| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiac Surgery, University of Catania, Catania
b Department of Cardiac Surgery, Papardo Hospital, Messina
c Department of Cardiology, University of Chieti, Chieti, Italy
Accepted for publication March 26, 2008.
* Address correspondence to Dr Calafiore, Division of Cardiac Surgery, University of Catania, Ferarrotto Hospital, Via Citelli, Catania, 95124, Italy (Email: calafiore{at}unich.it).
Presented at the Forty-fourth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: From November 1994 to December 2002, 4226 patients (EF > 0.30) underwent a first isolated CABG. Preoperative IMR was present in 1421 (33.6%, group IMR), of which 1254 had mild (1/4) and 167 had moderate (2/4). The remaining 2805 patients (66.4%, group no-IMR) showed no IMR. A nonparsimonious regression model was built to determine the propensity score. Ten-year freedom from death from any cause, cardiac death, and cardiac events was evaluated by the Kaplan-Meier method. Results of Cox analysis were adjusted by entering the propensity score as an independent variable.
Results: All patients had similar early mortality (2.1% no-IMR vs 2.5% IMR, p = 0.502) and morbidity (6.5% no-IMR vs 6.6% IMR, p = 0.840). In patients with EF of 0.31 to 0.40, but not in those ones with EF exceeding 0.40, IMR grade was an independent variable for worse long-term freedom from cardiac death (82.8 ± 3.2 vs 91.4 ± 2.4; Cox hazard ratio [HR], 2.1 [95% confidence interval (CI), 1.1 to 4.1]; p = 0.0324) and cardiac events (78.6 ± 3.5 vs 88.5 ± 2.7; Cox HR, 2.0 [95% CI, 1.1 to 3.7]; p = 0.0174).
Conclusions: Mild or moderate IMR in patients with an EF exceeding 0.30 undergoing first isolated CABG influences long-term outcome when EF is 0.31 to 0.40, but not when it exceeds 0.40.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
In a previous study [5], we demonstrated the prognostic value of uncorrected moderate IMR (2/4) in a cohort of patients with an ejection fraction (EF) of 0.30 or lower. However other studies, comparing patients with similar IMR grade who underwent coronary artery bypass grafting (CABG) alone and CABG with mitral valve annuloplasty, failed to demonstrate any clear superiority of mitral valve procedures [9–13], at least in terms of survival.
Other authors [8, 14] have recently supported the idea that even mild IMR, independently from left ventricular (LV) function, can impair mid- and long-term survival of patients undergoing isolated CABG. The purpose of this retrospective analysis was to evaluate if the presence of mild or moderate IMR could affect the long-term outcome of patients with EF exceeding 0.30 undergoing isolated CABG.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
|
Surgical Technique
Cardiopulmonary bypass was used in 2119 patients (50.1%), and myocardial revascularization was performed without cardiopulmonary bypass in the remaining 2107 (49.9%). The on- and off-pump techniques used have already been described [15].
End Points and Definitions
The primary end point was to evaluate the effect of mitral regurgitation on 10-year freedom from death any cause, cardiac death, cardiac events, and any event. The grade of IMR was differently assessed over the time: regurgitant jet area or jet/left atrium areas ratio, regurgitant volume, effective regurgitant orifice, and Doppler color flow mapping of the vena contracta [16–18]. The EF was calculated according to the biapical Simpson disk method [5]. A preoperative routine transthoracic echocardiogram was performed in all patients by a single expert in echocardiography (S. G.).
Cardiac death was defined as cardiac-related death or sudden death; cardiac event, as the occurrence of at least one of following: cardiac death, MI, heart failure (New York Heart Association functional class III to IV), or surgical or interventional reoperation; any event, as the occurrence of at least one of the following: death any cause, MI, heart failure, or surgical or interventional reoperation. MI was defined as enzymatic elevation, electrocardiograph sign of necrosis, new akinetic segment(s) at echocardiogram, or ventricular arrhythmias non-K+ related.
Follow-Up
All patients were followed up in our outpatient clinic at 3, 6, and 12 months postoperatively and thereafter at yearly intervals. The more recent information was obtained by calling the patient or the referring cardiologists. Follow-up was 100% completed from March to June 30, 2006. Median (25th to 75th percentiles) follow-up in months was 102 (83 to 125), consisting of 93 (76 to 118) in group no-MR and 112 (98 to 131) in group MR (p < 0.001).
