Transapical Beating-Heart Mitral Valve Repair With an Expanded Polytetrafluoroethylene Cordal Implantation DeviceClinical Perspective
Initial Clinical Experience
Background: Degenerative mitral valve (MV) disease is a common cause of severe mitral regurgitation (MR) and accounts for the majority of MV operations. Conventional MV surgery requires cardiopulmonary bypass, aortic cross-clamping, cardioplegia, and a thoracotomy or sternotomy and, therefore, is associated with significant disability, risks, and unpredictable rates of MV repair. Transesophageal echocardiography–guided beating-heart MV repair with expanded polytetrafluoroethylene cordal insertion has the potential to significantly reduce surgical morbidity. We report the first-in-human clinical experience with a novel preformed expanded polytetrafluoroethylene knot implantation device (Harpoon TSD-5) designed to treat degenerative MR.
Methods: Through a small left thoracotomy, the device was inserted into the heart and guided by transesophageal echocardiography to the ventricular surface of the prolapsed leaflet. Multiple expanded polytetrafluoroethylene cords were anchored in the leaflet and then adjusted to the correct length to restore MV leaflet coaptation and secured at the epicardium.
Results: Eleven patients with posterior leaflet prolapse and severe MR, with mean±SD age of 65±13 years and mean ejection fraction of 69±7%, were treated with 100% procedural success. Immediate postprocedural mean MR grade was trace. At 1 month, the mean MR grade was mild with significant decreases in end-diastolic volume (139 to 107 mL; P=0.03) and left atrial volume (118 to 85 mL; P=0.04).
Conclusions: A novel device used for beating-heart image-guided MV repair demonstrates a significant reduction in MR with favorable left ventricular and left atrial reverse remodeling. This approach has the potential to decrease invasiveness and surgical morbidity. Further follow-up is necessary to assess long-term efficacy.
Editorial, see p 198
Mitral valve disease is the most prevalent heart valve disorder and is 3 to 4 times more common than aortic valve disease. Significant mitral regurgitation (MR) affects ≈2% of the total population and 9% of the population >75 years of age.1 MR causes volume overload of the left ventricle (LV), which in turn can progress to ventricular dilation, decreased ejection performance, pulmonary hypertension, congestive heart failure, atrial fibrillation (AF), right ventricular dysfunction, and eventually death. Degenerative MR is the most common indication for mitral valve surgery in North America.2 Introduced clinically in 1986, mitral valve repair with expanded polytetrafluoroethylene (ePTFE) artificial cords has been shown to be equivalent to leaflet resection techniques3 and is increasingly being used in conventional mitral valve repair operations.4 Although effective, conventional mitral valve surgery requires cardiopulmonary bypass, aortic cross-clamping, cardioplegia, and a thoracotomy or sternotomy. It is therefore associated with significant disability and risks, including stroke, renal failure, widely variable rates of mitral valve repair versus replacement, and death.5 As a result, there is intense interest in the development of less invasive therapies for the treatment of MR. In this report, we describe the initial clinical experience with the safety and performance of a device designed to implant ePTFE artificial cords and to correct degenerative MR in the beating heart with echocardiographic image guidance.
Study Design and Oversight
We conducted this prospective, observational, early-feasibility trial to test the safety and performance of a novel ePTFE insertion device for mitral valve repair. The trial was conducted at 2 clinical centers (Jagiellonian University John Paul II Hospital, Krakow, and Institute of Cardiology, Warsaw, Poland) and was sponsored by Harpoon Medical Inc. The protocol was approved by the ethics committee at each participating institution and received central Polish Ministry of Health approval. All patients provided written informed consent before enrollment.
The authors had full access to the data and take full responsibility for the completeness and accuracy of the data and the analyses reported in this article.
Patients with severe degenerative MR resulting from isolated posterior leaflet prolapse were enrolled. Patients were selected if the predicted post-ePTFE artificial cord implantation coaptation surface was adequate to result in effective MR reduction in the judgment of the operating surgeon and the clinical trial team. This was a qualitative assessment based on 2- and 3-dimensional transesophageal echocardiography (TEE) that there was sufficient posterior leaflet tissue area in the prolapsed segment to occlude the regurgitant orifice when repositioned apically with the ePTFE cords.
