S2293 EUS-Gallbladder Drainage vs Percutaneous Cholecystostomy for Management of Acute Cholecystitis in the Non-Surgical Patient: EUS-GBD Again Shows Superiority

Introduction: Acute cholecystitis (AC) is a common disorder. Cholecystectomy is typically offered, unless patients are too acutely or chronically ill-suited for cholecystectomy. Percutaneous cholecystostomy (PC) has been historically offered in this setting, but EUS-gallbladder drainage (EUS-GBD) is emerging as a preferred technique. Methods: We analyzed prospectively collected data for EUS-GBD patients in our 5-hospital system. We also randomly selected and retrospectively analyzed 128 PC patients out of 1,179 patients who underwent PC in our system from 2015 until 2025. Results: Sixty-four patients were referred for EUS-GBD. Of the 128 PC patients, 66 patients (51.6%) were able to undergo eventual cholecystectomy, leaving 62 PC patients for comparison with EUS-GBD. Comparison of EUS-GBD to PC revealed similar technical success (95.3% vs 100%, P = 0.2). Three patients with EUS-GBD failures were due to intervening cancer, intervening varices, and one patient with anesthesia-induced bradycardia requiring termination of procedure. Of the 61 patients with successful EUS-GBD, 0% had recurrent cholecystitis. There was no difference in clinical success between transantral and transduodenal approaches. Length of stay after EUS-GBD and PC were equivalent (6.9 ± 7.8 days vs 6.9 ± 8.5 days, P = 0.9). Adverse events (AEs), were experienced in 11 patients (18.0%) of EUS-GBD patients with 12 overall adverse events, mostly post-procedural stent-related pain (14.8% of patients). One stent misdeployment occurred (1.6%) addressed with deployment of a second lumen-apposing metal stent (LAMS), and one patient experienced delayed LAMS migration treated with salvage endoscopic cholecystoduodenostomy and closure of the duodenal defect. PC patients, however, had a higher rate of AEs with 67.8% of patients experiencing 71 total adverse events, including 56.5% with drain dislodgment; 40.3% with leak; 30% with bleeding of at least 2 g/dL drop in hemoglobin (9.7% with liver hematoma rate overall); and 8.1% with biloma. Fifty percent had unplanned admissions and emergent unplanned re-interventions related to at least one drain-related adverse event, compared to 0% unplanned readmissions or unplanned interventions in the EUS-GBD cohort. Conclusion: EUS-GBD and PC demonstrate equivalent technical success in AC patients not undergoing cholecystectomy. However, the more favorable AE profile of EUS-GBD coupled with significantly higher unplanned admission and re-interventions in PC patients suggests EUS-GBD is the better option in this patient population.