Associations of prenatal exposure to per- and polyfluoroalkyl substances and fetal sex hormones in the Guangxi Zhuang Birth Cohort Study: Greater effect of long-chain PFAS.

ABSTRACT

Fetal sex hormone homeostasis disruption could lead to reproductive and developmental abnormalities. However, previous studies have reported inconsistent findings regarding the association of maternal per- and polyfluoroalkyl substances (PFAS) exposure with fetal sex hormone levels. A total of 277 mother-infant pairs from the Guangxi Zhuang Birth Cohort Study between 2015 and 2019 were selected. We quantified nine PFAS in maternal serum in early pregnancy, and detected three sex hormones, namely, estradiol (E2), progesterone (P4) and testosterone (TT), in cord blood. The generalized linear model (GLM) and Bayesian kernel machine regression (BKMR) model were used for single- and multiple-exposure analyses, respectively. In the GLM, there was no significant association between an individual PFAS and any hormone level or the E2/TT ratio, but a negative association between perfluorododecanoic acid (PFDoA) exposure and P4 levels in female infants was observed after stratification by sex. In the BKMR, a mixture of nine PFAS was positively associated with E2 levels and the E2/TT ratio, with the same main contributors, i.e., perfluoroundecanoic acid (PFUnA). And PFAS mixtures were not associated with P4 or TT levels. After stratification by infant sex, positive associations of PFAS mixtures with E2 levels and the E2/TT ratio were observed only in male infants, with the same main contributors, i.e., PFUnA. There was a positive association between PFAS mixtures and P4 levels in male infants, in which PFUnA was the main contributor; but a reverse association between PFAS mixtures and P4 levels in female infants, in which PFDoA was the main contributor. This study suggested that prenatal exposure to PFAS mixtures is associated with fetal sex hormones, and long-chain PFAS may play an important role in this association. Furthermore, sex differences in the association of maternal PFAS exposure with E2 and P4 levels need additional attention.