Author + information
- Received October 17, 2017
- Revision received October 24, 2017
- Accepted October 24, 2017
- Published online January 1, 2018.
- Vinayak Bapat, MBBS, MS, MCha,b,
- Vivek Rajagopal, MDc,
- Christopher Meduri, MD, MPHc,
- R. Saeid Farivar, MDd,
- Antony Walton, MDe,
- Stephen J. Duffy, MBBS, PhDe,
- Robert Gooley, MBBS, PhDf,
- Aubrey Almeida, MDf,
- Michael J. Reardon, MDg,
- Neal S. Kleiman, MDg,
- Konstantinos Spargias, MDh,
- Stratis Pattakos, MDh,
- Martin K. Ng, MBBS, PhDi,
- Michael Wilson, MDi,
- David H. Adams, MDj,
- Martin Leon, MDb,
- Michael J. Mack, MDk,
- Sharla Chenoweth, MSl,
- Paul Sorajja, MDd,∗ (, )
- for the Intrepid Global Pilot Study Investigators∗
- aSt. Thomas’ Hospital, London, United Kingdom
- bNew York Presbyterian/Columbia University Medical Center, New York, New York
- cPiedmont Heart Institute, Atlanta, Georgia
- dAbbott Northwestern Hospital, Minneapolis, Minnesota
- eCardiology Department, The Alfred, Melbourne, Australia
- fMonash Heart, Melbourne, Australia
- gHouston Methodist DeBakey Heart and Vascular Center, The Methodist Hospital, Houston, Texas
- hHygeia Hospital, Athens, Greece
- iRoyal Prince Alfred Hospital, Sydney, Australia
- jMount Sinai Medical Center, New York, New York
- kBaylor Scott & White Health, Plano, Texas
- lMedtronic, Minneapolis, Minnesota
- ↵∗Address for correspondence:
Dr. Paul Sorajja, Valve Science Center, Minneapolis Heart Institute Foundation, Abbott Northwestern Hospital, 800 East 28th Street, Minneapolis, Minnesota 55401.
Background Transcatheter mitral valve replacement (TMVR) is a potential therapy for patients with symptomatic, severe mitral regurgitation (MR). The feasibility of this therapy remains to be defined.
Objectives The authors report their early experience with TMVR using a new valve system.
Methods The valve is a self-expanding, nitinol valve with bovine pericardial leaflets that is placed using a transapical delivery system. Patients with symptomatic MR who were deemed high or extreme risk by the local heart teams were enrolled in a global pilot study at 14 sites (United States, Australia, and Europe).
Results Fifty consecutively enrolled patients (mean age: 73 ± 9 years; 58.0% men; 84% secondary MR) underwent TMVR with the valve. The mean Society for Thoracic Surgery score was 6.4 ± 5.5%; 86% of patients were New York Heart Association functional class III or IV, and the mean left ventricular ejection fraction was 43 ± 12%. Device implant was successful in 48 patients with a median deployment time of 14 min (interquartile range: 12 to 17 min). The 30-day mortality was 14%, with no disabling strokes, or repeat interventions. Median follow-up was 173 days (interquartile range: 54 to 342 days). At latest follow-up, echocardiography confirmed mild or no residual MR in all patients who received implants. Improvements in symptom class (79% in New York Heart Association functional class I or II at follow-up; p < 0.0001 vs. baseline) and Minnesota Heart Failure Questionnaire scores (56.2 ± 26.8 vs. 31.7 ± 22.1; p = 0.011) were observed.
Conclusions TMVR with the valve was feasible in a study group at high or extreme risk for conventional mitral valve replacement. These results inform trial design of TMVR in lower-risk patients with severe mitral valve regurgitation (Evaluation of the Safety and Performance of the Twelve Intrepid Transcatheter Mitral Valve Replacement System in High Risk Patients with Severe, Symptomatic Mitral Regurgitation – The Twelve Intrepid TMVR Pilot Study; NCT02322840)
Mitral regurgitation (MR) is the most common valvular lesion in the Western world (1). When left untreated, severe MR results in chronic volume overload and progressive dilatation of the left ventricle, leading to severe heart failure and impaired longevity (2–4). Mitral valve surgery remains the standard of care for patients with symptomatic severe mitral valve regurgitation. If mitral valve repair is not feasible then the valve is replaced with a mechanical or bioprosthetic heart valve.