Statistical Analysis
Results are expressed as mean value and standard deviation for normally distributed continue variables and as median (25th to 75th percentile) in case of nonnormally distributed continuous variables. The Kolmogorov-Smirnov statistic with a Lilliefors significance level was used for testing normality. Categoric variables are reported counts and percentages. Statistical analysis comparing the two groups was performed with unpaired two-tailed t test for continuous normally distributed variables, with the Mann-Whitney U test for continuous nonnormally distributed variables and with Pearson 
2 test (or the Fisher exact test) for categoric variables.
A nonparsimonious regression model was solved to generate a propensity score for each patient, representing the retrospective probability of being in group IMR [19] (Appendix). The model fit and predictive power were evaluated using the Hosmer-Lemeshow goodness-of-fit and C statistic, respectively. The model was validated in 500 bootstrap samples [20].
|
0.2 were then entered in the Cox regression. Final models were validated in 500 bootstrap samples. Factors appearing in 50% or more of the analyses were considered reliably statistically significant. The results of Cox analysis were reported as instantaneous relative risk ratio (hazard ratio [HR]), 95% confidence interval (CI), and p value. For all tests, a value of p < 0.05 was considered significant. Analysis was done with SPSS software (SPSS Inc, Chicago, IL). |
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Late Outcome
Of 4131 patients who survived the first postoperative month, 209 (5.1%) died during the follow-up, consisting of 124 in the no-IMR group, 72 in the mild IMR group, and 13 in moderate IMR group. Death in 105 patients (2.5%) was cardiac-related (64 in no-IMR group, 34 in mild IMR group, 7 in moderate IMR group). A new cardiac event occurred in 276 (6.7%), comprising 178 in the no-IMR group, 86 in the mild IMR group, and 12 in the moderate IMR group), whereas 380 (9.2%) had a new event (238 in the no-IMR group, 124 in the mild IMR group, and 18 in the moderate IMR group).
The 10-year freedom from death any cause, from cardiac death, from cardiac events, and from any event was 91.3 ± 0.5, 95.1 ± 0.4, 89.6 ± 0.6, and 85.8 ± 0.6, respectively. Cox analysis, adjusted for propensity score, identified the following independent variables for worse long-term outcome: age, extracardiac vasculopathy, chronic heart failure, chronic renal failure, EF (continuous), EF 0.31 to 0.40, diabetes mellitus, female sex, and left main disease; moreover, analysis confirmed that presence of IMR was not a risk factor for worse long-term clinical results (Table 2). IMR grade was also investigated by means of Cox analysis without achieving statistical significance as a risk factor.
|
The presence of IMR was an independent variable for worse cardiac outcome, with a HR of 2.1 (95% CI, 1.1 to 4.1) for cardiac death and 2.0 (95% CI, 1.1 to 3.7) for cardiac events. No difference was found between patients with or without IMR in case of an EF exceeding 0.40 (Fig 3, 4).
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
the available clinical data strongly suggest that the surgical treatment of IMR results in little, if any, survival benefit or reverse remodeling. It may be explained by the fact that slowly progressive MR due to any etiology represents a mild stimulus for remodeling that takes place to impact the ventricle, while a moderate sized MI is an intense and immediate stimulus for remodeling that is orders of magnitude more severe.
On the other side, Levine and colleagues [23] wrote:
MR, caused by altered geometry and function after acute MI, can itself initiate remodeling. MR alters LV loading; it increases diastolic wall stress, which can induce LV dilation and failure, and end systolic wall stress, with decreased contractility and increased end-systolic volume. Because of this vicious circle, MR begets more MR.
Most articles report on the natural history of untreated IMR in different settings—clinical [1, 2], interventional [24], and surgical [5–8, 14, 25]. In particular, Mallidi and colleagues [25], compared 163 patients undergoing isolated CABG with mild to moderate MR with 336 patients without MR. They found, 6 years after operation, a higher prevalence of heart failure symptoms and decreased cardiac event-free survival in patients with either mild or moderate IMR. A prospective study by Schroder and colleagues [14] showed that intraoperative mild or moderate IMR significantly impaired 5-year survival and event-free survival after isolated CABG. The same findings were confirmed at the 10-year follow-up by Grossi and colleagues [8]. Furthermore, there is clear evidence that CABG alone is not able to stabilize IMR, because its natural history seems to be independent from the revascularization [12, 26].
However, no study was able to confirm that correction of IMR could have a favorable impact on long-term outcome in comparable patients where IMR was treated or not treated. The main problem is survival. Unfortunately, early mortality in patients where mitral annuloplasty is performed for IMR is two times higher than in patients with a similar grade of IMR where isolated CABG is performed (5% vs 10%) [9, 10, 13, 26–30]. In these patients, where long-term survival is ruled by the EF, it is difficult to regain, in a relatively short period, 5 points of survival.
