Dietary folate, vitamin D, and inflammation modify associations between prenatal per- and polyfluoroalkyl substance (PFAS) exposures and fetal growth

Abstract

Background: Exposure to per- and polyfluoroalkyl substances (PFAS) is linked to reduced fetal growth, yet little is known about potentially modifiable factors that may mitigate these harmful effects. Aim: In this study, we assessed whether prenatal dietary factors, including folate, vitamin D, and dietary inflammation, may modify the inverse associations between prenatal PFAS exposure and birthweight for gestational age z-scores (BWZ). Methods: Participants were a subset of the prospective Atlanta African American Maternal-Child cohort (N=268, delivery dates between 2014–2018), which includes serum and dietary intake data. Serum collected at 8–14 weeks’ gestation was analyzed for four PFAS (PFHxS, PFOS, PFOA, PFNA) and total and free 25-hydroxy vitamin D (25(OH)D). A semi-quantitative food frequency questionnaire administered during early pregnancy was used to ascertain dietary folate equivalents (DFE) and dietary inflammation index (DII) over the last three months. We used linear regression stratified by dichotomized dietary measures (DFE, total and free 25(OH)D, and DII) to assess effect modification of the association between individual PFAS and BWZ. Mixture effects were estimated using quantile-based g-computation and Bayesian kernel machine regression (BKMR). Summary: A natural log increase in PFOS and PFOA was more strongly associated with lower BWZ only among participants with pro-inflammatory diets (DII ≥75th percentile) or low 25(OH)D (total 25(OH)D <75th percentile). Similar patterns were observed in PFAS mixture analyses (e.g., ΨDII≥75th percentile = −0.47, 95%CI= −0.77, −0.18, and ΨDII<75th percentile = −0.08, 95%CI= −0.24, 0.07). Overall, maternal dietary factors may influence susceptibility to PFAS-related fetal growth effects.