Endoscopic ultrasound-guided choledochoduodenostomy results in fewer complications than percutaneous drainage following failed ERCP in malignant distal biliary obstruction

Abstract

Background Percutaneous transhepatic biliary drainage (PTBD) and endoscopic ultrasound-guided biliary drainage (EUS-BD), including choledochoduodenostomy (EUS-CDS), are alternative methods for biliary drainage in patients with distal malignant biliary obstruction (MBO) after failed endoscopic retrograde cholangiopancreatography (ERCP). Data on long-term outcomes, adverse events (AEs), and quality of life (QoL) after EUS-CDS and PTBD are limited. Therefore, we created a registry to evaluate the outcomes of both drainage procedures. Methods Patients with distal MBO who underwent EUSCDS or PTBD after unsuccessful ERCP were included in this multicenter investigator-initiated prospective registry over an 18-month inclusion period. Primary end points were procedure-related AEs and mortality within 90 days postprocedure. Secondary end points included technical and clinical success, reinterventions, hospital stay, and QoL. Results 55 patients were included, with 12 patients undergoing PTBD (technical success 100%) and 43 patients EUSCDS (technical success 97.7%). Prior to ERCP, 7/12 patients in the PTBD group and 12/43 patients in the EUS-CDS group opted for best supportive care. The 90-day mortality rate was 66.7% in the PTBD group and 20.9% in the EUS-CDS group (P = 0.005). Furthermore, 11/12 patients (91.7%) in the PTBD group and 19/43 (44.2%) in the EUS-CDS group developed one or more AEs (P = 0.004). The median postprocedural hospital stay was 4 days (interquartile range [IQR] 2-6) in the PTBD group vs. 1 day (IQR 1-2) in the EUS-CDS group (P = 0.001). Conclusion When both modalities were available and technically feasible, gastroenterologists preferred EUS-CDS over PTBD. EUS-CDS seems to be associated with lower mortality and AE rates, shorter hospital admission, and fewer reinterventions, but a randomized controlled trial should confirm these observations.