Biol Trace Elem Res. 2026 Jul 28. doi: 10.1007/s12011-026-05261-9. Online ahead of print.
ABSTRACT
This study aimed to systematically investigate the relationship between mixed heavy metal exposure and serum C-reactive protein (CRP) level using an integrated multi-model statistical strategy. This study included 568 participants from the Manganese-Exposed Workers Healthy Cohort. Serum CRP and 20 blood metal concentrations were measured. Key metals were selected via LASSO regression and overall mixture effects and metal contributions were quantified by Quantile g-computation; and joint effects, nonlinearity, and interactions were evaluated using Bayesian Kernel Machine Regression (BKMR). LASSO regression identified 9 key metals (Calcium, Nickel, Copper, Titanium, Tin, Vanadium, Selenium, Arsenic, Rubidium). GLM revealed inverse linear associations for Ca, Se, Rb, and Ni, and a positive association for Cu with CRP. Quantile g-computation showed the overall mixture was significantly inversely associated with CRP (HR = 0.955, 95% CI: 0.918, 0.996), with calcium contributing the largest negative weight (-0.31). BKMR indicated that the overall mixture effect showed a monotonic decreasing trend across exposure quantiles, with a significant positive association at lower quantiles (0.25-0.5). BKMR also identified a significant positive interaction for tin (posterior mean = 0.051, 95% CI: 0.004, 0.098), with calcium showing the highest posterior inclusion probability (PIP = 0.972). This study suggests that metal mixtures exert exposure-level-dependent effects on CRP with complex interactions. Calcium may act as a central regulator, while tin shows amplified effects at higher co-exposure levels. These findings advance mechanistic understanding of mixture toxicology and inform exposure-level-specific risk assessment.
PMID:42521926 | DOI:10.1007/s12011-026-05261-9
