The Significance of Capsule Endoscopy (CE) Findings: Results from the CORI Database

Abstract

CE is frequently performed for the evaluation of obscure gastrointestinal bleeding. The significance of many findings, such as non-bleeding AVMs and red spots, is still unknown. Additionally, the influence of rate of bleeding (overt vs. occult) on CE findings is unknown. Methods: The results from CE exams entered into the CE module of a national database (CORI) were reviewed. Statistical analysis was performed by Pearson's chi-square with Yates' correction. Results: From 2/28/03-6/30/03 157 CE procedures in 148 patients were recorded from 8 sites by 12 physicians. 123 were for obscure gastrointestinal bleeding: overt (hematochezia, melena, hemorrhage) n=41, occult (iron deficiency anemia, occult positive stools) n=82. CE findings: mucosal abnormality 58%, AVM 34%, active bleeding 11%, nodule/polyp 10%, ulcer 6%, stricture 5%, tumor 2%. A potential cause for bleeding was found in 56% of the overt and 46% of the occult bleeding patients (p=NS). CE for overt bleeding was more likely to find active bleeding than in the occult bleeding patients: 22% vs. 6%, p=0.02. There were otherwise no significant differences between overt and occult groups. Compared to 24 patients undergoing CE for non-bleeding indications, AVMs were more commonly reported in those with bleeding: 12% vs. 34%, p=0.02. Patients with bleeding were also less likely to have a completely normal exam (no findings recorded): 17% vs. 55%, p=0.04. Conclusions: These data from the CORI multicenter database confirm that approximately half of patients undergoing CE will have a potential bleeding source identified. Patients presenting with overt bleeding are more likely to have active bleeding seen on CE. AVMs are more frequently seen in bleeding patients and are likely a clinically significant finding.