Association of Maternal Exposure to Persistent Organic Pollutants in Early Pregnancy With Fetal Growth

Marion Ouidir; Germaine M Buck Louis; Jenna Kanner; Katherine L Grantz; Cuilin Zhang; Rajeshwari Sundaram; Mohammad L Rahman; Sunmi Lee; Kurunthachalam Kannan; Fasil Tekola-Ayele; Pauline Mendola

doi:10.1001/jamapediatrics.2019.5104

. 2019 Dec 30;174(2):149–161. doi: 10.1001/jamapediatrics.2019.5104

Association of Maternal Exposure to Persistent Organic Pollutants in Early Pregnancy With Fetal Growth

Marion Ouidir ¹, Germaine M Buck Louis ², Jenna Kanner ¹, Katherine L Grantz ¹, Cuilin Zhang ¹, Rajeshwari Sundaram ³, Mohammad L Rahman ⁴, Sunmi Lee ⁵, Kurunthachalam Kannan ⁵, Fasil Tekola-Ayele ¹, Pauline Mendola ^1,^✉

¹Epidemiology Branch, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), National Institutes of Health, Bethesda, Maryland

²College of Health and Human Services, George Mason University, Fairfax, Virginia

³Biostatistics and Bioinformatics Branch, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland

⁴Department of Population Medicine, Harvard Pilgrim Health Care Institute, Harvard Medical School, Boston, Massachusetts

⁵Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, University at Albany School of Public Health, State University of New York, Albany

Accepted for Publication: July 27, 2019.

^✉

Corresponding Author: Pauline Mendola, PhD, Epidemiology Branch, Division of Intramural Population Health Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, 6710B Rockledge Dr, Room 3119, Bethesda, MD 20892 (pauline.mendola@nih.gov).

Published Online: December 30, 2019. doi:10.1001/jamapediatrics.2019.5104

Author Contributions: Drs Ouidir and Mendola had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ouidir, Buck Louis, Rahman, Tekola-Ayele, Mendola.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Ouidir, Lee.

Critical revision of the manuscript for important intellectual content: Buck Louis, Kanner, Grantz, Zhang, Sundaram, Rahman, Kannan, Tekola-Ayele, Mendola.

Statistical analysis: Ouidir, Sundaram, Rahman.

Obtained funding: Buck Louis, Zhang, Kannan.

Administrative, technical, or material support: Buck Louis, Kanner, Grantz, Zhang, Kannan, Mendola.

Supervision: Tekola-Ayele, Mendola.

Principal investigator for the NICHD Fetal Growth Study: Buck Louis.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was funded by contracts HHSN275200800013C, HHSN275200800002I, HHSN27500006, HHSN275200800003IC, HHSN275200800014C, HHSN275200800012C, HHSN275200800028C, HHSN275201000009C, and HHSN27500008 from the National Institutes of Health Intramural Research Program at the Eunice Kennedy Shriver NICHD.

Role of the Funder/Sponsor: The manuscript was cleared by the NICHD before submission, but this funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We acknowledge the research teams at all participating clinical study sites: Christina Care Health System; Columbia University; Fountain Valley Hospital; Long Beach Memorial Medical Center; New York Hospital, Queens; Northwestern University; University of Alabama at Birmingham; University of California, Irvine; Medical University of South Carolina; Saint Peters University Hospital; Tufts University; and Women and Infants Hospital of Rhode Island. The Wadsworth Center, Clinical Trials and Surveys Corporation, and The EMMES Company provided data and imaging support.

^✉

Corresponding author.

PMCID: PMC6990715 PMID: 31886849

This cohort study analyzes the concentration of toxic chemicals in racially/ethnically diverse women between 8 and 13 weeks of pregnancy and the associations of this exposure with fetal development.

Key Points

Question

Are maternal plasma levels of persistent organic pollutants in early pregnancy associated with fetal growth, and do maternal race/ethnicity or infant sex factor into this association?

Findings

In this cohort study of 2284 low-risk pregnant women from 4 racial/ethnic groups (white, black, Hispanic, and Asian), mixtures of organochlorine pesticides, dioxin-like polychlorinated biphenyls, and high levels of polybrominated diphenyl ethers were negatively associated with fetal growth measures. Although different patterns of associations by infant sex were found, the results did not vary substantively by maternal race/ethnicity.

Meaning

Findings of this study suggested that maternal exposure to specific persistent organic pollutant mixtures may reduce fetal growth, and this association is apparent even at low levels of exposure.

Abstract

Importance

Prenatal exposure to persistent organic pollutants (POPs) has been associated with birth size, but data on fetal growth and among racially/ethnically diverse pregnant women remain scarce.

Objectives

To assess the association between maternal plasma POPs in early pregnancy and fetal growth and by infant sex and maternal race/ethnicity.

Design, Setting, and Participants

This cohort study used the National Institute of Child Health and Human Development Fetal Growth Studies–Singleton cohort, which recruited nonobese, low-risk pregnant women before 14 weeks’ gestation between July 1, 2009, and January 31, 2013, in 12 community-based clinics throughout the United States. Participants self-identified their race/ethnicity, self-reported their behavioral risk factors, and were followed up throughout their pregnancy. Data were analyzed from July 31, 2018, to June 3, 2019.

Exposures

Levels of 76 POPs in early gestation plasma were measured: 11 perfluoroalkyl and polyfluoroalkyl substances, 1 polybrominated biphenyl, 9 polybrominated diphenyl ethers (PBDEs), 44 polychlorinated biphenyls (PCBs), and 11 organochlorine pesticides (OCPs). The bayesian kernel machine regression method was used to examine chemical class mixtures, and generalized additive mixed model was used to analyze individual chemicals.

Main Outcomes and Measures

Fourteen fetal biometrics were measured, including head circumference, abdominal circumference, and femur length, within 5 ultrasonography appointments.

Results

A total of 2284 low-risk pregnant women were included: 606 women (26.5%) self-identified as white with a mean (SD) age of 30.3 (4.4) years, 589 (25.8%) as black with a mean (SD) age of 25.5 (5.5) years, 635 (27.8%) as Hispanic with a mean (SD) age of 27.1 (5.5) years, and 454 (19.9%) as Asian with a mean (SD) age of 30.5 (4.5) years. A comparison between the 75th and 25th percentile of exposure revealed that the OCP mixture was negatively associated with most fetal growth measures, with a reduction of 4.7 mm (95% CI, −6.7 to −2.8 mm) in head circumference, 3.5 mm (95% CI, −4.7 to −2.2 mm) in abdominal circumference, and 0.6 mm (95% CI, −1.1 to −0.2 mm) in femur length. Higher exposure to the PBDE mixture was associated with reduced abdominal circumference (–2.4 mm; 95% CI, −4.0 to −0.5 mm) and femur length (−0.5 mm; 95% CI, −1.0 to −0.1 mm), and the dioxin-like PCB mixture was associated with reduced head circumference (–6.4 mm; 95% CI, −8.4 to −4.3 mm) and abdominal circumference (–2.4 mm; 95% CI, −3.9 to −0.8 mm). Associations with individual chemicals were less consistent. There were some interactions by fetal sex, although most of the results did not vary by maternal race/ethnicity. For example, oxychlordane (–0.98 mm; 95% CI, –1.60 to –0.36 mm; P for interaction <.001), trans-nonachlor (–0.31 mm; 95% CI, –0.54 to –0.08 mm; P for interaction = .005), and p,p’-dichlorodiphenyldichloroethylene (–0.19 mm; 95% CI, –0.22 to –0.09 mm; P for interaction = .006) were associated with shorter femur length among boys only.

Conclusions and Relevance

This study found that, among pregnant women with low POP levels, a mixture of OCPs was negatively associated with most fetal growth measures and that mixtures of PBDEs and dioxin-like PCBs were associated with reduced abdominal circumference. These findings suggested that, although exposures may be low, associations with fetal growth are apparent.

Introduction

Persistent organic pollutants (POPs) are toxic chemicals that can adversely affect human health. Efforts have been made to eliminate their production and use, but their long half-lives (which can be more than 10 years)¹ result in nearly ubiquitous exposure, including detection in pregnant women in the United States.² Current US sources of exposure are mostly dietary, including consumption of dairy products and fish. Chemical pollutants can cross the placenta and are present in umbilical cord blood,^3,4,5 and maternal plasma levels in early pregnancy reflect fetal exposure.⁶

Previous studies of the association between maternal POP exposure during pregnancy and birth weight have been largely inconsistent.^7,8,9,10 Most studies used birth weight and small for gestational age at birth as proxies for intrauterine growth.^11,12,13 However, birth weight can be an insufficient measure of fetal growth, failing to differentiate constitutionally small fetuses from growth-restricted fetuses of the same birth weight.¹⁴

The association between POPs and fetal growth may be modified by characteristics such as maternal race/ethnicity and fetal sex. For example, a negative association between perfluorononanoic acid and birth weight was observed in girls but not in boys,¹⁵ whereas perfluorooctanesulfonic acid was associated with low birth weight risk in girls but not in boys.⁸ One study concluded that an inverse association existed between polychlorinated biphenyl (PCB) concentrations and birth weight that varied by race/ethnicity.¹⁶

The Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies included low-risk pregnant women without major chronic diseases and from varied racial/ethnic backgrounds. The singleton cohort of the NICHD Fetal Growth Studies provides a notable opportunity to examine the association between chemicals and fetal growth along with potential effect modification by fetal sex and maternal race/ethnicity among low-risk pregnant women using well-implemented, standardized obstetrical ultrasonography protocols.¹⁷

Furthermore, previous studies have investigated POPs as individual chemicals, although exposure involves a mixture of chemicals simultaneously.² Consequently, assessing the association of chemicals with fetal growth using mixture analyses is needed¹⁸ to adequately consider the complexity of the implications of chemical mixtures for fetal growth, addressing highly correlated chemicals and their potential interaction as well as nonlinear associations.

