Rectal aberrant crypt foci identified using high-magnification-chromoscopic colonoscopy: Biomarkers for flat and depressed neoplasia

BACKGROUND: Aberrant crypt foci may represent preneoplastic lesions in the human colon. The prevalence of aberrant crypt foci detected using magnification chromoscopic colonoscopy is known to follow a stepwise progression from normal subjects to those with exophytic adenomas and colon cancer. No studies have addressed the prevalence of rectal aberrant crypt foci in patients with flat and depressed colonic lesions that cluster within the right hemi-colon and may undergo de novo neoplastic transformation. METHODS: All patients underwent total colonoscopy by a single endoscopist using the Olympus CF240Z magnifying colonoscope. Flat and depressed lesions were diagnosed using targeted indigo carmine chromoscopy. Prior to extubation, pan high-magnification-chromoscopy using indigo carmine was applied to the rectum and the distal 10 cm of mucosa examined using forward and retroflexed views. Aberrant crypt foci were defined as two or more crypts with dilated or slit-like openings that were raised above the adjacent mucosa. Using high-magnification chromoscopic colonoscopy we assessed the prevalence and dysplastic features of aberrant crypt foci in three groups: endoscopically "normal" subjects, patients with flat/depressed adenoma, and flat/depressed cancer. RESULTS: Two thousand five hundred and fifty-nine patients underwent colonoscopy of which 1,000 were eligible for inclusion. The median number of aberrant crypt foci per patient in the endoscopically normal, adenoma, and cancer group was 1 (range: 0-5), 9 (range: 0-22), and 38 (range: 14-64), respectively. The estimated relative risk of dysplastic aberrant crypt foci when comparing the flat adenoma group with the endoscopically "normal" group was 4.68 (95% CI: 2.23-9.91) with the relative risk for flat cancer versus endoscopically normal group being 21.8 (95% CI: 10.9-23.8). Patients with > 5 flat adenomas had higher crypt foci densities than those with < 5 adenomas (r = 0.53; p < 0.001). CONCLUSIONS: The number of aberrant crypt foci in normal patients, patients with flat adenoma, and flat cancer follow a stepwise incremental change as previously observed for exophytic adenomas and cancer. Detection of aberrant crypt foci in the rectum may be a useful biomarker for proximal colonic flat neoplasia and could be used at index flexible sigmoidoscopic screening to stratify risk of proximal colonic neoplasia. Patients with dysplastic aberrant crypt foci of high density should receive total colonoscopy.