Introduction: The implementation of bowel cancer screening in the UK requires the maintenance of high standards in colonoscopy. Part of this quality control requires the reliable documentation of complete colonoscopy that can be externally audited and assessed. It has been suggested that terminal ileal biopsy is the only definitive and reliable method of confirming caecal intubation, but it is not cost-effective and may now be contraindicated because of potential prion infection.
Objective: To determine how reliable routine terminal ileal images were as an independent predictor of complete colonoscopy and whether their interpretation was aided with water insufflation or indigo-carmine dye-spraying. Method Forty-nine histologically confirmed terminal ileal images were obtained from a single endoscopist's database; 19 were conventional white-light images, 15 were taken with water insufflation and 15 were taken using chromoscopy enhancement. The images were transferred onto CD-ROM and sent as a questionnaire to 42 colonoscopists who were asked to identify the images as terminal ileum or not.
Results: Twenty questionnaires were returned resulting in a total of 980 responses. Overall, the accuracy of positive identification was 53.4%. Water insufflation and chromoscopy improved the accuracy to 68.3% and 63% respectively. Experience of (> 1000 colonoscopies) did not increase overall accuracy. Less experienced endoscopists had an increased accuracy rate with dye-spraying (76.7%vs 59.3%, P < 0.05) but experienced endoscopists had an increased accuracy rate with water insufflation (67.4%vs 63.3%, P > .05).
Conclusion: Currently, terminal ileal imaging is not a reliable mode of documenting complete colonoscopy. Using water insufflation or dye-spraying coupled with modifications in image acquisition technique may improve its reliability but these methods require further investigation before they can replace the use of caecal landmarks as completion parameters.