The positive predictive value of guaiac faecal occult blood test in relation to the number of positive squares in two consecutive rounds of colorectal cancer screening

The aim of this study was to define the positive predictive values of a positive guaiac faecal occult blood test according to the number of positive squares, in two consecutive rounds of colorectal cancer mass screening in a French region. A total of 4172 colonoscopies were analyzed. Sex, age, number of positive squares, and colonoscopic and histopathologic findings were studied. In the results obtained, 76.6% of positive tests were positive with one or two squares. The number of positive squares was not related to sex, age and rank of participation. The positive predictive value for cancers and adenomas increased significantly with age, sex (male) and number of positive squares from 6.6% (one to two squares) to 27.6% (five to six squares) and from 15.2% to 22.2%, respectively. Cancer was diagnosed 211 times (54.1%) and advanced neoplasia was diagnosed 696 times (65.3%) following positive tests with one to two squares. The TNM stage of cancer increased significantly with the number of positive squares: 85.8% of stages 0-1-2 for one to two positive squares and 66.3% for five to six positive squares (P<0.001). Multivariate analysis showed an increased risk of cancer and advanced neoplasia for male patients and aged persons. The number of positive squares significantly increased the risk of cancer (odds ratio = 4.6 for five to six positive squares) and the risk of advanced neoplasia (odds ratio = 2.9). Age, sex and number of positive squares were independent predictive factors of positive guaiac faecal occult blood test. The proportion of TNM stages 3-4 was significantly higher in those with five to six positive squares. Performing a complete colonoscopy in every individual having a positive test, especially aged men with a high number of positive squares, should be a priority in any screening programme. European Journal of Cancer Prevention 20: 277-282 (C) 2011 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins.