Association of hepatic biomarkers with incident diabetes: a mediation analysis of the triglyceride-glucose index in a large Chinese cohort.

BACKGROUND

Diabetes disproportionately impacts low- and middle-income populations, exacerbating existing health disparities. The role of hepatic biomarkers, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), and the ALT/AST ratio, in predicting diabetes onset remains insufficiently elucidated. This research assessed how these biomarkers relate to diabetes risk, as well as assessed the mediating effect of the triglyceride-glucose (TyG) index.

METHODS

The secondary analysis utilized data from the Dryad public database, encompassing a cohort of 211,833 Chinese adults aged ≥ 20 years who underwent health examinations between 2010 and 2016. After applying rigorous exclusion criteria, 50,463 participants were included. Cox proportional hazards models were applied to examine how hepatic biomarkers and the TyG index influenced diabetes incidence. The mediation analysis was conducted to assess the TyG index's contribution to the hepatic biomarker-diabetes relationship.

RESULTS

Throughout the observational phase (mean 3.08 years), 1309 participants (2.59%) established diabetes. Increased levels of ALT, AST, and the ALT/AST ratio were all significantly related to a heightened diabetes risk, with the most significant correlation noted for the ALT/AST ratio (adjusted HR per unit increase: 1.04; 95% CI: 1.02-1.05; P < 0.001). Participants in the highest quartile of the ALT/AST ratio had nearly three times the risk of diabetes than the lowest quartile (HR: 2.94; 95% CI: 2.42-3.57; P < 0.001). Joint analysis revealed synergistic effects between elevated hepatic biomarkers and the TyG index, with the combination of high ALT/AST ratio and elevated TyG index yielding the greatest risk (HR: 5.23; 95% CI: 4.42-6.18; P < 0.001). The mediation analysis showed that the TyG index significantly mediated the associations, accounting for 40.25%, 36.45%, and 76.97% of the effects of ALT, AST, and the ALT/AST ratio, respectively, on diabetes risk.

CONCLUSION

Hepatic biomarkers, particularly the ALT/AST ratio, robustly predict diabetes risk in this large cohort, with the TyG index explaining most of this association. These insights reinforce the importance of integrating hepatic and metabolic assessment in preventive strategies to address the growing diabetes epidemic.