Inter-reader agreement of CT features of acute mesenteric ischemia. Issue 105 (August 2018)

• Record Type: Journal Article
• Authors: Copin, Pauline; Ronot, Maxime; Nuzzo, Alexandre; Maggiori, Léon; Bouhnik, Yoram; Corcos, Olivier; Vilgrain, Valérie
• Is Part Of: European journal of radiology. Issue 105(2018)
• Page Start: 87
• Page End: 95
• Publication Date: 2018-08
• Subjects: AMI acute mesenteric ischemia -- CT computed tomography -- DABE decreased or absent bowel wall enhancement; Mesenteric ischemia -- Mesenteric vascular occlusion -- Diagnostic imaging -- Emergencies -- CTA -- Bowel enhancement; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2018.05.027
• Languages: English
• ISSNs: 0720-048X

Abstract

Highlights: Inter-reader agreement for any arterial abnormality was substantial. Inter-reader agreement for arterial abnormality was slightly improved by arterial phase images. Inter-reader agreement for presence of unenhanced bowel wall hyperattenuation was almost perfect. Inter-reader agreement for decreased/absent bowel wall enhancement was moderate. Inter-reader agreement for decreased/absent bowel wall enhancement was improved by CT protocol. Abstract: Purpose: To evaluate the inter-reader agreement of the CT features of acute mesenteric ischemia (AMI). Methods: Between 2006 and 2014, 109 patients (57 men, 52%, mean age 50 years old [17–83]) admitted to our institution with a diagnosis of AMI were included. CT scans (42% were initially performed in our institution) were reviewed by two abdominal radiologists. Inter-observer agreement of the imaging features of vascular insufficiency and bowel ischemia was assessed by the percentage of agreement and the kappa value. Results: The final population included, Inter-observer agreement varied according to the different features (κ = 0.25-0.98). Inter-observer agreement for decreased/absent bowel wall enhancement was moderate (κ = 0.52), but was almost perfect (κ = 0.82) in the 47 patients (43%) with both unenhanced and arterial-phase images without positive oral contrast agent and excellent CT images quality. Conclusion: Inter-reader agreement was moderate to substantial for most CT features of AMI. Multiphasic CT scan protocol, including unenhanced, arterial phase and venous phase images, without positive oral contrast agent, and excellent CT images quality improve inter-observer agreement of imaging features of AMI, especially for decreased/absent bowel wall enhancement, and should be performed in patients with suspected AMI.