Cholesterol–high-density lipoprotein–glucose index versus triglyceride–glucose-derived indices for predicting 10-year cardiovascular mortality in the MASHAD cohort

Cardiovascular diseases (CVDs) remain a major global health challenge, accounting for substantial illness and death worldwide. Growing evidence suggests that insulin resistance (IR) has a substantial role in their development and progression. A recently introduced IR surrogate marker, the Cholesterol-HDL-Glucose (CHG) index, has been suggested as an index for identifying metabolic disturbances, but its value in predicting cardiovascular mortality has not been clearly established. This study set out to examine how well the CHG index predicts cardiovascular mortality when compared with the Triglyceride-Glucose (TyG) index and its variants combined with obesity measures. We analyzed data from middle-aged adults (35–65 years) in the Mashhad Stroke and Heart Atherosclerotic Disorder (MASHAD) cohort study, with data collected from 2011 to 2020. Cardiovascular and all-cause mortalities were tracked over at least 10-year follow-up. The CHG, TyG and its derived substitutes were calculated, and their associations with mortality outcomes were assessed using univariate and multivariate Cox regression models. Additionally, receiver operating characteristic (ROC) analysis, Harrell’s C-index, restricted cubic spline (RCS), net reclassification improvement (NRI) and integrated discrimination improvement (IDI), likelihood-based pseudo-R² and the share of explainable log-likelihood, decision curve analysis (DCA) for clinical utility, Kaplan-Meier survival curves, Fine and Grey subdistribution hazard models, E-value for robustness of associations, and subgroup analysis were performed in statistical analysis. Among 7467 participants who were analyzed, higher CHG values were strongly associated with a greater risk of both cardiovascular and all-cause mortality. Each standard deviation (SD) increase in CHG value, raise the risk of cardiovascular mortality by 35.4%, and 31.9% in competing-risk analysis, which was higher than TyG and its related markers. For all-cause mortality, 1-SD increase in CHG was associated with 21.4% higher risk. In predicting cardiovascular mortality, CHG outperformed TyG and its derivatives on ROC, C-index, IDI, pseudo-R², share of log-likelihood and DCA measures. For all-cause mortality, certain TyG-based indices performed better. RCS modeling indicated a linear relationship between CHG and cardiovascular mortality, and a non-linear relationship with all-cause mortality. Survival analysis, robustness checks, and subgroup analyses supported these findings, with no evidence of effect modification for cardiovascular mortality. Elevated CHG is independently associated with a higher risk of cardiovascular mortality and shows a consistent linear pattern. It provides stronger predictive value for cardiovascular death than TyG and its derivatives, supporting its role as a useful marker for assessing CVD risk.