Characterization of colonic polyps at conventional (nonmagnifying) colonoscopy after spraying with 0.2 % indigo carmine dye

Abstract

Japanese endoscopists have devised a classification system based on mucosal crypt patterns which is helpful for distinguishing between hyperplastic polyps, adenomas, and invasive cancers at colonoscopy. The aim of this study was to assess how well the various types of colonic polyp could be distinguished using conventional colonoscopes after spraying with 0.2 % indigo carmine dye. The endoscopic appearances of all colonic lesions were assessed in 476 unselected patients using normal-resolution, nonmagnifying colonoscopes after spraying with 0.2 % indigo carmine dye. A total of 709 lesions were found in the 476 patients, and histology was available for 673 of these lesions: 187 lesions were found to be non-neoplastic (128 hyperplastic, 2 juvenile, 30 inflammatory, and 27 classified as "others"); 467 lesions were adenomatous; and 19 lesions were carcinomas. Of the 467 adenomas, 377 were tubular, 77 were tubulovillous, 8 were villous and 5 were serrated; 423/467 were correctly identified (sensitivity 91 %). Of the 187 non-neoplastic lesions, 153 were correctly classified (specificity 82 %). A total of 343 of the 377 tubular lesions were correctly identified as tubular adenomas (sensitivity 90 %), and 46 of the 77 tubulovillous lesions were correctly identified. Standard colonoscopy with dye spraying can be used to differentiate colonic polyps. Magnification is not always necessary to distinguish neoplastic from nonneoplastic colonic lesions. This finding could result in resource savings in colonoscopic screening.