| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1997;113:134-148
© 1997 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Received for publication Jan. 24, 1996 revisions requested March 18, 1996; revisions received July 9, 1996; Accepted for publication July 22, 1996 Address for reprints: J-J. Goy, MD, Division of Cardiology, CHUV, CH-1011 Lausanne, Switzerland.
Abstract
Between 1979 and 1984, 321 patients received 354 St. Jude Medical prostheses (194 aortic, 94 mitral, 1 tricuspid, and 32 multiple valve replacements). Follow-up was 96% complete (2967 patient-years; mean 9.5 years per patient). Actuarial event-free rates at 10 years and linearized rates (in parentheses) of late complications were as follows: embolism, 85.0% ± 2.3% (2.3% per patient-year); anticoagulant-related hemorrhage, 74.8% ± 2.7% (3.3% per patient-year); cerebrovascular accident, 81.8% ± 2.5% (2.6% per patient-year); prosthesis thrombosis, 98.5% ± 0.7% (0.1% per patient-year); endocarditis, 97.2% ± 1.1% (0.4% per patient-year); prosthesis dysfunction, 97.1% ± 1.0% (0.4% per patient-year); hemolytic anemia, 98.5% ± 0.7% (0.1% per patient-year); reoperation, 97.4% ± 1.0% (0.4% per patient-year); overall mortality, 63.3% ± 2.7% (4.2% per patient-year); and valve-related death (including sudden death), 84.7% ± 2.2% (1.4% per patient-year). Independent preoperative risk factors were as follows: (1) for embolism, cardiac failure as indication for operation and history of prior systemic embolism; (2) for cerebrovascular accidents, the same two factors and age; (3) for endocarditis, diabetes, chronic alcoholism, and aortic valve replacement; (4) for overall mortality, age, ejection fraction (or cardiac index or cardiothoracic index), chronic alcoholism, and history of systemic embolism; and (5) for valve-related death, chronic alcoholism, degenerative cause of valve disease, and prosthetic diameter 23 mm or smaller. Ninety percent of survivors were in New York Heart Association functional class I or II at the end of follow-up. In conclusion, this study confirms the excellent durability of the St. Jude Medical valve and the remarkable functional benefit for the majority of the patients. However, prosthesis-related complications are still common, particularly for small-diameter prostheses. Outcome is strongly related to the patient's preoperative cardiac condition and to the adequacy of anticoagulation control
The St. Jude Medical valve is a bileaflet, low-profile, central flow design prosthesis known for its excellent hemodynamic properties. Follow-up studies have also demonstrated its mechanical durability,
1-18 and only anecdotal structural failures have been reported. 2,19,20 Nevertheless, these studies have also shown that, despite a thromboresistant pyrolytic carbon coating, anticoagulant therapy with its related risks is still required. In the present study, the incidence and preoperative determinants of every late complication are analyzed in patients followed up at least 10 years after the implantation of a St. Jude Medical prosthesis. Analysis of early events is also briefly presented.
Patients and methods
Patients.
Between July 1979 and December 1984, 519 patients received at least one St. Jude Medical valve in our hospital. To get the most complete follow-up we only selected for study the patients living in Switzerland at the time of operation. Thus 321 patients were included who received 354 St. Jude Medical valves. Clinical data are shown in Table I. Previous valve operation had been done in 34 patients (11%) and coronary artery bypass grafting in 2 patients (0.6%). The characteristics of the pathologic valves are shown in Table II. Hemodynamic and angiographic preoperative data for aortic or mitral valvulopathy are shown in Table III. The first indication for operation was cardiac failure in 226 cases (70.4%), chest pain in 16 (5%), syncope in 20 (6.2%), prosthesis dysfunction in 8 (2.5%), recurrent emboli in 1 (0.3%), and combination of two symptoms or more in 50 (15.6%).
|
|
|
Table IV.
|
Follow-up.
