Does completing advanced endoscopy fellowship improve outcomes after endoscopic mucosal resection?

Abstract

Background: It was reported that about 60% of the physicians in the USA believed that their Gastroenterology fellowship poorly prepared them for large polyp resection. The aim of this study was to compare endoscopic mucosal resection (EMR) efficacy and complication rates between skilled general gastroenterologists who perform high volume of EMR and advanced endoscopists.

Methods: We identified 140 patients with documented large colonic polyps treated by 4 providers using EMR technique at Carilion Clinic, in Roanoke, Virginia, USA between 01/01/2014-12/31/2017, with follow-up through 10-2018. Information on demographics, clinical and pathological features of high-risk polyps (i.e., size, histology, site, and degree of dysplasia), timing of surveillance endoscopies, tools used during resection, and skills of performing endoscopist's were extracted. The cumulative risks of polyp recurrence after first resection using EMR technique were estimated using Kaplan-Meier curves.

Results: One hundred and forty patients were identified (mean age, 64.1±11.2 years; 47.1% males). Fifty-five polyps (39.3%) were removed by 2 skilled gastroenterologists and 85 (60.7%) were removed by advanced endoscopists. Most of the polyps resected were located in the right colon (63.6%) and roughly half of the polyps were removed in piecemeal fashion. At follow-up endoscopy, the advanced endoscopy group had lower polyp recurrence rates. The median recurrence after polypectomy was significantly different between the groups (0.88 and 1.03 years for skilled gastroenterologists who did not complete and completed EMR hands-on workshops; respectively vs. 3.99 years for the advanced endoscopist who did not complete EMR hands-on workshop, P=0.03).

Conclusions: There is a need for additional EMR training since polyp recurrence was significantly different between the groups despite high rates of piecemeal resection in the advanced endoscopy groups.