| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:570-577
© 2005 The Society of Thoracic Surgeons
a Division of Cardiac Surgery, Massachusetts General Hospital, Boston, Massachusetts
c Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts
b Maritime Heart Centre, Halifax, Nova Scotia, Canada
Accepted for publication March 7, 2005.
* Address reprint requests to Dr Torchiana, Division of Cardiac Surgery, Massachusetts General Hospital, 55 Fruit St, BUL 209, Boston, MA02114 (Email: torchiana.david{at}mgh.harvard.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: We retrospectively reviewed 251 patients undergoing coronary artery bypass graft (CABG) surgery between 1991 and 2001 with 3+ ischemic MR, including 31 patients who had concomitant mitral annuloplasty. Univariate and multivariable testing was employed.
RESULTS: Actuarial 1-, 5-, and 10-year survival was 84.0%, 67.5%, and 37.1% in the overall group of 251 patients. Independent predictors of long-term mortality were age 70 years or more (hazard ratio 2.50 [95% confidence interval 1.82 to 3.44]), prior myocardial infarction (3.99 [2.15 to 7.39]), unstable angina (2.27 [1.69 to 3.04]), chronic renal failure (4.87 [3.13 to 7.58]), atrial fibrillation (2.21 [1.65 to 2.96]), left internal mammary artery to left anterior descending artery graft (0.28 [0.18 to 0.43]), preoperative ß-blocker (0.43 [0.28 to 0.67]), ejection fraction (0.71/10% [0.64 to 0.80]), left atrium size (0.88/mm [0.84 to 0.92]), diffuse wall motion abnormalities (2.83 [1.77 to 4.55]), and mitral leaflet restriction (3.85 [2.46 to 5.99]). The model controlled for the performance of annuloplasty, which did not emerge as an independent predictor. Patients undergoing annuloplasty did have less mean late MR than those having CABG alone (p = 0.005). Overall, 57.8% of patients (63 of 109) with follow-up echocardiograms had improvement in grade of MR compared with baseline. In 54 of 95 patients (56.8%), intraoperative transesophageal echocardiography downgraded the degree of MR compared with the preoperative study.
CONCLUSIONS: Patients with moderate ischemic MR undergoing CABG had relatively poor long-term survival, with significant differences when stratified according to preoperative characteristics. Performance of mitral annuloplasty reduced the degree of regurgitation but was not a predictor of long-term survival. Intraoperative transesophageal echocardiography frequently downgraded the degree of MR.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In addition to surgical revascularization, restrictive annuloplasty has become accepted as standard therapy for IMR [6]. While there is widespread agreement that trace or mild regurgitation is adequately treated by isolated CABG, and that mitral annuloplasty or replacement is appropriate for severe MR, the optimal surgical approach to moderate IMR remains controversial [7, 8]. It is in this gray zone of surgical decision-making that the long-term results of surgery and the influence of patient characteristics are most important to decipher. When the variability of IMR under different conditions of loading and ischemia is added to the fact that most studies group several grades of MR together with several approaches to therapy, it is not surprising that there is no clear consensus within the literature on what are the critical factors, including surgical approach, that influence outcomes.
In this report, we retrospectively reviewed patients undergoing CABG with or without annuloplasty who had moderate (3+) IMR and attempted to identify characteristics that influence long-term survival and degree of late MR.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Surgical details
The use of intraoperative transesophageal echocardiography (TEE), and the decision to perform a mitral annuloplasty were at the discretion of the surgeon. The prevailing strategy at the beginning of this interval favored performing isolated CABG particularly when TEE showed downgrading of MR. Annuloplasty rings were generally not undersized in the early part of the series.
Patient characteristics and follow-up
Baseline demographics, comorbidities, operative details, and short-term in-hospital outcomes were extracted retrospectively. Echocardiographic follow-up was restricted to studies performed at this institution and was obtained similarly from chart review. Follow-up echocardiograms were performed as clinically indicated. Long-term survival was the primary outcome and was assessed by searching a national social security death index; it excluded 9 patients (3.6%) from outside the United States who were censored at discharge. Survival follow-up ended on October 1, 2003.
