Abstract
BACKGROUND: The etiology of gastrointestinal bleeding remains unclear in a sizable minority of patients despite extensive evaluation. In these patients, obscure gastrointestinal bleeding (OGIB) poses a significant diagnostic and therapeutic challenge, resulting in higher patient morbidity and mortality, as well as increased utilization of health care resources. AIMS: In this study, we evaluate the diagnostic yield and clinical outcomes of patients with OGIB who have undergone balloon-assisted endoscopy (BAE). METHODS: We performed a large, retrospective cohort study of 311 consecutive patients who underwent BAE at our institution. Of these, 116 cases were identified to have had OGIB and were appropriate for inclusion in our analysis. Follow-up data were available for 110 patients, with median time to follow-up of 6.1 months (range 0.3 – 37.7). The pattern of bleeding, transfusion requirements, and frequency of OGIB control were captured. RESULTS: BAE identified bleeding sources in 79 patient cases (68.1%), of which, 66 (83.5%) had culprit small bowel pathology. Of these, vascular lesions were the most common finding, occurring in 39 patients (59.0%), followed by neoplasia (13.6%, 9 patients), and ulcerations/erosions (12.1%, 8 patients). Diagnostic yield was highest in those patients with active overt bleeding, compared to those with inactive or occult bleeding (80.0%, 20/25 patients vs 35.2%, 32/91 patients). Overall, OGIB was controlled in 58/110 patients (52.7%). Culprit vascular lesions demonstrated poorer control rates than ulcers/erosions, tumors/polyps, or unidentified pathology (38%, 14/37 patients vs 55.8%, 29/52 patients). Among those patients with uncontrolled OGIB, median 6-month red-cell transfusion requirements were lower post-procedure compared to those pre-procedure (0.69 units/month, range 0.00 – 50.7; vs 1.00 units/month, range 0.00 – 10.3). CONCLUSIONS: BAE is a useful diagnostic and therapeutic strategy in the management of patients with OGIB. Luminal evaluation of the small bowel with BAE should be undertaken as soon as OGIB is suspected, and preferably in proximity to active bleeding. Vascular lesions are particularly recalcitrant to endoscopic therapy and therefore warrant close monitoring. Controlled OGIB is defined as the clinically stable condition meeting the following criteria: (1) the absence of any overt re-bleeding, (2) no further requirement of red cell transfusions, and (3) the resolution or absence of persistent anemia. FUNDING AGENCIES: None