Author + information
- Received September 10, 2018
- Revision received November 14, 2018
- Accepted December 10, 2018
- Published online March 18, 2019.
- John G. Webb, MD∗ (, )@CHVI85209027,
- Dale J. Murdoch, MBBS,
- Robert H. Boone, MD,
- Robert Moss, MBBS,
- Adrian Attinger-Toller, MD,
- Philipp Blanke, MD,
- Anson Cheung, MD,
- Mark Hensey, MB, BCh, BAO,
- Jonathon Leipsic, MD,
- Kevin Ong, MD,
- Janarthanan Sathananthan, MBChB,
- David A. Wood, MD,
- Jian Ye, MD and
- Paolo Tartara, MD
- Centre for Heart Valve Innovation, St Paul’s Hospital, University of British Columbia, Vancouver, British Columbia, Canada
- ↵∗Address for correspondence:
Dr. John G. Webb, St. Paul’s Hospital, 1081 Burrard Street, Vancouver, British Columbia V6Z 1Y6, Canada.
Background Severe mitral regurgitation (MR) conveys significant morbidity and mortality, and surgical repair or replacement may not be a desirable option.
Objectives The purpose of this study was to evaluate the feasibility of a percutaneous transseptal transcatheter mitral valve replacement (TMVR) system.
Methods This first-in-human study was conducted between August 2017 and August 2018. The system comprises a nitinol dock, which encircles the chordae tendineae, and a balloon-expandable transcatheter heart valve. The dock and transcatheter heart valve form an ensemble, with the native mitral valve leaflets secured in between, thereby abolishing MR. Key inclusion criteria were severe symptomatic MR and high surgical risk; exclusion criteria included left ventricular ejection fraction <30% or screening suggesting unfavorable anatomy. The primary endpoint was technical success as defined by Mitral Valve Academic Research Consortium (MVARC) criteria at completion of the index procedure. The secondary endpoint was freedom from mortality, stroke, and device dysfunction (MR grade >1, mitral gradient >6 mm Hg, left ventricular outflow tract gradient >20 mm Hg) at 30 days.
Results Ten patients with severe MR of various etiologies (4 degenerative, 4 functional, and 2 mixed) were treated. The device was successfully implanted and the primary endpoint was achieved in 9 of 10 patients (90%). By transesophageal echocardiography, total MR was reduced to ≤ trivial in all implanted patients, and mean transmitral gradient was 2.3 ± 1.4 mm Hg. A pericardial effusion occurred in 1 patient: pericardiocentesis was performed, and the device was not implanted. Median length of hospital stay was 1.5 days. At 30 days, there was no stroke, myocardial infarction, rehospitalization, left ventricular outflow tract obstruction, device migration, embolization, or conversion to mitral surgery. One patient had recurrent regurgitation due to a paravalvular leak, treated with a closure device. All other treated patients had ≤1+ MR. No patients died.
Conclusions Percutaneous transvenous transseptal TMVR is feasible and safe in patients with severe MR who are at high risk for mitral valve surgery. Further evaluation is warranted.
Symptomatic mitral regurgitation (MR) conveys significant morbidity and mortality (1). However, many patients with severe MR are not treated with surgery due to advanced age, left ventricular (LV) dysfunction, or other comorbidities (2). This unmet clinical need has driven the development of safer, catheter-based treatments for mitral valve disease.
Transcatheter mitral valve repair can be safe and effective in patients with suitable anatomy (3). However, many patients have unsuitable anatomy and repair may be difficult, unsuccessful, or temporary (4,5). Transcatheter repair is associated with lesser degrees of MR reduction, which may be associated with poorer clinical outcomes and durability (6). Although surgical mitral valve repair is often preferred over surgical replacement (7), the relative benefits and risks in patients undergoing transcatheter replacement and repair are unknown.
Early experience with transcatheter mitral valve replacement (TMVR) has suffered, in part, due to the limitations of apical access and the associated thoracotomy. For patients with prior surgical valve replacement or repair valve-in-valve or valve-in-ring procedures are proven options, and for those with severe mitral annular calcification, valve-in mitral annular calcification is sometimes feasible (8,9). Building upon substantial experience with mitral valve-in-valve and valve-in-ring procedures utilizing the balloon-expandable Sapien 3 transcatheter heart valve (THV), we report the first-in-human experience with a percutaneous transseptal TMVR procedure (Edwards Sapien M3, Edwards Lifesciences, Irvine, California).