The aim of this cohort study was to assess the association between maternal prenatal plasma POP levels and measured fetal growth and to evaluate the potential effect modification by infant sex and maternal race/ethnicity.

Methods

The NICHD Fetal Growth Studies–Singleton cohort included nonobese pregnant women with low-risk antenatal profiles who were recruited between July 1, 2009, and January 31, 2013, to provide data on fetal growth for 4 US racial/ethnic groups (non-Hispanic white, non-Hispanic black, Hispanic, and Asian).¹⁹ The women enrolled were all between 8 weeks 0 days and 13 weeks 6 days of pregnancy and were from 12 clinical study sites across the United States (Christina Care Health System, Newark, Delaware; Columbia University, New York, New York; Fountain Valley Hospital, Fountain Valley, California; Long Beach Memorial Medical Center, Long Beach, California; New York Hospital, Queens, Flushing, New York; Northwestern University, Chicago, Illinois; University of Alabama, Birmingham, Alabama; University of California, Irvine, Orange, California; Medical University of South Carolina, Charleston, South Carolina; Saint Peters University Hospital, New Brunswick, New Jersey; Tufts University, Medford, Massachusetts; and Women and Infants Hospital, Providence, Rhode Island). All participants provided written informed consent. This cohort study was approved by the institutional review boards at the National Institutes of Health, clinical study sites, and data coordinating centers.

Race/ethnicity was self-identified by the participants. To ensure low-risk status, we excluded women with chronic disease; previous pregnancy complications; and self-reported behavioral risk factors such as use of cigarettes, illicit drugs, or alcohol in the months before pregnancy, as fully defined elsewhere.¹⁹ Among the 2334 pregnant women enrolled in the study, we excluded 15 (0.6%) found ineligible after enrollment, 28 (1.2%) who experienced a pregnancy loss, and 7 (0.3%) with missing ultrasonography data (eFigure 1 in the Supplement). The remaining 2284 pregnant women in the final analytic sample were similar in characteristics and POP exposure to the original NICHD Fetal Growth Studies cohort (eTables 1 and 2 in the Supplement).

Fetal Growth Data

After the first ultrasonography to confirm gestational age with self-reported last menstrual period, participants were randomized to 1 of 4 schedules for 5 ultrasonography appointments.¹⁷ This randomization scheme was designed to capture gestational weeks 16 to 40 without having to expose women to weekly ultrasonography. Measurement of fetal growth was performed using standardized obstetrical ultrasonography protocols and identical equipment (Voluson E8; GE Healthcare).¹⁷ All dedicated sonographers underwent an intensive training and evaluation period.²⁰ The quality control showed that measurements between site sonographers and experts had high correlation (>0.99) and a low coefficient of variation (<3%).²¹

Fourteen fetal growth biometrics were measuree in millimeters, including head measurements (circumference, biparietal diameter, occipital-frontal diameter, cerebral width [ie, transcerebellar diameter], and inner and outer orbit diameters), bone lengths (femur, humerus, radius, ulna, tibia, fibula, and foot), and body size (abdominal circumference and estimated fetal weight [in grams] using the Hadlock formula^22,23). Given that head, bone, and body size growth measures were generally similar, we focused the main findings on head circumference, femur length, and abdominal circumference (other measures are detailed in the eTables 8-21 and eFigures 2-6 in the Supplement).

Environmental Exposure Data

A total of 76 POPs were measured in plasma samples collected at the entry visit as described in a previous publication.¹⁰ These POPs included 11 persistent nonlipophilic chemicals (perfluoroalkyl and polyfluoroalkyl substances [PFASs], including N-methylperfluoro-1-octanesulfonamidoacetic acid, perfluorodecanoic acid, perfluorododecanoic acid, perfluorodecane sulfonate, perfluoroheptanoic acid, perfluorohexanesulfonic acid, perfluorononanoic acid, perfluorooctanoic acid, perfluorooctanesulfonic acid, perfluorooctanesulfonamide, and perfluoroundecanoic acid) and 65 persistent lipophilic chemicals (1 polybrominated biphenyl [153]; 9 polybrominated diphenyl ethers [PBDE] congeners 28, 47, 85, 99, 100, 153, 154, 183, and 209]; 44 polychlorinated biphenyls [PCBs], including tri- to deca-chlorobiphenyls; and 11 organochlorine pesticides [OCPs], including β-hexachlorocyclohexane, γ-hexachlorocyclohexane, hexachlorobenzene, oxychlordane, trans-chlordane, trans-nonachlor, p,p′-dichlorodiphenyldichloroethylene, o,p′- dichlorodiphenyldichloroethane, p,p′-dichlorodiphenyldichloroethane, p,p′- dichlorodiphenyltrichloroethane, and mirex).

Briefly, PFASs were quantified using 200 μL of plasma, and polybrominated biphenyl, PBDEs, PCBs, and OCPs were quantified using 1 mL of plasma and then shipped in dry ice to the Wadsworth Center, the public health laboratory of New York State.¹⁰ We measured lipids using plasma stored in −80°C freezers and isolated from nonfasting blood samples collected at enrollment.²⁴ Total cholesterol and triglyceride levels were directly measured in plasma (in nanograms per milliliter) using a chemistry analyzer (COBAS 6000; Roche Diagnostics). Total plasma lipids (in nanograms per milliliter) were calculated using the short formula: total lipids = 2.27 × total cholesterol + triglycerides +62.3.^25,26 Plasma cotinine levels (in nanograms per milliliter; to convert from nanograms per milliliter or micrograms per liter to nanomoles per liter, multiply by 5.675) at enrollment were measured using an ultraperformance liquid chromatography coupled with an electrospray triple quadrupole tandem mass spectrometry.¹⁰

Machine-measured concentrations were used to quantify POPs below the limit of quantification without substitution to minimize bias when estimating POP exposures and health end points.²⁷ Chemicals were log (1+chemical) transformed for most POPs and log (10+chemical) transformed for 11 POPs (ie, hexachlorobenzene; oxychlordane; trans-chlordane; trans-nonachlor; PBDE 47 and 99; and PCB 18/17, 31/28, 90/101/89, 138/158, and 153) to ensure positive minimum values. Log-transformed POPs were rescaled by their SD to generate results in interpretable units.

Statistical Analysis

Multiple imputation of exposure and covariates using fully conditional specification with 100 iterations was performed with the MICE package in R software.²⁸ We imputed POP data for 56 women (2.5%) and, when missing, for total plasma lipid concentrations (62 [2.7%]), infant sex (164 [7.2%]), marital status (3 [0.1%]), and maternal educational level (1 [0.04%]).

First, we estimated the overall association between pollutant mixtures and fetal growth with a bayesian kernel machine regression approach.²⁹ This method allows for highly correlated and nonlinear associations as well as potential interaction (synergy) between chemicals. For the mixture analysis, all OCPs, PFASs, and PBDEs contributed to their chemical class model. The PCBs were analyzed in 3 separate models according to their purported mechanism of action or their biological activity (eg, potentially estrogenic [PCBs 31/28, 44, 49/43, 52/73, 70/76, 90/101/89, 182/187, and 177], dioxin-like [PCBs 66/80, 74/61, 105/127, 118/106, 156, 167, 128, 138/158, and 170], and non–dioxin-like [PCBs 99, 153, 180, 196/203, and 183]).³⁰

Second, we investigated the association of individual POPs with fetal growth using a generalized additive mixed model with a smooth function for gestational age to account for the nonlinearity between fetal growth and gestational age.³¹

All models (bayesian kernel machine regression and generalized additive mixed model) had a random effect corresponding to the mother-child pair³¹ and were adjusted for the following potential confounders: maternal race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and Asian), maternal age (years, continuous), prepregnancy body mass index (calculated as weight in kilograms divided by height in meters squared; continuous), parity (0, 1, 2, or ≥3), highest educational level (<high school, high school diploma or General Education Development certificate or equivalent, some college or associate degree, bachelor’s degree, or master’s or advanced degree), marital status (not married vs living as married), infant sex (male or female), gestational age at the time of ultrasonography (continuous), total plasma lipids (except for PFASs, continuous), and log-transformed plasma cotinine level (log [x +1], continuous). We further assessed potential effect modification by infant sex and by maternal race/ethnicity.

Given the number of comparisons, we implemented the Benjamini and Yekutieli false discovery rate correction, which took into account the dependency of the multiple testing.³² Analyses were done using R software, version 3.4.3 (R Foundation for Statistical Computing). Descriptive statistics, χ², and analysis of variance were considered statistically significant at 2-tailed P < .05. Other significance testing was also based on 2-sided tests (ie, bayesian kernel machine regression and generalized additive mixed model). All P values were further adjusted for multiple comparisons to ascertain significance.

