BACKGROUND AND AIMS: Therapeutic endoscopy plays a critical role in the management of upper gastrointestinal (UGI) postsurgical leaks. Data is scarce regarding clinical success and safety. Our aim was to evaluate the effectiveness of endoscopic therapy for UGI postsurgical leaks and associated adverse events (AEs), as well as to identify factors associated with successful endoscopic therapy and AE occurrence.

METHODS: Retrospective, multicenter, international study of all patients who underwent endoscopic therapy for UGI postsurgical leaks between 2014 and 2019.

RESULTS: Two-hundred six patients were included. Index surgery most often performed was sleeve gastrectomy (39.3%), followed by gastrectomy (23.8%) and esophagectomy (22.8%). Median time between index surgery and commencement of endoscopic therapy was 16 days. Endoscopic closure was achieved in 80.1% of the patients after a median follow-up of 52 days (IQR, 33-81.3). Seven hundred seventy-five therapeutic endoscopies were performed. Multimodal therapy was needed in 40.8% of the patients. Cumulative success of leak resolution reached a plateau between third and fourth techniques (approximately 70%-80%); this was achieved after 125 days of endoscopic therapy. Smaller leak initial diameters, hospitalization in general ward, hemodynamic stability, absence of respiratory failure, previous gastrectomy, fewer numbers of therapeutic endoscopies performed, shorter length of stay and shorter times to leak closure were associated with better outcomes. Overall, 102 endoscopic therapy-related AEs occurred in 81 patients (39.3%), the vast majority being managed conservatively or endoscopically. Leak-related mortality rate was 12.4%.

CONCLUSION: Multimodal therapeutic endoscopy, despite being time-consuming and requiring multiple procedures, allows leak closure in a significant proportion of patients with a low rate of severe AEs.