Author + information
- Received May 3, 2010
- Revision received August 11, 2010
- Accepted August 20, 2010
- Published online January 1, 2011.
- Jan Harnek, MD⁎,⁎ (, )
- John G. Webb, MD‡,
- Karl-Heinz Kuck, MD¶,
- Carsten Tschope, MD#,
- Alec Vahanian, MD⁎⁎,
- Christopher E. Buller, MD§,
- Stefan K. James, MD†,
- Christiane P. Tiefenbacher, MD∥ and
- Gregg W. Stone, MD††
- ↵⁎Reprint requests and correspondence:
Dr. Jan Harnek, Department of Coronary Heart Disease, Skane University Hospital, Lund, Lund University, Getingev, SE-22185 Lund, Sweden
Objectives This study sought to assess the safety and efficacy of transcatheter valve annuloplasty in patients with mitral regurgitation (MR).
Background Mitral regurgitation is associated with a worsened prognosis in patients with dilated cardiomyopathy. Surgical mitral annuloplasty reduces the septal-lateral dimension of the mitral annulus resulting in improved leaflet coaptation with a reduction in regurgitation. Percutaneous annuloplasty with the MONARC device (Edwards Lifesciences, Irvine, California) implanted within the coronary sinus is designed to reduce mitral regurgitation through a similar mechanism.
Methods A total of 72 patients with MR grade ≥2 were enrolled at 8 participating centers in 4 countries. Clinical evaluation and transthoracic echocardiography were performed at baseline and at 3, 6, and 12 months. Multislice cardiac computed tomography and coronary angiography were performed at baseline and 3 months.
Results The MONARC device was implanted in 59 of 72 patients (82%). The primary safety end point (freedom from death, tamponade, or myocardial infarction at 30 days) was met in 91% of patients at 30 days and in 82% at 1 year. Computed tomography imaging documented passage of the great cardiac vein over an obtuse marginal artery in 55% of patients and was associated with angiographic coronary artery compression in 15 patients and myocardial infarction in 2 patients (3.4%). At 12 months, a reduction in MR by ≥1 grade was observed in 50.0% of 22 implanted patients with matched echocardiograms and in 85.7% of 7 patients with baseline MR grade ≥3.
Conclusions Implantation of the MONARC device in the coronary sinus is feasible and may reduce MR. However, coronary artery compression may occur in patients in whom the great cardiac vein passes over a coronary artery, necessitating strategies in future studies to avoid this occurrence.
Functional mitral regurgitation (MR) frequently develops in patients with heart failure and a dilated left ventricle due to papillary muscle displacement, leaflet tethering, and progressive mitral annular dilation. This may lead to a vicious cycle of worsening heart failure and MR. Late prognosis is determined, in part, by the severity of MR (1–3).
Treatment options for functional MR are limited. Pharmacological therapy and resynchronization pacing may reduce MR and improve outcome in selected patients (1,4,5). Although surgical repair of the mitral valve may reduce MR, mortality and morbidity are significant and recurrence is common (6,7). Patients with MR have a poor prognosis despite medical, resynchronization, or surgical therapy (8–12). Consequently, there is considerable interest in lesser invasive approaches to treat MR in patients with ischemic and dilated cardiomyopathy.
We recently reported our initial experience with a first-generation catheter-based coronary sinus (CS) device designed to address MR (13). We now report the safety and efficacy of a modified second-generation device in patients with clinical and echocardiographic follow-up for 1 year.
Patients were eligible if they had heart failure and dilated cardiomyopathy (ischemic or idiopathic) with MR grade ≥2 (on a scale of 0 to 4) as assessed by color flow criteria on transthoracic echocardiography. Coronary sinus anatomy was considered suitable for the MONARC device if the proximal anterior interventricular vein (AIV) was ≥3 mm and ≤6 mm in diameter, the ostial CS was ≥7 mm and ≤16 mm in diameter, and the length from the great cardiac vein (GCV) to the ostial CS was ≥14 cm and ≤18 cm.
Exclusion criteria included structural valve disease including endocarditis, rheumatic disease, or moderate-to-severe mitral valve prolapse. Patients with pre-existing implantable defibrillators, CS leads for biventricular pacing, left ventricular ejection fraction <25%, prior percutaneous coronary intervention within 3 months, moderate-to-severe mitral annular calcification, serum creatinine >2.0 mg/dl, or inability to comply with all protocol procedures were also excluded. Patients received aspirin 180 to 325 mg and clopidogrel 300 mg before the procedure and continued on daily doses of aspirin ≥81 mg indefinitely and clopidogrel 75 mg for 3 months. Written informed consent and hospital ethics committee approval were required.
