| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiac Surgery, Department of Surgery, Johns Hopkins Medical Institutions, Baltimore, Maryland
Accepted for publication May 5, 2008.
* Address correspondence to Dr Conte, Division of Cardiac Surgery, Department of Surgery, The Johns Hopkins Hospital, 600 N Wolfe St, Blalock 618, Baltimore, MD 21287-4618 (Email: jconte{at}csurg.jhmi.jhu.edu).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: We conducted a case-control study comparing patients with ischemic cardiomyopathy and ejection fraction less than 0.35 who underwent SVR + CABG versus CABG alone from June 2002 to December 2005. Patients who underwent SVR + CABG were compared with control patients who met criteria for SVR + CABG by ventriculogram or echocardiogram but received CABG alone. End points included survival, rehospitalization for heart failure, and New York Heart Association class.
Results: During the study period 120 patients underwent SVR + CABG (n = 62) versus CABG alone (n = 58). Patients in the SVR + CABG group were younger (60 versus 64 years; p = 0.04) and more likely to be New York Heart Association class III or IV preoperatively (98% versus 86%; p = 0.01). Operative mortality was similar between groups (6.4% versus 5.2%; p = 1.00). Ejection fraction was similar preoperatively (0.22 versus 0.24; p = 0.31) and postoperatively (0.34 versus 00.32; p = 0.40). The SVR + CABG patients experienced fewer rehospitalizations for heart failure (24% [13 of 54] versus 55% [24 of 44]; p = 0.006) but had similar 4-year survival (p = 0.60). At follow-up, 80% (50 of 62) of SVR + CABG versus 57% (27 of 47) of CABG alone patients improved to New York Heart Association class I or II (p = 0.01).
Conclusions: Patients with ischemic cardiomyopathy and ventricular enlargement experience similar early survival after SVR + CABG or CABG alone. However, SVR + CABG resulted in fewer rehospitalizations and better improvements in New York Heart Association class. Surgical ventricular restoration with CABG should be offered to eligible patients with ischemic cardiomyopathy and ventricular enlargement.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Ventricular enlargement is often the result of a maladaptive response to ischemic cardiac injury and results in the clinical syndrome of CHF. It has been reported that after a myocardial infarction, 26% of patients develop left ventricular dilation, leading to a spherically shaped ventricle with diminished contractile function and heart failure [3]. Surgical ventricular restoration (SVR) is an established therapy for restoring cardiac size, shape, and function in these patients [4–7]. Surgical ventricular restoration, in combination with coronary revascularization and correction of any existing valvular regurgitation, improves the morphologic characteristics of patients with ischemic cardiomyopathy and provides clinical outcomes superior to medical therapy alone [7–9]. Recent studies have also shown that SVR improves regional myocardial performance in nonischemic areas remote from the scar, improves ejection fraction, and reduces ventricular dyssynchrony [10–12]. Unfortunately, many patients who might benefit from SVR do not receive it because surgeons are often unaware of its potential benefit, are unfamiliar with the surgical techniques, or are reluctant to perform the procedure on patients with severe left ventricular dysfunction and advanced CHF.
Although SVR is an accepted therapy for the treatment of ischemic cardiomyopathy, controversy exists as to whether it is the coronary revascularization or the SVR that leads to postoperative improvements for these patients. Studies that have examined this issue have been limited by lack of appropriate control groups [13–15]. In this study, we sought to retrospectively compare patients who underwent SVR with CABG (SVR + CABG) to those who received CABG alone. To have appropriate control patients, we stratified patients undergoing CABG alone into two different groups. One group consisted of patients who were candidates for SVR at the time of their bypass procedure, but underwent only CABG, and the other group was composed of those who were not candidates for SVR. We hypothesized that patients who undergo SVR + CABG will show significant postoperative improvements compared with a cohort of CABG patients who were determined to be eligible for the SVR procedure but did not receive it.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
A total of 224 patients were included in the study. A single experienced SVR surgeon (J.V.C.) was blinded to patient identification and reviewed the ventriculograms and echocardiograms of all patients who underwent CABG with an EF less than 0.35 to determine their eligibility for SVR (Fig 1). Typical indications for SVR at our institution include an anteroseptal infarct, ventricular enlargement, akinesia or dyskinesia in the anterior wall, acceptable function of the basal portion of the heart and the lateral wall, and good right ventricular function [16]. All CABG patients were thus classified as either SVR candidates or noncandidates (Fig 1). Echocardiograms were the only preoperative images available for 5.6% of patients whereas ventriculograms were the only preoperative images available for 21.6% of patients. Both ventriculograms and echocardiograms were available for 72.8% of our study group.
