Date of Award

January 2011

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)



First Advisor

Hamid Mojibian

Subject Area(s)

Medical imaging and radiology, Medicine


The main diagnostic imaging modalities currently used to image the aortic root and ascending aorta include transthoracic echocardiography (TTE) and either ECG-gated or non-ECG-gated computed tomography (CT), each with its respective advantages and disadvantages. This study aimed to examine the reproducibility and agreement of inter-reader measurements of the aortic root made on all three imaging modalities. The results of this study revealed that inter-cardiologist measurements of the aortic root performed on echocardiography lack reproducibility (mean difference of 2.9 ± 5.76 mm). Aortic root measurements made on ECG-gated CT studies more closely agree with measurements made on echocardiography than do those made on non-ECG-gated CT studies (mean differences of 0.4 ± 3.93 mm versus 0.9 ± 5.92 mm in the axial dimension and -0.2 ± 3.82 mm versus 1.1 ± 5.46 mm in the orthogonal dimension). When a single radiologist reviews a set of both ECG-gated and non-ECG-gated CTs, the mean difference of axial aortic root measurements is small (-0.6 ± 1.99 mm). However, when aortic root measurements are made by two different radiologists, there is a much greater increase in mean difference for non-ECG-gated CT studies (-2.4 ± 6.32 mm) as compared with ECG-gated CT studies (0.3 ± 3.06 mm), suggesting that ECG-gating, which produces higher resolution images, buffers the amount of bias and variation that are introduced by inter-radiologist differences in measurement methods. Finally, review of radiologists' cardiac CT reports revealed poor standards and high non-uniformity in providing referring clinicians with relevant aortic measurements that have an important impact on patient care. A number of concluding recommendations are made and discussed to increase the value added by an institution's cardiac radiology service.