| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2008;135:1076-1080
© 2008 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease |
a Division of Cardiology, Texas Heart Institute at St Luke's Episcopal Hospital, Houston, Tex
c Division of Biostatistics, Texas Heart Institute at St Luke's Episcopal Hospital, Houston, Tex
d Division of Cardiovascular Surgery, Texas Heart Institute at St Luke's Episcopal Hospital, Houston, Tex
b Division of Cardiology, Baylor College of Medicine, Houston, Tex
e Division of Cardiovascular Surgery, Baylor College of Medicine, Houston, Tex
Received for publication September 4, 2007; revisions received November 2, 2007; accepted for publication November 13, 2007. * Address for reprints: Shun Kohsaka, MD, MC 1-133, PO Box 20345, Houston, TX 77225-0345. (Email: sk2798{at}columbia.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Prospectively collected long-term follow-up data from 469 consecutive patients who underwent aortic valve replacement between 1995 and 1998 were reviewed. The indexed effective orifice area was derived from the reference normal values of effective orifice area divided by the patient's body surface area. Outcome was stratified according to the severity of prosthesis–patient mismatch: moderate mismatch was defined as 0.65 to 0.85 cm2/m2 and severe mismatch as less than 0.65 cm2/m2. The Cox proportional-hazards model with propensity score adjustment was used to adjust for the observed differences in baseline characteristics between the mismatch groups.
Results: The degree of prosthesis–patient mismatch was minimal in 57% of patients, moderate in 39%, and severe in 4%. Predictors of clinically significant mismatch included small aortic valve sizes (19 and 21 mm), obesity, age greater than 65 years, and class III or IV heart failure. During a median follow-up period of approximately 7.9 years, overall survival was 77% in patients with minimal mismatch, 63% in those with moderate mismatch, and only 47% in those with severe mismatch (P < .001). Moderate or severe mismatch was a significant predictor of poorer survival (hazard ratio, 1.6; 95% confidence interval, 1.4–2.3; P < .01), even after adjustment for all significant clinical predictors (ie, propensity score; hazard ratio, 1.2; 95% confidence interval, 1.0–1.5; P = .05).
Conclusions: In a large aortic valve surgery population, prosthesis–patient mismatch occurred in 43% of patients, and those with significant mismatch had worse long-term outcomes than those with minimal mismatch.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The degree of PPM is defined by the effective orifice area (EOA) of the valve indexed to body surface area (BSA). Hemodynamic compromise occurs when indexed EOA is less than 75% of native EOA, leading to high postoperative transvalvular gradients and reduced regression of left ventricular hypertrophy.2 Significant mismatch has been reported in up to 60% of patients undergoing mechanical AVR.
However, the effect of PPM on survival remains controversial, possibly because the methods used to measure native EOA vary among studies. In some studies PPM has been shown to be an independent predictor of short-term survival in patients who have undergone AVR.3 Limited data suggest that the persistence of high gradients and limited regression might be an obstacle to improvement and might adversely affect both long-term survival and functional quality of life.2,4 We sought to determine whether PPM is associated with significantly reduced long-term survival in patients undergoing AVR.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We examined records of 492 adult patients (
18 years old) who underwent mechanical AVR for aortic stenosis at the Texas Heart Institute from January 1, 1993, through December 31, 1998, and agreed to participate in a telephone survey. In addition, non-Texas residents were excluded from the present study to obtain accurate long-term survival information from the state database. The decision to perform AVR was made by the treating physicians in conjunction with their patients. We excluded patients who had native or prosthetic endocarditis (defined as blood culture–positive or systemic sepsis with the clinical syndrome of infective endocarditis, culture-positive valvular vegetations, or diagnosed or suspected endocarditis that required antibiotic treatment; n = 21) or concomitant procedures other than coronary artery bypass grafting (n = 2). The remaining 469 patients were included in the present study.
