Associations between in utero exposure of per- and polyfluoroalkyl substances (PFAS) mixture and anthropometry measures at birth

Abstract

In utero exposure to per- and polyfluoroalkyl substances (PFAS), which are recognized developmental toxicants, potentially leads to decreased anthropometric measures in infants at birth. We analyzed 16 PFAS in 350 cord serum samples from Jinan, China, using ultra-high-performance liquid chromatography integrated with Orbitrap mass spectrometry. Birth length, birth weight, and head circumference were extracted from medical records and converted into z-scores (BL-z, BW-z and HC-z, respectively). Multivariable linear regression (MLR) models were employed to investigate the associations between individual PFAS and these birth anthropometric z-scores. To assess the cumulative effects of PFAS, quantile g-computation (QGC) and Bayesian kernel machine regression (BKMR) models were employed. Additionally, stratified analyses were performed to derive sex-specific estimates of the associations.

MLR analysis revealed significant associations between specific PFAS and reduced birth anthropometric measures varying by infant sex. For example, log2-transformed concentration of cord serum perfluorooctanoic acid (PFOA) was associated with reduced BL-z (β = −0.12 (−0.18, −0.06), p < 0.001) and BW-z (β = −0.20 (−0.31, −0.10), p < 0.001) in all infants. Perfluoroheptanesulfonic acid (PFHpS) was inversely associated with BL-z (β = −0.07 (−0.13, −0.02), p = 0.03) and HC-z (β = −0.06 (−0.11, −0.02), p = 0.01) exclusively in males. BKMR and QGC models suggested general negative dose-response pattern between exposure to PFAS mixtures and BL-z, BW-z, and HC-z in males. Conversely, these associations were not evident in females. The key PFAS identified as contributors to the joint effects, along with the directions of their estimated impacts as determined by the mixture methods, showed marginal consistency with the results obtained from the MLR models. Our study underscored that in utero exposure to certain PFAS was associated with reduced anthropometric measures at birth. Male infants were more susceptible to PFAS exposure, particularly to combined PFAS mixture effects.