The association between triglyceride-glucose index and all-cause/cardiovascular mortality in patients with different glucose metabolism statuses

The triglyceride-glucose index (TyG) is an emerging marker of metabolic health, yet its association with mortality across different glucose metabolism statuses remains unclear. This study aimed to investigate the relationship between the TyG and the risk of all-cause and cardiovascular mortality among individuals with normoglycemia, dysglycemia, and diabetes mellitus. Participants from nine cycles of the National Health and Nutrition Examination Survey (NHANES) were included and categorized into three groups: normoglycemia, dysglycemia, and diabetes. Cox regression and restricted cubic spline (RCS) analyses were performed to evaluate the linear and nonlinear associations between TyG and mortality. To assess the predictive power of TyG and the atherogenic index of plasma (AIP) for mortality, time-dependent receiver operating characteristic (ROC) curves were constructed. Subgroup analyses were conducted based on age, sex, and blood pressure status. During a median follow-up of 9.2 years, a total of 2199 all-cause deaths and 606 cardiovascular deaths were documented. In the normoglycemic group, a single standard unit increase in TyG was associated with a 35% higher risk of all-cause mortality and a 38% higher risk of cardiovascular mortality (HR: 1.35, 95% CI 1.17–1.56; HR: 1.38, 95% CI 1.04–1.84, respectively). Among participants with diabetes, RCS analysis revealed a U-shaped association between TyG and all-cause/cardiovascular mortality, with an inflection point at 9.1. No significant associations were observed in the dysglycemia group. TyG demonstrated superior predictive performance compared to the AIP for 3-year mortality in both normoglycemic and diabetic individuals. Subgroup analyses identified significant interaction effects of age and sex on the association between TyG and mortality. TyG was associated with an increased risk of all-cause and cardiovascular death in the normoglycemic subgroup, but not in the dysglycemic subgroup. In the diabetes subgroup, the association between the TyG and mortality was nonlinear. The predictive value of TyG across different glucose metabolism statuses provides new evidence for medical practice. A total of 23,103 participants with normoglycemia (n=10,078), dysglycemia (n=8748), and diabetes (n=4277) were included. The median follow-up time was 9.2 years. Weighted COX regression results (upper), restricted cubic spline results (lower left), and time-dependent (cut-off value of 36 months) receiver operating characteristic curve results (lower right).TyG, triglyceride glucose index; AIP, atherogenic index of plasma; RCS, restricted cubic spline; ROC, receiver operating characteristic. *The results of the Cox regression used TyG as a continuous variable.