The MONARC system consists of a 12-F guide catheter and dilator, a 9-F delivery catheter, and a nickel-titanium alloy (nitinol) implant. The implant has 3 sections: a distal self-expanding anchor, a springlike “bridge,” and a proximal self-expanding anchor. The distal anchor is deployed at the transition between the AIV and the GCV; the proximal anchor is deployed in the ostial CS (Fig. 1). The anchors are designed such that they remain firmly in place acutely and encourage tissue in-growth, which provides a secure platform at both ends of the device. A biodegradable component is incorporated within the springlike bridge and maintains the device in an elongated state at the time of implantation. After implantation, the biodegradable element absorbs over approximately 1 month, creating an active and sustained spring tension between the anchors. Shortening of the device is designed to reduce the septal-lateral dimension of the mitral valve and thereby improve leaflet coaptation (Fig. 1).
The first-generation MONARC device developed a fracture in the bridge segment that resulted in device separation during follow-up in 3 of 5 patients (13). Bridge separation did not cause adverse events but may have reduced the efficacy of the device. Therefore, a nonbiodegradable suture was added to reinforce the bridge segment and reduce the likelihood of separation of the device following implantation. This second-generation MONARC device was tested in the present study. Available device sizes during this study ranged in length from 120 to 159 mm, with proximal and distal anchor diameters of 10 to 18 mm and 6 to 8 mm, respectively.
The internal jugular vein is accessed percutaneously and a guiding catheter used to cannulate the CS (13). A steerable guidewire is advanced to the AIV over which a calibrated angiographic catheter (Cook Medical Inc., Bloomington, Indiana) is placed to facilitate measurement of the coronary venous anatomy. A MONARC device is selected with anchors 2 to 4 mm larger in diameter than the proximal AIV and CS and a length 2 cm shorter than the distance between these 2 sites. The MONARC delivery catheter is advanced over the guidewire and through the guiding catheter. The distal anchor is deployed at the junction of the AIV and GCV by retracting the outer restraining sheath. The device is tensioned slightly to remove slack, following which the sheath is retracted further to release the proximal anchor.
Study end points
The primary safety end point was the freedom from the composite occurrence of death, tamponade, or myocardial infarction (MI) at 30 days. A secondary safety end point was freedom from the composite occurrence of device embolization or migration from the target area, death, MI, CS thrombosis, or pulmonary embolism at 3 months. The primary efficacy end point was procedural success, defined as device deployment at the intended location without the occurrence of serious adverse events (death, tamponade, or MI). The secondary efficacy end point was a reduction in mitral regurgitation by ≥1 grade. Patients were followed at discharge and at 1, 3, 6, and 12 months.
Imaging and core laboratory analysis
An electrocardiogram was obtained before and after the implant procedure. Transthoracic echocardiography was performed before the implant, at discharge, and at 1, 3, 6, and 12 months of follow-up. Transesophageal echocardiography was performed before the implant and at 3-month follow-up and analyzed at a core laboratory (Cardiovascular Research Institute, Washington, D.C.). Coronary angiography was performed before the implant procedure and at 3-month follow-up and analyzed at a core laboratory (University of Colorado Health Science Center, Denver, Colorado). Computerized tomographic (CT) scans were performed before and 3 months after the implant procedure and analyzed at a core laboratory (Ohio State University Medical Center, Columbus, Ohio). Chest X-rays were obtained at discharge and at 1, 6, and 12 months of follow-up. Creatine kinase, creatine kinase-myocardial band, and troponins were measured after the procedure and as indicated for suspicion of recurrent ischemia.
Analysis of safety end points was performed in all patients. Analysis of efficacy was performed only in patients with implanted devices. Paired t tests and Wilcoxon signed rank tests were used to assess the differences from baseline to follow-up of continuous and ordinal variables, respectively. A p value of <0.05 was considered significant.
