作者
Daniel Selin,Viktor Oskarsson,John Maret‐Ouda,R Valente,Rickard Ljung,Bei Yang,Urban Arnelo,Mats Lindblad,Magnus Nilsson,Marcus Holmberg,Omid Sadr-Azodi
摘要
Importance: Same-admission cholecystectomy is recommended by guidelines for mild gallstone-related acute pancreatitis, yet surgery is often deferred. Endoscopic retrograde cholangiopancreatography (ERCP) is often used as an interim strategy, but its effectiveness compared with cholecystectomy is uncertain. Objective: To compare cholecystectomy, ERCP only, and no intervention in association with recurrent acute pancreatitis and other gallstone-related complications, accounting for death as a competing event. Design, Setting, and Participants: This population-based cohort study used nationwide Swedish registries (2006-2019). Adults with first episodes of gallstone-related acute pancreatitis and a length of hospital stay 10 days or less were included. Follow-up began on the day after discharge. Fine-Gray subdistribution hazard models were applied in prespecified time windows (≤7, 8-14, 15-30, 31-90, 91-365, and >365 days), adjusted for age, sex, socioeconomic factors, and comorbidities. These data were analyzed from September 2025 through January 2026. Exposures: Index-admission cholecystectomy, ERCP only, or no intervention; elective postdischarge cholecystectomy was modeled as a time-varying covariate. Main Outcomes and Measures: The primary outcome was recurrent acute pancreatitis. The secondary outcome was other gallstone-related complications (acute cholecystitis and/or choledocholithiasis). Results: Among 9593 patients (median [IQR], age 61 [44-75] years; 60.3% female and 39.7% male), 28.7% underwent cholecystectomy, 16.9% ERCP only, and 54.4% no intervention during index hospital stay. Recurrence of acute pancreatitis in the 3 groups was 3.4%, 4.9%, and 17.5%, respectively. Similar differences were seen in adjusted models (overall subdistribution hazard ratio [sHR] for ERCP only, 1.40; 95% CI, 1.02-1.92 and for no intervention, 6.06; 95% CI, 4.85-7.56 compared with cholecystectomy). The risk of recurrence peaked 8 to 14 days after discharge among patients treated with ERCP only; beyond 15 days, there was no evidence of higher recurrence after ERCP only compared with cholecystectomy. Other gallstone-related complications occurred in 1.6% in the cholecystectomy group, 19.9% in the ERCP only group, and 16.3% in the no intervention group. Conclusions and Relevance: In this study, same-admission cholecystectomy was associated with the lowest recurrence in acute pancreatitis and the lowest rate of other gallstone-related complications. ERCP only was associated with low long-term risk of recurrence but other gallstone-related complications remained common. These findings support prioritizing same-admission cholecystectomy in all individuals with acute gallstone-related pancreatitis, provided they are fit for surgery.