Predictability of Capsule Endoscopy Referred to a Tertiary Care Center for Double Balloon Enteroscopy: Should Some Capsules Be Re-read?: 1849

Introduction: Patients with obscure gastrointestinal bleeding (OGIB) with ‘positive’ findings on capsule endoscopy (CE) by gastroenterologists practicing in the community are often referred to our tertiary care center for double-balloon enteroscopy (DBE). We aim to explore the degree of concordance between these 2 procedures, which had been done in 2 different clinical settings. Methods: Patient data including demographics, referral diagnosis (on CE), and confirmatory diagnosis (on DBE) were collected and entered into a secure database. Concordance was calculated using kappa coefficient. A kappa of 0.60 or more suggests strong agreement, 0.40 to 0.60 moderate agreement, and less than 0.40 is indicative of low agreement. Results: A total of 73 patients with OGIB were referred to our center for DBE after undergoing CE outside. Ten of these patients (10/73 or 13.7%) had been found to have blood in small bowel on CE without any concrete diagnosis. Six of these 10 patients (60%) had a normal small bowel on DBE, while 2 patients each (20%) were found to have ulcerative and vascular pathology in the small bowel. Sixty-three patients had a diagnosis of either normal small bowel, or ulcerative, tumorous, or vascular etiology on CE. DBE revealed a source of bleeding in 17 out of the 22 patients (77.3%) with normal CE. On DBE, 35 patients (35/63 or 55.6%) had a vascular pathology, 11 had ulcerative disease (17.5%), 11 showed a normal small bowel (17.5%), while diverticulosis and small bowel tumor were found in 3 patients each (4.8%). The kappa coefficient for CE and DBE for the 63 patients was 0.28, suggesting poor agreement. Most patients with a referral diagnosis of vascular pathology were confirmed to have vascular disease on DBE (19/23 or 82.6%). Similarly, an initial diagnosis of ulcerative disease on CE was confirmed in 5 out of 6 patients on DBE. However, a diagnosis of mass on CE was only confirmed in 2 out of 12 cases (16.6%). Conclusion: Our study shows there is a poor concordance between capsule endoscopy done in the community and confirmatory DBE done at our tertiary care center. While a diagnosis of vascular or ulcerative pathology on CE has high likelihood of being confirmed on DBE, a diagnosis of normal small bowel or mass should prompt re-reading of the capsule study before DBE is pursued.