Data Sheet 1_Head-to-head comparison of stress hyperglycemia ratio versus triglyceride-glucose index for predicting mortality in heart failure: a retrospective cohort study.zip

BackgroundHeart failure (HF) is a life-threatening clinical syndrome characterized by high incidence and mortality, leading to considerable global health and economic burdens. Stress-induced hyperglycemia ratio (SHR) and triglyceride-glucose (TyG) index, as emerging biomarkers reflecting glucose metabolism, are closely associated with poor prognosis in many diseases. However, it remains unclear which of these two indicators possesses superior association and predictive value for prognosis in critically ill patients with HF.
MethodsA retrospective cohort study was conducted on critically ill HF patients, enrolled from the Medical Information Mart for Intensive Care IV (MIMIC-IV) version 3.1. The primary outcome was 180-day mortality, with 1-year (365-day), 90-day, and 30-day mortality as secondary outcomes. Baseline characteristics were compared between survivors and non-survivors. Cox regression, restricted cubic spline (RCS), Kaplan-Meier (K-M), and subgroup analyses were used to assess the association of SHR and TyG index with mortality. Discriminative performance of SHR versus TyG index was compared using ROC curves.
ResultsA total of 1,063 patients were enrolled. After adjusting for confounders, Cox regression analyses revealed that SHR was significantly associated with an increased risk of 180-day, 365-day, 90-day, and 30-day mortality, with hazard ratios (HRs) of 1.35, 1.26, 1.47, and 1.53, respectively. In contrast, TyG index was only associated with mortality risk at 180, 90, and 30 days (HRs: 1.20, 1.24, and 1.31, respectively), with no significant association observed at 1 year. Moreover, these associations were predominantly linear in nature. However, no statistically significant difference was observed in the predictive performance of SHR and TyG index for mortality at any time points (P>0.05).
ConclusionSHR and TyG index can be used as potential risk assessment tools for short-term (180-, 90-, and 30-day) mortality risk in critically ill HF patients, nevertheless, SHR is a more applicable and robust metabolic biomarker associated with 1-year mortality.