One-time follow-up data were obtained by questionnaires addressed to the physicians in charge of the patients (response rate 93%) and by telephone interview with all living patients or the patient's family in case of death; further information about complications was obtained from hospital reports. Death certificates were reviewed for each case of sudden death or death from unknown cause. Fourteen patients were lost to follow-up; follow-up was thus 96% complete and represented 2967 patient-years with a mean of 9.5 years per patient, a median of 10.8 years, and a maximum of 15.6 years.
Definitions.
We used the guidelines of The Society of Thoracic Surgeons and The American Association for Thoracic Surgery 21 for the definitions of morbid events and mortality and the recommendations of Bodnar and colleagues 22 to report thrombosis, embolism, and bleeding. We present the incidence of emboli according to the criteria of McGoon. 23 Some review of the precise terms used is helpful: hemorrhages referred to any episodes that necessitated hospital care (inpatient or outpatient care) for diagnostic or therapeutic interventions and to episodes that led to death. Ischemic cerebrovascular accidents were classified as follows: (1) transient ischemic attack: focal neurologic signs of vascular origin lasting less than 24 hours; (2) reversible neurologic deficit: neurologic deficit lasting more than 24 hours but resolving before hospital discharge; (3) major stroke: neurologic defects still present at hospital discharge; and (4) fatal stroke. Valve-related hemolytic anemia was defined as the presence of normocytic normochromic anemia (hemoglobin value <13 gm/100 ml for men, <12 gm/100 ml for women) with a lactate dehydrogenase elevation and/or a haptoglobin decrease and/or schistocytes, in the absence of any other known cause. Sudden death was defined as a nonaccidental, witnessed or unwitnessed death of a person for whom time and mode of death were unexpected. Anticoagulation at the time of events was considered to be efficient when the PT was between 15% and 30% and inefficient when the PT was greater than 30%; overanticoagulation was defined by a PT less than 15%. INR values were usually not used at the time under study.
Statistical analysis.
Fisher's exact test and 
2 test were done to detect the univariate risk factors for early death. The different late valve-related events were expressed in linearized form (percent per patient-year) for which all events were considered. Actuarial survival and event-free curves were calculated with the Kaplan-Meier method. To establish the composite event-free and survival curves, valve-related deaths were given precedence over nonfatal complications when both occurred. Univariate comparisons of subgroups were made with the log-rank test. The list of the screened variables is given in the appendix. Cox's proportional-hazards models were then used to identify independent prognostic factors. We introduced into the models variables having a value of p < 0.1 by log-rank test. When some covariates were highly correlated, we included the most significant one. Results of the multivariate analysis are given with estimates of the ß coefficient and as odds ratio (95% confidence interval), with the significance level p value for each variable.
Results
In-hospital (<30 days) mortality and complications.
In total, there were 54 in-hospital events in 50 patients (15.6% of patients): 10 deaths (3.1% of all patients; 2.6% with aortic valve replacement [AVR], 4.2% with mitral valve replacement [MVR], 3.1% with multiple valve replacement [multiple VR]) and 44 nonfatal events. Fifteen patients (4.7%) died or had long-term impairment after the operative period. Causes of in-hospital deaths and nonfatal events are shown in Table V. The following risk factors for early death were found by univariate analysis: emergency operation (p < 0.001), New York Heart Association (NYHA) class IV symptoms (p = 0.002), additional procedures without coronary artery bypass grafting (p = 0.007), additional procedures including coronary artery bypass grafting (p = 0.016), concomitant tricuspid anomaly (p = 0.025), acute endocarditis (p = 0.041), and cardiac index less than 2.5 L/m2 (p = 0.044).
|
(Table VI). Atrial fibrillation was not a significant risk factor in our series.
|
|
10%, 20; PT 11% to 14%, 10). Anticoagulation was in the target range in 18 and insufficient in 8. Actuarial freedom from hemorrhage at 5, 10, and 13 years is shown in Fig. 2. Freedom from hemorrhage tended to be smaller for the AVR group after 96 months, although the difference was not significant for the complete duration of follow-up (p = 0.365). Univariate analysis did not show any risk factor for hemorrhages among those listed in the appendix.
