Pitfalls of anatomical aortic valve area measurements using two-dimensional transoesophageal echocardiography and the potential of three-dimensional transoesophageal echocardiography

The aims of this study were to (i) investigate aortic annulus dynamics using two-dimensional (2D) speckle tracking echocardiography, (ii) determine optimal 2D short-axis view for the calculation of planimetric aortic valve area (AVA), and (iii) compare 2D planimetric AVA extracted from volumetric three-dimensional data sets using real-time 3DTEE (three-dimensional transoesophageal echocardiography) with standard 2DTEE planimetry. We studied 60 patients with aortic stenosis (AS) and 10 control subjects. AVA was calculated by standard 2DTEE planimetry method, volumetric 3DTEE method, and continuity equation (CE) from transthoracic echocardiography. In addition, aortic annular motion was studied using 2D speckle tracking. Aortic annulus moves cranially during early systole and subsequently moves caudally during the remainder of systole and isovolumic relaxation. Annulus again moved in the cranial direction during diastole in both groups. Although AVA correlated well between 2DTEE and 3DTEE methods (r = 0.95), 2DTEE showed a significantly larger AVA compared with 3DTEE method (1.26 +/- 0.39 vs. 1.10 +/- 0.39 cm(2), P < 0.001). In patients in whom aortic cusps were visible in 2DTEE short-axis images during systole only (n = 45), AVA using 2DTEE was still larger than that measured with 3DTEE. However, the bias in AVA was significantly lower compared with the remaining 15 patients (-0.13 +/- 0.11 vs. -0.26 +/- 0.12 cm(2), P < 0.005). Although both methods showed moderate correlation with AVA by CE (r = 0.78, 0.75), mean differences were significantly smaller by 3DTEE than 2DTEE (-0.01 +/- 0.25 vs. -0.17 +/- 0.27 cm(2), P < 0.001). Aortic annular motion affects the calculation of AVA using 2DTEE. Three-dimensional transoesophageal echocardiography has a potential for more accurate determination of anatomical AVA.