Evaluation of a Predictive Model and Clinical Outcomes in Mild Acute Pancreatitis: Towards Evidence-Based Early...

Given that the care for mild acute pancreatitis (AP) patients is often supportive, these patients may, due to the current strain on hospital care, be eligible for early hospital discharge. However, timely identification of this population and information about their clinical course is needed. This study aims to identify a reliable method for predicting a mild course at 24 and 48 h after hospitalization and to assess clinical outcomes to optimize future early discharge strategies.

This retrospective cohort study included consecutive AP patients admitted to a large teaching hospital in the Netherlands from January 2014 to September 2024, with 90-days follow-up. AP course was predicted using four methods involving C-reactive Protein (CRP) and Systemic Inflammatory Response Syndrome (SIRS). Clinical outcomes, including readmissions, were reported.

A total of 1313 patients were included, of whom 1154 (87.9%) had a confirmed mild course. Of these, 12.9% had AP-related readmissions and 1.8% died within 90 days. For the four prediction methods, positive predictive values ranged from 92.1 to 96.5%, and sensitivity from 51.1 to 79.4%. For predicted mild AP patients, readmission rates ranged from 14.0 to 15.2%, and mortality rates from 1.6 to 2.2%. Recurrent or ongoing AP was the main cause of readmission.

This study found that readmissions are common in mild AP patients, mainly due to recurrent or ongoing AP. CRP and SIRS are suitable in predicting a mild course, potentially within 24 h of admission.