Subscriber access provided by Northern Illinois University
Article
A novel method for calculating potency-weighted cumulative phthalates exposure with implications for identifying racial/ethnic disparities among U.S. reproductive-aged women in NHANES 2001-2012 Julia R Varshavsky, Ami R Zota, and Tracey J. Woodruff Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b00522 • Publication Date (Web): 31 Aug 2016 Downloaded from http://pubs.acs.org on August 31, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 36
Environmental Science & Technology
1
A novel method for calculating potency-weighted cumulative phthalates exposure
2
with implications for identifying racial/ethnic disparities among U.S. reproductive-
3
aged women in NHANES 2001-2012
4 5
Julia R. Varshavsky*1, Ami R. Zota2, and Tracey J. Woodruff3
6 7
1 Division
8
California, Berkeley, Berkeley, CA, 94720, USA
9
2 Department
of Environmental Health Sciences, School of Public Health, University of
of Environmental and Occupational Health, Milken Institute School of
10
Public Health, George Washington University, Washington DC, 20052, USA
11
3 Program
12
Gynecology, and Reproductive Sciences, Institute for Health Policy Studies, University
13
of California, San Francisco, San Francisco, California, 94143, USA
on Reproductive Health and the Environment, Department of Obstetrics,
14 15 16
*Corresponding author:
17 18 19 20 21 22 23
Julia Varshavsky, MPH Division of Environmental Health Sciences School of Public Health University of California, Berkeley 50 University Hall, Room 221 Berkeley, CA 94720 Email:
[email protected]; Phone: 510.384.0169
1 ACS Paragon Plus Environment
Environmental Science & Technology
24
TOC Abstract Art
Page 2 of 36
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
44 45 46 47 48
2 ACS Paragon Plus Environment
Page 3 of 36
Environmental Science & Technology
49 50 51
Abstract
52
affect reproduction and development. Multiple anti-androgenic phthalates
53
exposure during fetal development can have greater impacts than individual
54
exposure; thus, the National Academy of Sciences (NAS) recommends them for
55
cumulative assessment. Using National Health and Nutrition Examination Survey
56
data (NHANES, 2001-2012), we developed a potency-weighted sum of daily
57
intake (∑androgen-disruptor; µg/kg/day) of di-n-butyl phthalate (DnBP), diisobutyl
58
phthalate (DiBP), butyl benzyl phthalate (BBzP), and di(2-ethylhexyl) phthalate
59
(DEHP) based on NAS recommendations, and included diethyl phthalate (DEP)
60
and diisononyl phthalate (DiNP) in additional metrics (2005-2012). We compared
61
racial/ethnic differences in ∑androgen-disruptor among 2842 reproductive-aged
62
women. In sensitivity analyses, we assessed the influence of potency
63
assumptions, alternate urine dilution adjustment methods, and weighting
64
phthalate metabolites directly rather than daily intake estimates of parent
65
compounds. We found that DEHP contributed most to ∑androgen-disruptor (48-
66
64%), and that ∑androgen-disruptor decreased over time. Black women
67
generally had higher cumulative exposures than white women, although the
68
magnitude and precision of the difference varied by model specification. Our
69
approach provides a blueprint for combining chemical exposures linked to
70
common adverse outcomes, and should be considered in future exposure, risk,
71
and epidemiological studies.
Phthalates are ubiquitous chemicals linked to hormonal disruptions that
72
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 36
73 74 75
INTRODUCTION
76
of health outcomes. Fetal phthalate exposures cause a group of male
77
reproductive problems known as the “phthalate syndrome” in animals, which
78
includes birth defects of the testes and penis (e.g. cryptorchidism, hypospadias),
79
low sperm count, and infertility.1 Human studies support an association between
80
developmental phthalate exposures and male reproductive effects.2-6 Certain
81
phthalates, such as di-n-butyl phthalate (DnBP), diisobutyl phthalate (DiBP), butyl
82
benzyl phthalate (BBzP), di(2-ethylhexyl) phthalate (DEHP), and diisononyl
83
phthalate (DiNP), exert toxicity primarily through androgen disruption.1
84
Phthalates esters are hormonally active chemicals linked to a wide range
Phthalate exposures are ubiquitous due to their widespread use in myriad
85
consumer and personal care products.7, 8 Biomonitoring studies show the U.S.