|
Patient Variables
Data collection included demographic characteristics, preoperative risk factors, hemodynamic and operative data, and postoperative complications. Preoperative and postoperative EF and New York Heart Association (NYHA) functional class were also assessed. Rehospitalization for CHF was defined as a hospital admission longer than 30 days postoperatively in which CHF was the main diagnosis or the patient presented with heart failure symptoms and an EF less than 0.35. Mortality data were obtained from autopsy reports, physician's records, and the Social Security Death Index. For patients not regularly seen at our institution, follow-up information was collected by means of telephone interview. Telephone follow-up was 96.4% (8 of 224 patients were lost to follow-up) with a mean time of 21.8 ± 15.7 months. Standard discharge heart failure medications included angiotensin-converting enzyme inhibitors, β-blockers, and diuretics with no discernible differences in medications among groups.
Outcome Variables
The primary end point was overall mortality during the entire study period. Secondary outcomes included rehospitalization for heart failure, improvement in postoperative EF, improvement in NYHA functional class, length of stay, and other markers of morbidity including development of respiratory failure, renal insufficiency, stroke, need for intraaortic balloon counterpulsation, and infection.
Operative Technique
Our operative technique has been described previously [16, 17]. We prefer to perform SVR on the arrested heart whenever possible depending on the basis of cross-clamp time, concomitant procedures, and cardiac function. Coronary artery bypass grafting is normally performed first, followed by mitral valve repair or mitral valve replacement if indicated, and then SVR. We routinely use an intraventricular sizing device (Chase Medical, Richardson, TX; Bioventrix, Danville, CA) sized to 50 to 60 mL/m2 body surface area when placing the Fontan stitch and defining the new left ventricular apex. We then close the ventriculotomy defect primarily with a linear closure, unless the incision exceeds 2 to 3 cm, in which case we use a polyethylene terephthalate fiber (Dacron) patch.
Statistical Analysis
Descriptive statistics are presented as mean ± standard deviation unless otherwise stated. Chi square and the Student's t test were used to compare categorical and continuous variables, respectively. The Kaplan–Meier method was used to model survival, which was compared among the three groups by the Mantel-Cox log rank test. We used SPSS version 13 (SPSS Inc, Chicago, IL) software for statistical analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The SVR candidate group was well matched to the SVR + CABG cohort. The only statistically significant differences between SVR candidates and SVR + CABG patients were age, the number of preoperative percutaneous transluminal coronary angioplasty procedures, and the total number of patients who were NYHA class III or IV preoperatively (86% for SVR candidates versus 98% for SVR + CABG; p = 0.01; Table 1).
|
|
Postoperative complications and hospital stay among the groups were similar (Table 3).
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The multicenter RESTORE study evaluated the outcomes of SVR in 1,198 CHF patients, and found an excellent 5-year survival of 69% in patients with ischemic cardiomyopathy afer an anterior myocardial wall infarction [13]. Similarly, Di Donato and associates [7] retrospectively reviewed 245 SVR patients and found a 5-year survival of 85%. In patients with severe left ventricular dysfunction undergoing only CABG, studies have reported mortality rates between 60% and 80% [18, 19]. In our series of patients, 4-year survival for patients classified as noncandidates for SVR because of incompatible anatomy was 77%. For patients undergoing SVR + CABG, survival was 75% (Fig 2). In both cases, our results confirmed previous estimates. However, for individuals who we believed to be qualified for the SVR procedure yet did not receive it (SVR candidates), we report a 4-year survival of 62% (Fig 2). We have demonstrated that these patients were similar to SVR + CABG patients in almost every category (Table 1) and believe differences in their operative management affected their outcomes.
This study reviewed CABG patients with severe left ventricular systolic function to determine their eligibility for the SVR procedure. Previous studies relied on left ventricular internal diastolic dimensions or volume for comparison [14]. Yamaguchi and colleagues [15] showed that preoperative left ventricular end-systolic volume index predicts actuarial survival rate in patients with ischemic cardiomyopathy. Data from the global utilization of streptokinase and tissue plasminogen activator (t-PA) for occluded coronary arteries study demonstrated similar findings, showing that large preoperative ventricular volume is a strong predictor of early and late mortality [20]. However, a dilated cardiomyopathy does not necessarily correlate with eligibility for the SVR procedure. Current indications generally accepted for the SVR procedure include single-territory anterior myocardial infarction and left ventricular akinesia or dyskinesia [21], factors not necessarily present in all dilated cardiomyopathy cases. We believe that by reviewing preoperative ventriculograms and echocardiograms, we were able to accurately determine the individuals who could have benefited from SVR. Only after completing the patient classifications did we evaluate differences in preoperative characteristics and postoperative outcomes, and our preoperative analysis indicates that we created an appropriate comparison group (Table 1). Therefore, we believe our results accurately reflect the differences in postoperative outcomes for CABG versus SVR + CABG patients. Additionally, because the percentage of patients with three-vessel coronary artery disease or angina and the number of grafts in each group were similar, it is reasonable to consider that the SVR procedure itself played a significant role in the improved outcomes of the SVR + CABG cohort.
