| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2004;127:1068-1077
© 2004 The American Association for Thoracic Surgery
Cardiopulmonary support and physiology |
, MDc
a Department of Biomedical Engineering, Lerner Research Institute, Cleveland, Ohio, USA
b Department of Thoracic and Cardiovascular Surgery, Kaufman Center for Heart Failure, Cleveland, Ohio, USA
c Department of Cardiology, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
Read at the Eighty-third Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass, May 4-7, 2003.
Received for publication May 1, 2003; revisions received November 19, 2003; accepted for publication December 1, 2003.
* Address for reprints: Kiyotaka Fukamachi, MD, PhD, Department of Biomedical Engineering/ND20, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195, USA
fukamach{at}bme.ri.ccf.org
 |
Abstract |
|---|
 Top Abstract Material and methodsResultsDiscussionReferences |
|---|
METHODS: Functional mitral regurgitation with heart failure was induced in 7 dogs by rapid ventricular pacing. The Coapsys device, which consists of anterior and posterior epicardial pads connected by a subvalvular chord, was then implanted. Heart failure was maintained by continued pacing for 8 weeks. Hemodynamic and echocardiographic measurements were performed at pre- and postsizing and after 8 weeks. The Coapsys subvalvular chord was cut to verify that maintenance of valve competency was due to the device.
RESULTS: All implants were performed off-pump without atriotomy. Mitral regurgitation was reduced in all animals; mean mitral regurgitation grade was reduced from 2.9 ± 0.7 to 0.7 ± 0.8 (P = .00005) and was maintained at 0.8 ± 0.8 after 8 weeks, without hemodynamic compromise or structural damage to the mitral valve. Mitral regurgitation returned to 3.6 ± 0.8 (P = .102 versus presizing) after cutting the Coapsys subvalvular chord.
CONCLUSION: The Coapsys device consistently and chronically reduced functional mitral regurgitation. This device is in clinical trials in the United States.
|
Functional mitral regurgitation (MR) occurs with a structurally normal valve as a complication of left ventricular (LV) dysfunction, and as a result, almost half the patients with LV dysfunction have at least moderate MR.1 Recently, studies revealed that changes in the 3-dimensional relationships between the LV and mitral apparatus (mitral annular dilatation and papillary muscle displacement) induce functional MR.2-8 Based upon this information, 3-dimensional restoration of mitral valve (MV) geometry is an appropriate means to treat patients with functional MR. A novel device, the Coapsys Annuloplasty System (Myocor, Inc, Maple Grove, Minn), was developed to treat functional MR. The Coapsys device is designed to correct valve dysfunction without requiring the use of cardiopulmonary bypass (CPB) or an open-heart access method. This study was conducted to evaluate the acute and chronic effects of the Coapsys device on MR in a canine model of functional MR.
|
Material and methods |
|---|
|
TopAbstractMaterial and methodsResultsDiscussionReferences |
|---|
|
|
Animal model
Among several potential MR models, chronic rapid ventricular pacing was chosen in this study because the hemodynamic and echocardiographic changes are very similar to those found in human functional MR.11 Also, rapid ventricular pacing produces heart failure,4,11-13 which is a common comorbidity among this patient group.
The methodology of inducing and maintaining the model has been reported.12 Seven adult mongrel dogs (body weight 25.0 ± 1.1 kg) were paced via the right ventricular (RV) transvenous lead using rapid asynchronous ventricular pacing at 230 beats per minute for an average of 31 ± 4 days to induce functional MR with heart failure. No medication for heart failure, including diuretics, was given during the induction phase. At the end of the induction phase, the pacemaker rate was temporarily reduced to demand mode at 30 beats per minute so that the animal would resume normal sinus rhythm, and transthoracic 2-dimensional (2-D) echocardiograms were obtained to evaluate the progression of cardiac remodeling, the reduction of LV function, and development of MR and tricuspid regurgitation (TR; graded on a scale from 0+ to 4+). If MR was equal to or more than 2+ at the presizing data point then plans were made for Coapsys implantation.
