Author + information
- Received June 18, 2014
- Revision received August 20, 2014
- Accepted September 8, 2014
- Published online December 2, 2014.
- John Webb, MD∗∗ (, )
- Gino Gerosa, MD†,
- Thierry Lefèvre, MD‡,
- Jonathon Leipsic, MD∗,
- Mark Spence, MD§,
- Martyn Thomas, MD‖,
- Matthias Thielmann, MD¶,
- Hendrik Treede, MD#,
- Olaf Wendler, MD∗∗ and
- Thomas Walther, MD††
- ∗St. Paul’s Hospital, University of British Columbia, Vancouver, Canada
- †Policlinico Universitario, Padova, Italy
- ‡Institut Jacques Cartier, Massy, France
- §Royal Victoria Hospital, Belfast, Northern Ireland, United Kingdom
- ‖St. Thomas’ Hospital, London, United Kingdom
- ¶Universitätsklinikum Essen, Essen, Germany
- #Universitätsklinikum Eppendorf, Hamburg, Germany
- ∗∗King’s College Hospital, London, United Kingdom
- ††Kerckhoff Clinic, Bad Nauheim, Germany
- ↵∗Reprint requests and correspondence
: Dr. John Webb, St. Paul’s Hospital, 1081 Burrard Street, Vancouver, British Columbia V6Z 1Y6, Canada.
Background The SAPIEN 3 (Edwards Lifesciences Inc., Irvine, California) transcatheter valve incorporates features designed to address the well-known deficiencies of transcatheter aortic valve replacement (TAVR). An ultra–low-profile delivery system facilitates safe, controlled, and accurate implantation and an external seal minimizes paravalvular regurgitation.
Objectives The study evaluated whether TAVR with this third-generation valve would be a viable alternative to high- or intermediate-risk surgery for severe aortic stenosis.
Methods The prospective study enrolled 150 patients at 16 sites in Europe and Canada. Clinical and echocardiographic outcomes were assessed at baseline, post-procedure, and 30 days. New sizing recommendations were developed during the course of the study.
Results Patients were 83.6 ± 5.0 years of age, with multiple comorbidities reflected by a Society of Thoracic Surgeons score of 7.4 ± 4.5% and logistic EuroSCORE of 21.6 ± 12.3%. A transfemoral approach was chosen in 64.0% and alternative access (transapical/direct aortic) in the remainder. At 30 days, paravalvular regurgitation was none to mild in 96.4% and moderate in 3.5%. No patient had severe regurgitation. Transfemoral implantation was associated with low mortality (2.1%), no disabling stroke (0.0%), and fully percutaneous access and closure in 95.8%. Nontransfemoral alternative access was associated with higher rates of mortality (11.6%) and stroke (5.6%).
Conclusions This third-generation device addresses major deficiencies of earlier valves in terms of ease of use, accuracy of positioning, and paravalvular sealing. The rates of mortality and stroke with transfemoral access are among the lowest reported and support further evaluation as an alternative to open surgery in intermediate-risk patients. (Safety and Performance Study of the Edwards SAPIEN 3 Transcatheter Heart Valve [SAPIEN3]; NCT01808287)
Transcatheter aortic valve replacement (TAVR) has become a standard treatment for severe aortic stenosis when surgical risk is high. However, for TAVR to be accepted as a reasonable alternative to surgery in low- or intermediate-risk patients, improvements in outcomes beyond those achieved with early generation transcatheter heart valves (THVs) will be required.
TAVR was first performed with a balloon-expandable THV more than a decade ago. This prototypic device evolved into the SAPIEN and subsequently the SAPIEN XT THV (Edwards Lifesciences Inc., Irvine, California). The Edwards SAPIEN 3 (S3) (Edwards Lifesciences Inc.) represents the next generation of balloon-expandable THVs (1). The S3 THV system incorporates various new features to facilitate implantation while mitigating against vascular injury, stroke, suboptimal positioning, and paravalvular regurgitation.
We prospectively evaluated the safety and efficacy of the S3 THV in 150 patients with symptomatic, severe aortic stenosis who were enrolled at 16 sites in Europe and Canada. Patients provided informed consent, and approvals from ethics committees were obtained. The initial 50 patients were at high surgical risk (Society of Thoracic Surgeons [STS] risk score ≥8 or logistic EuroSCORE ≥15). The subsequent 100 patients were intermediate risk or higher (STS ≥4 or logistic EuroSCORE ≥10). Standardized definitions were used in accordance with the Valve Academic Research Consortium–2 consensus (2).
