Risk analyses for screening sigmoidoscopy based on a colorectal cancer (CRC) population

Objective. Although colonoscopy can be effective in the prevention of colorectal cancer (CRC), it requires many endoscopic facilities, has a high patient burden and risk of complications, and it is expensive. The aim of this study was to determine the risk for proximal CRC and to identify subgroups in which screening sigmoidoscopy can be effective. Material and methods. A database search was carried out on all patients who underwent endoscopy of the lower gastrointestinal (GI) tract between 1997 and 2005. All patients diagnosed with CRC were included. Variables including age, gender and the presence of distal colonic neoplasia were used for risk analyses. Results. In total, 783 patients were diagnosed with CRC. Tumour was located in the proximal colon in 68/255 (27%) of the patients<65 years. Of the patients<65 years, 22% (57/255) had proximal CRC without synchronous distal lesions and would thus have been missed by sigmoidoscopy screening. Among patients >65 years, 41% (216/528) were diagnosed with proximal CRC, significantly more often in women than in men (p<0.001). In 35% of patients (185/528) proximal CRC without distal colonic neoplasia was found, significantly more than in those under 65 years of age (p<0.001). Conclusions. Significantly more proximal localized CRC would have been missed by sigmoidoscopy screening in elderly patients, especially in women. In subjects<65 years of age, sigmoidoscopy screening allows detection of almost 80% of CRC cases and might suffice as a screening method.