- Email: [email protected]
Outcomes of Mitral Valve Repair Versus Replacement in the Elderly
Journal Pre-proof Outcomes of Mitral Valve Repair Versus Replacement in the Elderly Laura Seese, M.D., Ibrahim Sultan, M.D., Thomas Gleason, M.D., Yisi Wang, M.P.H., Floyd Thoma, B.S., Forozan Navid, M.D., Arman Kilic, M.D. PII:
S0003-4975(19)31296-2
DOI:
https://doi.org/10.1016/j.athoracsur.2019.07.057
Reference:
ATS 32994
To appear in:
The Annals of Thoracic Surgery
Received Date: 30 January 2019 Revised Date:
8 July 2019
Accepted Date: 15 July 2019
Please cite this article as: Seese L, Sultan I, Gleason T, Wang Y, Thoma F, Navid F, Kilic A, Outcomes of Mitral Valve Repair Versus Replacement in the Elderly, The Annals of Thoracic Surgery (2019), doi: https://doi.org/10.1016/j.athoracsur.2019.07.057. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 by The Society of Thoracic Surgeons
Outcomes of Mitral Valve Repair Versus Replacement in the Elderly Running Title: Mitral Valve Surgery in the Elderly
Laura Seese, M.D., Ibrahim Sultan, M.D., Thomas Gleason, M.D., Yisi Wang, M.P.H., Floyd Thoma, B.S., Forozan Navid, M.D., Arman Kilic, M.D.
From the Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA
Key Words: mitral valve; repair; replacement; elderly; recurrent mitral regurgitation
Word Count: 4,192
Correspondence and Reprint Requests: Arman Kilic, MD Division of Cardiac Surgery University of Pittsburgh Medical Center 200 Lothrop Street Suite C-700 Pittsburgh, PA 15213 Email: [email protected]
Abstract Background: This study evaluated outcomes of mitral valve surgery (MVS) in the elderly. Methods: Patients 75 years or older undergoing MVS at a single center between 2010-2018 were included. Patients were stratified into mitral valve repair (MVr) or replacement (MVR). The primary outcome was mortality. Secondary outcomes included postoperative complications, readmissions, and freedom from at least moderate mitral regurgitation. Multivariable Cox regression was used for risk adjustment. A subanalysis of isolated, non-reoperative MVS for degenerative disease was also performed. Results: 472 elderly patients underwent MVS: 301 (64%) MVr and 171 (36%) MVR. The majority of cases (68%) were performed for degenerative etiology. Survival was similar between MVr and MVR at 1-year (84% vs 86%, p=0.40) and 5-years (64% vs 64%, p=0.59). Postoperative complications were also comparable. Freedom from readmission was lower in the MVR cohort at 1-year (57% vs 68%, p=0.02) and 5-years (33% vs 43%, p=0.02). In isolated MVS for degenerative etiology, there was no difference in early or late mortality between MVr and MVR although readmissions were lower with MVr. Overall, freedom from mitral valve reoperation was comparable and the freedom from at least moderate regurgitation was 86% at 5-years in the MVr group. Conclusions: This study demonstrates that the outcomes of elderly patients undergoing MVS are acceptable with approximately 64% survival at 5 years. Our data suggests that MVr in the elderly may not confer a survival advantage but can be performed with durable results.
Abstract Word Count: 238
The benefits of mitral valve repair (MVr) over replacement (MVR) have been firmly established in degenerative mitral valve disease [1,2]. In fact, guidelines from the American Heart Association and American College of Cardiology (AHA/ACC) provide class I recommendations for MVr over MVR in chronic severe primary mitral regurgitation limited to the posterior leaflet [3]. In addition, class I recommendations are provided for chronic severe primary mitral regurgitation involving the anterior leaflet or both leaflets when a successful and durable repair can be achieved. The benefits of MVr in these settings relate to improved early and late survival, better preservation of left ventricular function, and freedom from prosthesisrelated complications such as bleeding events from anticoagulation, thromboembolic events, structural valve deterioration, and endocarditis [1-4]. The benefits of MVr are less established in non-degenerative etiologies as well as in patients with more limited life expectancy such as the elderly. Prior studies have evaluated outcomes of mitral valve surgery (MVS) specifically in elderly patients and have reported conflicting results [5,6]. The aim of this study was to review our longitudinal experience with MVS in the elderly population.
Patients and Methods Study Population Data were extracted from all patients who underwent MVS at a single institution between January 2010 and January 2018. Patients were included if aged 75 years or older. Isolated MVS as well as MVS with concomitant coronary artery bypass grafting (CABG), tricuspid valve surgery, or arrhythmia surgery were included. MVS with concomitant aortic procedures including aortic valve replacement were excluded. All etiologies of mitral valve disease were included. Patients with preoperative cardiogenic shock were excluded. Primary stratification was based on MVr versus MVR. The institutional review board approved this study.
Baseline Characteristics Baseline characteristics were compared between elderly patients who underwent MVr and MVR. Preoperative variables included age, sex, race, body mass index (BMI), ejection fraction, pulmonary artery systolic pressure, comorbidities, and laboratory values. Other variables included etiology of mitral regurgitation, severity of mitral regurgitation, operative status, cardiopulmonary bypass and ischemic times, annuloplasty or valve size, and Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM). All variable definitions were adopted from the STS database versions 2.73 and 2.81.
Outcomes The primary outcome was survival. Secondary outcomes included major postoperative complications, all-cause hospital readmission, recurrence of at least moderate mitral regurgitation, and mitral valve reoperation rates. Serial echocardiograms were typically obtained at 1 month, 6 months, and 1 year postoperatively. Major postoperative complications included prolonged ventilation >24 hours, acute renal failure, permanent stroke, deep sternal wound infection, sepsis, pneumonia, new-onset atrial fibrillation, and all-cause reoperation. The criteria for defining these complications were derived from the clinical definitions set forth by the STS. Postoperative blood product transfusion was defined the transfusion of packed red blood cells, fresh frozen plasma, platelets or cryoprecipitate.
Data Analysis Kaplan-Meier analyses were conducted to compare longitudinal outcomes between the MVr and MVR cohorts. Kaplan-Meier curves were compared using the log-rank test. Multivariable Cox regression analyses incorporating univariate preoperative predictors (inclusion criteria of two-tailed p<0.05) were conducted to evaluate the risk-adjusted impact of MVr versus MVR on survival as well as readmissions. In Kaplan-Meier analyses for readmission, censoring
occurred if the patient died during follow-up to address the competing events of death and readmission. As multiple readmissions can occur with each patient, the first readmission was considered the failure event where the patient was no longer free from readmission. Subanalyses were conducted for patients undergoing elective, non-reoperative, isolated MVS for degenerative disease and elderly patients undergoing urgent or emergent MVS. All continuous data are presented as mean ± standard deviation and all categorical data as number (percentage). Data were compared with the Mann-Whitney U test or student’s t-test. All statistical analyses were performed with version 9.4 SAS software (SAS Institute, Cary, NC).