Baseline features and implantation procedures
A total of 72 patients were enrolled between October 2005 and March 2007 at 8 participating centers in Canada, France, Germany, and Sweden. The mean age was 70.0 ± 9.8 years (range 37 to 90 years); 20 (18%) were women. Diabetes was present in 17 patients (24%), hypertension in 39 (54%), and hypercholesterolemia in 39 (54%). Most patients (68%) had underlying coronary artery disease, as manifested by a history of prior MI (57%), and/or coronary arterial bypass graft (47%) and/or percutaneous coronary intervention (40%). The baseline New York Heart Association (NYHA) class was I in 4%, II in 42%, III in 50%, and IV in 4%. The mean left ventricular ejection fraction was 37.8 ± 10.7% (range 23% to 67%). Mitral regurgitation at baseline as assessed by the core laboratory (transthoracic echocardiography in 59 patients and transesophageal echocardiography in 13 patients) was grade 1+ in 8%, grade 2+ in 54%, grade 3+ in 21%, and grade 4+ in 17%.
The device was implanted in 59 of 72 patients (82%). In the other 13 patients, a device was not implanted due to excessive tortuosity of the coronary venous anatomy or the lack of appropriately sized devices. In 1 of the patients with a tortuous sinus, there was an intraprocedural small dissection of the sinus during guidewire manipulation. The case was abandoned. Post-procedural echocardiography confirmed no pericardial effusion. Procedure time was 84 ± 59 min. Hospital stay was 4.1 ± 4.7 days.
Among the 72 enrolled patients, the primary safety end point of freedom from the occurrence of death, tamponade, or MI at 30 days was 91% at 30 days and 82% at 12 months. The secondary safety end point was achieved in 87% of patients at 3 months, 83% at 6 months, and 81% at 1 year.
Major adverse cardiovascular events are summarized in Table 1. Coronary sinus perforation resulted in 2 cases of tamponade early in the study from use of a stiff straight tipped wire; this complication did not recur with a change to J-tipped guidewires. Following discharge, MI occurred in 2 patients. One patient presented at day 16 with compression of a diagonal artery by the distal anchor. A second patient presented at day 19 with compression of an obtuse marginal artery. This site was stented with patency documented at 1-year follow-up.
There were no in-hospital deaths. None of the late deaths (8) were attributable to the procedure or device (Table 1). The survival rate was at 97%, 92%, 89%, and 87% at 1, 3, 6, and 12 months, respectively. Other adverse secondary safety end point events through 1-year follow-up included pulmonary embolism in 1 patient and device migration in 1 patient. There were no cases of device embolization or CS thrombosis.
Device malfunctions and coronary compressions
Routine follow-up coronary angiography at 3 months in 50 of 59 implants revealed 4 cases (8%) of separation of the bridge from the proximal anchor. No adverse clinical events were associated with anchor separation. However, only 1 of these 4 patients had an improvement in the degree of MR (compared with baseline) at follow-up.
Of 50 patients with implanted devices in whom follow-up angiography was performed, some degree of coronary artery compression was noted in 15 patients (30%): 5 were distal anchor-related, 9 were bridge-related, and 1 patient had both. The stenosis severity assessed by the angiographic core laboratory was >50% in 4 patients (8%) (Table 2). Among the 15 patients with angiographic coronary artery compression, 2 (13.3%) suffered a myocardial infarction.
Core laboratory analysis of CS anatomy by CT was available in 42 patients. The CS/GCV was observed to pass over a major obtuse marginal artery in 23 patients (55%), 8 of whom (34.8%) had angiographic coronary artery compression. In contrast, the CS/GCV did not pass over a major obtuse marginal artery in 19 patients, none of whom developed coronary compression.
Among the 59 patients in whom a device was implanted, the in-hospital procedural success rate was 93% (n = 55). In 4 patients, the proximal anchor was released outside the intended range of >0.5 to <2 cm from the CS ostium. Although this was not associated with adverse events, only 1 of these 4 patients had a reduction in MR at last follow-up. The secondary end point occurred in 87% of patients at 3 months, 83% at 6 months, and 81% at 1 year (Fig. 2).
The secondary efficacy end point (responder rate) is displayed in Figure 3 for 22 patients with matched data, in whom all transthoracic echocardiograms were completed within the pre-specified protocol windows (the results when all unmatched data were included were comparable—data not shown). In these 22 patients, the mean grade of MR was 2.5 ± 0.7 at baseline and 1.8 ± 0.8 at 12 months (p = 0.002). Mitral regurgitation decreased by ≥1 grade from baseline to follow-up in 40.9% of patients at 3 months and in 50.0% of patients at 6 months and 1 year. Moreover, the response rate appeared to be higher in the 7 patients with baseline grade ≥3+ MR compared with the 15 patients with only grade 2+ MR at baseline. The responder rate at 12 months was 85.7% in patients with severe MR at baseline (grade ≥3+) in whom the mean MR grade was reduced from 3.4 ± 0.5 at baseline to 2.1 ± 0.9 at last follow-up. Significantly fewer patients with a baseline MR grade of 2+ responded (Fig. 3).
