Impact of capsule endoscopy on outcome in mid-intestinal bleeding: a multicentre cohort study in 285 patients.

Background: Capsule endoscopy (CE) sensitively detects the bleeding source in the small bowel. However, the influence of CE on long-term outcome is not well established.

Methods: In five tertiary hospitals, all CE investigations were retrospectively identified dating back to 3 years. Patients with intestinal bleeding and negative bidirectional endoscopy were included, and relapse of bleeding was recorded.

Results: A bleeding source was detected in 219 of 285 patients (76.8%); CE provided the diagnosis in 175 of 219 (79.9%) and other, repeated investigations in 44 cases (20.1%). Follow-up (mean+/-SD=20.7+/-9.4 months) in 240 patients identified rebleeding in 65 (27.1%), and readmission to a hospital in 42 (17.5%). Hospital readmission was most frequent in patients with angiectasias (31.3%, relative risk (RR)=5.0; 95% confidence interval (CI)=2.4-10.4). Other risk factors included patients being older than 60 years of age (RR=3.8; 95% CI=1.5-9.5), and anticoagulant medication (RR=3.0; 95% CI=1.5-6.0). Therapeutic measures had a mean recurrence rate of 3.7% in surgical candidates (Meckel's diverticulum, tumor), 40% in endoscopically treated and 16% in medically treated patients. In case all the detected angiectasias had been cauterized, the relapse rate was low (11.8%), but in incompletely treated patients, it was high (85.7%). Bleeding relapse was never lethal.

Conclusion: CE guides therapeutic measures and predicts the risk of recurrent bleeding in small intestinal bleeding. High risk of rebleeding in angiectasias is significantly reduced by the cauterization of all demonstrable lesions.