Abstract
Diminutive colon polyps, defined as 5 mm or less, are encountered increasingly at colonoscopy. The risk of serious pathology in such polyps is low. There is a risk and cost of resecting all such polyps and sending tissue for pathologic evaluation. Enhancement of endoluminal imaging may enable discrimination of neoplastic vs non-neoplastic polyps. If this discrimination can be performed accurately with high confidence, it may be possible to either resect and discard diminutive adenomas, or inspect and do-not-resect diminutive hyperplastic polyps. In 2011, an expert group recommended thresholds of 90% negative predictive value for adenomas, and 90% accuracy in predicting appropriate surveillance intervals. Since 2011, criteria for polyp discrimination have been published and validated by experts and nonexperts. In vivo studies have been performed to compare endoscopic impression and pathologic diagnosis. An expert panel was convened in late 2014 to review the literature to determine if the proposed thresholds for discrimination can be attained and to recommend the next steps for introducing changes in clinical practice. We conclude that threshold levels can be achieved with several endoscopic image enhancements. The next steps to implementation of practice change include acquiring data on training and competence, determining best practices for auditing performance, understanding patient education needs, and the potential cost benefit of such changes.
