|
|
||||||||
a Department of Cardiothoracic Surgery, University Hospital Birmingham NHS Foundation Trust, Birmingham, United Kingdom
c Quality and Outcomes Research Unit (QuORU), University Hospital Birmingham NHS Foundation Trust, Birmingham, United Kingdom
b The Heart Hospital, London, United Kingdom
Accepted for publication July 10, 2009.
* Address correspondence to Dr Pagano, Consultant and Reader in Cardiothoracic Surgery, Clinical Director of QuORU, University Hospital Birmingham NHS FT, Edgbaston, Birmingham, B15 2TH, United Kingdom (Email: domenico.pagano{at}uhb.nhs.uk).
Presented at the Poster Session of the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
| Abstract |
|---|
|
|
|---|
Methods: Prospectively collected data on patients undergoing isolated AVR for aortic stenosis between January 1, 1997 and December 31, 2007 were analyzed. The projected effective valve orifice area from in vivo data was indexed to body surface area (EOAi). PPM was defined as moderate for EOAi of
0.85 cm2/m2 and severe if
0.6 cm2/m2. The reference group comprised patients with EOAi > 0.85 cm2/m2. The effect of PPM on postoperative survival was assessed by multivariate analysis.
Results: Of 801 patients, PPM was severe in 48 (6.0%), moderate in 462 (57.8%), and nonexistent in 291 (36.4%). Mismatch was associated with increasing age and female gender, thus resulting in an increase in the EuroSCORE (reference group, 4.9 ± 2.6; moderate PPM, 5.8 ± 2.4; and severe PPM, 6.1±2.1; p < 0.001). PPM did not significantly increase hospital mortality. Four deaths occurred in the reference group (1.4%), 12 in the moderate PPM (2.6%), and none in the severe PPM group (p = 0.311). The 5-year survival estimates were 83% in reference, 86% in moderate PPM, and 89% in severe PPM (p = 0.25). By multivariate analysis, PPM was not an independent risk factor for reduced in-hospital or late survival.
Conclusions: Moderate PPM is common in patients undergoing AVR for aortic stenosis, but severe mismatch is rare. Patients with PPM have similar early and late postoperative survival rate.
| Introduction |
|---|
|
|
|---|
The clinical significance of mismatch after AVR remains both controversial and important. Numerous studies have showed that PPM is a strong and independent risk factor for cardiac events and midterm death [3–8]. It has therefore been suggested that the predicted EOA indexed (EOAi) to body surface area (BSA) should be calculated preoperatively to determine the appropriate size of the prosthesis to be implanted [6, 9]. In aortic stenosis, especially in elderly and female patients, this may mean that an aortic root enlargement is required or a valve with a greater EOA should be chosen, such as a stentless valve.
These are technically more challenging procedures that lengthen the cardiopulmonary bypass and cross-clamp times, and although recent studies demonstrate they can be performed with similar operative risks [10], no late survival advantage has been demonstrated [11, 12]. In contrast, other studies have demonstrated no increased risk with mismatch [13–16]. More recently, it has been suggested that PPM is only important in specific patient groups, depending on the patient's age, size (body mass index), and ventricular function [17–19].
Consensus is growing that the predicted EOAi is the most appropriate measure [4, 9, 16, 20]. This variable has been repeatedly shown to correlate well with postoperative gradients, and some have suggested that it is highly predictive of adverse outcomes [3, 20]. Assessment of the effect of mismatch on survival has often been difficult because the patient cohorts analyzed have been very heterogeneous, including patients with both stenotic and regurgitant lesions and patients undergoing concomitant surgical procedures, including coronary artery bypass grafting. The aim of this study was to examine the frequency of PPM in patients with pure aortic stenosis undergoing isolated, first-time AVR and to examine the effect on late survival.
| Material and Methods |
|---|
|
|
|---|
Patient Population
The study included all patients undergoing first-time isolated AVR for severe aortic stenosis at our institution between January 1, 1997, and December 31, 2007. Some of the patients in this study had already been included in our previous analysis [13]. Severe aortic stenosis was defined as aortic valve area of less than 1 cm2 or a peak gradient exceeding 50 mm Hg. Patients with aortic incompetence or with more than mild aortic regurgitation, as defined by the American Society of Echocardiographers [21], on preoperative transthoracic echocardiography were excluded, as were patients with significant coronary artery disease who had undergone percutaneous coronary interventions or required concomitant coronary artery bypass grafting. Also excluded were patients undergoing mitral valve repair and ascending aorta replacement for post-stenotic dilatation.
