Introduction The role of endoscopic resection for colonic polyps previously destined for surgery is expanding. However, surgery remains appropriate in some cases. The aim of this study was to examine tertiary polyp referrals that did not undergo endoscopic polypectomy. The objectives were to determine (i) the proportion of polyps referred for polypectomy that were not endoscopically resected, (ii) the primary reason in this decision-making and (iii) factors associated with polyps that were not endoscopically resected.

Methods A prospective observational study of all polyps referred for endoscopic resection (ER) to a tertiary centre between January 2010 and August 2012 was performed. For each case, ER was either completed, abandoned or not attempted. The primary reason for abandoning or not attempting ER was documented. Demographics, polyp characteristics and histology were recorded and a comparative analysis (using chi-square test and independent-samples T test) was made between patients in whom ER was abandoned or not attempted with those in whom ER was completed.

Results ER was either abandoned (n/29) or not attempted (n/55) in 84 of 423 polyp referrals. This was most commonly because of suspected invasive cancer (45/84). Of these 45 polyps, 12 had characteristic macroscopic features of cancer on inspection. In 24/45, invasive cancer was suspected after advanced endoscopic examination (including surface morphology (Paris/NICE/Kudo) classification and forceps palpation). In 9/45, invasive cancer was only suspected during attempted ER, which was then abandoned. The remaining 41/84 polyps for which ER was abandoned or not attempted appeared benign. The positive and negative predictive values of endoscopic evaluation for the diagnosis of invasive cancer were 86% and 96% respectively. The benign-appearing polyps were not endoscopically resected because of (i) a high risk location (ie. overlying the appendix, IC valve or a diverticulum), n = 12; (ii) difficult access, n = 12; (iii) size ≥ 5 cm combined with other factors, n = 8; (iv) age/comorbidities, n = 4 or (v) poor tolerance of colonoscopy, n = 2. Forty-six percent of these benign polyps were in the caecum. In comparison with patients who underwent complete ER, those in whom ER was abandoned or not attempted were more likely to be female (56 vs. 37%, P < 0.001), had larger mean polyp size (4.7 cm vs 3.7 cm P < 0.001), and had a higher incidence of polyp cancer on histology (47 vs. 2.7% P < 0.001).

Conclusion Twenty percent of polyps referred to a tertiary institution for polypectomy may not be suitable for endoscopic resection. This is most commonly due to the presence of invasive cancer which can usually be recognised by endoscopic examination.

Disclosure of Interest None Declared.