Background: Indirect evidence and modeling analyses suggest that colonoscopy may be the most cost-effective way to screen the average-risk population for colorectal neoplasia. However, the success and safety of primary colonoscopic screening has not been prospectively evaluated in a multicenter trial.
Methods: Asymptomatic subjects age 50 to 75 years who had not undergone examination of the colon within 10 years were recruited from the general medicine clinics of 13 Department of Veterans Affairs Medical Centers. Eligible patients underwent colonoscopy by study coinvestigators, at which time all polyps were measured, photographed, and removed. Patients were contacted at 24 hours and 1 week to track procedure-related complications.
Results: Primary screening colonoscopy was performed in a cohort of 3196 asymptomatic subjects. A "good" preparation was reported in 81% of patients, and colonoscopy to the cecum was successful in 97.2% of cases. Mean insertion time to the cecum and total procedure times were 10.5 (8.7) and 30.6 (19.1) minutes, respectively. No preprocedural patient characteristics were identified that were predictive of an incomplete procedure. At least one polyp was resected in 1672 patients. There was no perforation and no death attributed to colonoscopy. Major morbidity considered to be definitely related to colonoscopy occurred in 9 of 3196 procedures (0.3%): lower GI bleeding requiring intervention (6), myocardial infarction and/or cerebrovascular accident (2), and thrombophlebitis (1). In subjects undergoing only diagnostic procedures, the major complication rate was 0.1%.
Conclusions: Screening colonoscopy can be performed in multiple centers with a high degree of success and safety in large numbers of asymptomatic, average-risk men.