Results Baseline Characteristics A total of 472 elderly patients (mean age 80 ± 4 years) underwent MVS during the study period, including 301 (64%) MVr and 171 (36%) MVR. Patients undergoing MVR were higher risk as evidenced by a higher STS-PROM (12.0% ± 8.2% vs 6.9% ± 6.1%; p<0.0001) (Table 1). There were more females, higher BMI, a higher proportion with prior open-heart surgery, and more patients with a history of arrhythmias, congestive heart failure, and prior mediastinal radiation in the MVR cohort (Table 1). In addition, there were more with rheumatic valve disease and moderate mitral regurgitation in MVR. Concomitant CABG was performed less frequently and concomitant tricuspid valve surgery more frequently in those undergoing MVR.
Outcomes Unadjusted operative mortality was similar amongst patients undergoing MVr and MVR (5.7% vs 7.0%, respectively; p=0.55) (Table 2). Readmission within 30 days of discharge from the index hospitalization was also similar for MVr (7.3%) and MVR (12.3%) patients (p=0.07). Furthermore, the rates of major postoperative complications were similar (Table 2) and these findings persisted irrespective of operative approach.
The median follow-up was 3.6 years (IQR 1.9 to 5.6 years), and was similar between the MVr and MVR groups. Unadjusted 1- and 5-year survival was similar (Figure 1). There was lower overall freedom from readmission in those undergoing MVR at 1- (68% vs 57%, p=0.02) and 5-years (43% vs 33%, p=0.02). The 5-year freedom from readmission for any cardiac cause was similar between MVR and MVr (49% MVR vs 43% MVr, p=0.26). The 5-year freedom from readmission specifically for heart failure was also similar between MVr (31%) and MVR (36%) patients (p=0.32). In risk-adjusted multivariable analysis, mitral valve repair versus replacement had no significant impact on the hazards for early or late mortality (Table 3). However, in elderly patients MVr had lower hazards for readmission at all time intervals, including a 38% lower adjusted hazard for readmission at 1-year and a 28% lower adjusted hazard for readmission at 5-years (Table 4). All patients had a postoperative echocardiogram. Overall freedom from at least moderate mitral valve regurgitation was significantly lower in the MVr group at 1-year (100% vs 93%, p=0.03) and 5-years (100% vs 86%, p=0.002) follow-up. The freedom from mitral valve reoperation at 5-years was also similar at 99% for MVR and 98% for MVr (log-rank p=0.44). The indications for reoperation were endocarditis (100%, n=2) in the MVR group and recurrent mitral regurgitation (75%, n=3) as well as endocarditis (25%, n=1) in the MVr group.
Sub-Analysis of Isolated MVS for Degenerative Etiology There were 112 elderly patients who underwent elective, non-reoperative, isolated MVS for degenerative mitral valve disease, including 37 MVRs and 75 MVrs. Unadjusted 1-year and 5-year survival was similar (Figure 2). The rates of operative mortality (2.7% MVr vs 5.4% MVR, p=0.60) and postoperative complications were also similar between the groups (Supplemental Table 1). Although the hazards for mortality at 5-years (hazard ratio [HR] for repair 1.04, 95% CI 0.46-2.37; p=0.92) were comparable between the groups, there was a 54% reduction in the hazards for readmission in elderly patients who underwent MVr compared to MVR
(Supplemental Table 2). Freedom from mitral valve reoperation at 5 years was comparable (MVR 100% versus MVr 99%; p=0.37), as was freedom from at least moderate mitral regurgitation (MVR 100% versus MVr 91%; p=0.07).
Sub-Analysis of Urgent and Emergent MVS There were 211 elderly patients who underwent urgent or emergent MVS during the study period, which included 129 MVrs and 82 MVRs. The rates of operative mortality in this subset were elevated but comparable between the groups (8.5% vs 8.5%, p=0.34). Additionally, the rates of postoperative complications were similar between MVrs and MVRs (Supplemental Table 3). Mitral valve repair had no impact on unadjusted or risk-adjusted mortality (Supplemental Table 4). Readmission at 5-year follow-up following urgent or emergent MVS was also not affected by MVr. The longitudinal survival was similar between the groups at 1year (79% MVR vs 81% MVr, p=0.72) and 5-years (55% MVR vs 50% MVr, p=0.79).
Comment Elderly patients are less frequently referred for MVS due to concerns of higher operative risk [7]. In addition, factors such as compromised tissue integrity, increased annular calcium burden, and higher proportion of concomitant coronary artery disease can alter the selection of MVr versus MVR more commonly in the elderly. In an analysis of the national STS database, mitral repair rates were 62% in patients less than 55 years of age and 45% in those 80 years of age or older (p<0.001) [8]. There is controversy regarding whether elderly patients who have more limited life expectancy benefit from the longer-term advantages of MVr. Some authors have demonstrated appreciable improvements in operative and longer-term outcomes with MVr, similar to what is observed in the general population particularly with degenerative disease [9,10]. Others have demonstrated comparable survival between MVr and MVR in the elderly [11,12].
Study Implications The current study adds to the literature by reporting a single center experience with MVS in the elderly. There are several implications of the study findings. Foremost, elderly patients undergoing MVR in this series are higher risk than MVr as demonstrated by a higher STSPROM, with higher risk demographics and comorbidities including female gender, prior congestive heart failure, redo surgery, and concomitant tricuspid valve regurgitation. This corresponds to findings from the national STS database for the general population where MVR patients are higher risk than those undergoing MVr [8]. Despite nearly double the STS-PROM in the MVR group, the unadjusted and riskadjusted operative mortality rates were comparable at 7.0% for MVR and 5.7% for MVr. Similarly, 5-year survival was comparable as well. Early and late survival were also comparable when limiting the analysis to isolated MVS for degenerative disease as well as in patients who underwent urgent or emergent MVS. In addition, rates of mitral valve reoperation were very low in both groups. These data collectively suggest that it is perhaps more justified to have a lower threshold for mitral valve replacement in the elderly. This is particularly true when considering that this is a higher risk patient subset, with more concomitant procedures being performed that translate into longer overall operative times. If mitral repair is not straightforward, for instance in complex bileaflet prolapse, cases of rheumatic valve disease or in decompensated patients undergoing urgent or emergent MVS, the risk of additional aortic cross-clamp time to perform a complex repair or the risk of additional cross-clamp to then replace the valve for a suboptimal initial attempt at repair, may be more detrimental than upfront replacement. This is evident in an analysis of 12,043 elderly patients undergoing MVS for degenerative valve disease that found the adjusted odds of mortality were 1.5-times higher in patients with MVR after failed
intraoperative MVr compared to MVRs without a repair attempt (Odds Ratio [OR] 1.52, 95% CI 1.03, 2.26) [13]. Unadjusted readmission rates were lower with mitral repair, particularly in those undergoing isolated MVS for degenerative disease, and this finding persisted after risk adjustment. In an analysis of over 21,000 Medicare beneficiaries, the 5-year freedom from allcause readmission was 22%, with higher rates of readmission for heart failure in patients undergoing MVR as opposed to MVr [14]. Possible reasons why readmission rates may be higher in MVR patients include readmissions for infective endocarditis, thromboembolic complications, structural valve deterioration or anticoagulation related complications. In the elderly, tissue valves are most commonly used as was observed in our study cohort. Some surgeons will utilize mechanical valves in elderly patients who are anticoagulated for other reasons. Although we have used mechanical valves in patients aged 70-75 years, we rarely use mechanical valves in patients over 75 years even if they are anticoagulated as these patients may need their anticoagulation stopped. The durability of mitral repair in our elderly cohort with degenerative disease was also similar to that reported in a general population of degenerative mitral valve patients, with over 90% durability at 5 years [15]. From this perspective, MVr is attractive when able to be performed efficiently and effectively in the elderly, for example in isolated MVS for technically straightforward degenerative disease. It is the cases of complex repairs for non-degenerative etiology or bileaflet disease where the additional benefit of repair may be questionable in the elderly.
