Ann Thorac Surg 2001;71:S261-S264
© 2001 The Society of Thoracic Surgeons
Valvular bioprostheses over 15 years
Late results of double-valve replacement with biologic or mechanical prostheses
Thierry Caus, MDa,
Philippe Rouvière, MDa,
Frédéric Collart, MDa,
Annick Mouly-Bandini, MDa,
Jean-Raoul Montiès, MDa,
Thierry Mesana, MD, PhDa
a Cardiac Surgery Service, CHU Timone, Marseille, France
Address reprint requests to Dr Caus, Cardiac Surgery Service, CHU Timone, Bd Jean Moulin, 13385 Marseille, France
e-mail: tcaus{at}ap-hm.fr
Presented at the VIII International Symposium on Cardiac Bioprostheses, Cancun, Mexico, Nov 3–5, 2000.
 |
Abstract
|
|---|
Background. We previously showed that the risk of reoperation for structural degeneration of bioprostheses was higher in cases involving patients older than 65 years (p = 0.003) and double-valve replacement (p = 0.02). The purpose of this study was to compare late outcome of mitral-aortic valve replacement using bioprostheses or mechanical valves.
Methods. The bioprosthesis group included all mainland France residents (n = 48) between 55 and 65 years old operated on between 1980 and 1995 for mitral-aortic valve replacement using bioprostheses. The mechanical valve group was obtained by matching each of these patients with a patient operated on using mechanical valves at approximately the same time during the study.
Results. In the bioprosthesis group, 10-year survival was 45% ± 8% versus 62% ± 7% in the mechanical valve group (not significant). The linearized reoperation rate was 6.8 per patient-year versus 1.1 per patient-year (p = 0.001), and the linearized reoperative mortality rate was 1.8 per patient-year and 0.7 per patient-year (not significant), respectively.
Conclusions. The reoperative mortality risk after mitral-aortic valve replacement using two bioprostheses does not significantly decrease overall survival after age 65 years.
|
Introduction
|
|---|
Any patient treated with a bioprosthesis may need reoperation for valve replacement if his or her life expectancy exceeds that of the valve. In a previous study, we reported that the risk of reoperative mortality was significantly higher in cases involving patients older than 65 years of age who had double-valve replacement (VR) [1]. On the basis of this finding, implantation of two bioprostheses would seem contraindicated in patients in whom structural degeneration requiring reoperation was likely to occur after age 65 years. However, alternative use of mechanical valves is associated with a significantly higher risk of potentially lethal hemorrhagic complications [2–4]. In an effort to resolve this dilemma, it appeared useful to compare late outcome of double VR using bioprostheses or mechanical valves. The purpose of this study was to determine whether the risk of reoperative mortality for structural degeneration of bioprostheses was greater than the cumulative rate of lethal hemorrhagic complications in patients with mechanical valves by comparing late results in two matched patient populations who underwent combined mitral-aortic VR using either two bioprostheses (TBP) or two mechanical valves (TMV). To heighten the sensitivity of this study, we included only patients at risk of requiring reoperation for structural degeneration after age 65 years.
|
Material and methods
|
|---|
To ensure a sufficient duration of follow-up, the TBP group included only mainland France residents operated on between the ages of 55 and 65 years for mitral-aortic VR using TBP between 1980 and 1995. Each patient in the TBP group was matched with a patient who underwent mitral-aortic VR using TMV at approximately the same time during the study. If several patients were suitable matches, inclusion was determined by drawing. The goal of the matching process was to obtain groups that were strictly comparable with regard to age, sex, preoperative clinical status, and duration of follow-up. To minimize the impact of associated coronary artery disease on late mortality, only patients with an isolated valvular disease at the time of the operation were included.
Data concerning the first operation were gathered by retrospectively reviewing computerized patient files stored in our department. Data concerning reoperations were gathered from the same files if reoperation took place in our department. Other data on survival, reoperations if performed elsewhere, and valve-related complications according to criteria defined by Edmunds and associates [5] were collected by telephone survey. Calls were made between October and December 1999 to the patient, family, referring physician, or attending cardiologist by an experienced investigator. Reoperated patients were maintained in their original study group until the end of the study on December 1999.
