Study Population

We retrospectively reviewed 227 consecutive patients with severe AS who underwent TAVR with a balloon-expandable SAPIEN or SAPIEN XT THV (Edwards Lifesciences, Irvine, CA) with intraprocedural 3D TEE from December 2010 to July 2012 at our institution. Twenty-one patients without 3D data of the aortic valve before TAVR, two patients without two-dimensional (2D) or 3D color Doppler transesophageal echocardiographic data after TAVR, and two patients with past aortic valve replacement were excluded. Thus, 202 patients were enrolled in this study. All characteristic data were retrieved from our computerized TAVR database. All procedural records pertaining to the procedural access route, the type and size of the aortic valve prosthesis, and ballooning after prosthetic deployment were reviewed. This study was approved by the institutional review board.

TAVR Procedure

The TAVR procedure was performed by an interdisciplinary heart team of interventional cardiologists, cardiac surgeons, echocardiographers, and anesthesiologists. The prosthetic valve size was selected on the basis of both multidetector computed tomographic and 3D echocardiographic measurements of the annular dimensions.

Two-Dimensional Transthoracic Echocardiography (TTE) and Measurements

All patients underwent comprehensive 2D TTE and Doppler studies at our institution <1 month before TAVR. The iE33 ultrasound system (Philips Medical Systems, Andover, MA) equipped with an S5-1 phased-array transducer (Philips Medical Systems) was used. In accordance with American College of Cardiology and American Heart Association guidelines, severe AS in this study was categorized quantitatively on the basis of aortic valve area < 1.0 cm ² , mean aortic gradient > 40 mm Hg, or aortic jet velocity > 4.0 m/sec. Left ventricular mass was calculated using the M-mode method. The ejection fraction was calculated by the biplane method of disks. Aortic valve area was derived from the continuity equation. The aortic valve area index, stroke volume index, and left ventricular mass index were calculated by indexing to body surface area.

Intraoperative 2D and 3D TEE

TEE was performed under general anesthesia using the iE33 ultrasound system equipped with an X7-2t transesophageal echocardiographic ultrasound probe, which provides a range frequency of 2.0 to 7.0 MHz and has a 3D matrix array (Philips Medical Systems). Before the procedure, the 3D data sets of the aortic valve were recorded mainly by 3D zoom mode and, if necessary, by full-volume mode with four electrocardiographically triggered sequential volumes of the aortic valve. All volumetric images were analyzed offline using QLAB 3DQ version 9.0 (Philips Medical Systems). Paravalvular regurgitation was assessed quantitatively by the sum of the cross-sectional area of the vena contracta using 2D or 3D color Doppler transesophageal echocardiographic data at the end of the TAVR procedure ( Figure 1 ). Grading of paravalvular regurgitation was determined using the following cutoffs: trace, 0 to 4 mm ² ; mild, 5 to 9 mm ² ; moderate, 10 to 29 mm ² ; and severe, ≥30 mm ² (according to Valve Academic Research Consortium-2 criteria ). This measurement was performed by paying meticulous attention to identify an appropriate annular plane for the measurement of vena contracta area using as many movies as possible. Significant paravalvular regurgitation was defined as at least a moderate grade.

Paravalvular regurgitation was assessed quantitatively by the sum of the cross-sectional area of the vena contracta in the short-axis view.

Three-Dimensional Analysis of AVC

We assessed the severity of AVC on a visual scale ranging from 0 to 3 in each cusp separately using 3D transesophageal echocardiographic data, defined as 0 if there was no high echo density, 1 if there was high echo density without acoustic shadow, 2 if there was high echo density with mild acoustic shadow partially visible below it, and 3 if there was high echo density with strong acoustic shadow completely invisible below it ( Figure 2 ). This scale was measured at the three different levels of the aortic valve apparatus: the “true” aortic annulus (a virtual ring defined by the three nadirs of the aortic cusps), the leaflets near the nadir, and the commissure.

Examples of visual scale of AVC. Score 0, no high echo density (A) ; score 1, high echo density without acoustic shadow (B) ; score 2, high echo density with mild acoustic shadow partially visible below it (C) ; score 3, high echo density with strong acoustic shadow completely invisible below it (D) . AVC , Aortic valve calcification.