Patient histories were obtained by means of interview at hospital or clinic presentation and were entered prospectively into the database. Hypertension was defined as either blood pressure greater than 130/90 mm Hg or the combination of a history of hypertension and current use of antihypertensive medications. Diabetes mellitus was defined as a fasting blood sugar level of greater than 125 mg/dL or the use of antidiabetic agents. Renal insufficiency was defined as a serum creatinine level of 2 mg/dL or greater. Obesity was defined as a body mass index of 30 or greater. Coronary artery disease was defined either as significant stenosis (>50%) in the left main artery or in all 3 major epicardial vessels. Peripheral vascular disease, cerebrovascular disease, and chronic lung disease were determined from each patient's medical history before the admission date. Previous myocardial infarction was deemed to be present if the patient had a history of myocardial infarction, coronary intervention, or coronary artery bypass surgery for myocardial infarction or if there were significant Q waves on the patient's surface electrocardiogram.
Patient size was represented by BSA, which was calculated from height and weight. The indexed EOA was derived from the reference normal values of EOA divided by the patient's BSA. The reference values for EOA were based on previously published data.3 The indexed EOAs for St Jude hemodynamic plus valves were obtained through direct communication with the manufacturer, and these numbers were found to be in concordance with previously published data.5,6 Outcome was stratified according to the severity of PPM: minimal mismatch was defined as an indexed EOA of greater than 0.85 cm2/m2, moderate mismatch as an indexed EOA of 0.65 to 0.85 cm2/m2, and severe mismatch as an indexed EOA of less than 0.65 cm2/m2.
The study end point was all-cause mortality. Survivorship was determined from the Texas State Department of Vital Statistics database, hospital records, and telephone follow-up. The Department of Vital Statistics keeps records of all births and deaths that have occurred in Texas from 1903 to the present. Consequently, follow-up information was available for all of the patients enrolled in the present study.
Most patient characteristics and event rates were recorded as binary categorical variables (ie, present or not present) and reported as percentages. The Pearson 
2 test was used to analyze discrete variables, and the Student t test was used to analyze continuous variables. Logistic regression models using a forward stepwise variable selection process were developed to determine which clinical and angiographic variables were associated with PPM.
Cumulative survival was estimated with the Kaplan–Meier method and compared between groups by using a log-rank test. Cox proportional-hazards regression models were developed to determine whether severe PPM was associated with survival after adjustment for potential confounders. The assumption of linearity was evaluated for the continuous measures by using restricted cubic spines. Adjusted analyses were performed for the study cohort overall and stratified by age and sex. Further analysis was conducted to control for selection bias related to PPM. Unconditional logistic regression was used to estimate a propensity score for each patient. This score was then incorporated into a proportional-hazards model as a continuous variable with all predictor variables significant at a 2-tailed nominal P value of less than .20 in the univariate analysis. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were computed with associated P values. All analyses were performed with SAS 6.09 software (SAS Institute, Cary, NC) for the VAX/VMS operating system.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
During this era, St Jude standard valves (St. Jude Medical, St Paul, Minn) were the preferred valve and accounted for 90.2% of our implantations (
Table 1). Nineteen- and 21-mm valves were placed in 27.5% of patients. The degree of PPM was minimal in 57%, moderate in 39%, and severe in only 4% of patients. Seventy-five percent of severe mismatch cases occurred after implantation of smaller (ie, 19 and 21 mm) mechanical aortic valves. This severe mismatch occurred in 11% of all patients who received 19- or 21-mm mechanical valves.
|
Table 2). Patient variables more commonly associated with greater degrees of mismatch included hypertension, increasing age, and higher BSA. Female sex was associated with more severe mismatch; however, this association is probably substantially confounded by the smaller prosthetic valve sizes often used in the female patients. No significant differences were found in other domains at baseline. Predictors of moderate or severe mismatch included small aortic valve sizes (19 and 21 mm), with an odds ratio (OR) of 8.3 (95% CI, 5.1–13.6; P < .01); obesity, with an OR of 3.4 (95% CI, 1.9–6.1; P < .01); and age greater than 65 years, with an OR of 1.9 (95% CI, 1.2–3.0; P < .01).