This first-in-human study was conducted at the Centre for Heart Valve Innovation, St. Paul’s Hospital, Vancouver, Canada, between August 2017 and August 2018. Patients included were age ≥18 years, with severe MR and symptoms of heart failure (New York Heart Association functional class ≥2). All patients were reviewed by a multidisciplinary heart team and considered: 1) suitable for mitral valve intervention; 2) at high or prohibitive surgical risk; and 3) not ideally suited to other available transcatheter mitral valve interventions (i.e., edge-to-edge repair, transapical TMVR). Exclusion criteria were LV end-diastolic diameter >70 mm and LV ejection fraction <30%. Anatomical feasibility was determined using transesophageal echocardiography (TEE) and computed tomography (CT) imaging: patients with severe mitral leaflet calcification, unsuitable chordal anatomy, high risk of left ventricular outflow tract (LVOT) obstruction, or unfavorable mitral valve anatomy were also excluded.
Utilizing percutaneous femoral venous access, a transseptal deflectable sheath is placed in the left atrium. A steerable catheter is then advanced just under the posteromedial mitral commissure (Figure 1). Using fluoroscopy and TEE guidance, the expandable polytetrafluoroethylene–covered nitinol “dock” is advanced under the anterior mitral leaflet. As the nitinol dock is advanced into the left ventricle, it assumes its pre-determined shape and encircles the chordae tendineae below the level of the mitral annulus. The dock is a single component with 3 distinct sections: a leading turn has a larger diameter (37 mm) to capture the chords, subsequent functional turns have a smaller diameter (25.5 mm outer diameter) and provide the anchor for the balloon-expandable THV. A polyethylene terephthalate braid covering the functional turns of the dock increases retention forces and prevents migration. A final atrial turn helps maintains dock position prior to THV deployment. While connected to the delivery catheter, the dock is fully retrievable.
After release of the dock, a balloon-expandable bovine pericardial leaflet THV is advanced through the transseptal sheath and deployed during rapid ventricular pacing. The transseptal TMVR system ensemble secures the native mitral valve leaflets between the dock and THV frame. Sealing occurs between the native mitral valve leaflets and a knitted (polyethylene terephthalate) cloth outside of the THV frame (Figure 2). The Sapien M3 valve is identical to the 29-mm diameter Sapien 3 aortic THV, with the addition of an external knitted PET seal which covers the entire outer surface of the valve frame. Implantation of the transseptal TMVR system is very similar to implantation of the Sapien 3 THV in failed mitral surgical bioprostheses or rings, a procedure with which there is extensive experience (Figure 3, Online Videos 1, 2, and 3).
Clinical follow-up was performed in-hospital and at 30 days for all patients. Transthoracic echocardiography was routinely performed pre-discharge and at 30 days, and CT between discharge and 30 days. All patients were treated with oral anticoagulation post-procedure. The primary endpoint for the study was technical success as defined by Mitral Valve Academic Research Consortium (MVARC) criteria at completion of the index procedure. The secondary endpoint was successful device implantation and freedom from mortality, stroke, and device dysfunction (MR grade >1, mitral gradient >6 mm Hg, LVOT gradient >20 mm Hg) at 30-day follow-up. Periprocedural complications were defined as per MVARC criteria (10,11).
Research was conducted in compliance with the Declaration of Helsinki for human investigation. Patients provided written informed consent. Research approval was granted by the institutional ethics review board.
Continuous data are presented as median (Q1, Q3) or mean ± SD; categorical variables are presented as count and percentage. Comparisons between baseline and 30-day parameters are made using the Wilcoxon signed-rank test.
H0: the median difference = 0
Ha: the median difference ≠ 0
The 2-sided test was performed at alpha = 0.05. Statistical analysis was performed using R and SAS.