Prior Studies of MVS in the Elderly Multiple studies have evaluated MVS in the elderly with conflicting results. A single institution series of 117 elderly patients undergoing MVS demonstrated improved early and late survival as well as lower postoperative stroke rates with mitral repair [9]. Authors of a
propensity matched analysis performed at 2 institutions that included 322 octogenarians undergoing MVS concluded that MVr was associated with better long-term survival than MVR in degenerative disease but that the survival benefit was more questionable in non-degenerative etiologies [10]. Another study demonstrated that mitral repair did not provide a long-term survival benefit in patients over the age of 60 years or in those patients requiring concomitant CABG [9]. A study of 59 octogenarians undergoing MVS for nonrheumatic mitral valve disease demonstrated comparable outcomes between MVr and MVR as well [12]. In contradistinction, a single center study of 341 patients aged 75 years or older undergoing MVS demonstrated improved 1, 2, and 5-year survival following mitral repair [16]. In a STS national database analysis that included 8,523 MVrs and 3,520 MVRs, operative mortality was reduced in the MVr cohort when compared to MVRs with and without chordal preservation [15]. Although this is the most robust retrospective study available in the literature to date, the focus is limited to early postoperative outcomes and does not evaluate longitudinal survival, which is a key factor in elderly patients who may have limited life expectancy.
Study Limitations This is a single institution retrospective study with all the inherent limitations of its design. Although our study is one the largest comparisons of elderly patients undergoing MVS, there is still a possibility of a type II error where there was a difference in outcomes between groups that simply was not detected due to sample size. There is also the potential for significant selection bias in choosing mitral repair versus replacement. We utilized the most current population in order to reduce any era effect; however, this limits the length of follow-up and may mask divergence of outcomes at longer follow-up. Furthermore, because of the specified time interval, the number of patients in the sub-analysis of patients with degenerative etiology is relatively small at 5-year follow-up, thus increasing the risk of a type II error.
Conclusions This study demonstrates that the outcomes of elderly patients undergoing MVS are acceptable with approximately 64% survival at 5 years. Our data suggests that mitral repair in the elderly does not confer a survival advantage but can be performed with durable results particularly in degenerative disease.
References 1. Daneshmand MA, Milano CA, Rankin JS, et al. Mitral valve repair for degenerative disease: a 20-year experience. Ann Thorac Surg. 2009; 88: 1828-37. 2. Shuhaiber J, Anderson RJ. Meta-analysis of clinical outcomes following surgical mitral valve repair or replacement. Eur J Cardiothorac Surg. 2007 Feb;31(2):267-75. 3. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC Focused Updat of the 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017 Jul 11;70(2):252-289. 4. Habib G, Thuny F, Avierinos JF. Prosthetic valve endocarditis: current approach and therapeutic options. Prog Cardiovasc Dis 2008; 50: 274-81. 5. Gaur P, Kaneko T, McGurk S, et al. Mitral valve repair versus replacement in the elderly: short-term and long-term outcomes. J Thorac Cardiovasc Surg. 2014 Oct;148(4):1400-6. 6. Ailawadi G, Swenson BR, Girotti ME, et al. Is mitral valve repair superior to replacement in elderly patients? Ann Thorac Surg. 2008 Jul;86(1):77-85. 7. Mirabel M, Lung B, Baron G, et al. What are the characteristics of patients with severe, symptomatic, mitral regurgitation who are denied surgery? Eur Heart J. 2007 Jun;28(11):1358-65. 8. Gammie JS, Sheng S, Griffith BP, et al. Trends in mitral valve surgery in the United States: results from the Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2009 May;87(5):1431-7. 9. Chikwe J, Goldstone AB, Passage J, et al. A propensity score-adjusted retrospective comparison of early and mid-term results of mitral valve repair versus replacement in octogenarians. Eur Heart J. 2011 Mar;32(5):618-26.
10. Thourani VH, Weintraub WS, Guyton RA, et al. Outcomes and long-term survival for patients undergoing mitral valve repair versus replacement: effect of age and concomitant coronary artery bypass grafting. Circulation. 2003 Jul 22;108(3):298-304. 11. DiGregorio V, Zehr KJ, Orszulak TA, et al. Results of mitral surgery in octogenarians with isolated nonrheumatic mitral regurgitation. Ann Thorac Surg. 2004; 78: 807-13. 12. Vassileva CM, Ghazanfari N, Spertus J, McNeely C, Markwell S, Hazelrigg S. Heart failure readmission after mitral valve repair and replacement: five year follow-up in the Medicare population. Ann Thorac Surg. 2014; 98(5): 1544-50. 13. Hendrix RJ, Bello RA, Flahive JM, et al. Mitral Valve Repair Versus Replacement in Elderly With Degenerative Disease: Analysis of the STS Adult Cardiac Surgery Database. Ann Thorac Surg. 2019 Mar;107(3):747-753. 14. David TE. Durability of mitral valve repair for mitral regurgitation due to degenerative mitral valve disease. Ann Cardiothorac Surg. 2015; 4(5): 417-21. 15. Silaschi M, Chaubey S, Aldalati O, et al. Is mitral valve repair superior to mitral valve replacement in elderly patients? Comparison of short- and long-term outcomes in a propensity-matched cohort. J Am Heart Assoc. 2016; 5(8): e003605.
Table 1. Baseline characteristics of elderly patients undergoing MVS stratified by mitral repair or replacement.
