Post written by Jason Erno, MD, and Don C. Rockey, MD, from the Digestive Disease Research Core Center, Medical University of South Carolina, Charleston, South Carolina, USA.

This cohort study examined patients with acute GI bleeding who underwent CT angiography (CTA) and compared its accuracy with that of subsequent endoscopy.

CTA has gained great popularity as a quick initial diagnostic study to localize a source of GI bleeding–and hospitalists and emergency department physicians frequently perform it. However, this practice is not evidence-based, and lack of endoscopic and clinical follow-up limits much of the published literature on accuracy of CTA.

Our study found that the sensitivity of CTA for identifying a bleeding lesion in patients with upper or lower GI bleeding was only 20%. Of the 78 patients with an initial negative CTA who underwent endoscopy, 52 had a lesion identified. Of those 52, 38 had a high-risk bleeding lesion requiring therapy.

Although CTA may be useful in specific patients (often those with complicated underlying intra-abdominal processes), the data here indicate that CTA is insensitive for detection of GI lesions in a typical patient population with GI bleeding.

These data comparing CTA with endoscopy are highly novel and indicate that CTA is generally not useful as a primary diagnostic test in GI bleeding. Rather, endoscopy should usually be the initial diagnostic (and therapeutic) approach in patients with GI bleeding.

Consort diagram of the patients included in the study. CTA, CT angiography; ICD, International Classification of Diseases; UGIB, upper GI bleeding; LGIB, lower GI bleeding; RBC, red blood cell.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.