86
population is exposed to multiple phthalates simultaneously.9-12 Animal studies
87
demonstrate that phthalate mixtures result in higher male reproductive risk than
88
individual phthalates, especially during fetal development.13-15 Human studies
89
also find higher risks from multiple phthalates.2, 5 In 2008, the National Academy
90
of Sciences (NAS) recommended phthalates and other anti-androgens for
91
cumulative risk assessment based on their ability to cause common adverse
92
outcomes (i.e. changes in testosterone concentration during development) rather
93
than shared mechanisms of action (i.e. how testosterone concentration is
94
disrupted).1 More recent mixture studies confirm NAS findings and further
95
highlight the need to consider the joint effects of co-occurring phthalates.16-19
4 ACS Paragon Plus Environment
Page 5 of 36
Environmental Science & Technology
96
However, advancements in the epidemiology of phthalate mixtures are limited by
97
current methods for characterizing cumulative exposures.20
98 99
Identifying high-risk subpopulations for cumulative exposure warrants examination since individual phthalate profiles may not accurately represent the
100
distribution of overall phthalate burden. Individual phthalate exposures have been
101
shown to vary by race/ethnicity and socioeconomic status (SES), often in
102
opposing directions.9, 21-24 For example, DnBP and BBzP are higher in low SES
103
groups, while DEHP exposure is lower in the same subpopulation.22, 23
104
Characterizing cumulative exposure inequalities may help elucidate health
105
disparities and identify solutions that better protect those at increased risk of
106
exposure and disease.25
107
In this paper, we develop a potency-weighted metric of androgen-
108
disrupting phthalates using 2008 NAS recommendations and examine
109
demographic differences in cumulative phthalates exposure among U.S.
110
reproductive-aged women. We focus on women of reproductive age because in
111
utero development has been identified as the most sensitive window for
112
phthalate toxicity,1 and previous work suggests that phthalate exposures among
113
reproductive-aged and pregnant women are similar12 but often distinct from other
114
subpopulations.26
115 116
MATERIALS AND METHODS
117
Study Population. We pooled data from six cycles (2001-2012) of the National
118
Health and Nutrition Examination Survey (NHANES), a nationally representative
5 ACS Paragon Plus Environment
Environmental Science & Technology
119
survey and physical examination of the civilian, non-institutionalized U.S.
120
population conducted by the U.S. Centers for Disease Control (CDC)
121
(http://www.cdc.gov/nchs/nhanes.htm). The study population was limited to
122
females aged 15-44 years (n=3480). Of these, we further excluded 544 women
123
who self-identified as a race/ethnicity other than non-Hispanic white, non-
124
Hispanic black, or Mexican American, who together comprise 86% of the total
125
population. We also excluded participants who did not have at least one
126
phthalate metabolite measurement (n=94), resulting in a final sample size of
127
2842 participants.
Page 6 of 36
128 129
Phthalate Metabolite Measurements. In each NHANES survey cycle, phthalate
130
metabolites are measured in one-third of survey participants. Analytical methods
131
are described elsewhere.27 Briefly, spot urine samples are collected as part of the
132
NHANES medical examination and analyzed at the CDC’s National Center for
133
Environmental Health (Atlanta, GA). Phthalate metabolites are quantified using
134
high performance liquid chromatography coupled with tandem mass
135
spectrometry.27, 28 Laboratory files were downloaded from the NHANES website
136
in March 2015 and included necessary impurity corrections for some previously
137
used analytical standards.29 Not all phthalate metabolites are measured in every
138
NHANES cycle, and limits of detection (LOD) for each metabolite can vary by
139
cycle. Thus, we used the maximum LOD for each phthalate metabolite and
140
substituted concentrations below the LOD with a value equal to the LOD divided
141
by the square root of two.11, 30
6 ACS Paragon Plus Environment
Page 7 of 36
Environmental Science & Technology
142 143
Cumulative Exposure Metric. To characterize cumulative exposure to androgen-
144
disrupting phthalates, we calculated a potency-weighted aggregate sum
145
(∑androgen-disruptor) of four phthalates (DnBP, DiBP, BBzP, and DEHP) that
146
are recognized as anti-androgenic by the NAS and whose metabolites were
147
measured in every cycle between 2001-2012 (Figure 1; model 1). The primary
148
hydrolytic metabolites for DnBP, DiBP, BBzP, and DEHP, respectively, are
149
mono-n-butyl phthalate (MnBP), mono-isobutyl phthalate (MiBP), mono-benzyl
150
phthalate (MBzP), and mono-(2-ethylhexyl) phthalate (MEHP). MEHP further
151
metabolizes to oxidative metabolites: mono-(2-ethyl-5-oxohexyl) phthalate
152
(MEOHP), mono-(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), and mono-(2-
153
ethyl-5-carboxypentyl) phthalate (MECPP).
