Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

P623. Forty-year analysis of colonoscopic surveillance program for ulcerative colitis


While it is well recognised that patients with ulcerative colitis (UC) have an increased risk of colorectal cancer (CRC), the precise magnitude and the proportion of patients progressing to CRC from pre-malignant neoplastic lesions is still being debated. This study reports on data collected from patients with endoscopically and histologically confirmed extensive UC over a 40-year period from a tertiary centre in the UK.


A retrospective analysis of UC patients enrolled in long-term surveillance at St Marks Hospital, UK was performed. Data were obtained from the prospective surveillance database, medical records, surgical, endoscopy and histology reports. The primary end point was defined as death, colectomy, withdrawal from surveillance, or census date (January 1, 2013).


Since 1971, a total of 1,375 patients underwent 8,623 (median = 5, IQR 3–8) colonoscopies during 17,444 patient-years of follow up. The median follow up duration was 10 years (IQR 6–15 years), and 857 patients were still on the surveillance program as of January 1, 2013. Median age of UC onset was 30 years (IQR 22–40). A total of 318 (23.1%) patients developed neoplasia of which 69 (5%) had CRC. The proportion of patients who had progressed to CRC for each type of neoplasia during surveillance was: adenoma (3/79, 3.8%), indefinite dysplasia (12/49, 24.5%), LGD (19/131, 14.5%), and HGD (8/27, 30%). Median age and disease duration at the time of CRC diagnosis was 55 years (IQR 49–63) and 22 years (IQR 14–29), respectively. The locations of CRCs were: distal to splenic flexure (40, 58%), proximal to splenic flexure (24, 35%), multifocal CRC involving both proximal and distal colon (3, 4%), or unidentified (2, 3%). Surgical specimen revealed that 21 (30%) of CRCs were accompanied by at least one or more focus of DALM, of which 6 (8.4%) were multi-focal CRCs affecting two or more segments of colorectum. Within surveillance, the cumulative incidence of CRC by disease duration was 0.2% at 10 years, 2.6% at 20 years, 7.3% at 30 years, 10.1% at 40 years and 13.1% at 50 years.


The overall incidence of CRC, even among patients with extensive UC at the tertiary referral centre, was considerably lower than reports from other earlier published studies. However, neoplasia in any form (i.e. adenoma, LGD, HGD, or CRC) was common, eventually affecting almost 1 in 4 patients undergoing surveillance. Given the variable rates of progression from dysplasia to CRC and relatively high incidence of multifocal lesions, patients with any grade of dysplasia require close monitoring and a careful pan-colonic examination to ensure that no lesions are missed.

  • Written by:

    C.H. Choi1, M. Rutter2, M. Kang3, J. Warusavitarne1, S. Thomas-Gibson1, M. Moorghen1, B. Saunders1, T. Graham4, A. Hart1, 1St Mark's Hospital, Department of IBD, Colorectal Surgery and Wolfson Unit of Endoscopy, London, United Kingdom, 2University Hospital of North Tees, Department of Gastroenterology, Stockton-on-Tees, Teesside, United Kingdom, 3Imperial College London, School of Medicine, London, United Kingdom, 4Queen Mary University of London, Centre for Tumour Biology, Barts Cancer Institute, London, United Kingdom