|
Intracranial hemorrhages were less frequent with 18 events in 14 patients; however, their prognosis was more severe with 9 deaths (mortality, 50%) because of intracerebral bleeding. Permanent impairment occurred in 1 case. Seven of the 18 events had been associated with recent head trauma. Diagnosis was made by computed tomographic scan in 14 cases, by lumbar puncture in 1, and clinically in 3. Overanticoagulation was documented in 5 of 10 events. Independent risk factors for all cerebrovascular accidents were history of systemic emboli, age, and cardiac failure 
(Table VI). When univariate analysis was done on the group with intracranial hemorrhages only, no risk factor could be identified. Actuarial freedom from cerebrovascular accident and intracranial hemorrhage and for the following five valve-related complications individually and all valve-related events is described in Table VII.
|
Endocarditis.
Ten patients had 11 events of endocarditis (recurrence rate, 10%); 10 events occurred in the AVR group and 1 in the MVR group (linearized rate: overall, 0.4%/pt-yr; AVR, 0.6%/pt-yr; MVR 0.1%/pt-yr). Two cases were fatal (mortality rate, 18%). No survivors needed reoperation. Independent risk factors for endocarditis were diabetes mellitus, alcoholism, and AVR (Table VI).
Prosthesis dysfunction.
No case of structural dysfunction was observed. Paravalvular leak was diagnosed 11 times in 9 patients and led to reoperation 7 times (recurrence rate, 11%). One case of aortic subvalvular stenosis because of fibrosis and muscular hypertrophy and one case of incomplete closure of an aortic prosthesis because of pannus entrapment also necessitated reoperation. Eleven aortic prostheses and two mitral prostheses were involved (linearized rates: for all dysfunctions, 0.4%/pt-yr; AVR, 0.6%/pt-yr; MVR 0.2%/pt-yr). The only risk factor identified univariately was age 45 years or younger (p = 0.030).
Hemolytic anemia.
Four cases of hemolytic anemia (linearized rate, 0.1%/pt-yr) were diagnosed. Three of them were a result of moderate or severe paravalvular leak in the aortic position and the condition normalized after reoperation. In the fourth case a significant leak was not found; the anemia was managed by medical treatment.
Reoperation.
Eleven reoperations were done in nine cases (linearized rate: overall, 0.4%/pt-yr, AVR, 0.5%/pt-yr; MVR, 0.2%/pt-yr). Paravalvular leaks necessitated prosthesis replacement in three cases and simple closure of the leak in four. Two cases of valve dysfunction and two of valve thrombosis necessitated valve replacement. No significant risk factor could be identified.
Overall valve-related complications.
Including all fatal and nonfatal emboli, anticoagulation-related hemorrhages, prosthetic dysfunctions, hemolytic anemias, endocarditis, valve thrombosis, and reoperations, valve-related complications occurred at a linearized rate of 7.0%/pt-yr for the whole group (AVR, 7.6%; MVR, 6.9%; multiple VR, 4.4%). The overall mortality rate for these complications was 9%. Actuarial estimates are given in Table VII. There was no statistical difference between operative groups when the entire follow-up was considered (p = 0.502), but beyond 8 years we observed a trend toward more complications in the AVR group.
Late deaths, overall deaths, and survival.
