Metals, Perfluoroalkyl Substance, and Birth Outcomes in the New Hampshire Birth Cohort Study: Beyond Single-Class Mixture Approaches

Background and Aim: Exposure to either metal or perfluoroalkyl substance (PFAS) mixtures during pregnancy has been associated with adverse birth outcomes. However, little is known about the potential joint effects of these two classes of chemicals. We aimed to investigate the joint, index-wise, and individual impacts of metals and PFAS exposures on birth outcomes in a prospective cohort study using both established and novel mixture modeling approaches. Methods: Study participants included 537 mother-child pairs from the New Hampshire Birth Cohort Study (NHBCS). Primary analyses included metals measured in maternal toenails collected six weeks postpartum, reflecting exposures during the prenatal period. PFAS were measured in maternal plasma collected during pregnancy. Birth weight (BW) and head circumference (HC) at birth were abstracted from medical records. Joint and individual associations between in utero metals and PFAS exposures and birth outcomes were evaluated using Bayesian Kernel Machine Regression (BKMR) and employing the recently developed Bayesian Multiple Index Models (BMIM) to additionally assess index-wise associations and interactions among the chemical classes (i.e. toxic metals, essential elements, and PFAS). Results: After controlling for potential confounders, the metals-PFAS mixture was associated with a larger HC at birth. This was driven by Mn (posterior inclusion probability [PIP]: 0.95). Similar associations were identified for the essential element group (including Mn) using BMIM (indexPIP: 0.95; posterior mean for Mn: 0.69; 95% credible interval: 0.00, 0.99). The positive relationship between Mn (essential element group) exposure and HC was stronger at higher levels of Hg (toxic metal group), consistently shown in both BKMR and BMIM. Prenatal co-exposure to metals and PFAS was not associated with BW. Conclusions: Our findings highlight the importance of simultaneously investigating multiple chemical classes in environmental mixture studies. Keywords: Perfluoroalkyl substance (PFAS); Toxic metals; Essential elements; Prenatal exposures; Bayesian kernel machine regression (BKMR); Bayesian multiple index model (BMIM)