. N: Azhdar. H: Alireza. A: Ali. K. (2008). Effects of low-level lead exposure on blood pressure and
function of the rat isolated heart. Indian J Pharmacol 40: 69-72.
Rizzi. E: Castro. MM: Fernandes. K: Barbosa. F. Jr: Arisi. GM: Garcia-Cairasco. N: Bendhack. LM: Tanus-Santos.
JE: Gerlach. RF. (2009). Evidence of early involvement of matrix metalloproteinase-2 in lead-induced
hypertension. Arch Toxicol 83: 439-449.
Robles. HV: Romo. E: Sanchez-Mendoza. A: Rios. AP: Soto. V: Avila-Casado. MC: Medina. A: Escalante. B.
(2007). Lead exposure effect on angiotensin II renal vasoconstriction. Hum Exp Toxicol 26: 499-507.
External Review Draft
4-160
DRAFT: Do Not Cite or Quote
-------
Rodriguez-Iturbe. B: Sindhu. RK: Ouiroz. Y: Vaziri. ND. (2005). Chronic exposure to low doses of lead results in
renal infiltration of immune cells, NF-kappaB activation, and overexpression of tubulointerstitial
angiotensin II. Antioxid Redox Signal 7: 1269-1274.
Roshan. YD: Assali. M: Moehaddam. AH: Hosseinzadeh. M: Mvers. J. (2011). Exercise training and antioxidants:
Effects on rat heart tissue exposed to lead acetate. Int J Toxicol 30: 190-196.
Rowan. WH. Ill: Campen. MJ: Wichers. LB: Watkinson. WP. (2007). Heart rate variability in rodents: Uses and
caveats in toxicological studies [Review]. Cardiovasc Toxicol 7: 28-51.
Roy. A: Bellinger. D: Hu. H: Schwartz. J: Ettinger. AS: Wright. RO: Bouchard. M: Palaniappan. K: Balakrishnaa
K. (2009). Lead exposure and behavior among young children in Chennai, India. Environ Health Perspect
117: 1607-1611.
Ruiz-Hernandez. A: Navas-Acien. A: Pastor-Barriuso. R: Crainiceanu. CM: Redon. J: Guallar. E: Tellez-Plaza. M.
(2017). Declining exposures to lead and cadmium contribute to explaining the reduction of cardiovascular
mortality in the US population, 1988-2004. Int J Epidemiol 46: 1903-1912.
Schober. SE: Mirel. LB: Graubard. BI: Brodv. DJ: Flegal. KM. (2006). Blood lead levels and death from all causes,
cardiovascular disease, and cancer: Results from the NHANES III Mortality Study. Environ Health
Perspect 114: 1538-1541.
Schwartz. J. (19911. Lead, blood pressure, and cardiovascular disease in men and women. Environ Health Perspect
91: 71-75.
Scinicariello. F: Abadin. HG: Edward Murray. H. (2011). Association of low-level blood lead and blood pressure in
NHANES 1999-2006. Environ Res 111: 1249-1257.
Scinicariello. F: Yesupriva. A: Chang. MH: Fowler. BA. (2010). Modification by ALAD of the association between
blood lead and blood pressure in the U.S. population: Results from the Third National Health and Nutrition
Examination Survey. Environ Health Perspect 118: 259-264.
Sharifi. AM: Darabi. R: Akbarloo. N: Lariiani. B: Khoshbaten. A. (2004). Investigation of circulatory and tissue
ACE activity during development of lead-induced hypertension. Toxicol Lett 153: 233-238.
Shin. JH: Lim. KM: Noh. JY: Bae. ON: Chung. SM: Lee. MY: Chung. JH. (2007). Lead-induced procoagulant
activation of erythrocytes through phosphatidylserine exposure may lead to thrombotic diseases. Chem Res
Toxicol 20: 38-43.
Shvachiv. L: Geraldes. V: Amaro-Leal. A: Rocha. I. (2018). Intermittent low-level lead exposure provokes anxiety,
hypertension, autonomic dysfunction and neuroinflammation. Neurotoxicology 69: 307-319.
Silva. MAS. C: de Oliveira. TF: Almenara. CCP: Broseghini-Filho. GB: Vassallo. DV: Padilha. AS: Silveira. EA.