There were 125 late deaths during the follow-up at a linearized rate of 4.2%/pt-yr (AVR, 4.7%; MVR, 3.1%; multiple VR, 4.1%). The causes of late death are reported in Table VIII. For univariate and multivariate analysis we considered early and late deaths together (overall mortality), or a total of 135 deaths. Actuarial survival curves for AVR, MVR, and multiple VR are shown in Fig. 3. Predictors of mortality identified by univariate and multivariate analysis are shown in Table IX. Despite the good correlation between mortality and extent of coronary artery disease (Fig. 4), this was not an independent risk factor when age and cardiac failure were considered. Because cardiac index, cardiothoracic index, and ejection fraction were measured only in subsets of patients, using these factors together in the multivariate analysis was not possible and a separate Cox's model was defined for each of them. Thus independence among these three covariates was not tested (the same applied to the multivariate analysis for sudden death, valve-related death, and cardiac non–valve-related death). Table IX only presents the independent risk factors for the model including ejection fraction. A high cardiothoracic index and a low cardiac index were independent risk factors for overall mortality in their respective models. Survival in the AVR group tended to be worse than that in the MVR group, although this difference did not reach significance (p = 0.069). When the analysis was done separately for the AVR group, the same risk factors for mortality were identified univariately. For the MVR group, the risk factors identified by log-rank test were coronary artery bypass grafting, age 60 years or older, cardiothoracic index, emergency operation, NYHA class III or IV, alcoholism, diabetes, and presence of cardiovascular risk factors. Cardiac index was borderline (p = 0.061). We did not find in either group a difference in survival between stenotic, regurgitant, or mixed lesions.
|
|
|
|
23 mm) was found in the valve-related death group. Comparing the sudden death group with the group of cardiac non–valve-related deaths, we noted only a greater frequency of preoperative aortic stenosis in the sudden death group (p = 0.04). The risk factors for sudden death and cardiac non–valve-related death were similar (Table IX). High cardiothoracic index and low cardiac index were independent factors for cardiac non–valve-related death in their models, and the latter was also a risk factor for sudden death. In comparison, the prognostic factors for the valve-related death group were different (Table IX). Actuarial freedom from valve-related deaths, with and without sudden deaths included, is shown in Fig. 5. Actuarial freedom rates for sudden death only at 5 years were 96.6% ± 1.1%, at 10 years 91.5% ± 1.8%, and at 13 years 91.0% ± 1.8%. Actuarial freedom rates for cardiac non–valve-related death only at 5 years were 95.4% ± 1.2%, at 10 years 87.3% ± 2.1%, and at 13 years 81.4% ± 2.8%. At the end of follow-up, 10% of patients in preoperative NYHA class I, 35% in class II, 48% in class III, and 55% in class IV had died after a mean survival time of 112, 83, 78, and 56 months, respectively.
|
|
Discussion
The focus of our comments in this discussion will be on particular points concerning late complications.
Embolism.
The linearized and actuarial event-free rates for embolism are in accordance with the findings of recent studies confirming the low thrombogenicity of this valve. Table X presents the range of incidence of the usual valve-related end points (expressed in linearized form) in 11 of the most important follow-up studies on the St. Jude Medical valve. Our data show that embolism is neither the most frequent nor the most dangerous valve-related complication in terms of survival. However, it is the most recurrent event. Most emboli are related to inadequate anticoagulation. This emphasizes the need for adequate anticoagulation after valve replacement with the St. Jude Medical prosthesis. Atrial fibrillation is not a significant incremental risk factor for embolism but is probably partially involved as shown in a later section of this discussion. As also demonstrated by others, 24-26 a history of prior embolism is an independent risk factor for recurrent events. The reason is not evident, but we nevertheless found atrial fibrillation and cardiovascular risk factors to be more frequent in the group with a history of emboli. Cardiac failure as the indication for operation is an important risk factor because it represents patients with a greater cardiothoracic index, larger atrial diameter, higher NYHA class, and more mitral abnormalities. These findings, as emphasized by Horstkotte and associates, 27 confirm the need to consider cardiac comorbidity in defining the best range for anticoagulation in individual patients.
