Introduction Although consistent risk factors for recurrence after resection of colorectal polyps have been identified in eastern series, there are few data on risk factors in western practice which may have different patient populations, experience, referral patterns and employ different techniques. Our unit is one of only a few specialist interventional endoscopy units in the UK. As a result, many referred patients have very large polyps, deeply scarred lesions from previous attempts at resection and several had been deemed too frail to undergo treatment. We aimed to determine the risk factors for recurrence after endoscopic resection of large colorectal polyps in this population.

Methods We analysed a series of endoscopic resections of large colorectal polyps. Several techniques were employed including EMR, ESD and hybrid techniques involving combinations of EMR, ESD, TEMS and transanal resection for particularly challenging polyps. Visible vessels were routinely coagulated. After resection, the area was scrutinised with magnification chromoendoscopy to check for residual polyp which was resected or ablated. Surveillance endoscopy was performed at 3 and 12 months.

Results 363 polyps with a mean size of 56 mm were resected in 326 patients who had a mean age of 71 years: 309 by EMR, 38 by ESD and 16 by hybrid procedures. Mean follow up was 12.2 months. 38% of polyps were deeply scarred. Recurrence occurred in 9.7% of patients, 17% of which were diminutive. 66% of recurrences were apparent on the first surveillance endoscopy.

Size > 30 mm, piecemeal resection, deeply scarred lesions and the use of argon plasma coagulation were associated with recurrence on univariate analysis. However, logistic regression revealed only piecemeal resection was independently associated with recurrence (OR 5.1, p = 0.03). Intraprocedural bleeding, old age, high grade dysplasia, rectal location and histological type were not significantly associated with recurrence. Furthermore, there were no significant differences in recurrence between lesions resected by ESD or EMR, or between lesions resected with traditional techniques and those using a hybrid of various endoscopic and minimally invasive surgical techniques.

Conclusion In contrast to other series, we found only piecemeal resection to be associated with recurrence. We feel that the routine use of techniques such as post-resection assessment using magnification chromoendoscopy to detect residual polyp and routine coagulation of visible vessels helps to eliminate some of the risk factors for recurrence. These techniques may also account for the success of hybrid procedures to resect large polyps in difficult locations with similar recurrence rates, which are an invaluable option in some challenging cases.

Disclosure of Interest None Declared