Diagnostic yield of double-balloon endoscopy in patients with obscure GI bleeding

Background Double-balloon endoscopy (DBE) is a new method that allows visualization, tissue sampling, and therapeutic intervention of a variety of pathologies throughout the small-intestinal tract. Objective In the present study, we evaluated the diagnostic yield of DBE and its impact on the final diagnosis, treatment, and clinical outcome of patients with obscure GI bleeding (OGIB). Design and Setting A hospital-based cross-sectional, follow-up study. Patients We studied 108 consecutive patients (66 men and 42 women) referred to our hospital from July 2003 to February 2007 for the evaluation of OGIB: 13 patients with overt-ongoing bleeding, 76 with overt-previous bleeding, and 19 with occult OGIB. Main Outcome Measurements Diagnostic yield, a final diagnosis, treatment, and clinical outcome were all analyzed in each group. Results DBE diagnostic rates for patients with overt-ongoing, overt-previous, and occult bleeding were 100.0%, 48.4% and 42.1%, respectively. The difference in diagnostic yields between the overt-ongoing group and the 2 other groups was statistically significant ( P < .005). The most common sources of bleeding were ulcers and tumor lesions. Small-intestinal lesions were identified in 52 of 108 patients; of which 36 patients (69.2%) were biopsied and 49 patients (94.2%) received treatment. Eight patients (7.4%) had recurrent bleeding during the mean follow-up period of 28.5 months. Sensitivity, specificity, and positive and negative predictive values of DBE in the diagnoses of small-intestinal lesions in patients with OGIB were 92.7%, 96.4%, 98.1%, and 87.1%, respectively. No serious complications were encountered. Conclusions DBE was proven to be a very useful diagnostic tool and had a therapeutic impact in the majority of patients with OGIB. The best candidates for the procedure were patients with overt-ongoing bleeding.