This analysis used the HPC Biowulf Cluster, the high-performance computational resources of the National Institutes of Health. Data analysis was conducted from July 31, 2018, to June 3, 2019.

Results

A total of 2284 low-risk pregnant women were included in the study. Among them, 606 women (26.5%) self-identified as white with a mean (SD) age of 30.3 (4.4) years, 589 (25.8%) as black with a mean (SD) age of 25.5 (5.5) years, 635 (27.8%) as Hispanic with a mean (SD) age of 27.1 (5.5) years, and 454 (19.9%) as Asian with a mean (SD) age of 30.5 (4.5) years. Prepregnancy body mass index, parity, educational level, and marital status also varied by maternal race/ethnicity, as did maternal plasma cotinine and total lipid levels (Table 1). Specifically, the mean (SD) maternal plasma cotinine and total plasma lipid levels by group were as follows: 1.0 (12.6) ng/mL and 611.6 (97.2) ng/mL for white, 3.7 (25.5) ng/mL and 578.5 (99.3) ng/mL for black, 0.3 (4.2) ng/mL and 628.2 (93.5) ng/mL for Hispanic, and 0.04 (0.2) ng/mL and 619.6 (93.7) ng/mL for Asian women.

Table 1. Description of the Study Cohort by Maternal Race/Ethnicity^a.

Variable	White (n = 606)	Black (n = 589)	Hispanic (n = 635)	Asian (n = 454)	P Value^b
At enrollment, mean (SD)
Gestational age, wk	12.5 (1.0)	12.5 (1.1)	12.9 (0.8)	12.9 (0.8)	<.001
Maternal age, y	30.3 (4.4)	25.5 (5.5)	27.1 (5.5)	30.5 (4.5)	<.001
Prepregnancy BMI, mean (SD)	23.2 (2.8)	24.1 (3.2)	24.4 (2.9)	22.3 (2.7)	<.001
Maternal levels at enrollment, mean (SD), ng/mL
Plasma cotinine	1.0 (12.6)	3.7 (25.5)	0.3 (4.2)	0.04 (0.2)	<.001
Total plasma lipids	611.6 (97.2)	578.5 (99.3)	628.2 (93.5)	619.6 (93.7)	<.001
Gestational age at delivery, mean (SD), wk	39.3 (1.5)	39.0 (2.0)	39.3 (1.7)	39.3 (1.3)	.003
Educational level
<High school	5 (1)	64 (11)	152 (24)	27 (6)	<.001
High school diploma or GED or equivalent	29 (5)	165 (28)	151 (24)	49 (11)
Some college or associate’s degree	112 (18)	225 (38)	234 (37)	93 (20)
Bachelor’s degree	249 (41)	87 (15)	75 (12)	145 (32)
Master’s degree or advanced degree	211 (35)	48 (8)	23 (4)	140 (31)
Parity: live births and stillbirths >20 wk
0	337 (56)	295 (50)	249 (39)	247 (54)	<.001
1	196 (32)	178 (30)	232 (37)	168 (37)
2	59 (10)	88 (15)	95 (15)	32 (7)
≥3	14 (2)	28 (5)	59 (9)	7 (2)
Marital status
Not married	38 (6)	301 (51)	168 (26)	40 (9)	<.001
Married or living as married	568 (94)	288 (49)	467 (74)	414 (91)	<.001
Infant sex
Male	333 (55)	295 (50)	334 (53)	225 (50)	.25
Female	273 (45)	294 (50)	301 (47)	229 (50)	.25

Open in a new tab

Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); GED, General Education Development.

SI conversion factors: To convert plasma cotinine level from ng/mL or μg/L to nmol/L, multiply by 5.675.

^{^a}

Missing data have been imputed. Values given as number (percentage) unless otherwise indicated.

^{^b}

P values derived from analysis of variance or χ² test comparing 4 groups.

Several median plasma POP concentrations differed by maternal race/ethnicity groups (Table 2). These concentrations were generally lower than the median levels in the National Health and Nutrition Examination Survey, a nationally representative sample of pregnant women in the United States.²

Table 2. Distribution of the Persistent Organic Pollutant (POP) Exposures at Enrollment by Maternal Race/Ethnicity^a.

