Abstract Endoscopic ultrasound-guided choledochoduodenostomy (EUS-CDS) has emerged as an alternative to endoscopic retrograde cholangiopancreatography (ERCP) for biliary drainage in patients with malignant distal biliary obstruction (MDBO). With cautery-enhanced lumen-apposing metal stents (CE-LAMSs), EUS-CDS has become more feasible, yet concerns remain regarding stent patency and safety. Comparative data between CE-LAMS-based and conventional self-expandable metal stent (SEMS)-based EUS-CDS are lacking. The aim of this analysis was to compare outcomes between the two EUS-CDS strategies using ERCP as a common comparator. A systematic review and network meta-analysis was conducted including randomized controlled trials (RCTs) comparing EUS-CDS using a CE-LAMS or SEMS with ERCP in MDBO. Procedural outcomes were technical success and procedure time; clinical outcomes included clinical success, adverse events (AEs), and cholangitis. Pairwise and network meta-analyses were performed using random-effects models. Ranking probabilities were assessed using the surface under the cumulative ranking curve (SUCRA). 6 RCTs (707 patients) were included in the analysis. EUS-CDS with CE-LAMS demonstrated significantly higher technical success compared with SEMS-based EUS-CDS (risk ratio RR 1.22, 95%CI 1.10 to 1.36) and ERCP (RR 1.18, 95%CI 1.10 to 1.27). It was also associated with shorter procedure time compared with SEMS-based EUS-CDS (standardized mean difference SMD −0.78, 95%CI −1.30 to −0.27) and ERCP (SMD −0.84, 95%CI −1.16 to −0.52). No significant differences in clinical success or AEs were found between techniques. According to SUCRA, CE-LAMS ranked highest for technical success and procedural time, while SEMS-based EUS-CDS showed the lowest risk of overall AEs and cholangitis. EUS-CDS with CE-LAMS offers superior procedural efficiency compared with ERCP and SEMS-based approaches with similar clinical success and safety, although concerns about cholangitis warrant further evaluation. PROSPERO registration number: CRD420251126733.
Spadaccini et al. (Fri,) studied this question.