Randomised Trial of NBI for Adenoma Detection

• Difference in number of patients with at least one histologically demonstrated adenoma between the two groups
  

• Number of adenomas detected in each arm
  
• number of advance adenomas (>20% villous elements, >10mm, high grade dysplasia) detected in each arm
  
• Total number of flat adenomas detected in each arm
  
• total number of non-neoplastic lesion in each arm
  
• total number of patients with 3 or more adenomas detected in each arm
  
• total number of patients with 5 or more adenomas detected in each arm
  
• assessment of video still for detection of neo-plastic or non-plastic nature of polyps, AFI, vs NBI vs NBI with magnification versus white light
  
• Sub-group analysis of primary end point according to indication for colonoscopy
  

Colorectal cancer is the second commonest cause of cancer death. In a majority of cases it is preceded by a precancerous lesion called an adenoma (polyp). Detection and removal of adenomas at colonoscopy has been shown to reduce the death rate from colorectal cancer. However, despite meticulous examination there is a "miss rate" for adenomas at colonoscopy which ranges from 66-25% in back-to-back colonoscopy studies. The nature of the polyps which as well as being pedunculated (cherry like) can be flat or depressed making them difficult to see, which may contribute to the "miss rate".

The factors, which influence the endoscopist detection are not well studied. Polyp detection rates wary widely, even among experts. Techniques that highlight lesions have advanced in recent years. Chromoendoscopy, the current gold standard technique, relies on spraying dye on the bowel lining, has been shown to help pick up more pre-cancerous polyps in two of three studies; however it is not widely used as it is time consuming and requires extra equipment and training. Narrow band imaging (NBI) is a technique that relies in light filters to improve contrast for the smallest blood vessels in the bowel lining which shows up adenomas as they have a richer vascular network. It is sometimes described as "digital chromoendoscopy" as the images produced are similar to chromoendoscopy, but it is much simpler and quicker to use. Autofluorescence endoscopy uses short wavelength light and light filters to produce a false colour image of the bowel lining where polyps stand out. These techniques have been used with some success in the oesophagus and stomach but little work is available for the colon.

We aim to see if NBI is better than standard colonoscopy for detecting precancerous polyps. This is likely as it is similar to chromoendoscopy which is already shown to help. If a polyp is found we will use other types of endoscopy, particularly NBI with magnification and autofluorescence to see if these techniques are helpful for discriminating between pre-cancerous and non pre-cancerous polyps.

May 2007 protocol minor amendment: additional viewing by endoscopists outside St Mark's ro allow assessment of inter- and intra-observer variability. No additional data collected.