OBJECTIVES/BACKGROUND:
Congenital aortic stenosis (AS) is the most common obstructive left heart lesion in the young adult population and often complicated by aortic dilatation. Our objective was to evaluate accuracy of aortic imaging with transthoracic echocardiography (TTE) compared with cardiac magnetic resonance (CMR).
METHODS:
Aortic diameters were measured at 4 levels by CMR and TTE. Agreement and concordance were assessed by Pearson's correlation and Bland-Altman analysis.
RESULTS:
Fifty-nine patients (age 33 ± 8 years; 66% male) with congenital AS and a bicuspid aortic valve (BAV) were included. Aortic diameters were generally smaller with TTE than with CMR. The best correlation was found at the level of the sinotubular junction (R(2) = 0.78) with a bias of 1.46 mm (limits of agreement: -5.47 to +8.39 mm). In patients with an aortic aneurysm >40 mm (n = 29) the correlation and agreement between TTE and CMR were found to be less good when compared with patients with normal aortic diameters, especially at the level of the proximal ascending aorta. The correlation and agreement between both imaging modalities were better in patients with type 1 BAV compared with type 2 BAV. Intra- and interobserver variability was smaller with CMR (1.8-5.9%) compared with TTE (6.9-15.0%).
CONCLUSIONS:
CMR was found to be superior to TTE for imaging of the aorta in patients with congenital AS, especially at the level of the proximal ascending aorta when an aortic aneurysm is present. Therefore, ideally CMR should be performed at least once to ensure an ascending aortic aneurysm is not missed.