Data Analysis
Univariate analyses
All statistical analysis was performed using SAS software (SAS Institute, Cary, North Carolina). Univariate analyses included two-tailed Wilcoxon rank sum or t tests for continuous variables and 
2 or Fisher’s exact tests for discrete variables. Survival was estimated by the Kaplan-Meier technique, with log-rank tests to compare stratified groups. Tests are considered statistically significant at p less than 0.05. No correction for multiple analyses was utilized.
Multivariable analyses
Time-specific effects were estimated with parametric, multiphase (mixture distribution components) hazard regression, using analytical tools available from the Cleveland Clinic Heart Center Hazard website (available at: http://www.clevelandclinic.org/heartcenter/hazard/default.htm). Covariates were modeled in both discrete and continuous forms after assessing logistic plots of continuous and categorical variables. Missing values were assigned the mean for that covariate. Variable selection for the multivariable model was performed by first considering each potential covariate in each phase of the model univariately. Those significant to p less than 0.10 were considered for inclusion in the model using automated stepwise selection, with manual oversight of each step, using entry and stay criteria of p less than 0.10 and p less than 0.05, respectively. While not statistically significant itself, mitral annuloplasty was forced into both phases of the final model as a clinically relevant variable; it was not an important confounder (less than 10% change in the effect estimates for the other variables) but did reduce the power of the model slightly. With this model we were able to distinguish between risk factors that acted during (two) different phases of follow-up, and to avoid assumptions about the proportionality of hazards inherent to Cox modeling. Model validation of the significant predictors (not including annuloplasty) was performed by calculating estimates for the final model using 1,000 datasets, each containing 251 patients selected from the original set with replacement (bootstrap method); the standard deviations of the means of the bootstrap estimates were used to test the original parameter estimates from the final model by means of a 2 test. Logistic regression was used for multivariable analysis of secondary echocardiographic outcomes.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
Late Echocardiographic Results
Follow-up echocardiograms were performed at our institution in 109 patients (Table 2). All patients had at least 1+ MR at a median of 1 year. Overall, the mean grade of MR was lower than preoperatively, and the annuloplasty subgroup had less mitral regurgitation (mean of 2.0 versus 2.6 for CABG alone, p
mean = 0.005). There was no association between death and either improvement (<3+, p = 0.40) or deterioration (>3+, p = 0.19) in the degree of MR at follow-up among the CABG only patients, and logistic regression analyses identified chronic renal failure (OR = 0.215; p = 0.02), left atrium size (0.87 per mm; p = 0.01), and inferior dysfunction (0.32; p = 0.06) as inversely associated with improvement in MR. There was no significant difference found in progression of the late grade of MR related to intraoperative downgrading (>3+ MR occurred in 2 of 6 [33.3%] nondowngrading patients versus 5 of 26 [19.2%] downgrading patients; p = 0.59).
Survival
Patients with 3+ IMR are sicker than the standard CABG population, and operative mortality was relatively high at 5.2% compared with standard CABG. There was no significant difference in mortality for concomitant annuloplasty compared with CABG only (p = 0.06; Table 3) although there was a trend, and annuloplasty patients did have more short-term complications, reoperation, and a longer postoperative length of stay. Over a median follow-up of 4.3 years, actuarial 1-, 5-, and 10-year survival was 84.0%, 67.5%, and 37.1% in the whole group of 251 patients. Two phases of hazard were resolved using a multiphase parametric model: a rapidly declining early phase of risk that gave way after the first year to a slowly rising late phase (Fig 1). Mortality was associated with a number of independent risk factors in the early and late phases: age 70 year or more, prior myocardial infarction, unstable angina, renal failure, atrial fibrillation, lack of left internal mammary artery to left anterior descending artery graft, lack of ß-blocker use, lower ejection fraction, smaller left atrium size, global wall motion abnormalities, and mitral leaflet restriction (Table 4). We controlled for mitral annuloplasty in this model, even though it was not independently associated with long-term survival (p = 0.52 for the early phase, p = 0.41 for the late phase). There were 83 deaths (37.7%) among CABG only patients and 13 (41.9%) in the annuloplasty group, so that 5- and 10-year actuarial survival estimates were 68.2% and 36.3% for CABG only, compared with 62.5% and 36.2% for concomitant annuloplasty (Fig 2). All of the significant predictors were validated by the bootstrap analysis.
