|
|
||||||||
Eur J Cardiothorac Surg 2006;29:133-138
© 2006 Elsevier Science NL
Division of Cardiovascular Surgery, Department of Surgery, Toronto General Hospital and University of Toronto, Toronto, Canada
Received 17 July 2005; received in revised form 8 November 2005; accepted 14 November 2005.
* Corresponding author. Address: Department of Cardiothoracic Surgery, King's College Hospital, London SE5 9RS, United Kingdom. Tel.: +44 207 3464341; fax: +44 207 3463433. (Email: andrew.sara{at}btinternet.com).
| Abstract |
|---|
|
|
|---|
Key Words: Aortic valve replacement Ventricular function Outcome
| 1. Introduction |
|---|
|
|
|---|
The long asymptomatic course of both AS and AI means that many patients have impaired ventricular function at diagnosis. Outcomes remain poorly characterised, however, for patients with impaired ventricular function who undergo AVR. Robust predictors of perioperative and long-term mortality remain elusive and the impact of valvular pathophysiology and other risk factors has yet to be clearly defined.
In order to assess the potential benefit of AVR in patients with impaired ventricular function, it is necessary to accumulate data on early as well as long-term survival. To this end, we evaluated the in-hospital outcomes and long-term results in a sub-group of patients with significant preoperative left ventricular dysfunction (ejection fraction < 40%) who underwent AVR with or without concomitant coronary artery bypass grafting (CABG). We sought to establish whether AVR could be performed at an acceptable operative and long-term risk in patients with significantly impaired ventricular function.
| 2. Materials and methods |
|---|
|
|
|---|
2.2 Operative technique
Standard anaesthesia and surgical techniques, cardiopulmonary bypass (CPB) and myocardial protection techniques were employed. CPB was instituted using ascending aortic cannulation and two-stage venous cannulation of the right atrium. Under mild systemic hypothermia (34 °C), cardiac arrest was induced and maintained using intermittent hyperkalaemic cold blood cardioplegia. The composition of the cardioplegic solution changed during the study period. In 1990, we employed 4:1 blood cardioplegia, gradually increasing the proportion of blood to crystalloid over the ensuing decade. Currently, we use blood supplemented with potassium and magnesium (Quest MPS® microplegia system; Quest Medical Inc., Allen, TX, USA). When CABG was required, proximal anastomoses were constructed before removal of the aortic cross clamp.
2.3 Follow-up
All patients were followed up prospectively at periodic intervals. Recent follow-up was conducted by either mailed questionnaire or telephone interview between January and July 2004. When patients could not be contacted via these methods, follow-up was obtained from their family physician. Follow-up of all patients was 100% complete.
2.4 Statistics
Data are expressed as mean ± standard deviation. Ordinal and nominal data were compared using the
2 test, or Fisher's exact test when appropriate. Continuous variables were compared with the unpaired Student's t-test. The preoperative variables tested for their univariate and multivariable association with outcomes included age, sex, previous cardiac surgery, endocarditis, preoperative renal failure, hypertension, coronary artery disease, preoperative arrhythmia, diabetes, peripheral vascular disease and valvular lesion (AS or AI).
Outcomes were evaluated multivariably by logistic regression analysis. Long-term outcomes were evaluated by Cox regression analyses. For the multivariable models, variables with a univariate p-value of less than 0.25, or of known biological significance, were examined. A stepwise procedure (backward Wald elimination test) was employed and a p-value of less than 0.05 used to enter and eliminate variables.
| 3. Results |
|---|
|
|
|---|
|
|
|
|
|
|
|
| 4. Discussion |
|---|
|
|
|---|
Chronic AI differs from AS in that both pressure and volume overload impose a combination of increased preload and afterload on the left ventricle. The excess (regurgitant) volume must be expelled into the high-pressure systemic circulation in early systole. Eccentric ventricular hypertrophy occurs in response to increased left ventricular end-diastolic wall stress [8] and the left ventricle compensates by means of a process called "afterload mismatch, preload reserve" [3]. Preload is increased so as to maintain forward flow according to the FrankStarling relationship and the ventricle dilates. Once the limit of preload reserve is reached, left ventricular systolic function declines and as LVEF falls, left ventricular end systolic volume increases leading to an increase in systolic wall stress (afterload) which further depresses pump function [8,9]. The adaptive process leads to myocardial fibrosis, possibly as a result of myocardial ischaemia, which occurs because of decreased coronary flow in diastole combined with increased myocardial oxygen consumption [10]. The proto-oncogene c-myc is also implicated in the pathogenesis of myocardial hypertrophy and fibrosis in chronic AI [11].