Statistical analysis was performed on a personal computer using SPSS 7.5 software for Windows (SPSS Inc, Chicago, IL). To determine the significance of differences between the two study groups, qualitative data were compared using the Pearson 
2 test and quantitative data, using the Student’s t test. Survival curves and freedom-from-complication curves were established for each group using the actuarial method and compared using the log-rank test.
|
Results
|
|---|
Preoperative clinical status of patients in the TBP (n = 48) and TMV (n = 48) groups is summarized in Table 1. There was no significant difference with regard to mean age, sex ratio, underlying cause, New York Heart Association status, cardiothoracic ratio, left ventricular ejection fraction, and incidence of pulmonary hypertension. However, significantly more patients in the TMV group had atrial fibrillation (AF) (p = 0.005), and more patients in the TBP group had a previous history of stroke (p = 0.017).
Concerning operative data, there were no significant differences in duration of cardiopulmonary bypass or the number of concomitant tricuspid procedures, five patients (10.4%) in the TBP versus eight patients (16.6%) in the TMV group. Likely owing to a longer preparation time before implantation of bioprostheses, mean duration of aortic cross-clamping was 97.6 ± 18 minutes in the TBP group versus 89.6 ± 21 minutes in the TMV group (p = 0.021).
Operative mortality in the first procedure was null because only surviving patients were included to study long-term outcome. Follow-up (mean duration 9.6 and 10.5 years in the TBP and TMV groups, respectively) was complete, and total cumulative follow-up was 858.4 patient-years. As shown in Figure 1, survival in the TBP group was 72.5% ± 6.1% at 5 years, 43.6% ± 8.3% at 10 years, and 29.7% ± 8.5% at 15 years versus 78.9% ± 6.2%, 62.3% ± 7.3%, and 35.6% ± 9.3%, respectively, in the TMV group (not significant [NS]). At the end of follow-up, 31 patients (64.6%) from the TBP versus 26 patients (54.2%) in the TMV group were dead (NS). Causes of late deaths were as follows: noncardiac mortality for 5 patients (16.6%) in the TBP versus 7 patients (27%) in the TMV group (NS), and respectively, cardiac non–valve-related mortality for 7 patients (22.6%) versus 4 patients (15.4%; NS) and valve-related mortality for 18 patients (58%) versus 11 patients (42.3%; NS).
View larger version (14K):
[in this window]
[in a new window]
|
Fig 1. Overall actuarial survival in the bioprostheses (TBP) and mechanical valve (TMV) groups. (Patients were not dropped out of the study after reoperation; not significant by log-rank test.)
|
|
Mean age at the time of the first reoperation was 68.3 years in the TBP group and 66.4 years in the TMV group (NS). The mortality rate of the first reoperation was 20% (4 of 20 patients) in the TBP group as compared with 37.5% (3 of 8 patients) in the TMV group (NS). As shown in Figure 2, the reoperative mortality rate was 5% at 5 years and 15% at 10 and 15 years in the TBP group versus 2% at 5 years, 6% at 10 years, and 12% at 15 years in the TMV group (NS). Taking into account all fatal valve-related complications, including thromboembolic events and hemorrhagic complications, survival free from fatal valve-related complications was 89% ± 9% at 5 years, 81% ± 11% at 10 years, and 64% ± 24% at 15 years for the TBP group and 91% ± 9%, 80% ± 13%, and 74% ± 16%, respectively, for the TMV group (NS).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 2. Actuarial survival curves showing freedom from fatal reoperation in the bioprostheses (TBP) and mechanical valve (TMV) groups. (Not significant by log-rank test.)
