Abstract
BACKGROUND: Patients with sepsis and diabetes exhibit complex pathophysiology that greatly increases the risk of adverse outcomes. This study aimed to investigate the association between CAR and 28-day all-cause mortality among intensive care unit (ICU) patients with sepsis and diabetes mellitus. METHODS: In this retrospective cohort study based on the eICU-CRD, we analyzed 1,800 adult patients with sepsis and diabetes mellitus, who were stratified into quartiles by admission CAR. Survival was assessed by Kaplan-Meier analysis. Covariates were selected via LASSO and stepwise regression. Multivariate Cox models evaluated CAR's independent association with mortality. Restricted cubic splines explored nonlinear relationships. ROC analysis and subgroup analyses were performed. RESULTS: Multivariate Cox proportional hazards analysis indicated that CAR was significantly correlated with the 28-day mortality risk in the ICU. Patients in the highest quartile (Q4) had a 4.68-fold greater risk of death compared with those in the lowest quartile [HR 4.68, 95% CI 2.55-8.61, p < 0.001]. CAR had an AUC of 0.664 for mortality prediction and was associated with a higher mortality risk in the norepinephrine group [HR 8.26, 95% CI 2.55-26.76, p < 0.001]. Higher CAR was also associated with a longer ICU length of stay (β = 1.06, p = 0.006). CONCLUSION: CAR was associated with increased 28-day mortality in ICU patients with sepsis and diabetes. These findings suggest that CAR may be useful for prognosis assessment and risk stratification.