Self-expanding metal stents or nonstent endoscopic therapy: Which is better for anastomotic leaks after total gastrectomy?

Background: Anastomotic leaks are a life-threatening complication of gastrectomies with high mortality after surgical reintervention. Endoscopic therapy using fibrin glue injection, endoclip, and other devices is an alternative to surgical intervention for anastomotic leaks. Recently, self-expanding metal stents (SEMS) were introduced to treat anastomotic leaks. The purpose of this study was to assess the clinical characteristics and therapeutic outcomes of SEMS and nonstent endoscopic therapy (NSET) for treatment of anastomotic leaks after total gastrectomy with the aim of assisting endoscopists in choosing a treatment method. Methods: Between July 2002 and March 2013, 13 patients treated with SEMS and 14 patients treated with NSET for anastomotic leaks after total gastrectomy were enrolled onto the study. Enrolled patients received 16 SEMS placement sessions and 21 NSET sessions. Results: No significant differences in baseline characteristics or clinical characteristics related to leakage were detected in patients with SEMS compared to NSET. The successful sealing rate at the first attempt by SEMS was significantly better than that of NSET (80.0 vs. 28.6 %, P = 0.036), whereas the successful sealing rate after multiple endoscopic treatments was not statistically different (80.0 vs. 64.3 %, P = 0.653). The main reason for reintervention with SEMS was complications and with NSET was nonseal (P = 0.004). Clinical outcomes including length of hospital stay, endoscopic treatment-related mortality, and all-cause mortality were not significantly different between the 2 groups. Conclusions: In terms of efficacy by single effort, SEMS was superior to other methods for treating anastomotic leaks after total gastrectomy. However, complications with SEMS should be considered when choosing an endoscopic treatment method.