Assessment of stress hyperglycemia ratio to predict mortality in critically ill patients with sepsis: a retrospective cohort study from the MIMIC-IV database

Introduction The stress hyperglycemia ratio (SHR) is a new insulin resistance assessment tool for patients, which has been linked to clinical adverse events. We aimed to explore the SHR–mortality relationship in critically ill patients with sepsis. Methods Patients diagnosed with sepsis, along with blood glucose and hemoglobin A1c levels measured within 24 hours of admission, were retrospectively included in the analysis from the MIMIC-IV database between 2008 to 2019. Patients were stratified into quartile groups (quartile 1 (Q1) to quartile 4 (Q4)) according to SHR level, with 28-day mortality as the primary outcome. The SHR and short term mortality association in patients with sepsis was investigated via Cox regression and Kaplan−Meier analyses. The robustness of the results was verified via multivariate adjustments, multicollinearity, least absolute shrinkage and selection operator (LASSO), and the Boruta algorithm method. The complex relationships among the SHR, short-term mortality were estimated via restricted cubic spline (RCS) analyses. Results 2407 sepsis patients were involved, with a median age of 67 years, and 59.5% were male. Overall, 28-day, 60-day and 90-day mortality were 17.49% (n=421), 21.31% (n=513) and 23.89% (n=575), respectively. After adjusting confounding variables, the SHR was associated with greater short-term mortality, including 28-day (hazard ratio (HR)=1.14, 95% confidence interval (CI)=1.04-1.24, p=0.005; Q4 vs. Q1 (reference group), HR=1.41, 95% CI=1.06-1.87, p=0.017, p_trend=0.005), 60-day (HR=1.12, 95% CI=1.02-1.70, p=0.015; Q4 vs. Q1, HR=1.32, 95% CI=1.02-1.72, p=0.037, p_trend=0.021) and 90-day (HR=1.11, 95% CI=1.02-1.22, p=0.019; Q4 vs. Q1, HR=1.32, 95% CI=1.03–1.68, p=0.027, p_trend=0.017) mortality. Furthermore, the RCS analysis revealed a quasi U-shaped relationship with regards to SHR and short-term mortality in sepsis. The mortality rate increased with a SHR value larger or smaller than 0.9. Conclusions Our research revealed that SHR could serve as a novel indicator for predicting short-term mortality in sepsis patients. SHR demonstrated a quasi U-shaped relationship with short-term mortality in sepsis.