Background
Surgical aortic valve replacement (SAVR) is the definitive proven therapy for patients with severe aortic stenosis who have symptoms or decreased left ventricular (LV) function. The development of transcatheter aortic valve implantation (TAVI) offers a viable and “less invasive” option for the treatment of patients with critical aortic stenosis at high risk with conventional approaches. The main objective of this study was the comparison of LV hemodynamic and structural modifications (reverse remodeling) between percutaneous and surgical approaches in the treatment of severe aortic stenosis.
Methods
Fifty-eight patients who underwent TAVI with the CoreValve bioprosthetic valve were compared with 58 patients with similar characteristics who underwent SAVR. Doppler echocardiographic data were obtained before the intervention, at discharge, and after 6-month to 12-month follow-up.
Results
Mean transprosthetic gradient at discharge was lower ( P < .003) in the TAVI group (10 ± 5 mm Hg) compared with the SAVR group (14 ± 5 mm Hg) and was confirmed at follow-up (10 ± 4 vs 13 ± 4 mm Hg, respectively, P < .001). Paravalvular leaks were more frequent in the TAVI group (trivial to mild, 69%; moderate, 14%) than in the SAVR group (trivial to mild, 30%; moderate, 0%) ( P < .0001). The incidence of severe prosthesis-patient mismatch (PPM) was significantly lower ( P < .004) in the TAVI group (12%) compared with the SAVR group (36%). At follow-up, LV mass and LV mass indexed to height and to body surface area improved in both groups, with no significant difference. In patients with severe PPM, only the TAVI subgroup showed significant reductions in LV mass. LV ejection fraction improved at follow-up significantly only in TAVI patients compared with baseline values (from 50.2 ± 9.6% to 54.8 ± 7.3%, P < .0001).
Conclusions
Hemodynamic performance after TAVI was shown to be superior to that after SAVR in terms of transprosthetic gradient, LV ejection fraction, and the prevention of severe PPM, but with a higher incidence of aortic regurgitation. Furthermore, LV reverse remodeling was observed in all patients in the absence of PPM, while the same remodeling occurred only in the TAVI subgroup when severe PPM was present.
Calcific aortic stenosis (AS) is a common disorder in elderly patients, with a prevalence of 2.5% at 75 years and up to 8% at 85 years. Surgical aortic valve replacement (SAVR) is the definitive proven therapy for patients with severe AS who have symptoms or objective consequences of left ventricular (LV) dysfunction. Operative mortality is quite low, even in elderly patients when properly selected, and long-term results have been shown to be satisfactory. However, the hemodynamic performance of prosthetic valves cannot be equivalent to that of normal native valves, and consequently, a substantial proportion of patients develop prosthesis-patient mismatch (PPM) after SAVR. This condition was first described by Rahimtoola, and it has been shown to predict adverse outcomes and a reduced regression of LV mass (LVM), being also correlated with postoperative resting and exercise transvalvular pressure gradients. Severe PPM is defined as an indexed effective orifice area (EOAi) ≤ 0.65 cm 2 /m 2 .
Recently, transcatheter aortic valve implantation (TAVI) has been proven to be feasible in patients with AS, using either self-expandable or balloon-expandable valved stents. The development of this technique for the treatment of critical AS offers a viable and “less invasive” option for patients at high risk with conventional approach.
However, TAVI and SAVR represent two different techniques; in the former, the native aortic valve is preserved, whereas in the latter, it is removed. Preservation of the native calcified valve in the endovascular procedure may be responsible for eventual complications, such as incomplete and/or irregular expansion of the prosthetic valve or perivalvular leakage. Use of the percutaneous approach is expected to increase in the future with promising preliminary acute and midterm hemodynamic results, showing low transprosthetic gradients and large prosthetic valve EOA.
The main objective of this study was the echocardiographic comparison of LV hemodynamic and structural modifications (reverse remodeling) between percutaneous self-expanding valve bioprostheses (Medtronic, Inc., Minneapolis, MN) and surgically implanted (stented) bioprostheses for the treatment of symptomatic AS.
