Reproducibility of the 2020 Society for Vascular Surgery/ Society of Thoracic Surgeons Reporting Standards for Uncomplicated Type B Aortic Dissection

Abstract

Purpose: To assess the inter-and intrareader agreement of the 2020 Society for Vascular Surgery/Society of Thoracic Surgeons (SVS/STS) reporting stadards in patients with acute uncomplicated type B aortic dissection (uTBAD). Materials and Methods: CT angiograms, obtained between January 2001 and December 2013, from 37 patients (mean age ± SD, 48.5 years ± 15.6; 24 male) with uTBAD were retrospectively reviewed by 14 cardiovascular radiologists and surgeons. The entry tear and proximal and distal extent of each dissection were allocated to aortic zones according to the 2020 SVS/STS reporting standards. Inter-and intrareader agreement was assessed using Cohen κ. Causes of discrepant classifications were analyzed. Results: Interreader agreement among all 14 readers (eight radiologists, four cardiothoracic surgeons, and two vascular surgeons) were fair for the entry tear (к = 0.33, 0.40, 0.22, and 0.40), poor to moderate for the proximal extent (к = 0.30, 0.37, 0.20, and 0.41), and moderate to substantial for the distal dissection extent (к = 0.65, 0.80, 0.41, and 0.77). Interreader agreement between individuals was poor to substantial for the entry tear (к = 0.04–0.71; 78% discrepancy) and proximal extent (к = 0.00–0.62; 68% discrepancy), and fair to excellent for the distal extent (к = 0.33–0.92; 35% discrepancy). The causes for discrepancies were unclear definitions of aortic zones, anatomic complexity of lesions, and poor image quality. Intrareader agreement was moderate for the entry tear (к = 0.41–0.57), fair to moderate for the proximal extent (к = 0.37–0.58), and substantial to excellent for the distal extent (к = 0.66–0.92). Conclusion: geons. The 2020 SVS/STS reporting standards for acute uTBAD were poorly reproducible among experienced cardiovascular radiologists and surgeons.