A37 EFFICACY OF CTA IN DIAGNOSING NON-TRAUMATIC NON-VARICEAL GASTROINTESTINAL BLEEDING PRIOR TO TRANSARTERIAL EMBOLIZATION AFTER ENDOSCOPIC FAILURE IN MANAGING ACUTE GASTROINTESTINAL BLEEDING

• Published in: Journal of the Canadian Association of Gastroenterology 2018-03, Vol.1 (suppl_1), p.66-66
• Main Authors: Wadhwani, A, Brunet, S, Nguyen, J, Herget, E, Beck, P
• Format: Article
• Language: English
• Subjects: Poster Presentations
• ISSN: 2515-2084; 2515-2092
• DOI: 10.1093/jcag/gwy008.038

Abstract Background Non-variceal gastrointestinal bleeding (NVGIB) is associated with a high mortality and morbidity. 10–30% of these patients tend to fail endoscopy and receive transarterial embolization (TAE) as an alternative. Studies suggest that performing pre-angiography computed tomography angiography (CTA) increases the positive yield of visceral angiography. Aims Our objective was to determine (1) the accuracy of CTA in diagnosing NVGIB following failed endoscopy and (2) the impact of CTA pre-TAE on the angiographic technique. Methods Data was collected from 83 consecutive patients who presented to the emergency department with acute NVGIB and received TAE after endoscopy failed to manage their NVGIB. Of these 83 patients, 40 underwent pre-angiography CTA. These CTA examinations were retrospectively reviewed by 2 radiology residents and 2 staff radiologists. These findings were compared to angiography, or/and surgery. Inter-reader reliability was evaluated with kappa coefficient (κ). Results Sensitivity, specificity, PPV, NPV, and accuracy of CTA in diagnosing NVGIB was 89%, 100%, 100%, 86%, and 93%, respectively. CTA was able to accurately diagnose the cause and source of NVGIB in 85% of the patients respectively. The inter-reader reliability coefficient for identifying the cause and source of NVGIB was κ=0.72 and κ=0.66 respectively. In 20 cases, in whom CTA localized NVGIB, no diagnostic catheter angiogram was required. In 6/20 cases, pre-TAE CTA enabled the identification of the bleeding site, which would not have been visualized on a routine diagnostic angiogram. When comparing patients that received CTA prior to their therapeutic embolization for NVGIB to the patients that did not receive a pre-embolization CTA, there was an overall reduction of 20 minutes of procedural time. Conclusions CTA is an accurate diagnostic modality in detecting NVGIB. Performing abdomen and pelvis CTA before TAE improves the localization of gastrointestinal bleeding and facilitates embolization by reducing the overall procedural time. Impact of pre-angiography CTA on reducing the overall number of imaging studies, amount of contrast administered, and overall mortality and morbidity needs to be further investigated. Funding Agencies None