Clinical Significance of the Triglyceride-Glucose Index in Patients Requiring Nonsurgical Intensive Care

Abstract

The evaluation of triglyceride-glucose (TyG) index has not been sufficient in patients requiring nonsurgical intensive care.A total of 3,906 patients who required intensive care were enrolled. We computed the TyG index using the value on admission by the following formula: ln [triglyceride (mg/dL) × glucose (mg/dL)/2]. Patients were divided into three groups according to the TyG index quartiles: low (quartile 1 [Q1]; TyG index !8.493, n = 977), middle (Q2/Q3; 8.494 !TyG index !9.536, n = 1,953), and high (Q4; TyG index > 9.537, n = 976). The median (interquartile range) TyG index was 9.00 (8.50-9.54); acute coronary syndrome (ACS) had the highest TyG index among all etiologies at 9.12 (8.60-9.68). A multivariate logistic regression model showed that ACS (odds ratio [OR], 2.133; 95% confidence interval [CI], 1.783-2.552) were independently correlated with high TyG index. A Cox proportional hazards regression model revealed that, in ACS, the Q2/Q3 and Q4 groups were independent predictors of 30-day all-cause mortality (hazard ratio [HR], 1.778; 95% CI, 1. HR, 2.986; 95% CI, respectively) and that in acute heart failure [AHF], the Q4 group was a converse independent predictor of 30-day all-cause mortality (HR, 0.488; 95% CI, 0.241-0.988).High TyG index was linked to ACS and negative outcomes in the ACS group; in contrast, low TyG index was associated with adverse outcomes in the AHF group.