Perfluoroalkyl/Polyfluoroalkyl Substances, Hemostatic/Inflammatory Biomarkers, and Incident Hypertension: The Study of Women's Health Across the Nation.

BACKGROUND

Perfluoroalkyl/polyfluoroalkyl substances (PFAS) are emerging clinical and public health concerns, but their role in cardiovascular disease risk remains unclear. Inflammation and thrombosis may link PFAS and cardiovascular disease.

OBJECTIVE

We investigated the association between serum PFAS concentrations and inflammatory and hemostatic markers in midlife women, and whether these markers mediated the association between PFAS and incident hypertension.

METHODS

We analyzed 1,387 participants from the Study of Women's Health Across the Nation from 1999 to 2000 to 2015 to 2016. Linear mixed-effects models estimated the association of 7 PFAS with high-sensitivity C-reactive protein (hs-CRP), fibrinogen, factor VII activity, plasminogen activator inhibitor-1 (PAI-1), and tissue plasminogen activator antigen. Quantile-g-computation evaluated mixture effects. Causal mediation analyses tested whether these markers mediated the PFAS-hypertension association.

RESULTS

Individual PFAS were positively associated with hs-CRP and PAI-1. For hs-CRP, per doubling of PFAS ranged from 3.59% (95% CI: 0.01%-7.30%) for branched perfluorooctane sulfonic acid isomers to 5.32% (95% CI: 1.67%-9.10%) for 2-(N-methyl-perfluorooctane sulfonamido) acetic acid. For PAI-1, per doubling of PFAS ranged from 2.41% (95% CI: -0.86% to 5.78%) for the linear perfluorooctane sulfonic acid isomer to 3.04% (95% CI: 0.24%-5.93%) for 2-(N-methyl-perfluorooctane sulfonamido) acetic acid. However, none of these associations remained statistically significant after false discovery rate correction. PFAS mixture effects were 6.15% (95% CI: -0.83% to 13.58%) for hs-CRP and 6.05% (95% CI: 0.29%-12.14%) for PAI-1. No associations were found with other markers, and biomarkers did not mediate the PFAS-hypertension association.

CONCLUSIONS

These findings suggest, but do not establish, that PFAS may be linked to inflammation and impaired fibrinolysis in midlife women.