P284 Non-ampullary duodenal adenomas: a tertiary centre experience

Introduction

Duodenal polyps are encountered in up to 5% of patients referred for gastroscopy.1 The optimal management of duodenal polyps has not been established and the risks of endoscopic resection are not well known. The aim of this retrospective study was to review the experience duodenal polypectomy in a tertiary referral centre with respect to safety profile and recurrence rates.

Methods

All patients undergoing endoscopic resection of duodenal polyps between August 2016 and November 21 were included. Electronic medical records were reviewed and demographics, lesion characteristics, procedural details and adverse events were recorded. Univariate and multivariable analysis were performed to determine risk factors for adenoma recurrence.

Results

44 patients were included in the study (mean age 61.9 years, 20 males, mean lesion size was 18.7mm). 16 were attempted under conscious sedation and 24 under propofol. 14 lesions were in D1, 24 in D2 and 2 in D3. 4 lesions were not attempted due to being circumferential and one would not lift. The remained were resected either enbloc (n=24) or piecemeal (n=14). 23% (9/39) were less than 10mm in size, 38% (15/39) were between 10 and 20mm in size, 35.9% (14/39) were between 20 and 50mm in size and one lesion was 60mm. 2/39 (5%) were readmitted with possible bleeding but no intervention was needed. Lesions removed piecemeal were significantly larger than those removed enbloc (median 20mm vs 12mm, p=0.036). Histology showed 24 were tubulovillous adenomas with low grade dysplasia, 3 were tubulovliious adenomas with high grade dysplasia, 3 were hamartomas, 1 was a T1 cancer (completely resected), 1 was a neuroendocrine tumour, 1 was hyperplastic, 1 was normal tissue and 5 were not retrieved. 11/39 had evidence of residual adenomatous tissue at follow up of which 4 were resected endoscopically. Logistic regression adjusting for lesion size, lesion location, lesion histology, type of sedation and enbloc/piecemeal resection showed piecemeal resection was independently associated with residual adenomatous tissue (adj OR 10.67, 2.0 - 55.5, p=0.0049).

Conclusions

This study shows endoscopy resection of duodenal polyps was feasible, with low adverse event rates and high complete resection rates even with large lesions. The main risk factor for incomplete resection was need for a piecemeal resection.

Reference

Jepsen JM, Persson M, Jakobsen NO, Christiansen T, Skoubo-Kristensen E, Funch-Jensen P, Kruse A, Thommesen P. Prospective study of prevalence and endoscopic and histopathologic characteristics of duodenal polyps in patients submitted to upper endoscopy. Scand J Gastroenterol. 1994 Jun;29(6):483-7. doi: 10.3109/00365529409092458. PMID: 8079103.