Triglyceride-glucose index is a significant predictor of hospital mortality in non-diabetic critically ill patients with ischemic stroke: a retrospective cohort study of the MIMIC-IV database.

Background and aims: The relationship between the TyG index and outcomes in non-diabetic patients with ischemic stroke admitted to intensive care unit (ICU) has not been validated. This study aims to investigate the correlation between the TyG index and mortality in non-diabetic ICU patients with ischemic stroke.

Methods: We collected data from ICU patients (≥ 18 years) with ischemic stroke and no diabetes. The primary outcome was hospital mortality, and the secondary outcomes were 30-day mortality following admission, hospital length of stay (LOS) and ICU LOS. Cox proportional hazards models and generalized linear models were employed to evaluate association between the TyG index and mortality and LOS, respectively. Nonlinear associations between the TyG index and outcomes were assessed using restricted cubic spline regression models.

Results: A total of 1122 eligible patients were included in this study. The hospital mortality was 10.61%, and 30-day mortality was 16.93%. Multivariate Cox proportional hazards models and generalized linear models revealed the higher of TyG was significantly associated with increased hospital mortality [adjusted HR (95% CI) 1.22 (1.02-1.46), P = 0.0264], 30-day mortality [adjusted HR (95% CI) 1.26 (1.10-1.44), P = 0.0011] and prolonged hospital LOS [adjusted difference (95% CI) 0.52 (0.06-0.97), P = 0.0276].

Conclusions: TyG index is a significant predictor of hospital mortality, 30-day mortality, and LOS in non-diabetic ICU patients with ischemic stroke, which could aid clinical decision-making in managing ischemic stroke.