At first, two orthogonal long-axis views of the aortic valve were extracted from the 3D data sets to be parallel to the aortic root ( Figures 3 B and 3C). A third plane perpendicular to both of the long-axis planes, the short axis of the aortic valve, was aligned ( Figure 3 A). In the midsystolic phase, fine adjustments of the cutting plane were performed to obtain the plane through the three nadirs of the aortic cusps for the evaluation of calcification at the aortic annulus. We moved the 2D cutting plane in 3D spaces to get an optimal long-axis view for each cusp and looked at both long-axis and short-axis images simultaneously to confirm the precise location of the calcification and its artifact. Next, we evaluated the calcification of the leaflets near the nadir, which can cause a prosthetic valve to be underexpanded ( Figure 3 B, red arrow). Finally, the short-axis plane was moved to the commissure level ( Figures 3 F and 3G), just beneath the highest position of the aortic valve where commissures between two cusps were clearly visualized, to evaluate the calcification of the commissure ( Figure 3 E). We obtained the sum of the AVC scales from the three different levels of the aortic valve apparatus separately. In addition, we assessed the shape of the calcification by measuring the radial and circumferential lengths of annular calcification ( Figures 4 A and 4B) and by focusing on any calcification protruding into the left ventricular outflow tract (LVOT) from the annular level ( Figure 4 E). To minimize and avoid the effect of the acoustic shadowing by the calcification, we analyzed multiple 3D volumes from different probe locations with multiple angles and directions in patients with severe acoustic shadowing.

Three-dimensional analysis of AVC. Example of the Philips QLAB quad screen displaying the 3D dataset of the aortic valve at the aortic annular level (A–D) and at the commissure level (E–H) . (A) The “true” aortic annular plane through the three nadirs of the aortic cusps ( red dot ) in the midsystolic phase. We moved the 2D cutting plane in 3D spaces to get an optimal long-axis view for each cusp and looked at both long-axis and short-axis images simultaneously (A,B) to confirm the precise location of the calcification and its artifact. Next we evaluated the calcification at the leaflets near the nadir ( B , red arrow ). Finally the short-axis plane was moved to the commissure level (E,F) to evaluate the calcification at the commissure. (C) and (G) show another long-axis view. (D) and (H) show 3D volume image. 2D , Two-dimensional; 3D , three-dimensional; AVC , aortic valve calcification.

Examples of the three specific features of AVC. Two representative cases with the specific features of AVC and subsequent significant paravalvular regurgitation after TAVR ( A–D and E–H ). Calcification at the annular level was measured with a radial length of 3.6 mm (A) and a circumferential length of 8.2 mm (B) , which caused significant paravalvular regurgitation (C,D) . Calcification protruding into the LVOT ( E,F , arrow ) from the annular level ( F , white dashed line ) caused significant paravalvular regurgitation (G,H) . AVC , Aortic valve calcification; LVOT , left ventricular outflow tract; TAVR , transcatheter aortic valve replacement.

Multidetector Computed Tomographic Image Acquisition and Measurement for Annular Size and Calcium Volume

Multidetector computed tomographic image acquisition was performed as described in our previous study. For annular size and leaflet calcium volume, curved multiplanar reconstruction analyses were performed using software specifically customized to valve analysis (3mensioValves version 4.1; 3mensio Medical Imaging BV, Bilthoven, The Netherlands). Leaflet calcium volume was also quantified using 3mensio Valves software for all available contrast scans, with a threshold for calcium detection set at 850 Hounsfield units. The distribution of AVC at the annulus for each cusp by real-time 3D TEE was compared with that by MDCT in another consecutive 26-patients who underwent TAVR at our institution. For real-time 3D transesophageal echocardiographic assessment, calcification was considered as absent if the AVC scale score at the annulus was 0 or 1 and as present if the score was 2 or 3. Furthermore, we compared total sum of AVC scale by real-time 3D TEE with leaflet calcium volume by MDCT.

Intraobserver and Interobserver Variability

For intra- and interobserver variability, 11 patients and 14 annular calcifications were randomly selected and measured by two independent observers at two separate times, without any sharing of their results with each other.

