Introduction Double balloon enteroscopy (DBE) offers the ability for diagnostic and therapeutic intervention in the small bowel. The procedure takes an average of an hour to carry out and can be uncomfortable for the patient. It also has an associated learning curve for the endoscopist. The depth of insertion is a subjective estimation of the total number of passes into the small bowel. The success/diagnostic yield of the procedure relies upon patient tolerability and locating the target lesion or bleeding point. The aim of this study was to assess the technical success rate for routine DBE over time in comparison to sedation dose used.

Methods A prospective review of the 290 DBE procedures done since the start of the service was conducted. The majority of patients underwent a capsule endoscopy either locally or at our centre prior to DBE which helped to guide the chosen route. Data was collected for sedation/analgesia used, procedure length, number of passes into the small bowel and diagnostic yield between the initial 145 (group 1) and latter 145 procedures (group 2). Similar comparisons were also done between the oral and anal routes of DBE.

Results The DBE procedures were performed from July 2006 to Nov 2012 by two endoscopists. Whilst the median doses of midazolam used between the two groups were similar (median 5 mg versus 6mg, p = 0.8), a greater amount of fentanyl was used in group 2 (median 50mcg versus 100mcg, p < 0.001). There was no difference in the procedure length or the number of passes recorded by the endoscopists. However there was a significant increase in the diagnostic yield in the latter group (32% versus 58%, p < 0.001).

A total of 165 oral DBE procedures were done. Whilst there were no differences in the procedures length or number of passes into the small bowel, a greater amount of fentanyl was used in the latter half of the procedures. The diagnostic yield improved significantly in the latter half of the oral DBE procedures (41% versus 61%, p = 0.01). There were 125 anal DBE procedures. There was no difference in procedure characteristics or diagnostic yield for the anal route over time despite higher doses of fentanyl.

Conclusion This study demonstrates an improved diagnostic yield for DBE over time particularly with the oral route and with a greater amount of fentanyl used. The absence of improvement in yield for the anal route in this study is consistent with the literature to date. The anal route remains a challenge for endoscopists due to difficulty achieving a stable platform in the terminal ileum to progress. More education on retrograde techniques would help overcome this limitation.

Disclosure of Interest None Declared.