Association Between Triglyceride‐Glucose Index and in‐Hospital Intensive Care Unit Readmission in Ischemic Stroke Patients: A Retrospective Analysis Based on the <scp>MIMIC</scp> ‐ <scp>IV</scp> Database

This retrospective cohort study utilized the Medical Information Mart for Intensive Care database to investigate the association between the triglyceride-glucose (TyG) index and in-hospital intensive care unit (ICU) readmission in patients with ischemic stroke (IS). Among 2929 patients, 24.0% experienced ICU readmission. Logistic regression analysis demonstrated that the continuous TyG index was associated with a 17% higher risk of ICU readmission (odds ratio [OR], 1.17; 95% confidence interval [CI], 1.01-1.36; p = 0.04). Compared with the lowest quartile (Q1), the higher TyG index quartiles showed a progressively increased risk of ICU readmission (Q2: OR, 1.49; 95% CI, 1.13-1.96; Q3: OR, 1.46; 95% CI, 1.10-1.95; Q4: OR, 1.53; 95% CI, 1.14-2.05; P for trend = 0.02). Restricted cubic spline analysis revealed a significant nonlinear relationship (P for nonlinearity = 0.027), with an inflection point at a TyG index value of 9.82 (P for log-likelihood ratio = 0.007). Incorporating the TyG index into conventional models significantly improved the predictive performance for ICU readmission, as evidenced by increases in the area under the curve, net reclassification improvement and integrated discrimination improvement. In conclusion, an elevated TyG index was associated with an increased risk of in-hospital ICU readmission in patients with IS and exhibited a nonlinear dose-response relationship. The TyG index enhanced the predictive accuracy of existing scoring tools, suggesting its potential clinical value for risk stratification and early intervention.