828 Comparison of ERCP Versus EUS-Guided Biliary Interventions in Patients With Surgically Altered Anatomy

www.giejournal.org Vol 828 Comparison of ERCP Versus EUS-Guided Biliary Interventions in Patients With Surgically Altered Anatomy Andrew S. Nett, Janak N. Shah, Kenneth Binmoeller, Steve D. Kane, Chris M. Hamerski, Yasser M. Bhat* Interventional Endoscopy Services (IES), California Pacific Medical Center, San Francisco, CA Background: Traditional ERCP can be difficult in patients with surgically altered anatomy. EUS-guided interventions (EUS-I) are an alternative means of facilitating therapy (via EUS-rendezvous ERCP or direct interventions) in patients with a difficult-to-access biliary tree. There are limited data on the use of EUS-I in patients with surgically altered anatomy. Aim: To study the impact of EUS-I on biliary endotherapy success in patients with surgically altered anatomy. Methods: At our center, EUS-I have been utilized to facilitate ductal access and therapy when ERCP is difficult, and are routinely used on a case-by-case basis at the discretion of the endoscopist. We retrospectively identified all patients with surgically altered anatomy undergoing attempted endoscopic biliary interventions over a 3-yr period. Procedural outcomes and therapeutic success (defined as completion of intended therapy) of ERCP versus EUS-I were compared. Type of endoscopes used for standard retrograde attempts were at the endoscopist’s discretion. EUS-I included: (1) rendezvous procedures: wire advanced deep into bilioenteric limb (roux-y anatomy) or into duodenum (Billroth 2) to facilitate retrograde insertion of endoscope and cannulation; or (2) direct anterograde interventions: over-the-wire anterograde dilation, stone extraction (e.g pushing stone across papilla), and/or anterograde stent placement. Results: Therapeutic biliary ERCPs were attempted in 80 patients (28 with roux-y bilioenteric reconstructions, 20 with roux-en-y gastric bypass with native papillae (RYGB) and 32 with prior Billroth 2). Duodenoscopes (nZ30) and gastroscopes (nZ2) were used in Billroth 2, colonoscopes (nZ19) and double-balloon enteroscopes (nZ9) in roux-y reconstructions, and colonoscopes (nZ12) and double-balloon enteroscopes (nZ8) in RYGB. EUS-I were attempted in 41 patients (8 with roux-y bilioenteric reconstructions, 28 with RYGB and 5 with Billroth 2). EUS-I techniques included rendezvous wire placement (nZ13) with subsequent retrograde attempts using duodenoscopes or colonoscopes, and direct anterograde interventions (nZ28). Therapeutic success was significantly higher for EUS-I compared to ERCP, especially for patients with RYGB (Table). Therapeutic success with ERCP (57/80; 71%) included 2 of 4 patients with success after initial failed EUS-I. Therapeutic success with EUS-I (36/41; 88%) included 12 of 14 with success after initial failed ERCP. There was no significant difference in immediate complication rates 1.3% for ERCP (1/80: clinically significant bleeding) and 2.4% for EUS-I (1/41: perforation requiring surgery), pZ1. Conclusion: In patients with post-surgical anatomy requiring biliary endotherapy, EUS-guided interventions were significantly more successful than standard ERCP, especially in Roux-en-Y gastric bypass patients. There was no difference in complication rates.