Early-pregnancy per- and polyfluoroalkyl substances and maternal post-pregnancy weight trajectory

Objective

The objective of this study was to evaluate associations of early-pregnancy plasma per- and polyfluoroalkyl substances (PFAS) with maternal post-pregnancy weight trajectory parameters.

Methods

We studied 1106 Project Viva participants with measures of early-pregnancy plasma concentrations of eight PFAS. We measured weight at in-person visits at 6 months and 3, 7, and 12 years after pregnancy and collected self-reported weight via annual questionnaires up to 17 years after pregnancy. Weight trajectory parameters were estimated via the Superimposition by Translation and Rotation model. We assessed individual and joint effects of PFAS with trajectory parameters using linear regression and Bayesian kernel machine regression (BKMR).

Results

Perfluorooctane sulfonate (PFOS) concentrations were positively associated with weight trajectory magnitude in both linear regression (0.8 kg [95% CI: 0.1 to 1.4] per doubling of PFOS) and BKMR analyses (2.6 kg [95% CI: 1.4 to 3.8] per increase from 25th to 75th percentile of PFOS concentrations). Conversely, in BKMR analyses, perfluorononanoate was negatively associated with trajectory magnitude (−2.0 kg [95% CI: −2.9 to −1.1]). In stratified linear regression, older-aged participants had more pronounced positive associations of PFOS, perfluorooctanoate, and 2-(N-ethyl-perfluorooctane sulfonamido) acetate with weight trajectory velocity. No associations were observed with the overall PFAS mixture.

Conclusions

Select PFAS, assessed in pregnancy, may affect maternal weight trajectories spanning 17 years after pregnancy, especially for older-aged individuals.