Methods
Patient Population
The present retrospective study included two subgroups of patients, all affected by symptomatic degenerative severe AS, who underwent the following procedures: (1) successful TAVI with the CoreValve ReValving system (CoreValve LLC, Irvine, CA) ( n = 58) and (2) successful SAVR with stented bioprostheses ( n = 58). These procedures were performed respectively in the hemodynamic and surgical sections of the Cardiac Thoracic and Vascular Department of the University of Pisa. Patients with moderate or severe mitral valve regurgitation were excluded from the study.
TAVI Group
These patients were referred to our department as being at high risk with standard SAVR or having been refused for surgical intervention. At baseline, all patients underwent structured assessments, including histories and physical examinations, electrocardiography, Doppler echocardiography, coronary and peripheral angiography, and computed tomography of the great arteries. Nineteen percent of TAVI patients had undergone prior percutaneous coronary revascularization, and 10% had undergone prior coronary artery bypass grafting. The baseline operative risk of patients was estimated by the logistic European System for Cardiac Operative Risk Evaluation score. High-risk patients were defined by consensus between an independent cardiologist and a cardiac surgeon regarding the morbidity and mortality associated with conventional surgery.
Inclusion criteria for TAVI were either compassionate use or the criteria described by Grube et al. already in use for the CoreValve implantation program in Italy. All patients gave written informed consent for the procedure.
Ninety consecutive patients who underwent TAVI were studied and selected, after excluding those who had unsuccessful TAVI (failure to implant the valve or procedural death, n = 8), those who died before 6-month to 1-year follow-up ( n = 4), and those who underwent TAVI <6 months previously ( n = 20).
All patients had complete clinical and echocardiographic evaluation at discharge and at 6-month to 12-month follow-up.
SAVR Group
TAVI patients were case matched with 58 patients of comparable sex, body mass index, aortic annular diameter, and LV ejection fraction (LVEF) who had undergone successful SAVR with stented bioprostheses. These 58 patients were selected from a prospective registry database including all patients who had undergone successful SAVR at the Cardiac Thoracic and Vascular Department. All clinical and echocardiographic data were collected prospectively at baseline, at discharge, and at 6-month to 12-month follow-up.
The stented bioprostheses used in this study were the Medtronic Mosaic porcine valve (Medtronic, Inc., Minneapolis, MN), the Carpentier-Edwards PERIMOUNT pericardial valve (Edwards Lifesciences, Irvine, CA), and the Mitroflow pericardial valve (Sorin Group, Saluggia, Italy). The porcine bioprosthetic valve consists of three porcine aortic valve leaflets cross-linked with glutaraldehyde and mounted on a metallic or polymer supporting stent. The pericardial valves were formed from strips of bovine pericardium mounted inside or outside a supporting stent.
SAVR was performed under general anesthesia, conducted with a propofol or remifentanil total intravenous technique. All procedures were performed with a standard median sternotomy under cardiopulmonary bypass with moderate (32°C) systemic hypothermia. Myocardial protection was achieved with antegrade and retrograde cold blood cardioplegia. Excision of the native aortic valve and annular debridement were performed in all cases before valve implantation. The size of the valve was determined by the diameter of the aortic annulus as measured by precalibrated cylindrical sizers and proprietary valve size. The valve was implanted in the supra-annular position with interrupted, radial, noneverting, pledget-supported sutures.
Seven percent of the SAVR group underwent coronary artery bypass grafting, while only 5% underwent concomitant coronary artery bypass grafting.
