S0811 Endoscopic Submucosal Dissection for Large Superficial Lesions in Ulcerative Colitis: A Matched Cohort Study

INTRODUCTION: Patients with inflammatory bowel disease (IBD) have an increased risk of colorectal neoplasia. Endoscopic submucosal dissection (ESD) to resect large colon neoplasms is used in non-IBD populations as an alternative to colorectal surgery. In this pilot study, we compared ESD outcomes of IBD-associated lesions to sporadic colorectal lesions in patients without IBD. METHODS: We conducted a retrospective, matched case series of patients with active ulcerative colitis (UC) undergoing ESD for resection of large rectal lesions. Matching criteria used were Charlson Comorbidity Index (CCI), lesion location (rectum), and lesion size (+/- 5 mm). Data regarding clinical and procedural characteristics was collected. Outcomes examined were curative resection and complications at the time of procedure, and local recurrence or colectomy within 1 year. Statistical analysis was performed using Fisher’s exact test and the Mann-Whitney U test. RESULTS: Five patients with UC and large rectal neoplasm were matched in a 1:1 fashion to patients without IBD with sporadic large rectal lesions. Patients had UC for a median duration of 13 years, with a median Simple Clinical Colitis Activity Index of 1. 60% of patients with UC had left-sided disease and 40% had pancolitis. 40% were treated with a 5-ASA, 20% with a biologic and 40% were not on IBD medication. After matching, clinical and procedural characteristics were similar (Table 1). Patients with UC had a younger age (57 vs 74 years, P = 0.05). No patients had previously undergone endoscopic mucosal dissection or lesion tattoo. All patients with UC had fibrosis. All resected lesions in the UC group were dysplastic with one carcinoma in the non-IBD group. No patients experienced complications and 80% of patients achieved a curative resection. One patient in the UC group with positive resection margins experienced recurrence at the ESD site after 4 months and underwent proctocolectomy, and one underwent colectomy for high-grade dysplasia separate from the ESD site. The non-IBD patient with carcinoma had >1 mm invasion to the submucosa and was considered non-curative but declined further surgical management. CONCLUSION: ESD for large colorectal lesions appears to be similarly effective in patients with IBD versus those without IBD, and may be an alternative to colorectal surgery. Further study in larger cohorts is needed to assess the utility of endoscopic management of neoplasia unresectable by routine polypectomy.Table 1