Polybrominated Diphenyl Ethers in Early Pregnancy and Racial/Ethnic Disparities in Pregnancy Outcomes: Findings from the NICHD Fetal Growth Studies-Singleton Cohort

Abstract

Women in the U.S. are ubiquitously exposed to a class of persistent organic pollutants known as polybrominated diphenyl ethers (PBDEs), which were used as flame retardants in a variety of consumer products. Although previous evidence has shown connections between certain PBDEs and risks of adverse pregnancy outcomes such as preterm birth (PTB) and gestational weight gain (GWG), several key research gaps remain: First, previous research evaluating the associations between PBDEs and PTB had generally focused only on individual chemicals, without accounting for the potential correlations and interactions between chemicals as mixtures. Second, there are pronounced racial and ethnic disparities in PTB in the U.S. population. Exposure levels to PBDEs also differ by race and ethnicity. It remains unclear whether and the extent to which PBDEs may be able to partially explain the racial and ethnic disparities in PTB. Third, very few studies have explored the associations of both individual and mixtures of PBDEs with various GWG measures throughout pregnancy using large population data. Thus, the major aims of this dissertation were to utilize data from a large, multi-site, multi-race and ethnicity cohort study of pregnant women in the U.S. to: (1) Examine the associations of exposure to individual and mixtures of PBDEs during early pregnancy with gestational age at delivery and risk of PTB; (2) Evaluate whether PBDEs may act as mediators for the racial and ethnic disparities in gestational age at delivery and PTB; (3) Assess the associations of exposure to individual and mixtures of PBDEs during early pregnancy with various measures of total and trimester-specific GWG. In Chapter 1, we used data from 2046 women without obesity and 396 women with obesity from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies – Singleton Cohort to explore the individual and joint effects of PBDEs on gestational age at delivery and PTB, using different statistical models accommodating both individual and mixture effects. We found significant associations of higher levels of the PBDE 153 congener (or a principal component summarizing most of PBDE 153 variability) with shorter gestational age at delivery, higher odds of very early/moderate PTB, and higher odds of spontaneous PTB among women without obesity, while no statistically significant association was found among women with obesity. We also reported that the associations among women without obesity were stronger among non-Hispanic black women, women who were overweight, and women who were of older age. Based on our results, we concluded that PBDEs, specifically PBDE 153, were associated with shorter gestation and higher risk of certain PTB subtypes among pregnant women without obesity. In Chapter 2, we applied multiple recently developed causal mediation methodologies incorporating PBDEs as potential mediators for the racial and ethnic disparities in gestational age at delivery and PTB, using data of 2008 women without obesity from the NICHD Fetal Growth Studies – Singleton Cohort. We demonstrated shorter gestation, higher risk of PTB, and higher exposure levels to PBDEs comparing non-Hispanic black to non-Hispanic white women in this study population. We found that a weighted index summarizing PBDEs as a mixture had a suggestive mediator role in the black vs. white disparity of gestational age at delivery which accounted for 11.8% of the total disparity. In addition, we found disparity in gestational age at delivery comparing certain Asian/Pacific Islander subgroups to non-Hispanic white women, although no statistically significant mediation via PBDEs was found. Our results point to the possibility of PBDE mixtures acting as a potential mediator for the racial and ethnic disparities in gestational age at delivery, while future research of larger size might be needed to validate these findings. In Chapter 3, we utilized data from 2052 women without obesity and 397 women with obesity from the NICHD Fetal Growth Studies – Singleton Cohort with weights measured throughout pregnancy. We applied multiple statistical approaches to evaluate the individual and joint associations of PBDEs with various measures of GWG. We found that among women with and without obesity, PBDE 153 was associated with lower total GWG and lower GWG velocity when adjusted for diet and other covariates in a subset of women with available baseline dietary information. Other PBDE congeners, including PBDE 47, 99, 100 and 154, were related to various GWG measures in opposite directions comparing women with and without obesity. Mixtures of all PBDE congeners among women without obesity were associated with lower total GWG, lower total and second trimester GWG velocity, higher odds of inadequate GWG and lower odds of excessive GWG when further adjusted for diet in the subset of women with available baseline dietary information, while mixtures of all PBDE congeners among women with obesity showed inconclusive results. Our results suggest that both individual and mixtures of PBDEs may alter GWG measures during pregnancy, and these effects might differ by pre-pregnancy obesity status. Further research might be needed to elucidate the potential mechanisms.