P535 Outcomes of endoscopic resections of large non-polypoid lesions inflammatory bowel disease: a single United Kingdom centre experience

Abstract

Introduction: Patients with colitis carry an increased risk for the development of dysplasia compared to those without1. The SCENIC consensus statement recommends endoscopic resection of all visible dysplasia2. Due to technical challenges and limited experience in the West of large colitis associated non-polypoid endoscopic resections, such patients are often subjected to colectomy. The King's Institute of Therapeutic Endoscopy (KITE) is a tertiary centre for endoscopic assessment and resection of large/challenging colorectal polyps. Here we present the largest single-centre case series of large non-polypoid resections associated with colitis. Aims & Methods: Adults with confirmed colitis (ulcerative colitis extending beyond the rectosigmoid junction and crohn's colitis affecting at least the left colon) with lesions at least 20mm in size within the colitis segment were included. Data including demographics, clinical history, lesion characteristics, method of resection and post-resection surveillance were collected prospectively in patients from January 2011 to November 2016. Resection techniques included endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD) and hybrid ESD. Surveillance of resection site with magnification chromoendoscopy (mCE) was performed at 3 months with pan colonic mCE at 1-year post resection and annually thereafter. Result(s): Thirteen lesions satisfied the inclusion criteria in 13 patients. Patient demographics and clinical data are presented in table 1. Mean lesion size was 47.3+/-22.4 (20-90) mm. All lesions were non polypoid with distinct margins and no ulceration. High-frequency mini-probe ultrasound confirmed intramucosal lesions in 5 cases where pit/vascular pattern was distorted due to inflammation. En bloc resection was achieved in 6 cases. 69% lesions were deeply scarred of which 66% had experienced prior instrumentation. Resection of a single lesion was abandoned due to intense fibrosis. Macrosocpic evidence of complete resection was achieved in all remaining cases. Endoscopic diagnosis of pre-cancerous lesions of less than 1000 mum submucosal invasion was confirmed histopathologically in 100% of resected lesions. Complete excision was confirmed in all en bloc resections. A single case of small perforation and another with delayed minor bleeding were both managed endoscopically. Mortality/hospital admission within 30 days post resection was 0%. Median follow up was 28 months (12- 35) with no recurrence. Alternative site dysplasia was detected in 2 patients. All lesions were sub 20mm and resected endoscopically. Two patients were referred for colectomy due to a concomitant diagnosis of neuroendocrine tumour and the second with alternate site advanced dysplasia. Conclusion(s): This cohort series demonstrates that endoscopic resection of large non-polypoid lesions in association with colitis is feasible using an array of resection methods, safe and has good long term outcomes in a western tertiary endoscopic centre (Table Presented).