Evaluation of an immunochemical test for faecal occult blood in screening for colorectal neoplasia in a high risk group.

A group of 1615 asymptomatic individuals presumed to be at increased risk of colorectal neoplasia were selected as the study group. All were tested by an immunochemical test for faecal occult blood. In the second half of the study, individuals who were occult blood negative were offered colonoscopy. Ninety-nine individuals (6.1%) returned stools which were occult blood positive. Investigation by full colonoscopy was possible in 90 cases, revealing nine patients (10%) with invasive cancers, four patients (4.4%) with carcinomata in situ, and 36 patients (40%) with adenomata. Non-neoplastic pathology capable of producing occult blood positive stools was found in 31 individuals (34.4%). No pathology was found in 10 instances (11.1%). Of the 53 occult blood negative subjects who underwent colonoscopy, eight were found to have adenomata. Only one of these was larger than 5 mm in diameter (18 mm). No carcinomata were found. The site within the large bowel of the tumour did not appear to significantly affect the occult blood status of the faeces but the size of the tumours detected suggests that the presence of blood within the stool is more likely to be associated with larger lesions. The frequency of detection of both carcinomata and adenomata was greatest in individuals who had a past history of colorectal neoplasia. The individual cost of this immunochemical test is nominal. The high diagnostic yield and low false positive and negative rates suggest that case follow-up, surveillance, or screening utilizing this test is justified.