Transcatheter aortic valve replacement is established as a standard of care in intermediate- or greater- risk patients with aortic stenosis (5,6). In contrast, transcatheter mitral valve replacement (TMVR) is an emerging therapy that may offer patients with severe symptomatic MR a less invasive alternative to open surgical treatment.
We prospectively examined the safety and function of a novel, self-expanding valve (Intrepid TMVR System, Medtronic, Inc., Redwood City, California) in patients with symptomatic MR that were deemed to be ineligible for conventional valve surgery.
Patients were recruited at 14 hospitals in Australia, Europe, and the United States (Online Table 1). Key inclusion criteria were age >18 years, symptomatic, severe MR, no or minimal mitral valve calcification, and left ventricular ejection fraction ≥20%. Key exclusion criteria were severe pulmonary hypertension (i.e., systolic pressure ≥70 mm Hg), need for coronary revascularization, hemodynamic instability, need for other surgical valvular therapy, severe renal insufficiency (serum creatinine >2.5 mg/dl), and prior mitral valve surgery or intervention. Online Table 2 contains complete enrollment criteria. A local heart team deemed all patients to be high or extreme risk for open mitral valve surgery. Institutional review board approval was obtained in all centers, and patients provided informed consent for study participation.
Patients underwent comprehensive imaging with transesophageal echocardiography (TEE) and cardiac contrast-enhanced, gated computed tomography (CT) to determine suitability of treatment with TMVR. Patient data were reviewed by an independent physician committee, leading to approval for study participation.
The TMVR system
The new TMVR system is composed of a self-expanding, nitinol valve and a transapical delivery system (Figure 1) (7). A circular outer fixation frame (43, 46, or 50 mm diameter) engages the dynamic mitral valve anatomy. A circular inner stent frame (27 mm) houses a tri-leaflet bovine valve. A flexible atrial brim is attached to the outer fixation frame to facilitate visualization under ultrasound during implantation. Fixation is achieved through the oversizing in combination with unique design features of the outer frame to wedge the valve in the subannular mitral space. The outer frame has a flexible atrial portion, which allows it to conform to the native annulus, and a stiffer ventricular portion wider than the native mitral annulus. Additionally, there are small cleats on the outer frame that act as frictional members to engage the native mitral leaflets. The dual-frame structure ensures that the inner frame remains circular throughout the cardiac cycle, independent of the shape and motion of the outer frame and mitral annulus. The conformable and symmetric design eliminates the need for rotational orientation and simplifies device implantation. The device profile is 17 to 18 mm, and thus helps minimize risk of left ventricular outflow tract (LVOT) obstruction. The TMVR valve’s delivery system is currently designed for transapical delivery using a 35-F catheter access sheath and a hydraulically actuated delivery catheter.
Using previously described methods, contrast-enhanced, cardiac CT of the mitral valve annulus is performed for choice of the prosthesis size (8,9). Prosthesis size that allows 10% to 30% oversizing in mitral annular perimeter, intercommissural diameter, and septal-lateral diameter, while minimizing risk of LVOT obstruction was chosen. In general, a predicted neo-LVOT of <1.3 cm2 was required for proceeding with device implantation.
The procedure is performed under general anesthesia with guidance by TEE and fluoroscopy (Figure 2, Online Videos 1, 2, and 3). A small left thoracotomy is performed, and guided by data from the pre-procedural cardiac CT. Left ventricular apical purse string sutures are placed at the optimal access site. The device’s access sheath is introduced over a guidewire, into the left ventricle. Following insertion of the delivery catheter into the access sheath, the delivery catheter is advanced into the left atrium using TEE guidance. The atrial brim is expanded using a hydraulic delivery mechanism, with confirmation on fluoroscopy. Under 2-dimensional and 3-dimensional TEE imaging, the brim is visualized to align the valve with the mitral annulus. A target zone is identified, followed by retracting the catheter to the desired location, and deploying the valve using a short run of rapid ventricular pacing. The delivery system is then withdrawn from the left ventricle and the apical access site is closed.
