LAUSR

BACKGROUND Evidence regarding the utility of endoscopic submucosal dissection (ESD) for neoplasia in patients with inflammatory bowel disease (IBD) is limited. This meta-analysis aims to understand the feasibility, safety, and long-term outcomes of ESD in IBD patients. METHODS Electronic databases were searched for observational and case-controlled studies. Primary endpoints were en bloc resection and margin-negative resection of neoplastic lesions. Secondary endpoints included procedure-related bleeding and perforation, local recurrence, and metachronous neoplasia. RESULTS We analyzed 25 studies with a total of 585 neoplastic lesions in 552 patients. The rates of en bloc resection and margin-negative resection were 0.88 [95% confidence interval (CI) 0.82-0.92] and 0.78 (95% CI 0.72-0.83), respectively. Meta-regression analysis showed longer disease duration was significantly associated with the higher rate of en bloc resection. The rates of procedure-related bleeding and perforation were 0.080 (95% CI 0.057-0.11) and 0.055 (95% CI 0.038-0.081), respectively. The rates of local recurrence and metachronous neoplasia were 0.008 events/person-year (95% CI 0.002-0.013) and 0.031 event/person-year (95% CI 0.016-0.046), respectively. Meta-analysis of case-controlled studies found no significant differences in the endpoints between IBD patients treated by ESD and those treated by endoscopic mucosal resection, or non-IBD patients treated by ESD. CONCLUSIONS ESD is a feasible and safe procedure to remove neoplastic lesions in IBD patients. Given there is a considerable risk of metachronous neoplasia development, postoperative surveillance colonoscopy with an appropriate consultation with surgeons is essential to monitor not only local recurrence but also neoplastic changes in the remaining colon.