Tracking early-life PFAS exposure in children at ages 4, 8, and 14 years: A longitudinal study from the INMA Spanish birth cohort

In recent times, the widespread use of PFAS, combined with their persistence and mobility, has led to global environmental contamination and human bioaccumulation. Mapping PFAS exposure in children at critical developmental stages (4, 8, and 14 years) is essential to assess early exposure and its relationship with living place, biological sex and potential health outcomes. In this work, plasma samples from two regions of Gipuzkoa have been analyzed from 207 children at early childhood, 241 children, and 167 adolescents in the INMA Gipuzkoa Spanish birth cohort (2011-2022) for 42 targeted PFAS. The results confirmed 18 different PFAS, including long and short-chain congeners, precursors, and emerging PFAS. The most detected PFAS were PFOS, PFOA, PFNA, PFDA, and PFHxS, with detection frequencies of 70-97 %. At 4 years, children showed the highest detection frequencies, while short-chain PFAS detection (PFBA, FPrPA, PFBS, PFPeA) increased when they reached puberty indicating a change on the environmental exposure patterns. The samples coming from the Urola region showed higher concentrations, exceeding the EFSA limit for chronic exposure to sum of 4 PFAS (PFNA, PFOA, PFOS, PFHxS) in 26-35 % of 14-year-old males and females. PFOA concentrations, in particular, increased between the 8- and 14-year follow-up in this region indicating an exposure between follow-ups. No significant differences were found between sexes for the total PFAS burden but for individual PFAS with overall higher levels in males at 14 years. Hazard quotient assessments employing points of departure taken from literature indicated no risk or moderate risk for hepatic and developmental effects in the general studied population. Finally, multivariate linear regression models showed that maternal PFAS levels and breastfeeding duration were positively associated with children's PFAS concentrations in early development, while environmental or regional factors became more influential with age. Parental socioeconomic status did not emerge as a significant predictor in the models.