Ann Thorac Surg 2008;86:769-773. doi:10.1016/j.athoracsur.2008.05.010
© 2008 The Society of Thoracic Surgeons
Original Articles: Adult Cardiac
Cerebrovascular Events After Stentless Aortic Valve Replacement During a 9-Year Follow-Up Period
Helmut Gulbins, MD*,
Ines Florath, MD,
Juergen Ennker, MD
Department of Cardiac Surgery, Heart Institute Lahr, Institute of Cardiovascular Medicine, University Witten-Herdecke, Hamburg, Germany
Accepted for publication May 5, 2008.
* Address correspondence to Dr Gulbins, University Hospital Eppendorf, University Heart Center, Martinistr. 52, Hamburg, D-20246, Germany (Email: h.gulbins{at}uke.de).
 |
Abstract
|
|---|
Background: One major advantage of biologic aortic valve prostheses is their low thrombogenicity compared with mechanical prostheses. The purpose of this study was to evaluate the incidence of cerebrovascular events during long-term follow-up after stentless aortic valve replacement.
Methods: Between 1996 and 2005, 1,014 patients (mean age, 73 years; range, 20 to 90 years) received stentless aortic valve replacement (Freestyle; Medtronic, Minneapolis, MN) and were included into the systematic follow-up that was closed in 2006 with a completeness of 94.7% and a mean follow-up interval of 3 years (range, 0.5 to 9.8 years). Predictors for freedom from cerebrovascular events were identified by Cox regression.
Results: Overall survival was 53% ± 5% after 8 years (mean, 6.8 ± 0.2 years). Permanent atrial fibrillation at time of surgery was a strong predictor of impaired survival during follow-up. Freedom from cerebrovascular events during follow-up was 68% ± 5% at 9 years of follow-up. Multivariate regression analysis revealed previous stroke, age at implant, diabetes mellitus, and carotid lesions as significant risk factors. Especially age older than 75 years was a strong risk factor for cerebrovascular events during follow-up (p = 0.004). Atrial fibrillation was not an independent risk factor for cerebrovascular events (p = 0.26) but was a strong predictor of poor survival (p < 0.001) during follow-up. There was no influence of technique of implantation (subcoronary versus full root; p = 0.41), sex (p = 0.35), additional bypass grafting (p = 0.65), and the size of the implanted prosthesis (p = 0.47).
Conclusions: The risk of cerebrovascular events during follow-up after stentless aortic valve replacement is related to the individual risk factors of the patients rather than to the valve prosthesis itself. Without additional risk factors, patients with these aortic valve prostheses showed an incidence of cerebrovascular events similar to those reported for a healthy population adjusted for age.
|
Introduction
|
|---|
One major advantage of biologic heart valve prostheses is their low thrombogenicity making a lifetime anticoagulative therapy with coumarin unnecessary [1, 2]. The main disadvantage is the limited durability, and therefore theses valve prostheses are usually recommended in patients older than 60 to 65 years [1, 3, 4]. These patients have an elevated risk for cerebrovascular events (CEs) even without artificial heart valve replacement or heart valve disease [5, 6]. In this study, we evaluated the incidence of postoperative CEs during follow-up in patients with stentless aortic valve replacement. Additionally, we evaluated the impact of preoperative risk factors for cerebral ischemic incidents during follow-up.
|
Patients and Methods
|
|---|
Patients
Between 1996 and 2005, 1,014 patients received a stentless aortic valve (Freestyle, Medtronic, Minneapolis, MN) for aortic valve replacement at our institution. All patients were included into the systematic follow-up. It started 6 months after surgery and was then continued on an every 2 year basis. The patients or their physicians were contacted directly. Follow-up was closed in April 2006; it was complete to 94.7% with a mean follow-up interval of 3 years (range, 0.5 to 9.8 years). For this study, the incidence of transitory ischemic attacks, prolonged neurologic deficits, and strokes were analyzed. Because of the difficulties in distinguishing among these entities, especially between prolonged neurologic deficit and stroke, all three were summarized as CEs. Confirmation of the event by neuroimaging was available in fewer than 60% of the patients; therefore, we could not use these results for differentiation. The study was approved by the ethic committee of the country's medical association of Baden-Württemberg, Germany. All patients gave informed consent to participate in the follow-up studies.