Technique of TAVI
The current third-generation (18-Fr) CoreValve ReValving system was used in all the TAVI procedures. The CoreValve aortic valve prosthesis consists of a trileaflet bioprosthetic porcine pericardial tissue valve, which is mounted and sutured in a self-expanding nitinol stent. The prosthetic frame (stent) is manufactured by laser cutting and has an overall length of 50 to 55 mm. Further details of the device have been described in previous studies. Vascular access was obtained either by percutaneous approach through the common femoral artery or by standard surgical cut-down of the subclavian artery. The procedure was performed with the patient under general anesthesia or local anesthesia in combination with a mild systemic sedative and analgesic treatment according to the patient’s needs. Balloon valvuloplasty with an 18-mm to 25-mm balloon (NuCLEUS PTV; NuMED Canada Inc., Cornwall, ON, Canada) under rapid pacing was performed before device placement, after which over a stiff guidewire, placed in the left ventricle, the device was deployed retrogradely under fluoroscopic guidance. Hemodynamic and echocardiographic outcomes were assessed serially during the procedure. After the procedure, all patients were transferred to the intensive care unit.
Two valve sizes, 26 and 29 mm, with an expanded diameter were available. The 26-mm valve was used with a corresponding native aortic annulus between 20 and 23 mm and the 29-mm valve with a native aortic annulus between 24 and 27 mm, according to measures on transthoracic and/or transesophageal echocardiography.
Transthoracic echocardiography is essential for preprocedural screening to measure the aortic valve annulus, LV outflow tract (LVOT), and ascending aorta, thus allowing an accurate selection of prosthetic valve size. Although the procedure can be performed under fluoroscopic guidance alone, transthoracic and/or transesophageal echocardiography can accurately evaluate the exact position of the deployment catheter and valve, which is critical for valve function.
Preoperative and Postoperative Conventional Two-Dimensional Color Doppler Echocardiography
All echocardiographic measurements were performed using a commercially available ultrasound system (Vivid 7; GE Healthcare, Milwaukee, WI) equipped with a harmonic 4.0-MHz variable-frequency phased-array transducer. All patients in the two groups underwent Doppler echocardiography at baseline before intervention, at hospital discharge, and at 6-month to 12-month follow-up.
LV end-diastolic diameter (LVEDD), LV end-systolic diameter (LVESD), end-diastolic ventricular septal thickness, and end-diastolic LV posterior wall thickness (EDPWth) were measured by M-mode echocardiography. According to these parameters, we calculated LVM and the relative LV wall thickness as follows: LVM (g) = 1.04 × [(LVEDD + EDPWth + EDPWth) 3 × LVEDD 3 ] × 13.6. Relative LV wall thickness was calculated as 2 × EDPWth/LVEDD. LVM index was determined by dividing the LVM measurement by body surface area (grams per square meter) (LVMbs) or by height 2.7 (LVMh). LV end-diastolic volume (LVEDV) and LV end-systolic volume (LVESV) were calculated from the apical two-chamber and four-chamber views using a modified Simpson’s method. LVEF was calculated as (LVEDV − LVESV)/LVEDV × 100.
The LVOT was measured in the parasternal long-axis view according to standard criteria. Cardiac index (liters per minute per square meter) was calculated as the product of LVOT area index, LVOT pulsed Doppler velocity-time integral, and heart rate. Transaortic peak velocity was measured by continuous-wave Doppler echocardiography, and the pressure gradient was calculated using the simplified Bernoulli equation. Aortic valve area (EOA) was obtained by the continuity equation method and was normalized for body surface area to obtain the aortic valve area index (EOAi). The occurence of severe PPM was defined as EOAi ≤ 0.65 cm 2 /m 2 . The presence, degree, and type (paravalvular vs transvalvular) of aortic regurgitation (AR) were classified as trivial, mild, moderate, or severe.
Definition of PPM
PPM was first proposed in 1987 by Rahimtoola. PPM occurs when the EOA of a normally functioning prosthesis is too small in relation to the patient’s body size (and therefore cardiac output requirements), resulting in abnormally high postoperative gradients. The most widely accepted and validated parameter for identifying PPM is EOAi (the prosthesis EOA divided by the patient’s body surface area).