Chemical	NHANES Median Concentration^b	POP Concentration < LOQ, %	Median Concentration (IQR)					P Value
Chemical	NHANES Median Concentration^b	POP Concentration < LOQ, %	Overall (n = 2284)	White (n = 606)	Black (n = 589)	Hispanic (n = 635)	Asian (n = 454)	P Value
OCPs, ng/g
β-HCH	NA	51	0 (0 to 4.51)	0 (0 to 1.84)	0 (0 to 1.36)	0.92 (0 to 4.78)	11.98 (3.15 to 71.57)	<.001
γ-HCH	NA	93	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	.003
HCB	NA	10	7.09 (4.05 to 10.64)	6.62 (4.17 to 9.39)	7.09 (3.50 to 10.68)	6.49 (3.71 to 10.02)	8.78 (5.19 to 15.88)	.46
Oxychlordane	NR^c	26	2.52 (0.75 to 4.41)	4.01 (2.39 to 5.80)	2.36 (0.47 to 4.04)	1.75 (0 to 3.19)	2.03 (0.19 to 4.09)	.002
Trans-chlordane	NA	71	0 (–0.86 to 1.20)	–0.12 (–0.83 to 0.49)	0.10 (–1.00 to 2.45)	0 (–0.81 to 1.51)	0 (–0.76 to 0.91)	.06
Trans-nonachlor	NR^c	10	4.57 (2.50 to 7.88)	5.87 (3.61 to 9.82)	4.76 (2.82 to 7.72)	3.45 (1.71 to 5.77)	4.49 (2.43 to 8.18)	<.001
p,p′-DDE	99.9	0	83.14 (52.34 to 170.68)	65.43 (47.97 to 93.05)	58.39 (42.26 to 87.91)	125.02 (65.90 to 265.36)	176.73 (91.36 to 470.56)	<.001
o,p′-DDD	NR^c	97	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	<.001
p,p′-DDD		72	0 (0 to 0.50)	0 (0)	0 (0 to 0.42)	0 (0 to 0.54)	0.44 (0 to 1.28)	<.001
p,p′-DDT	NR^c	35	1.27 (0 to 2.71)	0.44 (0 to 1.75)	0.81 (0 to 1.60)	1.33 (0 to 2.80)	4.20 (1.93 to 11.09)	<.001
Mirex	NA	68	0 (0 to 0.79)	0 (0 to 0.72)	0 (0 to 0.75)	0 (0 to 0.34)	0 (0 to 1.61)	<.001
PBBs, ng/g
PBB 153	1.1	99	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	.23
PCBs, ng/g
Di-CB
5/8	NA	84	0 (–0.6 to 0.53)	–0.06 (–0.48 to 0.34)	0 (–0.78 to 0.64)	0.07 (–0.59 to 0.56)	0.04 (–0.65 to 0.62)	.48
Tri-CB
18/17	NA	76	–0.08 (–0.96 to 0.77)	–0.17 (–0.80 to 0.42)	–0.16 (–1.35 to 0.87)	–0.04 (–0.91 to 0.91)	0.06 (–0.92 to 0.95)	.54
22	NA	98	–0.05 (–0.23 to 0.11)	–0.04 (–0.16 to 0.09)	–0.06 (–0.30 to 0.15)	–0.05 (–0.24 to 0.12)	–0.04 (–0.24 to 0.13)	.51
31/28	NA	60	0.44 (–0.80 to 1.55)	0.46 (–0.36 to 1.32)	0.33 (–1.18 to 1.67)	0.35 (–0.91 to 1.47)	0.68 (–0.68 to 1.91)	.17
33/20	NA	94	–0.1 (–0.44 to 0.19)	–0.07 (–0.33 to 0.13)	–0.12 (–0.62 to 0.24)	–0.09 (–0.47 to 0.21)	–0.10 (–0.46 to 0.21)	.44
37	NA	99	–0.04 (–0.16 to 0.08)	–0.04 (–0.14 to 0.06)	–0.05 (–0.20 to 0.10)	–0.03 (–0.15 to 0.10)	–0.03 (–0.17 to 0.08)	.78
Tetra-CB
41/64	NA	91	–0.13 (–0.57 to 0.27)	–0.09 (–0.46 to 0.22)	–0.19 (–0.81 to 0.34)	–0.16 (–0.62 to 0.27)	–0.12 (–0.50 to 0.26)	.18
44	2.0	91	–0.20 (–0.65 to 0.20)	–0.20 (–0.56 to 0.10)	–0.21 (–0.82 to 0.25)	–0.17 (–0.69 to 0.23)	–0.17 (–0.57 to 0.24)	.22
47/48/75	NA	95	0.03 (–0.24 to 0.26)	0.02 (–0.20 to 0.20)	0.04 (–0.31 to 0.27)	0 (–0.28 to 0.24)	0.10 (–0.16 to 0.36)	.68
49/43	1.3	93	–0.12 (–0.47 to 0.15)	–0.12 (–0.39 to 0.09)	–0.17 (–0.61 to 0.20)	–0.11 (–0.48 to 0.18)	–0.11 (–0.45 to 0.17)	.24
52/73	3.0	78	–0.37 (–1.55 to 0.64)	–0.38 (–1.21 to 0.36)	–0.54 (–2.17 to 0.82)	–0.34 (–1.47 to 0.75)	–0.26 (–1.41 to 0.81)	.054
66/80	1.2	84	0.25 (–0.07 to 0.59)	0.29 (0.03 to 0.55)	0.17 (–0.20 to 0.56)	0.17 (–0.13 to 0.50)	0.40 (0.06 to 0.83)	<.001
70/76	NA	88	–0.25 (–0.84 to 0.32)	–0.18 (–0.62 to 0.24)	–0.33 (–1.07 to 0.37)	–0.26 (–0.89 to 0.35)	–0.21 (–0.83 to 0.31)	.12
74/61	2.4	38	1.11 (0.52 to 1.92)	1.59 (1.00 to 2.63)	0.85 (0.38 to 1.46)	0.72 (0.27 to 1.28)	1.48 (0.80 to 2.33)	<.001
Penta-CB
90/101/89	1.8	80	–0.42 (–1.49 to 0.52)	–0.40 (–1.14 to 0.36)	–0.60 (–1.91 to 0.55)	–0.45 (–1.52 to 0.48)	–0.22 (–1.30 to 0.75)	.16
93/95	NA	82	–0.42 (–1.43 to 0.45)	–0.37 (–1.15 to 0.28)	–0.52 (–1.77 to 0.54)	–0.44 (–1.51 to 0.46)	–0.34 (–1.37 to 0.51)	.07
99	2.5	41	1.05 (0.47 to 1.86)	1.22 (0.69 to 2.08)	1.00 (0.36 to 1.71)	0.69 (0.25 to 1.31)	1.42 (0.78 to 2.64)	<.001
85/120	NA	100	0.02 (–0.08 to 0.10)	0.02 (–0.06 to 0.09)	0.01 (–0.10 to 0.10)	0.01 (–0.09 to 0.09)	0.03 (–0.08 to 0.11)	.29
110	1.3	91	–0.23 (–0.75 to 0.22)	–0.21 (–0.59 to 0.15)	–0.34 (–0.98 to 0.3)	–0.23 (–0.75 to 0.24)	–0.20 (–0.74 to 0.25)	.11
118/106	3.6	21	1.93 (1.00 to 3.26)	2.42 (1.47 to 3.81)	1.51 (0.69 to 2.63)	1.43 (0.69 to 2.48)	2.67 (1.55 to 4.60)	<.001
105/127	0.9	65	0.60 (0.29 to 1.05)	0.69 (0.42 to 1.12)	0.49 (0.16 to 0.84)	0.46 (0.21 to 0.81)	0.87 (0.50 to 1.56)	<.001
114/122	NA	99	0.13 (0.06 to 0.21)	0.20 (0.11 to 0.31)	0.11 (0.04 to 0.18)	0.09 (0.03 to 0.14)	0.15 (0.10 to 0.24)	<.001
Hexa-CB
128	<LOD	98	0.11 (0 to 0.21)	0.06 (0 to 0.17)	0.11 (0 to 0.20)	0.10 (0 to 0.18)	0.17 (0.05 to 0.32)	<.001
137	NA	93	0.24 (0.11 to 0.43)	0.35 (0.18 to 0.57)	0.22 (0.09 to 0.35)	0.15 (0.05 to 0.28)	0.34 (0.18 to 0.55)	<.001
138/158	7.3	6	4.90 (2.84 to 8.24)	6.31 (4.06 to 9.55)	4.06 (2.25 to 6.70)	3.56 (2.05 to 5.54)	7.16 (4.29 to 12.22)	<.001
146/161	1.1	63	0.58 (0.27 to 1.15)	0.67 (0.35 to 1.28)	0.47 (0.19 to 0.90)	0.40 (0.17 to 0.74)	1.17 (0.58 to 2.19)	<.001
153	8.8	6	5.77 (3.20 to 10.25)	7.56 (4.73 to 11.69)	4.73 (2.66 to 8.15)	4.08 (2.20 to 6.45)	9.19 (5.14 to 15.76)	<.001
156	1.2	65	0.60 (0.34 to 1.07)	0.92 (0.56 to 1.53)	0.46 (0.28 to 0.83)	0.39 (0.24 to 0.65)	0.78 (0.51 to 1.30)	<.001
157	NR^c	98	0.13 (0.07 to 0.25)	0.20 (0.10 to 0.35)	0.11 (0.05 to 0.20)	0.09 (0.05 to 0.15)	0.19 (0.11 to 0.32)	<.001
167	NR^c	95	0.21 (0.12 to 0.36)	0.26 (0.16 to 0.41)	0.16 (0.09 to 0.28)	0.16 (0.09 to 0.25)	0.33 (0.19 to 0.56)	<.001
Hepta-CB
170	2.5	27	1.37 (0.80 to 2.28)	1.86 (1.20 to 2.80)	1.03 (0.63 to 1.74)	0.99 (0.65 to 1.52)	1.98 (1.30 to 3.31)	<.001
172/192	<LOD	94	0.15 (0.05 to 0.32)	0.19 (0.06 to 0.38)	0.10 (0.02 to 0.21)	0.10 (0.03 to 0.20)	0.28 (0.15 to 0.58)	<.001
177	0.7	90	0.24 (0.11 to 0.44)	0.27 (0.14 to 0.44)	0.18 (0.06 to 0.36)	0.17 (0.07 to 0.31)	0.44 (0.23 to 0.90)	<.001
180	6.8	5	3.35 (1.93 to 5.76)	4.80 (3.06 to 6.93)	2.41 (1.47 to 4.34)	2.39 (1.51 to 3.76)	4.97 (3.24 to 9.25)	<.001
182/187	2.5	35	1.24 (0.57 to 2.36)	1.49 (0.85 to 2.39)	1.01 (0.35 to 1.91)	0.87 (0.34 to 1.53)	2.32 (1.14 to 4.46)	<.001
183	0.9	73	0.51 (0.28 to 0.86)	0.60 (0.37 to 0.92)	0.42 (0.19 to 0.72)	0.39 (0.20 to 0.63)	0.79 (0.45 to 1.35)	<.001
Octa-CB
194	1.2	62	0.63 (0.35 to 1.07)	0.90 (0.57 to 1.32)	0.49 (0.29 to 0.86)	0.44 (0.26 to 0.71)	0.87 (0.52 to 1.53)	<.001
195	NA	97	0.17 (0.09 to 0.29)	0.23 (0.14 to 0.32)	0.14 (0.07 to 0.24)	0.13 (0.07 to 0.21)	0.23 (0.13 to 0.43)	<.001
196/203	1.1	56	0.73 (0.40 to 1.23)	0.95 (0.61 to 1.48)	0.62 (0.34 to 1.06)	0.51 (0.29 to 0.83)	1.01 (0.57 to 1.89)	<.001
199	1.1	62	0.63 (0.32 to 1.16)	0.82 (0.49 to 1.39)	0.52 (0.25 to 0.97)	0.41 (0.22 to 0.69)	0.99 (0.51 to 1.95)	<.001
202	NA	94	0.18 (0.07 to 0.35)	0.24 (0.11 to 0.38)	0.16 (0.04 to 0.31)	0.11 (0.03 to 0.21)	0.31 (0.13 to 0.67)	<.001
Nona-CB
206	0.9	82	0.39 (0.21 to 0.67)	0.48 (0.30 to 0.73)	0.38 (0.2 to 0.65)	0.26 (0.14 to 0.45)	0.51 (0.26 to 0.97)	<.001
208	NA	96	0.15 (0.06 to 0.28)	0.18 (0.10 to 0.30)	0.15 (0.06 to 0.27)	0.09 (0.03 to 0.18)	0.22 (0.09 to 0.46)	<.001
Deca-CB
209	NR^c	91	0.28 (0.17 to 0.47)	0 (0.19 to 0.46)	0.26 (0.16 to 0.43)	0.22 (0.13 to 0.35)	0.42 (0.25 to 0.74)	<.001
PBDEs, ng/g
28	1.3	63	0 (0 to 1.09)	0 (0 to 0.66)	0 (0 to 1.26)	0 (0 to 1.27)	0 (0 to 0.94)	.17
47	23.7	7	8.84 (3.89 to 17.84)	6.84 (3.35 to 13.22)	12.54 (6.81 to 25.60)	9.99 (4.62 to 19.67)	6.65 (2.51 to 13.36)	<.001
85	<LOD	97	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	.65
99	5.1	38	2.17 (0 to 5.44)	1.34 (0 to 3.87)	3.26 (0 to 8.92)	2.43 (0 to 5.88)	1.60 (0 to 4.25)	.42
100	6.6	29	2.17 (0 to 4.45)	1.63 (0 to 3.58)	3.11 (1.22 to 6.55)	2.38 (0.58 to 4.82)	1.33 (0 to 3.17)	<.001
153	7.8	58	0 (0 to 7.12)	0 (0 to 11.73)	2.38 (0 to 12.10)	0 (0 to 4.13)	0 (0 to 3.31)	<.001
154	NR^c	61	0.45 (0 to 2.86)	0 (0 to 3.20)	0.83 (0 to 2.74)	0 (0 to 2.80)	1.16 (0 to 2.70)	.93
183	<LOD	100	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	.23
209	NA	100	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	.46
PFASs, ng/mL
N-MeFOSAA	<LOD	20	0.06 (0.03 to 0.12)	0.06 (0.04 to 0.12)	0.09 (0.05 to 0.17)	0.04 (0.02 to 0.08)	0.04 (0.02 to 0.09)	<.001
PFDA	<LOD	1	0.25 (0.16 to 0.42)	0.29 (0.19 to 0.47)	0.21 (0.13 to 0.38)	0.20 (0.14 to 0.29)	0.36 (0.22 to 0.59)	<.001
PFDoDA	NA	49	0.03 (0.01 to 0.05)	0.03 (0.01 to 0.05)	0.03 (0.01 to 0.06)	0.02 (0.01 to 0.03)	0.05 (0.02 to 0.10)	.047
PFDS	NA	92	0.01 (0 to 0.03)	0.02 (0.01 to 0.03)	0.02 (0.01 to 0.04)	0.01 (0 to 0.02)	0.02 (0.01 to 0.04)	<.001
PFHpA	<LOD	58	0.02 (0.01 to 0.06)	0.03 (0.01 to 0.07)	0.02 (0.01 to 0.05)	0.02 (0.01 to 0.06)	0.03 (0.01 to 0.05)	.004
PFHxS	1.2	0	0.71 (0.44 to 1.23)	1.18 (0.74 to 1.83)	0.73 (0.44 to 1.26)	0.52 (0.34 to 0.82)	0.60 (0.42 to 0.91)	<.001
PFNA	0.7	0	0.77 (0.54 to 1.17)	0.93 (0.64 to 1.33)	0.67 (0.47 to 1.07)	0.67 (0.48 to 0.94)	0.93 (0.63 to 1.30)	<.001
PFOA	2.6	0	2.00 (1.31 to 3.01)	2.93 (1.97 to 4.26)	1.73 (1.10 to 2.57)	1.71 (1.16 to 2.54)	1.91 (1.38 to 2.69)	<.001
PFOS	12.0	0	5.16 (3.39 to 7.98)	6.91 (4.63 to 10.08)	5.50 (3.62 to 8.28)	3.58 (2.47 to 5.09)	5.62 (3.64 to 8.39)	<.001
PFOSA	<LOD	99	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	<.001
PFUnDA	<LOD	2	0.19 (0.09 to 0.35)	0.21 (0.12 to 0.32)	0.16 (0.09 to 0.34)	0.13 (0.08 to 0.20)	0.42 (0.18 to 0.79)	<.001

