No abstract

Colonoscopy is the established method of surveillance of subjects at high risk of developing colorectal neoplasia but the procedure is expensive, time consuming and occasionally hazardous. Faecal occult blood tests can be prepared at home and are cheap, simple and safe. Hemeselect is an immunological faecal occult blood test that is more sensitive for colorectal cancer than Haemoccult. The aim of this study was to determine the sensitivity of the Hemeselect test for asymptomatic colorectal neoplasia in subjects at high risk of the disease who were undergoing colonoscopy, thus assessing its suitability as an alternative means of screening high-risk groups. A total of 919 asymptomatic subjects were asked to complete Hemeselect tests. These were completed satisfactorily by 808 individuals (compliance rate 88 per cent) and were positive in 164 patients (20 per cent). At colonoscopy 11 cancers were detected in ten patients (seven Hemeselect positive) and 36 (16 Hemeselect positive) had at least one adenoma 1 cm or more in diameter. The test sensitivites of Hemeselect for carcinoma and large (1 cm or more) adenomas were 70 and 44 per cent respectively. In a subset of 417 subjects who also completed Haemoccult tests, the sensitivities were 33 and 18 per cent. Hemeselect specificity is 88 per cent compared with 98 per cent for Haemoccult. While the sensitivity of Hemeselect is higher than that of Haemoccult, it is still insufficient to replace colonoscopy in high-risk groups.

(Gut 1997; 40: 110-112) Keywords: faecal occult blood tests, colorectal cancer.cancer (HNPCC) but by flexible sigmoidoscopy or FOBT for lesser degrees of risk. We wished to investigate the sensitivity and specificity of HemeSelect in a group at high risk undergoing colonoscopy and the effect of adding HS to flexible sigmoidoscopy in people at lesser risk. Methods PATIENTSHO and HS were examined in two groups of subjects, detailed below. Subjects were asked to complete HO and HS tests on each of three daily bowel motions before endoscopy. Rehydration of HO test cards was not performed. A blue discolouration at 30 seconds was taken as a positive reaction after the application of two drops of 1% w/w hydrogen peroxide.3 Erythrocyte agglutination at a 1:8 dilution was taken as a positive reaction for HS.3 No dietary restriction was enforced. Participants were recruited, to the study, from an open access clinic for relatives of colorectal cancer patients and a number of case finding initiatives ongoing in our unit. Overall compliance, with endoscopy was 73% at the time of study, but many people were self referring volunteers. The male:female ratio was 1:1-58. Subjects who presented with symptoms were excluded from analysis.

This study is a 2-year follow-up of an average-risk population offered screening with both Haemoccult and Hemeselect tests to determine the interval cancer rate, and thus sensitivity. The effect on compliance with Hemeselect of testing over 1 day rather than 3 days was investigated in a separate cohort. In the first study, 3948 subjects received tests; 1489 (37.7 per cent) completed both tests and 148 had a positive result, 17 (1.1 per cent) were Haemoccult positive and 145 (9.7 per cent) were Hemeselect positive. Investigation of 142 patients revealed ten with cancer (Dukes stage A, seven; B, one; C, two). All were detected by Hemeselect but only one was Haemoccult positive. After a median follow-up of 35 (range 26-43) months, seven further patients developed colorectal cancer (stage A, one; B, three; C, three) but none followed a negative Hemeselect test (100 per cent sensitivity). In the second study 2703 subjects were offered Hemeselect tests. Compliance for testing over 1 day (48.6 per cent) was significantly better than that over 3 days (43.1 per cent) (chi 2 = 8.1, 1 d.f., P < 0.01). Hemeselect is a promising screening test for the early detection of colorectal cancer.