Transthoracic echocardiography was obtained before hospital discharge in all patients. Post-procedure anticoagulation was instituted for a minimum of 3 months with warfarin (international normalized ratio target range: 2.5 to 3.5) and a single antiplatelet agent, consisting of either aspirin (75, 81, or 100 mg daily) or clopidogrel (75 mg daily).
Clinical follow-up, 6-min-walk testing, and echocardiography were performed at 1 month, 3 months, 6 months, and every 6 months thereafter. The Minnesota Living With Heart Failure Questionnaire was assessed at baseline and at 1 year.
Data analysis and endpoints
The analysis cohort includes the initial 50 consecutively enrolled patients who were approved for inclusion in the study (NCT02322840) between May 6, 2015, and July 21, 2017. All echocardiographic data were examined in a core echocardiography laboratory (Atif Qasim, MD, University of California, San Francisco, California). Grading of valvular and paravalvular MR was performed using standard criteria and classified into none, mild, moderate, or severe (10). Left ventricular ejection fraction was assessed with biplane Simpson method of discs using transthoracic echocardiography. Procedure time was defined as the duration from initial skin incision to final skin closure. Apical access time was defined as the duration from insertion of apical access needle to apical closure. Deployment time was defined as the duration from delivery catheter insertion into the left ventricle to completion of implantation of the valve.
The primary endpoint was assessment of the nature, severity, and frequency of complications associated with the delivery or implantation of the valve. An independent physician committee reviewed adverse clinical events for reporting including all-cause mortality, stroke, myocardial infarction, bleeding, cardiovascular rehospitalization, and reoperation. Standard definitions for clinical events reported were used in accordance with Mitral Valve Academic Research Consortium criteria (11).
Data are reported as medians with interquartile ranges (IQRs) or mean ± SD, unless otherwise noted. Adverse events are reported as the proportion of enrolled patients with an event. Survival at 1 year is reported as Kaplan-Meier estimates. The change from baseline to last follow-up for New York Heart Association (NYHA) functional class, Minnesota Living With Heart Failure Questionnaire, 6-min walk distance, and echocardiographic parameters were calculated using the Wilcoxon signed-rank test. All statistical analysis was performed using statistical software SAS version 9.4 (SAS Institute, Inc., Cary, North Carolina).
Role of the sponsor
The present investigation was made possible through financial and regulatory support of the sponsor (Medtronic Inc.). Data on patient demographics, medical history, echocardiographic findings, and clinical events were warehoused with the sponsor, who provided statistical analysis for this report.
Table 1 lists the baseline characteristics of the study population (mean age: 73 ± 9 years; 58% men). Debilitating heart failure was common, with 43 patients (86.0%) having NYHA functional class III or IV symptoms, and hospitalization for heart failure having occurred within the prior year in 29 patients (58%). Severe comorbidities were frequent; overall, the Society for Thoracic Surgery Predicted Risk of Mortality was 6.4 ± 5.5% and the EuroSCORE II was 7.9 ± 6.2%.
MR, as assessed by the echocardiographic core laboratory, was severe in 47 patients (95.9%), with the predominant mechanism classified as either secondary (n = 42; 84%) or primary (n = 8; 16%) MR. Two patients were treated initially for severe MR, and were subsequently determined to have moderate severity during further formal review. Overall, the mean left ventricular ejection fraction at baseline was 43 ± 12% (range: 20% to 70%); only 15 of 49 patients (30.6%) had a left ventricular ejection fraction >50%. Five patients had well-functioning aortic valve prostheses, and 22 patients (44.9%) had concomitant moderate or severe tricuspid regurgitation.