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
All of the available studies of CABG for IMR are retrospective and observational and therefore have flaws. They fall into three broad categories: (1) studies of CABG alone or with mitral surgery; (2) results of isolated CABG, with or without a non-IMR control group; and (3) annuloplasty series (Table 5).
Coronary artery bypass graft surgery with and without mitral surgery for IMR
One problem with prior reports examining CABG with and without annuloplasty is that eligibility was not limited to a single preoperative grade of MR, with the degree of MR being invariably lower in the CABG only patients. Unfortunately, this systematic difference between these two groups produces an important bias that in a retrospective study may not always be correctable by statistical adjustment. Furthermore, when a range of MR is included in such studies, the predominant approach selected tends to alter the results of surgery for the two groups. For example, where annuloplasty is aggressively favored, patients with lower degrees of MR and risk tend to fall into the annuloplasty group, making the outcomes in this group more favorable. Surgeons who are convinced of the value of this approach can then point to improving results as justification, basically a self-fulfilling prophecy. On the other hand, when annuloplasty is undertaken only at a higher threshold, the opposite is true—risk is higher in the annuloplasty group and the outcomes poorer—again affirming the wisdom of the original bias. By including only patients with 3+ MR, our study has less tendency toward bias in these two very important regards.
Despite these issues, our findings are generally consistent with the prior literature with regard to the impact of patient characteristics and of mitral annuloplasty. Most studies report high rates of late mortality, demonstrated the detriment of poor ventricular function and comorbidities, and did not find any difference in survival with and without annuloplasty [12, 13, 15, 16]. In contrast, Prifti and associates [17] showed improved 3-year survival with annuloplasty. This study, however, was restricted to a narrow range of low EF (17% to 30%) with 2 to 3+ IMR and is complicated by including complex repairs and replacement, as well as structural valve disease (prolapse in 8% to 10%, which was otherwise untreated in the CABG alone group).
Isolated CABG series
There are numerous studies that fall within the realm of one-operation reports, including those reporting the results of CABG in patients with and without IMR. While there are some compelling data that MR following CABG has a deleterious effect on survival [18], the results in the literature are mixed. In general, controlling for left ventricular function appeared to lessen or eliminate the effect of MR on survival: while on an unadjusted basis, IMR significantly increased 5-year mortality after CABG [19], several matched studies—all of which included ejection fraction or left ventricular grade as a matching variable—found no effect of MR on survival [10, 11, 20]. Similarly, Paparella and associates [21] found that MR did not emerge as an independent predictor after multivariable adjustment or stratification by left ventricular function. These results support the notion that ventricular function, not the manifestation (or absence) of MR itself, may be the critical operand influencing survival. In our study group, while there was an overall reduction in regurgitation after CABG, a fair degree of recurrent MR occurred in some patients. We did not find a convincing correlation between the improvement in MR and survival. There are small series of patients with IMR that report improved symptoms, EF, and grade of MR after CABG [22, 23].
Annuloplasty series
Recent experiences demonstrating low operative mortality for annuloplasty (as low as 4% [24, 25]) have been reported (versus more than 10% in other reports [18, 26]) in diverse patients including those with cardiomyopathy [27] and IMR [28]. Comparison with these studies is more difficult since CABG was not always performed, owing to combined repair and replacement procedures, or because of other differences in the patients studied including a broad range of severity of MR. The translation into a long-term survival advantage is also lacking.
