BACKGROUND
Per- and polyfluoroalkyl substances (PFAS) may impact ovarian folliculogenesis and steroidogenesis, but whether prenatal exposure may impact offspring reproductive health is unknown. This study examines the extent to which maternal PFAS plasma concentrations during pregnancy are associated with polycystic ovary syndrome (PCOS) and related characteristics in female offspring.
METHODS
We studied 322 mother-daughter pairs in Project Viva, a Boston-area longitudinal pre-birth cohort enrolled 1999-2002. We examined associations of maternal prenatal (median: 9.6 weeks gestation) plasma concentrations of six PFAS (log2 transformed) with PCOS and related characteristics among daughters during mid-to-late adolescence. We estimated the associations of single PFAS and PFAS as a mixture with each outcome, using logistic regression and quantile g-computation, respectively, adjusting for parity, and maternal sociodemographic and other lifestyle/health factors.
RESULTS
Among the 322 mother-daughter pairs, the majority of mothers identified as non-Hispanic White and had a college degree, and 13% of daughters had either self-reported PCOS or probable PCOS based on irregular menstrual cycles and clinical or biochemical markers of hyperandrogenism. Among all daughters, there were 27% with irregular menstrual cycles, 34% with hirsutism, and 6% with moderate-to-severe acne. When fully adjusted for confounders, per doubling of maternal 2-(N-ethyl-perfluorooctane sulfonamido) acetate (EtFOSAA) concentration was associated with higher odds of self-reported PCOS [OR (95% CI) = 2.66 (1.18, 5.99)], and per doubling of maternal perfluorononanoate (PFNA) concentration was associated with higher odds of moderate-to-severe acne [OR (95% CI) = 2.33 (1.09, 4.99)] in daughters with or without irregular menstrual cycles. We found no associations of the mixture of six PFAS with PCOS or related traits.
CONCLUSION
Our findings suggest a positive association between maternal concentrations of EtFOSAA and PCOS in their daughters during mid-to-late adolescence, although future studies with larger sample size and extended follow-up across the reproductive life-course are needed.