8%) were done in the <3-day cohort and only 7 (21.2%) in the >3-day cohort. Additionally, of the 22 patients who had an angioectasia without active bleeding, 14 examinations (63.6%) were done in the <3-day cohort and only 8 (36.3%) in the >3-day cohort. Successful therapeutic intervention was performed in 18.9% of patients (17 of 90) in the

<3-day group: 12 therapeutic deep enteroscopies for coagulation of angioectasia, 2 therapeutic EGDs with coagulation of an angioectasia (n = 1) and clipping of a Dieulfoy lesion (n = 1), 2 therapeutic colonoscopies with ZD1839 chemical structure coagulation of an angioectasia (n = 1) and clipping of a Dieulfoy lesion (n = 1), and 1 surgical resection for Meckel’s diverticulum. This is in contrast to only 7.4% of patients

(4 of 54) in the >3-day cohort (P = .046) ( Fig. 5), which entailed 3 therapeutic deep enteroscopies for coagulation of angioectasia and 1 therapeutic colonoscopy with hemostasis of a solitary cecal ulcer. Blood transfusion requirement for the two inpatient cohorts was calculated to see whether the higher yield of VCE in the <3-day cohort was confounded by an increased severity of GI bleeding in this cohort. We found the blood transfusion requirements between the two cohorts to be very similar, with a mean number of 4.48 ± 0.96 units packed red blood cells transfused in the <3-day cohort versus 4.43 ± 1.12 units transfused in the >3-day cohort. Two patients in the <3-day cohort were excluded from this analysis because data were not available, and 3 patients in the >3-day cohort were excluded because they required >45 units packed red blood cells because of other comorbidities: Alectinib solubility dmso 1 because of bleeding while anticoagulated for mechanical valve, RVX-208 1 to ongoing bleeding because of ischemic ileal ulcerations, and 1 to systemic lupus erythematosus with purpura fulminans. Comorbid conditions between the two inpatient cohorts were very similar, as outlined in Table 3. No significant difference

were found in anticoagulant, anti-inflammatory, or antiplatelet use (nonsteroidal anti-inflammatory drugs, clopidrogel, and warfarin). There was also a similar distribution of those with coronary disease, diabetes, renal disease, and cirrhosis. Findings of VCE for outpatients are also presented in Table 2. Detection of active bleeding and/or angioectasia for the outpatient cohort was 25.8% (30 of 116). Two capsules showed evidence of both an active bleed and angioectasia. Successful therapeutic intervention was performed in 10.3% of patients (12 of 116): 10 therapeutic deep enteroscopies and 2 therapeutic EGDs. Two capsules were retained in the ulcerated stricture of the small bowel, one of which required operative intervention. It was notable that the diagnostic yield for detecting an active bleed for the >3-day cohort (13%) and the outpatient cohort (12.9%) was statistically similar (P = .8) ( Fig. 2).