Intraoperative Downgrading of MR
In our study a majority of patients downgraded to less than 3+MR, including 71% in the CABG only subgroup. After surgery, many (48%) went on to have persistent 3+ or greater MR on follow-up echocardiography when treated with CABG alone. Aklog and coworkers [29] previously reported an 89% frequency of downgrading followed by 40% recurrence of at least 3+ MR within 6 weeks in patients undergoing CABG with 3+ IMR preoperatively. This phenomenon has important implications for research and clinical practice, and the high frequency of its occurrence is worth emphasizing.
Limitations
Our analyses are limited by important considerations. Owing to our selective eligibility criteria (particularly the need for a preoperative echocardiogram at our institution), many patients who underwent CABG or mitral annuloplasty, or both, for IMR at Massachusetts General Hospital during this interval were not included in this report. This study was retrospective, which complicates risk adjustment, as statistical methods cannot adjust for unmeasured characteristics. Furthermore, the study design makes inferences from comparisons between nonrandomized CABG and CABG plus annuloplasty patients problematic as there may be some residual bias after multivariable adjustment. Therefore, even though it did not appear to be a significant predictor or confounder in our analyses, we still chose to include this variable in our model. It is possible that the annuloplasty patients would have fared worse, were it not for the benefit of their annuloplasty. In addition, by performing nonrestrictive annuloplasty during much of this era, annuloplasty was perhaps only partly effective and left more residual MR than one would like. Whether this has implications for late survival remains uncertain, particularly in light of animal studies that do not demonstrate benefit to prophylactic annuloplasty [30]. Not all patients had follow-up echocardiography studies, and there is a risk of measurement bias, with symptomatic patients and those with physical findings of heart failure more likely to be restudied, which would lead to a systematic overestimate of the severity of late MR. On the other hand, one could argue that the most severely afflicted patients might have died without returning for reevaluation, giving the opposite bias. Another limitation is small sample size for the annuloplasty group, true also of our other exploratory post-hoc analyses. Finally, this study did not examine rehospitalization, quality of life, functional status, or other important outcomes besides survival.
Patient Selection for Surgery for Moderate IMR
These results contribute to the ongoing discussion of the management of IMR by providing an analysis of the results of CABG restricted to patients with only one level of severity (3+) of IMR. Overall, long-term survival after revascularization in this population was poor, and MR recurred or persisted in many patients. Factors such as ventricular function and patient comorbidities appeared to be the most important predictors of long-term mortality, and survival was the same whether annuloplasty was performed or not. The challenge facing the surgeon and future investigators is to better identify those patients for whom each surgical strategy is most beneficial and to continue to develop novel, more effective approaches for treating IMR.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. B. Prucz, E. S. Weiss, N. D. Patel, L. U. Nwakanma, A. S. Shah, and J. V. Conte The Impact of Surgical Ventricular Restoration on Mitral Valve Regurgitation Ann. Thorac. Surg., September 1, 2008; 86(3): 726 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Milano, M. A. Daneshmand, J. S. Rankin, E. Honeycutt, M. L. Williams, M. Swaminathan, L. Linblad, L. K. Shaw, D. D. Glower, and P. K. Smith Survival Prognosis and Surgical Management of Ischemic Mitral Regurgitation Ann. Thorac. Surg., September 1, 2008; 86(3): 735 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Calafiore, V. Mazzei, A. L. Iaco, M. Contini, A. Bivona, M. Gagliardi, P. Bosco, S. Gallina, and M. Di Mauro Impact of ischemic mitral regurgitation on long-term outcome of patients with ejection fraction above 0.30 undergoing first isolated myocardial revascularization. Ann. Thorac. Surg., August 1, 2008; 86(2): 458 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Calafiore, A. L. Iaco, M. Contini, A. Bivona, E. Varone, P. Greco, and S. Scandura Mitral Valve Repair for Ischemic Mitral Regurgitation Angiology, August 1, 2008; 59(2_suppl): 89S - 92S. [Abstract] [PDF]
|
|
|
|
|
|
E. C. Waisbren, L.-M. Stevens, E. G. Avery, M. H. Picard, G. J. Vlahakes, and A. K. Agnihotri Changes in mitral regurgitation after replacement of the stenotic aortic valve. Ann. Thorac. Surg., July 1, 2008; 86(1): 56 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Agricola, M. Oppizzi, M. Pisani, A. Meris, F. Maisano, and A. Margonato Ischemic mitral regurgitation: mechanisms and echocardiographic classification Eur J Echocardiogr, March 1, 2008; 9(2): 207 - 221. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Braun, N. R. van de Veire, R. J.M. Klautz, M. I.M. Versteegh, E. R. Holman, J. J.M. Westenberg, E. Boersma, E. E. van der Wall, J. J. Bax, and R. A.E. Dion Restrictive Mitral Annuloplasty Cures Ischemic Mitral Regurgitation and Heart Failure Ann. Thorac. Surg., February 1, 2008; 85(2): 430 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Y. Szeto, R. C. Gorman, J. H. Gorman III, and M. A. Acker Ischemic Mitral Regurgitation Card. Surg. Adult, January 1, 2008; 3(2008): 785 - 802. [Full Text]
|
|
|
|
|
|
J. I. Fann, N. B. Ingels Jr., and D. C. Miller Pathophysiology of Mitral Valve Disease Card. Surg. Adult, January 1, 2008; 3(2008): 973 - 1012. [Full Text]
|
|
|
|
 |
|
 S. Sirivella and I. Gielchinsky Results of Coronary Bypass and Valve Operations for Mitral Valve Regurgitation Asian Cardiovasc Thorac Ann, October 1, 2007; 15(5): 396 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Srivastava, A. Banerjee, S. Jacob, and J. Dunning Should patients undergoing coronary artery bypass grafting with mild to moderate ischaemic mitral regurgitation also undergo mitral valve repair or replacement? Interactive CardioVascular and Thoracic Surgery, August 1, 2007; 6(4): 538 - 546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Jones Adding Mitral Valve Annuloplasty to Surgical Revascularization Does Not Benefit Patients With Functional Ischemic Mitral Regurgitation J. Am. Coll. Cardiol., June 5, 2007; 49(22): 2202 - 2203. [Full Text] [PDF]
|
|
|
|
|
|
T. Mihaljevic, B.-K. Lam, J. Rajeswaran, M. Takagaki, M. S. Lauer, A. M. Gillinov, E. H. Blackstone, and B. W. Lytle Impact of Mitral Valve Annuloplasty Combined With Revascularization in Patients With Functional Ischemic Mitral Regurgitation J. Am. Coll. Cardiol., June 5, 2007; 49(22): 2191 - 2201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sirivella and I. Gielchinsky Clinical outcomes of surgery of mitral valve regurgitation and coronary artery bypass grafting Interactive CardioVascular and Thoracic Surgery, August 1, 2006; 5(4): 392 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Grossi, G. A. Crooke, P. L. DiGiorgi, C. F. Schwartz, U. Jorde, R. M. Applebaum, G. H. Ribakove, A. C. Galloway, J. B. Grau, and S. B. Colvin Impact of Moderate Functional Mitral Insufficiency in Patients Undergoing Surgical Revascularization Circulation, July 4, 2006; 114(1_suppl): I-573 - I-576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Jones The Year in Cardiovascular Surgery J. Am. Coll. Cardiol., May 16, 2006; 47(10): 2094 - 2107. [Full Text] [PDF]
|
|
|
|
|
|
M. A. Borger, A. Alam, P. M. Murphy, T. Doenst, and T. E. David Chronic Ischemic Mitral Regurgitation: Repair, Replace or Rethink? Ann. Thorac. Surg., March 1, 2006; 81(3): 1153 - 1161. [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 |