Coapsys implant surgery
On the date of study, the pacemaker was again reduced to demand mode, and the animal was anesthetized with thiopental (15 mg/kg) and intubated. The anesthesia was maintained with isoflurane (0.5%-2.5%). The following hemodynamic and echocardiographic measurements were taken as the baseline closed-chest data point. A catheter with 2 Millar pressure sensors (Model SPC-562; Millar Instruments, Inc, Houston, Tex) was placed via the carotid artery to record aortic (AoP) and LV (LVP) pressures. A thermistor-tip balloon catheter (Criticath SP5507 TD catheter; Becton Dickinson Infusion Therapy System Inc, Sandy, Utah) was passed from the jugular vein to the pulmonary artery to monitor the pulmonary artery pressure and the central venous pressure. Cardiac output (COSG) and stroke volume (SVSG) were measured by thermodilution method. Hemodynamic indices were digitized in real time at a sampling rate of 200 Hz with a data acquisition system (PowerLab; AD Instruments, Inc, Mountain View, Calif) and stored on a hard disk for subsequent analysis. Changes in LV volumes were assessed before and after sizing the Coapsys device by a 3-dimensional (3-D) ultrasound system (Volumetrics Medical Imaging Inc, Durham, NC). Subsequently, LV end-diastolic (LVEDV) and end-systolic (LVESV) volumes were determined by rotated apical 6-plane views with multiplanar Simpson method.14 Stroke work (stroke work = [LVEDV – LVESV] x [LV end-systolic pressure – LV end-diastolic pressure] x 0.0136) and ejection fraction (ejection fraction = [LVEDV – LVESV]/LVEDV x 100) were calculated. Transthoracic 2-D images were obtained to assess MR, TR, and aortic regurgitation (AR) by color Doppler imaging. The mitral annular septal-lateral (S-LMA) dimension was measured using the long-axis view and the mitral annular commissure-commissure (C-CMA) dimension was measured using the 2-chamber view at end-diastole and end-systole. The LV cross-sectional dimensions in septal-lateral (S-LLV) and commissure-commissure (C-CLV) planes were measured at midpapillary muscle level. MR volume and effective regurgitant orifice area (EROA) were calculated using the continuity equation.15 MR jet area, the ratio of MR jet area to left atrial area (MR/LA ratio), annular area, and peak MV pressure gradient were also measured. A sternotomy was performed and the chest was opened. Just prior to opening the chest, a continuous infusion of lidocaine was started at 1 mg/kg/h to prevent ventricular arrhythmias. A 14-mm Transonic flow probe (model no. 14A165, Transonic Systems Inc, Ithaca, NY) was placed on the ascending aorta and connected to a flow meter (Model T106, Transonic Systems Inc) to measure cardiac output (COFP) and stroke volume (SVFP). A Millar catheter (model SPC-350; Millar Instruments, Inc) was inserted into the LA to monitor LA pressure (LAP).
The Coapsys device was surgically implanted. The appropriate sites for the Coapsys device placement were identified through a combination of external landmarks and 2-D echocardiographic visualization of internal structures to avoid compromise of papillary muscles, mitral apparatus, or main coronary artery branches. The position of the posterior pad was approximately 2.5 cm from the atrioventricular groove and midway between the papillary muscles. The anterior pad was positioned at the apical edge of the base of the RV outflow tract and approximately 2 cm to the RV side of the left anterior descending artery. Following site identification, the delivery instrument was fixed to the heart utilizing vacuum assistance. A needle assembly, consisting of a stylet and hollow catheter, was then inserted through the delivery instrument from the anterior wall. Following passage of the needle assembly, the stylet was removed and the subvalvular chord threaded through the catheter and pulled through both walls. The subvalvular chord was drawn through the ventricle until the posterior pad rested on the LV wall. The anterior pad was inserted over the anterior side of the subvalvular chord. Hemodynamic and echocardiographic measurements were recorded as the presizing data point. Following Coapsys device placement, the subvalvular chord leader was then secured to the sizing instrument. The Coapsys device was then tightened by drawing the posterior leaflet and annulus toward the anterior leaflet. The deployment level was selected when MR was minimized or eliminated as assessed by color Doppler images. The device size (epicardial to epicardial measurement) was obtained as the presizing and postsizing measurements (measured as percent linear reduction from presizing measurement). The Coapsys device was fixed at the deployment level. Hemodynamic and echocardiographic measurements were repeated as the postsizing data point. The chest was then closed in the standard fashion.
