Positive association of exposure to per- and polyfluoroalkyl substances with arterial stiffness parameters and its interaction with sex hormone levels

To examine the associations of per- and polyfluoroalkyl substances (PFAS) with estimated pulse wave velocity (ePWV) and assess their interactions with estradiol and testosterone.

Based on the data of healthy adults with measurements of 5 PFAS from NHANES 2013-2016, we assessed linear and nonlinear associations of exposure to a single PFAS with ePWV using weighted linear regression and restricted cubic spline (RCS) models; weighted quantile sum (WQS) regression was used to estimate the association of PFAS mixture exposure with ePWV with additional age-stratified WQS analyses, and ePWVs between high and low mixture exposure groups were compared using doubly robust estimation. Additive and multiplicative interactions between PFAS and sex hormones were evaluated.

After covariate adjustment, all the 5 PFAS were positively associated with ePWV. RCS analyses indicated that only PFOS had a significant nonlinear association with ePWV. WQS regression showed a positive association between PFAS mixture exposure and ePWV. Perfluorooctane sulfonate (PFOS) and perfluorohexane sulfonate (PFHS) contributed the most to the WQS index with weights of 47.5% and 24.3%, respectively. Age-stratified WQS analyses showed that positive association was mainly observed in middle-aged and older adults, with PFOS and PFHS retaining high weights across different age groups. In doubly robust analysis, ePWV was 1.284 m/s higher in the high-than in low-mixture exposure group (95%CI: 1.078-1.490; P<0.001). Multiplicative interaction analyses suggested weaker positive PFAS-ePWV associations in participants with high estradiol or testosterone levels. Both of the sex hormones showed negative interactions with PFHS; estradiol also interacted negatively with PFOS and PFNA, whereas testosterone interacted negatively with PFOA.

PFAS exposure is associated with an elevated ePWV, with PFOS and PFHS as the key driving components. The positive association between PFAS and ePWV is relatively weaker in individuals with high levels of sex hormones.