MV Repair N= 301 Age (years) Female Caucasian Race BMI (kg/m2) BSA (m2) Dyslipidemia Diabetes Mellitus Hypertension Chronic Lung Disease Preoperative Dialysis Serum Creatinine (mg/dL) Serum Albumin (g/dL) Total Bilirubin (mg/dL) Immunosuppression Peripheral Arterial Disease Cerebrovascular Disease Mediastinal Radiation Previous PCI Previous CABG Previous Valve Surgery Family History of CAD Previous MI NYHA Class Class I Class II Class III Class IV Prior Congestive Heart Failure Atrial Fibrillation Etiology of Mitral Valve Disease Degenerative Rheumatic Ischemic Endocarditis Degree of Mitral Regurgitation*
79.66 ± 3.58 142 (47.18%) 289 (96.01%) 26.83 ± 5.36 1.88 ± 0.24 223 (74.09%) 76 (25.25%) 262 (87.04%) 65 (21.59%) 3 (1.00%) 1.15 ± 0.46 3.60 ± 0.54 0.92 ± 0.54 18 (5.98%) 44 (14.62%) 83 (27.57%) 8 (3.04%) 63 (20.93%) 26 (8.64%) 6 (1.99%) 43 (14.29%) 100 (33.22%)
MV Replacement N=171 79.93 ± 3.68 103 (60.23%) 167 (97.66%) 27.66 ± 5.47 1.86 ± 0.20 127 (74.27%) 50 (29.24%) 149 (87.13%) 47 (27.49%) 1 (0.58%) 1.18 ± 0.67 3.61 ± 0.50 0.89 ± 0.48 10 (5.85%) 20 (11.70%) 53 (30.99%) 12 (7.45%) 40 (23.39%) 28 (16.37%) 43 (25.15%) 19 (11.11%) 57 (33.335)
109 (36.21%) 34 (11.30%) 98 (32.56%) 60 (19.93%) 152 (50.50%) 67 (22.26%)
56 (32.16%) 14 (8.19%) 53 (30.99%) 48 (28.07%) 114 (66.67%) 76 (44.44%)
209 (69.44%) 4 (1.33%) 85 (28.23%) 3 (1.00%)
118 (69.01%) 23 (13.45%) 28 (16.37%) 2 (1.17%)
P Value
0.43 0.006 0.34 0.06 0.55 0.97 0.35 0.98 0.15 1.00 0.72 0.81 0.37 0.95 0.37 0.43 0.04 0.53 0.01 <0.0001 0.33 0.98 0.28
0.0007 <0.0001 <0.0001
<0.0001
Moderate Severe Mitral Stenosis Number of Diseased Coronary Vessels 0 1 2 3 LVEF (%) LVEF Category < 25% 25%-40% 41%-59% ≥ 60% Pulmonary Artery Systolic Pressure (mmHg) Preoperative Intra-Aortic Balloon Pump Operative Status Elective Urgent Cardiopulmonary Bypass Time (min) Ischemic Time (min) Valve Type Mechanical Biologic Ring or Valve Size (mm) Surgical Approach Full Sternotomy Right Thoracotomy (Robotic or Minimally Invasive) Concomitant Procedures CABG Tricuspid Valve Surgery Ablation procedure STS Predicted Risk of Mortality % (STS PROM)
25 (8.31%) 270 (91.53%) 14 (4.75%)
36 (21.05%) 132 (78.57%) 76 (44.97%)
100 (33.33%) 48 (16.00%) 54 (18.00%) 98 (32.67%) 49.35 ± 13.03
68 (40.24%) 28 (16.57%) 24 (14.20%) 49 (28.99%) 53.26 ± 11.21
27 (8.97%) 52 (17.28%) 142 (47.18%) 80 (26.58%) 49.70 ± 15.70
4 (2.34%) 28 (16.37%) 77(45.03%) 62 (36.26%) 56.42 ± 14.99
2 (0.66%)
2 (1.17%)
172 (57.14%) 129 (42.86%) 147.69 ± 53.87 110.13 ± 42.09
89 (52.05%) 82 (47.95%) 156.77 ± 62.41 122.11 ± 52.06
NA NA 31.00 ± 3.56
18 (11.0%) 145 (89.0%) 29.19 ± 2.23
220 (73.09%)
131 (76.61%)
79 (26.24%)
40 (23.39%)
<0.0001 0.41
0.002 0.01
<0.0001
0.62 0.53
0.21 0.04
<0.0001 0.40
0.0002 133 (44.19%) 22 (7.31%) 29 (9.63%)
41 (23.98%) 21 (12.28%) 20 (11.70%) <0.0001
6.86% ± 6.05
12.02% ± 8.23
Abbreviations: BMI, Body Mass Index; BSA, Body Surface Area; CABG, coronary artery bypass grafting; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; STS, society for thoracic surgeons.
*From an available sample of 295 repairs and 168 replacements.
Table 2. Postoperative outcomes stratified by mitral repair or replacement.
Operative Mortality (30-day or inhospital) Blood Product Transfusion Prolonged Ventilation >24 Hours Acute Renal Failure Deep Sternal Wound Infection Sepsis Pneumonia Permanent Stroke Reoperation (In-Hospital, All-Cause) New-Onset Atrial Fibrillation
Mitral Valve Repair (n=301)
Mitral Valve Replacement (n=171)
P Value
17 (5.65%)
12 (7.02%)
0.55
156 (51.83%) 46 (15.28%) 19 (6.31%) 1 (0.33%) 7 (2.33%) 21 (6.98%) 7 (2.33%) 13 (4.32%) 52 (30.41%)
104 (60.82%) 35 (20.47%) 10 (5.85%) 2 (1.17%) 7 (4.09%) 10 (5.85%) 9 (5.26%) 14 (8.19%) 100 (33.22%)
0.06 0.15 0.84 0.30 0.28 0.63 0.09 0.08 0.53
Table 3. Predictors of mortality in multivariable Cox regression analysis.
Hazard Ratio Operative Mortality Mitral Repair (vs Replacement) Immunosuppression Hypertension Etiology Degenerative Rheumatic Ischemic Endocarditis 5-Year Mortality Mitral Repair (vs Replacement) Chronic Obstructive Pulmonary disease Immunosuppression Serum Albumin (increasing, per 1g/dL)
95% Confidence Interval
P Value
0.78 3.16 0.34
(0.34, 1.78) (1.01, 9.91) (0.13, 0.84)
0.56 0.048 0.02
Reference 0.54 1.12 19.0
Reference (0.07, 4.45) (0.20, 6.36) (2.73, 132.4)
Reference 0.56 0.22 0.003
0.86 1.78 1.95 0.40
(0.57, 1.30) (1.19, 2.67) (1.10, 3.46) (0.25, 0.68)
0.49 0.02 0.02 <0.0001
Non-significant predictors included in adjusted 30-day and 5-year mortality multivariable models: Age, preoperative intra-aortic balloon pump, sex, serum albumin, serum bilirubin, dialysis, number of diseased vessels, previous myocardial infarction, left ventricular ejection fraction, dialysis dependence, New York Heart Association class, diabetes, cerebrovascular disease and concomitant procedures.
Table 4. Significant predictors of readmission at 1-year and 5-years in multivariable Cox regression analysis using a competing risk method for mortality.
Hazard ratio
95% Confidence Interval
P Value
0.495, 0.937
0.02
1-Year Readmission Unadjusted MV Repair (vs Replacement)
0.68 Risk-Adjusted
MV Repair (vs Replacement)
0.62
0.440, 0.874
0.006
Creatinine Immunosuppression
1.28 2.38
1.048, 1.560 1.333, 4.238
0.02 0.003
0.553, 0.945
0.02
0.506, 0.902 0.295, 0.935 1.066, 1.574 1.027, 1.851
.008 0.03 .009 0.03
5-Year Readmission Unadjusted MV Repair (vs Replacement)
0.72 Risk-Adjusted
MV Repair (vs Replacement) Caucasian Race Creatinine Cerebrovascular Disease
0.68 0.53 1.30 1.38
Non-significant predictors included in adjusted 1-year and 5-year readmission multivariable models: number of diseased vessels, concomitant procedure, left ventricular ejection fraction, New York Heart Association Class, previous myocardial infarction, cerebrovascular disease, serum albumin, and concomitant procedures.
Figure Legends
Figure 1. Overall survival (A) and overall freedom from all-cause readmission (B) stratified by mitral valve repair versus replacement in the elderly.