Ten patients were included in the study cohort, mean age 76.1 ± 5.3 years (range 69 to 87 years). A total of 5 patients (50%) were men. Patients were judged to be at high or prohibitive surgical risk by at least 2 experienced cardiac surgeons. Mean Society of Thoracic Surgeons (STS) predicted risk of mortality was 3.8 ± 2.4% (range 1.2% to 9.8%) and EuroSCORE II was 5.9 ± 2.1%. Baseline characteristics and comorbidities are listed in Table 1.
MR was graded severe (4 of 4) in all 10 patients. MR etiology was degenerative in 4 (40%), functional in 4 (40%), and mixed in 2 (20%). Of those with degenerative MR, the underlying pathology was prolapse (1), flail segment (2), leaflet retraction and flail segment (1), leaflet perforation (prior endocarditis) and flail segment (1), and calcification (1). LV systolic function was moderately impaired (LV ejection fraction 30% to 50%) in 6 (60%), and regional wall motion abnormalities were present in 4 (40%). Median (Q1, Q3) LV end-diastolic diameter was 60 mm (52 to 63.75 mm; range 45 to 65 mm). Mitral annular area by CT was 12 ± 2.3 cm2 (range 8.5 to 15.1 cm2).
The primary endpoint was achieved in 9 of 10 patients (90%) (Central Illustration). Total MR was reduced to ≤ trivial (0 of 4) in all implanted patients, and mean mitral gradient by TEE was 2.3 ± 1.4 mm Hg (Figure 4) (Online Videos 2 and 3). Concomitant paravalvular leak closure was not required in any patients, and iatrogenic atrial septal defect (ASD) closure was performed in 3 patients (30%) for ASD ≥10 mm or bidirectional interatrial flow. Total procedure time was 220 ± 45 min, fluoroscopy time was 57 ± 24 min, and total contrast volume was 72 ± 38 ml.
A pericardial effusion occurred in 1 patient during dock deployment: the dock was removed successfully, pericardiocentesis performed, and procedure terminated. The patient was discharged on day 2 to continue on medical management. No other procedural complications occurred. Median length of stay was 1.5 days.
The secondary endpoint (successful device implantation and freedom from mortality, stroke, and device dysfunction [MR grade >1, mitral gradient >6 mm Hg, LVOT gradient >20 mm Hg]) was met in 7 patients (70%). At 30 days, there was no death, myocardial infarction, stroke, rehospitalization, or LVOT obstruction.
In patients receiving the device, mitral regurgitation was ≤ mild in 8 (89%) and severe in 1. At 1 month, 1 patient had severe paravalvular regurgitation secondary to a leaflet or chordal tear (A1 scallop) adjacent to an area of calcification. A 12-mm Amplatzer Vascular Plug II (Abbott Vascular, Minneapolis, Minnesota) was implanted with MR reduced to moderate. No other device-related complications occurred: there was no device migration, embolization, or conversion to mitral surgery (Table 2).
Median transmitral gradient at 30 days was 6 mm Hg (Q1, Q3: 5, 6 mm Hg). One patient had an elevated mitral gradient at 30 days (10 mm Hg) with normal leaflet motion, no MR, and no thrombus detected on multiphase CT. No patients had clinical or echocardiographic evidence of LVOT obstruction. Pre-procedure and post-procedure LVOT gradient by TTE doppler assessment were not different (3.5 [Q1, Q3: 3.25, 6.00) vs. 5 [Q1, Q3: 2, 6]; p = 1.00) (Table 3).
One patient remained in the hospital after the index procedure due to oropharyngeal bleeding related to TEE and required blood transfusion (2 U; MVARC minor bleeding) and antibiotics for ventilator-acquired pneumonia.
Functional status was improved at 30 days (median New York Heart Association functional class III vs. II; p = 0.020). The 6-min walk distance and Kansas City cardiomyopathy questionnaire (KCCQ-12) were unchanged from baseline (median 288 vs. 350; p = 0.812; and 51 vs. 73.45; p = 0.195, respectively).
Post-procedural CT demonstrated a well-seated transseptal TMVR system ensemble in all cases, with no evidence of malpositioning or migration. Neo-LVOT area was 5.1 ± 2.3 cm2, and no patients had a neo-LVOT area <1.5 cm2.
