Introduction:
Acute pancreatitis (AP) is a severe inflammatory disease causing abdominal pain and organ failure, potentially leading to necrosis and organ dysfunction. This study aimed to create a nomogram to predict mortality in intensive care unit (ICU) patients with AP.

Material and methods:
We conducted a cohort study using the Medical Information Mart for Intensive Care IV (MIMIC-IV) and data from the First People’s Hospital of Changde City, selecting variables via univariate logistic regression and constructing a nomogram with multivariate logistic regression. The nomogram’s performance was evaluated using the area under the curve (AUC) calculated from the receiver operating characteristic curve (ROC), reclassification improvement (NRI) and integrated discrimination improvement (IDI), while clinical utility was assessed with a calibration curve and decision curve analysis (DCA) validated the predictive model’s efficacy.

Results:
A total of 1100 patients were analyzed, with ICU mortality rates of 7.9% in the training set, 9.2% in the internal validation set, and 13.3% in the external validation set. From the 32 extracted variables, five were ultimately selected: hemoglobin, urea nitrogen, SAPS II score, OASIS score, and myocardial infarction. We subsequently developed and validated a nomogram. The AUC, NRI, and IDI of the nomogram were superior to those of the traditional SAPS II and OASIS scoring systems. Calibration curves and the Hosmer-Lemeshow test revealed satisfactory alignment between predicted and observed outcomes, while DCA substantiated the clinical utility of the model across a wide range of threshold probabilities.

Conclusions:
The nomogram showed good performance in predicting mortality among ICU patients with AP and may serve as a valuable tool for clinicians to identify high-risk patients and optimize medical decision-making.