In this study, we used the normal EOA values assessed in our echocardiography laboratory from data measured at 6 to 12 months after the operation. The normal EOA values were then divided by body surface area, and we considered severe PPM as an EOAi ≤ 0.65 cm 2 /m 2 ; this cutoff value was selected according to the results of previous studies. Other investigators have used slightly different definitions to indicate the presence of PPM and its severity: their threshold values of EOAi to identify PPM and to quantify its severity were mild (>0.85 cm 2 /m 2 ), moderate (≤0.85 and >0.65 cm 2 /m 2 ), and severe (≤0.65 cm 2 /m 2 ).
Statistical Analysis
Categorical variables were compared using χ 2 tests and are presented as frequencies. Continuous variables are presented as mean ± SD. Two-tailed unpaired Student’s t tests were used for comparisons between groups, and paired Student’s t tests were used for intragroup comparisons. Comparisons between groups were analyzed with one-way analysis of variance with Bonferroni’s posttests. A P value < .05 was considered statistically significant. The data were analyzed using Systat version 11 (Systat Software Inc., Chicago, IL).
Results
Baseline and Procedural Data of Patients
Baseline clinical and echocardiographic data are shown in Table 1 . Mean age and logistic European System for Cardiac Operative Risk Evaluation score were significantly higher in TAVI than in SAVR patients ( P < .001 and P < .01, respectively). There were no significant differences in body surface area values, body mass index, and mean aortic annular diameter ( Table 2 ). All patients had a critical AS, with a mean pressure gradient of 60.1 ± 19.3 mm Hg and a mean EOA of 0.55 ± 0.20 cm 2 . Baseline LVEFs were similar and preserved in the two groups. Furthermore, no significant differences in LVM, LVMh, and LVMbs were observed among groups at baseline.
| TAVI | SAVR | ||
|---|---|---|---|
| Variable | ( n = 58) | ( n = 58) | P |
| Age (years) | 82.2 ± 5.4 | 75.24 ± 4.7 | <.001 |
| Logistic EuroSCORE (%) | 22.3 ± 12.4 | 16.1 ± 7.7 | .01 |
| Women | 25 (43%) | 24 (41%) | NS |
| Body surface area (m 2 ) | 1.71 ± 0.19 | 1.81 ± 0.16 | NS |
| Body mass index (kg/m 2 ) | 24.3 ± 3.6 | 23.7 ± 3.9 | NS |
| PAS (mmHg) | 129.2 ± 17.4 | 133.4 ± 17.3 | NS |
| PAD (mm Hg) | 70.2 ± 10.9 | 71.5 ± 10.6 | NS |
| Prior CABG | 6 (10%) | 4 (7%) | NS |
| Aortic annular diameter (mm) | 21.7 ± 2.1 | 21.4 ± 2.1 | NS |
| LVEDD (cm) | 51.2 ± 5.3 | 52.5 ± 5.9 | NS |
| LVESD (cm) | 31.6 ± 7.9 | 34.2 ± 6.2 | NS |
| LVEDV (mL) | 101.9 ± 39.3 | 104.2 ± 32.4 | NS |
| LVESV (mL) | 51.9 ± 27.6 | 50.6 ± 21.6 | NS |
| LVEF (% | 50.2 ± 9.6 | 52.9 ± 7.6 | NS |
| Cardiac index (L/min/m 2 ) | 2.71 ± 0.85 | 2.62 ± 0.94 | NS |
| Peak pressure gradient (mm Hg) | 93.4 ± 25.3 | 92.6 ± 27.6 | NS |
