Background
The percutaneously implanted Edwards SAPIEN valve (Edwards Lifesciences, Irvine, CA) consists of cusps mounted within a stent. The individual impact of the stent and cusps on transvalvular flow and its implications for the echocardiographic assessment of valve function have not been previously reported.
Methods
The study group consisted of 40 patients who underwent successful implantation with the SAPIEN valve. Pulsed Doppler was recorded with sample volumes immediately proximal to the stent (prestent), within the stent but proximal to the cusps (in-stent precusp), and distal to the cusps (in-stent postcusp). The Doppler velocity index and effective orifice area were calculated using both prestent and in-stent precusp velocities to represent “subvalvular” flow and continuous-wave recordings of the left ventricular outflow tract and aortic valve to represent postvalvular flow.
Results
In all patients, there was flow acceleration at two levels: in-stent precusp and in-stent postcusp. The mean in-stent precusp peak velocities were significantly higher than the prestent values (1.5 ± 0.2 vs 1.0 ± 0.2 m/sec, P < .0001). Effective orifice area and Doppler velocity index calculated using the prestent versus in-stent precusp velocities were also significantly different (1.79 ± 0.34 vs 2.54 ± 0.46 cm 2 , P < .0001, and 0.48 ± 0.12 vs 0.73 ± 0.13, P < .0001, respectively).
Conclusions
The SAPIEN valve demonstrates flow acceleration at two levels, representing contributions of both the stent and valve cusps to the total valve gradient. Failure to recognize this phenomenon may result in inappropriate selection of the in-stent precusp pulsed Doppler spectrum to represent “subvalvular” flow, thereby overestimating the effective orifice area and Doppler velocity index.
In recent years, transcatheter aortic valve implantation has emerged as a promising new therapy for patients with severe aortic stenosis who are considered at high risk for surgical aortic valve replacement. One of the most frequently implanted transcatheter heart valves, the SAPIEN valve (Edwards Lifesciences, Irvine, CA), is a balloon-expandable valve that consists of bovine pericardial valve leaflets mounted inside a cylindrical stainless steel stent. Although modeled after surgically implanted valves, the SAPIEN transcatheter heart valve differs from surgically implanted aortic valve bioprostheses in that the stent is a cylinder that extends for a much longer distance (14.1 and 15.8 mm for 23-mm and 26-mm valves, respectively) than does a conventional surgical sewing ring. This creates the substrate for a complex hemodynamic profile with the potential for flow acceleration at both the inlet to the stent and again at the level of cusps, a phenomenon that has not previously been reported.
Echocardiography is the most widely used method for the initial evaluation and serial follow-up of aortic prosthetic valves. As recommended by the American Society of Echocardiography and European Association of Echocardiography guidelines for the assessment of prosthetic valves, key elements of the assessment are the calculation of effective orifice area (EOA) using the continuity equation and the Doppler velocity index (DVI). Each is based on the ratio of velocities proximal and distal to the prosthesis, both of which are derived with Doppler echocardiography, with EOA of the prosthesis calculated as (left ventricular outflow tract [LVOT] cross-sectional area × LVOT velocity-time integral [VTI])/VTI across the valve and DVI defined as the ratio of the velocities proximal to and distal to the valve. The guidelines emphasize that the cross-sectional area of the LVOT is derived assuming a circular geometry and using a diameter measurement just underneath the prosthesis using the parasternal long-axis view and that the LVOT VTI is measured proximal to the prosthesis, avoiding the region of subvalvular flow acceleration.
Should the flow characteristics of the SAPIEN valve differ from those of traditional surgical valves, this could have a significant impact on the echocardiographic calculation of EOA and DVI for these valves. The purpose of this study, therefore, was to use echocardiography to define the flow characteristics of the SAPIEN valve and to assess the impact, if any, of the cylindrical SAPIEN valve design on the echocardiographic calculation of EOA and DVI.