Patients were excluded if they had functional MR, a Society of Thoracic Surgeons predicted risk of mortality (for repair) of >6%, severe pulmonary hypertension (systolic pulmonary artery pressure >60 mm Hg), or severe LV dysfunction (ejection fraction <40%). Full inclusion and exclusion criteria are available at https://clinicaltrials.gov/ct2/show/NCT02432196.
Device Design and Implantation Procedure
The TSD-5 device (Harpoon Medical, Inc) is a 3-mm-diameter shafted instrument designed to anchor ePTFE cords on the prolapsed mitral valve leaflet. A small (2–3 cm) anterolateral left thoracotomy is performed in the fourth or fifth intercostal space. The pericardium is opened; heparin is administered to maintain an activated clotting time of >350 seconds; and concentric pledgeted purse-string sutures are placed at the insertion site on the epicardium ≈3 to 4 cm basal to the apex of the heart (at the level of the base of the papillary muscles) and just lateral to the left anterior descending coronary artery. A purpose-designed 14F valved introducer is inserted over a guidewire into the ventricle and secured. A TSD-5 is inserted into the introducer and, with TEE guidance (simultaneous orthogonal [x plane] views at midcommissural and long-axis planes), is steered to the ventricular surface of the prolapsed leaflet. The target site on the prolapsed leaflet is stabilized with the device by applying pressure to the ventricular side of the leaflet with the end-effector of the device. Once proper positioning is confirmed in the orthogonal TEE planes, the device is actuated, which results in perforation of the leaflet by a specially designed 21-gauge needle wrapped with 50 coils of ePTFE in a preformed knot configuration (Figure 1). There is no need to grasp or catch the moving leaflet, and sutures are not limited to the free edge, allowing the operator to anchor ePTFE cords anywhere on the leaflet.
As the needle is automatically withdrawn, a double-helix coiled ePTFE knot is formed on the atrial surface of the leaflet, securing the associated pair of ePTFE artificial cords to the leaflet. The TSD-5 device is withdrawn through the valved introducer, and the ePTFE cords are exteriorized through the introducer. Additional TSD-5 devices are deployed to anchor the desired number of ePTFE cordal pairs to the leaflet (typically 3 or 4 pairs). Once insertions are complete, the valved introducer is withdrawn and the purse-string suture is tied. The ePTFE cords emanating from the ventricle are passed through separate holes in a single Teflon pledget and with a tourniquet are simultaneously titrated to an optimal length (defined as maximal coaptation on echocardiography and absent MR on color Doppler interrogation) with TEE guidance. Each pair is then tied on the single pledget, and the procedure is completed. Movie I in the online-only Data Supplement demonstrates the TSD-5 ePTFE artificial cord implantation procedure. Aspirin (325 mg) is administered after the operation and daily thereafter.
Study End Points
The prespecified primary performance outcome was procedural success, defined as successful implantation of ≥1 ePTFE artificial cords on the mitral valve, and demonstration of MR reduction from severe to moderate or less at the conclusion of the procedure and at 30 days. The primary safety end point was freedom from serious adverse events during the ePTFE implantation procedure, at discharge, and at 30 days after the procedure. We assessed the severity of MR using intraoperative TEE and predismissal, 30-day, and 6-month transthoracic echocardiography. Perioperative mortality was defined as the greater of 30-day mortality or in-hospital mortality. Adverse events were defined according to the Society of Thoracic Surgeons Adult Cardiac Surgery Database definitions.2 This study represents an interim analysis of an ongoing early-feasibility study. The common date for all analyses was January 1, 2016. The mean length of follow-up was 186 days (range, 64–319 days).
Echocardiographic analyses were performed independently by a core laboratory (Massachusetts General Hospital). LV dimensions, left atrial volumes (biplane area length), and LV volumes were measured (biplane method of disks) per American Society of Echocardiography chamber quantification guidelines.6 The degree of MR was graded as none, trace, mild, moderate, or severe (corresponding to numerical grading of 0 to 4+, respectively). with the use of integrative criteria as specified by the American Society of Echocardiography.7 Septal-lateral mitral annular dimension was measured at end diastole in the apical 4-chamber view.
