Endoscopic mucosal resection defect inspection for predicting recurrences: International image-based survey

Background and study aims Endoscopic mucosal resection (EMR) is a safe and minimally invasive procedure to remove colorectal non-pedunculated polyps. Recurrence rates are relatively high and differ among endoscopists. We aimed to evaluate whether endoscopists are able to predict recurrence based on thorough inspection of images of mucosal defects after an assumed complete EMR. Methods We developed an online survey in which endoscopists were invited to indicate whether they expected recurrence to develop when inspecting 30 post-EMR defect images. All EMRs were considered to be complete resections by the performing endoscopist. Participating endoscopists were scored based on the number of correct answers regarding presence or absence of recurrence found at first surveillance colonoscopy. Results A total of 140 endoscopists responded to the survey (response rate 25%). A total of 124 respondents with a mean age of 46.5 years evaluated the 30 images. The overall score in the cohort was 70%, indicating that respondents were able to correctly predict recurrence in three-quarters of cases with an overall level of certainty of 53.4%. When comparing results of experienced and less experienced endoscopists based on the number of endoscopic submucosal dissections and/or EMRs performed yearly, no difference (71% versus 69%, P = 0.23) was found. Conclusions This study shows that recurrences after presumed complete EMR can reasonably well be predicted by both experienced and less experienced endoscopists when evaluating images with mucosal defects. Thorough inspection of the post-EMR defect may reduce recurrence rates by recognizing and subsequent treatment of suspect areas.