(2015). Exposure to a low lead concentration impairs contractile machinery in rat cardiac muscle. Biol
Trace Elem Res 167: 280-287.
Silveira. EA: Lizardo. JHF: Souza. LP: Stefanon. I: Vassallo. DV. (2010). Acute lead-induced vasoconstriction in
the vascular beds of isolated perfused rat tails is endothelium-dependent. Braz J Med Biol Res 43: 492-499.
Simoes. MR: Aguado. A: Fiorim. J: Silveira. EA: Azevedo. BF: Toscano. CM: Zhenvukh. O: Briones. AM: Alonso.
MJ: Vassallo. DV: Salaices. M. (2015). MAPK pathway activation by chronic lead-exposure increases
vascular reactivity through oxidative stress/cyclooxygenase-2-dependent pathways. Toxicol Appl
Pharmacol 283: 127-138.
Simoes. MR: Ribeiro Junior. RF: Vescovi. MVA: de Jesus. HC: Padilha. AS: Stefanon. I: Vassallo. DV: Salaices.
M: Fioresi. M. (2011). Acute lead exposure increases arterial pressure: Role of the renin-angiotensin
system. PLoS ONE 6: el8730.
Skoczvnska. A: Stoiek. E. (2005). The impact of subchronic lead poisoning on the vascular effect of nitric oxide in
rats. Environ Toxicol Pharmacol 19: 99-106.
Skroder. H: Hawkesworth. S: Moore. SE: Wagatsuma. Y: Kippler. M: Vahter. M. (2016). Prenatal lead exposure
and childhood blood pressure and kidney function. Environ Res 151: 628-634.
External Review Draft
4-161
DRAFT: Do Not Cite or Quote
-------
Teve. SO: Yanoskv. JD: Cuffee. Y: Weng. X: Luauis. R: Farace. E: Wang. L. i. (2020). Association between blood
lead levels and blood pressure in American adults: Results from NHANES 1999-2016. Environ Sci Pollut
Res Int 27: 45836-45843.
Tonelli. M: Wiebe. N: Bello. A: Field. CJ: Gill. JS: Hemmelgarn. BR: Holmes. DT: Jindal. K: Klarenbach. SW:
Manns. BJ: Thadhani. R: Kinniburgh. D. (2018). Concentrations of trace elements and clinical outcomes in
hemodialysis patients: A prospective cohort study. Clin J Am Soc Nephrol 13: 907-915.
Tsao. DA: Yu. HS: Cheng. JT: Ho. CK: Chang. HR. (2000). The change of beta-adrenergic system in lead-induced
hypertension. Toxicol Appl Pharmacol 164: 127-133.
Tsoi. MF: Lo. CWH: Cheung. TT: Cheung. BMY. (2021). Blood lead level and risk of hypertension in the United
States National Health and Nutrition Examination Survey 1999-2016. Sci Rep 11: 3010.
U.S. EPA (U.S. Environmental Protection Agency). (1990). Air quality criteria for lead: Supplement to the 1986
addendum [EPA Report]. (EPA/600/8-89/049F). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2006a). Air quality criteria for lead (Final report, 2006):
Volume I of II [EPA Report], (EPA/600/R-05/144aF). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for ozone and related
photochemical oxidants. (EPA/600/R-05/004aF-cF). Research Triangle Park, NC.
U.S. EPA (U.S. Environmental Protection Agency). (2013a). Integrated science assessment for lead [EPA Report].
(EPA/600/R-10/075F). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2013b). Integrated science assessment for lead (contains errata
sheet created 5/12/2014) [EPA Report], (EPA/600/R-10/075F). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2015). Preamble to the Integrated Science Assessments [EPA
Report]. (EPA/600/R-15/067). Research Triangle Park, NC: U.S. Environmental Protection Agency, Office
of Research and Development, National Center for Environmental Assessment, RTP Division.