The rationale for surgical treatment of mild to moderate IMR depends on the natural history of untreated patients. Our study shows that in a large series of patients with a long mean follow-up, mild or moderate IMR impairs the long-term outcome of patients with an EF between 0.31 and 0.40, but not in patients with EF exceeding 0.40. This finding is coherent with other studies. Ellis and colleagues [24] reported 3-year survival of 4221 patients undergoing percutaneous coronary interventions. The 3-year actuarial survival differed significantly among patients with differed IMR grades (92.3% no IMR, 84.5% mild IMR, 74.6% moderate IMR, and 68.6% severe IMR, respectively, p < 0.001). This difference was more notable in patients with EF of 0.40 or less (75.7%, 66.9%, 53.7%, and 46.5%; p < 0.001) and less pronounced in patients with an EF exceeding 0.40 (93.5%, 89.6%, 82.2%, and 86.5%; p < 0.001) but with a lesser hazard gradient.
In a previous article [5], we demonstrated the negative impact of untreated moderate IMR on long-term outcome when EF was 0.30 or lower. The findings of both studies move the problem of untreated IMR toward the lowest values of EFs. In these patients, very likely the contribution of IMR, even if of low grade, to the LV remodelling process is concrete and inversely proportional to the EF; as a consequence, treating IMR can be beneficial.
We are aware that the main limitation of our study is the retrospective nature of the analysis. The adjustment of results by means of balancing score (propensity score) allowed us to reduce the misleading power of the selection bias. Another limitation of this study is the varying MR assessment methodology over time.
Moreover, we are not able to provide echocardiographic controls to support the hypothesis that untreated mild or moderate IMR increases in patients with a lower EF, even if this concept is clearly reported in the literature. At the same time, the lack of postoperative echocardiographic late controls does not allow us to speculate on the prognostic role of MR progression.
In conclusion, our study focuses the negative impact of mild or moderate IMR in patients with an EF of 0.31 to 0.40, but not in patients with an EF that exceeds 0.40. A randomized study comparing clinical outcome and MR progression between MV repair and isolated revascularization is mandatory to clearly identify the gold standard treatment in this subset of patients.
|
Appendix |
|---|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
And then the second question would be did you have intraoperative echo to document that you still had mitral regurgitation at the time of surgery?
DR CALAFIORE: Perhaps the term ischemic mitral regurgitation is not a perfect term, because we think that the patient can have a transient mitral regurgitation. All these patients had mitral regurgitation that followed a complete acute myocardial infarction. So the regurgitation was not really ischemic, but we can call it postischemic. All the patients had, anyway, in the operating room, a transesophageal echocardiography that documented the presence of mitral regurgitation.
DR MARTIN: And just a little follow-up to that. Were all of these patients transmural or were some of these subtransmural?
DR CALAFIORE: Yes, transmural. Absolutely.
DR DANIEL H. DRAKE (Traverse City, MI): First, I wish to congratulate Dr Calafiore on an excellent paper. I think this is a hallmark paper that will likely shape our practices for many years.
I have two comments. The presentation specifically identifies that the etiology of the mitral regurgitation is "ischemic." It would be difficult to conclude that this diagnosis was uniformly responsible for the observed mitral insufficiency. Certainly there would have been patients with lesions such as minor congenital abnormalities, mild rheumatic disease, et cetera, that would result in 1+ to 2+ mitral regurgitation but are causally unrelated to coronary ischemia. To justify the use of the term "ischemic," there should be clear evidence of wall motion abnormalities, leaflet displacement below the plane of the annulus during systole, and/or a reduced coaptive surface in an otherwise normal appearing valve. This diagnosis is usually established through reasonably high-resolution anatomic imaging such as transesophageal echo or gated magnetic resonance imaging. This level of detail was not presented and would be very difficult to uniformly obtain at most medical centers. I would further point out that most of your patients had a normal ejection fraction and relatively few had been documented as having previous myocardial infarctions. Therefore the use of the term "ischemic" mitral regurgitation does not appear to be justified.
Interestingly, removal of the term "ischemic" from the manuscript does not diminish the clinical importance of this work. In fact, it may enhance it. In practice the exact etiology of 1+ to 2+ mitral regurgitation in the coronary bypass population with ejection fractions greater than 30% is usually unknown and may be clinically unimportant. As long as significant Carpentier type II dysfunction (prolapse) is excluded, surgical revascularization should probably proceed as you conclude. Please comment on the applicability of your findings independent of direct documentation of the regurgitation etiology.