| Mean pressure gradient (mm Hg) | 59.3 ± 18.1 | 56.6 ± 22.8 | NS |
| Peak aortic jet velocity (m/sec) | 4.8 ± 0.62 | 4.8 ± 0.66 | NS |
| EOA (cm 2 ) | 0.53 ± 0.19 | 0.58 ± 0.19 | NS |
| EOAi (cm 2 /m 2 ) | 0.31 ± 0.12 | 0.30 ± 0.15 | NS |
| TAVI | SAVR | ||
|---|---|---|---|
| Variable | ( n = 58) | ( n = 58) | P |
| LVEDD (cm) | |||
| Discharge | 51.5 ± 4.1 | 51.1 ± 5.2 | NS |
| Follow-up | 51.6 ± 5.6 | 50.3 ± 4.9 | NS |
| LVESD (cm) | |||
| Discharge | 30.88 ± 6.23 | 33.76 ± 6.38 | NS |
| Follow-up | 29.52 ± 6.05 | 32.86 ± 6.21 | NS |
| LVEDV (mL) | |||
| Discharge | 100.4 ± 29.8 | 91.2 ± 28.6 | NS |
| Follow-up | 95.4 ± 27.2 | 93.7 ± 26.4 | NS |
| LVESV (mL) | |||
| Discharge | 51.6 ± 26.1 | 43.9 ± 18.1 | NS |
| Follow-up | 46.1 ± 20.5 | 43.7 ± 18.8 | NS |
| LVEF (%) | |||
| Discharge | 52.1 ± 7.1 | 51.9 ± 6.6 | NS |
| Follow-up | 54.8 ± 7.3 | 52.8 ± 7.4 | NS |
| Cardiac index (L/min/m 2 ) | |||
| Discharge | 2.63 ± 0.94 | 2.53 ± 0.77 | NS |
| Follow-up | 2.67 ± 0.81 | 2.68 ± 0.99 | NS |
| Peak pressure gradient (mm Hg) | |||
| Discharge | 15.5 ± 8.1 | 26.8 ± 8.7 | .001 |
| Follow-up | 19.1 ± 7.0 | 27.1 ± 13.9 | .001 |
| Mean pressure gradient (mm Hg) | |||
| Discharge | 10.4 ± 4.8 | 13.6 ± 5.2 | .003 |
| Follow-up | 10.2 ± 3.7 | 13.5 ± 4.3 | .001 |
| Peak aortic jet velocity (m/sec) | |||
| Discharge | 2.19 ± 0.46 | 2.38 ± 0.42 | .03 |
| Follow-up | 2.10 ± 0.41 | 2.39 ± 0.44 | .04 |
| EOA (cm 2 ) | |||
| Discharge | 1.67 ± 0.43 | 1.48 ± 0.32 | .025 |
| Follow-up | 1.67 ± 0.57 | 1.37 ± 0.45 | .009 |
| EOAi (cm 2 /m 2 ) | |||
| Discharge | 0.92 ± 0.27 | 0.81 ± 0.18 | .015 |
| Follow-up | 1.01 ± 0.31 | 0.76 ± 0.27 | .0001 |
| Severe PPM | |||
| Discharge | 5 (9%) | 17 (30%) | .015 |
| Follow-up | 7 (12%) | 21 (36%) | .004 |
Prosthesis size distribution according to type of aortic bioprosthesis and native aortic valve is shown in Figure 1 . Thirty-eight patients (65%) in the TAVI group received 26-mm valves, and 20 patients (35%) received 29-mm valves. The 21-mm and 23-mm valves were the most commonly implanted in the SAVR group (63.5%). In the SAVR group, about 82% of patients received a Medtronic Mosaic porcine valves, and the other 18% received Carpentier-Edwards PERIMOUNT pericardial valve or Mitroflow pericardial valves.
Postoperative Conventional Two-Dimensional Color Doppler Echocardiography
Doppler Echocardiographic Data at Hospital Discharge
Doppler echocardiographic data at hospital discharge are shown in Table 2 . No significant differences in LVEDD, LVESD, LVEDV, and LVESV were observed among the groups. Peak and mean transprosthetic gradients at discharge were lower in the TAVI group than in the SAVR group, while EOA and EOAi were larger in the TAVI group than in the SAVR group ( Figure 2 ). The incidence of severe PPM was higher ( P < .015) in the SAVR group (30%) than in the TAVI group (9%).
