Dual-source computed tomography angiography in aortic stenosis: comparison with transthoracic echocardiography

Abstract

The diagnostic performance of dual-source computed tomography (DSCT) angiography was evaluated in this study and compared to transthoracic echocardiography (TTE), which is commonly used as the reference standard, in the assessment of aortic valve morphology, measurement of aortic valve area (AVA) and classification of aortic valve stenosis. Forty subjects were recruited, with 20 diagnosed with aortic stenosis (AS) and the remaining 20 identified with normal aortic valve. Morphological findings of the aortic valve were compared between these two modalities. Maximum AVA was measured planimetrically in mid-systole with DSCT angiography (AVACT) and compared with TTE-derived AVA (AVATTE). There was more than 90% concordance of valve morphology between DSCT angiography and TTE (sensitivity 91.7%, specificity 100%). There was excellent correlation between AVACT and AVATTE (r = 0.98, < 0.001). Bland Altman analysis showed good inter-modality agreement (95% confidence interval [CI]: -0.26 – 0.51 cm2) with slight overestimation of AVA (+ 0.13 cm2) by DSCT. Very good correlation was observed between AVACT and the mean as well as peak transvalvular pressure gradients (r² = 0.82 and 0.84 respectively). It was concluded that DSCT angiography can be used as a reliable, non-invasive and accurate imaging modality to evaluate the aortic valve morphology.