Introduction Double Balloon Enteroscopy (DBE) is widely used in clinical practice worldwide. A DBE service at South Tyneside District Hospital was commenced in January 2010 to complement the existing capsule endoscopy (CE) service. We present the results of an audit of prior investigation before DBE.

Methods Clinical records were examined for patients referred for DBE with a diagnosis of suspected Crohn's disease. Information was gathered regarding: place of referral, previous imaging and endoscopy, findings and histology.

Results 28/37 (77%) of referrals were from outside our hospital. 15/37 referred for investigation of Crohn's disease, 75% of these were from outside our hospital. Seven patients with known Crohn's were referred for investigation of recurrent symptoms or for possible stricturing disease. Eight patients were referred with possible Crohn's based on clinical symptoms and signs. All patients had been previously investigated with multiple endoscopic or imaging modalities. Most common method of prior imaging for patients being investigated for Crohn's disease was Barium follow through (BaFT) 42%, followed by CE 33%, CT 12.5%, MRE 12.5%. 87% had a colonoscopy prior to referral. 11/15 had abnormal imaging, 5 (33.3%) having inflammatory changes seen on CE. Of these histology was taken in three and found: Crohn's (1), non-specific inflammation (1), normal (1). 3/5 cases were normal at DBE. 4/15 had entirely normal previous investigations. Of the eight patients with suspected Crohn's, two patients with abnormal radiology had DBE findings consistent with Crohn's. Of the three patients with SB ulceration on CE only one had an abnormal DBE and histology obtained was inconclusive. See Abstract PMO-210 table 1.

Conclusion A large proportion of our referrals are to investigate Crohn's disease. Most have had multiple endoscopic and imaging modalities prior to referral. DBE is more likely to have positive findings when associated with abnormal imaging rather than abnormal VCE findings. Only those with abnormalities on imaging other than VCE were confirmed to have Crohn's disease; it may be that subtle inflammatory changes on VCE are over reported or that findings were beyond the reach of DBE. Our figures although small would suggest that in those with normal radiological imaging there is little improvement in diagnostic yield with DBE.

Competing interests None declared.