Capsule endoscopy is used to investigate the small bowel in patients with GI bleeding of obscure etiology. Capsule endoscopy was compared prospectively with push enteroscopy in 20 patients with GI hemorrhage.

Twenty patients (8 men, 12 women; mean age 65.5 years, range 38–80 years) were enrolled in the study. All had undergone non-diagnostic EGD, colonoscopy, and barium contrast radiography of the small bowel. All patients underwent capsule endoscopy followed by push enteroscopy. The physician performing the enteroscopy (senior endoscopist) interpreted the capsule endoscopy in an unblinded manner, while a second blinded reviewer (endoscopy fellow) interpreted the capsule endoscopy to establish interinterpreter reliability.

There was complete agreement between the blinded and the unblinded physicians in 18 of 20 cases; minor disparities were noted in the remaining two cases. In the small bowel, capsule endoscopy identified positive findings in 14 (70%) patients, whereas, push enteroscopy identified positive findings in 5 (25%) patients. Despite these results, the findings were definitive in only 6 of the 20 patients by using capsule endoscopy, and in two of 20 patients with push enteroscopy.

When strict standards of interpretation were used, capsule endoscopy resulted in more positive findings than push enteroscopy, but the number of definitive findings for both imaging methods was low. There was a high degree of reliability between a novice and an experienced endoscopist with respect to the interpretation of capsule endoscopy.