Figure 2. Overall survival in isolated, elective non-reoperative mitral valve surgery for degenerative disease (A) and overall survival for elderly patients undergoing urgent or emergent mitral valve surgery (B) stratified by mitral valve repair versus replacement.
S0003-4975(19)31296-2
DOI:
https://doi.org/10.1016/j.athoracsur.2019.07.057
Reference:
ATS 32994
To appear in:
The Annals of Thoracic Surgery
Received Date: 30 January 2019 Revised Date:
8 July 2019
Accepted Date: 15 July 2019
Please cite this article as: Seese L, Sultan I, Gleason T, Wang Y, Thoma F, Navid F, Kilic A, Outcomes of Mitral Valve Repair Versus Replacement in the Elderly, The Annals of Thoracic Surgery (2019), doi: https://doi.org/10.1016/j.athoracsur.2019.07.057. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 by The Society of Thoracic Surgeons
Outcomes of Mitral Valve Repair Versus Replacement in the Elderly Running Title: Mitral Valve Surgery in the Elderly
Laura Seese, M.D., Ibrahim Sultan, M.D., Thomas Gleason, M.D., Yisi Wang, M.P.H., Floyd Thoma, B.S., Forozan Navid, M.D., Arman Kilic, M.D.
From the Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA
Key Words: mitral valve; repair; replacement; elderly; recurrent mitral regurgitation
Word Count: 4,192
Correspondence and Reprint Requests: Arman Kilic, MD Division of Cardiac Surgery University of Pittsburgh Medical Center 200 Lothrop Street Suite C-700 Pittsburgh, PA 15213 Email: [email protected]
Abstract Background: This study evaluated outcomes of mitral valve surgery (MVS) in the elderly. Methods: Patients 75 years or older undergoing MVS at a single center between 2010-2018 were included. Patients were stratified into mitral valve repair (MVr) or replacement (MVR). The primary outcome was mortality. Secondary outcomes included postoperative complications, readmissions, and freedom from at least moderate mitral regurgitation. Multivariable Cox regression was used for risk adjustment. A subanalysis of isolated, non-reoperative MVS for degenerative disease was also performed. Results: 472 elderly patients underwent MVS: 301 (64%) MVr and 171 (36%) MVR. The majority of cases (68%) were performed for degenerative etiology. Survival was similar between MVr and MVR at 1-year (84% vs 86%, p=0.40) and 5-years (64% vs 64%, p=0.59). Postoperative complications were also comparable. Freedom from readmission was lower in the MVR cohort at 1-year (57% vs 68%, p=0.02) and 5-years (33% vs 43%, p=0.02). In isolated MVS for degenerative etiology, there was no difference in early or late mortality between MVr and MVR although readmissions were lower with MVr. Overall, freedom from mitral valve reoperation was comparable and the freedom from at least moderate regurgitation was 86% at 5-years in the MVr group. Conclusions: This study demonstrates that the outcomes of elderly patients undergoing MVS are acceptable with approximately 64% survival at 5 years. Our data suggests that MVr in the elderly may not confer a survival advantage but can be performed with durable results.
Abstract Word Count: 238
The benefits of mitral valve repair (MVr) over replacement (MVR) have been firmly established in degenerative mitral valve disease [1,2]. In fact, guidelines from the American Heart Association and American College of Cardiology (AHA/ACC) provide class I recommendations for MVr over MVR in chronic severe primary mitral regurgitation limited to the posterior leaflet [3]. In addition, class I recommendations are provided for chronic severe primary mitral regurgitation involving the anterior leaflet or both leaflets when a successful and durable repair can be achieved. The benefits of MVr in these settings relate to improved early and late survival, better preservation of left ventricular function, and freedom from prosthesisrelated complications such as bleeding events from anticoagulation, thromboembolic events, structural valve deterioration, and endocarditis [1-4]. The benefits of MVr are less established in non-degenerative etiologies as well as in patients with more limited life expectancy such as the elderly. Prior studies have evaluated outcomes of mitral valve surgery (MVS) specifically in elderly patients and have reported conflicting results [5,6]. The aim of this study was to review our longitudinal experience with MVS in the elderly population.
Patients and Methods Study Population Data were extracted from all patients who underwent MVS at a single institution between January 2010 and January 2018. Patients were included if aged 75 years or older. Isolated MVS as well as MVS with concomitant coronary artery bypass grafting (CABG), tricuspid valve surgery, or arrhythmia surgery were included. MVS with concomitant aortic procedures including aortic valve replacement were excluded. All etiologies of mitral valve disease were included. Patients with preoperative cardiogenic shock were excluded. Primary stratification was based on MVr versus MVR. The institutional review board approved this study.
Baseline Characteristics Baseline characteristics were compared between elderly patients who underwent MVr and MVR. Preoperative variables included age, sex, race, body mass index (BMI), ejection fraction, pulmonary artery systolic pressure, comorbidities, and laboratory values. Other variables included etiology of mitral regurgitation, severity of mitral regurgitation, operative status, cardiopulmonary bypass and ischemic times, annuloplasty or valve size, and Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM). All variable definitions were adopted from the STS database versions 2.73 and 2.81.
Outcomes The primary outcome was survival. Secondary outcomes included major postoperative complications, all-cause hospital readmission, recurrence of at least moderate mitral regurgitation, and mitral valve reoperation rates. Serial echocardiograms were typically obtained at 1 month, 6 months, and 1 year postoperatively. Major postoperative complications included prolonged ventilation >24 hours, acute renal failure, permanent stroke, deep sternal wound infection, sepsis, pneumonia, new-onset atrial fibrillation, and all-cause reoperation. The criteria for defining these complications were derived from the clinical definitions set forth by the STS. Postoperative blood product transfusion was defined the transfusion of packed red blood cells, fresh frozen plasma, platelets or cryoprecipitate.
Data Analysis Kaplan-Meier analyses were conducted to compare longitudinal outcomes between the MVr and MVR cohorts. Kaplan-Meier curves were compared using the log-rank test. Multivariable Cox regression analyses incorporating univariate preoperative predictors (inclusion criteria of two-tailed p<0.05) were conducted to evaluate the risk-adjusted impact of MVr versus MVR on survival as well as readmissions. In Kaplan-Meier analyses for readmission, censoring
occurred if the patient died during follow-up to address the competing events of death and readmission. As multiple readmissions can occur with each patient, the first readmission was considered the failure event where the patient was no longer free from readmission. Subanalyses were conducted for patients undergoing elective, non-reoperative, isolated MVS for degenerative disease and elderly patients undergoing urgent or emergent MVS. All continuous data are presented as mean ± standard deviation and all categorical data as number (percentage). Data were compared with the Mann-Whitney U test or student’s t-test. All statistical analyses were performed with version 9.4 SAS software (SAS Institute, Cary, NC).
Results Baseline Characteristics A total of 472 elderly patients (mean age 80 ± 4 years) underwent MVS during the study period, including 301 (64%) MVr and 171 (36%) MVR. Patients undergoing MVR were higher risk as evidenced by a higher STS-PROM (12.0% ± 8.2% vs 6.9% ± 6.1%; p<0.0001) (Table 1). There were more females, higher BMI, a higher proportion with prior open-heart surgery, and more patients with a history of arrhythmias, congestive heart failure, and prior mediastinal radiation in the MVR cohort (Table 1). In addition, there were more with rheumatic valve disease and moderate mitral regurgitation in MVR. Concomitant CABG was performed less frequently and concomitant tricuspid valve surgery more frequently in those undergoing MVR.