The S3 THV incorporates an optimized cobalt chromium alloy frame (which allows for an extremely low-crimped profile with high radial strength), bovine pericardial leaflets, and an adaptive external polyethylene terephthalate fabric seal (Figure 1). The S3 valve is longer than previous generation balloon-expandable valves; the 23, 26, and 29 mm S3 valves have expanded heights of 18, 20, and 22.5 mm.
The transfemoral Commander delivery catheter provides a stable platform allowing for controlled coaxial alignment and accurate positioning of the THV within the native valve. The system is compatible with 14-F (≤26 mm diameter THVs) and 16-F (29 mm THV) expandable introducer sheaths. The transapical/direct aortic Certitude (Edwards Lifesciences Inc.) delivery catheter has an ergonomic design, which allows for single-operator THV implantation using an 18-F (≤26 mm THV) or 21-F sheath (29 mm THV). Both delivery systems incorporate a central balloon marker, which is the primary landmark for positioning during THV expansion.
Echocardiographic and computed tomography analysis
Transthoracic echocardiograms performed at baseline, prior to hospital discharge, and at 30 days were read by an independent core laboratory (Cleveland Clinic, Cleveland, Ohio) according to standard criteria that incorporate regurgitant volume, regurgitant fraction, and effective regurgitant orifice area (2,3).
Patients were evaluated with 3-dimensional multidetector computed tomography (CT). Measurements were performed in accordance with Society of Cardiovascular Computed Tomography guidelines by a centralized core laboratory (St. Paul’s Hospital, Vancouver, Canada) (4). Aortic valve calcification was graded semiquantitatively: none, mild (small isolated spots), moderate (multiple larger spots), and severe (extensive calcification of all cusps). If present, the distribution of calcification and extension into the left ventricular outflow tract (LVOT) was also assessed as follows: mild, 1 nodule of calcium extending <5 mm and covering <10% of the perimeter of the annulus; moderate, 2 nodules or 1 extending >5 mm or covering >10% of the perimeter of the annulus; and severe, multiple nodules or a single focus extending >1 cm in length or covering >20% of the perimeter of the annulus (5).
The S3 THV was supplied with labeled diameters of 23, 26, and 29 mm. When fully expanded with an appropriately sized balloon, these THVs had predicted diameters of 22.75, 25.71, and 28.75 mm with externals of 406, 519, and 649 mm2, respectively.
The selection of an appropriately sized THV was at the discretion of the individual operator, informed by procedural transesophageal echocardiography and annular measurements and recommendations provided by the CT core laboratory. Sizing recommendations were based on annular area measurements with the percentage of oversizing (positive percentage) or undersizing (negative percentage) calculated using the formula: % oversizing = (THV nominal area/MDCT annular area – 1) × 100.
Prior to this study, sizing guidelines established for the earlier SAPIEN XT THV were utilized for the S3 THV. These recommended selecting a THV with an expanded area at least 5% larger than that of the annulus (5% oversizing) (1,6). During the course of this study, the requirement for routine oversizing was discontinued so as to allow selection of a THV up to 5% smaller than the CT-derived annulus area (5% undersizing).
Continuous variables were summarized as mean ± SD or as medians (minimum, maximum) as appropriate, and were compared using Student t test or Wilcoxon rank sum test. Categorical variables were compared using the Fisher exact test. Additionally, New York Heart Association functional class and Canadian Cardiovascular Society angina values at 30 days were compared to baseline values using the Bowker’s test of symmetry to assess the shifts across 4 ordinal classes. An alpha level of 0.05 was used for all hypothesis testing. No multiplicity adjustment was performed. All statistical analyses were performed using SAS software version 9.3 (SAS Institute, Cary, North Carolina).
Mean age was 83.6 ± 5.0 years and 54% of patients were female. As shown in Table 1, patients had multiple comorbidities with a predicted risk of operative mortality of 7.4 ± 4.5% by STS and by 21.6 ± 12.3% by logistic EuroSCORE estimates.