Methods
Study Group
The study group consisted of 40 consecutive patients with severe aortic stenosis who underwent successful transapical or transfemoral transcatheter aortic valve implantation with the SAPIEN 23-mm or 26-mm valve and who had analyzable digitally stored Doppler transthoracic echocardiograms. The 23-mm valve was implanted in patients with annular diameters of 18 to 21 mm, and the 26-mm valve was implanted in those with annular diameters of 21 to 25 mm. The implantation approach was transfemoral in 15 patients (38%) and transapical in 35 (62%), using the technique described in detail in previous reports. All subjects gave informed consent, and the study was approved by the institutional committee on human research.
Echocardiography
After implantation, patients underwent complete Doppler transthoracic echocardiographic examinations at discharge and between 6 and 12 months of follow-up. For each patient, the technically best study was selected for evaluation.
The LVOT diameter ( d ) was measured in systole immediately proximal to the ventricular border of the stent in the parasternal long-axis view, and the LVOT cross-sectional area was calculated using the formula π( d /2) 2 . Pulsed-wave Doppler was recorded using the apical five-chamber or three-chamber view with multiple sample volumes, including those immediately proximal to the stent (prestent), within the stent but proximal to the cusps (in-stent precusp), and distal to the cusps (in-stent postcusp). Spectra were optimized to eliminate spectral broadening and display modal velocities, and the pulsed-wave Doppler sample volume was minimized. A zoomed full-screen display of the pulsed-wave Doppler sample volume location was recorded preceding the capture of the spectral Doppler. Modal pulsed-wave Doppler VTI and peak velocity were measured in each location. In a large majority of patients, these parameters could not be measured at postcusp locations, because of aliasing of the pulsed-wave Doppler spectra. Thus, as is the convention, continuous-wave Doppler was used to capture the velocities distal to the valve. All measurements were made in triplicate by a single observer (S.S.) and averaged.
The DVI ( V 1 / V 2 ) and EOA ([LVOT cross-sectional area × VTI 1 ]/VTI 2 ) were calculated using both prestent and in-stent precusp velocities to represent “subvalvular” flow ( V 1 and VTI 1 ) and aligned continuous-wave recordings of the LVOT and aortic valve to represent postvalvular flow ( V 2 and VTI 2 ).
The mean transvalvular gradient was calculated from continuous-wave Doppler echocardiography using the Bernoulli equation, and left ventricular ejection fraction was calculated using Simpson’s biplane method. Aortic insufficiency was graded as trivial, mild, moderate, or severe.
Statistical Analysis
The sample size was chosen in part for convenience and in part on the basis of the following considerations: for a continuous measurement, a sample size of 20 would yield 80% power to detect a mean difference that is ≥62.6% of its corresponding standard deviation at a 5% significance level. A sample size of 40 would yield 80% power to detect a mean difference that is ≥44.3% of its corresponding standard deviation at a 5% significance level. On the basis of previously reported data from the Placement of Aortic Transcatheter Valve (PARTNER) cohort, the standard deviation for EOA for the SAPIEN valve is approximately 0.5 cm 2 . Because preliminary assessment of study subjects indicated that the choice of LVOT sampling position would result in a difference in calculated EOA of ≥0.4 cm 2 , the selected sample size provided adequate power for analysis of the entire cohort ( n = 40) and the groups with 23-mm versus 26-mm valves ( n = 20 each).
Categorical variables are expressed as percentages and continuous variables as mean ± SD. The comparison between the measurements with each of the sampling sites was carried out by paired two-sided t tests, and the comparison between two groups was carried out using unpaired two-sided t tests. P values < .05 were considered significant.
