Serum Per- and Polyfluoroalkyl Substances Are Associated with Increased Hearing Impairment: A Re-Analysis of the National Health and Nutrition Examination Survey Data

Although studies have shown that per- and polyfluoroalkyl substances (PFAS) are potential environmental ototoxicants, epidemiologic study has been limited. I conducted a cross-sectional study to re-examine the associations between PFAS and hearing impairment. Data were obtained from the National Health and Nutrition Examination Survey (NHANES) 1999–2000, 2003-06, 2009-12, and 2015-16. Perfluorooctanoic acid (PFOA), perfluorooctane sulfonic acid (PFOS), perfluorohexane sulfonic acid (PFHxS), and perfluorononanoic acid (PFNA) were measured in serum samples. Participants were divided into quartiles for each PFAS. Air conduction pure-tone audiometry was administered. Hearing impairment (1: yes, 0: no) was defined as a hearing threshold of more than 25 dB at 500, 1000, 2000, 4000, and 8000 Hz in the worse ear. I assessed the relation of serum PFAS with hearing impairment by the generalized linear mixed model with a logit link and binary distribution. Tests for linear trend across quartiles of serum PFAS were conducted using the median serum PFAS in each quartile as a continuous variable. After adjusting for age, sex, body mass index, education, ethnicity group, and family income, I found positive correlations between PFOA and hearing impairment at 2000 Hz (p-trend < 0.01) and 3000 Hz (p-trend = 0.02); between PFOS and hearing impairment at 500 Hz (p-trend < 0.01), 2000 Hz (p-trend < 0.0001) and 3000 Hz (p-trend = 0.02); between PFNA and hearing impairment at 2000 Hz (p-trend = 0.05), 3000 Hz (p-trend < 0.01), 4000 Hz (p-trend = 0.02), and 8000 Hz (p-trend < 0.01); between PFHxS and hearing impairment at 500 Hz (p-trend = 0.04), 1000 Hz (p-trend = 0.03), and 2000 Hz (p-trend < 0.01). However, some of the findings were not significant when only comparing the highest with the lowest quartile of PFASs. In conclusion, several background serum PFASs are positively correlated with hearing impairment in the United States adult population.