Data Collection and Study End Points
Prospectively collected data on 801 patients undergoing AVR at our institution were collated and analyzed. The predicted EOAi was derived from the published normal in vivo EOA values for each model and size of prosthesis implanted [13]. In this study, we defined moderate mismatch as an EOAi of less than 0.85 cm2/m2, and severe mismatch as an EOAi of less than 0.6 cm2/m2, as described and validated in recent studies [3, 4].
In-hospital mortality was tracked from our database, and postdischarge survival data were obtained from the National Central Cardiac Audit Database, which is linked to the Office of National Statistics (census date Jan 6, 2008). In-hospital mortality was defined as death within 30 days from operation or at any time within the same hospital admission.
Subgroup analyses were also performed for those patients in whom mismatch has specifically been demonstrated to have an adverse affect on survival, specifically those aged younger than 60 years at the time of operation, those with a BSA exceeding 2.1 m2, and patients with impaired systolic function, defined as left ventricular ejection fraction (LVEF < 0.50), as assessed by Simpson's method on preoperative transthoracic echocardiography.
Statistical Analysis
Descriptive data are expressed as mean ± 1 standard deviation. The level of statistical significance (
) was set at 0.05 (two-sided). The additive European System for Cardiac Operative Risk Evaluation (EuroSCORE), the most commonly used tool to assess the risk profile of patients undergoing cardiac operations in the United Kingdom, contains variables known to influence outcomes, including age, gender, and ventricular function. The additive EuroSCORE was calculated for each patient and incorporated into the statistical modeling. In-hospital survival was analyzed as previously noted using nonlinear modeling, including the EuroSCORE as a patient-level covariate and the surgeon as a random affect. Late survival was analyzed with a Cox regression model, again including EuroSCORE and surgeon. We also analyzed the addition of mismatch, as previously defined, to these models [22]. All analysis was performed using SAS 9.1 software (SAS Institute, Cary NC).
| Results |
|---|
|
|
|---|
|
|
The 5-year survival estimates were similar: 83% in reference, 86% in moderate PPM, and 89% in severe PPM (p = 0.25; Fig 1). The EuroSCORE was again predictive of late survival (hazard ratio [HR], 4.32; 95% CI 3.9 to 4.3; p < 0.0001), but the addition of moderate PPM (HR, 1.12; 95% CI, 0.34 to 4.67; p = 0.4) or severe PPM (HR, 0.94; 95% CI, 0.11 to 9.12; p = 0.92) was not.
|
Systolic function was impaired (LVEF < 0.50) in 191 patients, of whom 16 (8.4%) had severe mismatch and 89 (46.6%) had moderate mismatch. In the reference group, there were 4 in-hospital deaths (4.7%), in the moderate PPM group there were 2 in-hospital deaths (2.2%), and in the severe PPM group there were no in-hospital deaths (p = 0.497). The 5-year survival estimates were again similar: 82% in reference, 85% in moderate PPM, and 86% in severe PPM (p = 0.601).
| Comment |
|---|
|
|
|---|
A large number of studies have examined the effect of PPM on survival after AVR [3, 5, 7, 13–16, 23, 24]. Initial reports suggested a step-wise increase in in-hospital and late mortality rates associated with mismatch. These studies also noted a similar association with increasing age and female gender, as in this study, and acknowledged that the decrease in late survival may have been due to this increased risk profile [2, 3]. Further analyses taking into account the different risk profiles found no adverse effect on survival associated with both moderate PPM and severe PPM for the cohort as whole [13–15], but subgroup analyses from these more recent studies have suggested that younger patients or patients with impaired ventricular function may still be at risk of reduced survival with severe PPM [14–19, 25]. We have looked at the effect of severe mismatch in younger patients and those with impaired ventricular function. Again, we found no increased incidence of late events in these groups, but the numbers of patients analyzed in these groups were small.