Results
Study Group
The study group consisted of 20 men and 20 women (mean age, 85.4 ± 7.7 years), with equal numbers of patients receiving 23-mm and 26-mm valves. More men (16 [40%]) received the 26-mm valve compared with women (four [19%]). The mean measured LVOT diameter was smaller in the group with 23-mm valves compared with the group with 26-mm valves (20.4 ± 0.7 vs 22.7 ± 0.7 mm, P < .0001). Additional information concerning the study group is provided in Tables 1 and 2 .
| Variable | Total ( n = 40) | Group 1 (23-mm valve) ( n = 20) | Group 2 (26-mm valve) ( n = 20) |
|---|---|---|---|
| Age (y) | 85.4 ± 7.7 | 87.7 ± 6.7 | 83.1 ± 8.2 |
| Men | 20 (50%) | 4 (20%) | 16 (80%) |
| Time to echocardiography (mo) ∗ | 7.2 ± 6.7 | 5.2 ± 6.0 | 9.4 ± 6.9 |
| Transfemoral approach | 25 (63%) | 9 (45%) | 16 (80%) |
∗ Time from the procedure to the follow-up echocardiographic study.
| Variable | Total ( n = 40) | Group 1 (23-mm valve) ( n = 20) | Group 2 (26-mm valve) ( n = 20) |
|---|---|---|---|
| LVOT diameter (mm) | 21.6 ± 1.4 | 20.4 ± 0.7 | 22.7 ± 0.7 |
| Left ventricular ejection fraction (%) | 60 ± 10 | 61 ± 10 | 60 ± 9 |
| Mean transaortic valve gradient (mm Hg) | 10 ± 3 | 9 ± 3 | 10 ± 4 |
| Peak aortic valve velocity (m/sec) | 2.1 ± 0.4 | 2.0 ± 0.3 | 2.2 ± 0.4 |
| Aortic regurgitation ∗ | |||
| None/trivial/mild | 38 | 19 | 19 |
| Moderate | 2 | 1 | 1 |
Hemodynamic Profile
In all patients, there was flow acceleration at two levels: within the stent but proximal to the cusps and distal to the cusps. Thus, there were three discrete pulsed-wave Doppler spectra: prestent, in-stent precusp, and in-stent postcusp ( Figure 1 ), although, as previously noted, aliasing precluded adequate capture of postcusp spectra in a large majority of subjects (95%). Although transesophageal spectra were not used for analytic purposes, they do provide images that better delineate the position of the cusps relative to the sites of flow acceleration ( Figure 2 ). Additionally, the phenomenon of flow acceleration at two sites can also be captured by close inspection of the continuous-wave spectra of the valves ( Figure 3 ). This reveals two spectra within the continuous-wave envelope corresponding to prestent and in-stent precusp velocities. The mean in-stent precusp pulsed Doppler peak velocities and VTIs were significantly higher than the prestent values (1.5 ± 0.2 vs 1.0 ± 0.2 m/sec, P < .0001, and 30.5 ± 7.1 vs 21.5 ± 5.6 cm, P < .0001, respectively; Table 3 ). The mean EOA calculated using the prestent VTI values as VTI 1 was significantly lower than that calculated using in-stent precusp VTI values as VTI 1 (1.79 ± 0.34 vs 2.54 ± 0.46 cm 2 , P < .0001). The DVI was also significantly lower when calculated using the peak pulsed-wave Doppler prestent velocities as V 1 versus in-stent precusp velocities as V 1 (0.48 ± 0.12 vs 0.73 ± 0.13, P < .0001).
| Variable | Sample volume prestent ∗ | Sample volume in-stent precusp † | P |
|---|---|---|---|
| V 1 max (m/sec) | 1.0 ± 0.2 | 1.5 ± 0.2 | <.0001 |
| VTI 1 (cm) | 21.5 ± 5.5 | 30.5 ± 7.1 | <.0001 |
| DVI | 0.48 ± 0.12 | 0.73 ± 0.13 | <.0001 |
| EOA (cm 2 ) | 1.79 ± 0.34 | 2.54 ± 0.46 | <.0001 |
| Indexed stroke volume (mL/m 2 ) | 44.7 ± 11.7 | 64.0 ± 16.2 | <.0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
