Indigo Carmine Staining and Narrow Band Imaging (NBI) of Colonic Mucosa for Detection of Dysplasia in Ulcerative Colitis

Specific Aims:

Primary:

1. To assess intra- and inter-observer agreement for the detection of abnormalities seen
by white light colonoscopy followed by indigo carmine staining and white light
colonoscopy followed by NBI imaging.

Secondary:

2. To estimate the sensitivity and specificity of these modalities for dysplasia among UC
patients who undergo surveillance colonoscopy.

3. To assess intra- and inter-observer agreement for abnormalities identified by confocal
laser microscopy among patients with chronic ulcerative colitis.

4. To determine the specificity and sensitivity of confocal microscopy in the detection of
dysplasia among patients with ulcerative colitis.

50 consecutive patients with clinically quiescent UC presenting for surveillance colonoscopy
will be randomized to white light followed by NBI OR white light followed by indigo carmine
staining. Identified abnormalities will then be targeted for confocal imaging in 40
patients. Immediately prior to confocal imaging, fluorescein 2.5 ml of 10% solution will be
administered intravenously. For these patients, confocal images will also be obtained of the
area around the lesion and of the colon segments as in the parent study and correlated with
pathology. Gross activity will be determined by white light colonoscopy for each colon
segment. At least 32 biopsies for surveillance will be obtained with additional biopsies of
abnormalities present on white light, NBI or indigo carmine. Photographs will be obtained of
all abnormalities prior to biopsy and of random mucosa without abnormalities. Intra- and
inter-observer agreement will be based on the first 10 patients at each site through review
of de-identified endoscopic photographs. If modifications are necessary because of poor
agreement, re-assessment will be made after the next 10 patients. Otherwise each site will
enroll 40 additional patients with agreement summarized and sensitivity and specificity of
these techniques for detecting dysplasia

Significance: If intra and inter-observer agreement for mucosal abnormalities detected by
indigo carmine and NBI modalities is shown to be good, in ulcerative colitis, then larger
studies will be planned on UC surveillance populations. If confocal techniques are feasible
then these will be integrated into larger studies.

Inclusion Criteria:

1. Clinically quiescent (UC) disease will be defined as a colitis activity index <8 at
the time of enrollment.

2. Scheduled for routine, clinically-indicated colonoscopy surveillance

Exclusion Criteria:

1. Patients unable to give informed consent,

2. Pregnant or female patients of child bearing age who are sexually active an not
practicing birth control methods. (in this group a pregnancy test is required as per
routine clinical care)

3. Poor colonic preparation (<90% colon visualization)

4. Documented allergy to indigo carmine

5. Coagulopathy that can not be corrected (INR> 1.5 and/or platelet count<50,000)