154
To calculate unitless relative potency factors (RPFs) for the four
155
phthalates, we used NAS phthalate-specific benchmark doses (BMDs)
156
(mg/kg/day), which are statistically derived doses related to a predefined change
157
from controls in benchmark response (BMR) (in this case a 5% reduction in
158
testosterone concentration, or a BMR equal to 5%) (Table 1).1, 31 BMDs were
159
used rather than the lower limit of a one-sided 95% confidence interval (CI) on
160
the BMD (BMDL) because we are comparing across chemicals.31 In the NAS
161
analysis, DnBP had the highest potency, or lowest BMD. Therefore, we derived
162
RPFs by dividing the DnBP reference BMD by that of each phthalate.
163
RPF i = (BMD Reference / BMD i ); i = individual phthalate (Eq. 1)
7 ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 36
164
Because BMDs are based on phthalate doses administered to laboratory animals
165
and not metabolite concentrations, we estimated the daily intake of parent
166
phthalate compounds (μg/kg/day) from measured urinary metabolites (ng/mL =
167
μg/L) using a pharmacokinetic modeling equation adapted from previous
168
studies.32, 33
169 170 171 172
where ME i is the urinary concentration of metabolite per gram of creatinine (μg/g)
173
for each phthalate, CE is the creatinine excretion rate normalized by body weight
174
(mg/kg/day), F UE,i is the molar fraction of urinary excreted metabolite related to
175
parent compound for each phthalate (unitless), 1000 is a conversion factor (i.e.
176
mg/g), and MW p and MW m are the molecular weights of the parent phthalates
177
and metabolites, respectively.
178
Daily Intake i = (ME i x CE) x (MW p ) (Eq. 2) (F UE,i x 1000) x (MW m )
Although creatinine excretion rates vary across racial/ethnic groups, we
179
initially assumed a uniform creatinine excretion rate of 18 mg/kg/day34 for all
180
participants. We calculated creatinine-adjusted concentrations (ME) for each
181
participant by dividing their measured urinary concentrations (μg/L) by their
182
measured urinary creatinine concentration (g/L). The fractional urinary excretion
183
values, or F UE , are 0.69, 0.69, and 0.73 for DnBP, DiBP, and BBzP,
184
respectively.35, 36 The values for MEHP, MEHHP, and MEOHP are 0.062, 0.149,
185
and 0.109, respectively. MECPP was not included as a DEHP metabolite for
186
2001-2012 analyses because it was not measured in all survey cycles, but it was
187
included in 2005-2012 analyses with an F UE of 0.132.37 Potency-weighted daily
188
intake estimates of ∑androgen-disruptor, expressed in μg/kg/day, were then 8 ACS Paragon Plus Environment
Page 9 of 36
Environmental Science & Technology
189
calculated by summing the products of RPFs and daily intakes for each phthalate
190
(See Figure S1 for sample calculations).
191
∑androgen-disruptor = ∑(Daily Intake i x RPF i ) (Eq. 3)
192
In addition to our original metric (Figure 1; model 1), we calculated a
193
second version of ∑androgen-disruptor by adding diethyl phthalate (DEP) and
194
DiNP to model 1 (Figure 1; model 2a). While DEP does not exhibit anti-
195
androgenic properties in laboratory studies,35 several epidemiologic studies
196
report an association between MEP and male reproductive health endpoints,
197
although findings are inconsistent.3, 5, 6 Therefore, we assumed DEP to have a
198
low but non-zero potency equal to one order of magnitude less than the least
199
potent phthalate in our model (RPF = 1/10 multiplied by 0.24, or 0.024). Mono-
200
ethyl phthalate (MEP) is the primary metabolite of DEP, and DEP was assumed
201
to have the same F UE as DnBP.32 We also added DiNP to the metric since it has
202
been recognized as an anti-androgenic phthalate.1, 35 Since DiNP’s primary
203
metabolite, mono(carboxy-isooctyl) phthalate (MCOP), was not measured before
204
2005, model 2a was examined in 2005-2012 participants only (N=1723). DiNP
205
was assumed to be 2.3 times less potent than DEHP.18 We used an F UE value of
206
0.099 for MCOP.37
207 208
Data Analysis. We conducted analyses using SAS software, Version 9.4 (SAS
209
Institute Inc., Cary, NC). Since we combined multiple cycles of data, we
210
calculated new sample population weights according to NHANES analytical
211
guidelines.38 All analyses were adjusted for sample population weights and the
9 ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 36
212
NHANES clustered sample design. We defined statistical significance at p