Postoperative analgesics and antibiotics were administrated, but no anticoagulants or antiplatelet drugs were administrated. To avoid the natural recovery of the LV and MV performance after cessation of pacing, rapid pacing at 190 beats per minute was started 3 days after surgery and maintained for 8 weeks until the terminal study.12 Each dog was given furosemide (40 mg/d) during the postoperative period. One dog was unable to eat or drink after the implantation surgery due to severe heart failure. This dog was treated with hyperalimentary therapy and discontinued pacing for 1 week (data from this animal were included in all data points). Another dog died on postoperative day 14 due to an ascending aortic rupture, which was induced by the chronically implanted aortic flow probe (data from this animal were included in baseline closed-chest, presizing, and postsizing data points).
Terminal study
Eight weeks after the Coapsys implantation, the animals were again anesthetized and hemodynamic monitoring lines were replaced. Hemodynamic and echocardiographic measurements were performed as terminal closed-chest data point. After obtaining all data, the chest was opened, and hemodynamic and echocardiographic measurements were performed as terminal open-chest data point. Under 2-D echocardiographic guidance, the Coapsys subvalvular chord was cut utilizing a laparoscopic scissor introduced into the LV through a purse-string suture on the apex to verify that maintenance of the valve competency was due to the device (model verification). Hemodynamic and echocardiographic measurements were again performed and recorded as subvalvular chord-cut data point.
All animals were put to death and complete autopsies were performed. The hearts were excised, and the device's relationship to the coronary vessels and intraventricular structures was examined.
Statistical analysis
Data were expressed as mean ± standard deviation. For the acute evaluation, there was 1 pair of data: presizing and postsizing. For the chronic evaluation, there were 3 pairs of data: closed-chest (baseline and terminal) and open-chest (presizing and terminal open-chest and postsizing and terminal open-chest). For the model verification, there were 2 pairs of data: terminal open-chest and subvalvular chord-cut, and presizing and subvalvular chord-cut. A paired t test was used for each paired data.
|
Results |
|---|
|
TopAbstractMaterial and methodsResultsDiscussionReferences |
|---|
Table 1 summarizes the echocardiographic findings. By 2-D echocardiography, MR was reduced in all animals. MR decreased from 2.9 ± 0.7 at presizing to 0.7 ± 0.8 at postsizing (P = .00005) and from 2.5 ± 0.8 at baseline closed-chest to 0.8 ± 0.8 at terminal closed-chest (P = .002). MR increased from 0.8 ± 0.8 at terminal closed-chest to 1.9 ± 0.7 at terminal open-chest (P = .041). MR returned to 3.6 ± 0.8 after cutting the Coapsys subvalvular chord (P = .001 vs terminal open-chest). There was no significant difference in MR grade between presizing and subvalvular chord-cut (Figure 3). End-diastolic and end-systolic S-LMA dimensions decreased significantly at postsizing and terminal closed-chest and increased significantly after cutting the subvalvular chord (Figure 4). The C-CMA dimensions were unchanged (Figure 5). End-diastolic and end-systolic S-LLV dimensions also decreased significantly at postsizing and terminal closed-chest and increased significantly after cutting the subvalvular chord. The end-diastolic C-CLV dimension at postsizing increased significantly and end-systolic C-CLV dimension decreased significantly after cutting the subvalvular chord. There were no significant differences in S-LLV and C-CLV dimensions between presizing and terminal subvalvular chord-cut. At terminal open-chest, both S-LMA and S-LLV dimensions decreased significantly compared with those at presizing and increased significantly compared with those at postsizing. There were significant decreases in EROA and MR volume between presizing and postsizing and baseline closed-chest and terminal closed-chest. EROA and MR volume increased from terminal closed-chest to terminal open-chest (P = .048 and P = .028, respectively) and from terminal open-chest to subvalvular chord-cut (P = .023 and P = .054, respectively). MR jet area and MR/LA area ratio improved after the Coapsys implantation and was maintained for 8 weeks. Annular area decreased from 4.7 ± 0.6 cm2 at presizing to 3.3 ± 0.5 cm2 at postsizing; P = .0003). Peak MV pressure gradient was unchanged and revealed no mitral stenosis. There were no significant changes in TR and AR.
