RESUMEN
BACKGROUND: Due to the increased risk of colorectal cancer, colonoscopic surveillance is recommended for patients with ulcerative and Crohn's colitis. Because surveillance intervals differ considerably between the recently updated American Gastroenterological Association (AGA) and British Society of Gastroenterology (BSG) guidelines, we compared the neoplasia yield and colonoscopic workload of these guidelines. METHODS: Patients with inflammatory bowel disease undergoing surveillance were identified using medical records. Patients were stratified according to the BSG and AGA guidelines, and corresponding colonoscopic workload was calculated based on the risk factors present during follow-up. The incidence of colitis-associated neoplasia (CAN), defined as a low-grade dysplasia in flat mucosa or a non-adenoma-like mass, high-grade dysplasia, or colorectal cancer was compared between the risk groups of either guidelines. RESULTS: In total, 1018 patients with inflammatory bowel disease who underwent surveillance were identified. Using the AGA surveillance intervals, 64 patients (6%) were assigned to annual and 954 patients (94%) to biannual surveillance, resulting in 541 colonoscopies per year. The yield of CAN was 5.3% and 20.3% in the low- and high-risk groups, respectively (P = 0.02). Using the BSG surveillance intervals, 204 patients received surveillance annually (20%), 393 patients every 3 years (39%), and 421 patients every 5 years (41%), resulting in 420 colonoscopies per year, which is 22% lower than the AGA guidelines. The yield of CAN was 3.6%, 6.9%, and 10.8%, for the low-, intermediate-, and high-risk groups, respectively (P = 0.26). CONCLUSIONS: Although the BSG surveillance intervals offer the advantage of a lower colonoscopic workload, the risk stratification of the AGA seems superior in distinguishing patients at higher risk of CAN.