REAL-TIME FEEDBACK TO IMPROVE COLONOSCOPY

ABSTRACT Colorectal cancer is the second leading cause of cancer-related deaths in the US, claiming about 50,000 lives in 2015. Colonoscopy is currently the preferred screening modality for colorectal cancer; in theory colonoscopy should prevent most colorectal cancers. However, recent data suggest that there is a significant miss-rate associated with colonoscopy for the detection of even large polyps and cancers. Indeed, reports from Canada and Germany fail to show any protection of colonoscopy for right-sided and only about 70% protection for left-sided colorectal cancer. Reports from the US show that prevention of colorectal cancer related death is at best 53-68% in carefully controlled longitudinal studies. Furthermore, patients of endoscopists with the highest rate of detection of premalignant polyps had the lowest rate of colorectal cancers in subsequent years. In 2015, the American College of Gastroenterology and the American Society of Gastroenterology renewed consensus guidelines defining a good quality colonoscopy but adherence to these guidelines varies among endoscopists and there is growing concern that these guidelines do not reflect true quality. We hypothesize that real-time objective feedback during the withdrawal phase of colonoscopy will improve quality of colonoscopy. We have created software that in real-time evaluates the technique of the endoscopist, categorizes technique into different degrees of quality (similar to software in airplanes that warns pilots against stalling) within milliseconds and via heads-up display technology can inform the endoscopist about the level of measured quality. In other words, we have created a system that has the potential to improve endoscopic technique during a procedure in a live patient. Using this system, we propose to address two Specific Aims. First, to determine the optimal method of real-time feedback for four features of colonoscopy related to the amount of stool in the colon and the effort of the endoscopist to inspect as much as possible all mucosa of the colon. And second, to test whether real-time feedback during colonoscopy improves quality of colonoscopy and the adenoma detection rate in three endoscopy centers. Successful evaluation and implementation of the proposed real-time analysis and feedback system has the potential to improve the quality of care of over 14 million US citizens ? the approximate number of people undergoing colonoscopy ? on an annual basis. In addition, the technology lends itself for rapid adaptation to other endoscopic medical procedures such as upper gastrointestinal endoscopy, cystoscopy, arthroscopy and bronchoscopy.