Association of coexposure to per- and polyfluoroalkyl substances and heavy metals with thyroid function across varied age pregnant women and the potential role of hemoglobin

Background

Previous studies have examined the relationship between exposure to one chemical mixture type and thyroid function in pregnant women; however, the combined impact of coexposure to two or more chemical mixture types is limited. Additionally, age-specific studies are insufficient.

Objectives

This study aimed to confirm the relationships between coexposure to nine per- and polyfluoroalkyl substances (PFAS) and heavy metals (HMs) mixtures and thyroid function in pregnant women in different age groups, and to explore the potential role of hemoglobin in these association.

Methods

A total of 156 pregnant women from a cross-sectional study at one hospital in Liaoning Province, China, were included in this study. High-performance liquid chromatography-electrospray tandem mass spectrometry and inductively coupled plasma-tandem mass spectrometry were used to measure the levels of PFAS and HMs, respectively, in cord blood. Generalized linear regression (GLR), quantile-based g-computation (qgcomp), weighted quantile sum (WQS) regression, and Bayesian kernel machine regression (BKMR) were performed to assess the individual and combined effects of PFAS and HMs on thyroid hormones levels. Furthermore, mediation analyses were used to estimate whether these combined effects were mediated by hemoglobin levels.

Results

Mixed analysis revealed that coexposure to nine PFAS and HMs was negatively correlated with FT3 in the GLR, qgcomp, WQS and BKMR models for all participants, and similar results were obtained for the nonadvanced age group. Single analysis explored significantly negative associations between exposure to Co, Ba or Pb and FT3 levels in all participants by GLR analysis. In the nonadvanced age groups, further single analysis explored that PFOA was significantly and negatively associated with FT3 by GLR and BKMR analysis, and contributed most substantially (PIP = 0.90). It also had the largest negative weight (0.36) by further qgcomp analysis. Moreover, hemoglobin masked almost 15 % of the correlation between coexposure to PFAS and HMs and FT3 in the nonadvanced age group.

Conclusions

Our study demonstrated a negative association between PFAS and HMs coexposure and FT3 levels, which was even more pronounced and partly masked by hemoglobin in the nonadvanced age group of pregnant women. PFOA was the primary pollutant affecting this association. Focusing on hemoglobin as a potential target for reducing the harmful effects of exposure to PFAS and HMs mixtures on thyroid function in pregnant women might be valuable.