Mean patient age at the time of aortic valve replacement was 73 years (range, 20 to 90 years), sex distribution was 1:1 (50.1% male patients), and 385 patients received additional aortocoronary bypass grafting (38%). The majority of prostheses were implanted in a subcoronary fashion (83%, n = 846); the others were implanted as total root (17%, n = 168) with reimplantation of the coronary arteries. No patient had additional antifibrillation surgery such as Maze procedures. Patient characteristics were summarized in Table 1. Postoperatively, patients received heparin until complete mobilization was reached. Anticoagulation with warfarin was done only in patients with persistent atrial fibrillation. Patients with additional coronary artery bypass grafting were treated with acetylsalicylic acid with a daily dosage of 100 mg.
Statistical Analysis
Statistical analysis was performed using the software package SPSS (SPSS Inc, Chicago, IL). Values were expressed in real numbers (%) or mean values ± one standard deviation. The missing value rate was below 5%. Missing values were completed by the mean for continuous variables and by the most frequent event for categorical variables. Survival and freedom from CEs were estimated by the Kaplan–Meier method, expressed in mean ± one standard error and univariately compared by the log-rank test.
Predictors for freedom from CEs and survival time were identified by Cox regression. For variable selection of both multivariate regression models, the Akaike Information Criterion (AIC = deviance of the model + 2 x number of included variables) was calculated for variables showing a difference for the outcome variable with a probability value smaller than or equal to 0.25. The variables were included into multiple regression models in a stepwise way. The AIC was calculated each time a variable was included. The final model is reached when no more reduction in AIC is observed.
To verify the assumption that the effect of the covariate is linear in the log hazard, we plotted the estimated coefficients for the designated variable formed from quartiles versus the midpoints of the groups (Fig 1). If the polygon connecting the points is nearly linear, then the correct scale for the analyzed variable is linear in the log hazard. Otherwise a transformation is necessary.
View larger version (7K):
[in this window]
[in a new window]
|
Fig 1. Relationship between the estimated coefficients of the variable age transformed in quartiles and the midpoints of each age group.
|
|
Data were collected on all patients as a part of a national quality assessment trial of cardiothoracic surgery (Quadra: Quality Assurance Data Review Analysis) using standardized protocols of the German Society of Thoracic and Cardiovascular Surgery. The following variables were considered as potential confounders: age; sex; body mass index; body surface area; body height; body weight; concomitant coronary artery bypass grafting; emergency procedure; re-replacement of the aortic valve; New York Heart Association classification; physiologic subgroup as aortic stenosis, aortic insufficiency, or combined aortic valve disease (aortic stenosis and insufficiency, both with grade II or more); history of congestive heart failure; previous myocardial infarction (<90 days); history of syncope; history of venous or arterial embolism; hyperlipidemia (treated); hypertension (treated); diabetes mellitus (on diet or medication); previous coronary artery bypass grafting; peripheral occlusive arterial disease (intermittent claudication or previous vascular surgery); internal carotid artery stenosis (>70%); neurologic disorders (affecting ambulation or day-to-day functioning); history of renal disease; chronic pulmonary disease; atrial fibrillation before surgery and before discharge; and anticoagulation therapy at discharge and follow-up time.
|
Results
|
|---|
Overall survival was 53% ± 5% years after 8 years with a mean survival time of 6.8 ± 0.2 years. The perioperative incidence of CE was 1.8%. Two hundred thirty-one patients (23%) were on long-term anticoagulation mainly owing to permanent atrial fibrillation (19%). Survival at 5 and 8 years was 55% ± 7% and 29% ± 10% versus 75% ± 3% and 59% ± 6%, respectively, for patients with and without preoperative atrial fibrillation. Permanent atrial fibrillation significantly impaired survival when it was present at the time of surgery (Fig 2) and was also identified as an independent risk factor (hazard ratio [HR], 1.75; 95% confidence interval [CI], 1.3 to 2.36) after adjustment of further risk factors such as age (HR, 1.57; 95% CI, 1.24 to 2.0 per 10 years), congestive heart failure (HR, 1.42; 95% CI, 1.04 to 1.94), diabetes mellitus (HR, 1.45; 95% CI, 1.07 to 1.97), ejection fraction (HR, 0.99; 95% CI, 0.98 to 1.00), liver insufficiency (HR, 1.83; 95% CI, 1.19 to 2.81), serum creatinine (HR, 1.3; 95% CI, 1.11 to 1.53), and body mass index (HR, 0.96; 95% CI, 0.92 to 0.99). Freedom from thromboembolic events during follow-up was 68% ± 5% at 9-year follow-up. Preoperative risk factors influencing the incidence of CEs during follow-up are listed in Table 2. Multivariate regression analysis revealed previous stroke, age at implant, diabetes mellitus, and carotid lesions as significant risk factors for CEs during follow-up (Table 3). To verify the linear assumption for the variable, age was divided into quartiles and the coefficients of the newly designed variable estimated by the Cox-regression were plotted versus the midpoint of each group (Fig 1). The figure shows that age is not linear in the log hazard, and the relationship looks nearly binary with a reasonable cutoff point at 75 years. Taking 75 years of age as the cutoff point, the impact of age increases as an independent risk factor (p = 0.004 vs p = 0.046). Atrial fibrillation was not an independent risk factor (p = 0.26) for CEs, but it was a strong predictor of poor survival (p < 0.001) during follow-up. The technique of implantation had no impact on CEs during follow-up (p = 0.41) as was the case for sex (p = 0.35), additional coronary artery bypass grafting (p = 0.65), and the size of the implanted prostheses (p = 0.47).