Baseline characteristics and clinical outcomes were described using counts and percentages for categorical variables and mean±SD, sometimes supported by ranges, for continuous measures. Statistical testing of pre-post echocardiography measurements was done with matched-pair t tests. Statistical significance was based on a significance level of P=0.05. Analyses were performed with JMP 8.0 statistical software (SAS Institute Inc).
From February 2015 through October 2015, 30 patients were screened and 11 consecutive patients were enrolled in the trial. All implanting surgeons were trained on a bench-top simulator and in an animal laboratory. The mean age of the patients was 65±13 years (range, 42–89 years), and 91% were men. The baseline characteristics of the patients who underwent beating-heart mitral valve repair are shown in Table 1. Most patients were low risk for conventional surgical mitral valve repair, and all but 1 patient had class 1 indications for mitral valve repair.8
There was a 100% procedural success rate. An average of 3.6±0.7 (range, 3–5) pairs of ePTFE artificial cords were implanted, with reduction of MR during the procedure from severe to none/trace in 8 patients and to mild in 3 patients. Total procedure time averaged 108±30 minutes (range, 72–167 minutes), and the total time that the introducer was in the ventricle averaged 38±14 minutes (range, 20–58 minutes). No patient required intraoperative inotropic or vasopressor support. There was no perioperative mortality and no intraoperative conversion to conventional cardiac surgery. There were no perioperative strokes, new-onset renal failure, postoperative AF, or myocardial infarction. No blood transfusions were required during the procedure or hospitalization. At 1 month, all 4 patients with preoperative paroxysmal AF were in sinus rhythm, and 1 patient with preoperative persistent AF remained in AF. There were 2 in-hospital adverse events: 2 patients required reoperation (subxiphoid window) for delayed pericardial effusion on postoperative days 5 and 13. In both cases, serosanginous fluid was evacuated (150 and 500 mL), and both patients recovered uneventfully and were discharged home in good condition.
One patient required conventional cardiac surgical operation on postoperative day 72 for recurrent symptomatic severe MR. Moderate MR was recognized on the predismissal echocardiogram. Recurrent severe MR was identified 60 days postoperatively and was associated with recurrence of dyspnea. At reoperation, all 3 ePTFE knots were intact and located on the atrial surface of the P2 segment of the posterior leaflet with evidence of early endothelialization (Figure 2). There was no evidence of leaflet damage. One of the 3 pairs of ePTFE artificial cords was found to have become untied from the epicardial apical pledget, and this end of the suture pair was free within the ventricle but remained anchored at the knot to the leaflet.
In addition, 1 native edge cord to A2 was ruptured. A quadrangular resection incorporating the P2 segment of the posterior leaflet and all 3 ePTFE knots was performed with insertion of a size 34 complete semirigid annuloplasty ring. An ePTFE cord was placed to A2. The patient was discharged from the hospital in good condition with no MR.
The severity of MR at the conclusion of the procedure, before dismissal, and at 30 days is reported in Table 2.
Clinical and echocardiographic follow-up was 100% complete. There was no evidence of mitral stenosis at 1 month after the procedure, with peak and mean transmitral gradients of 3.2 and 1.5 mm Hg, respectively. Four patients have had 6-month follow-up echocardiography. All are in New York Heart Association class I. In all cases, the ePTFE artificial cords and knots are in stable positions. Three patients have none/trace MR, and 1 patient has moderate MR.
Preoperative and 30-day echocardiographic measurements are reported in Table 3.
There was evidence of early ventricular remodeling, with a decrease in the end-diastolic dimension of 11% (54±3 to 48±5 mm), end-diastolic volumes of 18% (142±41 to 116±36 mL), and the septal-lateral dimension from 40.0 to 37.9 mm. There were no important differences in ejection fraction and no wall motion abnormalities after the procedure. There was favorable early remodeling of the left atrium with a decrease in anterior-posterior left atrial dimension and volumes. Figure 3 demonstrates representative preprocedural and postprocedural TEE images, and Figure 4 is a representative 6-month transthoracic follow-up study.
