Abstract
BACKGROUND: Epidemiological studies linking metabolic syndrome (MetS) and exposure to perfluoroalkyl substances (PFASs) are limited, and the observations gleaned thus far are inconclusive. The study was performed to explore the association of serum PFASs both singly and in a mixed manner with MetS, and meanwhile to examine whether this association was mediated by serum albumin in a US national population. METHODS: Total 8108 participants from the National Health and Nutrition Examination Survey, 2007-2018 were included. Four PFASs - including perfluorohexane sulfonic acid (PFHxS), perfluorononanoic acid (PFNA), perfluoromethylheptane sulfonic acid (PFOS), and perfluorooctanoic acid (PFOA), were selected. Weighted quantile sum regression was used to evaluate mixed PFAS exposure and MetS. Multivariable logistic regression models were used to calculate odd ratio (OR) and 95 % confidence interval (95 % CI). Mediating analyses were used to evaluate the mediating effects of albumin. RESULTS: Comparing the highest with lowest quartile yielded a multivariable-adjusted OR (95 % CI) of 1.40 (1.14-1.72) for PFHxS, 1.36 (1.09-1.70) for PFNA, 1.26 (1.00-1.58) for PFOA, and 1.50 (1.19-1.88) for PFOS when associating MetS. Per unit increment in ln-transformed PFHxS, PFNA, PFOA, and PFOS concentrations was significantly associated with 16 %, 17 %, 13 %, and 15 % increased risk of MetS, respectively. When stratified by sex, the significant association between four PFASs and MetS was only noted in females. Mixed PFAS exposure was inversely associated with MetS, and 8.1 % of this association was mediated by serum albumin (P < 0.001). CONCLUSIONS: Our findings indicate a significant and independent association of serum PFASs with MetS, and importantly this association was dose-dependent, sex-specific, and possibly mediated by serum albumin.