View larger version (29K):
[in this window]
[in a new window]
|
Fig 2. Overall survival with and without atrial fibrillation (AF). Survival was significantly (p < 0.05) worse in patients with persistent atrial fibrillation.
|
|
Patients with none of these risk factors had an incidence of CE of 1.37% per patient-year during follow-up. Patients with more than one risk factor had a fourfold risk for CEs (5.18% per patient-year) during follow-up. Freedom from CEs at 5 and 7 years was 95% ± 3% and 86% ± 5%; 89% ± 2% and 76% ± 7%; and 74% ± 6% and 57% ± 19% for patients with no risk factors, with one risk factor, and with more than one risk factor, respectively (Fig 3).
View larger version (34K):
[in this window]
[in a new window]
|
Fig 3. Impact of the identified risk factors (RF) on freedom from cerebrovascular events. The incidence of thromboembolic events per patient-year was nearly threefold in the presence of additional risk factors. Without risk factors, the incidence was 1.25% per patient-year. During the first 3 years, however, the incidence was zero, underlining age as the most powerful risk factor for cerebrovascular events (CE).
|
|
|
Comment
|
|---|
These results demonstrate the low thrombogenicity of stentless biologic aortic valves. In patients without additional risk factors, the incidence of CEs was similar to that reported in a healthy population adjusted for age [5, 6]. The results are similar to those reported for stented pericardial valves [7, 8]. The biologic tissue, regardless of the glutaraldehyde-fixation and its lack of an endothelial cell coverage, obviously has a low thrombogenicity. In addition, the movements of the biologic valve cusps are smoother, avoiding unnecessary stress to the blood, like the phenomenon of cavitation that was reported for mechanical heart valves [9, 10]. Cavitation causes a negative pressure for several milliseconds during closure of mechanical leaflets. This results in formation of small microbubbles imploding after pressure equalization, thus damaging corpuscular blood components.
Patients without additional risk factors showed a nearly normal incidence of CEs during follow-up. The plot underlines the impact of age on this complication during long-term follow-up: with increasing follow-up time the plot becomes steeper, reflecting the increased age of the patients at risk. Age as the strongest risk factor when older than 75 years was also reported in the literature [5, 6, 11]. In these patients, no anticoagulative therapy is recommended currently, but the increased risk suggests a potential benefit of anticoagulation in these patients [12].
Coumarin therapy itself was an independent risk factor as it was administered not only in patients with atrial fibrillation but also in those with other vascular or cardiac disorders, increasing the risk for stroke. One shortcoming of the present study was that we could not distinguish between hemorrhagic and ischemic stroke because these data were not available on enough patients to allow rigorous analysis. Therefore we can only suppose that the higher incidence of CEs in patients on chronic coumarin therapy was influenced by a high-risk situation for thromboembolic and ischemic events as well as by hemorrhagic strokes. The incidence of bleeding complications, however, also increases with age [13, 14] to a level between 2% and 4% per patient-year. The incidence of 1.25% of CEs per patient-year in those without additional risk factors was markedly lower compared with the estimated incidence of severe bleeding complications under coumarin therapy in this collective. This suggests no benefit of this anticoagulation therapy in older patients without additional risk factors. Aspirin has also been shown to reduce the incidence of CEs with less bleeding complication compared with coumarin. A large randomized and prospective multicenter study could evaluate a potential effect of prophylactic aspirin therapy in older patients without additional risk factors for CEs.