Effects on left ventricular structure and function
Serial changes in quantitative echocardiographic variables are shown in Table 3. Although not powered or tested for statistical significance, directional reductions were present from baseline to follow-up in mitral valve annular diameter, effective regurgitant orifice area, regurgitant volume, left atrial volume, left ventricular end diastolic volume, left ventricular end systolic volume, and left ventricular ejection fraction.
Changes in NYHA functional class
In the paired analysis, 26 of 47 surviving patients at 6 months (55.3%) had at least 1 NYHA functional class improvement (including 12 patients who were NYHA class I), and 5 patients improved by at least 2 NYHA classes. NYHA functional class progressed at 6 months in only 2 patients. At 12 months, 19 of 41 surviving patients (46.3%) had at least 1 NYHA class improvement (including 9 patients in NYHA class I), and 6 patients improved by at least 2 NYHA classes. Only 1 patient had functionally deteriorated.
Other late clinical events
Five patients who did not have a clinically adequate reduction in MR following MONARC device implantation subsequently underwent mitral valve surgery at days 70, 110, 164, 222, and 232, respectively (ring annuloplasty in 2 patients and mitral valve replacement in 3 patients). Surgery was not influenced by the presence of the study implant. One patient died after surgery from multiorgan failure, and 4 patients had an uncomplicated post-operative course with significant reduction in MR.
Five patients who had persistent heart failure subsequently underwent implantation of a coronary sinus lead for biventricular pacing at days 109, 149, 170, 222, and 357 after device implantation. The placement of the lead was not hampered by the presence of the MONARC implant (14) and served as a convenient landmark for the CS ostium.
The 12-month results of the multicenter prospective, nonrandomized EVOLUTION phase I trial (Clinical Evaluation of the Edwards Lifesciences Percutaneous Mitral Annuloplasty System for the Treatment of Mitral Regurgitation) demonstrate the feasibility of a gradually foreshortening indirect acting mitral annuloplasty system delivered into the coronary sinus via a percutaneous jugular venous approach to reduce MR in patients with dilated and ischemic cardiomyopathy. Compared with other percutaneous mitral valve repair technologies, the MONARC procedure is relatively straightforward and has a short learning curve and procedure time. A majority of patients appeared to have a reduction in MR during follow-up, with the percentage of responders gradually increasing from baseline to 6 months and thereafter stabilizing. Patients with severe MR (grade ≥3+) seemed most likely to benefit.
Major procedural complications included 2 cases of CS perforation occurring due to navigation of the GCV in the early phase of the trial with a straight-tip guidewire. However, the potential for external coronary artery compression when the CS/GCV runs over the obtuse marginal epicardial vessels (which occurs in >50% of patients ) emerged as the most important consideration for use of this device. Angiographic follow-up demonstrated some degree of coronary artery narrowing in 30% of patients, which was significant (diameter stenosis >50%) in 8% of patients. This resulted in coronary occlusion in 3 patients, 2 of whom presented with a MI (3.4% of all patients receiving an implant). Patients in whom the anatomic relationship of the coronary venous and arterial anatomy is not a concern may include those with small, occluded, or protected (patent bypass graft) circumflex branches. The present study has also demonstrated that noninvasive screening with CT is able to identify those patients in whom the obtuse marginal vessels course under the CS/GCV who are at risk. Nonetheless, if the potential safety benefits of the MONARC are to be realized, prospective studies are required to demonstrate that procedural and late complications can be avoided.
Whether the gradual nature of MONARC device activation is beneficial compared with acute CS tensioning (16) cannot be answered by this study. The constant and continuous foreshortening of the device over 4 to 6 weeks may allow the heart to progressively remodel. The observation that left ventricular structure and function continued to improve between 3 and 6 months (well after which maximal device contraction has occurred) suggests that the gradual reduction in MR in patients might interrupt the progressive cycle of increasing heart failure and MR, although controlled studies in greater numbers of patients are required to confirm this hypothesis. Finally, compared with other CS technologies, the greater length of the MONARC device with placement of the distal anchor in the AIV may be important for left ventricular remodeling, as it tends to lift or retract the anterior wall of the ventricle.
