Deciphering the impact of heavy metal mixed exposure on lipid metabolism using three statistical models

Abstract

Lipid metabolism disorders pose a significant threat to human health. However, the relationship between heavy metal mixed exposure and lipid metabolism remains poorly understood. This study recruited 1717 residents living near a chromium factory in northeast China. The concentrations of blood Cr, Mn, Cd, Pb, V, and serum CHOL, TG, LDL and HDL levels were measured. Generalized linear model (GLM), quantile g-computation (Qg-comp), and Bayesian kernel machine regression (BKMR) were simultaneously employed to investigate the associations between heavy metal mixed exposure and lipid markers levels. GLM analysis revealed significant associations between blood Cr concentration and HDL (β = -0.07; 95%CI: -0.09, -0.05), LDL (β = -0.06; 95%CI: -0.11, -0.02), and CHOL (β = 0.07; 95%CI: 0.01, 0.12) levels. V concentration was positively associated with HDL (β = 0.12; 95%CI: 0.06, 0.18) and LDL (β = 0.17; 95%CI: 0.04, 0.30) levels. Qg-comp analysis indicated a negative association between heavy metal mixed exposure and HDL (β = -0.040; 95%CI: -0.073, -0.006) level. BKMR model further confirmed the negative relationship between heavy metal mixed exposure and HDL, with the interaction between blood Cr (> 1.05 μg/L) and blood V (> 5.16 μg/L) contributing to decreased HDL levels. Our findings suggested that heavy metal mixed exposure had impacts on HDL and CHOL levels, and the Cr and V may mutually play a predominant role in the observed abnormal HDL levels.

Keywords: Bayesian kernel machine regression (BKMR); Heavy metals; High-density lipoprotein (HDL); Lipid metabolism; Quantile g-computation (Qg-comp).