Mean glucose AUC as a new predictor of in-hospital mortality in ICU patients with acute pancreatitis: A retrospective cohort study

Glycemic control plays a crucial role in the outcomes of acute pancreatitis (AP), yet traditional glucose metrics have shown limited prognostic value. This study investigates the association between mean glucose area under the curve (AUC) and mortality outcomes in AP patients. We conducted a retrospective cohort study of 1085 intensive care unit patients with AP using data from the Medical Information Mart for Intensive Care IV database. Patients were stratified into tertiles based on mean glucose AUC. Multivariable Cox regression and restricted cubic spline analyses were employed to assess the relationship between mean glucose AUC and mortality outcomes. Predictive performance was quantified using receiver operating characteristic curve analysis, and robustness was further tested through comprehensive subgroup and sensitivity analyses. A higher mean glucose AUC was significantly associated with increased mortality risk. In the fully adjusted model, the hazard ratio for in-hospital mortality was 1.10 (95% confidence interval: 1.08–1.12; P < .001). Compared with the lowest tertile, patients in the highest tertile demonstrated an 11.12-fold increased mortality risk (95% confidence interval: 5.78–21.41; P < .001). Mean glucose AUC showed superior discriminative capacity for predicting in-hospital mortality (AUC = 0.783) compared with other glucose metrics. The association remained robust across various subgroups and sensitivity analyses. Mean glucose AUC serves as a powerful, independent predictor of in-hospital mortality in AP patients. This novel metric outperforms conventional glucose parameters in prognostic accuracy, suggesting its potential utility in clinical risk stratification and treatment optimization.