Statistical Analysis

Continuous data are presented as mean ± SD and categorical variables as frequencies or percentages, as appropriate. Differences between patients with and without significant paravalvular regurgitation were assessed by unpaired Student t tests or Mann-Whitney U tests for continuous variables and by χ ² tests or Fisher exact probability tests for categorical variables, as appropriate. The association between the sum of the AVC scales at the three different levels of the aortic valve apparatus and the grade of paravalvular regurgitation was assessed by the Spearman correlation coefficient. Sum of the AVC scale at the annulus, specific calcification features, and area cover index by real-time 3D TEE were included in a multivariate logistic regression analysis with the stepwise forward selection method to identify independent variables that were associated with significant paravalvular regurgitation. Two-tailed P values < .05 were considered statistically significant. For the AVC scale, intra- and interobserver reproducibility was evaluated using Bland-Altman analysis. Intra- and interobserver reproducibility for the measurements of calcification length was described by the percentage of mean absolute difference using Bland-Altman analysis and by intraclass correlation coefficients. The reproducibility for the AVC scale was assessed using the κ statistic and the percentage of mean absolute difference. Analyses were performed using SPSS version 19.0 (SPSS, Inc, Chicago, IL).

Three-Dimensional Analysis of AVC

We assessed the severity of AVC on a visual scale ranging from 0 to 3 in each cusp separately using 3D transesophageal echocardiographic data, defined as 0 if there was no high echo density, 1 if there was high echo density without acoustic shadow, 2 if there was high echo density with mild acoustic shadow partially visible below it, and 3 if there was high echo density with strong acoustic shadow completely invisible below it ( Figure 2 ). This scale was measured at the three different levels of the aortic valve apparatus: the “true” aortic annulus (a virtual ring defined by the three nadirs of the aortic cusps), the leaflets near the nadir, and the commissure.

Examples of visual scale of AVC. Score 0, no high echo density (A) ; score 1, high echo density without acoustic shadow (B) ; score 2, high echo density with mild acoustic shadow partially visible below it (C) ; score 3, high echo density with strong acoustic shadow completely invisible below it (D) . AVC , Aortic valve calcification.

At first, two orthogonal long-axis views of the aortic valve were extracted from the 3D data sets to be parallel to the aortic root ( Figures 3 B and 3C). A third plane perpendicular to both of the long-axis planes, the short axis of the aortic valve, was aligned ( Figure 3 A). In the midsystolic phase, fine adjustments of the cutting plane were performed to obtain the plane through the three nadirs of the aortic cusps for the evaluation of calcification at the aortic annulus. We moved the 2D cutting plane in 3D spaces to get an optimal long-axis view for each cusp and looked at both long-axis and short-axis images simultaneously to confirm the precise location of the calcification and its artifact. Next, we evaluated the calcification of the leaflets near the nadir, which can cause a prosthetic valve to be underexpanded ( Figure 3 B, red arrow). Finally, the short-axis plane was moved to the commissure level ( Figures 3 F and 3G), just beneath the highest position of the aortic valve where commissures between two cusps were clearly visualized, to evaluate the calcification of the commissure ( Figure 3 E). We obtained the sum of the AVC scales from the three different levels of the aortic valve apparatus separately. In addition, we assessed the shape of the calcification by measuring the radial and circumferential lengths of annular calcification ( Figures 4 A and 4B) and by focusing on any calcification protruding into the left ventricular outflow tract (LVOT) from the annular level ( Figure 4 E). To minimize and avoid the effect of the acoustic shadowing by the calcification, we analyzed multiple 3D volumes from different probe locations with multiple angles and directions in patients with severe acoustic shadowing.

Three-dimensional analysis of AVC. Example of the Philips QLAB quad screen displaying the 3D dataset of the aortic valve at the aortic annular level (A–D) and at the commissure level (E–H) . (A) The “true” aortic annular plane through the three nadirs of the aortic cusps ( red dot ) in the midsystolic phase. We moved the 2D cutting plane in 3D spaces to get an optimal long-axis view for each cusp and looked at both long-axis and short-axis images simultaneously (A,B) to confirm the precise location of the calcification and its artifact. Next we evaluated the calcification at the leaflets near the nadir ( B , red arrow ). Finally the short-axis plane was moved to the commissure level (E,F) to evaluate the calcification at the commissure. (C) and (G) show another long-axis view. (D) and (H) show 3D volume image. 2D , Two-dimensional; 3D , three-dimensional; AVC , aortic valve calcification.

