| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Thoracic and Cardiovascular Surgery, Heart and Diabetes Center North-Rhine Westfalia, University Hospital of Ruhr University Bochum, Bad Oeynhausen, Germany
Accepted for publication July 29, 2008.
* Address correspondence to Dr Maleszka, Department of Thoracic and Cardiovascular Surgery, Heart and Diabetes Center NRW, University Hospital of Ruhr University Bochum, Georgstrabe 11, Bad Oeynhausen, 32545, Germany (Email: amaleszka{at}hdz-nrw.de).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Methods: We performed a retrospective analysis of 55 patients aged 80 years and over with double valve replacement who were operated on at our institution between 2001 and 2005. Thirty-day mortality and 1-year survival were assessed.
Results: For most of the patients, stenosis was the cause of aortic valve surgery, whereas regurgitation was the cause of mitral valve replacement in the majority of patients. In one third of the patients, cardiac surgery had to be performed on an urgent/emergency basis. A large number of patients had concomitant diagnoses such as atrial fibrillation (73%), coronary artery disease (44%), renal insufficiency (29%), chronic obstructive pulmonary disease (20%), and diabetes mellitus (15%). In total, 16 patients (29%) died during follow-up. Survival rates at 30 days and 1 year were 91% and 71%, respectively. As determined by multivariable logistic regression analysis, Karnofsky performance status (hazard ratio: 0.899 per % increase; 95% confidence interval: 0.811 to 0.996; p = 0.043) and bypass time (hazard ratio: 1.062 per min; 95% confidence interval: 1.006 to 1.120; p = 0.028) were independent predictors of 30-day mortality. Beside these factors, additional independent predictors of 1-year mortality were preoperative stroke and postoperative intestinal failure (p = 0.008 and 0.003, respectively).
Conclusions: Our data demonstrate that, for selected octogenarians, double valve replacement can be performed with acceptable outcome. A better performance status of the patients at the time of cardiac surgery will probably improve 1-year survival.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
Although previous studies have demonstrated a higher risk for octogenarians undergoing heart surgery compared with younger patient groups [3–4], recent studies have shown that isolated aortic valve replacement in selected octogenarians can be performed with acceptable hospital and midterm mortality [5].
For multivalvular disease, it is reported in the literature that the overall operative risk of simultaneous aortic and mitral valve replacement lies between 4% and 13% and is, therefore, much higher than the risk of either aortic valve replacement or mitral valve replacement alone [6–8]. Recent reports suggest an even higher perioperative risk for octogenarians undergoing simultaneous aortic and mitral valve replacement [9, 10]. However, reliable data are scarce. Therefore, it is the aim of this study to determine whether simultaneous aortic and mitral valve replacement is a viable therapeutical option for multivalvular disease in this age group.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
All patients had preoperative coronary angiography and echocardiography or cardiac catheterization, or both, to assess hemodynamic data. All coronary lesions of at least 50% were considered for bypass surgery. Forty-two patients were admitted to our heart center for cardiac surgery from other clinics, whereas 13 patients were ambulatory. The choice of prosthetic valve was based on the patient's preference. However, all but 1 patient with atrial fibrillation wished implantation of a bioprosthetic instead of mechanical valves, as most of these patients wanted to avoid the lifelong strict oral anticoagulation regimen to prevent thromboembolic events of aortic and mitral valve prostheses. Postoperative antithrombotic treatment with intravenous unfractionated heparin was started for all patients in the intensive care unit. On the first postoperative day, oral anticoagulation therapy was started with phenprocoumon with a target international normalized ratio (INR) of 2.8 to 3.5 until discharge. Thereafter, patients checked INR values with their cardiologists. For bioprostheses, oral anticoagulation therapy was discontinued after 12 weeks postoperatively when other indications (eg, atrial fibrillation) for anticoagulation were absent.
The Ruhr University Ethics Committee, Bochum, Germany, approved the study, and the need for individual informed consent was waived.
