Abstract
Based on the 206 cases of colorectal cancer we have encountered during the past seven years, we have evaluated the usefulness of testing for occult blood in feces in the screening of colorectal cancers; we have also evaluated the sensitivity of colonoscopy in the detection of colorectal neoplasm. These 206 cases of colorectal cancer were divided into 2 groups: the asymptomatic group (145 cases) and the symptomatic group (61 cases). Ninety six per cent of the asymptomatic group had positive occult blood in feces. The asymptomatic group had early cancers in 85% of the cases (mucosal invasion: 55%; submucosal invasion: 30%) or Dukes A lesions in 87%, and received endoscopic resection in 64%, whereas the symptomatic group had advanced cancers in 72% of the cases, Dukes C lesions in 38%, or were inoperable in 11%, and received surgical resection in 69%. Of 750 cases in which a total of 1,406 lesions had received endoscopic resection, the initial colonoscopy done elsewhere had overlooked 60 lesions (4.3%) of over 5mm in size. Of these, 7 lesions were of over 1cm in size and 2 (other than the previous 7) were early cancers. The overlooked lesions were mostly smaller than 1 cm and of flat shape, and were often found in the sigmoid and ascending colon, but the highest rate (12.8%) of overlooking was noted in the cecum. Of 57 cases of early cancer, 23 cases (40.3%), including 3 submucosal cancers, could not be correctly diagnosed as malignancy through endoscopic biopsy done elsewhere prior to endoscopic resection. These results suggest that testing for occult blood in feces is very useful for the screening of colorectal cancer in its curable stage and that colonoscopy should be performed very carefully because it is not perfect, despite the fact that it is considered to be the most excellent method of detecting colorectal neoplasm.
