The TyG index-based predictive model achieved a C-index of 0.93, indicating high accuracy for assessing in-hospital mortality risk in older AMI patients with multimorbidity.
Does a predictive model incorporating the triglyceride-glucose (TyG) index accurately predict in-hospital mortality in older acute myocardial infarction patients with multimorbidity?
A predictive model incorporating the TyG index accurately assesses in-hospital mortality risk in older patients with AMI and multimorbidity.
Absolute Event Rate: 0% vs 0%
This study aims to develop a predictive model based on the triglyceride-glucose (TyG) index to assess in-hospital mortality risk in older acute myocardial infarction (AMI) patients with multimorbidity. This retrospective study included 479 patients aged ≥ 65 years with AMI and multimorbidity, hospitalized at Qilu Hospital of Shandong University from September 2017 to March 2022. Patients were randomly divided into a training set (n = 384) and an internal validation set (n = 95). In addition, 90 patients admitted after April 2022 were included as an external validation set. Univariate Cox regression and least absolute shrinkage and selection operator regression were employed to select potential predictive variables, and a predictive model was constructed using Cox regression. Model performance was assessed using receiver operating characteristic (ROC) curves, concordance (C)-index, time-dependent ROC curves, calibration plots, and decision curve analysis (DCA). The predictive performance of the TyG-incorporated model was compared with models excluding TyG, as well as the global registry of acute coronary events and thrombolysis in myocardial infarction risk scores. During an average hospitalization of 9.8 days, 7.85% of patients experienced all-cause mortality. Key predictors included the TyG index, occupation, ≥ 2 Killip classification, hepatic insufficiency, angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker, Nicorandil and white blood cell. The TyG-incorporated model demonstrated superior predictive performance, with C-index values of 0.93 (95% CI 0.89, 0.97; P < 0.001) in the training set, 0.85 (95% CI 0.74, 0.96; P < 0.001) in the internal validation set, and 0.89 (95% CI 0.81, 0.97; P < 0.001) in the external validation set, indicating high predictive accuracy. Calibration plots demonstrated good model calibration, and DCA results indicated superior clinical applicability of the model. The predictive model based on the TyG index provides an effective tool for assessing in-hospital mortality risk in older AMI patients with multimorbidity, demonstrating strong predictive performance and significant clinical value.
Lin et al. (Wed,) reported a other. The TyG index-based predictive model achieved a C-index of 0.93, indicating high accuracy for assessing in-hospital mortality risk in older AMI patients with multimorbidity.