The second point that I ask you to address has to do with your definition of adverse "cardiac events." There is no specific mention of late follow-up with surface echocardiography. It is therefore conceivable that patients with 3+ or 4+ mitral regurgitation, but only New York Heart Association class II symptoms, might not be classified as having a "cardiac event" or otherwise adverse outcome. As those patients would generally be considered to have failed conservative mitral management, they should be identified as having a "cardiac event." Do you have any data to support the long-term stability of the observed 1+ to 2+ mitral regurgitation? Again, I commend you for this important study and the excellent presentation.
DR CALAFIORE: I must tell you that in this data it was difficult to have as being a long-lasting experience. And this is one of the limitations of a retrospective study. We can follow these patients very carefully by telephone or just looking at them. But most of them will not come to the hospital. There are no more echo studies. So we have not exactly all the information we need. So I think that the issue of the progression of the regurgitation cannot be addressed by this paper.
Coming back to what you said before, I understand that the problem of the etiology of the mitral regurgitation can be an issue, and I agree with you. However, in these cases we can see with the presence of any kind of reduced motion of the leaflet as one point that addressed us to the ischemic etiology of the mitral regurgitation. Anyway, I agree with you that sometimes we can have moderate mitral regurgitation that the etiology really is not ischemic but can be all other disease.
DR ROBERT A. E. DION (Genk, Belgium): Antonio, you know my interest in ischemic mitral regurgitation. And your paper threatens to induce the idea that you should not correct moderate ischemic mitral regurgitation.
I have two questions. First, if you face a mild to moderate ischemic mitral regurgitation, I believe that you should submit the patient to a stress test, either preoperatively as proposed by Pierard and his group, or either intraoperatively as we published earlier.
Did you ask in these patients a preoperative stress test, and in how many cases did the MR worsen?
My second question is about patient selection. Ischemic mitral regurgitation is going to be more present between 30% and 40% than above 40%. Did you specifically consider this group of patients, and are the general conclusions of your paper also applicable to them?
DR CALAFIORE: In coming to your first question, I agree that this is a limitation of the retrospective study. So during 14 years of experience, you do not follow exactly all the same concepts. So at the beginning we were a little bit more reluctant to operate on patients with mild or moderate mitral regurgitation. We have a certain number here. With time, things change. So I can really answer what I believe today, after 14 years of experience, I agree with you that mild or moderate mitral regurgitation today need, perhaps, to be treated if it is demonstrated. You utilize a stress test; I personally check more carefully the mitral anatomy. So if I see that the patient has any huge septolateral distance and the anatomy of the mitral valve is abnormal, now I undersize the mitral valve because I think that this will be helpful for the patient. But this is an evolving concept. And honestly, I cannot say that I did all those things in these patients.
Coming to the second question of yours ...
DR DION: Between 30% and 40%.
DR CALAFIORE: Yes. This is another gray zone of this presentation. I think that if we could go carefully looking at all the small subgroups of ejection fraction, very likely the patients between 30% and 40% needed to be more carefully evaluated. Unfortunately, again, this is one of the limitations. We have not so many patients to have such a complete division in small subgroups. But I think that even if the general concept is the concept I said, patients between 30% and 40% needs today to be more carefully evaluated, what we did not do in the past.
DR ROBERT S. D. HIGGINS (Chicago, IL): Dr Calafiore, have these observations changed your practice?
DR CALAFIORE: I think that there is another component that perhaps needs to be put in this scenario. The volumes are also not considered in this paper. So we can have a 35% patient with mild or moderate mitral regurgitation and a big heart, or we have a normal heart. It depends on this. Anyway, this is a strength of the problem that we have now. So many of us are doing surgery because they believe it is true. But anyway, the long-term data, the long-term survival, certain surgical patients show today that it is still something that needs to be better evaluated. Under certain aspects it changed my practice, because surely for patients with 40% or higher ejection fraction, I do not treat moderate mitral regurgitation. I am more careful for the patient with the lower ejection fraction, even if this paper didn't demonstrate any real advantage. But today we have more accurate possibility to diagnose the mitral regurgitation that we had not 14 years ago.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. Makhija and J. Desai Early and mid-term functional and survival benefits in ischaemic versus degenerative mitral valve repair using Duran flexible ring: a single surgeon series Interactive CardioVascular and Thoracic Surgery, September 1, 2009; 9(3): 471 - 475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Silberman, M. W. Klutstein, T. Sabag, A. Oren, D. Fink, O. Merin, and D. Bitran Repair of Ischemic Mitral Regurgitation: Comparison Between Flexible and Rigid Annuloplasty Rings. Ann. Thorac. Surg., June 1, 2009; 87(6): 1721 - 1727. [Abstract] [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 |