In this study, we describe the initial evaluation of the Harpoon ePTFE artificial cord implantation device in a first-in-human series of 11 consecutive patients with severe degenerative MR. The device was used to implant artificial ePTFE cords in the posterior leaflet of the beating heart with TEE image guidance. Procedural success occurred in all patients, with the insertion of between 3 and 5 pairs of ePTFE artificial cords resulting in a reduction in MR from severe to moderate or less in all patients at 30 days. MR reduction was stable in 3 of 4 patients followed up to 6 months.
Expanded ePTFE sutures were first described for use in surgical mitral valve repair by Zussa et al4 and introduced clinically by David.9 In a large series of 606 consecutive patients with degenerative MR repaired with ePTFE artificial cords with a mean clinical follow-up of 10.1 years, the freedom from recurrent severe MR at 18 years was 91% and the freedom from reoperation on the mitral valve was 90%.10 One randomized trial compared results of conventional resection techniques with ePTFE cordal replacement techniques and found equivalent early results, with evidence of greater coaptation surface in the ePTFE group.3 Recent studies have suggested that LV performance is superior in patients with degenerative MR repaired with ePTFE artificial cordal techniques compared with resection methods.11 In North America, there is increasing use of ePTFE in mitral valve repair: Between 2011 and 2014, among 20 523 patients undergoing mitral valve repair for degenerative disease, ePTFE cordal replacement was used in 31% (6370).2
We reasoned that a shafted instrument with a narrow diameter that could reliably anchor ePTFE artificial cords to a prolapsed mitral valve leaflet with the use of transapical access and TEE guidance would enable reliable and durable mitral valve repair through a small nonsternotomy incision on the beating heart. We found that the Harpoon TSD-5 device performed as designed and enabled the operator to anchor ePTFE artificial cords in precise targeted locations on the prolapsed segment of the mitral valve. The procedure was efficient, with an average procedural time of 108±30 minutes and an introducer dwell time of 38±14 minutes, and was characterized by complete hemodynamic stability. Transapical mitral valve repair with a device designed to grasp the free edge of the prolapsed mitral valve leaflet and secure an ePTFE cord (Neochord, Inc, Minneapolis, MN) has previously been reported, with intermediate-term results from patients treated after an initial learning curve suggesting a strong safety profile and an effective reduction in MR.12 Compared with the Neochord device, the Harpoon device is characterized by a smaller-diameter shaft (3 versus 8 mm), a valved introducer to minimize intraprocedural bleeding, the ability to insert the ePTFE cords anywhere on the leaflet rather than within 4 mm of the free edge, and a fundamentally different anchoring mechanism.
One challenge of conventional mitral valve repair surgery with ePTFE artificial cords is correct determination of the optimal length of the ePTFE cords on the open and arrested heart. Beating-heart image-guided transapical mitral valve repair with the Harpoon TSD-5 device enabled the surgeon to precisely titrate the length of the ePTFE artificial cords in real time on the beating, working heart. In all cases, the operator was able to tighten or loosen the ePTFE cords to maximize leaflet coaptation and to minimize MR. The only intracardiac implant in this procedure is the ePTFE suture and knot, analogous to conventional ePTFE-based surgical repair, which has a long clinical record of safety and effectiveness. Preclinical testing with fresh human autopsy hearts has demonstrated that the Harpoon preformed knot has anchoring forces equivalent to conventionally sutured ePTFE.
One patient in this initial clinical experience required reoperation for recurrent MR. The cause of recurrence was multifactorial and included unfastening and detachment of 1 pair of ePTFE cords from the pledget on the epicardium; an unrecognized and untreated anterior leaflet, which prolapsed 1 to 2 mm above the plane of the mitral annulus; and progression of disease with rupture of a native anterior leaflet edge cord. Improved surgical training, better methods of securing the ePTFE cords to the epicardial pledget, deployment of ePTFE cords on the anterior leaflet, and increased experience with patient selection based on analysis of reconstructed 3-dimensional TEE data sets should all contribute to minimizing the likelihood of this type of failure in future patients.