Data Collection and Definitions
Preoperative, perioperative, and postoperative data were collected and entered into a computerized database. The 30-day mortality was registered. Variables for the short-term follow-up were demographic, preoperative and postoperative hemodynamic data, timing and comorbidities, surgical procedure, postoperative bleeding, and neurologic, renal, abdominal, and pulmonary complications. Renal dysfunction preoperatively and postoperatively was defined as a serum creatinine level greater than 2.0 mg/dL. Neurologic events were defined as the occurrence of a transient ischemic attack (fully reversible symptoms of short duration) or a stroke (central neurologic deficit persisting for more than 72 hours). Pulmonary complication was defined as prolonged mechanical ventilation (longer than 24 hours). Intestinal failure comprised clinical ileus and ileus with abnormal permeability diagnosed clinically, by ultrasonography or roentgenographic imaging, or both, leading to conservative or operative therapy. Low cardiac output was considered when cardiac index was below 1.8 L · min–1 · m–2. We calculated the logistic EuroSCORE (European System for Cardiac Operative Risk Evaluation) [11] for all patients. In addition, we assessed the Karnofsky performance status [12] preoperatively and postoperatively.
In May 2007, the surviving patients, their relatives, and their cardiologists or general practitioners were interviewed by telephone to assess 1-year survival and morbidities retrospectively. According to published guideline for reporting mortality after cardiac valve interventions [13], we classified causes of death as all-cause mortality, valve-related mortality, sudden unexpected death, and cardiac death.
Surgical Procedure
Valve replacements and concomitant procedures were performed using standard extracorporal circulation and median sternotomy, cold crystalloid cardioplegia (Bretschneider), and mild systemic hypothermia (30° to 32°C). For coronary artery bypass graft surgery, the left internal mammary artery and saphenous vein grafts were used. Tricuspid anuloplasty was performed as a modification of the De Vega technique in all cases.
Statistical Analysis
We report categorical variables using the number (n) and percent of observations. Continuous variables are expressed as mean and standard deviation or median and interquartile range (IQR) when appropriate. The Mann-Whitney U test was used for comparative evaluations of continuous variables. All p values less than 0.05 were considered statistically significant.
Survival rates were calculated with the Kaplan-Meier product-limit estimator. We also tested associations of risk factors to 30-day mortality and 1-year mortality. Risk factors were first tested in a univariate logistic regression model (30-day mortality) and a univariate Cox regression model (1-year mortality). As only preoperative and intraoperative risk factors are relevant for operative death, the following variables were tested in the univariate model for 30-day mortality: age, sex, body mass index, diabetes mellitus, logistic EuroSCORE, Karnofsky performance status, atrial fibrillation, stroke, myocardial infarction, renal insufficiency, coronary artery disease, patchplasty, redo, concomitant coronary artery bypass graft surgery, chronic obstructive pulmonary disease, and urgent/emergency operation. To assess predictors of 1-year mortality, we also included postoperative complications such as renal failure, gastrointestinal failure, respiratory failure, neurologic complications, low cardiac output syndrome, and rethoracotomy in the univariate model. Because the number of variables that can be included for multivariable testing is limited and depends on the number of events [14], we retained only variables with a p value less than 0.01 in the univariable analysis for multivariable proportional hazard analysis. We used the software SPSS, version 14 (Chicago, Illinois), to perform the analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
|
Fifty-four bioprostheses (34 porcine and 20 pericardial prostheses) and 1 mechanical bileaflet prostheses were implanted in aortic position; for the mitral position, 54 porcine bioprostheses and 1 bileaflet mechanical prosthesis were chosen. In the aortic position, 1 19-mm prosthesis (2%), and 16 21-mm (29%), 24 23-mm (44%), 11 25-mm (20%), and 3 27-mm prostheses (5%) were implanted. In mitral position, 1 25-mm prosthesis (2%), and 28 27-mm (51%), 19 29-mm (34%), and 7 31-mm (13%) prostheses were implanted. Clamping time and bypass time were 84.6 ± 18.2 minutes and 129.9 ± 36.0 minutes, respectively. During follow-up, no bleeding event or valve endocarditis occurred. No reintervention was necessary. Neurologic events occurred in 4 patients: a stroke in 2 patients, 1 of whom died in postoperative month 11, and transient ischemic attack in 2 patients. The frequency of other postoperative complications is presented in Table 2.
|
|
The univariate logistic regression analysis revealed that preoperative Karnofsky performance status and bypass time were predictors of 30-day survival. Both variables remained independent predictors of 30-day mortality in the multivariable logistic regression analysis (Karnofsky performance status, hazard ratio: 0.899 per percent increase; 95% CI: 0.811 to 0.996, p = 0.043; bypass time, hazard ratio: 1.062 per minute; 95% CI: 1.006 to 1.120, p = 0.028).