|
|
|
|
Despite the significant decreases in LV volumes, there were no significant changes in COFP (1.71 ± 0.55 L/min to 1.63 ± 0.49 L/min; P = .144) or SVFP (19.2 ± 8.3 mL to 18.4 ± 7.7 mL; P = .196) between presizing and postsizing. There were no significant changes in COSG and SVSG chronically. The mean AoP (77 ± 20 mm Hg to 73 ± 18 mm Hg; P = .011) and systolic LVP (91 ± 20 mm Hg to 86 ± 18 mm Hg; P = .012) decreased slightly but statistically significantly at postsizing. There were no significant changes in mean AoP and systolic LVP chronically. The heart rate increased at terminal closed-chest (100 ± 20 beats per minute to 116 ± 18 beats per minute; P = .046), and the heart rate at terminal open-chest increased significantly compared with pre- and postsizing. The LAP at terminal open-chest decreased significantly compared with pre- and postsizing. In other hemodynamic parameters, there were no significant differences between presizing and postsizing and baseline closed-chest and terminal closed-chest. There were no significant differences between terminal open-chest and terminal subvalvular chord-cut and between presizing and subvalvular chord-cut in all hemodynamic parameters (Table 2).
|
|
Discussion |
|---|
|
TopAbstractMaterial and methodsResultsDiscussionReferences |
|---|
On the other hand, some groups have suggested ventricular shape changes including papillary muscle repositioning should be used to reduce functional MR.5,18,19 Surgical LV remodeling was effective in reducing ischemic MR in an animal infarct model.5 Also in animals, an external device that repositioned papillary muscles reduced MR without compromising LV function in an ischemic MR model.18 Menicanti and associates19 reported that papillary muscle imbrication in patients undergoing ventricular reconstruction surgery reduced functional MR, which supports the concept of LV reconstruction surgery used to reduce MR. We believe that both annular reduction and papillary muscle repositioning will create the most ideal, chronically effective treatment for functional MR. In accordance with this belief, we employed a novel device to treat functional MR. The posterior pad of the Coapsys device has two heads: one to move the mitral annulus, the other to reposition the papillary muscles.
From terminal closed chest to terminal open chest, MR grade, EROA, MR volume, MR area, MR/LA area ratio, end-diastolic S-LMA, and end-systolic S-LMA increased and LVEDV and LVESV decreased. The significant decrease in volume is likely due to hemodynamic changes caused by opening the chest, freeing the heart of surrounding adhesions, and redistributing circulating volume, which in turn results in a reduction of annular change induced by the device and the commensurate increase in MR indices. Also, there is a possibility that this manipulation moved the Coapsys device. Additionally, continued pacing for 8 weeks after implantation may have led to progressive LV dysfunction and MR. We cut the Coapsys subvalvular chord to verify that maintenance of the valve competency was due to the device. After cutting the Coapsys subvalvular chord, all parameters regarding MR, which were improved by the Coapsys device, returned to baseline. The mitral annular and LV geometric changes also returned to the presizing condition. These results suggest that the Coapsys device maintained the reduction of functional MR.