Mitral valve regurgitation is a common occurrence in individuals with a dilated cardiomyopathy. The longitudinal and transverse elongation of the ventricle causes lateral displacement and widening of the distance between the papillary muscles hindering leaflet coaptation [22–27]. Concomitant revascularization with mitral valve repair and SVR has been shown to improve left ventricular and mitral valve function [28]. In our analysis, we noted no significant difference in preoperative mitral grade. However, we did find a greater number of mitral valve repairs in the SVR + CABG patient group when compared with the SVR candidate group. This signified that the SVR + CABG patients had a more severe cause of disease at the time of operation; however, as we have shown, they had better results in both our primary and secondary outcomes. To ensure that the mitral valve operations were not confounding our results, we removed all individuals undergoing either mitral valve repair or replacement at the time of their bypass operation and found no changes in either primary or secondary end points.
We believe that SVR is a useful tool to treat ischemic cardiomyopathy in eligible patients but is underutilized. We sought to illustrate this point by reviewing our own institutional experience and have demonstrated that many eligible patients did not receive SVR. The reasons can range from the primary surgeon not appreciating the potential benefit or being unfamiliar with candidate selection, the lack of training in the surgical technique, or simply the politics of patient referral. During the period under review at our institution, all of these were in play at some point. Undoubtedly, this contributed significantly to our having a control group of SVR-eligible patients. One of our hopes is that this paper will raise the awareness of SVR so that in the future eligible patients will be referred to surgeons qualified to perform the restorative procedure.
Our retrospective study is limited by a relatively short follow-up interval, a strong reliance on patient history, and the potential for selection bias. Classifications were based on the review of preoperative imaging and were thus inherently subjective. We tried to mitigate the subjectivity of these classifications by having a single cardiac surgeon review all available echocardiograms and ventriculograms. The surgeon was blinded to all preoperative characteristics and postoperative variables; therefore, decisions were based entirely on heart function, dilation, and wall motion abnormalities. Despite these limitations, we created a comparison group for SVR patients that differed only by age and percutaneous transluminal coronary angioplasty. Longer follow-up and a larger patient cohort will enable us to make better long-term conclusions about the benefits of SVR, but our initial results indicate the importance of its application to potential candidates. Current studies such as the Surgical Treatment for Ischemic Heart Failure (STICH) trial [29] and longer follow-up on recently completed prospective studies [30] will also help to elucidate the benefits of SVR.
In conclusion, we have shown that SVR + CABG patients had a greater improvement in EF and NYHA class, and were less likely to be rehospitalized for CHF postoperatively compared with those individuals who were eligible for the SVR procedure but did not receive it. Therefore, the combined SVR + CABG procedure should be offered to eligible patients with ischemic cardiomyopathy and ventricular enlargement. Future studies will determine long-term outcomes in all three patient groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
DR WEISS: Well, I think I can answer that question, and I will invite Dr Conte to correct me if I am wrong. But my feeling is that based on the results of this study as well as our whole experience with SVR, we now believe that if a patient is a suitable candidate, SVR is a superior procedure to CABG (coronary artery bypass grafting) alone for symptomatic improvement. Our surgeons at Johns Hopkins are really well knowledgeable about the benefits of SVR and are able to make a decision about whether or not a patient is a candidate, and if they themselves do not perform SVR, patients are usually referred to Dr Conte or one of the surgeons at our institution who will do the procedure.
DR ERNESTO R. SOLTERO (Ponce, PR): What percentage of your patients do you refer for AICD (automatic implanted cardioverter defibrillator) after surgery?
DR WEISS: That is a question that I don't know the answer to and I would ask Dr Conte if you know the answer. I don't know that specific number.
