Association between high triglyceride-glucose index status and one-year all-cause mortality in patients with thoracic aortic aneurysm: a MIMIC-IV cohort study

Abstract

The triglyceride-glucose (TyG) index is widely used as a surrogate marker of insulin resistance and has been associated with a range of adverse cardiovascular outcomes. However, its role in the longer-term prognosis of patients with thoracic aortic aneurysm (TAA), particularly one-year mortality, remains unclear. This study aimed to evaluate the association between the TyG index and one-year mortality in patients with TAA and to assess its potential prognostic value. This study used data from the MIMIC-IV database and included 554 patients with TAA. Patients were classified into low- and high-TyG groups according to the median TyG index. The primary outcome was one-year all-cause mortality. Multivariable logistic regression was used to assess the association between the TyG index and one-year mortality, and restricted cubic spline (RCS) analysis was performed to examine potential nonlinear relationships. Kaplan–Meier survival curves and Cox proportional hazards models were used for time-to-event analyses. The predictive performance, clinical net benefit, and calibration of models incorporating the TyG index were evaluated using receiver operating characteristic (ROC) curves, decision curve analysis, and calibration curves. In addition, sensitivity and subgroup analyses were conducted to examine the robustness of the findings. In the overall TAA cohort, after full adjustment, the high-TyG group was significantly associated with an increased risk of one-year mortality (OR = 1.829, 95% CI: 1.083–3.091, P = 0.024). Similar findings were observed among ICU patients (OR = 3.273, 95% CI: 1.128–9.498, P = 0.029). RCS analysis suggested a potential nonlinear relationship between the TyG index and one-year mortality. Kaplan–Meier curves showed a lower one-year survival rate in the high-TyG group than in the low-TyG group, although the log-rank test did not reach statistical significance. In Cox proportional hazards models, high TyG status was associated with an increased risk of one-year mortality in both the overall cohort (HR = 1.651, 95% CI: 1.041–2.618, P = 0.033) and ICU patients (HR = 2.922, 95% CI: 1.154–7.397, P = 0.024). ROC analysis showed that adding high TyG status increased the model AUC from 0.686 to 0.702 in the overall cohort and from 0.809 to 0.844 in ICU patients; however, these incremental improvements were not statistically significant. Decision curve analysis and calibration curves indicated that the model incorporating high TyG status provided a slightly greater net benefit and a marginally lower Brier score. Sensitivity analyses showed that the association between high TyG status and increased one-year mortality remained directionally consistent when the TyG index was categorized into tertiles, after excluding patients with ruptured TAA, and after excluding patients who died during hospitalization. In patients with TAA, high TyG status was independently associated with an increased risk of one-year all-cause mortality. This association was consistent across sensitivity analyses and time-to-event analyses. Incorporating high TyG status into clinical models provided modest incremental prognostic information, but the improvement in predictive performance was limited, suggesting that it should not be used as a standalone clinical prediction tool. Further prospective studies are needed to validate these findings and clarify its clinical utility.