|
Figure 1). Patients with moderate or severe mismatch had significantly poorer survival than patients with minimal mismatch (P = .0076).
|
Table 3). Similarly, indexed EOA as a continuous variable was a significant predictor of long-term mortality in univariate analysis (HR, 0.3; 95% CI, 0.2–0.5; P < .01) but not in multivariate analysis (HR, 0.9; 95% CI, 0.6–1.3; P = .54). Because the relationship between the indexed EOA and the risk of mortality was not linear, we also performed a logarithmic transformation of the indexed EOA and found a similar trend.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous short- and intermediate-term survival analyses have not consistently identified PPM as an independent predictor of adverse outcomes.7,8 It is therefore agreed that severe PPM increases early mortality, whereas its effect on late results is less clear. Clinically, severe PPM appears to be associated with a higher incidence of late symptoms of heart failure and less regression of left ventricular hypertrophy, as determined by means of echocardiographic analysis.9-11 Recently, an analysis from the Mayo Clinic identified severe PPM as an independent predictor of long-term mortality in patients with small aortic valve prostheses.12 However, it is important to note that only patients with small valves (19 and 21 mm) were included in that study. Current mechanical valves provide low systolic gradients and acceptable orifice areas in patients with a small aortic root, but these devices necessitate anticoagulation, thus incurring a risk of hemorrhagic complication in older patients.13 The study also excluded all short-term deaths, which might have biased the results because short-term mortality is higher in patients with moderate or severe PPM. It is important to note that in our study patients undergoing aortic root enlargement concomitant with AVR were not excluded. In fact, throughout our study period, an attempt was made to upsize the aortic root to accommodate larger valve sizes. As a consequence, severe PPM was found in only 4% of patients.
Age greater than 65 years was an independent risk factor for long-term mortality in our patients. Because the population is aging, the incidence of degenerative aortic valve disease continues to grow, and the potential adverse effect of PPM on long-term survival in the elderly population is of increasing concern. Some studies have suggested that the effect of PPM is more important in younger patients14 and that elderly patients simply might not live long enough to manifest a survival decrement from significant PPM.15 However, we found no interaction between age and PPM. Instead, PPM predicted long-term mortality regardless of patient age.
Prosthesis size in our population was defined by the in vivo functional dimensions of the prosthesis and not by geometric dimensions. Geometric dimensions are usually obtained from the valve size reported on the label by the manufacturer, but this number variably refers to the diameter of the external sewing ring and to the diameter of the internal orifice. Pibarot and Dumesnil2 have suggested that rather than using geometric prosthesis dimensions, projected EOA, which is computed from in vivo EOA, should be used because it is a more physiologic measure.
In the current study the association between moderate-to-severe PPM and long-term survival was only modestly significant after adjustment for age and sex (P = .05) and after stepwise modeling with propensity scoring (P = .04). One possible explanation for this finding is that our study only included 469 patients, which might not have been enough to adequately examine 20 variables. Thus our results might underestimate the true independent association between substantial PPM and long-term outcomes. It is quite possible that if we had more patients, the association between moderate-to-severe PPM and long-term survival would become more significant.
Our study has other limitations. First, this was not a randomized trial; therefore surgical procedures were performed at the discretion of the individual surgeon, and patient characteristics in the 3 mismatch groups were intrinsically different. Propensity score adjustment was used to reduce treatment selection bias, allowing for a fairer comparison. Yet without treatment randomization, the possibility that there were unmeasured confounding variables cannot be excluded. The small number of patients in the severe mismatch group also limits our ability to draw firm conclusions about such patients or other types of patients with smaller valves.
Second, more than 90% of the valves used in our cohort were older St Jude Medical models, and the newer prostheses have better hemodynamic performance and are associated with a lower prevalence and severity of PPM, which might translate into better outcomes. Thus our results might not be generalizable to patients with newer prosthetic valves.
