Colorectal cancer surveillance by colonoscopy in a prospective, population-based long-term Swiss screening study - outcomes, adherence, and costs.

BACKGROUND: The success of colorectal cancer (CRC) screening depends mainly on screening quality, patient adherence to surveillance, and costs. Consequently, it is essential to assess the performance over time. METHODS: In 2000, a closed cohort study on CRC screening in individuals aged 50 to 80 was initiated in Uri, Switzerland. Participants who chose to undergo colonoscopy were followed over 18 years. We investigated the adherence to recommended surveillance and collected baseline characteristics and colonoscopy data. Risk factors at screening for the development of advanced adenomas were analyzed. Costs for screening and follow-up were evaluated retrospectively. RESULTS: 1278 subjects with a screening colonoscopy were included, of which 272 (21.3%; 69.5% men) had adenomas, and 83 (6.5%) had advanced adenomas. Only 59.8% participated in a follow-up colonoscopy, half of them within the recommended time interval. Individuals with advanced adenomas at screening had nearly five times the risk of developing advanced adenomas compared to individuals without adenomas (24.3% vs. 5.0%, OR 4.79 CI 2.30-9.95). Individuals without adenomas developed advanced adenomas in 4.9%, including four cases of CRC; three of them without control colonoscopy. The villous component in adenomas smaller than 10 mm was not an independent risk factor. Costs for screening and follow-up added up to CHF 1'934'521 per 1'000 persons screened, almost half of them for follow-up examinations; 60% of these costs accounted for low-risk individuals. CONCLUSION: Our findings suggest that follow-up of screening colonoscopy should be reconsidered in Switzerland; in particular, long-term adherence is critical. Costs for follow-up could be substantially reduced by adopting less expensive long-term screening methods for low-risk individuals.

Prospective Swiss pilot study of Endocuff-assisted colonoscopy in a screening population.

BACKGROUND AND STUDY AIMS: The adenoma detection rate (ADR) is one of the quality measures in screening colonoscopy and is crucial for reducing colorectal cancer morbidity and mortality. Up to 25 % of adenomas are missed during colonoscopy. Endocuff is an easy-to-use device that is attached like a cap to the distal tip of the colonoscope in order to optimize visualization behind the folds of the colon and increase the ADR. This is the first prospective study of Endocuff-assisted colonoscopy (EC) in a screening population with follow-up to determine the ADR and adverse events of EC. PATIENTS AND METHODS: We prospectively enrolled asymptomatic patients referred for screening colonoscopy during the 4-month study period. We documented the Boston Bowel Preparation Scale (BBPS) score, cecal intubation rate, polyp detection rate, ADR, number of advanced adenomas, and number of adverse events. Colonoscopies were performed by five board-certified gastroenterologists. During follow-up, the patients were called 4 to 12 weeks after EC. RESULTS: A total of 104 EC procedures were performed. Cecal intubation was achieved in 99 % of the patients, with a median intubation time of 6 minutes. The polyp detection rate and ADR in our study were 72 % and 47 %, respectively, and 13.5 % of the lesions were advanced adenomas. A significant number of adenomas were detected in the right side of the colon. Considering all the adenomas and hyperplastic polyps above the sigmoid, we recommended that nearly 60 % of our patients repeat an endoscopic follow-up according to the existing Swiss guidelines. We noted no perforations or other serious adverse events, even in the patients with extensive diverticulosis. CONCLUSIONS: EC is feasible with the most commonly available colonoscopes without severe adverse events. EC seems to be a safe and effective device for increasing the ADR, including small adenomas in the right side of the colon. Therefore, this technique may be recommended in the future to increase the ADR in a screening population.