Levels and determinants of exposure to perfluoroalkyl substances in French pregnant women from the SEPAGES cohort.

BACKGROUND

Identifying the primary determinants of exposure to contaminants such as poly- and perfluoroalkyl substances (PFAS) is essential for establishing efficient regulations. We aimed to identify determinants of serum PFAS during pregnancy, a sensitive exposure window.

METHODS

This study relied on 450 pregnant women enrolled in a French cohort from 2014 to 2017. Socio-demographic, diet, water consumption and lifestyle factors were collected through questionnaires. Twenty-six PFAS were measured in maternal serum collected around 19 weeks of gestation (median). Multivariable linear or multinomial regressions were used to study the associations between determinants and individual PFAS, as well as with a three-category variable derived from clustering (categories representing women with the lowest (N = 163), moderate (N = 236) and highest (N = 51) concentration).

RESULTS

Seven PFAS were quantified in 83 % or more of the samples, while the remaining PFAS were quantified in less than 20 %. The highest median concentrations were observed for PFOS (3.95 ng/mL), PFOA (1.02 ng/mL) and PFHxS (0.69 ng/mL). Maternal age was positively associated with all PFAS, and parous women had lower levels of PFOA, PFNA, PFHxS, PFOS and PFHpS than nulliparous. Fish and liver consumption were consistently and positively associated with PFUnDA, PFDA and PFNA. Use of hormonal contraception was found to be associated with a higher PFOA concentration, where a high level of education (5 years after high school) was associated with higher PFUnDA and PFDA levels. Multinomial models on clusters of PFAS exposures showed the same associations for maternal characteristics.

CONCLUSION

Among the factors associated with higher PFAS concentrations in our cohort, those related to the consumption of specific food items may offer actionable levers to reduce exposure. Strengthened regulation should be prioritized over individual choices, as the latter can be strongly influenced by socio-economic status, potentially limiting the feasibility of individual dietary changes.