|
|
Reoperation-free survival was 83.1% ± 10.2% at 5 years, 34.4% ± 15.2% at 10 years, and 28.1% ± 18.1% at 15 years for the TBP group versus 88% ± 8%, 80.6% ± 12.2%, and 76% ± 14.6%, respectively, for the TMV group (p < 0.001). Figure 3 shows nonfatal reoperation-free survival. The main indication for reoperation in the TBP group was structural degeneration (18 of 24 patients), whereas all reoperations in the TMV group were undertaken for nonstructural problems, including thrombosis, endocarditis, or periprosthetic leakage. The cumulative incidence of thromboembolic and hemorrhagic events was 8% at 5 years, 12.5% at 10 years, and 66.6% at 15 years in the TBP group versus 8% at 5 years, 23.6% at 10 years, and 70.6% at 15 years in the TMV group (NS). Overall freedom from reoperation and thromboembolic or hemorrhagic events was 84% ± 11% at 5 years, 34% ± 16% at 10 years, and 20% ± 16% at 15 years in the TBP group versus 87% ± 10%, 68% ± 16%, and 45% ± 23%, respectively, in the TMV group (p = 0.01).
View larger version (13K):
[in this window]
[in a new window]
|
Fig 3. Actuarial survival curves showing freedom from nonfatal reoperation in the bioprostheses (TBP) and mechanical valve (TMV) groups. (p < 0.001 by log-rank test.).
|
|
Concerning quality of life of patients surviving at the end of study, there was no difference between the two groups in terms of New York Heart Association functional status. The incidence of reoperation was 9 of 17 patients (52.9%) in the TBP group versus 2 of 22 patients (9%) in the TMV group (p = 0.01), but the incidence of valve complications not involving deterioration or reoperation was respectively 1 of 17 patients (5.9%) versus 6 of 22 patients (27.3%; NS). The overall incidence of chronic AF was 9 of 17 patients (52.9%) in the TBP group and 13 of 22 patients (59.5%) in the TMV group (NS). As a result, 50% of patients in the TBP group exhibited need for long-term anti-vitamin K therapy during follow-up.
|
Comment
|
|---|
There is a paucity of studies comparing bioprostheses and mechanical valves for double VR [3, 4, 6]. The purpose of this study was to obtain accurate data on long-term survival in patients undergoing combined mitral-aortic VR using either bioprostheses or mechanical valves. To heighten the sensitivity of this study, only patients likely to require reoperation after age 65 years were included as age has been shown to be a major risk factor for reoperation [1, 7]. The need for sufficient minimum follow-up and zero dropouts led us to select only mainland France residents operated on before 1995.
Unlike previously published series [3, 4, 6], patients in this study were carefully matched with regard to age, preoperative clinical status, surgical procedures, and duration of follow-up. Because mitral-aortic VR was performed more often using mechanical valves than bioprostheses in our experience, all patients treated with bioprostheses were included, and the TMV group was achieved by matching each patient in the TBP group with a patient treated using mechanical valves. If more than one patient was a suitable match, inclusion was determined by drawing so as to obtain the same number of patients in the two groups. Comparison of both groups revealed significantly more patients with AF in the TMV group and significantly more patients with a history of stroke in the TBP group. These discrepancies are understandable insofar as chronic AF is a factor in favor of mechanical valve replacement because of the need for anticoagulation therapy and history of stroke is a factor in favor of bioprosthetic valve replacement to reduce the risk of secondary brain damage caused by anticoagulation therapy.
Another feature of this study is expression of overall survival in function of the initial goal of treatment. This was performed to highlight the consequences of the initial valve choice on the survival of the patient. Further justifying this approach is the fact that the survival period of the study was much longer than the lifespan of the bioprostheses. Therefore, follow-up of patients treated with bioprostheses was not artificially shortened in this study. Hence, we found that overall actuarial survival was slightly shorter for patients undergoing mitral-aortic VR using two bioprostheses (NS). This difference cannot be explained either by shorter life expectancy at the time of inclusion in the TBP group—although this was a retrospective study—or by an increase in the number of deaths from valve-related complications in the TBP group. However, closer analysis of causes of late death disclosed a higher incidence of deaths owing to heart-related causes despite good valve function in the TBP group (22.6% versus 15.4%). Because we excluded all patients with associated coronary disease, this result could be caused by myocardial alterations induced by reoperations, which were more common in the TBP group.