The goal of the procedure was to mechanically coapt the MV leaflets by changing the MV annular dimensions and papillary muscle position. This fact was reflected in the reduced MR measured by echocardiography, the primary end point of this study. Echocardiographic reduction of MR is, of course, also the clinical standard for judging MV repair. We also secondarily sought to determine if there were hemodynamic changes associated with this MR reduction. The methodologies of this type of hemodynamic measurement are more difficult to analyze and some conflict exists between the results obtained by COSG, COFP, and those calculated by 3-D echo volumetrics. By direct LAP measurement, there was a significant reduction in LAP (from 16.5 ± 5.8 mm Hg at presizing to 12.5 ± 2.6 mm Hg at terminal open-chest, P = .022). Also, COFP increased from 1.71 ± 0.55 L/min at presizing to 2.34 ± 0.72 L/min at terminal open-chest but did not reach significance. Whether these are accurate observations, or even important, are different questions. Not all patients with cardiomyopathy undergoing MV repair have an immediate improvement in hemodynamic parameters. Heart failure is more complex than just measurements of forward cardiac output. Also, unlike the Myosplint (a completely different device and concept),20,21 we were treating MR and so we did not try to obtain major changes in ventricular size or dimensions. There were, however, significant reductions in acute LV volumes. Nevertheless, as with all such animal experimental work, we need a careful randomized clinical study in a much larger population to sort through the clinical benefits of increasing MV coaptation and decreasing MR.
Study limitations
In this animal study, we used the rapid ventricular pacing-induced functional MR model with LV dysfunction. In this model, there were usually 2 MR jets: a lateral and a medial jet. This is similar to the pattern seen in human ischemic cardiomyopathy (separated jets), whereas most patients with idiopathic dilated cardiomyopathy have a single central MR jet.8 The effect of the Coapsys device may be different in these patients. We continued rapid pacing at 190 beats per minute for 8 weeks after implantation, so somewhat mild or moderate remodeling might have occurred during that time.
There were some inconsistencies in measurements between the COSG or COFP and the cardiac output calculated from 3-D echo-derived LVEDV, LVESV, and 2-D echo-derived MR volume. These inconsistencies are probably due to the limitation of volumetric analysis using 3-D echo. The 3-D echo data were shown to give estimates of LV volume changes with this device.
Anatomic measurements, including depth of coaptation and length of coaptation surface, would probably be a much more accurate way to judge how much to tighten the device rather than merely by interpreting color Doppler echo, and that is certainly something that will be looked at carefully in the clinical cases.
In conclusion, the Coapsys device reduced functional MR and improved parameters regarding MR acutely and chronically. In addition, we did not observe any negative impact on hemodynamic function in this animal model. The Coapsys device may be a promising new therapy for functional MR, and clinical trials are under way.
Discussion
Dr Farzan Filsoufi (New York, NY). Fifty-two years ago, the pioneering cardiac surgeon, Charles Bailey, published a paper on the surgical treatment of mitral insufficiency on a beating heart using a strip of pericardium passed through the ventricle in a fashion similar to that of this novel device. This places this interesting study in historical context.
To understand the potential applicability of this device to patients with mitral regurgitation secondary to ischemic heart disease or dilated cardiomyopathy, it is important to emphasize the pathophysiology of this condition. These patients have restricted leaflet motion during systole resulting from ventricular dilatation and papillary muscle displacement. This corresponds to type IIIb dysfunction in Carpentier's functional classification of MR. Most patients also have some degree of secondary annular dilatation, primarily in the septolateral dimension. This annular dilatation is an important factor in the pathogenesis and progression of type IIIb mitral regurgitation.
We believe the current standard of care for this condition is a downsized remodeling annuloplasty, which reshapes the annulus and increases the surface of coaptation by decreasing the septolateral dimension. Because it only treats the annular component and not the ventricular component of the dysfunction, the remodeling annuloplasty technique is associated with a small incidence of residual or recurrent mitral regurgitation.
The Coapsys device is unique in that it attempts to address both the annular and ventricular components of type IIIb MR. This is a well-designed and well-executed study in a difficult animal model with interesting results.
I have the following questions for you. My first question relates to durability. Although you demonstrate a significant reduction in MR, this decrease did not correlate with expected anatomic findings. For example, on 2-D echocardiography, the mitral valve area and the septolateral dimension actually increased by 33% and 20% over the 2-month study period. This inconsistency between the physiologic correction of the MR and the anatomic findings should be addressed. Are you concerned about this early increase in valve area and septolateral dimension and its potential impact on the long-term durability of the procedure?