|
Table X, and this cannot be attributed only to variations in the anticoagulation target range or in the mode of control of anticoagulation. Because every bleeding episode that necessitated hospital care was counted (among which many cases were treated on an outpatient basis), we certainly included more benign events than most reports. Horstkotte and associates 27 observed in a prospective study a linearized rate of 6.7%/pt-yr taking into account all benign and severe hemorrhagic events. This indicates that rates in retrospective studies are probably underestimated (the same conclusion applies to embolic events). It has been clear for the past few years that the anticoagulation target for the St. Jude Medical prosthesis should be lower than was previously advised to reduce hemorrhages without an increased risk of thromboemboli. 18,27,28 The recently published guidelines of the European Society of Cardiology recommend an INR of 2.5 to 3.0 after AVR and 3.0 to 3.5 after MVR with the second-generation valves. 29 In other respects, hemorrhages seem slightly more frequent in the AVR group beyond 9 years (Fig. 2); we do not have any satisfactory explanation for that.
Cerebrovascular accidents.
Eighty-six percent of all emboli and 18% of all hemorrhagic events were cerebral; the former were lethal in 5% and the latter in 50%. Because of the predominance of ischemic cerebrovascular accidents, risk factors for cerebrovascular accidents are similar to those for embolism, but age plays a more significant part for cerebrovascular accidents.
Other valve-related complications.
Our results happen to be within the ranges presented in Table X. Valve thrombosis is a threatening complication with a 50% mortality rate and is strongly linked to deficient anticoagulation. Endocarditis is significantly more frequent in patients with diabetes or alcoholism or when the underlying valve disease was aortic stenosis, particularly that of a degenerative cause. Alcoholism, diabetes, and AVR are independent factors. If degenerative disease is a well-known risk factor of native valve endocarditis, especially in the elderly, to our knowledge it has not been described as a persistent risk factor after valve replacement. This may be because of the persistence or the relapse of calcific depositions and areas of epithelial denudation, which promote the development of vegetations. Furthermore, patients in this group are older and age is a major risk factor for all events. Multiple VR was not associated with an increased risk of endocarditis in the present report but this remains controversial in the literature.
Overall mortality and survival.
The linearized rates of mortality and actuarial survival at 10 years are comparable with those given by Ibrahim, 17 Khan, 13 and Fernandez 12 and their coworkers. Baudet and associates 18 showed higher survival rates in a comparable sample of patients, but freedom from valve-related death was the same. Horstkotte and associates 27 also observed a significantly better overall survival at 10 years (AVR, 85%; MVR, 78%; double VR, 72%). However, the population was different with more female patients, younger patients, and a prospective follow-up that could help identify complications earlier and contribute to a better prognosis. With respect to risk factors for global mortality, coronary artery disease was not independent when considered with age and cardiac failure. Cardiothoracic index and ejection fraction are simple and useful parameters to determine the best time for operation and to predict the outcome for the patient. Syncope, as an indication for operation, has a favorable effect on late survival; this has already been described in previous series of aortic stenosis. History of prior systemic emboli was an independent risk factor for overall mortality and cardiac death; it was not related to an increased incidence of fatal cerebral emboli as might be expected and, indeed, the first cause of death in this subgroup was cardiac failure with no valve-related death. Patients with a history of prior emboli show various negative characteristics, some well known (atrial fibrillation, increased risk of atherosclerosis, and increased frequency of past cardiac operations) and possibly others that are unknown. Eliminating this risk factor from the multivariate analysis did not make others become significant. In other respects, the poorer survival for the AVR group (Fig. 3) was somewhat surprising and not in keeping with findings of most past studies. Only Kratz and colleagues 11 showed similar findings. In the present study, this trend was caused by an increased incidence of various valve-related complications, which led to more frequent valve-related deaths, particularly for degenerative sclerosing valve disease. The increased sex ratio; the increased frequency of coronary artery disease, coronary artery bypass grafting, and hypertension; and the higher proportion of older patients (>70 years) in the AVR group compared with the MVR and multiple VR groups together do not completely explain this point: the patient characteristics and the surgical technique used for AVR did not differ notably from what has previously been reported. Although a selection bias cannot be excluded, we perhaps missed some critical factors in the history or in the general, cardiac, and valvular state of these patients that could explain this discrepancy.
