Early Pregnancy Plasma Per- and Polyfluoroalkyl Substances (PFAS) and Maternal Midlife Adiposity

Abstract

Abstract Context Evidence suggests that exposure to per- and polyfluoroalkyl substances (PFAS) increases the risk of developing cardiometabolic disease risk factors. Limited research has evaluated associations between PFAS, assessed during pregnancy, a sensitive window for maternal endocrine effects, and long-term maternal adiposity. Objective Estimate associations of early pregnancy measures of individual PFAS, and PFAS mixtures, with maternal adiposity in midlife. Methods We studied 547 Project Viva participants with measures of early pregnancy (mean gestation 10.0 weeks; mean age 32.5 years) plasma concentrations of 6 PFAS and midlife adiposity outcomes (mean follow-up 17.7 years; mean age 50.7 years), including weight, waist circumference (WC), trunk fat mass (TFM), and total body fat mass (TBFM). We used linear regression and Bayesian Kernel Machine Regression (BKMR). Results Linear regression estimated higher midlife weight per doubling of perfluorooctane sulfonate (PFOS) (3.8 kg [95% CI: 1.6, 5.9]) and 2-(N-ethyl-perfluorooctane sulfonamido) acetate (2.3 kg [95% CI: 0.9, 3.7]). BKMR analyses of single PFAS plasma concentrations (comparing the 25th percentile concentration to the 75th percentile) showed a positive association between PFOS and midlife adiposity (weight: 7.7 kg [95% CI: 4.0, 11.5]; TFM: 1.2 kg [95% CI: 0.0, 2.3]; TBFM: 3.0 kg [95% CI: 0.8, 5.2]), but inverse associations with perfluorononanoate (weight: −6.0 kg [95% CI: −8.5, −3.5]; WC: −1.8 cm [95% CI: −3.2, −0.3]; TFM: −0.8 kg [95% CI: −1.5, −0.1]; TBFM: −1.4 kg [95% CI: −2.7, −0.3]) and perfluorohexane sulfonate (TFM: −0.8 kg [95% CI: −1.5, −0.1]; TBFM: −1.4 kg [95% CI: −2.6, −0.2]). No associations were observed with the overall PFAS mixture. Conclusion Select PFAS, assessed in pregnancy, may differentially affect maternal midlife adiposity, influencing later-life maternal cardiometabolic health.