Accuracy of detection of colorectal neoplasia using an immunochemical occult blood test in symptomatic referred patients: comparison of retrospective and prospective studies.

Objective: In this study the sensitivity and specificity of immunochemical tests for colorectal neoplasia were evaluated in retrospective and prospective studies.

Methods: Four types of fecal blood tests--a chemical test (Hemoccult II) and three different immunochemical tests including a test which detects hemoglobin and transferrin- were performed in the retrospective study. In the prospective study the test for hemoglobin and transferrin was used for all patients that underwent total colonoscopy.

Patients: One hundred seven patients with colorectal neoplasia, 57 with gastroduodenal bleeding, and 62 with normal digestive tracts were examined retrospectively. One thousand two hundred and ninety-eight nonspecifically symptomatic patients whose endoscopic examination was negative for hemorrhagic lesions in the upper digestive tract were examined prospectively.

Results: In the retrospective study, sensitivities for the detection of colorectal cancers and adenomas with diameters > or =10 mm using the tests which detect hemoglobin and transferrin were 98% and 89%, respectively. These were the highest sensitivity among the four tests. The specificity of this test was 97%, which was higher than that of the Hemoccult II test. In the prospective study, the sensitivities of the tests for hemoglobin and transferrin for the detection of colorectal cancers and adenomas with diameters > or =10 mm were 79% and 33%, respectively. The specificity was 95%.

Conclusions: The test for hemoglobin and transferrin showed the highest sensitivity and specificity for colorectal neoplasia in the retrospective study. The sensitivity and specificity of this test were not so high in the prospective study, but they may be clinically applicable in the evaluation of patients with various nonspecific symptoms.