Although previous studies have examined the hepatotoxicity of single metal exposure, the associations between metal mixture and non-alcoholic fatty liver disease (NAFLD) or fibrosis remain unclear. This study investigated the associations of urinary metal mixture with the risks of NAFLD and liver fibrosis in US adults using data from the National Health and Nutrition Examination Survey (NHANES) from 2017.01 to 2020.03. Vibration-controlled transient elastography was used to detect the controlled attenuation parameter (CAP) and liver stiffness measurement (LSM), which are indicators of NAFLD and liver fibrosis respectively. Three novel mixture modeling approaches including the Bayesian kernel machine regression (BKMR), weighted quantile sum (WQS) regression and quantile g-computation (qgcomp) were used to estimate the associations of the urinary fourteen-metal mixture with Ln CAP and Ln LSM. There were 2283 adults aged over 18 years (1209 women and 1074 men) were included. Among women, urinary metal mixture was positively associated with Ln CAP in the BKMR and qgcomp models (both P < 0.05). However, no significantly associations of urinary metal mixture with Ln CAP were observed among men in all models (all P > 0.05). The metal mixture was not associated with Ln LSM in the three models regardless of genders (all P > 0.05). In conclusion, we observed sex-specific associations between urinary metal mixture and the prevalence of NAFLD in US adults. These findings emphasize the role of environmental heavy metal exposure in the development of NAFLD, and confirm the need for more prospective cohort studies on sex-specific manner.
Keywords: BKMR; Heavy metal mixture; NAFLD; Qgcomp; WQS regression.
Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.