作者
            
                Taro Shibuki,Norio Fukami,Tsuyoshi Igarashi,Shota Yamaguchi,Kanae Inoue,Tomonao Taira,Tomoyuki Satake,Kazuo Watanabe,Mitsuhito Sasaki,Hiroshi Imaoka,Shuichi Mitsunaga,Naoto Gotohda,Masafumi Ikeda            
         
                    
            摘要
            
            ABSTRACT Background and objective Endoscopic retrograde cholangiopancreatography (ERCP) is the gold standard for managing malignant biliary obstruction (MBO), but it is challenging in patients with surgically altered anatomy (SAA). EUS-guided hepaticogastrostomy (EUS-HGS) and enteroscopy-assisted ERCP are alternative techniques, but their comparative efficacy and safety in this population are not well-researched. Herein, we evaluated the outcomes of EUS-HGS versus single-balloon–assisted enteroscopy ERCP (SBE-ERCP) in MBO patients with SAA. Patients and methods This retrospective study included 70 patients with MBO who underwent EUS-HGS ( n = 29) or SBE-ERCP ( n = 41) at a single center between April 2015 and June 2024. Data on the patient characteristics, procedure details, time to recurrent biliary obstruction (TRBO), and adverse events (AEs) were analyzed. Results The EUS-HGS group showed a higher technical success rate (100% vs. 68%, P = 0.002), shorter procedure time (23 vs. 50 min, P < 0.001), and longer TRBO (251 vs. 103 d, P = 0.020) as compared with the SBE-ERCP group. Subgroup analysis revealed that pancreatobiliary cancer and SBE-ERCP were significantly associated with a shorter TRBO. When stratified by stent type, the median TRBO in the SBE-ERCP with PS group was 49 days, serving as the reference. In comparison, the median TRBO was 140 days in the SBE-ERCP with the self-expandable metallic stent group (hazard ratio [HR], 0.46; P = 0.181), 127 days in the EUS-HGS with PS group (HR, 0.38; P = 0.055), and 701 days in the EUS-HGS with the self-expandable metallic stent group (HR, 0.13; P = 0.014). The overall incidence of AEs was comparable between the 2 groups (20.6% vs. 17.1%), and there were no severe AEs. Regarding reintervention, there was no significant difference in technical success rate and procedure time between EUS-HGS and SBE-ERCP. Conclusion EUS-HGS was associated with a superior technical success rate, shorter procedure time, and longer stent patency, without an increased risk of AEs, as compared with SBE-ERCP. Thus, EUS-HGS should be appropriate for managing MBO patients with SAA.