Transcatheter alternative access (transapical and direct aortic) patients, as compared to transfemoral patients, were a higher-risk group (Table 1). They had significantly more moderate to severe mitral regurgitation (41.0% vs. 13.3%; p = 0.0018), peripheral vascular disease (38.9% vs. 16.7%; p = 0.0032), and myocardial infarction (27.8% vs. 11.5%; p = 0.0142) as well as trends toward more coronary disease (p = 0.1561), atrial fibrillation (p = 0.1247), and bypass surgery (p = 0.0557), and a higher logistic EuroSCORE (p = 0.0215). Transcatheter alternative access patients were more likely to undergo general anesthesia (100.0% vs. 63.5%; p < 0.0001) and anesthesia time was longer (153.0 min vs. 128.9 min; p = 0.0007), while they received less contrast (96.9 ml vs. 148.7 ml; p < 0.0001) and had a shorter fluoroscopy time (7.6 min vs. 17.5 min; p < 0.0001) (Table 2).
CT findings at baseline are shown in Table 3. Moderate or severe calcification was commonly seen; it was annular in 31.7%, subannular in 25.9%, and in the LVOT in 23.9%. Mean annular area in mid-systole was 490.9 ± 79.8 mm2. Mean diameter from CT was 24.5 ± 2.1 mm (Table 3), as compared to a 2-dimensional transesophageal echocardiography diameter of 23.2 ± 2.3 mm (Table 2). For patients undergoing transfemoral access, the minimum access artery diameter was 6.6 ± 1.2 mm with moderate or severe calcification and tortuosity in 40.6% and 52.2%, respectively.
Access was transfemoral in 96 patients (64.0%). Alternative access (transapical or direct aortic) was utilized in 54 patients (36.0%). Implanted valves were 23 mm in 26.7% of patients, 26 mm in 53.3%, and 29 mm in 20.0%. The catheter system was routinely able to deliver the valve to and across the native valve in an atraumatic manner, to achieve accurate coaxial positioning, and to facilitate controlled stable THV expansion. Malposition occurred in 1 patient (0.7%), as did annular rupture. CT analysis suggested excessive oversizing (approximately 25% by area) and the presence of severe subannular calcification. This patient underwent conversion to open heart surgery. Procedural characteristics are detailed in Table 2.
Outcomes at 30 days
All-cause mortality was 2.1% in the transfemoral access cohort and 11.1% in the alternative access cohort (p = 0.03). Strokes were reported in 1% of transfemoral patients and 5.6% of alternative access patients (p = 0.13).
Pacemakers were required in 13.3% of all patients. One patient with THV malposition underwent repeat intervention with implantation of a second S3 THV during follow-up. No patient suffered coronary obstruction, device embolization, or rehospitalization. Clinical outcomes are shown in Table 4.
Transthoracic echocardiography at 30 days showed a reduction in mean gradient from 45.2 ± 14.5 mm Hg to 10.6 ± 4.7 mm Hg and an increase in valve area from 0.6 ± 0.2 cm2 to 1.5 ± 0.4 cm2 (Figure 2). Paravalvular regurgitation was absent or trace in 74.3%, mild in 22.1%, and moderate in 3.5%. No patient developed severe paravalvular regurgitation (Figure 3).
Patients in New York Heart Association functional class ≥3 fell from 88.1% at baseline to 6.6% at 30 days. Similarly at follow-up, 97% of patients were in Canadian Cardiovascular Society angina class ≤1. There were significant improvements in quality of life as documented by both visual analogue and 6-min walk testing (Figure 4).
The 30-day mortality rate of 2.1% for patients undergoing transfemoral access is among the lowest ever reported in a large multicenter series (Central Illustration) (7-12). The contrasting mortality of 11.1% in patients undergoing an alternative access procedure is consistent with a higher risk cohort with higher rates of peripheral and coronary arterial disease. Consistent with this, multisystem failure played a role in none of the transfemoral deaths, but did in all of the alternative access deaths. By way of clinical comparison, in the STS/American College of Cardiology Transcatheter Valve Therapy high-risk registry, despite a similar STS-predicted risk profile, transfemoral mortality rates of 5.0% were more than double that reported here, while nontransfemoral mortality was similar at 10.8% (13).
