TTE

TTE was performed 5 months after TAVI, from June to October 2010, using the X5-1 PureWave microbeamforming xMATRIX probe (Philips Medical Systems). This probe presents all the facilities of the 2D S5-1 probe with 3D capabilities as full volume, Live 3D, and Live 3D color acquisitions, from one to four beats. The full volume combines a series of four subvolumes acquired with electrocardiographic gating to create the final reconstructed image. This acquisition mode is essential for the measurement of volumes and left ventricular ejection fraction by 3D volumetric assessment. Live 3D and Live 3D color provide real-time 3D volumetric motion, without electrocardiographically gated reconstruction. However, these acquisition modes can be optimized for a higher frame rate by performing near-real-time images, with electrocardiographically gated reconstruction, with up to four heartbeats. In this study, we used Live 3D without electrocardiographically gated reconstruction and Live 3D color images acquired with four heartbeats. We used iCrop, a new flexible tool that allows direct analysis on Live 3D and Live 3D color images, for interpretation while performing the exams.

First, conventional 2D TTE was performed, in which the prosthesis area was calculated using the continuity equation. Apical five-chamber, three-chamber, and parasternal long-axis and short-axis color views were recorded, and AR jets were described following international standard recommendations. We systematically assessed the vena contracta width in the parasternal long-axis or apical long-axis view using a Nyquist limit of 50 to 60 cm/sec ( Figure 1 ). AR volume was calculated as the difference between 3D TTE–derived total left ventricular stroke volume and stroke volume through the right ventricular outflow tract. The right ventricular outflow tract diameter and velocity-time integral at the site were measured in the short-axis parasternal view, and the right ventricular stroke volume was calculated by multiplication of the cross-sectional right ventricular outflow tract area by the velocity-time integral. AR was graded using an integrative approach considering the recommended semiquantitative Doppler parameters and the AR volume as mild (AR volume < 30 mL) or moderate (30 mL ≤ AR volume ≤ 59 mL), assessed as previously described.

Paravalvular AR assessed by 2D echocardiography in the parasternal long-axis view. This image illustrates a vena contracta width of 2 mm. It was acquired from the same patient shown in Figure 6 . Ao , Aorta; LA , left atrium; LV , left ventricle.

Full-volume and Live 3D color images were acquired from one to four heartbeats in the apical four-chamber view. At the time of acquisition, we used the 3D TTE and two orthogonal 2D transthoracic echocardiographic views for guidance, as illustrated in Figures 2 A and 3 . If necessary, we adjusted the lateral and the elevation width to be certain that all of the structures we intended to analyze were included in the volumetric images acquired. Full-volume acquisitions, obtained by combining four electrocardiographically gated subvolumes, were used to measure left ventricular volumes and ejection fraction by direct volumetric analysis ( Figure 2 ). We assessed the aortic prosthesis using Live 3D views ( Figure 4 ). Using Live 3D color images, with the narrowest sector possible for frame rate optimization, the aortic prosthesis was isolated with the iCrop tool. The entire circumference of the prosthesis, and the paravalvular region, were viewed from the aortic root and also from the ventricular aspect, as illustrated in Figure 3 .

(A) Three-dimensional transthoracic echocardiographic four-heartbeat full-volume acquisition. (B) Left ventricular ejection fraction and stroke volume determination by direct volumetric analysis.

Paravalvular AR assessed with Live 3D color, using the iCrop tool. Two orthogonal 2D views and the 3D reconstruction are obtained simultaneously. (A) Paravalvular AR view from the left ventricle (LV) ( arrows ). (B) Paravalvular AR view from the aortic root ( arrows ). Depending on the location of the cut plane in relation to the aortic prosthesis, the area of the paravalvular AR jet varies as viewed in the short-axis orientation.

Live 3D transthoracic echocardiographic multiplanar reconstruction for aortic prosthesis (AP) ring assessment. LA , Left atrium; LV , left ventricle.

The iCrop tool constructs an image from two orthogonal square views, whose size can be adjusted for the proposed region of analysis. The anatomic findings can be viewed from each of the sides of the square ( Figure 3 ); consequently, in the same acquisition, the prosthetic valve can be viewed from the left ventricular outflow tract and also from the aortic root. Depending on the cutting plane, the area of the regurgitant jet varies, because the view from the aortic root shows the isovelocity surface area proximal to the regurgitant orifice, while the view from the left ventricular outflow tract shows the beginning of the color jet as blood entrains and travels into the left ventricle. Consequently, iCrop allowed the identification and localization of the AR jet, but it was not used to measure AR jet size, which was performed using the multiplanar reconstruction tools. To describe AR localization systematically, we designed a reference diagram in which the parasternal short-axis view was chosen as a reference and the paravalvular region divided into 12 sections according to the hours of a clock face ( Figure 5 ).

Representation of paravalvular AR jets location according to the face of a clock. Conventional 2D transthoracic echocardiographic views are represented. (A) Parasternal short-axis view. (B) Parasternal long-axis view. (C) Apical five-chamber view. (D) Apical three-chamber view.

For AR vena contracta planimetry by 3D TTE, Live 3D color data sets were analyzed using the multiplanar reconstruction tool. First, the best frame for AR jet visualization in two orthogonal long-axis views was selected. Then we cropped the data set along the axis of the AR jet, from the aortic side to the level of the vena contracta, in a plane that was exactly perpendicular to the long axis of the AR jet. Subsequently, we performed planimetry ( Figure 6 ) using the methodology previously validated by Fang et al. In patients with more than one independent leak, the vena contracta was taken as the sum of all individual vena contracta areas.

AR Live 3D color analysis using multiplanar reconstruction tools. After selecting the best frame for AR jet visualization in two orthogonal long axis views (A,B) , the data set was cropped through the perpendicular plane of the AR jet long axis, from the aortic side to the level of the vena contracta (C) . (A) Coronal axis view. (B) Sagittal axis view. (C) Axial axis view, with planimetry of the vena contracta. (D) Multiplanar simultaneous axis view. Ao , Aorta; LV , left ventricle.

This protocol was approved by the local ethics committee, and all patients gave written informed consent for participation.

Reproducibility

Two experienced operators blinded to 2D echocardiographic results analyzed separately 3D color data sets, worked through the multiplanar reconstruction tools, and performed the vena contracta planimetry in patients with AR.

Statistical Analysis

Categorical variables are expressed as percentages and continuous variables as mean ± SD unless otherwise specified. Continuous variables were compared between groups using unpaired t tests (for normally distributed variables) or the Mann-Whitney U test (for variables not normally distributed). Kendall’s τ (nonparametric test) was calculated for the assessment of correlations between vena contracta measurements by 3D TTE and 2D TTE and AR volume. The diagnostic efficiency of vena contracta width and planimetry was assessed using receiver operating characteristic curve analysis. All reported P values are two tailed, and P values < .05 were considered to indicate statistical significance. Analyses were performed using SPSS version 16.0 (SPSS, Inc., Chicago, IL).