Factors associated with positive findings from capsule endoscopy in patients with obscure gastrointestinal bleeding

Abstract

Background & aims: Capsule endoscopy (CE) is used most frequently to identify causes of obscure gastrointestinal bleeding (OGIB). Identifying factors associated with the detection of lesions by CE could improve resource utilization and thereby improve patient selection for CE examination. We sought to identify clinical factors associated with positive findings from CE in patients with OGIB.

Methods: We analyzed data from 698 CE procedures performed between December 2001 and April 2011 at St Paul's Hospital, Vancouver, Canada (50.3% of patients were female; mean age, 63.4 years). A positive finding was defined as a lesion that was believed to be the source of the bleeding (ulceration, mass lesion, vascular lesion, or visible blood). Univariate and multivariate logistic regression analyses were used to correlate demographic and clinical parameters with positive findings.

Results: A lesion believed to be the cause of bleeding was identified in 42% of cases. In univariate analysis, the number of esophagogastroduodenoscopies (EGDs), the presence of connective tissue disease or diabetes with end-organ damage, Charlson comorbidity index scores, and increasing transfusion requirements were significantly associated with identification of causative pathology from CE (all P < .027). In multivariate analysis, increasing number of EGDs (odds ratio [OR], 1.17; 95% confidence interval [CI], 1.00-1.37), increasing transfusion requirements (3-9 units: OR, 1.70; 95% CI, 1.08-2.66, and ≥10 units: OR, 2.72; 95% CI, 1.69-4.37), and connective tissue disease (OR, 2.24; 95% CI, 1.14-4.41) were all significantly associated with identification of positive findings by using CE (all P < .045).

Conclusions: Patients with a higher number of precapsule EGDs or transfusions, or connective tissue disease, are superior candidates for analysis of OGIB by CE.