Histopathology of Small Polyps Removed in the Videoendoscopic Era

Abstract

Background: With the advent of CT-colonography, the importance (or lack of importance) of small (<1.0 cm) colorectal polyps has become increasingly controversial. At least 90% of colorectal adenomas are <1.0 cm in size. The natural history of small adenomas is unknown. Aim: We sought to describe the histopathology of small colorectal adenomas removed via videoendoscopy and the level of interobserver variation. Methods and Statistical Analysis: Consecutive polyps (10,780) removed during consecutive colonoscopies (10,034) at a single center were reviewed between January 1999 to January 2004. Patients with a history of inflammatory bowel disease (IBD) were excluded. A subset of 393 histopathology slides of consecutive polyps <1.0 cm in size were reviewed by three expert GI pathologists from outside our center. Each was blinded to the interpretations of other pathologists. Advanced adenomas were defined as having features of tubulovillous, villous, serrated adenomas, or high-grade dysplasia. Cochran-Mantel-Haenszel tests were used for the comparisons between the pathologists. Results: 8,798 (81.6%) of polyps were <6.0 mm and (based on the interpretations of our center's pathologists) 4,279 (48.6%) within this group were adenomas and 111 (1.26%) were advanced adenomas. 1,282 (11.9%) polyps were between 6.0 - 9.9 mm and 753 (58.7%) within this group were adenomas and 73 (5.7%) were advanced adenomas. There were large variations between expert pathologists in the interpretation of villous elements and high-grade dysplasia. Thus, for polyps < 6.0 mm in size in the sample reviewed by outside experts, the range of identification of villous elements between pathologists was 0.45% - 3.8%, and the range of high-grade dysplasia was 1.3% - 9.5%. For adenomas between 6.0 - 9.0 mm in size, the range of villous elements was 0.79% - 11.3%, and the range of interpretations of high-grade dysplasia was 1.9% - 27.7%. Conclusion: If CT colonography policy is that polyps < 1.0 cm or ≤ 5 mm are ignored or not referred for polypectomy, the substantial majority (91% and 78% respectively) of colorectal neoplasms will be left in place. This would be a substantial paradigm shift in colorectal cancer prevention strategy. Estimating the clinical significance of small polyps by their histology is complicated by wide variation in interpretation of villous elements and high-grade dysplasia, even among experts.