Open in a new tab

Abbreviations: CB, chlorobiphenyl; HCB, hexachlorobenzene; β-HCH, β-hexachlorocyclohexane; γ-HCH, γ-hexachlorocyclohexane; IQR, interquartile range; LOD, limit of detection; LOQ, limit of quantification; NA, not applicable; N-MeFOSAA, N-methylperfluoro-1-octanesulfonamidoacetic acid; NHANES, National Health and Nutrition Examination Survey; NR, not reported; OCP, organochlorine pesticide; o,p′-DDD, o,p′-dichlorodiphenyldichloroethane; p,p′-DDE, p,p′-dichlorodiphenyldichloroethylene; p,p′-DDD, p,p′-dichlorodiphenyldichloroethane; p,p′-DDT, p,p′-dichlorodiphenyltrichloroethane; PBB, polybrominated biphenyl; PBDE, polybrominated diphenyl ether; PCB, polychlorinated biphenyl; PFAS, perfluoroalkyl and polyfluoroalkyl substances; PFDA, perfluorodecanoic acid; PFDoDA, perfluorododecanoic acid; PFDS, perfluorodecane sulfonate; PFHpA, perfluoroheptanoic acid; PFHxS, perfluorohexanesulfonic acid; PFNA, perfluorononanoic acid; PFOA, perfluorooctanoic acid; PFOS, perfluorooctanesulfonic acid; PFOSA, perfluorooctanesulfonamide; PFUnDA, perfluoroundecanoic acid.

^{^a}

All POP concentrations were based on machine-measured concentrations without substitution of concentrations below the LOQ. Concentrations of OCPs, PBBs, PCBs, and PBDEs were adjusted for total plasma lipids. Missing data were imputed. P values were derived from Kruskal-Wallis nonparametric test comparing the median of the 4 racial/ethnic groups.

^{^b}

Median from the pregnant women in the 2003-2004 NHANES.²

^{^c}

Estimate was not reported in the NHANES study because it was less than the maximum LOD.

Mixture Analysis

In the mixture analysis, higher exposure tended to reflect a pattern of diminished growth. As a chemical class, OCPs were negatively associated with all fetal growth measures with the exception of estimated fetal weight. Organochlorine pesticides and dioxin-like PCBs were negatively associated with head circumference (Figure 1) such that the 75th percentile exposure to OCPs was 4.7 mm (95% CI, −6.7 to −2.8 mm) smaller and to dioxin-like PCBs was 6.4 mm (95% CI, −8.4 to −4.3 mm) smaller compared with the 25th percentile exposures. In addition, OCPs, high level of PBDEs, and dioxin-like PCBs were negatively associated with abdominal circumference, with reductions of 3.5 mm (95% CI, −4.7 to −2.2 mm) occurring with OCPs, 2.4 mm (95% CI, −4.0 to −0.5 mm) with high level of PBDEs, and 2.4 mm (95% CI, –3.9 to −0.8 mm) with dioxin-like PCBs when comparing the 75th and 25th percentile exposures. Femur length was negatively associated with OCPs (−0.6 mm; 95% CI, −1.1 to −0.2 mm) and high level of PBDEs (−0.5 mm; 95% CI, −1.0 to −0.1 mm) when comparing the 75th percentile against the 25th percentile exposure level). Associations of POP mixtures with other fetal growth measures are presented in eFigures 2-6 in the Supplement. The contributions of each individual chemical within the mixture are presented in eTables 3-6 in the Supplement.

Figure 1. — The 5% increase in exposure is shown in quantiles, and the 25th percentile is used as the reference point. The model was adjusted for maternal race/ethnicity, age, prepregnancy body mass index, parity, highest educational level, and marital status as well as infant sex, gestational age at the time of ultrasonography, total plasma lipids (except for perfluoroalkyl and polyfluoroalkyl substances [PFASs]), and log-transformed plasma cotinine concentration, with repeated measurements of fetal growth. Dots indicate the estimates; vertical lines, the 95% CIs; dashed horizontal lines, the null; OCPs, organochlorine pesticides; PBDEs, polybrominated diphenyl ethers; PCBs, polychlorinated biphenyls.

Associations with abdominal circumference appeared stronger among infant boys than infant girls for OCPs (−3.78 mm [95% CI, −5.56 to −2.00 mm] vs −2.77 mm [95% CI, −4.51 to −1.02 mm]) and dioxin-like PCBs (−7.07 mm [95% CI, −12.44 to −1.70 mm] vs −1.18 mm [95% CI, −3.62 to 1.25 mm]). The association for estrogenic PCBs appeared stronger among girls than boys (−1.84 mm [95% CI, −3.14 to −0.53 mm] vs −0.38 mm [95% CI, −1.80 to 1.04]) (eFigure 7 in the Supplement). Patterns of associations with abdominal circumference appeared similar by maternal race/ethnicity (eFigure 8 in the Supplement).

Single Chemical Analysis

In the overall cohort, each SD increase in PBDE 154, PCB 157, and PCB 167 concentrations was negatively associated with head circumference (−0.34 mm [95% CI, −0.51 to −0.17 mm] for PBDE 154, −0.36 mm [95% CI, −0.55 to −0.17 mm] for PCB 157, and −0.41 mm [95% CI, −0.62 to −0.19 mm] for PCB 167) in contrast with the positive association between perfluoroheptanoic acid concentration and head circumference (0.39 mm; 95% CI, 0.21-0.56 mm) (Figure 2; eTable 7 in the Supplement). Results were generally similar for the other fetal head measurements (eTables 8 and 9 in the Supplement). No individual POP was significantly associated with abdominal circumference (Figure 2; eTable 10 in the Supplement). Specific PBDE concentrations were significantly associated with decreased femur length, whereas inconsistent associations were observed between femur length and OCPs (−0.9 mm [95% CI, −0.14 to −0.04 mm] for p,p′-dichlorodiphenyldichloroethylene vs 0.16 mm [95% CI, 0.11 to 0.21 mm] for β-hexachlorocyclohexane), PCBs (from −0.15 mm [95% CI, −0.22 to −0.08 mm] for PCB 208 to 0.10 mm [95% CI, 0.05-0.14 mm] for PCB 31/28), and PFASs (from −0.13 mm [95% CI, −0.18 to −0.09 mm] for perfluorooctane sulfonamide to 0.16 mm [95% CI, 0.11 to 0.20 mm] for perfluoroheptanoic acid) (Figure 2).

