Associations of serum metals and metalloids with noninvasive hepatic steatosis indices: a seasonal repeated-measures metabolome-wide study.

Background

Evidence linking metals and metalloids to hepatic steatosis rests mainly on single measurements. We examined these associations and accompanying metabolic pathways using four seasonal repeated serum measurements.

Methods

This Hefeng cohort included 172 adults with 541 observations after excluding visits with current alcohol consumption or liver stiffness ≥ 9.6 kPa. ICP-MS quantified 18 serum metals and metalloids, and HR LC-MS yielded 1,285 metabolite features. Analyses combined linear and generalized additive mixed models, metabolome-wide association analysis, meet-in-the-middle screening, mediation analysis and weighted quantile sum (WQS) regression using the hepatic steatosis index (HSI) and fatty liver index (FLI).

Results

HSI was positively associated with Se, Mn and Cu, and negatively with Cd. FLI was positively associated with Zn, Cd, Pb and V, and negatively with Mo. Of three metabolites shared by all five FLI-associated elements, only γ-aminobutyric acid (GABA) was credibly annotated. DL-glutamic acid was a Zn/V/Mo-selective hub. In the Mo-FLI model, the indirect-to-total ratio for DL-glutamic acid was 30.6%, whereas Zn and V showed indirect effects opposite to their total associations. WQS showed positive mixture-FLI associations in winter and spring.

Conclusions

Environmental risk assessment for hepatic steatosis should consider seasonal variation and the choice of surrogate index. Manganese and selenium were associated with HSI after season adjustment; FLI associations were sensitive to this adjustment. Mixture-FLI associations were confined to winter and spring. GABA and DL-glutamic acid emerged as candidate metabolic links requiring prospective confirmation because exposures, metabolites and outcomes were measured at the same visits.