The S3 low-profile transfemoral system was associated with a low (4.2%) rate of major vascular complications and no access-related mortality. Notably, prior studies commonly utilized open surgical arterial access; in the current study, percutaneous access and closure were routine (95.8% of patients). Transfemoral patients had a minimum access arterial diameter of only 6.6 ± 1.2 mm despite frequent tortuosity and calcification, suggesting that an increasingly broad spectrum of patients will be candidates for a fully percutaneous transfemoral procedure.
The 30-day stroke rate of 2.7% (femoral and nontransfemoral combined) compares favorably with the 4.3% and 4.9% rates seen in recent PARTNER 2 (Placement of AoRTic TraNscathetER Valves Trial) SAPIEN XT and CoreValve high-risk randomized studies (10,14). More dramatic is the low rate of stroke (0% disabling, 1.0% nondisabling) associated with transfemoral access.
Transcranial Doppler studies have demonstrated that cerebral emboli are most common as the THV crosses, or is positioned or repositioned within, the native valve (15). Repeated implantation attempts, valve dislodgement, and repositioning have been associated with an increase in stroke (16). Balloon dilation, performed after initial THV expansion to reduce paravalvular regurgitation, also has been associated with an increase in embolic stroke (17-19). A low-profile delivery system that facilitates atraumatic implantation, potentially without pre-dilation and with very low rates of post-dilation (3.3%) reported here, might be expected to be associated with a relative reduction in stroke risk.
New onset left bundle branch block was observed in 18% of patients, similar to the 12% to 30% rate reported following TAVR with balloon-expandable valves and lower than the 29% to 65% rate seen with current self-expanding valves (20). New left bundle branch block has been associated with an increased need for pacemakers, reduced left ventricular function, and poorer functional status (20). Similarly, the rate of new pacemaker implantation was 13.3%, a rate higher than generally reported with balloon-expandable valves, but lower than that seen with self-expanding valves (9,11,14). Early in this experience, operators tended to use a positioning technique similar to the SAPIEN XT THV, although the S3 THV is 3 to 4 mm longer than earlier balloon-expandable THVs. As the trial progressed it was recommended that the S3 be deployed slightly more aortic; that is, with the middle balloon marker centered at or just above, rather than below, the annular plane. Early clinical experience suggests that this approach may reduce the need for pacemaker implantation.
Paravalvular regurgitation occurs commonly with TAVR and has been associated with poorer late survival. Comparing rates of paravalvular regurgitation between series, or between core laboratories, is difficult. However, the 3.5% rate of at least moderate paravalvular regurgitation is extremely low by any estimate. At 1 month, 96.4% of patients had no, trivial, or mild leaks. None had severe regurgitation. Importantly, the low rate and severity of paravalvular regurgitation was realized with modest area oversizing of <7%, which is well below the required area oversizing for prior balloon expandable prostheses.
Effective annular sealing requires accurate positioning of the THV within the native annulus. Accurate positioning is facilitated by the stable THV delivery platform; deflection controls that facilitate coaxial alignment; and slow, controlled balloon expansion. Almost all valves (99.3%) were placed at the intended location. Importantly, the external adaptive sealing cuff appeared to allow for effective sealing even in areas the implant frame was not directly apposed to the annulus.
Prior to this study, the sizing guidelines established for the earlier SAPIEN XT THV were utilized for the S3 THV. The guidelines recommended selecting a THV with an expanded area at least 5% larger than the annulus to achieve optimal paravalvular sealing (1,6). This approach may have been the result of routine underestimation of annular dimensions by 2-dimensional echocardiography. With only 3 THV sizes available, the implanting physician was sometimes faced with choosing between a smaller THV with the potential for paravalvular leak or the next larger THV with the potential for annular injury or rupture. In such circumstances, underexpansion of an oversized THV by underfilling the deployment balloon was sometimes employed to reduce the risk of annular rupture (21). However, there are concerns that excessively underexpanded THVs may suffer from reduced leaflet durability.
During our early S3 experience, it became apparent that the absence of oversizing, or even modest undersizing, was compatible with effective sealing when enabled by the external sealing cuff and accurate positioning. However, undersizing assumes accurate 3-dimensional CT estimates of annular dimensions (22). New CT-based area guidelines were developed that recommended an acceptable lower range of –5% (undersizing).
