Our understanding of the natural history of upper gastrointestinal (GI
) involvement in familial adenomatous polyposis (FAP) is still evolvin
g, although we know that the main cause of death after colectomy in FA
P is upper GI malignancy, affecting 5% of patients. The aim of duodena
l surveillance is to target high risk individuals and identify cancers
early. We have screened 200 patients prospectively and have observed
that duodenal polyposis progresses slowly, but there are some young pe
ople who have severe disease who merit close observation. We pay parti
cular attention to endoscopic technique and histological detail, and u
se a duodenal staging system. Patients are offered randomisation to st
udies of chemopreventive agents, and those with advanced disease are c
onsidered for surgery. Successful management is inhibited by our defic
ient knowledge of the natural history of upper gastrointestinal polypo
sis, and by our inability to identify high risk individuals with histo
logical markers rather than because of any technological deficiencies
in endoscopic equipment.