Outcomes Unadjusted operative mortality was similar amongst patients undergoing MVr and MVR (5.7% vs 7.0%, respectively; p=0.55) (Table 2). Readmission within 30 days of discharge from the index hospitalization was also similar for MVr (7.3%) and MVR (12.3%) patients (p=0.07). Furthermore, the rates of major postoperative complications were similar (Table 2) and these findings persisted irrespective of operative approach.
The median follow-up was 3.6 years (IQR 1.9 to 5.6 years), and was similar between the MVr and MVR groups. Unadjusted 1- and 5-year survival was similar (Figure 1). There was lower overall freedom from readmission in those undergoing MVR at 1- (68% vs 57%, p=0.02) and 5-years (43% vs 33%, p=0.02). The 5-year freedom from readmission for any cardiac cause was similar between MVR and MVr (49% MVR vs 43% MVr, p=0.26). The 5-year freedom from readmission specifically for heart failure was also similar between MVr (31%) and MVR (36%) patients (p=0.32). In risk-adjusted multivariable analysis, mitral valve repair versus replacement had no significant impact on the hazards for early or late mortality (Table 3). However, in elderly patients MVr had lower hazards for readmission at all time intervals, including a 38% lower adjusted hazard for readmission at 1-year and a 28% lower adjusted hazard for readmission at 5-years (Table 4). All patients had a postoperative echocardiogram. Overall freedom from at least moderate mitral valve regurgitation was significantly lower in the MVr group at 1-year (100% vs 93%, p=0.03) and 5-years (100% vs 86%, p=0.002) follow-up. The freedom from mitral valve reoperation at 5-years was also similar at 99% for MVR and 98% for MVr (log-rank p=0.44). The indications for reoperation were endocarditis (100%, n=2) in the MVR group and recurrent mitral regurgitation (75%, n=3) as well as endocarditis (25%, n=1) in the MVr group.
Sub-Analysis of Isolated MVS for Degenerative Etiology There were 112 elderly patients who underwent elective, non-reoperative, isolated MVS for degenerative mitral valve disease, including 37 MVRs and 75 MVrs. Unadjusted 1-year and 5-year survival was similar (Figure 2). The rates of operative mortality (2.7% MVr vs 5.4% MVR, p=0.60) and postoperative complications were also similar between the groups (Supplemental Table 1). Although the hazards for mortality at 5-years (hazard ratio [HR] for repair 1.04, 95% CI 0.46-2.37; p=0.92) were comparable between the groups, there was a 54% reduction in the hazards for readmission in elderly patients who underwent MVr compared to MVR
(Supplemental Table 2). Freedom from mitral valve reoperation at 5 years was comparable (MVR 100% versus MVr 99%; p=0.37), as was freedom from at least moderate mitral regurgitation (MVR 100% versus MVr 91%; p=0.07).
Sub-Analysis of Urgent and Emergent MVS There were 211 elderly patients who underwent urgent or emergent MVS during the study period, which included 129 MVrs and 82 MVRs. The rates of operative mortality in this subset were elevated but comparable between the groups (8.5% vs 8.5%, p=0.34). Additionally, the rates of postoperative complications were similar between MVrs and MVRs (Supplemental Table 3). Mitral valve repair had no impact on unadjusted or risk-adjusted mortality (Supplemental Table 4). Readmission at 5-year follow-up following urgent or emergent MVS was also not affected by MVr. The longitudinal survival was similar between the groups at 1year (79% MVR vs 81% MVr, p=0.72) and 5-years (55% MVR vs 50% MVr, p=0.79).
Comment Elderly patients are less frequently referred for MVS due to concerns of higher operative risk [7]. In addition, factors such as compromised tissue integrity, increased annular calcium burden, and higher proportion of concomitant coronary artery disease can alter the selection of MVr versus MVR more commonly in the elderly. In an analysis of the national STS database, mitral repair rates were 62% in patients less than 55 years of age and 45% in those 80 years of age or older (p<0.001) [8]. There is controversy regarding whether elderly patients who have more limited life expectancy benefit from the longer-term advantages of MVr. Some authors have demonstrated appreciable improvements in operative and longer-term outcomes with MVr, similar to what is observed in the general population particularly with degenerative disease [9,10]. Others have demonstrated comparable survival between MVr and MVR in the elderly [11,12].
Study Implications The current study adds to the literature by reporting a single center experience with MVS in the elderly. There are several implications of the study findings. Foremost, elderly patients undergoing MVR in this series are higher risk than MVr as demonstrated by a higher STSPROM, with higher risk demographics and comorbidities including female gender, prior congestive heart failure, redo surgery, and concomitant tricuspid valve regurgitation. This corresponds to findings from the national STS database for the general population where MVR patients are higher risk than those undergoing MVr [8]. Despite nearly double the STS-PROM in the MVR group, the unadjusted and riskadjusted operative mortality rates were comparable at 7.0% for MVR and 5.7% for MVr. Similarly, 5-year survival was comparable as well. Early and late survival were also comparable when limiting the analysis to isolated MVS for degenerative disease as well as in patients who underwent urgent or emergent MVS. In addition, rates of mitral valve reoperation were very low in both groups. These data collectively suggest that it is perhaps more justified to have a lower threshold for mitral valve replacement in the elderly. This is particularly true when considering that this is a higher risk patient subset, with more concomitant procedures being performed that translate into longer overall operative times. If mitral repair is not straightforward, for instance in complex bileaflet prolapse, cases of rheumatic valve disease or in decompensated patients undergoing urgent or emergent MVS, the risk of additional aortic cross-clamp time to perform a complex repair or the risk of additional cross-clamp to then replace the valve for a suboptimal initial attempt at repair, may be more detrimental than upfront replacement. This is evident in an analysis of 12,043 elderly patients undergoing MVS for degenerative valve disease that found the adjusted odds of mortality were 1.5-times higher in patients with MVR after failed
intraoperative MVr compared to MVRs without a repair attempt (Odds Ratio [OR] 1.52, 95% CI 1.03, 2.26) [13]. Unadjusted readmission rates were lower with mitral repair, particularly in those undergoing isolated MVS for degenerative disease, and this finding persisted after risk adjustment. In an analysis of over 21,000 Medicare beneficiaries, the 5-year freedom from allcause readmission was 22%, with higher rates of readmission for heart failure in patients undergoing MVR as opposed to MVr [14]. Possible reasons why readmission rates may be higher in MVR patients include readmissions for infective endocarditis, thromboembolic complications, structural valve deterioration or anticoagulation related complications. In the elderly, tissue valves are most commonly used as was observed in our study cohort. Some surgeons will utilize mechanical valves in elderly patients who are anticoagulated for other reasons. Although we have used mechanical valves in patients aged 70-75 years, we rarely use mechanical valves in patients over 75 years even if they are anticoagulated as these patients may need their anticoagulation stopped. The durability of mitral repair in our elderly cohort with degenerative disease was also similar to that reported in a general population of degenerative mitral valve patients, with over 90% durability at 5 years [15]. From this perspective, MVr is attractive when able to be performed efficiently and effectively in the elderly, for example in isolated MVS for technically straightforward degenerative disease. It is the cases of complex repairs for non-degenerative etiology or bileaflet disease where the additional benefit of repair may be questionable in the elderly.