As determined by univariate Cox regression analysis, preoperative stroke, the logistic EuroSCORE, Karnofsky performance status, bypass time, postoperative respiratory failure, intestinal failure, and low cardiac output syndrome were predictors of 1-year mortality. As determined by multivariable analysis, however, only preoperative stroke, preoperative Karnofsky score, bypass time, and postoperative intestinal failure remained independent predictors of 1-year mortality (Table 3).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
It was a surprising finding that bypass time and intestinal failure were independent risk factors of 1-year survival, whereas preoperative left ventricular dysfunction was not. Kuwaki and coworkers [15] have identified preoperative ventricular dysfunction as the main predictor of survival in a group of 123 patients aged 53.8 years undergoing double valve replacement. We only may speculate that in our series the reason for the lack of left ventricular dysfunction as risk factor is due to the relatively small number of patients and a strong selection bias leading to a homogeneous group of patients. The bypass time, after the more complex surgical procedure of double valve replacement compared to isolated aortic valve replacement, may underline the important role of sufficient myocardial protection in this context. This finding underlines the need to prevent organ dysfunction, especially in this patient group.
Recent studies of octogenarians undergoing aortic valve replacement alone showed that the valve replacement on an urgent or emergency basis was an independent predictor of mortality, reflecting the failing heart during decompensation [16–18]. In our series, procedures were performed on urgent or emergency basis in one third of the patients. However, this classification did not predict mortality in our study cohort. Nevertheless, we cannot rule out that an urgent/emergency operation has influenced outcome indirectly by leading to longer bypass times. It may also be that other factors may reflect clinical status more appropriately than classification of cardiac surgery as elective, urgent, or emergent does. This assumption is in line with the finding that the preoperatively assessed Karnofsky performance status and the presence of stroke were independent predictors of mortality.
Earlier investigations of octogenarians with aortic valve replacement alone have shown that patients with concomitant coronary artery disease had more often left ventricular dysfunction after myocardial infarction, leading to a worse midterm and long-term outcome than patients with isolated aortic valve disease [1, 19–21]. Moreover, in an earlier series of 170 patients aged 50.5 years, concomitant tricuspid valve and aortocoronary bypass surgery were related to poorer survival rates after double valve replacement [8]. However, our data do not confirm the assumption that those factors are important independent risk factors for survival in cardiac surgery patients. Of the 16 patients who died within the first postoperative year, roughly 50% had concomitant surgical procedures such as tricuspid valve and aortocoronary bypass surgery or patchplasty of the aortic anulus.
In Germany, the mean life expectancy of an 82-year-old woman is 7.5 years, and for an 82-year-old man, it is 6.3 years [22]. These data indicate that clinical outcome was acceptable in the present study cohort with double valve replacement. The postoperative improvements in Karnofsky performance status for the survivors of our study demonstrate that double valve replacement not only added years to life but also reduced functional impairment in these patients. Nevertheless, it is also obvious from our data that 1-year mortality should be reduced.
Our study has some limitations. Firstly, this investigation has the general limitations of a retrospective study. Secondly, the study group included different types of bioprostheses (porcine and pericardial) and even one pair of bileaflet mechanical valves. That may have influenced study results independent of age and surgery procedure. Thirdly, the cardiologists referring the patients may have executed some form of bias. Patients considered too ill may have been denied referral to a surgical institution altogether. Finally, the number of variables in the multivariable analyses was limited owing to statistical restrictions. The use of more covariates may sometimes come to opposite conclusions [23]. From the clinical point of view, however, it is likely that preoperative functional impairment (as indicated by Karnofsky performance status) and unexpected perioperative and postoperative complications can significantly influence 1-year survival.
In summary, our data demonstrate that double valve replacement for selected patients aged 80 years and older can be performed with an acceptable outcome. Therefore, octogenarians should not be denied complex valvular surgery because of their age alone. A better performance status of the patients at the time of cardiac surgery would probably improve 1-year mortality.
|
Footnotes |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Lee, S. Li, J. S. Rankin, S. M. O'Brien, J. S. Gammie, E. D. Peterson, P. M. McCarthy, F. H. Edwards, and The Society of Thoracic Surgeons Adult Cardiac Sur Fifteen-Year Outcome Trends for Valve Surgery in North America Ann. Thorac. Surg., March 1, 2011; 91(3): 677 - 684. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
