RESUMO
BACKGROUND: This prospective study examined the appropriate use of colonoscopy in an open-access system with the American Society for Gastrointestinal Endoscopy guidelines and determined whether the American Society for Gastrointestinal Endoscopy guidelines were associated with relevant endoscopic findings. METHODS: In a cohort of 2221 consecutive patients referred for colonoscopy, the proportion of patients who underwent colonoscopy for appropriate indications was prospectively assessed. The relationship between appropriateness and the presence of clinically relevant endoscopic diagnoses was assessed by calculating (1) the positive and negative likelihood ratio of the indications; and (2) the change in the probability of relevant endoscopic diagnoses in the presence of the American Society for Gastrointestinal Endoscopy criteria. RESULTS: The rate for 'generally not indicated' colonoscopies was 37%. Relevant endoscopic diagnoses were present in 28.5% of cases with American Society for Gastrointestinal Endoscopy indications versus 20.1% of patients without appropriate indications. However, the risk of finding relevant diagnoses was significantly increased by American Society for Gastrointestinal Endoscopy criteria application (odds ratio (OR) 1.58; 99% CI 1.20-2.07; p<0.01). Furthermore, in both endoscopic situations (appropriate and not appropriate), the likelihood ratio, positive and negative, varied very little, suggesting a low predictivity for serious pathologies by the appropriate procedure. CONCLUSIONS: The use of an appropriateness evaluation system makes it possible to increase the probability of finding relevant endoscopic diseases. However, the exclusive use of such a system for selecting patients to undergo colonoscopy involves a relatively high risk of colorectal neoplasms going undetected.
