Associations of Serum Perfluoroalkyl Substance and Vitamin D Biomarker Concentrations in the National Health and Nutrition Examination Survey (NHANES), 2003-2010

Abstract

Perfluoroalkyl substances (PFAS) are persistent endocrine disrupting chemicals found in industrial and commercial products. Previous research has shown that other endocrine disrupting chemicals such as phthalates and bisphenol A may alter circulating levels of vitamin D; however, no research has examined associations between PFAS and vitamin D biomarkers. We conducted a cross-sectional analysis of 7,040 individuals aged 12 years and older participating in the 2003-2010 cycles of the United States National Health and Nutrition Examination Survey (NHANES). Concentrations of four PFAS and total 25-hydroxyvitamin D [25(OH)D] were measured in serum samples. We used multivariable linear regression to estimate covariate-adjusted differences in total 25(OH)D per log 10-unit increase in PFAS concentrations. We assessed potential effect measure modification by gender, age, and race/ethnicity in stratified models. PFAS were detected in over 98% of the samples. After adjusting for confounders, a 10-fold increase in perfluorohexane sulfonic acid was associated with a 2.6 nmol/L increase (95% CI: 1.0, 4.2) in total 25(OH)D. Age, gender, and race/ethnicity did not modify this association. A 10-fold increase in perfluorooctane sulfonic acid was associated with a 2.8 nmol/L decrease (95% CI: 0.6, 4.9) in total 25(OH)D, with associations significantly stronger among whites (β: -5.3; 95% CI: -8.4, -2.3) and individuals older than 60 years of age (β: -5.8; 95% CI: -9.7, -1.8). Other PFAS were not associated with total 25(OH)D. Our results suggest that some perfluoroalkyl substances may be associated with altered vitamin D levels in the United States population, and associations may vary by chemical, age, and race/ethnicity.