The long asymptomatic course of AS and AI means that many patients have impaired left ventricular function at diagnosis, but the balance of risks and benefits of AVR versus alternative treatments is not well known in this subgroup of patients. Impaired left ventricular function has long been recognised as a predictor of adverse outcome after AVR [2,8] and some previous studies have even suggested that impaired ventricular function should be considered a contraindication to AVR [12].
The increased operative risk must, however, be seen in the context of the dismal natural history of AS and AI when associated with impaired ventricular function [3,4]. The expected survival, for example, is less than two years in patients with severe AS [3] or AR [4] in association with congestive heart failure. In addition, we have previously reported that whilst there has, in recent years, been an increase the proportion of patients undergoing AVR who are high-risk by virtue of impaired ventricular function, there has been no associated increase in morbidity or mortality after AVR [13].
The current study suggests that AVR can be performed with very low perioperative morbidity and mortality in patients with poor left ventricular function. Operative mortality was 2.3% overall; 1.5% in the AS group and 0.8% in the AI group. Similarly, our long-term results suggest that the outlook after surgery is excellent (Table 5). These results do, however, contrast with those reported in several contemporary studies in which operative mortality ranged from 8 to 21% [10,1417]. Other series have, however, reported outcomes more consistent with the findings in this study. Bevilacqua et al. [18], for example, reported a 5.7% operative mortality when patients with severe AS and LVEF < 35% underwent AVR, whilst Rediker et al. [19] reported no operative deaths in 42 patients with a mean aortic valve area of 0.7 cm2 and LVEF < 45%. Klodas et al. [20] also reported 0% operative mortality for AVR in patients with severe AI and left ventricular end diastolic diameters greater than 80 mm.
The published long-term outcomes show similar variation. Nonetheless many contemporary series have demonstrated long-term outcomes more in keeping with our results. Several authors have reported 5-year survival rates between 64 and 84% [2,10,17,21].
The not very ideal methods of myocardial protection, which were used in the early series probably account, at least in part, for some of the poor outcomes observed, but the problem of patient selection, particularly in AS, may also have played a part. Carabello et al. [22], in a study published in 1980, were the first to highlight the poor outcome of AVR performed for severe AS in patients with marked left ventricular dysfunction and low transvalvular gradients. These authors reported a 75% operative mortality and Carabello's results prompted surgeons to try better to define which subgroups with AS and ventricular dysfunction might benefit from AVR.
It is now understood that patients with impaired ventricular function who undergo AVR for severe AS fall roughly into one of three groups. The first group comprises patients who have high systolic left ventricular pressures and transvalvular gradients; these patients will generally benefit from AVR. The other two groups consist of patients, such as those in the present study, who have low transvalvular gradients. In these situations, the mean pressure gradient generated by the left ventricle may be low despite the presence of severe stenosis. Where the low gradient is because of "afterload mismatch", which limits myocardial fibre shortening and thereby apparently reduces left ventricular systolic function, AVR is associated with good outcomes [8]. Other patients with severe AS and low transvalvular gradients will, however, respond poorly after AVR. These patients are assumed to have a coexisting cardiomyopathy in addition to the myocardial dysfunction which attends "afterload mismatch". It is in this last group that the effects of aortic valve surgery are least well established.
A heterogeneous mixture of patients drawn from these three different groups may account for the wide variation in clinical outcomes seen in patients with AS and poor ventricular function. In addition, some patients with low transvalvular gradients and an irreversible cardiomyopathy may be incorrectly classified as having severe AS because of limitations in the techniques used to calculate aortic valve area [23]. Even when dobutamine stress echocardiography is employed preoperatively, 20% of patients with low transvalvular gradients and AS will defy classification into either the "afterload mismatch" or cardiomyopathy categories.
