Yield of high-grade dysplasia based on polyp size detected at colonoscopy: a series of 2295 examinations following a positive fecal occult blood test in a population-based study.

PURPOSE

The aim was to determine the rate of high-grade dysplasia among patients with all adenomas, and its prevalence in patients with adenomas of different sizes in a well-defined population-based study.

POPULATION AND METHODS

We performed a secondary analysis of the 2295 colonoscopies performed following a positive fecal occult blood test result during the first round of colorectal cancer screening in one French district. The rates of high-grade dysplasia were calculated for 3 size categories of adenoma (diminutive, <or=5 mm; small, 6-9 mm; large, >or=10 mm). Predictive factors for high-grade dysplasia were assessed by univariate and multivariate analyses.

RESULTS

A total of 1284 adenomas were detected in 784 subjects. High-grade dysplasia was present in 32.1% of the 784 subjects and in 2.7%, 16.0%, and 51.1% of those whose adenomas were diminutive, small, and large, respectively. Among subjects with no more than 2 small adenomas, the proportion of those with high-grade dysplasia was 12.4%. Both adenoma size and a villous component within adenomas were found to be independent predictive factors for high-grade dysplasia by multivariate analysis.

CONCLUSIONS

Because of the high rate of high-grade dysplasia among small adenomas, our results reinforce the need to remove all small adenomas detected at colonoscopy. Furthermore, the results suggest that opting for CT colonography surveillance instead of colonoscopic removal among subjects with one or 2 small polyps revealed by CT colonography would have led to missed high-grade dysplasia in 12.4% of them.