This study investigated the feasibility of a new transcatheter mitral valve replacement system in patients with severe MR who were at high risk with conventional surgery and unsuitable for other available transcatheter procedures (edge-to-edge repair or transapical TMVR). The important findings of this study are: 1) transseptal TMVR is feasible with an early procedural success rate of 90%; 2) results were achieved across a variety of mitral regurgitation etiologies, both functional and degenerative (chordal, leaflet, calcification); 3) the procedure is safe, with few procedural complications in this high-risk cohort and no deaths at 30 days; and 4) early discharge is possible with median length of stay of 1.5 days.
The transseptal TMVR procedure has several potentially desirable features. First, a transfemoral transseptal approach is less invasive with less morbidity and recovery time than conventional surgery or transapical TMVR. Similar to percutaneous mitral plication, patients remain hemodynamically stable for long periods, blood loss is minimal, and the left ventricle is not compromised. Of interest, MR was often reduced after the first turn of the dock encircled the mitral chords, consistent with a variable annuloplasty effect.
The transseptal TMVR valve represents a relatively minor modification of the 29-mm Sapien 3 aortic THV, with its well-documented reliability, hemodynamic function, and durability. As anchoring of the THV relies primarily on the 26-mm diameter dock encircling the mitral leaflets and chords, mitral annular dimensions are relatively less important, and the single 29-mm diameter transseptal TMVR system is suitable for a relatively broad range of ventricular anatomies.
As with all prosthetic mitral valves, LVOT obstruction is a concern and must be screened for. However, there are 2 major factors that mitigate against this risk: 1) the axis of the transcatheter valve is routinely biased toward the shorter posterior native mitral leaflet and away from the longer anterior leaflet; and 2) the encircling dock pulls the anterior leaflet posterior away from the LVOT. Focused echocardiography and routine pre- and post-implant invasive assessment did not find evidence of LVOT obstruction.
This is a single-center study with a small cohort of patients. Each case was carefully reviewed to ensure clinical and technical suitability. The 30-day outcomes are reported, and longer-term clinical and valve durability outcomes will require further study.
Percutaneous transseptal transcatheter mitral valve replacement is feasible and may offer a relatively safe option in patients with MR who are at high risk for surgery. Early safety and efficacy are acceptable, and further study in a larger cohort of patients is warranted.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In the first-in-human trial of percutaneous transseptal mitral valve replacement in 10 patients with severe MR of various types at high surgical risk, technical success was achieved in all but 1 case, and patients were safe at 30 days.
TRANSLATIONAL OUTLOOK: Further studies are required to determine the long-term safety and efficacy of percutaneous mitral valve replacement.
Drs. Webb, Blanke, Leipsic, Wood, and Ye are consultants to and have received research support from Edwards Lifesciences. Dr. Moss has received consultancy payments and travel assistance from Edwards Lifesciences. Dr. Blanke has served as a consultant for Neovasc, Tendyne, and Circle Cardiovascular Imaging. Dr. Cheung is a consultant to Abbott and Medtronic. Dr. Leipsic has received support through his institutional core laboratory from Edwards, Medtronic, Abbott, and Neovasc; and has served as a consultant for and has stock options in CIRCL CVI and Heartflow. Dr. Tartara is an employee of and has stock ownership in Edwards Lifesciences. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- left ventricular outflow tract
- mitral regurgitation
- transcatheter heart valve
- transcatheter mitral valve replacement
- Received September 10, 2018.
- Revision received November 14, 2018.
- Accepted December 10, 2018.
- 2019 American College of Cardiology Foundation
- Stewart M.H.,
- Jenkins J.S.
- Beigel R.,
- Wunderlich N.C.,
- Kar S.,
- Siegel R.J.
- Grasso C.,
- Popolo Rubbio A.,
- Capodanno D.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Yoon S.H.,
- Whisenant B.K.,
- Bleiziffer S.,
- et al.
- Cheung A.,
- Webb J.G.,
- Barbanti M.,
- et al.
- Stone G.W.,
- Vahanian A.S.,
- Adams D.H.,
- et al.
- Stone G.W.,
- Adams D.H.,
- Abraham W.T.,
- et al.