Outcomes of Endoscopic Management of Familial Adenomatous Polyposis Associated Duodenal Adenomas: A Tertiary Center Experience

Introduction: Familial adenomatous polyposis (FAP) is associated with a high prevalence of duodenal adenomas necessitating frequent endoscopic surveillance and resection. Our study aims to describe resection patterns and to determine the short- and long-term adverse events from endoscopic management. Methods: We performed a retrospective review for consecutive FAP patients who underwent endoscopic resection of duodenal lesions at Indiana University hospital between January 2000 and December 2021. We excluded patients < 18 years of age and those who had prior duodenal malignancies or duodenal resections prior to index endoscopy. The cohort was divided into 2 groups - those with ampullary involvement (AI) and those with no ampullary involvement (NAI). Results: One hundred and ten patients with FAP were included. Patient, lesion, and resection characteristics are found in Table 1. 47% had ampullary involvement at their index endoscopy and 73% of patients had high-risk adenomas (HRA) defined as adenomas larger than 10mm, had high grade dysplasia or were tubulovillous histology. Median lesion size was 10mm. The AI group had a higher incidence of intraprocedural adverse events (12% vs 10% in the NAI group), with intraprocedural bleeding requiring intervention being the most common complication. Perforations occurred in 1% of all patients (2% in AI vs 0% in NAI group). There was a higher rate of early adverse events requiring readmission within 30 days after the endoscopy in the AI group (35% vs 4% in the NAI group). Post-ERCP pancreatitis occurred in 3% of all patients (6% in AI vs 0% in NAI group). 10% of those with intraprocedural and early adverse events required hospitalization with a median hospital stay of 3 days and 3% required additional endoscopic procedures for complication management. Only 3 patients from the AI group developed complications beyond 30 days of the initial endoscopy including orifice stenosis and biliary strictures. No post-procedure deaths were reported. Conclusion: This study is one of the largest cohorts of FAP patients undergoing endoscopic resection for duodenal lesions to date. Endoscopic resection carries notable per-procedure risks, which are more common with ampullary adenoma resection, but most are not serious and do not require hospitalization. Post-resection pancreatitis (in ampullary lesions) and perforation are feared short-term adverse events but are uncommon in expert hands. Long-term complications are rare and are more common in ampullary resections. Table 1. - Patient, Lesion and Resection Characteristics and Intraprocedural and Early Adverse Events Related to Endoscopic Resection Patient Characteristics All patients, n (%)(n=110) Ampullary, n (%)(n=52) Non-ampullary, n (%)(n=58) Age in years (Median, (IQR)) 39 (28,51) 44 (35,57) 37 (22,49) Gender Female 57 (52%) 30 (58%) 27 (47%) Male 53 (48%) 22 (42%) 31 (53%) Ethnicity Caucasian 102 (93%) 48 (92%) 54 (93%) Black/African American 5 (5%) 3 (6%) 2 (3%) Hispanic/Latino 1 (1%) - 1 (2%) Asian - - - Other/Undocumented 2 (1%) 1 (2%) 1 (2%) Endoscopic polyp size, mm (Median, (IQR)) 10 (5,15) 10 (10,18) 7(3,13) Histologic polyp size,mm(Median, (IQR)) 8.5 (5.5,12) 12 (8,15) 7 (4,9) En-bloc resection Yes 91 (83%) 39 (75%) 52 (90%) Polyp morphology Sessile 95 (86%) 37 (71%) 58 (100%) Semisessile - - - Pedunculated 1 (1%) 1 (2%) - Unknown/undocumented 14 (13%) 14 (27%) - Polyp histology Non-advanced Adenomas 48 (44%) 14 (27%) 34 (59%) Advanced Adenomas Tubulovillous adenoma 13 (12%) * 10 (19%) * 3 (5%) * Adenomas >= 10mm 60 (55%) * 38 (73%) * 22 (38%) * High grade dysplasia 7 (6%) 7 (13%) - Normal duodenal tissue 3 (3%) - 3 (5%) Not documented 3 (3%) 3 (6%) - Tool used Hot snare 60 (55%) 48 (92%) 12 (20%) Cold snare 40 (36%) 3 (6%) 37 (64%) Biopsy forceps 7 (6%) 1 (2%) 6 (10%) Intraprocedural Adverse Events Perforation 1 (1%) 1 (2%) 0 (0%) Bleeding requiring intervention 11 (10%) 5 (10%) 6 (10%) Early Adverse Events (< 30 days) Post-sphincterotomy bleed 4(4%) 4 (8%) - Abdominal pain 2 (2%) 1 (2%) 1 (2%) Pancreatitis 3 (3%) 3 (6%) 0 (0%) Hospitalization 11 (10%) 10 (19%) 1 (2%) Median number of days) 3 3 1.5 Need for repeat intervention related to adverse event 3 (3%) 3 (6%) 0 (0%) *Some adenomas overlap in terms of high-risk features.