One patient did not undergo TMVR implantation because of site bleeding during apical access. Successful implantation of the valve occurred in 48 of 49 remaining patients (98.0%), with a median procedure time of 100 min (IQR: 80 to 124 min) (Online Table 3). The median time for device deployment was 14 min (IQR: 12 to 17 min). The only failure to implant successfully was related to sizing miscalculation and subsequent malpositioning of the valve. There were no incidences of device malfunction, device failure, or conversions to open cardiac surgery. In 5 patients, an intra-aortic balloon pump was placed, with 2 patients having prophylactic placement and 3 others having placement for hemodynamic management. One patient also had extracorporeal membrane oxygenation performed before device implantation because of evidence of pulmonary hypertension that had worsened significantly during induction of anesthesia. Two additional patients were placed on extracorporeal membrane oxygenation while conducting apical repairs for bleeding complications.
The median clinical follow-up duration for the entire cohort was 173 days (IQR: 54 to 342 days) (Central Illustration and Figure 3). There were 7 deaths (14.0%) within 30 days (Table 2, Online Table 4); 3 deaths were related to apical access site bleeding at or immediately after the initial procedure, 1 occurred in the patient with malposition, and 3 other patients died because of refractory heart failure early after the procedure (<30 days). Four additional patients died after 30 days (between days 54 and 122) but there were no deaths after 4 months (Figure 3); 3 of these late deaths were caused by sudden cardiac arrest and 1 by intracranial hemorrhage in the setting of an unwitnessed fall. Among these 4 late deaths, the absence of structural valve degeneration was documented either at autopsy or before death with echocardiography within the prior month. Other adverse events are listed in Table 2.
In 5 patients, reoperation was performed for bleeding observed in the immediate post-operative period. In 1 instance, minor bleeding was observed at the apical site and was controlled with a single stitch; in 4 other patients, chest wall bleeding was identified.
Among all alive patients with 30-day echocardiographic follow-up (n = 42), significant reductions in MR severity and improvements in symptoms occurred (Figure 4). MR was either absent (i.e., grade 0) or mild (i.e., grade 1) at the last clinical evaluation in patients who received implants (n = 42; 100%), with no evidence of LVOT obstruction (change in peak LVOT gradient, 0.4 ± 4.5 mm Hg) or mitral stenosis (final mean gradient: 4.1 ± 1.3 mm Hg) (Online Table 5). Among patients with MR, all cases were mild; paraprosthetic MR occurred in 3 patients (7.1%), whereas prosthetic MR occurred in 8 patients (19.0%). For patients with paired data, pulmonary artery systolic pressure, by transthoracic echocardiography, decreased from 46.7 ± 14.7 mm Hg to 37.2 ± 9.7 mm Hg at follow-up (p < 0.0001). There also was a decrease in left ventricular ejection fraction (43.6 ± 12.1% vs. 36.2 ± 10.2%; p < 0.0001), and an increase in left ventricular end-systolic dimension (4.8 ± 1.0 cm at baseline vs. 5.1 ± 0.9 cm at follow-up; p = 0.0007), without significant changes in left ventricular end-diastolic dimension (5.9 ± 0.7 cm at baseline vs. 5.9 ± 0.8 cm at follow-up; p = 0.94). In all implanted patients, there were no instances of hemolysis, device embolization, or thrombosis.
Among the patients with NYHA functional class follow-up at 30 days (n = 42), mild or no symptoms of heart failure (i.e., NYHA functional class I or II) were present in 79% (p < 0.0001 vs. baseline). During follow-up, a nonsignificant change in 6-min walk distance was observed (248.3 ± 103.6 m vs. 267.6 ± 110.0 m; p = 0.31). Significant improvements in Minnesota Living With Heart Failure Questionnaire scores (56.2 ± 26.8 vs. 31.7 ± 22.1; p = 0.011) occurred.
The principal findings of this pilot study were: 1) the valve was successfully implanted in 48 of 49 patients (98%), with reduction in MR to mild or none in all patients who received implants; 2) the procedure was short and reproducible with a minimal learning curve, with no device malfunction or structural valve degeneration occurring acutely or during clinical follow-up (median: 173 days); and; 3) improvements in symptom status and quality of life were observed in procedural survivors. These observations in a cohort of patients deemed to be ineligible for conventional open mitral valve surgery are provocative (Central Illustration).