U.S. EPA (U.S. Environmental Protection Agency). (2019). Integrated Science Assessment (ISA) for particulate
matter (final report, Dec 2019) [EPA Report]. (EPA/600/R-19/188). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2020). Integrated Science Assessment (ISA) for ozone and
related photochemical oxidants (final report, Apr 2020) [EPA Report]. (EPA/600/R-20/012). Washington,
DC.
van Bemmel. DM: Li. Y: McLean. J: Chang. MH: Dowling. NF: Graubard. B: Raiaraman. P. (2011). Blood lead
levels, ALAD gene polymorphisms, and mortality. Epidemiology 22: 273-278.
Vaziri. ND: Ding. Y: Ni. Z: Gonick. HC. (1997). Altered nitric oxide metabolism and increased oxygen free radical
activity in lead-induced hypertension: Effect of lazaroid therapy. Kidney Int 52: 1042-1046.
Vaziri. ND: Liang. K: Ding. Y. (1999). Increased nitric oxide inactivation by reactive oxygen species in lead-
induced hypertension. Kidney Int 56: 1492-1498.
Wan. H: Chen. S: Cai. Y: Chen. Y: Wang. Y: Zhang. W: Chen. C: Wang. N: Guo. Y: Lu. Y. (2021). Lead exposure
and its association with cardiovascular disease and diabetic kidney disease in middle-aged and elderly
diabetic patients. Int J Hyg Environ Health 231: 113663.
Wang. H: Li. X: Li. RJ: Yan. J: Lan. Z: Chen. J: Zhang. L. (2020). Associations of exposure to metals with the risk
of hypertension among an older population aged 40-75 years in rural southwest China. J Appl Toxicol 40:
1076-1086.
Weaver. VM: Ellis. LR: Lee. BK: Todd. AC: Shi. W: Ahn. KD: Schwartz. BS. (2008). Associations between patella
lead and blood pressure in lead workers. Am J Ind Med 51: 336-343.
Weisskopf. MG: Jain. N: Nie. HL: Sparrow. D: Vokonas. P: Schwartz. J: Hu. H. (2009). A prospective study of
bone lead concentration and death from all causes, cardiovascular diseases, and cancer in the department of
veterans affairs normative aging study. Circulation 120: 1056-1064.
External Review Draft
4-162
DRAFT: Do Not Cite or Quote
-------
Wildemann. TM: Weber. LP: Siciliano. SD. (20151. Combined exposure to lead, inorganic mercury and
methylmercury shows deviation from additivity for cardiovascular toxicity in rats. J Appl Toxicol 35: 918-
926.
Xu. C: Shu. Y: Fu. Z: Hu. Y: Mo. X. (20171. Associations between lead concentrations and cardiovascular risk
factors in U.S. adolescents. Sci Rep 7: 9121.
Xu. J: Engel. LS: Rhodea J: Jackson. WB. II: Kwok. RK: Sandler. DP. (2021). The association between blood
metals and hypertension in the GuLF study. Environ Res 202: 111734.
Xu. LH: Mu. FF: Zhao. JH: He. O: Cao. CL: Yang. H: Liu. O: Liu. XH: Sun. SJ. (2015). Lead induces apoptosis and
histone hyperacetylation in rat cardiovascular tissues. PLoS ONE 10: e0129091.
Yan. LP: Rouzier. V: Pierre. JL: Lee. MH: Muntner. P: Parsons. PJ: Apollon. A: St-Preux. S: Malebranche. R:
Pierre. G: Emmanuel. E: Nash. D: Kingerv. J: Walsh. KF: Smith. CE: Metz. M: Tvmeiczvk. O:
Deschamps. M: Pape. JW: ... McNairv. ML. (2022). High lead exposure associated with higher blood
pressure in Haiti: A warning sign for low-income countries. Hypertension 79: 283-290.
Yang. WY: Zhang. ZY: Thiis. L: Cauwenberghs. N: Wei. FF: Jacobs. L: Luttun. A: Verhamme. P: Kuznetsova. T:
Nawrot. TS: Staessen. JA. (2017). Left ventricular structure and function in relation to environmental
exposure to lead and cadmium. J Am Heart Assoc 6: e004692.
Yao. B: Lu. X: Xu. L: Wang. Y: Ou. H: Zhou. H. (2020). Relationship between low-level lead, cadmium and
mercury exposures and blood pressure in children and adolescents aged 8-17 years: An exposure-response
analysis of NHANES 2007-2016. Sci Total Environ 726: 138446.