Multivariate analysis has demonstrated that the risk of reoperative mortality after structural degeneration of bioprosthesis is threefold higher in patients older than 65 years [1]. Notwithstanding, the current data showed no significant difference in reoperative mortality despite a 5 times higher incidence of reoperations in the TBP group than the TMV group. Indeed even though the incidence of reoperations was significantly lower in the TMV group, reoperative mortality was higher. On the other hand, the number of fatal hemorrhagic or thromboembolic complications was comparable in the two groups because the incidence of chronic AF was the same regardless of the type of prosthesis used. At the end of this study, 53% of patients in the TBP group were receiving anti-vitamin K treatment.
Although overall late survival was similar between the TBP and the TMV groups, the actuarial rate of nonfatal valve morbidity differed significantly between them. The incidence of reoperation was 4 times higher in the TBP group, and the incidence of nonfatal hemorrhagic (p < 0.001) and thromboembolic events was 9 times higher in the TMV group (p < 0.05). The same difference in morbidity has been described in previous studies [3, 4, 6]. At the end of the present study, the incidence of reoperation was higher in the TBP than TMV group, but the incidence of valve complications not related to structural degeneration was lower.
Based on our findings, we conclude that overall long-term survival after combined mitral-aortic VR in patients between the ages of 55 and 65 years is independent of the type of prosthesis used even though the risk of reoperation is higher in patients treated with bioprostheses. The overall risk of valve complications is the same regardless of the type of prosthesis used. However patients treated with two bioprostheses who experience chronic AF are exposed to a double long-term mortality risk from reoperation and hemorrhagic or thromboembolic complications.
|
Acknowledgments
|
|---|
We thank Dr Roch Giorgi for his contribution to the statistical analysis of the data. This work received a Young Researcher Harvard 2000 from the Société de Chirurgie Thoracique et Cardiovasculaire de Langue Française as well as the Clinical Research Prize 2000 from the Assistance Publique de Hôpitaux de Marseille.
|
References
|
|---|
-
Caus T., Albertini J.N., Chi Y., Collart F., Monties J.R., Mesana T. Multiple valve replacement increases the risk of reoperation for structural degeneration of bioprostheses. J Heart Valve Dis 1999;8:376-383.[Medline]
-
Schaff H.V., Marsh D.H. Multiple valve disease. In: Edmunds L.H., Jr, ed. Cardiac surgery in the adult. New York: McGraw-Hill, 1997:1071-1099.
-
Munro A.I., Jamieson W.R.E., Burr L.H., Ling H., Miyagishima R.T., Germann E. Comparison of porcine bioprostheses and mechanical prostheses in multiple valve replacement operations. Ann Thorac Surg 1995;60:S459-S463.
-
Brown P.S., Jr, Roberts C.S., McIntosh C.L., Swain J.A., Clark R.E. Relation between choice of prostheses and late outcome in double-valve replacement. Ann Thorac Surg 1993;55:631-640.[Abstract/Free Full Text]
-
Edmunds L.H., Jr, Clark R.E., Cohn L.H., Grunkemeier G.L., Miller D.C., Weisel R.D. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1996;62:932-935.[Abstract/Free Full Text]
-
Bortolotti U., Milano A., Testolin L., Tursi V., Mazzucco A., Gallucci V. Influence of type of prosthesis on late results after combined mitral-aortic valve replacement. Ann Thorac Surg 1991;52:84-91.[Abstract/Free Full Text]
-
Piehler J.M., Blackstone E.H., Bailey K.R., et al. Reoperation on prosthetic heart valve: patient-specific estimates of early events. J Thorac Cardiovasc Surg 1995;109:30-48.[Abstract/Free Full Text]
Top
Abstract
Introduction
