BACKGROUND: Duodenal polyposis is a manifestation of adenomatous polyposis that predisposes to duodenal or ampullary adenocarcinoma. Duodenal polyposis is monitored by upper GI endoscopies and may require iterative resections and prophylactic radical surgical treatment when malignancy is threatening. OBJECTIVE: The purpose of this study was to evaluate severity scoring for surveillance and treatment in a large series of duodenal polyposis. DESIGN: From 1982 to 2014, every patient surveyed by upper GI endoscopies for duodenal polyposis was included. SETTINGS: The study was conducted at a single tertiary care center. PATIENTS: We performed 1912 upper GI endoscopies in 437 patients (median = 3; interquartile range, 2-6 endoscopies). MAIN OUTCOME MEASURES: Conservative treatment was performed in 103 patients (159 endoscopic and 17 surgical resections), whereas radical surgical treatment (Whipple procedure or duodenectomy) was required in 52 (median age, 47.5 y; range, 43.0-57.3 y) because of high-grade dysplasia or unresectable lesions. RESULTS: Genes involved were APC (n = 274; 62.7%) and MUTYH (n = 21; 4.8%). First upper GI endoscopies (median age, 32 y; range, 21-44 y) revealed duodenal polyposis in 190 (43.5%). Rates of low-grade dysplasia, high-grade dysplasia, and duodenal or ampulary adenocarcinoma at 5 years were 65% (range, 61.7%-66.9%), 12.1% (range, 10.3%-13.9%), and 2.4% (range, 1.5%-3.3%), whereas 10-year rates were 75.8% (range, 73.1%-78.5%), 20.8% (range, 18.2%-23.4%), and 5.4% (range, 3.8%-7.0%). The rate of ampullary abnormalities rose during surveillance from 18.3% at the first upper GI endoscopies to 47.4% at the fourth. Predictive factors for high-grade dysplasia were age at first upper GI endoscopy, type and age of colorectal surgery, Spigelman score, presence of an ampullary abnormality, and number of endoscopic treatments. In multivariate analysis, only age at first upper GI endoscopy and presence of an ampullary abnormality were independent predictive factors. Histologic analysis after radical surgical treatment showed high-grade dysplasia in 30 patients and duodenal or ampulary adenocarcinoma in 11 (4 patients had lymph node involvement). LIMITATIONS: The study was limited by its retrospective analysis of a prospective database. CONCLUSIONS: More than 20% of patients developed high-grade dysplasia with duodenal polyposis after 10 years. Iterative endoscopic resections allowed extended control, but surgery remained necessary in 12% of the patients and happened too late in many cases; 20% of those operated had developed duodenal or ampulary adenocarcinoma, whereas 8% exhibited malignancy with lymph node involvement. The trigger for prophylactic surgery required a more accurate predictive score leading to closer endoscopic surveillance. Modifying the Spigelman score by accounting for ampullary abnormalities should be considered as a means to increase compliance with closer endoscopic follow-up in high-risk patients. See Video Abstract at http://links.lww.com/DCR/A430.