Age Stratified Analyses of TyG Index eGDR and Additive Effects on T2DM Outcomes in Prevalence and Mortality

Although the triglyceride-glucose (TyG) index and estimated glucose disposal rate (eGDR) have emerged as potential biomarkers, the age-specific predictive value of these markers for diabetes progression and long-term outcomes has not been clearly established. We sought to quantify the associations of TyG and eGDR with T2DM incident and all-cause mortality, and further examine whether these associations vary by age. In this cross-sectional and cohort study, a total of 15,457 eligible participants was included, among whom 2,328 had type 2 diabetes mellitus, participants were stratified into two age groups: Young (< 65 years) and old (≥ 65 years) adults. Logistic regression models to evaluate the associations between TyG index, eGDR index and their additive effect with the risk of type 2 diabetes incidence, Kaplan–Meier survival analysis and Cox proportional hazards regression models were utilized to assess the relationships between TyG/eGDR indices and all-cause mortality. TyG index was significantly positively correlated with the risk of type 2 diabetes across different ages. eGDR showed a significant inverse association with type 2 diabetes risk. High TyG and low eGDR were associated with the highest risk of type 2 diabetes. The restricted cubic spline analysis revealed a U-shaped relationship between TyG index and all-cause mortality, a L-shaped curve relationships between eGDR and all-cause mortality in the total and young type 2 diabetes, no significant association was found in the old type 2 diabetes subgroup. The High TyG and low eGDR demonstrated the highest risk of all-cause mortality in the younger type 2 diabetes, but no additive effect was observed in the older type 2 diabetes. TyG and eGDR positively correlated with the risk of type 2 diabetes, and the combination of TyG and eGDR indices improved the identification of the risk and adverse outcome of diabetes, whereas their association with mortality of type 2 diabetes is not significant in the elderly population even in an additive model.