A study has found that, rather than routinely performing computed tomography (CT) and CT angiography (CTA) on all patients suspected of having acute gastrointestinal bleeding, selectively applying these tests to patients with a higher likelihood of bleeding can improve diagnostic efficiency.

The research team led by Professors Hyun Hye‑kyung and Heo Cheol‑woong of the Division of Gastroenterology at Yonsei University Yongin Severance Hospital, Professor Jung Da‑hyun of the Division of Gastroenterology at Severance Hospital, and Professor Son Nak‑hoon of the Department of Statistics at Keimyung University analyzed 12 years of accumulated clinical data to evaluate the real‑world diagnostic performance of CTA and to propose criteria for selective testing. The study results were published in the journal of the American Society for Gastrointestinal Endoscopy.

Acute gastrointestinal bleeding is a common condition encountered in emergency departments, and the speed and accuracy with which the bleeding site is identified affect treatment strategy and patient prognosis. CTA is increasingly used because it is non‑invasive, quick to perform, and can identify the site of bleeding. Some clinical guidelines recommend it as a first‑line test.

However, many existing recommendations are based on small‑scale studies, and uncertainty has remained in actual clinical practice about how capable CTA truly is as a diagnostic tool and which patients should be prioritized for the test. The burden of radiation exposure and contrast agent use must also be taken into account.

The research team analyzed data from 1,770 patients who met the study criteria, selected from among 5,525 adult patients who underwent CTA as a first‑line test in the emergency departments of Yongin Severance Hospital and Severance Hospital between October 2011 and December 2023 due to suspected acute gastrointestinal bleeding.

The sensitivity of CTA in detecting actual bleeding was 44.9%. The positive predictive value, which is the probability that bleeding is actually present when CTA is interpreted as positive, was only 52.4%, indicating limitations when using CTA alone as a first‑line test.

By contrast, the specificity, which reflects how accurately the test identifies patients without bleeding, was 88.3%, and the negative predictive value, which is the probability that there is truly no bleeding when CTA is negative, was 84.9%. This indicates that CTA is relatively useful for ruling out the possibility of active bleeding.

The research team also identified seven clinical factors that increase the likelihood of a positive CTA finding: male sex, use of antithrombotic agents, anemia, elevated blood urea nitrogen (BUN), hematemesis, hematochezia, and hemodynamic instability, each of which was found to be an independent predictor.

Based on these factors, the team developed a tool to predict the probability of a positive CTA result for individual patients. When CTA was selectively performed in patients at higher risk of bleeding according to the risk criteria set in the study, diagnostic sensitivity increased to as high as 69.5%.

Professor Hyun stated, “This study shows that, rather than performing CTA uniformly in all patients suspected of gastrointestinal bleeding, it is more rational to select only those patients for whom it is truly necessary based on clinical risk,” adding, “It is expected to help improve diagnostic efficiency while reducing unnecessary radiation and contrast exposure and easing the cost burden.” The study evaluated the diagnostic performance of CTA using not only endoscopy but also various confirmatory methods such as angiography, surgery, and red blood cell scans. The research team explained that similar results were confirmed in sensitivity analyses applying multiple conditions.

Popular News