What is the learning curve associated with double-balloon enteroscopy? Technical details and early experience in 6 US tertiary care centers

Background: Performance parameters for double-balloon enteroscopy (DBE) have not been described. Objective: To determine the learning curve for DBE. Design: Prospective cohort study. Setting: Six U.S. tertiary centers. Patients: A total of 188 subjects undergoing 237 DBE procedures; 130 (69%) with obscure GI bleeding. Interventions: Performance parameters from each center's initial 10 cases were compared to the subsequent examinations. Main Outcome Measurements: Exam duration, depth of insertion, and findings on DBE examination. Results: DBE was introduced by mouth in 149 (63%) cases, by rectum in 77 (33%) cases, and through a stoma in 6 (2.5%) patients. The mean (+/- SD) duration was 109.1 +/- 44.6 minutes for the first 10 cases and 92.4 +/- 37.6 minutes for subsequent cases (P =.005) but did not change for rectal DBE procedures. There was no change in mean depth of insertion, but the mean fluoroscopy time declined significantly (P =.025). Diagnostic or therapeutic maneuvers were performed in 64% of cases; DBE led to a diagnosis in 81 (43%) patients. A total of 78% of patients had prior capsule endoscopy (CE) with significant agreement between DBE and CE (kappa = 0.74). One perforation occurred (0.4%). Per-rectal cases failed to reach the small bowel in 24 (31%) cases. Limitations: All patients did not undergo initial CE. The therapeutic DBE scope was not available for the initial 8 months of the study. Conclusions: There was a significant decline in overall procedural time and fluoroscopy time after the initial 10 DBE cases. There was no improvement in performance parameters when DBE was performed via the rectal approach despite increased, but limited, operator experience.