It is unlikely that ePTFE knots placed on the mitral leaflet with the Harpoon ePTFE cordal implantation device will compromise subsequent surgical mitral valve repair. We found that to be the case in this patient, in whom at reoperation the ePTFE knots did not appreciably change the structure of the mitral valve leaflet at 2 months after implantation and a resectional repair was successful. Although in this case the surgeon chose to perform a resectional repair, a nonresectional repair would have been equally feasible. This is in contrast to experience with the MitraClip device, in which a fibrous reaction to the clip commonly precludes subsequent successful surgical repair in most patients.13
MR reduction was excellent in all cases, with 30-day MR grading of mild or less in all but 2 patients, 1 patient who required reoperation and another with evidence of modest reprolapse of the posterior leaflet. In the patient who developed moderate MR by the 6-month follow-up (the second patient ever treated with the device), we identified progressive prolapse of a lateral segment of P2 that was not targeted for ePTFE resuspension during the initial procedure. Now that we have gained experience, in retrospect, we would have specifically targeted this segment for ePTFE cord implantation. In the remaining 3 patients with 6 month follow-up, the posterior leaflet position was stable with a generous surface of coaptation. As expected after successful surgical mitral valve repair, the MR reduction was associated with favorable LV, mitral annular, and left atrial remodeling, with decreases in LV dimensions and volumes, septal-lateral mitral annulus dimension, and left atrial volumes.
This initial clinical experience supports the conclusion that surgical-grade MR reduction is possible with the device and that the repair is durable at the 1-month follow-up. We do not believe that annular dilation per se mandates annular downsizing with an annuloplasty ring but rather that the key predictor of successful mitral valve repair is the ratio of total leaflet tissue to mitral orifice area, which if adequate allows sufficient coaptation. The ePTFE artificial cords placed in this experience are not implanted on the tip of the papillary muscles as in conventional mitral valve surgery but rather are inserted through the myocardium on the anterior surface of the LV in proximity to the left anterior descending coronary artery. The resulting force vectors on the leaflet therefore include both apical and anterior forces, the latter of which serves to decrease and stabilize the septal-lateral (anterior-posterior) dimension of the mitral annulus.6 Reduction of MR with the Harpoon TSD-5 resulted in modest but significant early reductions in septal-lateral dimensions as a component of postprocedural LV remodeling.
Mitral valve repair is clearly superior to replacement for degenerative disease, with an operative risk of approximately half compared with replacement, improved LV function, and superior freedom from reoperation, structural valvular deterioration, thromboembolism, anticoagulation-related complications, and infective endocarditis.14,15 Current American College of Cardiology/American Heart Association guidelines indicate mitral valve repair in preference to replacement for degenerative disease (Class I recommendation) and prohibit (Class III: harm) mitral replacement for severe primary MR limited to less than one half of the posterior leaflet.16 Despite clear evidence of the short- and long-term benefits of mitral valve repair for degenerative MR, at present, mitral valve repair rates are between 60% and 85% in North America, and there is substantial variability based on center and surgeon experience, reflecting the complexity of mitral valve repair for degenerative disease. As a result of real-time echocardiographic titration of ePTFE cordal lengths on the loaded working heart, image-guided beating-heart mitral valve repair for degenerative disease with the Harpoon TSD-5 has the potential to increase mitral valve repair rates.
As expected, transapical beating-heart mitral valve repair had an excellent safety profile, with no perioperative mortality and little morbidity. We believe that compared with conventional mitral valve surgery, beating-heart mitral valve repair will be associated with a shorter hospital length of stay, substantially less resource use, and faster return to preoperative functional status.
The Harpoon TSD-5 device is suitable for both low-risk degenerative MR patients (those currently referred for surgical therapy) and high- and extreme-risk patients who currently are not candidates for conventional surgery. We anticipate continued procedural refinement and, as these learnings are implemented, continued improvement in clinical results.
This trial did not have a control arm of patients undergoing conventional mitral valve repair surgery; thus, we cannot make definitive statements comparing this approach with conventional mitral valve operation. This report represents an interim analysis of an ongoing study. This initial series was limited to patients with posterior mitral leaflet prolapse only. However, as additional experience is achieved, we plan on broadening this approach to bileaflet prolapse and complex mitral degenerative pathology.
