Introduction:
Previous studies have found that patients with recurrent acute pancreatitis (RAP) may be at reduced risk for a clinically severe course and have reduced mortality. However, there is still a lack of data related to RAP patients admitted to the intensive care unit (ICU).

Material and methods:
Baseline characteristics of patients diagnosed with initial and recurrent acute pancreatitis from the Medical Information Mart for Intensive Care/MIMIC-IV database were extracted. In-hospital mortality and length of hospital/ICU stay were identified as outcomes. Binomial logistic regression analysis was performed to clarify the independent risk factors for in-hospital mortality in both groups, and we determined the best scoring system for prognosis prediction by plotting the receiver operating characteristic (ROC) curves and the decision curve analysis (DCA) curves.

Results:
The in-hospital mortality rate was 13.96% in patients with initial acute pancreatitis (IAP) and 3.57% in patients with RAP, and there was no statistically significant difference between the two groups regarding length of hospital/ICU stay. For IAP, the Charlson Comorbidity Index, the Bedside Index for Severity in Acute Pancreatitis (BISAP) score, and the Systemic Inflammatory Response Syndrome (SIRS) score on the first day of admission were independent risk factors for in-hospital mortality. Age, gender, Charlson Comorbidity Index, BISAP score, SIRS score, and obesity were not independent risk factors for in-hospital mortality in patients with RAP. For patients with IAP, the areas under the ROC curves (AUCs) of the four scoring systems (the BISAP, the Logistic Organ Dysfunction System (LODS), the Oxford Acute Severity of Illness Score (OASIS), and the Simplified Acute Physiology Score II (SAPS II)) were 0.720, 0.847, 0.808, and 0.845, respectively, but the results of the Z test showed no significant difference between LODS and SAPS II; The DCA showed that at the threshold of 0.2–0.6, SAPS II score almost always showed a higher net clinical benefit than the other scoring systems, but when the threshold exceeded 0.6, none of the four scoring systems showed a net clinical benefit. For patients with RAP, the AUCs of the four scoring systems (BISAP, LODS, OASIS, and SAPS II) were 0.944, 0.861, 0.681, and 0.829, respectively, but the AUC value of BISAP was only significantly different from that of LODS; the DCA showed that in the threshold range of 0–0.25, BISAP score almost always showed a higher net clinical benefit than the other scoring systems, but in other threshold ranges, none of the four scoring systems showed a net clinical benefit.

Conclusions:
RAP is less severe and has a lower risk of in-hospital mortality than IAP. The Charlson Comorbidity Index, the BISAP, and the SIRS score on the first day of admission were all independent risk factors for in-hospital mortality in patients with IAP. The SAPS II score was a better scoring system for predicting in-hospital mortality in patients with IAP. In contrast, the BISAP score showed potential for predicting in-hospital mortality in patients with RAP.