My next question relates to your technique of intraoperative sizing. You rely on intraoperative echo to size the device. However, our group and others have clearly documented the downgrading of type IIIb MR by intraoperative echocardiography as a result of downloading effects of general anesthesia. Do you feel that anatomic measurements such as septolateral dimension and coaptation depth and length would be more appropriate parameters for sizing in the clinical setting?
My third question relates to the clinical indications for the device. There is a broad spectrum of patients with type IIIb MR with regard to the severity of mitral regurgitation and the degree of left ventricular dilatation. Which patients within this spectrum do you feel are likely to have a successful, durable outcome with this device? Specifically, do you believe that it will work as sole therapy for patients with severe MR and dilated ventricles or do you see it as an adjunct to annuloplasty in these patients?
Dr McCarthy. First, we assess carefully all of the changes that we see regarding durability. One explanation for some of the differences in measurements—which, by the way, were only about 3 mm—is that some of those measurements were made in the acute setting with the chest open, versus later measurements with the chest closed. So there are certainly differences in loading conditions. Also, measurements are influenced by the difference in echo techniques in that the measurement of the annular area was by short axis, the measurement of the septolateral by long axis, and the commissure-to-commissure measurement by 2-chamber views. So it is hard in 2-D echoes to really recreate the total annular area and correlate it to the other 2 measurements because of the different views. Furthermore, this is a pacing model of heart failure that we had developed a few years ago, and there is ongoing pacing at a rate of 190 beats per minute. So we expect that there would be some continued remodeling during that time, which may account for some of the small changes. Nevertheless, durability is the most important issue for the clinical trials.
In terms of the technique, I completely agree that anatomic measurements, including coaptation depth and length, would probably be a much more accurate way to judge how to place the device than merely by interpreting color Doppler echo. This is certainly something that will be looked at carefully in the clinical cases.
In terms of the clinical indications, I don't know yet exactly what role this is going to have. Coapsys just recently began a randomized clinical trial in the United States with groups of patients undergoing coronary artery bypass grafting (CABG) with ring annuloplasty, patients undergoing CABG alone, and CABG with the Coapsys device. So we will be looking at how this holds up in patients with ischemic cardiomyopathy and then we can determine the proper role for Coapsys.
|
Footnotes |
|---|
. This study was financially supported by Myocor, Inc (Maple Grove, Minn). Patrick M. McCarthy, MD, is a consultant to Myocor, Inc (Maple Grove, Minn).
|
References |
|---|
|
TopAbstractMaterial and methodsResultsDiscussionReferences |
|---|
ZB, Qin JX, Gillinov MA, et al. Geometric differences of the mitral apparatus between ischemic and dilated cardiomyopathy with significant mitral regurgitation: real-time three-dimensional echocardiography study. Circulation. 2003;107:1135–1140 ZB, Doi K, Schenk S, Nemeh H, et al. Off-pump mitral valve repair using the Coapsys device: a pilot study in a pacing-induced mitral regurgitation model. Ann Thorac Surg. 2004;77:688-92This article has been cited by other articles:
|
|
 |
|
  M. Akiyama, Z. B. Popovic, K. Kamohara, F. Cingoz, M. Daimon, C. Ootaki, Y. Ootaki, M. Martin, J. Liu, M. W. Kopcak Jr, et al. Acute reduction of functional mitral regurgitation in canine model using an epicardial device. Ann. Thorac. Surg., May 1, 2008; 85(5): 1771 - 1775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Langer, F. Rodriguez, S. Ortiz, A. Cheng, T. C. Nguyen, M. K. Zasio, D. Liang, G. T. Daughters, N. B. Ingels, and D. C. Miller Subvalvular Repair: The Key to Repairing Ischemic Mitral Regurgitation? Circulation, August 30, 2005; 112(9_suppl): I-383 - I-389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Levine and E. Schwammenthal Ischemic Mitral Regurgitation on the Threshold of a Solution: From Paradoxes to Unifying Concepts Circulation, August 2, 2005; 112(5): 745 - 758. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