Yu. CG: Wei. FF: Zhang. ZY: Thiis. L: Yang. WY: Muiai. B: Feng. YM: Boggia. J: Roels. HA: Struiiker-Boudier.
HAJ: Nawrot. TS: Verhamme. P: Staessen. JA. (2020). Central hemodynamics in relation to low-level
environmental lead exposure. Blood Press 29: 157-167.
Zeller. I: Knoflach. M: Seubert. A: Kreutmaver. SB: Stelzmtiller. ME: Wallnoefer. E: Blunder. S: Frotschnig. S:
Messner. B: Willeit. J: Debbage. P: Wick. G: Kiechl. S: Laufer. G: Bernhard. D. (2010). Lead contributes
to arterial intimal hyperplasia through nuclear factor erythroid 2-related factor-mediated endothelial
interleukin 8 synthesis and subsequent invasion of smooth muscle cells. Arterioscler Thromb Vase Biol 30:
1733-1740.
Zhang. A: Hu. H: Sanchez. BN: Ettinger. AS: Park. SK: Cantonwine. D: Schnaas. L: Wright. RO: Lamadrid-
Figueroa. H: Tellez-Roio. MM. (2012). Association between prenatal lead exposure and blood pressure in
female offspring. Environ Health Perspect 120: 445-450.
Zhang. A: Park. SK: Wright. RO: Weisskopf. MG: Mukheriee. B: Nie. H: Sparrow. D: Hu. H. (2010). HFE H63D
polymorphism as a modifier of the effect of cumulative lead exposure on pulse pressure: The Normative
Aging Study. Environ Health Perspect 118: 1261-1266.
Zhang. LF: Peng. SO: Sheng. W: Li. BL: Gang. H: Dong. YS. (2007). Direct effects of lead (Pb2+) on the relaxation
of in vitro cultured rat aorta to acetylcholine. Toxicol Lett 170: 104-110.
Zhang. LF: Peng. SO: Wang. S. (2005). Influence of lead (Pb2+) on the reactions of in vitro cultured rat aorta to 5-
hydroxytryptamine. Toxicol Lett 159: 71-82.
Zhang. LF: Peng. SO: Wang. S. (2009). Decreased aortic contractile reaction to 5-hydroxytryptamine in rats with
long-term hypertension induced by lead (Pb2+) exposure. Toxicol Lett 186: 78-83.
Zhang. M: Liu. T: Wang. G: Buckley. JP: Guallar. E: Hong. X: Wang. MC: Wills-Karp. M: Wang. X: Mueller. NT.
(2021). In utero exposure to heavy metals and trace elements and childhood blood pressure in a U.S. urban,
low-income, minority birth cohort. Environ Health Perspect 129: 67005.
Zheutlin. AR: Hu. H: Weisskopf. MG: Sparrow. D: Vokonas. PS: Park. SK. (2018). Low-level cumulative lead and
resistant hypertension: A prospective study of men participating in the Veterans Affairs Normative Aging
Study. J Am Heart Assoc 7: e010014.
Zhu. G: Chen. Z: Dai. B: Zheng. C: Jiang. H: Xu. Y: Sheng. X: Guo. J: Dan. Y: Liang. S: Li. G. (2018). Chronic
lead exposure enhances the sympathoexcitatory response associated with P2X4 receptor in rat stellate
ganglia. Environ Toxicol 33: 631-639.
External Review Draft
4-163
DRAFT: Do Not Cite or Quote
-------
Zhu. G: Dai. B: Chen. Z: He. L: Guo. J: Dan. Y: Liang. S: Li. G. (2019). Effects of chronic lead exposure on the
sympathoexcitatory response associated with the P2X7 receptor in rat superior cervical ganglia. Auton
Neurosci 219: 33-41.
Zota. AR: Shenassa. ED: Morello-Frosch. R. (2013). Allostatic load amplifies the effect of blood lead levels on
elevated blood pressure among middle-aged U.S. adults: A cross-sectional study. Environ Health 12: 64
External Review Draft
4-164
DRAFT: Do Not Cite or Quote
------- |