Serum lipids mediate the relationship of multiple polyaromatic hydrocarbons on non-alcoholic fatty liver disease: A population-based study

Abstract

Polyaromatic hydrocarbons (PAHs) are recognized as organic pollutants with liver toxicity. However, the relationship between PAHs and nonalcoholic fatty liver disease (NAFLD) is unclear in humans. The aim of this study was to investigate the levels of PAHs in the US population and their association with the risk of NAFLD. We investigated urinary levels of nine PAHs in 2436 participants from the National Health and Nutrition Examination Survey (NHANES) between 2005 and 2012, including 1-Hydroxynapthalene (1-OHN), 2-Hydroxynapthalene (2-OHN), 3-Hydroxyfluorene (3-OHF), 2-Hydroxyfluorene (2-OHF), 3-Hydroxyphenanthrene (3-OHPhe), 1-Hydroxyphenanthrene (1-OHPhe), 2-Hydroxyphenanthrene (2-OHPhe), 1-Hydroxypyrene (1-OHPyr), 9-Hydroxyfluorene (9-OHF). Logistic regression models were used to estimate the relationship between single PAH and NAFLD. Assessment of the overall effect of multiple PAH mixtures on NAFLD using Bayesian kernel machine regression (BKMR) model. There were 698 participants diagnosed with NAFLD in the study group. After adjusting for related covariates such as sex, age, race, education, marital status, poverty income ratio (PIR), body mass index (BMI), total energy intake, smoking, hypertension, and diabetes, logistic regression analysis showed that compared to the low tertile (T1), the odds ratio of the high tertile (T3) was 1.70 (95%CI: 1.26-2.29, p = 0.001) for total PAHs, 1.50 (95%CI: 1.11-2.03, p = 0.008) for 2-OHN, 1.75 (95%CI: 1.31-2.34, p < 0.001) for 2-OHPhe, 1.59 (95%CI: 1.18-2.14, p = 0.002) for 9-OHF and 0.63 (95%CI: 0.46-0.87, p = 0.004) for 3-OHF. In the BKMR model, we found that the overall effect of the nine PAH mixtures was positively associated with the risk of NAFLD. Mediation analysis showed that HDL and TG mediated the association between PAHs and NAFLD. Our study suggests that multiple PAHs mixtures exposure may induce NAFLD by mediating serum lipids in human metabolism.

Keywords: Bayesian kernel machine regression; NHANES; Non-alcoholic fatty liver disease; Polyaromatic hydrocarbons; Serum lipids.