DR JOHN V. CONTE (Baltimore, MD): Well, in the U.S. it is difficult to refer patients for an AICD after surgery because of the way payments are. The defibrillators will not be paid for if they are to be put in immediately after. Early in our experience, however, we did go and do routine EP (electrophysiologic) testing and found very few patients who actually were inducible. So what we typically do now is when they are discharged, and they are usually discharged to our heart failure clinic for follow-up, at 3 months' time if their ejection fraction is not above 35%, they would have a test performed to see if they do need a defibrillator. Most of the time that isn't the case. These patients do have significant improvement. So they don't meet typical MADIT II (Multicenter Automatic Defibrillator Implantation Trial II) criteria.
DR ADRIAN JEREMY LEVINE (Stoke-on-Trent, UK): I have done over 85 SVRs. I have just got a few questions to ask. Firstly, what tests of viability do you use for your SVR candidates? Secondly, how often do you see ventricular tachydysrhythmia after your SVR procedures? And finally, what measures of RV (right ventricular) performance do you use preoperatively to assess whether someone is an SVR candidate or not?
DR WEISS: Again, those are sort of technical questions that are difficult for me to answer. I will probably ask Dr Conte to comment.
DR CONTE: Eric is a third-year surgery resident in the lab, so some of these are a bit above his level of training.
For most of our patients, they would have an MRI (magnetic resonance imaging); that is our viability test of choice. If they don't, they would have had a thallium scan or a dobutamine echo[cardiography] in most cases.
What was your second question?
DR LEVINE: The percentage of patients who will have any form of ventricular tachydysrhythmia after SVR.
DR CONTE: In contradistinction to some other people who have reported that, we have had a very low incidence of that. In fact, Marisa Di Donato from Italy has actually put together a group of clinical investigators in a study that she has called "electricity" to actually look at that very question. But our clinical incidence is very low after SVR and no higher than it is after routine CABGs, even those with low ejection fraction CABGs.
DR LEVINE: My final question was about RV performance.
DR CONTE: We really haven't done much in terms of evaluating RV performance because that hasn't been a problem, even in those patients who have calculated low RV stroke work indices, which early on we were very concerned about that, particularly when early on Dr Dor and Dr Menicanti had actually talked about that as being a contraindication to doing this procedure. We had not found that to be a major problem. So in fact we don't even look at that anymore.
DR BOBBY KONG (Ann Arbor, MI): A question for Dr Conte. Is your SVR technique a Dor procedure or do you have some linear reduction procedure? And if you use the Dor procedure, how do you size the LV postop[eratively] or intraop[eratively] and how do you estimate the size of your patch?
DR CONTE: I routinely have used ventricular sizers. There are a couple that are commercially available and I have used both of those almost equally, and I shoot for a volume of between 50 and 60 mL/m2. That is basically the way that I had learned it from Lorenzo Menicanti and I do that most of the time. Most of the repairs are a Dor type repair; about 40% are with patches, the remainder without patches. Usually it is the larger ones that require patches so that we don't narrow down, or reduce, the size of the anterior part of the ventricle too much, and then by patching, it seems that you reduce the tension you put on the papillary muscles. So we reduce the amount of MR (mitral regurgitation) afterwards in those patients with the big ventricles that we patch.
DR MEHMET HAKAN AKAY (Houston, TX): I was wondering how often you need to have an assist device available; how often you place these patients on an assist device?
DR CONTE: Balloon pumps are about 15%, ventricular assist devices in a handful of patients, but it is certainly a very real incidence. The incidence is most common in those people who have had infarctions within 30 days of the time of their operation. Those are the ones who we found are most at risk. And the other group of patients who are at high risk for need for an assist device are those people who do not have clear-cut evidence of ischemia on preoperative testing. So those patients who we revascularize and we improve their myocardial blood flow and having documented that they needed improvement in their blood flow tend to be the patients that require it more commonly. But it is really just a handful of patients.
DR ROHIT SHAHANI (Poughkeepsie, NY): Thank you for an interesting paper. My question is, do you have any data regarding ventricular volumes and what you use as objective measurements so as to reduce the size of the ventricle and how much do you do? Most of the literature reports the data as a measure of the ventricular volumes. I saw you presented data on ventricular dimensions in end diastole, but I didn't see any objective data on volumes. Could you please tell us something about that?
DR CONTE: Preoperative volumes or postoperative volumes?
DR SHAHANI: Pre and post, when you say that this is how much I need to reduce the size of the ventricle to.
DR CONTE: Well, one of the nice things with the sizing devices, whichever ones that you would choose to use, is that you calculate the size of the ventricle based on their idealized body surface area, and so the volume that we aim for postoperatively is that volume. Regardless of whether you use a patch or not, that is the volume you should attempt to reach.
