Patient and endoscopic characteristics and clinical outcomes in subjects with non-variceal GI bleeding referred for transarterial embolization: a single-center experience

• Published in: Abdominal imaging Vol. 47; no. 11; pp. 3883 - 3891
• Main Authors: Ephraim Joseph, Kripalini, Devane, Aron M., Abrams, Gary A.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.11.2022; Springer Nature B.V
• Subjects: Bleeding; Blood cells; Clinical outcomes; Embolization; Embolization, Therapeutic - methods; Endoscopy; Extravasation; Gastroenterology; Gastrointestinal Hemorrhage - diagnostic imaging; Gastrointestinal Hemorrhage - therapy; Hepatology; Humans; Hypotension; Imaging; Interventional Radiology; Medicine; Medicine & Public Health; Mortality; Patients; Radiology; Retrospective Studies; Stenosis; Treatment Outcome; Vascular Surgical Procedures - adverse effects; Therapeutic endoscopy; Transesophageal angiography; Embolization; Interventional radiology; Non-variceal upper gastrointestinal bleeding
• ISSN: 2366-0058; 2366-004X; 2366-0058
• DOI: 10.1007/s00261-022-03650-x

Purpose Management of massive non-variceal upper gastrointestinal bleeding (NV-UGIB) can be challenging. Transarterial Embolization (TAE) is often the first therapeutic approach when endoscopic therapy fails before surgery. The purpose of this study is to analyze the technical success, and outcome for our patients with an NV-UGIB referred for TAE. Method This retrospective analysis included 74 consecutive patients with an NV-UGIB in whom TAE was performed after endoscopic treatment between February 2016 to May 2019 at Prisma Health-Upstate Greenville Memorial Hospital. Results TAE was 98.7% technically successful, with a failure due to severe celiac stenosis, and 85.1% clinically successful. Most TAEs were performed empirically due to lack of extravasation yet were clinically as effective as targeted TAE. We noted a 30-day rebleeding rate and mortality rate of 14.8% and 13.5%, respectively. No complications were reported during the angiographic procedure. Subjects with coagulopathy had more rebleeding (45.5% vs. 17.5%, p  = 0.040), and mortality (30% vs 7.4%, p  = 0.012). Mortality was also associated with the number of transfused packed blood cells (13.6 ± 8.4 vs. 6.1 ± 5.4, p  = 0.020) units and hypotension on admission (27.8% vs. 8.9%, p  = 0.043). Interestingly, subjects that underwent left gastric artery (LGA) compared to non-LGA embolization had a higher rebleeding rate of (37.5% vs. 8.6%, p  = 0.004) and a greater mortality rate of (37.5% vs. 6.9%, p  = 0.002). Conclusion TAE is clinically effective in the presence or absence of contrast extravasation to treat uncontrolled or high-risk NV-UGIB. Less effective clinical outcomes regarding TAE targeting the LGA warrant further investigation. Graphical abstract