* = Presenting author

P340. Has the risk of developing colorectal cancer in patients with ulcerative colitis been overstated? A meta-analysis


C. Castaño-Milla1, M. Chaparro1, J.P. Gisbert1

1Hospital De La Princesa, IP, Gastroenterology and CIBEREHD, Madrid, Spain



Background: It is known that patients with ulcerative colitis (UC) have an increased risk of developing colorectal cancer (CCR), but there is controversy about the magnitude of this effect.

Aim: To conduct a meta-analysis of studies evaluating the incidence of CCR in patients with UC.

Methods: Selection of studies: studies describing the incidence and prevalence of CCR in UC. Search strategy: a MEDLINE search using the keywords: colorectal cancer, colon cancer, dysplasia, ulcerative colitis and inflammatory bowel disease, up to November 2011 was performed. Data synthesis: Cumulative incidence and incidence rates of CCR were combined and meta-analyzed using the generic inverse variance method. Subanalises: subanalyses were performed depending on the time of follow-up, the extension of UC, the study design, and the year of publication.

Results: 67 studies, including a total of 57391 patients, met the inclusion criteria. The overall cumulative incidence of CCR in patients with UC was 18.8/1000 patients (95% CI, 15.8–22; I2 = 86%). The overall incidence rate of CCR was 1.67 per 1000 patient-years (py) (1.41–1.92; I2 = 83%). When considering those studies that only included patients with extensive colitis, the incidence rate was higher: 3.7/1000 py (2.40–5; I2 = 39%). 20 studies reported results stratified into 10 years of disease duration intervals: in the first decade, the incidence rate was 1.01/1000 py (0.71–1.31; I2 = 14%), in the second decade it was 3.75/1000 py (2.51–4.98; I2 = 83%), and it was 5.85/1000 py (3.73–8; I2 = 83%) in the third decade. When only population-based studies were considered, the incidence rate was 1.24/1000 py (0.97–1.5; I2 = 79%). When results were pooled classified by the publication year, the incidence rates were as follow: 4.29/1000 py (0.95–7.64; I2 = 59%) in the 50s, 3.70/1000 py (1.98–5.42; I2 = 73%) in the 60s, 3.86/1000 py (1.24–6.74; I2 = 90%) in the 70s, 2.67/1000 py (1.75–3.57; I2 = 87%) in the 80s, 1.16/1000 py (0.75–1.57; I2 = 70%) in the 90s, and 1.09/1000 py (0.81–1.38; I2 = 80%) from 2000 to 2011.

Conclusions: The risk of developing CCR in UC patients seems to be lower than previously suggested. The incidence rate of CCR was, overall, of only 1.67 per 1000 py, and this figure was even lower in population studies. The extension and the duration of the disease increase the risk of CCR. It seems to be a decreasing trend in the incidence of CCR in UC patients in the last decades.