Figure 2. — Generalized additive mixed model was used and adjusted for maternal race/ethnicity, age, prepregnancy body mass index, parity, highest educational level, and marital status as well as infant sex, gestational age at the time of ultrasonography, total plasma lipids (except for perfluoroalkyl and polyfluoroalkyl substances [PFASs]), and log-transformed plasma cotinine concentration, with repeated measurements of fetal growth. FDR indicates false discovery rate; OCPs, organochlorine pesticides; PBDEs, polybrominated diphenyl ethers; and PCBs, polychlorinated biphenyls. See Table 2 legend for additional expansion of the abbreviations.

Associations between PCBs and long bones differed by fetal sex and maternal race/ethnicity (eTables 11-17 in the Supplement). Among boys, PCB 66/80 (0.16 mm; 95% CI, 0.09-0.24 mm; P for interaction = .02), 93/95 (0.42 mm; 95% CI, 0.09-0.76 mm; P for interaction = .03), 110 (0.17 mm; 95% CI, 0.06-0.29 mm; P for interaction = .03), and 114/122 (0.10 mm; 95% CI, 0.02-0.17 mm; P for interaction = .04) were associated with longer femur lengths. Among black mothers, PCB 18/17 (1.92 mm; 95% CI, 1.16-2.68 mm; P for interaction <.001), 22 (0.29 mm; 95% CI, 0.16-0.42 mm; P for interaction = .002), 31/28 (0.27 mm; 95% CI, 0.14-0.40 mm; P for interaction = .005), 33/20 (0.30 mm; 95% CI, 0.17-0.44 mm; P for interaction <.001), and 37 (0.32 mm; 95% CI, 0.20-0.44 mm; P for interaction <.001) were positively associated with femur length. Among Asian mothers, PCB 194 (–0.19 mm; 95% CI, –0.30 to –0.08 mm; P for interaction = .003), 195 (–0.15 mm; 95% CI, –0.25 to –0.06 mm; P for interaction = .005), 196/203 (–0.21 mm; 95% CI, –0.30 to –0.11; P for interaction = .002), 199 (–0.2 mm; 95% CI, –0.29 to –0.11 mm; P for interaction <.001), 202 (–0.14 mm; 95% CI, –0.22 to –0.07 mm; P for interaction = .005), 206 (–0.32 mm; 95% CI, –0.43 to –0.22 mm; P for interaction <.001), 208 (–0.31 mm; 95% CI, –0.41 to –0.21 mm; P for interaction <.001), and 209 (–0.26 mm; 95% CI, –0.38 to –0.15; P for interaction <.001) were associated with shorter femur lengths.

Regarding OCPs, oxychlordane (–0.98 mm; 95% CI, –1.60 to –0.36 mm; P for interaction <.001), trans-nonachlor (–0.31 mm; 95% CI, –0.54 to –0.08 mm; P for interaction = .005), and p,p′-dichlorodiphenyldichloroethylene (–0.19 mm; 95% CI, –0.22 to –0.09 mm; P for interaction = .006) were associated with shorter femur lengths among boys. Regarding PFASs, association between femur length and perfluorodecanoic acid was positive among white infants (0.53 mm; 95% CI, 0.26-0.45 mm) and black infants (0.14 mm; 95% CI, 0.05-0.23 mm) but was negative among Asian infants (–0.31 mm; 95% CI, –0.41 to –0.21 mm; P for interaction <.001); results for perfluorooctanesulfonic acid were similar (0.25 mm [95% CI, 0.15-0.34 mm] for white infants, 0.16 mm [95% CI, 0.07-0.26 mm] for black infants, and –0.33 mm [95% CI, –0.43 to –0.23 mm] for Asian infants; P for interaction <.001). Associations of single chemicals with other fetal measurements are presented in eTables 18-21 in the Supplement.

Discussion

This cohort study found an association between reduced, ultrasonographically measured fetal growth and specific chemical mixtures in a racially/ethnically diverse cohort of women with low-risk antenatal profiles who were prospectively followed up throughout their pregnancy. The concentrations of POPs were generally lower in this cohort compared with those in pregnant women participating in the National Health and Nutrition Examination Survey, a finding that may be a function of lower exposures to chemicals over time or the selection of only healthy pregnant women, which is associated with optimal fetal growth.² As a mixture, OCPs were consistently negatively associated with fetal growth measurements with the exception of estimated fetal weight, which is not directly measured but derived from other fetal measurements. Dioxin-like PCBs were negatively associated with head circumference, whereas high levels of the PBDE mixture were negatively associated with bone length.

To our knowledge, this study is the first to comprehensively examine POPs in relation to fetal growth parameters. We used a bayesian kernel machine regression method to study the association of mixtures with potential collinearity, interaction, and nonlinear associations.^33,34 We found a stronger negative association with fetal growth using mixture analyses compared with results of individual chemical analyses, particularly for OCPs. This finding may reflect the differences in risk owing to the complex interactions in each chemical class as demonstrated by the variation in results for individual chemicals.

Most past studies investigated POPs in relation to anthropometric measures at birth and reported inconsistent results.^7,8,9,10 Only 1 study, conducted among 2407 Spanish pregnant women in the INMA (Infancia y Medio Ambiente—Environment and Childhood) cohort, used fetal growth measures (abdominal circumference, biparietal diameter, and femur length) for up to 8 routine ultrasonography appointments to create fetal growth curves.¹¹ The INMA study reported negative associations between maternal serum PCB 138, 153, and 180 and femur length after 20 weeks’ gestation, supporting our findings.

Findings from the present study are inconsistent with those of anthropometric measures at birth in the NICHD Fetal Growth Studies–Singleton cohort,¹⁰ underscoring the differences between fetal growth assessments and birth measures. For example, PFASs that were analyzed as single chemicals were negatively associated with bone lengths at birth,¹⁰ whereas this study found that some PFASs were associated with longer fetal bone length and others were associated with shorter bone length. Moreover, no significant positive associations were observed between POPs and any of the anthropometric measures at birth in our cohort. Similarly, the INMA cohort study found no consistent pattern of results comparing anthropometric measures at birth with fetal growth from ultrasonography measurements.^11,35,36

Persistent organic pollutants are endocrine-disrupting chemicals that have different and potentially multiple mechanistic pathways. Some chemicals affect thyroid hormones that are essential for normal fetal growth,³⁷ and exposure has been associated with altered thyroid hormone levels during pregnancy.^38,39 However, chemicals such as dichlorodiphenyltrichloroethane, polybrominated biphenyl, and some PCBs can have estrogenic properties that are known to promote fetal growth.⁴⁰ Furthermore, dichlorodiphenyltrichloroethane metabolites have different implications for health, p,p′-dichlorodiphenyldichloroethylene (a persistent metabolite) has an androgen insensitivity effect that has been associated with increased risk of being small for gestational age and decreased birth weight,^41,42 and o,p′-dichlorodiphenyltrichloroethane is estrogenic but quickly metabolized.⁴³ Other studies have reported a negative association between OCPs and birth weight,^3,44,45 further supporting the present finding in the overall population that maternal exposure to the OCP mixture was associated with decreased fetal growth.

Racial/ethnic and socioeconomic disparities in POP exposure have been reported,^46,47 but most of the findings in the current study did not vary substantively by maternal race/ethnicity. However, our findings suggest that the fetuses of Asian women were more sensitive to higher-chlorinated PCBs and PFASs compared with fetuses of non-Asian women, resulting in smaller bone length. Most studies on POPs have been conducted among white populations⁴⁸ and achieved largely null findings. Consistent with the current study of individual chemicals, a Chinese study reported a negative association between birth length and perfluorooctanoic acid.⁴⁹

This study revealed different patterns of associations by infant sex. In the mixture analysis, we observed a stronger association of PCBs and PBDEs with abdominal circumference among boys. However, in the single chemical analysis, the association of PBDEs with femur length and estimated fetal weight was greater among girls than boys. In addition, lower-chlorinated PCBs had positive associations with bone length among boys, whereas higher-chlorinated PCBs had negative associations. One study conducted among 448 British girls found a positive (but nonsignificant) association between birth length and PCB 118 but negative (but nonsignificant) associations with PCB 153 and 187,⁹ whereas another study found no differences between boys and girls.⁵⁰

Limitations and Strengths

This study has some limitations. Although we had only 1 exposure assessment early in pregnancy, the persistent nature of the exposure suggested it was a reasonable reflection of prenatal exposure. The concentrations were quite low compared with levels in a nationally representative sample of pregnant women, so the findings may be conservative and may not reflect the risk for pregnant women with occupational or other high POP exposures. In addition, using the machine values for measurements below the limit of quantification may lead to some nondifferential measurement errors that would likely bias the results toward the null.⁵¹ In contrast, substituting concentrations below the limit of quantification can systematically bias toward the null when estimating human health outcomes. We also relied on maternal self-reporting of race/ethnicity and behavioral risks, such as smoking and drug use, which may be subject to error, although the cotinine test results were consistent with self-reported smoking status.

The strengths of this work included the standardized repeated measurements of fetal growth with an established protocol in a racially/ethnically diverse population. In addition, the low-risk antenatal profiles of the women participants minimized other competing risk factors for altered fetal growth.

