Feasibility of endoscopic suturing to prevent adverse events and hospitalization after endoscopic submucosal dissection

• Published in: Endoscopy International Open Vol. 8; no. 9; pp. E1212 - E1217
• Main Authors: Han, Samuel, Wani, Sachin, Edmundowicz, Steven A., Soetikno, Roy, Hammad, Hazem
• Format: Journal Article
• Language: English
• Published: Stuttgart · New York Georg Thieme Verlag KG 01.09.2020
• Subjects: Original article
• ISSN: 2364-3722; 2196-9736
• DOI: 10.1055/a-1197-6534

Abstract Background and study aim  Endoscopic submucosal dissection (ESD) enables en bloc removal of gastrointestinal epithelial lesions but can leave a large mucosal defect, which can lead to inpatient observation and delayed bleeding or perforation. The aim of this study was to examine the safety and effectiveness of endoscopic suturing in closing ESD defects to prevent adverse events. Patients and methods  In this single-center prospective cohort study, endoscopic suturing was performed to close ESD defects in the stomach or rectum. Suturing was performed in the antegrade position starting from the edge most distal to the endoscope insertion site, moving from right to left, left to right manner before ending at the edge most proximal to the endoscope insertion site. Results  In total, 31 patients (mean age 65.6, 71 % male) received endoscopic suturing after gastric (58.1 %) or rectal (41.9 %) ESD. Mean lesion size was 27.4 ± 16.2 mm and mean suturing time was 13.4 ± 5.9 min. Complete closure was achieved in all patients. Same-day discharge occurred in 58.1 % of patients; the remainder were hospitalized with mean length of stay of 1 ± 0.6 day. There were no instances of delayed bleeding or delayed perforation (0 %, 95 % CI: 0–11.5 %). No recurrences were found on surveillance endoscopy. Conclusions  Based on this small prospective study, endoscopic suturing of post-ESD defects in the stomach and rectum appears to be feasible, safe, and potentially effective in preventing bleeding or perforation. Further larger controlled studies, however, are needed to validate these findings.