Am Fam Physician. 2002;65(4):725-726
The subject of colorectal cancer screening has filled many pages in the recent medical and lay press. Some authors have suggested combining use of stool cards for occult blood testing with sigmoidoscopy to increase the sensitivity of colon cancer detection. Lieberman and colleagues previously reported on the results of screening asymptomatic patients with colonoscopy to determine the rate at which examination of the distal colon detects advanced neoplasia. The authors now report on a separate analysis of the same study population, this time describing the results of patients who had stool card examinations before colonoscopy.
The authors examined 17,732 patient records to find 3,196 patients who were asymptomatic for any condition that might suggest bowel pathology (i.e., abdominal pain, rectal bleeding, change in stool habits) and had no serious comorbid conditions or previous colon cancer screening. Stool card results before colonoscopy were available for 2,885 patients (92.4 percent). The cards were rehydrated before testing, which has been shown to increase the sensitivity of occult blood detection. The location and size of all lesions seen on colonoscopy were recorded, and lesions within the rectum and sigmoid colon served as a proxy for what would have been obtained by flexible sigmoidoscopy.
Advanced colonic neoplasia, defined as a tubular adenoma of 10 mm or more in diameter, any villous adenoma, adenoma with high-grade dysplasia, or invasive cancer was seen in approximately 11 percent of study participants. The combination of stool cards plus sigmoidoscopy would have identified 76 percent of all cases of advanced neoplasia. In the remaining 24 percent of cases, no lesions were within reach of the sigmoidoscope nor was there a positive stool card result (either of which factors would typically prompt a full colonoscopic examination). Although performing stool card testing before endoscopic examination would have reduced the total number of endoscopies needed to find all of the neoplasia cases (those with a positive stool card test would skip sigmoidoscopy and proceed to colonoscopy), the authors stated that it did not add significantly to the number of cases that would have been detected by sigmoidoscopy alone.
The authors conclude that performing stool card testing before sigmoidoscopy does not significantly increase the detection of colon cancer, and about one fourth of advanced colonic neoplasia cases would not be detected by this combined regimen.