Third, although they have been validated multiple times in the past,2,16 most of our "reference values" for aortic valve prosthesis EOA were derived from the results of a single study and were therefore crucially dependent on the methods used in that study.
Finally, our models do not include the cause of death, and thus our models might be based on mortality unrelated to aortic valve disease. However, all-cause mortality is an appropriate end point to follow because it accounts for both cardiac and systemic disease and is unaffected by the reporting and misclassification bias potentially introduced by a physician's filing of a death report.17
In conclusion, in a large aortic valve surgery population, substantial PPM occurred in 43% of patients. Moderate or severe mismatch was most likely to occur in patients with larger BSA, older age, and smaller prosthesis size. The patients with substantial mismatch had significantly worse long-term outcome than the patients with minimal mismatch.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Pisano, T. D'Amico, C. Palmeri, R. Franchino, K. Fattouch, G. Bianco, and G. Ruvolo Valve prosthesis-patient mismatch: hemodynamic, echocardiographic and clinical consequences Interact CardioVasc Thorac Surg, December 1, 2011; 13(6): 606 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-A. Clavel, J. Rodes-Cabau, E. Dumont, R. Bagur, S. Bergeron, R. De Larochelliere, D. Doyle, E. Larose, J. G. Dumesnil, and P. Pibarot Validation and Characterization of Transcatheter Aortic Valve Effective Orifice Area Measured by Doppler Echocardiography J. Am. Coll. Cardiol. Img., October 1, 2011; 4(10): 1053 - 1062. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. J. Stewart and L. L. Rodriguez Avoiding Patient-Prosthesis Mismatch: Is TAVI a New Solution? J. Am. Coll. Cardiol. Img., October 1, 2011; 4(10): 1063 - 1066. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bleiziffer, A. Ali, I. M. Hettich, D. Akdere, R. P. Laubender, D. Ruzicka, J. Boehm, R. Lange, and W. Eichinger Impact of the indexed effective orifice area on mid-term cardiac-related mortality after aortic valve replacement Heart, June 1, 2010; 96(11): 865 - 871. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Flameng, M. C. Herregods, M. Vercalsteren, P. Herijgers, K. Bogaerts, and B. Meuris Prosthesis-Patient Mismatch Predicts Structural Valve Degeneration in Bioprosthetic Heart Valves Circulation, May 18, 2010; 121(19): 2123 - 2129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Jilaihawi, D. Chin, T. Spyt, M. Jeilan, M. Vasa-Nicotera, J. Bence, E. Logtens, and J. Kovac Prosthesis-patient mismatch after transcatheter aortic valve implantation with the Medtronic-Corevalve bioprosthesis Eur. Heart J., April 1, 2010; 31(7): 857 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Tanaka, K. Tajima, Y. Takami, N. Okada, S. Terazawa, A. Usui, and Y. Ueda Early and Late Outcomes of Aortic Valve Replacement in Dialysis Patients Ann. Thorac. Surg., January 1, 2010; 89(1): 65 - 70. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Urso, R. Sadaba, and G. Aldamiz-Echevarria Is patient-prosthesis mismatch an independent risk factor for early and mid-term overall mortality in adult patients undergoing aortic valve replacement? Interact CardioVasc Thorac Surg, September 1, 2009; 9(3): 510 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J G Dumesnil, J Magne, N Girerd, and P Pibarot Moderate patient-prosthesis mismatch can impact on mortality after aortic valve replacement Heart, April 1, 2009; 95(7): 592 - 592. [Full Text] [PDF]
|
|
|
|
|
|
J Mascherbauer and H Baumgartner The authors' reply: Heart, April 1, 2009; 95(7): 592 - 593. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Mohty, J. G. Dumesnil, N. Echahidi, P. Mathieu, F. Dagenais, P. Voisine, and P. Pibarot Impact of Prosthesis-Patient Mismatch on Long-Term Survival After Aortic Valve Replacement: Influence of Age, Obesity, and Left Ventricular Dysfunction J. Am. Coll. Cardiol., January 6, 2009; 53(1): 39 - 47. [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 |