Sudden death and valve-related death.
Above all, sudden death was linked with a low preoperative ejection fraction or with coronary artery disease, suggesting a non–valve-related cause. We agree with Butchart 30 that "it is illogical to attribute every sudden death in a prosthetic valve series to the prosthesis itself, particularly when a high proportion of the patients have coronary disease"; this realization led us to present our data with sudden deaths separately or integrated in valve-related deaths. Alcoholism was a risk factor for sudden death and for valve-related death probably because of anticoagulation-related hemorrhages.
The incidence of valve-related deaths was higher with small prostheses (23 mm), but small valves were used only for aortic replacements, and use of small valves was thus a demonstrated risk factor only in this category. Small diameter is significantly correlated with female sex, small body surface area, and increased age, but this does not fully explain the increased risk; the small prostheses may be submitted to higher forces that lead to paravalvular leaks and create turbulences in blood flow that favor thrombosis, embolisms, and vegetations. Alcoholism and a degenerative cause of valve disease were also independent factors of valve-related death but, importantly, age was not.
Cardiac non–valve-related death.
Cardiac death remained the primary cause of death and was directly related to the degree of preoperative coronary disease, myocardial dysfunction, and cardiac enlargement despite surgical revascularization at the time of valve replacement. This emphasizes the need for early valvular operation.
Functional improvement.
For survivors the functional result was excellent and sustained with most of patients being in NYHA class I or II after 10 years. That 77% of the nonretired patients were professionally active has, of course, an important positive impact in socioeconomic terms.
Limitations of the study.
Although we tried to minimize the effect of the retrospective collection of data by a thorough examination of hospital files and by precise questionnaires, events such as transient ischemic attacks, for example, may be missed and information on hemodynamic values and on anticoagulation control at the time of events was not complete. Differences in the definitions of endpoints made comparisons among studies difficult. The guidelines of The Society of Thoracic Surgeons and The American Association for Thoracic Surgery have only partially solved this problem, particularly concerning embolic events, valve-related hemorrhages, and sudden death. We used several definitions in the methods section that may not correspond to those used in other reports. Statistical power, when analyzing covariates in subgroups such as operative groups or types of valvular disease, was low because of small numbers of patients or events.
In conclusion, this study confirms that with St. Jude Medical valves, prosthesis-related complications are still common and sometimes dangerous. About 20% of patients, at 10 years, will be dead of valve-related causes or have a permanent impairment related to the prosthesis. The preoperative status of the patients is a major determinant of the long-term outcome. Anticoagulation should be optimal and adapted to the individual patient taking into account the various risk factors for emboli and hemorrhages. In patients without complications the functional improvement is immediate, sustained, and warrants a normal professional life in the great majority of cases.
Appendix
The following table shows the risk factors screened by log-rank test for each valve-related complication, for overall mortality, for valve-related death, sudden death, cardiac non–valve-related death, and for combined events (mortality and morbidity, mortality and permanent morbidity) with and without sudden death included. After this first screening, we tested, when needed, one or more specific physiopathologic anomalies against the others. The same was done for the different causes of valve disease or for the distinct groups of valvular replacement.
|
Acknowledgments
We thank the Swiss Office Fédéral de la Statistique for the information received on death certificates for some of our patients. Dr. Debétaz is infinitely grateful to his wife and children for their help and huge patience during all the time necessary to achieve this study.
Footnotes
Service of Cardiothoracic Surgeryb 
Institute for Social and Preventive Medicinec University Hospital, Lausanne, Switzerland.
*References – 1,2,4,9,10,12-15,17,18.
