Characterization and classification of duodenal adenomas in patients with familial adenomatous polyposis by using narrow band imaging.

Familial adenomatous polyposis (FAP) is an inherited autosomal dominant
condition, in which multiple gastrointestinal adenomas are found. Aside from
colonic polyps, duodenal adenomas have a reported prevalence of nearly 100%.
Since these adenomas have a malignant potential, duodenal surveillance
endoscopies are recommended to overcome the occurrence of duodenal cancer.
Until now, no accurate method has been identified for the detection of patients
at high risk for the development of duodenal cancer. We evaluate the use of
Narrow Band Imaging (NBI) as a novel endoscopic imaging technique to define
groups at high risk for cancer development.

The aims of the study are: 1) to characterize the mucosal morphology and
vascular pattern of neoplastic and non neoplastic mucosa by NBI and standard
White Light Endoscopy (WLE) in the upper gastrointestinal tract; 2) to compare
the accuracy of NBI and WLE for the identification of high grade neoplastic and
villous duodenal adenomas; 3) to compare NBI and WLE in staging duodenal
adenomatosis according to Spigelman's classification.

Patients with FAP will undergo an upper endoscopy with WLE, followed by a
second inspection with NBI. The stomach will be investigated for fundic gland
polyposis and suspicious polyps. In the duodenum the number and size of
detected polyps will be noted. Suspicion of duodenal adenomas is based on size
> 10 mm, central depression, irregular surface or irregular vascular pattern.
Overview and focused WLE and NBI images of all suspicious adenomas and 2
control areas are taken before sampling. Histopathology will be linked to
corresponding images. All images are reviewed according to a systematic image
evaluation process, consisting of an unblinded assessment of an exploratory set
of images and a blinded evaluation of a learning set and validation set.
Histopathological outcome will be used as gold standard for diagnosis.

In comparison to the procedure for regular patient care, the endoscopy time for
study patients takes about 5 minutes longer. Secondly, additional biopsies are
taken in study patients, although sampling by a biopsy forceps has no risk for
perforation.