Recurrence rate and risk factors following cold snare endoscopic mucosal resection of polyps ≥20 mm in size

Document Type

Conference Proceeding

Publication Date


Publication Title

Gastrointestinal Endoscopy


Introduction: Traditionally, endoscopic mucosal resection (EMR) for polyps ≥20mm has been performed with snare cautery. Due to adverse events (AE) associated with cautery, such as bleeding and perforation, the use of cold snare EMR has increased. These AE are not routinely seen with cold EMR, due to the lack of cautery. Data evaluating adenoma/sessile polyp recurrence rates (ARR) and risk factors for recurrence after cold snare EMR of polyps ≥20mm has not been fully assessed. The aim of this study is to define ARR for polyps ≥20mm removed via cold snare EMR and risk factors for recurrence. Methods: A retrospective chart review was conducted from 1/2015 to 7/2019 at a single tertiary care center. In this period, 469 patients underwent piecemeal cold snare EMR of polyps ≥20mm by 5 endoscopists. Complete resection of the polyp was documented in all cases. Of these, 310 had a surveillance colonoscopy and were included in the study, however, 159 were lost to follow up. Data including age, gender, race, history of predisposal to colon cancer, smoking and alcohol history, polyp location, histology, dysplasia at index, and size of the polyp were collected. Results: 108 (34.8%) patients had evidence of recurrence on pathology at follow up colonoscopy. The average time for follow in the ARR group was 5.7 months. There was no association between recurrence and gender, personal history of polyposis syndrome, family history of colon cancer, smoking, alcohol use, number of polyps removed at index colonoscopy, and the polyp location. Patients with recurrence were found to be older (p=0.008), Asian or African American compared to Caucasian (p=0.02); and required the use of endoscopic clips (p=0.017) at index procedure. Recurrence rates were higher in larger polyps (34 vs 27 mm, p=<0.001). Also, ARR increased with polyps >30mm (26.7% in polyps 20-30mm vs 76.9% in polyps >50mm). There was a significant association between polyp histology and ARR; tubulovillious adenomas (p=<0.001) and high-grade dysplasia (p=0.003) more likely to recur while tubular adenomas and sessile serrated polyps had lower ARR. Among patients with follow-up procedures, there was no interval cancer found. Conclusion: To our knowledge this is the largest, if not only, cohort looking at ARR for polyps ≥20mm removed via piecemeal cold snare EMR. Surprisingly, the ARR in the cold EMR group was 34.8%, which is higher than the average ARR commonly seen with hot EMR (∼20%). Factors that influenced the ARR were older age, Asian or African American race, endoclip use, polyp size and histology. Cold snare EMR has been shown to be a safe procedure but the ARR and factors that affect this rate must be considered when managing these patients. Prospective trials are needed to further validate these findings.





First Page


Last Page


This document is currently not available here.