Prior Studies of MVS in the Elderly Multiple studies have evaluated MVS in the elderly with conflicting results. A single institution series of 117 elderly patients undergoing MVS demonstrated improved early and late survival as well as lower postoperative stroke rates with mitral repair [9]. Authors of a
propensity matched analysis performed at 2 institutions that included 322 octogenarians undergoing MVS concluded that MVr was associated with better long-term survival than MVR in degenerative disease but that the survival benefit was more questionable in non-degenerative etiologies [10]. Another study demonstrated that mitral repair did not provide a long-term survival benefit in patients over the age of 60 years or in those patients requiring concomitant CABG [9]. A study of 59 octogenarians undergoing MVS for nonrheumatic mitral valve disease demonstrated comparable outcomes between MVr and MVR as well [12]. In contradistinction, a single center study of 341 patients aged 75 years or older undergoing MVS demonstrated improved 1, 2, and 5-year survival following mitral repair [16]. In a STS national database analysis that included 8,523 MVrs and 3,520 MVRs, operative mortality was reduced in the MVr cohort when compared to MVRs with and without chordal preservation [15]. Although this is the most robust retrospective study available in the literature to date, the focus is limited to early postoperative outcomes and does not evaluate longitudinal survival, which is a key factor in elderly patients who may have limited life expectancy.
Study Limitations This is a single institution retrospective study with all the inherent limitations of its design. Although our study is one the largest comparisons of elderly patients undergoing MVS, there is still a possibility of a type II error where there was a difference in outcomes between groups that simply was not detected due to sample size. There is also the potential for significant selection bias in choosing mitral repair versus replacement. We utilized the most current population in order to reduce any era effect; however, this limits the length of follow-up and may mask divergence of outcomes at longer follow-up. Furthermore, because of the specified time interval, the number of patients in the sub-analysis of patients with degenerative etiology is relatively small at 5-year follow-up, thus increasing the risk of a type II error.
Conclusions This study demonstrates that the outcomes of elderly patients undergoing MVS are acceptable with approximately 64% survival at 5 years. Our data suggests that mitral repair in the elderly does not confer a survival advantage but can be performed with durable results particularly in degenerative disease.
References 1. Daneshmand MA, Milano CA, Rankin JS, et al. Mitral valve repair for degenerative disease: a 20-year experience. Ann Thorac Surg. 2009; 88: 1828-37. 2. Shuhaiber J, Anderson RJ. Meta-analysis of clinical outcomes following surgical mitral valve repair or replacement. Eur J Cardiothorac Surg. 2007 Feb;31(2):267-75. 3. Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC Focused Updat of the 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017 Jul 11;70(2):252-289. 4. Habib G, Thuny F, Avierinos JF. Prosthetic valve endocarditis: current approach and therapeutic options. Prog Cardiovasc Dis 2008; 50: 274-81. 5. Gaur P, Kaneko T, McGurk S, et al. Mitral valve repair versus replacement in the elderly: short-term and long-term outcomes. J Thorac Cardiovasc Surg. 2014 Oct;148(4):1400-6. 6. Ailawadi G, Swenson BR, Girotti ME, et al. Is mitral valve repair superior to replacement in elderly patients? Ann Thorac Surg. 2008 Jul;86(1):77-85. 7. Mirabel M, Lung B, Baron G, et al. What are the characteristics of patients with severe, symptomatic, mitral regurgitation who are denied surgery? Eur Heart J. 2007 Jun;28(11):1358-65. 8. Gammie JS, Sheng S, Griffith BP, et al. Trends in mitral valve surgery in the United States: results from the Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2009 May;87(5):1431-7. 9. Chikwe J, Goldstone AB, Passage J, et al. A propensity score-adjusted retrospective comparison of early and mid-term results of mitral valve repair versus replacement in octogenarians. Eur Heart J. 2011 Mar;32(5):618-26.
10. Thourani VH, Weintraub WS, Guyton RA, et al. Outcomes and long-term survival for patients undergoing mitral valve repair versus replacement: effect of age and concomitant coronary artery bypass grafting. Circulation. 2003 Jul 22;108(3):298-304. 11. DiGregorio V, Zehr KJ, Orszulak TA, et al. Results of mitral surgery in octogenarians with isolated nonrheumatic mitral regurgitation. Ann Thorac Surg. 2004; 78: 807-13. 12. Vassileva CM, Ghazanfari N, Spertus J, McNeely C, Markwell S, Hazelrigg S. Heart failure readmission after mitral valve repair and replacement: five year follow-up in the Medicare population. Ann Thorac Surg. 2014; 98(5): 1544-50. 13. Hendrix RJ, Bello RA, Flahive JM, et al. Mitral Valve Repair Versus Replacement in Elderly With Degenerative Disease: Analysis of the STS Adult Cardiac Surgery Database. Ann Thorac Surg. 2019 Mar;107(3):747-753. 14. David TE. Durability of mitral valve repair for mitral regurgitation due to degenerative mitral valve disease. Ann Cardiothorac Surg. 2015; 4(5): 417-21. 15. Silaschi M, Chaubey S, Aldalati O, et al. Is mitral valve repair superior to mitral valve replacement in elderly patients? Comparison of short- and long-term outcomes in a propensity-matched cohort. J Am Heart Assoc. 2016; 5(8): e003605.
Table 1. Baseline characteristics of elderly patients undergoing MVS stratified by mitral repair or replacement.
