Standardized Protocol to Analyze Computed Tomography Imaging of Type B Aortic Dissections

Abstract

Purpose: To propose a standard measuring protocol for type B aortic dissections so as to improve comparability between studies reporting aortic dimensions. Methods: Fifteen computed tomography (CT) scans of type B aortic dissections were measured with a standard protocol by 2 independent observers using postprocessing software. The following parameters were assessed: true, false, and total lumen diameter; true and false lumen volume; and entry tear size, location, and number. Diameters were measured in a perpendicular plane at 2, 10, and 20 cm from the left subclavian artery and 5 cm from the most distal renal artery. True lumen volume was assessed from the left subclavian artery to the aortic bifurcation, while the false lumen volume was from the start to end up to the aortic bifurcation. Entry tear location was assessed in relation to the left subclavian artery. Intra- and interobserver repeatability and agreement were evaluated using the Bland-Altman method, an a priori set of acceptable differences, and Lin’s concordance correlation coefficient (LCCC). Results: Intra- and interobserver mean differences for aortic diameter and true and false lumen volumes were generally within the limits of agreement and the a priori differences; the LCCC showed excellent agreement. Entry tear location, size, and number were difficult to measure in a repeatable manner, with inconsistent correlation coefficients, especially between the 2 observers. Conclusion: This protocol showed acceptable repeatability for aortic diameter and aortic volume measurements. Assessment of entry tears proved challenging and associated with less favorable results. Additionally, investigators are urged to be more transparent regarding the measurement methodology used in studies describing aortic dimensions.