Diagnostic Yield of Video Capsule Endoscopy for Small Bowel Bleeding: Eight Consecutive Years of Experience at the VA Caribbean Healthcare System.

OBJECTIVE To evaluate the diagnostic yield of video capsule endoscopy (VCE) in patients with small bowel gastrointestinal bleeding and examine the impact of this diagnostic technology on the clinical management of this complaint. METHODS This was a retrospective study in which all patients who underwent VCE (May 7, 2003 - December 31, 2011) were included. Records were reviewed for the type of bleeding (overt vs. occult; when present), demographic data, lab results, and capsule endoscopy findings. Information regarding medical treatment (i.e., endoscopic intervention, surgical therapy, or both) was also recorded. RESULTS A total of 229 subjects were included in the study. Most were men; the mean age of all the subjects was 69.8 years. Of the 229 VCEs, 154 (67.3%) were done because of occult bleeding and 75 (32%) because of overt bleeding. VCEs were normal in 34 (14.9%) cases and non-diagnostic in 15 (6.6%). Angiodysplasia, erosions, and ulcers were the most common findings (48.5%, 24.5%, and 10.92% respectively). Active bleeding was reported in 7 cases (3%). Nearly 20% of the 229 cases required either endoscopic or surgical intervention. CONCLUSION In our study, VCE achieved a diagnostic yield of 78.6%. In 1 of every 5 subjects, video capsule endoscopy led to the identification of small bowel lesions that required either endoscopic or surgical resection, rather than conservative treatment with iron replacement. VCE proved to be a very useful investigative tool, not only for establishing the source of bleeding but also, most importantly, for directing the appropriate therapy for lesions that would otherwise have been missed by conventional studies.