| [ Summary ] |
We have conducted 93,491 total colonoscopic sessions on 28,035 Japanese to screen for colorectal lesions during our annual medical health check-up from 1983 to 2003. A total of 101 cases of invasive cancers (0.11%), 177 of severe displasia (0.19%), 1,067 of adenomas≥10 mm (1.1%), and 12,701 of adenomas >= 10 mm (13.6%) were found. When compared with the results of immunochemical fecal occult blood tests (FOBT) performed blindly just before colonoscopies, sensitivity of FOBT for invasive cancer, severe dysplasia, and adenoma >=10 mm were 64%, 25%, and 17%, respectively, leading to a rates of 21% of sensitivity for advanced colorectal lesions, in total. The risk ratios for advanced lesions, compared to the risk at the first examinations, progressively increased in relation to the interval after the recent examination, from an odds ratio of 0.21 at 1 year (p<0.001),to 0.39 at 2 years (p<0.001), 0.47 at 3 years (p<0.001), 0.64 at 4years (p=0.04), 0.84 at 5years (p=0.49), 0.86 at 6-8 years (0.51), to 2.1 at 9-11 years (p=0.01). The difference in risk ratios between 1 year and 2 year intervals was also significant. An increase in invasive lesions was found at the two year interval. In summary, screening total colonoscopy could detect many more significant colorectal lesions than FOBT. The effects of screening total colonoscopy carried out over at least a four year period and were not seen over a ten year period. However, annual colonoscopies seemed to provide the maximum benefit when the problem of cost was not considered. |