Conclusions

This cohort study found that POP mixtures were negatively associated with fetal growth measures, although the associations were less consistent for individual chemicals. We believe these findings provide insight into the implications of POPs for fetal growth when exposures are low and suggest that, even if exposures could be successfully minimized, these associations may persist. Overall, bone lengths were more often associated with mixture and individual chemical exposures, whereas PFAS exposure had the most inconsistent associations. The strongest associations were seen with the OCP mixture, which was consistently associated with reductions in most fetal growth measures. Although this study found different patterns of associations by infant sex, it showed that most results did not vary substantively by maternal race/ethnicity.

Supplement.

eFigure 1. Flowchart, NICHD Fetal Growth Studies– Singletons (n = 2284)

eFigure 2. Changes in Head Measurements for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 3. Changes in Arm Measurements for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 4. Changes in Leg Measurements for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 5. Changes in Cerebral Width, Inner Orbit Diameter and Outer Orbit Diameter for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 6. Changes in Estimated Fetal Weight for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 7. Changes in Longitudinal Head Circumference, Abdominal Circumference and Femur Length by Fetal Sex for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eFigure 8. Changes in Longitudinal Head Circumference, Abdominal Circumference and Femur Length by Maternal Race/Ethnicity for Each 5% Increase (Quantile) of POPs Mixture Exposure Using the 25th Percentiles as the Reference Point, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 1. Comparison Between the Original Cohort and the Study Population (Pregnant Women from the NICHD Fetal Growth Studies – Singletons

eTable 2. Comparison of the POPs Exposures (Median (25th, 75th percentile) at Enrollment Between Included and Excluded Pregnant Women from the NICHD Fetal Growth Studies – Singletons.

eTable 3. Organochlorine Pesticides – Estimation of the Contribution of Individual Chemical to the Fetal Growth Changes When the Chemical of Interest Changes from the 25th to the 75th Percentile While All of the Other Chemicals in the Mixture Are Fixed to Their 25th Percentile, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 4. Polybrominated Diphenyl Ethers – Estimation of the Contribution of Individual Chemical to the Fetal Growth Changes When the Chemical of Interest Changes from the 25th to the 75th Percentile While All of the Other Chemicals in the Mixture Are Fixed to Their 25th Percentile, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 5. Poly-and-Perfluorinated Alkyl Substances – Estimation of the Contribution of Individual Chemical to the Fetal Growth Changes When the Chemical of Interest Changes from the 25th to the 75th Percentile While All of the Other Chemicals in the Mixture Are Fixed to Their 25th Percentile, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 6. Polychlorinated Biphenyl Congeners – Estimation of the Contribution of Individual Chemical to the Fetal Growth Changes When the Chemical of Interest Changes from the 25th to the 75th Percentile While All of the Other Chemicals in the Mixture Are Fixed to Their 25th Percentile, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 7. Association Between EDCs and Longitudinal Head Circumference Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 8. Association Between EDCs and Longitudinal Biparietal Diameter Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 9. Association Between EDCs and Longitudinal Occipital-Frontal Diameter Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 10. Association Between EDCs and Longitudinal Abdominal Circumference Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 11. Association Between EDCs and Longitudinal Humerus Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 12. Association Between EDCs and Longitudinal Radial Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 13. Association Between EDCs and Longitudinal Ulnar Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 14. Association Between EDCs and Longitudinal Femur Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 15. Association Between EDCs and Longitudinal Tibia Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 16. Association Between EDCs and Longitudinal Fibula Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 17. Association Between EDCs and Longitudinal Foot Length Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 18. Association Between EDCs and Longitudinal Estimated Fetal Growth Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 19. Association Between EDCs and Longitudinal Cerebral Width Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 20. Association Between EDCs and Longitudinal Inner Orbit Diameter Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

eTable 21. Association Between EDCs and Longitudinal Outer Orbit Diameter Using a Generalized Additive Mixed Model, NICHD Fetal Growth Studies – Singletons (n = 2284)

