Abstract

Background and Aims: Almost all patients with familial adenomatous polyposis (FAP) develop duodenal adenomas, with a 4% to 18% risk of progression into duodenal cancer. Prophylactic endoscopic resection of duodenal adenomas may prevent cancer and is considered safer than surgical alternatives; however, data are limited. Therefore, the aim of this study was to assess safety and effectiveness of endoscopic duodenal interventions in patients with FAP. Methods: We performed a historical cohort study including patients with FAP who underwent an endoscopic duodenal intervention between 2002 and 2018. Safety was defined as adverse event rate per intervention and effectiveness as duodenal surgery–free and duodenal cancer–free survival. Change in Spigelman stage was assessed as a secondary outcome. Results: In 68 endoscopy sessions, 139 duodenal polypectomies were performed in 49 patients (20 men; median age, 43). Twenty-nine patients (14 men; median age, 49) underwent a papillectomy. After polypectomy, 9 (13%) bleedings and 1 (2%) perforation occurred, all managed endoscopically. Six (21%) bleedings (endoscopically managed), 4 (14%) cases of pancreatitis, and 1 (3%) perforation (conservatively treated) occurred after papillectomy. Duodenal surgery–free survival was 74% at 89 months after polypectomy and 71% at 71 months after papillectomy; no duodenal cancers were observed. After a median of 18 months (interquartile range, 10-40; range, 3-121) after polypectomy, Spigelman stages were significantly lower (P <.01). Conclusions: In our FAP patients, prophylactic duodenal polypectomies were relatively safe. Papillectomies showed substantial adverse events, suggesting its benefits and risk should be carefully weighted. Both were effective, however, because surgical interventions were limited and none developed duodenal cancer.

Original languageEnglish
Pages (from-to)457-466
Number of pages10
JournalGastrointestinal Endoscopy
Volume93
Issue number2
Early online date2020
DOIs
Publication statusPublished - Feb 2021

Cite this