Recent data demonstrate that extreme area oversizing (>20%) is a predictor of annular injury (5%). Undersizing (as much as –5%) may be preferable to extreme oversizing to reduce this risk. Similarly calcification involving the LVOT is also a predictor of annular injury (5). Consequently, in this setting, an even narrower sizing range of –5% (undersizing) to +10% (oversizing) may be desirable (5). Proposed CT area-based sizing guidelines are shown in Table 5. These guidelines will require validation in future studies.
This study was limited primarily by the learning curve associated with the use of a new device and by the sample size.
The S3 THV addresses major deficiencies of earlier transcatheter valves in terms of ease of use, accuracy of positioning, and paravalvular sealing. The low-profile delivery system permits safe and fully percutaneous arterial access in the majority of patients. The low rates of mortality, stroke, vascular complications, and paravalvular regurgitation achieved with transfemoral access are among the lowest reported to date and support further evaluation of this new device as an alternative to surgery in intermediate risk patients.
COMPETENCY IN PATIENT CARE: TAVR has become an accepted option for patients with severe, symptomatic aortic valve stenosis at high risk of complications associated with surgical aortic valve replacement but technological improvements must be demonstrated to accept a catheter-based alternative to surgery in low- or intermediate-risk patients.
TRANSLATIONAL OUTLOOK: The low rates of stroke, vascular complications, paravalvular regurgitation, and mortality associated with transfemoral delivery of the SAPIEN 3 (Edwards Lifesciences Inc., Irvine, California) aortic valve prosthesis warrant further study of TAVR in intermediate-risk patients.
Drs. Webb, Spence, and Thomas have received consulting fees from Edwards Lifesciences. Dr. Gerosa has received honoraria and consulting fees from Edwards Lifesciences, AstraZeneca, St. Jude Medical, Heartware, Jarvik, and Sorin; and has received research support from and is on the advisory board for Edwards Lifesciences. Dr. Lefèvre has served as a proctor and consultant for Edwards Lifesciences. Dr. Thomas has served as a consultant for and received research support from Edwards Lifesciences. Dr. Wendler has received speakers’ honoraria from, has been a proctor for, and has been a consultant for Edwards Lifesciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- computed tomography
- left ventricular outflow tract
- Society of Thoracic Surgeons
- transcatheter aortic valve replacement
- transcatheter heart valve
- Received June 18, 2014.
- Revision received August 20, 2014.
- Accepted September 8, 2014.
- American College of Cardiology Foundation
- Binder R.K.,
- Rodés-Cabau J.,
- Wood D.A.,
- et al.
- Zoghbi W.,
- Chambers J.,
- Dumesnil J.,
- et al.
- Achenbach S.,
- Delgado V.,
- Hausleiter J.,
- Schoenhagen P.,
- Min J.,
- Leipsic J.
- Barbanti M.,
- Yang T.H.,
- Rodés Cabau J.,
- et al.
- Binder R.K.,
- Webb J.G.,
- Willson A.B.,
- et al.
- Webb J.G.,
- Chandavimol M.,
- Thompson C.,
- et al.
- Thomas M.,
- Schymik G.,
- Walther T.,
- et al.
- ↵Leon MB, for the PARTNER II investigators. A randomized evaluation of the SAPIEN XT transcatheter valve system in patients with aortic stenosis who are not candidates for surgery: PARTNER II, inoperable cohort. Paper presented at the 62nd Annual Scientific Session of the American College of Cardiology; March 10, 2013; San Francisco, CA.
- Wendler O. 30-Day Outcomes from the SOURCE XT TAVI Post Approval Study. Paper presented at: EuroPCR 2012; May, 2012; Paris, France.
- Kahlert P.,
- Al-Rashid F.,
- Dottger P.,
- et al.
- Tamburino C.,
- Capodanno D.,
- Ramondo A.,
- et al.
- Nombella-Franco L.,
- Rodés-Cabau J.,
- DeLarochellière R.,
- et al.
- Daneault B.,
- Koss E.,
- Hahn R.T.,
- et al.
- Hahn R.T.,
- Pibarot P.,
- Webb J.,
- et al.
- Urena M.,
- Mok M.,
- Serra V.,
- et al.
- Barbanti M.,
- Leipsic J.,
- BInder R.,
- et al.
- Yang T.,
- Webb J.,
- Blanke P.,
- et al.