A further confounding factor is the heterogeneous populations previously studied. In addition to patients with pure aortic stenosis, most studies included patients with aortic regurgitation and mixed valve disease. Patients with aortic regurgitation tend to have a larger aortic annulus, which allows a larger valve to be placed and increases the postoperative EOA. These patients have dilated ventricles with eccentric hypertrophy, and studies have shown that the reverse remodeling that occurs postoperatively is different than that of patients with aortic stenosis [26] and that these differences may affect late outcome. Other studies, including our previous work, have also included patients with significant coronary artery disease requiring concomitant bypass grafting, a group again known to have a reduced late survival [26]. In this study, we therefore focused on patients with isolated aortic stenosis to avoid these potentially negative confounding effects on survival.
The median follow-up for the patients in this study was 52 months. A late adverse effect on survival has been hypothesized, but the reverse remodeling seen on serial postoperative echocardiography after AVR appears to be complete by 2 years in all but patients with the most profound of residual gradients [12]. The advantage of a higher EOA may be the rate of LV mass regression [27], and therefore, we might expect to see a survival difference during this time frame.
Although the in vivo data have been shown to consistently predict postoperative gradients [2, 3, 9, 20], there is a growing appreciation that in some patients, severe concentric hypertrophy leads to a small fixed stroke volume with preserved systolic function but severe diastolic impairment [14, 26]. The low stroke volumes in these patients may not generate significant transvalvular gradients, and therefore, the low predicted EOAi may be still be adequate to facilitate adequate reverse remodeling after AVR. Although the importance of the effects of LV hypertrophy on survival are well documented [28, 29], there is renewed appreciation that the degree of hypertrophy and rate of its regression is influenced by a number of genetic and environmental factors, including angiotensin-converting enzyme gene polymorphism, insulin-like growth factor-1, and hypertension [30–32].
This study has some limitations. The follow-up data obtained from the UK Central Cardiac Audit Database provides knowledge only of survival status, with no cause of death, and this does not allow differentiation between cardiovascular and other causes of death. Although we have no postoperative echocardiographic data, the EOAi, as discussed, has been shown to correlate well with late hemodynamics. Mismatch may not lead to an increase rate of late mortality, but it has been hypothesized to attenuate the symptomatic benefit that patients derive from the operation [33]. We were not able to address this question because our study did not include data on postoperative symptom relief, exercise capacity, or residual heart failure.
In conclusion, this study analyzing the incidence of patient-prosthesis mismatch after aortic valve replacement for aortic stenosis suggests that although moderate mismatch is common, severe mismatch is rare. Severe mismatch is associated with increasing age, female gender, and an increasing operative risk profile as calculated by the EuroSCORE. Despite this, we found no increase in early or late death, irrespective of age, patient size, or LF function. Aggressive over-sizing or root enlargement strategies may therefore be unwarranted. Further studies are indicated specifically to examine the effect of mismatch on symptomatic improvement and exercise tolerance after aortic valve replacement.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Lorusso, S. Gelsomino, F. Luca, G. De Cicco, G. Bille, R. Carella, E. Villa, G. Troise, M. Vigano, C. Banfi, et al. Type 2 Diabetes Mellitus Is Associated With Faster Degeneration of Bioprosthetic Valve: Results From a Propensity Score-Matched Italian Multicenter Study Circulation, January 31, 2012; 125(4): 604 - 614. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Sato, E. Suenaga, S. Koga, and H. Kawasaki Aortic valve replacement with smaller valve size Asian Cardiovasc Thorac Ann, October 1, 2011; 19(5): 314 - 319. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Cotoni, R. T. Palac, L. J. Dacey, and D. J. O'Rourke Defining Patient-Prosthesis Mismatch and Its Effect on Survival in Patients With Impaired Ejection Fraction Ann. Thorac. Surg., March 1, 2011; 91(3): 692 - 699. [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 |