Factors affecting serum PFAS concentrations among US females with surgically and naturally induced menopause: data from NHANES 2003-2018.

Sex hormones influence excretion of the biopersistent per-and polyfluoroalkyl substances (PFAS) in rodents, but such influences in human studies are less clear. Data from National Health and Nutrition Examination Survey (NHANES) for 2003-2018 for US females aged ≥ 20 years who reported having hysterectomy (HYST, N=1064) and who reported being in natural menopause (MENOP, N=1505) were analyzed for associations of ever use of birth control pills, past pregnancies, live births, and other factors with serum concentrations of six per- and polyfluoroalkyl substances (PFAS). For both HYST and MENOP, PFAS concentrations computed as adjusted geometric means (AGM) were higher among those who took female replacement hormone therapy (HRT) compared to nonusers in multivariable adjusted models, for example PFOS in HRT takers (10.70 ng/mL; 95% C.I. 9.46-12.11) vs. 8.70 ng/mL (95% C.I. 8.07-9.37) in nonusers (p<0.01), and PFOA in HRT users was 2.85 ng/mL (95% C.I. 2.53-3.21) vs. 2.44 ng/mL (95% C.I. 2.32-2.36) in nonusers (p=0.01), with similar findings across race/ethnicity stratifications. HYST participants with retained ovaries sometimes had higher serum PFAS than those without ovaries in post-HYST participants not taking HRT, but results had overlapping confidence intervals in all cases and were inconsistent. PFASs were inversely associated with obesity and directly associated with higher SES as reflected in poverty income ratio (PIR) in most cases, yet HRT results for the entire population are robust to adjustments for obesity and PIR. The results suggest the hypothesis that exogenous hormone use, and specifically estrogen hormones, are associated with higher serum PFAS in postmenopausal women. We discuss potential explanations for the findings, including data from other populations that estrogens may delay the onset of kidney disease, a finding which might paradoxically increase serum PFAS among the HRT population to explain some or all of our findings in a menopausal population.