In this first-in-human experience, the Harpoon TSD-5 enabled less invasive beating-heart image-guided transapical mitral valve repair with 100% procedural success and safe and reliable reduction in MR. This initial experience will require follow-up to ensure durability and effectiveness in larger numbers of patients.
We are grateful for the support and input of Jolanta Rzucidlo-Resil, MD, Radoslaw Litwinowicz, MD, and Anton Chrustowicz, MD.
Sources of Funding
Funding for this study was provided by Harpoon Medical, Inc.
Drs Gammie, D’Ambra, and Bartus and Mr. Wilson have stock in and/or options to purchase stock in Harpoon Medical, Inc. Dr Ghoreishi is a coinventor of the technology, which has been licensed from the University of Maryland by Harpoon Medical Inc. The other authors report no conflicts.
Sources of Funding, see page 196
Circulation is available at http://circ.ahajournals.org.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.116.022010/-/DC1.
- Received February 12, 2016.
- Accepted June 13, 2016.
- © 2016 American Heart Association, Inc.
- 2.↵STS Adult Cardiac Surgery National Database. Circulation. 2014;129:2440–2492.doi: 10.1161/CIR.0000000000000029.
- Falk V,
- Seeburger J,
- Czesla M,
- Borger MA,
- Willige J,
- Kuntze T,
- Doll N,
- Borger F,
- Perrier P,
- Mohr FW
- Lang RM,
- Bierig M,
- Devereux RB,
- Flachskampf FA,
- Foster E,
- Pellikka PA,
- Picard MH,
- Roman MJ,
- Seward J,
- Shanewise JS,
- Solomon SD,
- Spencer KT,
- Sutton MS,
- Stewart WJ
- Zoghbi WA,
- Enriquez-Sarano M,
- Foster E,
- Grayburn PA,
- Kraft CD,
- Levine RA,
- Nihoyannopoulos P,
- Otto CM,
- Quinones MA,
- Rakowski H,
- Stewart WJ,
- Waggoner A,
- Weissman NJ
- Nishimura RA,
- Otto CM,
- Bonow RO,
- Carabello BA,
- Erwin JP 3rd.,
- Guyton RA,
- O’Gara PT,
- Ruiz CE,
- Skubas NJ,
- Sorajja P,
- Sundt TM 3rd.,
- Thomas JD
- Russo A,
- Grigioni F,
- Avierinos JF,
- Freeman WK,
- Suri R,
- Michelena H,
- Brown R,
- Sundt TM,
- Enriquez-Sarano M
- Wang A,
- Athan E,
- Pappas PA,
- Fowler VG Jr.,
- Olaison L,
- Paré C,
- Almirante B,
- Muñoz P,
- Rizzi M,
- Naber C,
- Logar M,
- Tattevin P,
- Iarussi DL,
- Selton-Suty C,
- Jones SB,
- Casabé J,
- Morris A,
- Corey GR,
- Cabell CH
- Nishimura RA,
- Otto CM,
- Bonow RO,
- Carabello BA,
- Erwin JP 3rd.,
- Guyton RA,
- O’Gara PT,
- Ruiz CE,
- Skubas NJ,
- Sorajja P,
- Sundt TM 3rd.,
- Thomas JD
What Is New?
This report describes an initial clinical experience with a new device designed to implant expanded polytetrafluoroethylene artificial cords on the prolapsed mitral valve and to correct degenerative mitral regurgitation in the beating heart with the use of transesophageal echocardiographic image guidance.
The main findings of this study were that in an initial 11-patient clinical experience, surgeons used the device to successfully implant multiple expanded polytetrafluoroethylene artificial cords and to effectively reduce mitral regurgitation in all patients.
The results were stable at 30 days, and the procedure was safe, with no acute conversions to open heart surgery, blood transfusions, stroke, or death.
What Are the Clinical Implications?
This study suggests that the Harpoon TSD-5 has the potential to facilitate less invasive beating-heart therapy for degenerative mitral regurgitation.
This initial experience will require follow-up to ensure durability and effectiveness in larger numbers of patients.