We do have the volumes. These were all large ventricles, all were class III and IV heart failures, so they were the typical big ventricles.
DR SHAHANI: The mean end-diastolic dimension was 56 and 58 mm, which was in that big range but not the giant range that you normally see where we lean toward doing an SVR. That is why the question on what objective evidence we should use on when to do the SVR.
DR CONTE: Well, I don't think anybody can really tell you that there is an ideal volume. Any volume used as a guide should be indexed to the body surface area. In fact, based on your question, I would say you should probably do it when you think it is abnormal. If you clearly have an area of akinesis or dyskinesis in a ventricle that is enlarged at all, I think we have shown that you have less heart failure readmission with comparable mortality. So just from the standpoint of treating their heart failure, if you have a patient that has got heart failure and their ventricle is enlarged with a discernible area of akinesis or dyskinesis, you are going to do the patient, and the health-care system, a benefit by doing that operation and keeping them out of the hospital more commonly than you would without it.
DR SINISA GRADINAC (Belgrade, Serbia): I would like to ask Dr Conte if the patients that were described here were a part of the RESTORE (Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis) study, because the NIH (National Institutes of Health), as you know, sponsored the STICH (Surgical Treatment for Ischemic Heart Failure) trial in which we have randomized 1,000 patients of which 500 received SVR and CABG and the other 500 only CABG. This randomization was finished 12 months ago and now we are awaiting the NIH results. Thank you.
DR CONTE: I think we are all anxiously awaiting the results of the STICH trial. Some of these patients were enrolled in STICH; there was a handful of them. I think one of the things that you have an advantage of in a prospective randomized trial is obviously the fact that you are going to have large volumes and you are going to exclude any single-center biases that happen. On the other hand, if you are going to truly evaluate a surgical procedure, having a single operator will certainly induce some controls in the conduct of a surgical operation that you are not going to have in a randomized multicenter trial. So I think trials from single institutions add some information, but certainly we are looking forward to the results from the STICH trial.
DR NEIL J. THOMAS (Chicago, IL): I have just a quick question to ask your group about the mitral valve. I couldn't help but notice on one of your slides, which showed that in the SVR and CABG group; you were doing a higher proportion of mitral valve annuloplasty in combination for what I presume was associated MR. I have seen a number of patients come back with clinically significant MR, unfortunately, one of whom I performed SVR and CABG on. And I have seen several other patients, operated on by others, who have come back after SVR with increasing mitral insufficiency. Have you given any thought or has your group given any thought to looking at doing a mitral valve ring or posterior annuloplasty band in all of those patients in an adjunctive manner; sort of along the lines of the concept that an annuloplasty is a type of its own SVR?
DR CONTE: It is a very good question, and over the years I have gone back and forth on that. There actually were some patients with mitral regurgitation that we have just gone and done reduction of the interpapillary muscle distances to reduce the tethering on the papillary muscles and the chordal apparatus. We are also very careful when we are going and putting in the anterior purse string. If we go and tighten that too much, we are going to inadvertently stretch those papillary muscles and cause them to splay out a bit, and in fact you can increase the amount of MR. In fact, there was an abstract about 5 years ago from a Japanese center—and I apologize for not remembering which one it was—and they showed about a 30% incidence of that in patients who had primary anterior closures and did not patch patients. So I think that is critically important when you do that, and I think that is one of the advantages that using one of these sizing devices will add to it.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Isomura, J. Hoshino, Y. Fukada, A. Kitamura, S. Katahira, T. Kondo, T. Iwasaki, G. Buckberg, and from the RESTORE Group Volume reduction rate by surgical ventricular restoration determines late outcome in ischaemic cardiomyopathy Eur J Heart Fail, April 1, 2011; 13(4): 423 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. W. Skelley, J. G. Allen, G. J. Arnaoutakis, E. S. Weiss, N. D. Patel, and J. V. Conte The Impact of Volume Reduction on Early and Long-Term Outcomes in Surgical Ventricular Restoration for Severe Heart Failure Ann. Thorac. Surg., January 1, 2011; 91(1): 104 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Prucz, E. S. Weiss, and J. V. Conte Reply Ann. Thorac. Surg., August 1, 2009; 88(2): 709 - 709. [Full Text] [PDF]
|
|
|
|
|
|
T. Nakamura Surgical Ventricular Restoration: Who Really Benefits? Ann. Thorac. Surg., August 1, 2009; 88(2): 708 - 709. [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 |