MV Repair N= 301 Age (years) Female Caucasian Race BMI (kg/m2) BSA (m2) Dyslipidemia Diabetes Mellitus Hypertension Chronic Lung Disease Preoperative Dialysis Serum Creatinine (mg/dL) Serum Albumin (g/dL) Total Bilirubin (mg/dL) Immunosuppression Peripheral Arterial Disease Cerebrovascular Disease Mediastinal Radiation Previous PCI Previous CABG Previous Valve Surgery Family History of CAD Previous MI NYHA Class Class I Class II Class III Class IV Prior Congestive Heart Failure Atrial Fibrillation Etiology of Mitral Valve Disease Degenerative Rheumatic Ischemic Endocarditis Degree of Mitral Regurgitation*
79.66 ± 3.58 142 (47.18%) 289 (96.01%) 26.83 ± 5.36 1.88 ± 0.24 223 (74.09%) 76 (25.25%) 262 (87.04%) 65 (21.59%) 3 (1.00%) 1.15 ± 0.46 3.60 ± 0.54 0.92 ± 0.54 18 (5.98%) 44 (14.62%) 83 (27.57%) 8 (3.04%) 63 (20.93%) 26 (8.64%) 6 (1.99%) 43 (14.29%) 100 (33.22%)
MV Replacement N=171 79.93 ± 3.68 103 (60.23%) 167 (97.66%) 27.66 ± 5.47 1.86 ± 0.20 127 (74.27%) 50 (29.24%) 149 (87.13%) 47 (27.49%) 1 (0.58%) 1.18 ± 0.67 3.61 ± 0.50 0.89 ± 0.48 10 (5.85%) 20 (11.70%) 53 (30.99%) 12 (7.45%) 40 (23.39%) 28 (16.37%) 43 (25.15%) 19 (11.11%) 57 (33.335)
109 (36.21%) 34 (11.30%) 98 (32.56%) 60 (19.93%) 152 (50.50%) 67 (22.26%)
56 (32.16%) 14 (8.19%) 53 (30.99%) 48 (28.07%) 114 (66.67%) 76 (44.44%)
209 (69.44%) 4 (1.33%) 85 (28.23%) 3 (1.00%)
118 (69.01%) 23 (13.45%) 28 (16.37%) 2 (1.17%)
P Value
0.43 0.006 0.34 0.06 0.55 0.97 0.35 0.98 0.15 1.00 0.72 0.81 0.37 0.95 0.37 0.43 0.04 0.53 0.01 <0.0001 0.33 0.98 0.28
0.0007 <0.0001 <0.0001
<0.0001
Moderate Severe Mitral Stenosis Number of Diseased Coronary Vessels 0 1 2 3 LVEF (%) LVEF Category < 25% 25%-40% 41%-59% ≥ 60% Pulmonary Artery Systolic Pressure (mmHg) Preoperative Intra-Aortic Balloon Pump Operative Status Elective Urgent Cardiopulmonary Bypass Time (min) Ischemic Time (min) Valve Type Mechanical Biologic Ring or Valve Size (mm) Surgical Approach Full Sternotomy Right Thoracotomy (Robotic or Minimally Invasive) Concomitant Procedures CABG Tricuspid Valve Surgery Ablation procedure STS Predicted Risk of Mortality % (STS PROM)
25 (8.31%) 270 (91.53%) 14 (4.75%)
36 (21.05%) 132 (78.57%) 76 (44.97%)
100 (33.33%) 48 (16.00%) 54 (18.00%) 98 (32.67%) 49.35 ± 13.03
68 (40.24%) 28 (16.57%) 24 (14.20%) 49 (28.99%) 53.26 ± 11.21
27 (8.97%) 52 (17.28%) 142 (47.18%) 80 (26.58%) 49.70 ± 15.70
4 (2.34%) 28 (16.37%) 77(45.03%) 62 (36.26%) 56.42 ± 14.99
2 (0.66%)
2 (1.17%)
172 (57.14%) 129 (42.86%) 147.69 ± 53.87 110.13 ± 42.09
89 (52.05%) 82 (47.95%) 156.77 ± 62.41 122.11 ± 52.06
NA NA 31.00 ± 3.56
18 (11.0%) 145 (89.0%) 29.19 ± 2.23
220 (73.09%)
131 (76.61%)
79 (26.24%)
40 (23.39%)
<0.0001 0.41
0.002 0.01
<0.0001
0.62 0.53
0.21 0.04
<0.0001 0.40
0.0002 133 (44.19%) 22 (7.31%) 29 (9.63%)
41 (23.98%) 21 (12.28%) 20 (11.70%) <0.0001
6.86% ± 6.05
12.02% ± 8.23
Abbreviations: BMI, Body Mass Index; BSA, Body Surface Area; CABG, coronary artery bypass grafting; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; STS, society for thoracic surgeons.
*From an available sample of 295 repairs and 168 replacements.
Table 2. Postoperative outcomes stratified by mitral repair or replacement.
Operative Mortality (30-day or inhospital) Blood Product Transfusion Prolonged Ventilation >24 Hours Acute Renal Failure Deep Sternal Wound Infection Sepsis Pneumonia Permanent Stroke Reoperation (In-Hospital, All-Cause) New-Onset Atrial Fibrillation
Mitral Valve Repair (n=301)
Mitral Valve Replacement (n=171)
P Value
17 (5.65%)
12 (7.02%)
0.55
156 (51.83%) 46 (15.28%) 19 (6.31%) 1 (0.33%) 7 (2.33%) 21 (6.98%) 7 (2.33%) 13 (4.32%) 52 (30.41%)
104 (60.82%) 35 (20.47%) 10 (5.85%) 2 (1.17%) 7 (4.09%) 10 (5.85%) 9 (5.26%) 14 (8.19%) 100 (33.22%)
0.06 0.15 0.84 0.30 0.28 0.63 0.09 0.08 0.53
Table 3. Predictors of mortality in multivariable Cox regression analysis.
Hazard Ratio Operative Mortality Mitral Repair (vs Replacement) Immunosuppression Hypertension Etiology Degenerative Rheumatic Ischemic Endocarditis 5-Year Mortality Mitral Repair (vs Replacement) Chronic Obstructive Pulmonary disease Immunosuppression Serum Albumin (increasing, per 1g/dL)
95% Confidence Interval
P Value
0.78 3.16 0.34
(0.34, 1.78) (1.01, 9.91) (0.13, 0.84)
0.56 0.048 0.02
Reference 0.54 1.12 19.0
Reference (0.07, 4.45) (0.20, 6.36) (2.73, 132.4)
Reference 0.56 0.22 0.003
0.86 1.78 1.95 0.40
(0.57, 1.30) (1.19, 2.67) (1.10, 3.46) (0.25, 0.68)
0.49 0.02 0.02 <0.0001
Non-significant predictors included in adjusted 30-day and 5-year mortality multivariable models: Age, preoperative intra-aortic balloon pump, sex, serum albumin, serum bilirubin, dialysis, number of diseased vessels, previous myocardial infarction, left ventricular ejection fraction, dialysis dependence, New York Heart Association class, diabetes, cerebrovascular disease and concomitant procedures.
Table 4. Significant predictors of readmission at 1-year and 5-years in multivariable Cox regression analysis using a competing risk method for mortality.
Hazard ratio
95% Confidence Interval
P Value
0.495, 0.937
0.02
1-Year Readmission Unadjusted MV Repair (vs Replacement)
0.68 Risk-Adjusted
MV Repair (vs Replacement)
0.62
0.440, 0.874
0.006
Creatinine Immunosuppression
1.28 2.38
1.048, 1.560 1.333, 4.238
0.02 0.003
0.553, 0.945
0.02
0.506, 0.902 0.295, 0.935 1.066, 1.574 1.027, 1.851
.008 0.03 .009 0.03
5-Year Readmission Unadjusted MV Repair (vs Replacement)
0.72 Risk-Adjusted
MV Repair (vs Replacement) Caucasian Race Creatinine Cerebrovascular Disease
0.68 0.53 1.30 1.38
Non-significant predictors included in adjusted 1-year and 5-year readmission multivariable models: number of diseased vessels, concomitant procedure, left ventricular ejection fraction, New York Heart Association Class, previous myocardial infarction, cerebrovascular disease, serum albumin, and concomitant procedures.
Figure Legends
Figure 1. Overall survival (A) and overall freedom from all-cause readmission (B) stratified by mitral valve repair versus replacement in the elderly.
Figure 2. Overall survival in isolated, elective non-reoperative mitral valve surgery for degenerative disease (A) and overall survival for elderly patients undergoing urgent or emergent mitral valve surgery (B) stratified by mitral valve repair versus replacement.