Exposure to perfluoroalkyl substances and thyrotropin levels in an exposed young adult population in the Veneto Region

Abstract

Background: Perfluoroalkyl substances (PFASs) are persistent and widespread environmental pollutants. People living in the North-East of Italy have been exposed from the late 1970s to 2013 to high concentrations of PFASs through drinking water. The role PFASs could play on thyroid function is still controversial and studies that focused on thyrotropin (TSH) showed inconsistent results. The aim of this study was to evaluate the association between serum PFASs and TSH levels and its dose-response shape in a large exposed population. Methods: A cross-sectional study was conducted in 14,993 individuals aged 20-39 living in the contaminated area. In the main analysis, patients with thyroid pathologies and pregnant women were excluded. Serum levels of perfluorooctanoic acid (PFOA), perfluorooctanesulfonic acid (PFOS), perfluorohexanesulfonic acid (PFHxS) and perfluorononanoic acid (PFNA) were measured by HPLC-MS/MS. Generalized Additive Models were used to evaluate the association between TSH levels and serum PFASs, using thin plate spline smooth terms to model the potential non-linear relationship. Models were stratified by gender and adjusted for potential confounders. A secondary analysis was conducted to evaluate the association between PFASs and the prevalence of self-reported thyroid disorders. Results: In both males and females, a statistically significant non-linear association was found with PFOS. The shape of the slopes shows a decrease in TSH concentration at low levels of PFOS but an increase at higher serum PFOS. PFOA in females was instead found to have a modest linear positive association with TSH. No association was found between TSH and PFNA or PFHxs. We also found higher odds of reporting thyroid disease with increased measured PFNA concentrations in women. Conclusions: Serum PFASs concentrations were modestly associated with thyroid function in this highly exposed young adults population suggesting this substance to be a possible endocrine disruptor. There is evidence of a non-linear association with TSH for PFOS.