References
This article has been cited by other articles:
|
|
 |
|
  O. Lund and M. Bland Risk-corrected impact of mechanical versus bioprosthetic valves on long-term mortality after aortic valve replacement J. Thorac. Cardiovasc. Surg., July 1, 2006; 132(1): 20 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Mistiaen, Ph. Van Cauwelaert, Ph. Muylaert, S. U. Sys, F. Harrisson, and H. Bortier Thromboembolic events after aortic valve replacement in elderly patients with a Carpentier-Edwards Perimount pericardial bioprosthesis J. Thorac. Cardiovasc. Surg., April 1, 2004; 127(4): 1166 - 1170. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vink, R. A. Kraaijenhagen, B. A. Hutten, R. B. A. van den Brink, B. A. de Mol, H. R. Buller, and M. Levi The optimal intensity of vitamin k antagonists in patients with mechanical heart valves: A meta-analysis J. Am. Coll. Cardiol., December 17, 2003; 42(12): 2042 - 2048. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. L. Grunkemeier and Y. Wu "Our complication rates are lower than theirs": Statistical critique of heart valve comparisons J. Thorac. Cardiovasc. Surg., February 1, 2003; 125(2): 290 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gelsomino, G. Morocutti, P. Da Col, R. Frassani, R. Carella, G. Minen, and U. Livi Preliminary experience with the St. Jude Medical Regent mechanical heart valve in the aortic position: early in vivo hemodynamic results Ann. Thorac. Surg., June 1, 2002; 73(6): 1830 - 1836. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Milano, M. De Carlo, G. Mecozzi, A. D'Alfonso, G. Scioti, C. Nardi, and U. Bortolotti Clinical outcome in patients with 19-mm and 21-mm St. Jude aortic prostheses: comparison at long-term follow-up Ann. Thorac. Surg., January 1, 2002; 73(1): 37 - 43. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Perleth and R. Busse Health technology assessment of oral anticoagulation following heart valve replacement Eur. Heart J. Suppl., December 1, 2001; 3(suppl_Q): Q60 - Q64. [Abstract] [PDF]
|
|
|
|
|
|
F. P. Casselman, M. L. Bots, W. Van Lommel, P. J. Knaepen, R. Lensen, and F. E.E. Vermeulen Repeated thromboembolic and bleeding events after mechanical aortic valve replacement Ann. Thorac. Surg., April 1, 2001; 71(4): 1172 - 1180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Lund, S. L. Nielsen, H. Arildsen, L. B. Ilkjaer, and H. K. Pilegaard Standard aortic St. Jude valve at 18 years: performance profile and determinants of outcome Ann. Thorac. Surg., May 1, 2000; 69(5): 1459 - 1465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Lund, H. K. Pilegaard, L. B. Ilkjaer, S. L. Nielsen, and H. Arildsen Performance profile of the Starr-Edwards aortic cloth covered valve, track valve, and silastic ball valve Eur. J. Cardiothorac. Surg., October 1, 1999; 16(4): 403 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. K. Rosengart, M. O'Hara, S. J. Lang, W. Ko, N. Altorki, K. H. Krieger, and O. W. Isom Outcome analysis of 245 CarboMedics and St. Jude valves implanted at the same institution Ann. Thorac. Surg., November 1, 1998; 66(5): 1684 - 1691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-P. Remadi, P. Bizouarn, O. Baron, O. Al Habash, P. Despins, J.-L. Michaud, and D. Duveau Mitral valve replacement with the St. Jude medical prosthesis: a 15-year follow-up Ann. Thorac. Surg., September 1, 1998; 66(3): 762 - 767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Ruchat, M. Hurni, A. P. Fischer, and H. Sadeghi Semicontinuous Suture Technique for All Prosthetic Valve Insertions: The "Hoist" Technique Ann. Thorac. Surg., March 1, 1998; 65(3): 859 - 860. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ninet, F. Tronc, J. Robin, A. Curtil, I. Aleksic, and G. Champsaur Mechanical versus biological isolated aortic valvular replacement after the age of 70: equivalent long-term results Eur. J. Cardiothorac. Surg., January 1, 1998; 13(1): 84 - 89. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