Ventricular systole results in significant torquing and bending of CS implants. Bridge fractures with separation, which occurred with the first-generation device (13), were not observed in the present study, attesting to the success of design modifications. However, fractures in the proximal anchor were noted in 4 patients in the present series, which may have contributed to lack of device efficacy. Extending the bridge-section reinforcement to this region of the device may reduce this fracture mode.
Importantly, MONARC device implantation did not appear to interfere with subsequent mitral valve surgery (n = 5) or CS lead placement (n = 5). One patient had previously undergone a Mobius (Edwards Lifesciences) clip implant (17) (an investigational device that used a percutaneous suture to achieve a double outlet mitral valve) and was additionally treated with the MONARC device 3 months later without difficulty. Surgical experience suggests that mitral annuloplasty and edge-to-edge leaflet repair (18) may be synergistic in the treatment of MR.
Additional studies are required to determine the echocardiographic and clinical benefits that may be achieved with the present device. The average 1-grade reduction in MR with the MONARC device is less than what is obtained with surgical mitral valve repair, but may be of some benefit in patients with functional MR, particularly if accomplished with minimal periprocedural morbidity. Moreover, patient inclusion and exclusion criteria may affect the likelihood of a favorable response. The response rate was highest in the patients with more severe grades of MR. More sophisticated echocardiographic assessment may provide additional insight into which patients are likely to benefit with a CS annuloplasty device (1). Surgical series have suggested that improved cardiac performance after mitral valve repair occurs only when the pre-operative left ventricular end diastolic diameter is <65 mm (19).
The present report is the largest study to date of CS annuloplasty. Nonetheless, limitations of this phase I study include the modest sample size, heterogeneity of the study population, incomplete follow-up, and lack of a control group. Analyzable echocardiographic quantitative parameters were not available from all patients at all times. Functional measures of heart failure improvement such as the 6-min walk time and quality-of-life measures beyond NYHA class were not assessed.
The limitations notwithstanding, the present study indicates that implantation of the MONARC device in patients with functional MR is feasible, can be performed at relatively low procedural risk, and may result in a reduction in MR. However, coronary artery compression may occur in patients in whom the CS/GCV courses over the circumflex vessels and may result in late MI. A multicenter evaluation of the MONARC device (EVOLUTION II) is underway to determine the safety and efficacy of this device (with a reinforced proximal anchor) in patients with MR. This study includes pre-procedural CT screening to exclude those at risk for coronary compression of a major epicardial vessel, a concurrent control group, and serial measures of 6-min walk time and quality of life. Ultimately, a randomized controlled trial will be necessary to establish a potential role of the MONARC device in the treatment of patients with MR.
For a description of the EVOLUTION I study organization and list of participants, please see the online version of this article.
Supported in part by a grant from Edwards Lifesciences, Inc. Drs. Harnek, Webb, Kuck, and Vahanian are consultants to Edwards Lifesciences, Inc. Dr. Harnek has received honoraria from Edwards Lifesciences. Dr. Vahanian is a consultant for Medtronic. Dr. Buller is a consultant to Abbott Vascular. Dr. Stone received honoraria from Edwards Lifesciences, and he has equity in Guided Delivery Systems. All other authors report that they have no relationships to disclose.
- Abbreviations and Acronyms
- anterior interventricular vein
- coronary sinus
- computerized tomography
- great cardiac vein
- myocardial infarction
- mitral regurgitation
- New York Heart Association
- Received May 3, 2010.
- Revision received August 11, 2010.
- Accepted August 20, 2010.
- American College of Cardiology Foundation
- Grigioni F.,
- Enriquez-Sarano M.,
- Zehr K.J.,
- Bailey K.R.,
- Tajik A.J.
- Daubert C.,
- Gold M.R.,
- Abraham W.T.,
- et al.,
- REVERSE Study Group
- Bonow R.O.,
- Carabello B.A.,
- Kanu C.,
- et al.
- Bursi F.,
- Enriquez-Sarano M.,
- Nkomo V.T.,
- et al.
- Mihaljevic T.,
- Lam B.K.,
- Rajeswaran J.,
- et al.
- Webb J.G.,
- Harnek J.,
- Munt B.I.,
- et al.
- Maselli D.,
- Guarracino F.,
- Chiaramonti F.,
- et al.
- Schofer J.,
- Siminiak T.,
- Haude M.,
- et al.
- Feldman T.,
- Kar S.,
- Rinaldi M.,
- et al.,
- EVEREST Investigators
- Braun J.,
- Bax J.J.,
- Versteegh M.I.,
- et al.