Examples of the three specific features of AVC. Two representative cases with the specific features of AVC and subsequent significant paravalvular regurgitation after TAVR ( A–D and E–H ). Calcification at the annular level was measured with a radial length of 3.6 mm (A) and a circumferential length of 8.2 mm (B) , which caused significant paravalvular regurgitation (C,D) . Calcification protruding into the LVOT ( E,F , arrow ) from the annular level ( F , white dashed line ) caused significant paravalvular regurgitation (G,H) . AVC , Aortic valve calcification; LVOT , left ventricular outflow tract; TAVR , transcatheter aortic valve replacement.

Multidetector Computed Tomographic Image Acquisition and Measurement for Annular Size and Calcium Volume

Multidetector computed tomographic image acquisition was performed as described in our previous study. For annular size and leaflet calcium volume, curved multiplanar reconstruction analyses were performed using software specifically customized to valve analysis (3mensioValves version 4.1; 3mensio Medical Imaging BV, Bilthoven, The Netherlands). Leaflet calcium volume was also quantified using 3mensio Valves software for all available contrast scans, with a threshold for calcium detection set at 850 Hounsfield units. The distribution of AVC at the annulus for each cusp by real-time 3D TEE was compared with that by MDCT in another consecutive 26-patients who underwent TAVR at our institution. For real-time 3D transesophageal echocardiographic assessment, calcification was considered as absent if the AVC scale score at the annulus was 0 or 1 and as present if the score was 2 or 3. Furthermore, we compared total sum of AVC scale by real-time 3D TEE with leaflet calcium volume by MDCT.

Intraobserver and Interobserver Variability

For intra- and interobserver variability, 11 patients and 14 annular calcifications were randomly selected and measured by two independent observers at two separate times, without any sharing of their results with each other.

Statistical Analysis

Continuous data are presented as mean ± SD and categorical variables as frequencies or percentages, as appropriate. Differences between patients with and without significant paravalvular regurgitation were assessed by unpaired Student t tests or Mann-Whitney U tests for continuous variables and by χ ² tests or Fisher exact probability tests for categorical variables, as appropriate. The association between the sum of the AVC scales at the three different levels of the aortic valve apparatus and the grade of paravalvular regurgitation was assessed by the Spearman correlation coefficient. Sum of the AVC scale at the annulus, specific calcification features, and area cover index by real-time 3D TEE were included in a multivariate logistic regression analysis with the stepwise forward selection method to identify independent variables that were associated with significant paravalvular regurgitation. Two-tailed P values < .05 were considered statistically significant. For the AVC scale, intra- and interobserver reproducibility was evaluated using Bland-Altman analysis. Intra- and interobserver reproducibility for the measurements of calcification length was described by the percentage of mean absolute difference using Bland-Altman analysis and by intraclass correlation coefficients. The reproducibility for the AVC scale was assessed using the κ statistic and the percentage of mean absolute difference. Analyses were performed using SPSS version 19.0 (SPSS, Inc, Chicago, IL).

Baseline Clinical Characteristics and TAVR Procedure

Paravalvular regurgitation after TAVR was graded as none in 47 patients (23%), trivial in 60 patients (30%), mild in 58 patients (29%), moderate in 37 patients (18%), and severe in no patients. We could evaluate AVC in 202 patients, including the anterior side (i.e., corresponding to the right coronary cusp side on TEE), except in 31 patients (15%) with severe acoustic shadows. The data on baseline clinical characteristics and the TAVR procedure for the 37 patients with significant paravalvular regurgitation and 165 patients with no significant paravalvular regurgitation are shown in Table 1 . Compared with the group with no significant paravalvular regurgitation, the group with significant paravalvular regurgitation had more male patients, taller patients, and more patients with chronic kidney disease. There were no significant differences in TAVR procedure between the two groups.

Table 1

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Arterial Wave Reflection and Aortic Valve Calcification in an Elderly Community-Based Cohort
2. Three-Dimensional Transesophageal Echocardiography in Degenerative Mitral Regurgitation
3. The Relationship between Time from Myocardial Infarction, Left Ventricular Dyssynchrony, and the Risk for Ventricular Arrhythmia: Speckle-Tracking Echocardiographic Analysis
4. ASE 2015: Boston is Wicked Awesome!

Stay updated, free articles. Join our Telegram channel

Join