Click here for additional data file.^{(4.5MB, pdf)}

References

1.Ritter R, Scheringer M, MacLeod M, Moeckel C, Jones KC, Hungerbühler K. Intrinsic human elimination half-lives of polychlorinated biphenyls derived from the temporal evolution of cross-sectional biomonitoring data from the United Kingdom. Environ Health Perspect. 2011;119(2):225-231. doi: 10.1289/ehp.1002211 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Woodruff TJ, Zota AR, Schwartz JM. Environmental chemicals in pregnant women in the United States: NHANES 2003-2004. Environ Health Perspect. 2011;119(6):878-885. doi: 10.1289/ehp.1002727 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Govarts E, Iszatt N, Trnovec T, et al. . Prenatal exposure to endocrine disrupting chemicals and risk of being born small for gestational age: pooled analysis of seven European birth cohorts. Environ Int. 2018;115:267-278. doi: 10.1016/j.envint.2018.03.017 [DOI] [PubMed] [Google Scholar]
4.Jacobson JL, Fein GG, Jacobson SW, Schwartz PM, Dowler JK. The transfer of polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs) across the human placenta and into maternal milk. Am J Public Health. 1984;74(4):378-379. doi: 10.2105/AJPH.74.4.378 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Vizcaino E, Grimalt JO, Fernández-Somoano A, Tardon A. Transport of persistent organic pollutants across the human placenta. Environ Int. 2014;65:107-115. doi: 10.1016/j.envint.2014.01.004 [DOI] [PubMed] [Google Scholar]
6.Manzano-Salgado CB, Casas M, Lopez-Espinosa MJ, et al. . Transfer of perfluoroalkyl substances from mother to fetus in a Spanish birth cohort. Environ Res. 2015;142:471-478. doi: 10.1016/j.envres.2015.07.020 [DOI] [PubMed] [Google Scholar]
7.Bach CC, Bech BH, Brix N, Nohr EA, Bonde JP, Henriksen TB. Perfluoroalkyl and polyfluoroalkyl substances and human fetal growth: a systematic review. Crit Rev Toxicol. 2015;45(1):53-67. doi: 10.3109/10408444.2014.952400 [DOI] [PubMed] [Google Scholar]
8.Manzano-Salgado CB, Casas M, Lopez-Espinosa MJ, et al. . Prenatal exposure to perfluoroalkyl substances and birth outcomes in a Spanish birth cohort. Environ Int. 2017;108:278-284. doi: 10.1016/j.envint.2017.09.006 [DOI] [PubMed] [Google Scholar]
9.Patel JF, Hartman TJ, Sjodin A, Northstone K, Taylor EV. Prenatal exposure to polychlorinated biphenyls and fetal growth in British girls. Environ Int. 2018;116:116-121. doi: 10.1016/j.envint.2018.04.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Buck Louis GM, Zhai S, Smarr MM, et al. . Endocrine disruptors and neonatal anthropometry, NICHD Fetal Growth Studies - Singletons. Environ Int. 2018;119:515-526. doi: 10.1016/j.envint.2018.07.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Lopez-Espinosa MJ, Murcia M, Iñiguez C, et al. . Organochlorine compounds and ultrasound measurements of fetal growth in the INMA Cohort (Spain). Environ Health Perspect. 2016;124(1):157-163. doi: 10.1289/ehp.1408907 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Lopez-Espinosa MJ, Costa O, Vizcaino E, et al. . Prenatal exposure to polybrominated flame retardants and fetal growth in the INMA cohort (Spain). Environ Sci Technol. 2015;49(16):10108-10116. doi: 10.1021/acs.est.5b01793 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Snijder CA, Roeleveld N, Te Velde E, et al. . Occupational exposure to chemicals and fetal growth: the Generation R Study. Hum Reprod. 2012;27(3):910-920. doi: 10.1093/humrep/der437 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Gardosi J. Customized fetal growth standards: rationale and clinical application. Semin Perinatol. 2004;28(1):33-40. doi: 10.1053/j.semperi.2003.12.002 [DOI] [PubMed] [Google Scholar]
15.Bach CC, Bech BH, Nohr EA, et al. . Perfluoroalkyl acids in maternal serum and indices of fetal growth: the Aarhus Birth Cohort. Environ Health Perspect. 2016;124(6):848-854. doi: 10.1289/ehp.1510046 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Casas M, Nieuwenhuijsen M, Martínez D, et al. . Prenatal exposure to PCB-153, p,p′-DDE and birth outcomes in 9000 mother-child pairs: exposure-response relationship and effect modifiers. Environ Int. 2015;74:23-31. doi: 10.1016/j.envint.2014.09.013 [DOI] [PubMed] [Google Scholar]
17.Buck Louis GM, Grewal J, Albert PS, et al. . Racial/ethnic standards for fetal growth: the NICHD Fetal Growth Studies. Am J Obstet Gynecol. 2015;213(4):449.e1-449.e41. doi: 10.1016/j.ajog.2015.08.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Gore AC, Chappell VA, Fenton SE, et al. . Executive summary to EDC-2: the Endocrine Society’s second Scientific Statement on endocrine-disrupting chemicals. Endocr Rev. 2015;36(6):593-602. doi: 10.1210/er.2015-1093 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Grewal J, Grantz KL, Zhang C, et al. . Cohort profile: NICHD Fetal Growth Studies-Singletons and Twins. Int J Epidemiol. 2018;47(1):25-25l. doi: 10.1093/ije/dyx161 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Fuchs KM, D’Alton M. 23: Can sonographer education and image review standardize image acquisition and caliper placement in 2D ultrasounds? experience from the NICHD Fetal Growth Study. Am J Obstet Gynecol. 2012;206(1):S15-S16. doi: 10.1016/j.ajog.2011.10.049 [DOI] [Google Scholar]
21.Hediger ML, Fuchs KM, Grantz KL, et al. . Ultrasound quality assurance for singletons in the National Institute of Child Health and Human Development Fetal Growth Studies. J Ultrasound Med. 2016;35(8):1725-1733. doi: 10.7863/ultra.15.09087 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements—a prospective study. Am J Obstet Gynecol. 1985;151(3):333-337. doi: 10.1016/0002-9378(85)90298-4 [DOI] [PubMed] [Google Scholar]
23.Grantz KL, Hediger ML, Liu D, Buck Louis GM. Fetal growth standards: the NICHD fetal growth study approach in context with INTERGROWTH-21st and the World Health Organization Multicentre Growth Reference Study. Am J Obstet Gynecol. 2018;218(2S):S641-655.e28. doi: 10.1016/j.ajog.2017.11.593 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Bao W, Dar S, Zhu Y, et al. . Plasma concentrations of lipids during pregnancy and the risk of gestational diabetes mellitus: a longitudinal study. J Diabetes. 2018;10(6):487-495. doi: 10.1111/1753-0407.12563 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Bernert JT, Turner WE, Patterson DG Jr, Needham LL. Calculation of serum “total lipid” concentrations for the adjustment of persistent organohalogen toxicant measurements in human samples. Chemosphere. 2007;68(5):824-831. doi: 10.1016/j.chemosphere.2007.02.043 [DOI] [PubMed] [Google Scholar]
26.Phillips DL, Pirkle JL, Burse VW, Bernert JT Jr, Henderson LO, Needham LL. Chlorinated hydrocarbon levels in human serum: effects of fasting and feeding. Arch Environ Contam Toxicol. 1989;18(4):495-500. doi: 10.1007/BF01055015 [DOI] [PubMed] [Google Scholar]
27.Schisterman EF, Vexler A, Whitcomb BW, Liu A. The limitations due to exposure detection limits for regression models. Am J Epidemiol. 2006;163(4):374-383. doi: 10.1093/aje/kwj039 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.van Buuren S, Groothuis-Oudshoorn K. mice: multivariate imputation by chained equations in R. JSS. 2011;45(3):67. doi: 10.18637/jss.v045.i03 [DOI] [Google Scholar]
29.Bobb JF, Valeri L, Claus Henn B, et al. . Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures. Biostatistics. 2015;16(3):493-508. doi: 10.1093/biostatistics/kxu058 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Wolff MS, Camann D, Gammon M, Stellman SD. Proposed PCB congener groupings for epidemiological studies. Environ Health Perspect. 1997;105(1):13-14. doi: 10.1289/ehp.9710513 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Cantonwine DE, Ferguson KK, Mukherjee B, et al. . Utilizing longitudinal measures of fetal growth to create a standard method to assess the impacts of maternal disease and environmental exposure. PLoS One. 2016;11(1):e0146532. doi: 10.1371/journal.pone.0146532 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Benjamini Y, Yekutieli D. The control of the false discovery rate in multiple testing under dependency. Ann Stat. 2001;29(4):1165-1188. https://www.jstor.org/stable/2674075 [Google Scholar]
33.Chiu YH, Bellavia A, James-Todd T, et al. ; EARTH Study Team . Evaluating effects of prenatal exposure to phthalate mixtures on birth weight: a comparison of three statistical approaches. Environ Int. 2018;113:231-239. doi: 10.1016/j.envint.2018.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Lazarevic N, Barnett AG, Sly PD, Knibbs LD. Statistical methodology in studies of prenatal exposure to mixtures of endocrine-disrupting chemicals: a review of existing approaches and new alternatives. Environ Health Perspect. 2019;127(2):26001. doi: 10.1289/EHP2207 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Basterrechea M, Lertxundi A, Iñiguez C, et al. ; INMA Project . Prenatal exposure to hexachlorobenzene (HCB) and reproductive effects in a multicentre birth cohort in Spain. Sci Total Environ. 2014;466-467:770-776. doi: 10.1016/j.scitotenv.2013.07.053 [DOI] [PubMed] [Google Scholar]
36.Lopez-Espinosa MJ, Murcia M, Iñiguez C, et al. . Prenatal exposure to organochlorine compounds and birth size. Pediatrics. 2011;128(1):e127-e134. doi: 10.1542/peds.2010-1951 [DOI] [PubMed] [Google Scholar]
37.Blazer S, Moreh-Waterman Y, Miller-Lotan R, Tamir A, Hochberg Z. Maternal hypothyroidism may affect fetal growth and neonatal thyroid function. Obstet Gynecol. 2003;102(2):232-241. doi: 10.1016/s0029-7844(03)00513-1 [DOI] [PubMed] [Google Scholar]
38.Alvarez-Pedrerol M, Guxens M, Ibarluzea J, et al. . Organochlorine compounds, iodine intake, and thyroid hormone levels during pregnancy. Environ Sci Technol. 2009;43(20):7909-7915. doi: 10.1021/es9007273 [DOI] [PubMed] [Google Scholar]
39.Ballesteros V, Costa O, Iñiguez C, Fletcher T, Ballester F, Lopez-Espinosa MJ. Exposure to perfluoroalkyl substances and thyroid function in pregnant women and children: a systematic review of epidemiologic studies. Environ Int. 2017;99:15-28. doi: 10.1016/j.envint.2016.10.015 [DOI] [PubMed] [Google Scholar]
40.Kaijser M, Granath F, Jacobsen G, Cnattingius S, Ekbom A. Maternal pregnancy estriol levels in relation to anamnestic and fetal anthropometric data. Epidemiology. 2000;11(3):315-319. doi: 10.1097/00001648-200005000-00015 [DOI] [PubMed] [Google Scholar]
41.Kelce WR, Stone CR, Laws SC, Gray LE, Kemppainen JA, Wilson EM. Persistent DDT metabolite p,p′-DDE is a potent androgen receptor antagonist. Nature. 1995;375(6532):581-585. doi: 10.1038/375581a0 [DOI] [PubMed] [Google Scholar]
42.Longnecker MP, Klebanoff MA, Zhou H, Brock JW. Association between maternal serum concentration of the DDT metabolite DDE and preterm and small-for-gestational-age babies at birth. Lancet. 2001;358(9276):110-114. doi: 10.1016/S0140-6736(01)05329-6 [DOI] [PubMed] [Google Scholar]
43.Juberg DR, Loch-Caruso R. Investigation of the role of estrogenic action and prostaglandin E2 in DDT-stimulated rat uterine contractions ex vivo. Toxicology. 1992;74(2-3):161-172. doi: 10.1016/0300-483X(92)90136-3 [DOI] [PubMed] [Google Scholar]
44.Toichuev RM, Zhilova LV, Paizildaev TR, et al. . Organochlorine pesticides in placenta in Kyrgyzstan and the effect on pregnancy, childbirth, and newborn health. Environ Sci Pollut Res Int. 2018;25(32):31885-31894. doi: 10.1007/s11356-017-0962-6 [DOI] [PubMed] [Google Scholar]
45.Woods MM, Lanphear BP, Braun JM, McCandless LC. Gestational exposure to endocrine disrupting chemicals in relation to infant birth weight: a Bayesian analysis of the HOME Study. Environ Health. 2017;16(1):115. doi: 10.1186/s12940-017-0332-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.James-Todd TM, Chiu YH, Zota AR. Racial/ethnic disparities in environmental endocrine disrupting chemicals and women’s reproductive health outcomes: epidemiological examples across the life course. Curr Epidemiol Rep. 2016;3(2):161-180. doi: 10.1007/s40471-016-0073-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Zota AR, Adamkiewicz G, Morello-Frosch RA. Are PBDEs an environmental equity concern? exposure disparities by socioeconomic status. Environ Sci Technol. 2010;44(15):5691-5692. doi: 10.1021/es101723d [DOI] [PubMed] [Google Scholar]
48.Zheng T, Zhang J, Sommer K, et al. . Effects of environmental exposures on fetal and childhood growth trajectories. Ann Glob Health. 2016;82(1):41-99. doi: 10.1016/j.aogh.2016.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Wu K, Xu X, Peng L, Liu J, Guo Y, Huo X. Association between maternal exposure to perfluorooctanoic acid (PFOA) from electronic waste recycling and neonatal health outcomes. Environ Int. 2012;48:1-8. doi: 10.1016/j.envint.2012.06.018 [DOI] [PubMed] [Google Scholar]
50.Robledo CA, Yeung E, Mendola P, et al. . Preconception maternal and paternal exposure to persistent organic pollutants and birth size: the LIFE study. Environ Health Perspect. 2015;123(1):88-94. doi: 10.1289/ehp.1308016 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Richardson DB, Ciampi A. Effects of exposure measurement error when an exposure variable is constrained by a lower limit. Am J Epidemiol. 2003;157(4):355-363. doi: 10.1093/aje/kwf217 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials