Similar Survival After Mitral Valve Replacement or Repair for Ischemic Mitral Regurgitation: A Meta-Analysis Victor Dayan, MD, Gerardo Soca, MD, Leandro Cura, MD, and Carlos A. Mestres, MD, PhD, FETCS Centro Cardiovascular, Hospital de Clinicas, Facultad de Medicina, Montevideo, Uruguay; Department of Cardiovascular Surgery, Hospital Clinico, University of Barcelona, Barcelona, Spain
Background. Ischemic mitral regurgitation (IMR) occurs in 20% of patients after myocardial infarction. There is no agreement as to the best surgical option. With no prospective randomized controlled trials available, our objective was to perform a meta-analysis comparing replacement and repair. Methods. A literature search was performed in PubMed, EMBASE, Ovid, and Google Scholar. The following keywords were included: “ischemic mitral regurgitation” and “repair or replacement.” Inclusion and exclusion criteria were used to reflect current surgical practice (subvalvular preservation, ring annuloplasty). Primary outcomes of interest were operative mortality and survival. Secondary outcomes analyzed were change in ejection fraction (EF), left ventricular (LV) dimensions, New York Heart Association (NYHA) class, reoperation rate, and 2D or greater recurrence of mitral regurgitation. Results. Of 280 articles, only 12 satisfied all inclusion and exclusion criteria. These articles included 2,508
patients, 64% of whom received valve replacement. Operative mortality was lower after repair (odds ratio [OR], 0.56; 95% confidence interval [CI], 0.38–0.85; p [ 0.001); no difference was found when only articles with patients operated on mainly after 1998 were included (OR, 0.70; 95% CI, 0.44 –1.12; p [ 0.14). Survival was similar (hazard ratio [HR], 0.86; 95% CI, 0.66–1.13; p [ 0.28). No differences in EF, ventricular dimensions, NYHA class, and reoperation were found. Regurgitation recurrence was higher in the repair group (OR, 7.51; 95% CI, 3.7–15.23; p < 0.001). Conclusions. Mitral valve repair is associated with lower operative mortality but higher recurrence of regurgitation in patients with ischemic mitral regurgitation. No differences were found regarding survival, NYHA class, and functional indicators.
I
Previous meta-analyses concluded that mitral valve repair (MVP) is associated with better short- and longterm survival when compared with mitral valve replacement (MVR) [12]. There are no prospective randomized trials comparing MVP with MVR in patients with chronic IMR. Until the results of an ongoing trial are available [13], meta-analysis is the only tool to evaluate the accumulated evidence. The purpose of this study was to perform a meta-analysis of the available evidence of the best operative option for IMR.
schemic mitral regurgitation (IMR) is present in 20% to 30% of patients after an acute myocardial infarction [1]. When IMR is severe, survival can be as low as 60% at 1 year [2]. Patient survival after operation for IMR is worse than for organic MR [3]. Patients with IMR and coronary artery disease who undergo isolated coronary artery bypass grafting have worse survival, and MR persists in these patients [4]. Current guidelines consider surgical intervention for IMR with an ejection fraction (EF) 30% or greater as a class I recommendation; a class IIb recommendation is when the EF is 30% or less [5]. The durability of repair of degenerative MR over the long term is well established [6], but recurrence of functional MR is common because of the progressive nature of the underlying ventricular disease (30%–40% annually) [7]. The best operation for IMR is controversial. Some studies recommend repair [8, 9], others replacement [10], and others show no differences because both options had the same outcomes [11]. Accepted for publication Oct 11, 2013. Address correspondence to Dr Dayan, 26 de Marzo 3459/602, Montevideo, Uruguay; e-mail: [email protected].
Ó 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;-:-–-) Ó 2013 by The Society of Thoracic Surgeons
Material and Methods A literature search was performed in PubMed, EMBASE, Ovid, and Google Scholar for studies published between 1965 and 2013 without language restriction according to the following criteria: exp *Mitral Valve Insufficiency/ or ischemic mitral regurgitation.mp. or exp *Ventricular Dysfunction, Left/or Mitral Valve Insufficiency/ su [Surgery] or exp *Ventricular Dysfunction, Left/or functional mitral regurgitation.mp. AND mitral valve replacement.mp. or mitral valve repair.mp. or exp *Mitral Valve Annuloplasty/. The related articles function was used to broaden the search. The Cochrane library was 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.10.044
2
MVR ¼ HTN ¼ hypertension;
p < 0.05 between MVP and MVR. A and B after surgical era represent their inclusion in the analysis as either after or before 1998.
AMI ¼ acute myocardial infarction; COPD ¼ chronic obstructive pulmonary disease; mitral valve replacement; NS ¼ not significant.
a
LVEF ¼ left ventricular ejection fraction;
MVP ¼ mitral valve repair;
1984–1994B 1993–2003A 1995–2002B 1993–2002B 1995–2003A 1986–2006B 1979–2002B 1995–2008A 2001–2009A 1993–2007A 1996–2011A 2001–2010A NS 44 40 NS 27 43a NS 40a 35 34 35 37 NS 45 44 NS 27 35 NS 45 35 34 35 37 NS NS NS NS NS NS NS NS 9 NS NS NS NS NS NS NS NS NS NS NS 11 NS NS NS 48 NS NS 21a NS NS NS 48a NS 4 NS 54a 43 NS NS 24 NS NS NS 36 NS 3 NS 32 NS 7 14 29 NS 2 NS 28a 6 9 30 NS NS 7 11 26 NS 5 NS 15 7 10 28 NS 98a NS NS 49a NS 45 NS 76 49 91 NS NS 89 NS NS 82 NS 51 NS 57 53 85 NS NS NS 19 15 14 NS 9 NS 22 23 NS 21 NS NS 13 16 13 NS 15 NS 17 19 NS 20 NS NS 15 21 8 NS 26a NS 39 32 26 35 NS NS 26 22 17 NS 39 NS 33 29 34 36 NS NS 51 NS 57 NS 60a NS 51 74 68 41 NS NS 54 NS 50 NS 74 NS 58 72 71 41 NS 52 46 NS 7 NS 53 NS 40 44 37 31 34 45 33 NS 26 NS 45 NS 31 36 32 27 31 68.6a 68 69 67a NS 67 NS 66 72 70 66 69 65.5 68 67 62 NS 66 NS 66 71 70 66 67 56 41 56 14 15 106 54 184 106 85 244 65 94 61 54 38 34 416 125 186 112 302 244 65 Cohn et al [10] Mantovani et al [20] Reece et al [18] Silberman et al [9] Bonacchi et al [19] Milano et al [8] Ngaage et al [24] Magne et al [22] Qiu et al [21] Maltais et al [17] Lorusso et al [11] Chan et al [23]
MVP MVR MVP MVP MVP MVR MVP MVR MVP MVR MVP MVR MVP
Female (%) Age Patients
Table 1. Demographics of Patients in Included Studies
HTN
Diabetes
COPD
NYHA III-IV
RF
Emergency Recent AMI
LVEF
Era of MVR MVP MVR MVP MVR MVP MVR MVP MVR Operation
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
also searched using the same terms. All the review articles whose subject was MVR or MVP, as well as their reference lists, were also reviewed. VD and GS performed the literature review, article selection, and data abstracts independently.
Inclusion and Exclusion Criteria Articles were included if there was a direct comparison in outcomes between MVP and MVR in severe IMR. The following exclusion criteria were used to select the final articles for the meta-analysis: (1) ischemic cause in only a subset of the patients with outcomes not specifically provided, (2) lack of annuloplasty ring in more than 20% of cases, (3) beating heart procedures, (4) concomitant surgical ventricular restoration, (5) no description regarding subvalvular apparatus preservation or no preservation in greater than 20% of cases, and (6) articles written in languages besides English, Spanish, or Portuguese.
Outcomes of Interest The main outcomes of interest were operative mortality and long-term survival (considered to be 1 year after operation). Operative mortality was defined as death within 30 days after operation or in-hospital death [14]. Secondary outcomes of interest were change in left ventricular ejection fraction (LVEF), change in New York Heart Association (NYHA) class, change in LV dimensions, reoperation, MR 2þ or more at follow-up, and freedom from valve-related morbidity and mortality.
Statistics For each study, hazard ratios (HRs) reflecting long-term survival and freedom from valve-related morbidity and mortality along with their corresponding variances were calculated. When only the survival curves were available, they were inspected and the survival rates were estimated using 6-month intervals using the method of Parmar and colleagues [15]. Overall log hazard ratios using inverse variances as weights were then calculated for each study. The odds ratio (OR) was used as the summary statistic for mortality, reoperation, and recurrence. For LVEF and change in NYHA status, the summary statistic chosen was the mean difference. Changes in LV dimensions were analyzed using standardized mean difference. Heterogeneity was examined using Cochran’s Q test as well as the I2 statistic. The degree of heterogeneity was graded as low (< 25%), moderate (25%–75%), and high (> 75%) [16]. Because of patient and treatment procedure heterogeneity in the included studies, random effects models were used to calculate the summary statistics and their 95% confidence intervals (CIs). Subgroup analysis and meta-regression was performed to assess covariates using the Enter method. The year 1998 was selected as a cut point to distinguish 2 “surgical eras” in mitral valve surgical procedures. Articles were assigned an era (before or after 1998) based on the time frame in which the study patients were included. For example, articles in which patients were operated on between 1995 and 2005 would be assigned to the “after 1998 era” because most of the
Ann Thorac Surg 2013;-:-–-
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
3
Table 2. Procedures Performed in Included Studies MVR Prosthesis (%)
MVR Subvalvular Preservation (%)
Mechanical
Biological
Anterior þ Posterior
Posterior
MVP Partial/Suture Annuloplasty (%)
MVP Ring Annuloplasty (%)
MVP Undersizing
Cohn et al [10]
29
71
0
100
15
NS
Mantovani et al [20]
76
34
0
100
0
34 CEC 51 Duran 100 CEC
Reece et al [18]
NS
NS
NS
NS
0
100 Rigid
Silberman et al [9] Bonacchi et al [19] Milano et al [8] Ngaage eat al [24] Magne et al [22]
100 NS 72 NS 79
0 NS 28 NS 21
NS 0 NS NS 20
NS 100 NS NS 66
0 17 9 9 0
100 Duran 83 CEC 91 St. Jude 81 Flexible 100 78 CEC 19 CEMc 3 DA
Qiu et al [21] Maltais et al [17]
38 46
62 54
11 NS
89 NS
0 8
100 92
Lorusso et al [11]
47
53
NS
NS
0
0
100
42
58
0
100 15 CEC 37 CEP 30 CEPII 3 CEMc 2 Geo 100 75 MF 10 DA 8 CEP 6 CEAS
Chan et al [23]
4.9 (26 mm) 19.7 (28 mm) 34.5 (30 mm) 27.8 (32 mm) 9.9 (34 mm) 3.2 (36 mm) 28 mm males 26 mm females 28 1.8 mm NS NS NS 9 (24 mm) 32 (26 mm) 3 (27 mm) 36 (28 mm) 11 (30 mm) 8 (31–33 mm) 30 mm 42 (24–28 mm) 36 (30–34 mm) 27 (26 mm) 52 (28 mm) 13 (30 mm) 6 (32 mm) 1 (34 mm) 1 (36 mm) 28.4 1.6 mm
CEAS ¼ Crosgrove-Edwards annuloplasty system; CEC ¼ Carpentier Edwards Classic; CEMc ¼ Carpentier Edwards McCarthyETlogix; CEP ¼ Carpentier Edwards Physio; CEPII ¼ Carpentier Edwards Physio II; DA ¼ Duran AnCore; Geo ¼ St Jude Geoform; MF ¼ Medtronic Future Band; MVP ¼ mitral valve repair; MVR ¼ mitral valve replacement.
patients were operated on after 1998 (a similar number of patients per year was assumed). Values of p less than 0.05 were considered significant. Funnel plots were inspected to assess the potential of publication bias. Meta-analysis results are displayed in forest plots. Analysis was conducted using Review Manager, version 5.2 (The Cochrane Collaboration, Update Software, Oxford).
Results Two hundred eighty studies published between 1965 and 2013 were identified. On the basis of title and abstracts, 36 articles were selected and reviewed in full. Twelve articles met the inclusion and exclusion criteria [8–11, 17–24]. All were retrospective and nonrandomized studies. They included a total of 2,509 patients, 1611 (64%) of whom underwent MVP. Patient characteristics in eligible studies and procedures performed are summarized in Tables 1 and 2, respectively.
Hospital Mortality A total of 11 studies (2,416 patients) reported hospital mortality (Fig 1A). ORs ranged from 2.32 (favoring replacement) [17] to 0.16 (favoring repair) [18]. The summary OR was 0.56 (95% CI, 0.38–0.85; p ¼ 0.001). Studies in which patients were operated on mainly after 1998 [11, 17, 19–23] had an OR of 0.70 (95% CI, 0.44–1.12; p ¼ 0.14] (Fig 1B), whereas the remaining studies had an OR of 0.41 (95% CI, 0.25–0.68; p < 0.001). When both results were compared using a z test, no differences were found (p ¼ 0.16).
Survival Eleven studies reported long-term survival (Fig 2A). The study HR was 0.86 (95% CI, 0.66–1.13; p ¼ 0.28). Heterogeneity was moderate (I2 ¼ 65%). To explain heterogeneity, subgroup analysis was performed based on surgical era and propensity-matched studies. Only 4 studies
4
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
Fig 1. Operative mortality forest plots. (A) Global mortality: OR ¼ 0.56 (95% CI, 0.38–0.85; p ¼ 0.006). (B) Including only articles in which patients were operated on mainly after 1998: OR, 0.7 (95% CI, 0.44–1.12; p ¼ 0.14). (CI ¼ confidence interval; df ¼ degrees of freedom; M-H ¼ Mantel-Haenszel.)
included propensity-matched patients [11, 17, 22, 23]. When only these studies were included, heterogeneity was null (I2 ¼ 0%), and the pooled HR was 1.05 (95% CI, 0.92 –1.19; p ¼ 0.46) (Fig 2B). Subgroup analysis based on surgical era (with patients operated on mainly after 1998) revealed low heterogeneity (I2 ¼ 0%) [11, 17, 19–22]. After meta-regression including surgical era and propensity matches as covariates, only surgical era resulted as a probable explanation for heterogeneity, with a coefficient Fig 2. Survival forest plots. (A) Global survival: hazard ratio (HR), 0.86 (95% CI, 0.66–1.13; p ¼ 0.28). (B) Survival in which only studies with propensity-matched controls were included: HR, 1.05 (95% CI, 0.92–1.19; p ¼ 0.51). (CI ¼ confidence interval; IV ¼ inverse variance; SE ¼ standard error.)
B of 0.637 (p ¼ 0.035) (Table 3; Fig 3A). When outliers were removed [13], surgical era as covariate gained much more strength in meta-regression analysis (B ¼ 0.97; p < 0.001) (Fig 3B).
LVEF, LV End Diastolic Diameter, and NYHA Classification Follow-up echocardiograms and NYHA assessments were done after a mean of 50 [21], 36 [20], and 6 [19]
Ann Thorac Surg 2013;-:-–-
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Recurrence of MR
Table 3. Meta-Regression Analysis for Predictors of Heterogeneity in Survival Predictor Surgical era Propensity-matched studies
5
B Coefficient
p Value
0.637 0.041
0.035 0.95
months after operation. Pooled analysis revealed a mean difference in LVEF of 0.48% (95% CI, –3.25 to 4.22%; p ¼ 0.8). Although Chan and colleagues [23] did not report mean LVEF before and after operation, they described similar improvement in LV function between both groups at follow-up. No differences in LV end diastolic diameter (–0.09; 95% CI, –0.35–0.16; p ¼ 0.47) and NYHA change (0.01; 95% CI, –0.60 to 0.61; p ¼ 0.98) were found (Fig 4).
Reoperation Reoperation because of endocarditis or recurrent MR (caused by paravalvular leak or failed repair) was reported in 4 studies [11, 19, 20, 24]. The OR was 1.49 (95% CI, 0.91–2.46; p ¼ 0.16] (Fig 5A).
Information was gathered regarding recurrence of MR 2þ or greater on follow-up. Studies concluded that MVP results in an increase in recurrence of MR during the follow-up. The pooled OR was 7.51 (95% CI, 3.7–15.23; p < 0.001) (Fig 5B).
Freedom From Valve-Related Morbidity and Mortality Only 2 articles reported data regarding valve-related morbidity and mortality. Because of the small number of patients included, we decided to exclude this measurement.
Comment The superiority of MVP has been established for patients with degenerative MR [6]. The advantages of MVP include lower operative mortality, better preservation of ventricular function, fewer valve-related complications—including thromboembolism, endocarditis, and anticoagulation-related bleeding events—and improved long-term survival [25–28]. The main disadvantage of MVP is recurrence of MR, which affects survival mainly in patients with MR 2þ or greater [29]. The incidence of 2þ or greater MR varies from 11% [30] at 16 months to 72% [31] at 60 months. Predictors of recurrence at followup have been described [3, 22, 32]. Although the main advantage of MVR is the low incidence of recurrence of MR, its Achilles heel is the increased thromboembolic risk compared with MVP [33]. Debate exists about the benefit of MVR or MVP regarding hospital mortality, survival, ventricular remodeling, and changes in LVEF, among other functional indicators in patients with IMR. Previous metaanalyses [12, 34] confirmed that MVR is associated with increased hospital mortality and lower survival. We have included a total of 12 studies that address the outcomes of MVP versus MVR for IMR. These studies were selected after rigorous inclusion and exclusion criteria, aiming to reflect current surgical practice. The advantages of ring annuloplasty over suture or bovine pericardium annuloplasty have been established [35]. Even though a minority of specialized surgical centers obtain excellent short- and long-term results with suture annuloplasty [36, 37], we excluded articles in which more than 20% of procedures lacked an annuloplasty ring. Because of the evident short- and long-term benefit of preservation of the subvalvular apparatus at MVR [27], we excluded articles in which there was no preservation in more than 20% of cases.
Mortality
Fig 3. Regression analysis for survival and surgical era. (A) Global analysis. (B) Excluding outliers (Cohn et al[10]).
Our results suggest that MVP is associated with lower inhospital mortality, with a pooled OR similar to that in other individual studies [12, 34]. When we included only studies in which the majority of patients were operated on after 1998, no difference between MVP and MVR were found. Assuming a similar patient risk profile before and after 1998, hospital mortality is mainly related to surgical
6
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
Fig 4. Forest plots. (A) Change in left ventricular (LV) ejection function. (B) LV end diastolic dimensions. (C) New York Heart Association (NYHA) class forest plots. (CI ¼ confidence interval; df ¼ degrees of freedom; IV ¼ inverse variance; SD ¼ standard deviation; Std. ¼ standard.)
technique (subvalvular preservation), myocardial preservation (cardioplegia), extracorporeal circulation, and postoperative management. A number of changes have been introduced over time in all of these areas, which could explain the decrease in hospital mortality in MVR after 1998.
Survival There was no difference in long-term survival. This metaanalysis reports similar long-term survival between MVP and MVR for IMR. Probable explanations for our results compared with those of Vassileva and coworkers [12] and Rao and colleagues [34], which reported a survival advantage of MVP, are the following: (1) Both studies included results from Al-Radi and associates [38], who reported the lack of preservation of the subvalvular apparatus in more than 20% of cases. Al-Radi and Fig 5. Forest plots. (A) Mitral regurgitation recurrence 2þ or greater. (B) Reoperation forest plots. (CI ¼ confidence interval; df ¼ degrees of freedom; M-H ¼ Mantel-Haenszel.)
associates confirmed a survival advantage of MVP over MVR at 1 year, whereas no differences were found in the long term. This difference in survival could probably be explained by the higher hospital mortality in patients with MVR (OR of 17.2 against MVR), which might be related to the lack of preservation of the subvalvular apparatus. (2) Vassileva and colleagues [12] did not include results from Cohn and colleagues [10], which favored MVR over MVP, in their analysis. (3) Rao and associates [34] included the study from Gillinov and coworkers [39], who performed a subgroup analysis of survival according to preoperative risk. There is no information in their report regarding global survival or mortality. Therefore, this contribution was not included in the analysis. (4) These meta-analyses did not include the contributions of Chan and coworkers [23], Lorusso and associates [11], Maltais and colleagues [17] and Qiu
Ann Thorac Surg 2013;-:-–-
and colleagues [21], reporting no survival difference between treatments.
Functional Factors and Recurrence of MR Both techniques provide similar benefit in cardiac remodeling, EF improvement, and NYHA class. Although the incidence of reoperation was similar in both treatment modalities, MVP increased the odds of recurrence of MR ( 2þ) 7 times. This latter finding is extremely important. As mentioned earlier, recurrence of MR 2þ or greater has an impact on survival [29]. Similar reoperation rates between both groups suggest that surgeons who perform MVP establish a high threshold for reintervention in these patients.
Limitations This meta-analysis includes observational studies with the inherent bias of retrospective reviews, among them the recent study of Lorusso and associates [14], which represents a substantial proportion of the total number of patients included in the analysis and could therefore dominate the results. Few studies compared changes in LVEF, LV diastolic dysfunction, and NYHA class, which weakened the conclusions. Inspection of funnel plots revealed fairly symmetrical distributions and did not raise any major concerns about the potential of publication bias. However, the possibility of such bias still exists and should be taken into account when considering the results. The lack of patient selection in the studies based on predictive factors of recurrence of MR, as well as low performance of adjunctive repair techniques, could negatively affect the long-term results of MVP reported in our study.
Conclusions Hospital mortality is lower in patients treated with MVP compared with those treated with MVR. MR recurrence is higher with MVP. There is no difference in long-term survival between the 2 groups. Recurrence of MR and thromboembolic risk are the main factors regarding survival in MVP and MVR, respectively. Therefore, the choice of MVP or MVR should be individualized in each patient, taking into account predictors of recurrence and thromboembolic risk.
References 1. Bursi F, Enriquez-Sarano M, Nkomo VT, et al. Heart failure and death after myocardial infarction in the community: the emerging role of mitral regurgitation. Circulation 2005;111: 295–301. 2. Hickey MS, Smith LR, Muhlbaier LH, et al. Current prognosis of ischemic mitral regurgitation: implications for future management. Circulation 1988;78(3 Pt 2):I51–9. 3. Calafiore AM, Iac o AL, Tash A, Abukudair W, Di Mauro M. Mitral valve surgery for functional mitral regurgitation in patients with chronic heart failure –update of the results. Thorac Cardiov Surg 2010;58:131–5.
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
7
4. Aklog L, Filsoufi F, Flores KQ, et al. Does coronary artery bypass grafting alone correct moderate ischemic mitral regurgitation? Circulation 2001;104(2 suppl 1):I68–75. 5. Bonow RO, Carabello BA, Chatterjee K, et al. 2008 focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease). Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons.). J Am Coll Cardiol 2008;52:e1–142. 6. Braunberger E, Deloche A, Berrebi A, et al. Very long-term results (more than 20 years) of valve repair with Carpentier’s techniques in nonrheumatic mitral valve insufficiency. Circulation 2001;104:I8–11. 7. McGee EC, Gillinov AM, Blackstone EH, et al. Recurrent mitral regurgitation after annuloplasty for functional ischemic mitral regurgitation. J Thorac Cardiovasc Surg 2004;128:916–24. 8. Milano CA, Daneshmand MA, Rankin JS, et al. Survival prognosis and surgical management of ischemic mitral regurgitation. Ann Thorac Surg 2008;86:735–44. 9. Silberman S, Oren A, Klutstein MW, et al. Does mitral valve intervention have an impact on late survival in ischemic cardiomyopathy? Isr Med Assoc J 2006;8:17–20. 10. Cohn LH, Rizzo RJ, Adams DH, et al. The effect of pathophysiology on the surgical treatment of ischemic mitral regurgitation: operative and late risks of repair versus replacement. Eur J Cardiothorac Surg 1995;9:568–74. 11. Lorusso R, Gelsomino S, Vizzardi E, et al. Mitral valve repair or replacement for ischemic mitral regurgitation? The Italian Study on the Treatment of Ischemic Mitral Regurgitation (ISTIMIR). J Thorac Cardiovasc Surg 2013;145:128–39. 12. Vassileva CM, Boley T, Markwell S, Hazelrigg S. Metaanalysis of short-term and long-term survival following repair versus replacement for ischemic mitral regurgitation. Eur J Cardiothorac Surg 2011;39:295–303. 13. Perrault LP, Moskowitz AJ, Kron IL, et al. Optimal surgical management of severe ischemic mitral regurgitation: to repair or to replace? J Thorac Cardiovasc Surg 2012;143: 1396–403. 14. Jacobs JP, Mavroudis C, Jacobs ML, et al. What is operative mortality? Defining death in a surgical registry database: a report of the STS Congenital Database Taskforce and the Joint EACTS-STS Congenital Database Committee. Ann Thorac Surg 2006;81:1937–41. 15. Parmar MKB, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1988;17:2815–34. 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60. 17. Maltais S, Schaff HV, Daly RC, et al. Mitral regurgitation surgery in patients with ischemic cardiomyopathy and ischemic mitral regurgitation: factors that influence survival. J Thorac Cardiovasc Surg 2011;142:995–1001. 18. Reece TB, Tribble CG, Ellman PI, et al. Mitral repair is superior to replacement when associated with coronary artery disease. Ann Surg 2004;239:671–5. 19. Bonacchi M, Prifti E, Maiani M, Frati G, Nathan NS, Leacche M. Mitral valve surgery simultaneous to coronary revascularization in patients with end-stage ischemic cardiomyopathy. Heart Vessels 2006;21:20–7. 20. Mantovani V, Mariscalco G, Leva C, Blanzola C, Cattaneo P, Sala A. Long-term results of the surgical treatment of chronic ischemic mitral regurgitation: comparison of repair and prosthetic replacement. J Heart Valve Dis 2004;13:421–8. 21. Qiu Z, Chen X, Xu M, et al. Is mitral valve repair superior to replacement for chronic ischemic mitral regurgitation with left ventricular dysfunction? J Cardiothorac Surg 2010;5:107.
8
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
22. Magne J, Girerd N, Senechal M, et al. Mitral repair versus replacement for ischemic mitral regurgitation: comparison of short-term and long-term survival. Circulation 2009;120(11 suppl):S104–11. 23. Chan V, Ruel M, Mesana TG. Mitral valve replacement is a viable alternative to mitral valve repair for ischemic mitral regurgitation: a case-matched study. Ann Thorac Surg 2011;92:1358–65. 24. Ngaage DL, Daly RC, Rosales G, et al. Mitral regurgitation surgery in heart failure due to ischemic cardiomyopathy: a 24-year experience. J Heart Valve Dis 2008;17:251–9. 25. Grossi EA, Galloway AC, Miller JS, et al. Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated? J Thorac Cardiovasc Surg 1998;115: 389–94. 26. Kawachi Y, Oe M, Asou T, Tominaga R, Tokunaga K. Comparative study between valve repair and replacement for mitral pure regurgitation—early and late postoperative results. Jpn Circ J 1991;55:443–52. 27. Okita Y, Miki S, Ueda Y, Tahata T, Sakai T, Matsuyama K. Comparative evaluation of left ventricular performance after mitral valve repair or valve replacement with or without chordal preservation. J Heart Valve Dis 1993;2:159–66. 28. Sand ME, Naftel DC, Blackstone EH, Kirklin JW, Karp RB. A comparison of repair and replacement for mitral valve incompetence. J Thorac Cardiovasc Surg 1987;94:208–19. 29. Crabtree TD, Bailey MS, Moon MR, et al. Recurrent mitral regurgitation and risk factors for early and late mortality after mitral valve repair for functional ischemic mitral regurgitation. Ann Thorac Surg 2008;85:1537–43. 30. Glower DD, Tuttle RH, Shaw LK, Orozco RE, Rankin JS. Patients survival characteristics after routine mitral valve
Ann Thorac Surg 2013;-:-–-
31.
32.
33.
34. 35. 36. 37. 38. 39.
repair for ischemic mitral regurgitation. J Thorac Cardiovasc Surg 2005;129:860–8. Gelsomino S, Lorusso R, De Cicco G, et al. Five-year echocardiographic results of combined undersized mitral ring annuloplasty and coronary artery bypass grafting for chronic ischemic mitral regurgitation. Eur Heart J 2008;29:231–40. Ciarka A, Braun J, Delgado V, et al. Predictors of mitral regurgitation recurrence in patients with heart failure undergoing mitral valve annuloplasty. Am J Cardiol 2010;106: 395–401. Russo A, Grigioni F, Avierinos JF, et al. Thromboembolic complications after surgical correction of mitral regurgitation incidence, predictors, and clinical implications. J Am Coll Cardiol 2008;51:1203–11. Rao C, Murphy MO, Saso S, et al. Mitral valve repair or replacement for ischaemic mitral regurgitation: a systematic review. Heart Lung Circ 2011;20:555–65. Czer LS, Maurer G, Trento A, et al. Comparative efficacy of ring and suture annuloplasty for ischemic mitral regurgitation. Circulation 1992;86:II46–52. Calafiore AM, Di Mauro M, Gallina S, et al. Mitral valve surgery for chronic ischemic mitral regurgitation. Ann Thorac Surg 2004;77:1989–97. Hausmann H, Siniawski H, Hetzer R. Mitral valve reconstruction and replacement for ischemic mitral insufficiency: seven years’ follow up. J Heart Valve Dis 1999;8:536–42. Al-Radi OO, Austin PC, Tu JV, David TE, Yau TM. Mitral repair versus replacement for ischemic mitral regurgitation. Ann Thorac Surg 2005;79:1260–7. Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg 2001;122:1125–41.
Background. Ischemic mitral regurgitation (IMR) occurs in 20% of patients after myocardial infarction. There is no agreement as to the best surgical option. With no prospective randomized controlled trials available, our objective was to perform a meta-analysis comparing replacement and repair. Methods. A literature search was performed in PubMed, EMBASE, Ovid, and Google Scholar. The following keywords were included: “ischemic mitral regurgitation” and “repair or replacement.” Inclusion and exclusion criteria were used to reflect current surgical practice (subvalvular preservation, ring annuloplasty). Primary outcomes of interest were operative mortality and survival. Secondary outcomes analyzed were change in ejection fraction (EF), left ventricular (LV) dimensions, New York Heart Association (NYHA) class, reoperation rate, and 2D or greater recurrence of mitral regurgitation. Results. Of 280 articles, only 12 satisfied all inclusion and exclusion criteria. These articles included 2,508
patients, 64% of whom received valve replacement. Operative mortality was lower after repair (odds ratio [OR], 0.56; 95% confidence interval [CI], 0.38–0.85; p [ 0.001); no difference was found when only articles with patients operated on mainly after 1998 were included (OR, 0.70; 95% CI, 0.44 –1.12; p [ 0.14). Survival was similar (hazard ratio [HR], 0.86; 95% CI, 0.66–1.13; p [ 0.28). No differences in EF, ventricular dimensions, NYHA class, and reoperation were found. Regurgitation recurrence was higher in the repair group (OR, 7.51; 95% CI, 3.7–15.23; p < 0.001). Conclusions. Mitral valve repair is associated with lower operative mortality but higher recurrence of regurgitation in patients with ischemic mitral regurgitation. No differences were found regarding survival, NYHA class, and functional indicators.
I
Previous meta-analyses concluded that mitral valve repair (MVP) is associated with better short- and longterm survival when compared with mitral valve replacement (MVR) [12]. There are no prospective randomized trials comparing MVP with MVR in patients with chronic IMR. Until the results of an ongoing trial are available [13], meta-analysis is the only tool to evaluate the accumulated evidence. The purpose of this study was to perform a meta-analysis of the available evidence of the best operative option for IMR.
schemic mitral regurgitation (IMR) is present in 20% to 30% of patients after an acute myocardial infarction [1]. When IMR is severe, survival can be as low as 60% at 1 year [2]. Patient survival after operation for IMR is worse than for organic MR [3]. Patients with IMR and coronary artery disease who undergo isolated coronary artery bypass grafting have worse survival, and MR persists in these patients [4]. Current guidelines consider surgical intervention for IMR with an ejection fraction (EF) 30% or greater as a class I recommendation; a class IIb recommendation is when the EF is 30% or less [5]. The durability of repair of degenerative MR over the long term is well established [6], but recurrence of functional MR is common because of the progressive nature of the underlying ventricular disease (30%–40% annually) [7]. The best operation for IMR is controversial. Some studies recommend repair [8, 9], others replacement [10], and others show no differences because both options had the same outcomes [11]. Accepted for publication Oct 11, 2013. Address correspondence to Dr Dayan, 26 de Marzo 3459/602, Montevideo, Uruguay; e-mail: [email protected].
Ó 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;-:-–-) Ó 2013 by The Society of Thoracic Surgeons
Material and Methods A literature search was performed in PubMed, EMBASE, Ovid, and Google Scholar for studies published between 1965 and 2013 without language restriction according to the following criteria: exp *Mitral Valve Insufficiency/ or ischemic mitral regurgitation.mp. or exp *Ventricular Dysfunction, Left/or Mitral Valve Insufficiency/ su [Surgery] or exp *Ventricular Dysfunction, Left/or functional mitral regurgitation.mp. AND mitral valve replacement.mp. or mitral valve repair.mp. or exp *Mitral Valve Annuloplasty/. The related articles function was used to broaden the search. The Cochrane library was 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.10.044
2
MVR ¼ HTN ¼ hypertension;
p < 0.05 between MVP and MVR. A and B after surgical era represent their inclusion in the analysis as either after or before 1998.
AMI ¼ acute myocardial infarction; COPD ¼ chronic obstructive pulmonary disease; mitral valve replacement; NS ¼ not significant.
a
LVEF ¼ left ventricular ejection fraction;
MVP ¼ mitral valve repair;
1984–1994B 1993–2003A 1995–2002B 1993–2002B 1995–2003A 1986–2006B 1979–2002B 1995–2008A 2001–2009A 1993–2007A 1996–2011A 2001–2010A NS 44 40 NS 27 43a NS 40a 35 34 35 37 NS 45 44 NS 27 35 NS 45 35 34 35 37 NS NS NS NS NS NS NS NS 9 NS NS NS NS NS NS NS NS NS NS NS 11 NS NS NS 48 NS NS 21a NS NS NS 48a NS 4 NS 54a 43 NS NS 24 NS NS NS 36 NS 3 NS 32 NS 7 14 29 NS 2 NS 28a 6 9 30 NS NS 7 11 26 NS 5 NS 15 7 10 28 NS 98a NS NS 49a NS 45 NS 76 49 91 NS NS 89 NS NS 82 NS 51 NS 57 53 85 NS NS NS 19 15 14 NS 9 NS 22 23 NS 21 NS NS 13 16 13 NS 15 NS 17 19 NS 20 NS NS 15 21 8 NS 26a NS 39 32 26 35 NS NS 26 22 17 NS 39 NS 33 29 34 36 NS NS 51 NS 57 NS 60a NS 51 74 68 41 NS NS 54 NS 50 NS 74 NS 58 72 71 41 NS 52 46 NS 7 NS 53 NS 40 44 37 31 34 45 33 NS 26 NS 45 NS 31 36 32 27 31 68.6a 68 69 67a NS 67 NS 66 72 70 66 69 65.5 68 67 62 NS 66 NS 66 71 70 66 67 56 41 56 14 15 106 54 184 106 85 244 65 94 61 54 38 34 416 125 186 112 302 244 65 Cohn et al [10] Mantovani et al [20] Reece et al [18] Silberman et al [9] Bonacchi et al [19] Milano et al [8] Ngaage et al [24] Magne et al [22] Qiu et al [21] Maltais et al [17] Lorusso et al [11] Chan et al [23]
MVP MVR MVP MVP MVP MVR MVP MVR MVP MVR MVP MVR MVP
Female (%) Age Patients
Table 1. Demographics of Patients in Included Studies
HTN
Diabetes
COPD
NYHA III-IV
RF
Emergency Recent AMI
LVEF
Era of MVR MVP MVR MVP MVR MVP MVR MVP MVR Operation
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
also searched using the same terms. All the review articles whose subject was MVR or MVP, as well as their reference lists, were also reviewed. VD and GS performed the literature review, article selection, and data abstracts independently.
Inclusion and Exclusion Criteria Articles were included if there was a direct comparison in outcomes between MVP and MVR in severe IMR. The following exclusion criteria were used to select the final articles for the meta-analysis: (1) ischemic cause in only a subset of the patients with outcomes not specifically provided, (2) lack of annuloplasty ring in more than 20% of cases, (3) beating heart procedures, (4) concomitant surgical ventricular restoration, (5) no description regarding subvalvular apparatus preservation or no preservation in greater than 20% of cases, and (6) articles written in languages besides English, Spanish, or Portuguese.
Outcomes of Interest The main outcomes of interest were operative mortality and long-term survival (considered to be 1 year after operation). Operative mortality was defined as death within 30 days after operation or in-hospital death [14]. Secondary outcomes of interest were change in left ventricular ejection fraction (LVEF), change in New York Heart Association (NYHA) class, change in LV dimensions, reoperation, MR 2þ or more at follow-up, and freedom from valve-related morbidity and mortality.
Statistics For each study, hazard ratios (HRs) reflecting long-term survival and freedom from valve-related morbidity and mortality along with their corresponding variances were calculated. When only the survival curves were available, they were inspected and the survival rates were estimated using 6-month intervals using the method of Parmar and colleagues [15]. Overall log hazard ratios using inverse variances as weights were then calculated for each study. The odds ratio (OR) was used as the summary statistic for mortality, reoperation, and recurrence. For LVEF and change in NYHA status, the summary statistic chosen was the mean difference. Changes in LV dimensions were analyzed using standardized mean difference. Heterogeneity was examined using Cochran’s Q test as well as the I2 statistic. The degree of heterogeneity was graded as low (< 25%), moderate (25%–75%), and high (> 75%) [16]. Because of patient and treatment procedure heterogeneity in the included studies, random effects models were used to calculate the summary statistics and their 95% confidence intervals (CIs). Subgroup analysis and meta-regression was performed to assess covariates using the Enter method. The year 1998 was selected as a cut point to distinguish 2 “surgical eras” in mitral valve surgical procedures. Articles were assigned an era (before or after 1998) based on the time frame in which the study patients were included. For example, articles in which patients were operated on between 1995 and 2005 would be assigned to the “after 1998 era” because most of the
Ann Thorac Surg 2013;-:-–-
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
3
Table 2. Procedures Performed in Included Studies MVR Prosthesis (%)
MVR Subvalvular Preservation (%)
Mechanical
Biological
Anterior þ Posterior
Posterior
MVP Partial/Suture Annuloplasty (%)
MVP Ring Annuloplasty (%)
MVP Undersizing
Cohn et al [10]
29
71
0
100
15
NS
Mantovani et al [20]
76
34
0
100
0
34 CEC 51 Duran 100 CEC
Reece et al [18]
NS
NS
NS
NS
0
100 Rigid
Silberman et al [9] Bonacchi et al [19] Milano et al [8] Ngaage eat al [24] Magne et al [22]
100 NS 72 NS 79
0 NS 28 NS 21
NS 0 NS NS 20
NS 100 NS NS 66
0 17 9 9 0
100 Duran 83 CEC 91 St. Jude 81 Flexible 100 78 CEC 19 CEMc 3 DA
Qiu et al [21] Maltais et al [17]
38 46
62 54
11 NS
89 NS
0 8
100 92
Lorusso et al [11]
47
53
NS
NS
0
0
100
42
58
0
100 15 CEC 37 CEP 30 CEPII 3 CEMc 2 Geo 100 75 MF 10 DA 8 CEP 6 CEAS
Chan et al [23]
4.9 (26 mm) 19.7 (28 mm) 34.5 (30 mm) 27.8 (32 mm) 9.9 (34 mm) 3.2 (36 mm) 28 mm males 26 mm females 28 1.8 mm NS NS NS 9 (24 mm) 32 (26 mm) 3 (27 mm) 36 (28 mm) 11 (30 mm) 8 (31–33 mm) 30 mm 42 (24–28 mm) 36 (30–34 mm) 27 (26 mm) 52 (28 mm) 13 (30 mm) 6 (32 mm) 1 (34 mm) 1 (36 mm) 28.4 1.6 mm
CEAS ¼ Crosgrove-Edwards annuloplasty system; CEC ¼ Carpentier Edwards Classic; CEMc ¼ Carpentier Edwards McCarthyETlogix; CEP ¼ Carpentier Edwards Physio; CEPII ¼ Carpentier Edwards Physio II; DA ¼ Duran AnCore; Geo ¼ St Jude Geoform; MF ¼ Medtronic Future Band; MVP ¼ mitral valve repair; MVR ¼ mitral valve replacement.
patients were operated on after 1998 (a similar number of patients per year was assumed). Values of p less than 0.05 were considered significant. Funnel plots were inspected to assess the potential of publication bias. Meta-analysis results are displayed in forest plots. Analysis was conducted using Review Manager, version 5.2 (The Cochrane Collaboration, Update Software, Oxford).
Results Two hundred eighty studies published between 1965 and 2013 were identified. On the basis of title and abstracts, 36 articles were selected and reviewed in full. Twelve articles met the inclusion and exclusion criteria [8–11, 17–24]. All were retrospective and nonrandomized studies. They included a total of 2,509 patients, 1611 (64%) of whom underwent MVP. Patient characteristics in eligible studies and procedures performed are summarized in Tables 1 and 2, respectively.
Hospital Mortality A total of 11 studies (2,416 patients) reported hospital mortality (Fig 1A). ORs ranged from 2.32 (favoring replacement) [17] to 0.16 (favoring repair) [18]. The summary OR was 0.56 (95% CI, 0.38–0.85; p ¼ 0.001). Studies in which patients were operated on mainly after 1998 [11, 17, 19–23] had an OR of 0.70 (95% CI, 0.44–1.12; p ¼ 0.14] (Fig 1B), whereas the remaining studies had an OR of 0.41 (95% CI, 0.25–0.68; p < 0.001). When both results were compared using a z test, no differences were found (p ¼ 0.16).
Survival Eleven studies reported long-term survival (Fig 2A). The study HR was 0.86 (95% CI, 0.66–1.13; p ¼ 0.28). Heterogeneity was moderate (I2 ¼ 65%). To explain heterogeneity, subgroup analysis was performed based on surgical era and propensity-matched studies. Only 4 studies
4
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
Fig 1. Operative mortality forest plots. (A) Global mortality: OR ¼ 0.56 (95% CI, 0.38–0.85; p ¼ 0.006). (B) Including only articles in which patients were operated on mainly after 1998: OR, 0.7 (95% CI, 0.44–1.12; p ¼ 0.14). (CI ¼ confidence interval; df ¼ degrees of freedom; M-H ¼ Mantel-Haenszel.)
included propensity-matched patients [11, 17, 22, 23]. When only these studies were included, heterogeneity was null (I2 ¼ 0%), and the pooled HR was 1.05 (95% CI, 0.92 –1.19; p ¼ 0.46) (Fig 2B). Subgroup analysis based on surgical era (with patients operated on mainly after 1998) revealed low heterogeneity (I2 ¼ 0%) [11, 17, 19–22]. After meta-regression including surgical era and propensity matches as covariates, only surgical era resulted as a probable explanation for heterogeneity, with a coefficient Fig 2. Survival forest plots. (A) Global survival: hazard ratio (HR), 0.86 (95% CI, 0.66–1.13; p ¼ 0.28). (B) Survival in which only studies with propensity-matched controls were included: HR, 1.05 (95% CI, 0.92–1.19; p ¼ 0.51). (CI ¼ confidence interval; IV ¼ inverse variance; SE ¼ standard error.)
B of 0.637 (p ¼ 0.035) (Table 3; Fig 3A). When outliers were removed [13], surgical era as covariate gained much more strength in meta-regression analysis (B ¼ 0.97; p < 0.001) (Fig 3B).
LVEF, LV End Diastolic Diameter, and NYHA Classification Follow-up echocardiograms and NYHA assessments were done after a mean of 50 [21], 36 [20], and 6 [19]
Ann Thorac Surg 2013;-:-–-
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Recurrence of MR
Table 3. Meta-Regression Analysis for Predictors of Heterogeneity in Survival Predictor Surgical era Propensity-matched studies
5
B Coefficient
p Value
0.637 0.041
0.035 0.95
months after operation. Pooled analysis revealed a mean difference in LVEF of 0.48% (95% CI, –3.25 to 4.22%; p ¼ 0.8). Although Chan and colleagues [23] did not report mean LVEF before and after operation, they described similar improvement in LV function between both groups at follow-up. No differences in LV end diastolic diameter (–0.09; 95% CI, –0.35–0.16; p ¼ 0.47) and NYHA change (0.01; 95% CI, –0.60 to 0.61; p ¼ 0.98) were found (Fig 4).
Reoperation Reoperation because of endocarditis or recurrent MR (caused by paravalvular leak or failed repair) was reported in 4 studies [11, 19, 20, 24]. The OR was 1.49 (95% CI, 0.91–2.46; p ¼ 0.16] (Fig 5A).
Information was gathered regarding recurrence of MR 2þ or greater on follow-up. Studies concluded that MVP results in an increase in recurrence of MR during the follow-up. The pooled OR was 7.51 (95% CI, 3.7–15.23; p < 0.001) (Fig 5B).
Freedom From Valve-Related Morbidity and Mortality Only 2 articles reported data regarding valve-related morbidity and mortality. Because of the small number of patients included, we decided to exclude this measurement.
Comment The superiority of MVP has been established for patients with degenerative MR [6]. The advantages of MVP include lower operative mortality, better preservation of ventricular function, fewer valve-related complications—including thromboembolism, endocarditis, and anticoagulation-related bleeding events—and improved long-term survival [25–28]. The main disadvantage of MVP is recurrence of MR, which affects survival mainly in patients with MR 2þ or greater [29]. The incidence of 2þ or greater MR varies from 11% [30] at 16 months to 72% [31] at 60 months. Predictors of recurrence at followup have been described [3, 22, 32]. Although the main advantage of MVR is the low incidence of recurrence of MR, its Achilles heel is the increased thromboembolic risk compared with MVP [33]. Debate exists about the benefit of MVR or MVP regarding hospital mortality, survival, ventricular remodeling, and changes in LVEF, among other functional indicators in patients with IMR. Previous metaanalyses [12, 34] confirmed that MVR is associated with increased hospital mortality and lower survival. We have included a total of 12 studies that address the outcomes of MVP versus MVR for IMR. These studies were selected after rigorous inclusion and exclusion criteria, aiming to reflect current surgical practice. The advantages of ring annuloplasty over suture or bovine pericardium annuloplasty have been established [35]. Even though a minority of specialized surgical centers obtain excellent short- and long-term results with suture annuloplasty [36, 37], we excluded articles in which more than 20% of procedures lacked an annuloplasty ring. Because of the evident short- and long-term benefit of preservation of the subvalvular apparatus at MVR [27], we excluded articles in which there was no preservation in more than 20% of cases.
Mortality
Fig 3. Regression analysis for survival and surgical era. (A) Global analysis. (B) Excluding outliers (Cohn et al[10]).
Our results suggest that MVP is associated with lower inhospital mortality, with a pooled OR similar to that in other individual studies [12, 34]. When we included only studies in which the majority of patients were operated on after 1998, no difference between MVP and MVR were found. Assuming a similar patient risk profile before and after 1998, hospital mortality is mainly related to surgical
6
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
Ann Thorac Surg 2013;-:-–-
Fig 4. Forest plots. (A) Change in left ventricular (LV) ejection function. (B) LV end diastolic dimensions. (C) New York Heart Association (NYHA) class forest plots. (CI ¼ confidence interval; df ¼ degrees of freedom; IV ¼ inverse variance; SD ¼ standard deviation; Std. ¼ standard.)
technique (subvalvular preservation), myocardial preservation (cardioplegia), extracorporeal circulation, and postoperative management. A number of changes have been introduced over time in all of these areas, which could explain the decrease in hospital mortality in MVR after 1998.
Survival There was no difference in long-term survival. This metaanalysis reports similar long-term survival between MVP and MVR for IMR. Probable explanations for our results compared with those of Vassileva and coworkers [12] and Rao and colleagues [34], which reported a survival advantage of MVP, are the following: (1) Both studies included results from Al-Radi and associates [38], who reported the lack of preservation of the subvalvular apparatus in more than 20% of cases. Al-Radi and Fig 5. Forest plots. (A) Mitral regurgitation recurrence 2þ or greater. (B) Reoperation forest plots. (CI ¼ confidence interval; df ¼ degrees of freedom; M-H ¼ Mantel-Haenszel.)
associates confirmed a survival advantage of MVP over MVR at 1 year, whereas no differences were found in the long term. This difference in survival could probably be explained by the higher hospital mortality in patients with MVR (OR of 17.2 against MVR), which might be related to the lack of preservation of the subvalvular apparatus. (2) Vassileva and colleagues [12] did not include results from Cohn and colleagues [10], which favored MVR over MVP, in their analysis. (3) Rao and associates [34] included the study from Gillinov and coworkers [39], who performed a subgroup analysis of survival according to preoperative risk. There is no information in their report regarding global survival or mortality. Therefore, this contribution was not included in the analysis. (4) These meta-analyses did not include the contributions of Chan and coworkers [23], Lorusso and associates [11], Maltais and colleagues [17] and Qiu
Ann Thorac Surg 2013;-:-–-
and colleagues [21], reporting no survival difference between treatments.
Functional Factors and Recurrence of MR Both techniques provide similar benefit in cardiac remodeling, EF improvement, and NYHA class. Although the incidence of reoperation was similar in both treatment modalities, MVP increased the odds of recurrence of MR ( 2þ) 7 times. This latter finding is extremely important. As mentioned earlier, recurrence of MR 2þ or greater has an impact on survival [29]. Similar reoperation rates between both groups suggest that surgeons who perform MVP establish a high threshold for reintervention in these patients.
Limitations This meta-analysis includes observational studies with the inherent bias of retrospective reviews, among them the recent study of Lorusso and associates [14], which represents a substantial proportion of the total number of patients included in the analysis and could therefore dominate the results. Few studies compared changes in LVEF, LV diastolic dysfunction, and NYHA class, which weakened the conclusions. Inspection of funnel plots revealed fairly symmetrical distributions and did not raise any major concerns about the potential of publication bias. However, the possibility of such bias still exists and should be taken into account when considering the results. The lack of patient selection in the studies based on predictive factors of recurrence of MR, as well as low performance of adjunctive repair techniques, could negatively affect the long-term results of MVP reported in our study.
Conclusions Hospital mortality is lower in patients treated with MVP compared with those treated with MVR. MR recurrence is higher with MVP. There is no difference in long-term survival between the 2 groups. Recurrence of MR and thromboembolic risk are the main factors regarding survival in MVP and MVR, respectively. Therefore, the choice of MVP or MVR should be individualized in each patient, taking into account predictors of recurrence and thromboembolic risk.
References 1. Bursi F, Enriquez-Sarano M, Nkomo VT, et al. Heart failure and death after myocardial infarction in the community: the emerging role of mitral regurgitation. Circulation 2005;111: 295–301. 2. Hickey MS, Smith LR, Muhlbaier LH, et al. Current prognosis of ischemic mitral regurgitation: implications for future management. Circulation 1988;78(3 Pt 2):I51–9. 3. Calafiore AM, Iac o AL, Tash A, Abukudair W, Di Mauro M. Mitral valve surgery for functional mitral regurgitation in patients with chronic heart failure –update of the results. Thorac Cardiov Surg 2010;58:131–5.
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
7
4. Aklog L, Filsoufi F, Flores KQ, et al. Does coronary artery bypass grafting alone correct moderate ischemic mitral regurgitation? Circulation 2001;104(2 suppl 1):I68–75. 5. Bonow RO, Carabello BA, Chatterjee K, et al. 2008 focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease). Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons.). J Am Coll Cardiol 2008;52:e1–142. 6. Braunberger E, Deloche A, Berrebi A, et al. Very long-term results (more than 20 years) of valve repair with Carpentier’s techniques in nonrheumatic mitral valve insufficiency. Circulation 2001;104:I8–11. 7. McGee EC, Gillinov AM, Blackstone EH, et al. Recurrent mitral regurgitation after annuloplasty for functional ischemic mitral regurgitation. J Thorac Cardiovasc Surg 2004;128:916–24. 8. Milano CA, Daneshmand MA, Rankin JS, et al. Survival prognosis and surgical management of ischemic mitral regurgitation. Ann Thorac Surg 2008;86:735–44. 9. Silberman S, Oren A, Klutstein MW, et al. Does mitral valve intervention have an impact on late survival in ischemic cardiomyopathy? Isr Med Assoc J 2006;8:17–20. 10. Cohn LH, Rizzo RJ, Adams DH, et al. The effect of pathophysiology on the surgical treatment of ischemic mitral regurgitation: operative and late risks of repair versus replacement. Eur J Cardiothorac Surg 1995;9:568–74. 11. Lorusso R, Gelsomino S, Vizzardi E, et al. Mitral valve repair or replacement for ischemic mitral regurgitation? The Italian Study on the Treatment of Ischemic Mitral Regurgitation (ISTIMIR). J Thorac Cardiovasc Surg 2013;145:128–39. 12. Vassileva CM, Boley T, Markwell S, Hazelrigg S. Metaanalysis of short-term and long-term survival following repair versus replacement for ischemic mitral regurgitation. Eur J Cardiothorac Surg 2011;39:295–303. 13. Perrault LP, Moskowitz AJ, Kron IL, et al. Optimal surgical management of severe ischemic mitral regurgitation: to repair or to replace? J Thorac Cardiovasc Surg 2012;143: 1396–403. 14. Jacobs JP, Mavroudis C, Jacobs ML, et al. What is operative mortality? Defining death in a surgical registry database: a report of the STS Congenital Database Taskforce and the Joint EACTS-STS Congenital Database Committee. Ann Thorac Surg 2006;81:1937–41. 15. Parmar MKB, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1988;17:2815–34. 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60. 17. Maltais S, Schaff HV, Daly RC, et al. Mitral regurgitation surgery in patients with ischemic cardiomyopathy and ischemic mitral regurgitation: factors that influence survival. J Thorac Cardiovasc Surg 2011;142:995–1001. 18. Reece TB, Tribble CG, Ellman PI, et al. Mitral repair is superior to replacement when associated with coronary artery disease. Ann Surg 2004;239:671–5. 19. Bonacchi M, Prifti E, Maiani M, Frati G, Nathan NS, Leacche M. Mitral valve surgery simultaneous to coronary revascularization in patients with end-stage ischemic cardiomyopathy. Heart Vessels 2006;21:20–7. 20. Mantovani V, Mariscalco G, Leva C, Blanzola C, Cattaneo P, Sala A. Long-term results of the surgical treatment of chronic ischemic mitral regurgitation: comparison of repair and prosthetic replacement. J Heart Valve Dis 2004;13:421–8. 21. Qiu Z, Chen X, Xu M, et al. Is mitral valve repair superior to replacement for chronic ischemic mitral regurgitation with left ventricular dysfunction? J Cardiothorac Surg 2010;5:107.
8
DAYAN ET AL REPLACEMENT VERSUS REPAIR IN MITRAL REGURGITATION
22. Magne J, Girerd N, Senechal M, et al. Mitral repair versus replacement for ischemic mitral regurgitation: comparison of short-term and long-term survival. Circulation 2009;120(11 suppl):S104–11. 23. Chan V, Ruel M, Mesana TG. Mitral valve replacement is a viable alternative to mitral valve repair for ischemic mitral regurgitation: a case-matched study. Ann Thorac Surg 2011;92:1358–65. 24. Ngaage DL, Daly RC, Rosales G, et al. Mitral regurgitation surgery in heart failure due to ischemic cardiomyopathy: a 24-year experience. J Heart Valve Dis 2008;17:251–9. 25. Grossi EA, Galloway AC, Miller JS, et al. Valve repair versus replacement for mitral insufficiency: when is a mechanical valve still indicated? J Thorac Cardiovasc Surg 1998;115: 389–94. 26. Kawachi Y, Oe M, Asou T, Tominaga R, Tokunaga K. Comparative study between valve repair and replacement for mitral pure regurgitation—early and late postoperative results. Jpn Circ J 1991;55:443–52. 27. Okita Y, Miki S, Ueda Y, Tahata T, Sakai T, Matsuyama K. Comparative evaluation of left ventricular performance after mitral valve repair or valve replacement with or without chordal preservation. J Heart Valve Dis 1993;2:159–66. 28. Sand ME, Naftel DC, Blackstone EH, Kirklin JW, Karp RB. A comparison of repair and replacement for mitral valve incompetence. J Thorac Cardiovasc Surg 1987;94:208–19. 29. Crabtree TD, Bailey MS, Moon MR, et al. Recurrent mitral regurgitation and risk factors for early and late mortality after mitral valve repair for functional ischemic mitral regurgitation. Ann Thorac Surg 2008;85:1537–43. 30. Glower DD, Tuttle RH, Shaw LK, Orozco RE, Rankin JS. Patients survival characteristics after routine mitral valve
Ann Thorac Surg 2013;-:-–-
31.
32.
33.
34. 35. 36. 37. 38. 39.
repair for ischemic mitral regurgitation. J Thorac Cardiovasc Surg 2005;129:860–8. Gelsomino S, Lorusso R, De Cicco G, et al. Five-year echocardiographic results of combined undersized mitral ring annuloplasty and coronary artery bypass grafting for chronic ischemic mitral regurgitation. Eur Heart J 2008;29:231–40. Ciarka A, Braun J, Delgado V, et al. Predictors of mitral regurgitation recurrence in patients with heart failure undergoing mitral valve annuloplasty. Am J Cardiol 2010;106: 395–401. Russo A, Grigioni F, Avierinos JF, et al. Thromboembolic complications after surgical correction of mitral regurgitation incidence, predictors, and clinical implications. J Am Coll Cardiol 2008;51:1203–11. Rao C, Murphy MO, Saso S, et al. Mitral valve repair or replacement for ischaemic mitral regurgitation: a systematic review. Heart Lung Circ 2011;20:555–65. Czer LS, Maurer G, Trento A, et al. Comparative efficacy of ring and suture annuloplasty for ischemic mitral regurgitation. Circulation 1992;86:II46–52. Calafiore AM, Di Mauro M, Gallina S, et al. Mitral valve surgery for chronic ischemic mitral regurgitation. Ann Thorac Surg 2004;77:1989–97. Hausmann H, Siniawski H, Hetzer R. Mitral valve reconstruction and replacement for ischemic mitral insufficiency: seven years’ follow up. J Heart Valve Dis 1999;8:536–42. Al-Radi OO, Austin PC, Tu JV, David TE, Yau TM. Mitral repair versus replacement for ischemic mitral regurgitation. Ann Thorac Surg 2005;79:1260–7. Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg 2001;122:1125–41.