The valve design was shown to be advantageous. Implantation and fixation for anchoring was self-centering and symmetric, and the separation of the outer anchoring stent from the inner stent housing the functional valve components proved to be adaptable to a variety of anatomic situations. The lower profile of the valve also allowed implantation in previously contraindicated conditions, including prior prosthetic aortic valve replacement and patients, particularly women, with smaller ventricular size and who may be at risk for LVOT obstruction (9).
We targeted a high- or extreme-risk study group of patients who were deemed to be ineligible for open mitral valve surgery, which may help to explain the high 30-day mortality. Apical bleeding was a concerning issue and was associated with mortality in 3 patients, emphasizing the need for diligent technical care and willingness to abort the procedure if the access site is believed to be unsafe. Three patients died because of exacerbation of heart failure suggesting the need for more intensive perioperative heart failure management. This experience highlights the recognition that patients undergoing TMVR may experience a complex recovery period and the need for careful patient selection. We were surprised by the high incidence of reoperation for chest wall bleeding, although this may have been partially caused by overly aggressive institution of anticoagulant therapies. Emphasis on chest wall hemostasis will be important, because this technology currently relies on a mini-thoracotomy for transapical access. Ongoing development of the new TMVR system for transfemoral, transseptal implantation, avoiding the need for anterior thoracotomy and resultant apical access complications, will likely further lower the morbidity of this technology in the future.
Our results support the further exploration of image-guided beating heart TMVR as an alternative to open heart mitral valve surgery in appropriately selected patients. There was a decrement in left ventricular ejection fraction (44% to 36%) following implantation of the prosthesis, similar to what has been reported in other TMVR series (12). In the CTSN trial of severe ischemic MR, whose study group shares some characteristics with ours, a decrement in left ventricular ejection fraction was minimally present, if at all, in the chordal-replacement arm (40.0% at baseline vs. 37.8% at 1-year follow-up) and the patients who had repair (42.4% at baseline vs. 41.5% at 1-year follow-up). Further evaluation of the significance of such change in ejection fraction among patients who receive TMVR, which is a chordal-sparing treatment, is needed (13). Of note, in the 3 sudden deaths that occurred in follow-up despite successful TMVR, none of the patients had been treated with an implantable cardioverter-defibrillator despite having clinical indications for the therapy. In comparison with the most recent registry data regarding transcatheter mitral valve repair in patients with functional MR that reported a relatively high rate of moderate or severe MR at 1 year (14), TMVR with the valve was found to be highly effective in eliminating significant mitral valve regurgitation. Although the results of our initial feasibility trial demonstrate the promise of TMVR with the novel valve system, further clinical investigations are required to define other patient populations with severe MR that may benefit from this procedure.
The present investigation is an examination of the feasibility of a novel prosthesis for TMVR, which is a rapidly evolving field. Although there were no instances of known device thrombosis, prospective screening with adjunctive multimodality imaging (e.g., TEE or CT) was not performed and the incidence of subclinical abnormalities is not known. Measurement of MR severity was performed with transthoracic echocardiography in the resting state with no attempt to alter the hemodynamic milieu (e.g., raising or lowering blood pressure), although data on the effects of any subtle hemodynamic alterations on MR severity were not available. In addition, anticoagulation with warfarin was recommended for >3 months, but the rate of continuation or discontinuation was unknown. The target population for the study was high- or extreme-risk patients (Online Table 6). Extenuating circumstances (e.g., frailty) were used to justify therapeutic intervention with TMVR, yet formal testing or objective measurement of such variables was not consistently performed or available. Such information would be of benefit for optimizing patient selection for TMVR therapy.
TMVR with the new TMVR system resulted in the correction of MR in symptomatic patients deemed to be at high or extreme risk for open heart surgery. Stable valve function was observed longitudinally, and most patients experienced significant improvement in their clinical symptoms and functional class. Further investigations will determine the role of this therapeutic option in a broader population of patients with mitral valve disease compared with both surgical mitral valve replacement and transcatheter mitral valve repair techniques.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In a pilot study, TMVR with the valve system reduced or corrected mitral regurgitation resulting in symptomatic improvement in patients at high or extreme surgical risk.