In contrast to AS, the problem of when to intervene in AI is now relatively clear [1]. Although left ventricular dysfunction also increases operative risks in AI, it remains possible to achieve excellent results even in the face of extreme left ventricular dilatation and apparently irreversible ventricular dysfunction [20]. The increased mortality risk after AVR of patients with AI and left ventricular systolic dysfunction is primarily due to postoperative congestive heart failure. Congestive heart failure was responsible for as many as 78% of late deaths in the era before the importance of surgical intervention before the onset of ventricular dysfunction was appreciated [24].
Multivariable analysis did not establish predictors either of early or late mortality in this study. In the case of operative mortality, the very low death rate meant that logistic regression was not deemed appropriate to evaluate predictors of perioperative mortality. Cox regression analysis was performed to evaluate late outcomes. The failure to identify independent predictors of mortality distinguishes this study from many others, but these studies show no consistency with respect to the variables identified. Rothenburger et al. [17], for example, identified elevated creatinine, NYHA III or IV, concomitant CABG and left ventricular end systolic diameter > 54 mm as independent predictors of mortality in a cohort of patients with both AS and AI with impaired left ventricular function. Sharony et al. [10], however, found that only renal disease and advanced age predicted mortality, but Duarte et al. [21] identified diuretic use, male sex, re-operation, age > 60 years and AI as independent predictors of late death. The same study found that LVEF < 30%, mitral regurgitation, concomitant CABG and emergency or redo surgery predicted hospital mortality which was reported as of 12.5% [21].
The explanation for these discrepant reports is, once again, likely to be the heterogeneous nature of the groups studied both in terms of valvular and ventricular pathophysiology as well differences in such factors as myocardial protection and the era during which the patients underwent surgery.
Overall, the 10-year-survival for the patients in this series approached 55%, which compares favourably with heart transplantation and particularly with medical therapy. Pereira et al. [25], for example, found that in AS with severe left ventricular dysfunction, 1- and 4-year survival rates were similar to our findings and were markedly improved in surgical patients as compared with medically treated patients (82% vs 78% at 1 year and 41% vs 15% at 4 years).
| 5. Limitations |
|---|
|
|
|---|
We have emphasised the importance of accurate characterisation of valvular pathophysiology particularly in low gradient AS. In this study, however, few patients underwent dobutamine stress echocardiography which we have only used to risk stratify patients and not to include or exclude surgical candidacy. Dobutamine stress echocardiography is not performed routinely in our unit, but only when other pre-existing comorbidity indicates a need to demonstrate recoverable myocardium. In the absence of recruitable myocardium, we are then better able to inform patients of the high-risk nature of their surgery and, conversely, can reassure patients when recoverable myocardium is seen.
Causes of ventricular dysfunction other than aortic valve pathology with or without concomitant coronary artery disease cannot be excluded in the patient population studied.
| 6. Conclusions |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. R. Kamath, P. Varadarajan, R. Turk, U. Sampat, R. Patel, S. Khandhar, and R. G. Pai Survival in Patients With Severe Aortic Regurgitation and Severe Left Ventricular Dysfunction Is Improved by Aortic Valve Replacement: Results From a Cohort of 166 Patients With an Ejection Fraction <=35% Circulation, September 15, 2009; 120(11_suppl_1): S134 - S138. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Nicolini and T. Gherli Alternatives to transplantation in the surgical therapy for heart failure Eur. J. Cardiothorac. Surg., February 1, 2009; 35(2): 214 - 228. [Abstract] [Full Text] [PDF] |
||||
![]() |
R.-M. B.G.E. Breukers, R. J. Trof, R. B.P. de Wilde, P. C.M. van den Berg, J. W.R. Twisk, J. R.C. Jansen, and J. Groeneveld Relative value of pressures and volumes in assessing fluid responsiveness after valvular and coronary artery surgery Eur. J. Cardiothorac. Surg., January 1, 2009; 35(1): 62 - 68. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A. Grossi, C. F. Schwartz, P.-J. Yu, U. P. Jorde, G. A. Crooke, J. B. Grau, G. H. Ribakove, F. G. Baumann, P. Ursumanno, A. T. Culliford, et al. High-Risk Aortic Valve Replacement: Are the Outcomes as Bad as Predicted? Ann. Thorac. Surg., January 1, 2008; 85(1): 102 - 107. [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 |