TRANSLATIONAL OUTLOOK: Future studies are needed to better define the indications for TMVR for treatment of mitral regurgitation.
The authors thank Gan Dunnington, MD, David Lee, MD, David Liang, MD, Jack Boyd, MD, Neil Schwartz, MD, and Ronald Witteles, MD, for expert patient review and study oversight. They also thank the investigators from John Paul II Hospital in Krakow, Poland (Boguslaw Kapelak, MD, Jerzy Sadowski, MD, Krzysztof Bartus, MD and Andrzej Gackowski, MD) for their early work in performing the first-in-human transcatheter mitral valve replacement cases and providing invaluable experience to move the therapy forward. Colleen Gilbert, PharmD, and Jessica Dries-Devlin, PhD, provided editorial support including creation of tables and figures and manuscript formatting. Both are employees of Medtronic.
The Intrepid Global Pilot Study was funded by Medtronic (Minneapolis, Minnesota). Dr. Bapat has received personal fees for consultancy and speaker service from Medtronic. Dr. Rajagopal has received personal fees for speaking from Medtronic, Boston Scientific, and Abbott Vascular; and is on the screening committee for the APOLLO trial sponsored by Medtronic. Dr. Meduri has received grant support paid to his institution from Medtronic; and is a consultant to Boston Scientific and Mitralign. Dr. Walton has received proctor and advisory board fees from Medtronic unrelated to the submitted work. Dr. Duffy has received research support paid to his institution from Medtronic; and receives proctor fees from Medtronic unrelated to the submitted work. Dr. Gooley has received proctor fees from Medtronic unrelated to the submitted work. Dr. Reardon has received personal fees from Medtronic and Boston Scientific outside the submitted work. Dr. Kleiman has received proctoring fees from Medtronic. Dr. Spargias has received grant support and personal fees for travel from Medtronic. Dr. Pattakos has received grant support from Medtronic and Edwards. Dr. Ng reports fees paid to his institution for research participation. Dr. Adams has received grant support from Medtronic; and royalty agreements through Mount Sinai School of Medicine with Medtronic and Edwards Lifesciences. Dr. Mack is on the Executive Board for the APOLLO trial sponsored by Medtronic; and is coprincipal investigator for the Partner 3 and COAPT trials sponsored by Edwards and Abbott Vascular. Ms. Chenoweth is an employee of Medtronic. Dr. Sorajja has received grant support from Medtronic for participation in the Steering Committee of the APOLLO trial. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Deepak L. Bhatt, MD, MPH, and Ted Feldman, MD, served as Guest Editors.
- Abbreviations and Acronyms
- computed tomography
- interquartile range
- left ventricular outflow tract
- mitral regurgitation
- New York Heart Association
- transesophageal echocardiogram
- transcatheter mitral valve replacement
- Received October 17, 2017.
- Revision received October 24, 2017.
- Accepted October 24, 2017.
- 2018 American College of Cardiology Foundation
- Nkomo V.T.,
- Gardin J.M.,
- Skelton T.N.,
- et al.
- Grigioni F.,
- Enriquez-Sarano M.,
- Zehr K.J.,
- Bailey K.R.,
- Tajik A.J.
- Grigioni F.,
- Detaint D.,
- Avierinos J.F.,
- et al.
- Baumgartner H.,
- Falk V.,
- Bax J.J.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Meredith I.,
- Bapat V.,
- Morriss J.,
- McLean M.,
- Prendergast B.
- Blanke P.,
- Divr D.,
- Cheung A.,
- et al.
- Blanke P.,
- Naoum C.,
- Dvir D.,
- et al.
- Zoghbi W.A.,
- Adams D.,
- Bonow R.O.,
- et al.
- Stone G.W.,
- Adams D.H.,
- Abraham W.T.,
- et al.
- Muller D.W.M.,
- Farivar R.S.,
- Jansz P.,
- et al.,
- Tendyne Global Feasibility Trial Investigators
- Sorajja P.,
- Vemulapalli S.,
- Feldman T.,
- et al.