The identified risk factors are not surprising as they represent the typical factors influencing CEs. Patients with the identified risk factors additional to age presented with a nearly threefold incidence of CEs during follow-up. When excluding those already on coumarin therapy, again the question can be raised whether these patients may benefit from a prophylactic inhibition of thrombocyte aggregation during follow-up. In our analysis, we could not support such a benefit, but the aspirin therapy was not randomized and mostly associated with additional coronary artery disease and not with the probability of CEs.
Atrial fibrillation is known to be an independent risk factor for stroke [14–16]. In our analysis, however, it was not. One reason might be the consequent anticoagulation in these patients, avoiding this complication. This, together with the rather low number of patients at risk in the present study (only 19% had permanent atrial fibrillation), might have avoided atrial fibrillation becoming an independent risk factor. This underlines the benefit of a consequent anticoagulative therapy in patients with permanent atrial fibrillation [12, 15]. Another explanation is the lack of information concerning the definite cause of death because most of the patients who died during follow-up had no autopsy. Cerebrovascular events causing disastrous stroke might have been partly responsible for the poorer survival during follow-up in patients with atrial fibrillation. Recently, Ngaage and colleagues [17] also found atrial fibrillation to be an independent risk factor for poor survival after aortocoronary bypass grafting. Since 2006, we started to perform left atrial ablation procedures in patients with atrial fibrillation. The results for these patients, however, are not included in the presented study.
Because age was the strongest risk factor for CEs, the patients who would profit most from biologic heart valve prostheses were patients younger than 65 years of age with nearly no risk for CEs. After mechanical heart valve replacement, their risk increases as a result of thromboembolic events or hemorrhagic complications during follow-up compared with a normal healthy population. The limited durability of biologic heart valves diminishes this advantage as the risk of reoperation reduces the benefit of a lower thrombogenicity. There are, however, studies going on to evaluate biologic heart valve prostheses in younger patients. These studies are important to calculate the real risk and complication rate for both types of heart valve prostheses. If the xenogeneic heart valves perform equally to allogeneic homografts in these younger patients [18, 19], this will influence the strategy of valve choice in the future.
Altogether, the stentless porcine aortic valve performed very well regarding the stroke incidence during follow-up [20, 21]. Traditionally, all CEs after valvular surgery were attributed to the valve, although many of them were surely influenced by the patient's individual risk profile. These risk factors were also identified for stented bioprostheses [8]. The low incidence of thromboembolic events in patients without risk factors, however, underlines the generally low thrombogenicity of biologic valves and the lack of necessity for postoperative anticoagulation.
The risk of CEs during follow-up after stentless aortic valve replacement is related to the individual risk factors of the patients rather than to the valve prosthesis itself. Without additional risk factors, patients with these aortic valve prostheses showed an incidence of CEs similar to those reported for a healthy population adjusted for age.
|
References
|
|---|
- Bonow RO, Carabello BA, Kanu C, et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to revise the 1998 Guidelines for the Management of Patients with Valvular Heart Disease): developed in collaboration with the Society of Cardiovascular Anesthesiologists: endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons Circulation 2006;114:e84-e231.[Free Full Text]
- Mistiaen W, Van Cauwelaert P, Muylaert P, Sys SU, Harrisson F, Botier H. Thromboembolic events after aortic valve replacement in elderly patients with a Carpentier-Edwards Perimount pericardial prosthesis J Thorac Cardiovasc Surg 2004;127:1166-1170.[Abstract/Free Full Text]
- Wheatley DJ. The ‘threshold age’ in choosing biological versus mechanical prostheses in western countries J Heart Valve Dis 2004;13(Suppl):S91-S94.[Medline]
- Kaempchen S, Guenther T, Toschke M, Grunkemeier GL, Wottke M, Lange R. Assessing the benefit of biological valve prostheses: cumulative incidence (actual) vs. Kaplan-Meier (actuarial) analysis Eur J Cardiothorac Surg 2003;23:710-713.[Abstract/Free Full Text]
- Caradang R, Seshadri S, Beiser A, et al. Trends in incidence, lifetime risk, severity, and 30-day mortality of stroke over the past 50 years JAMA 2006;296:2939-2946.[Abstract/Free Full Text]
- Seshadri S, Beiser A, Kelly-Hayes M, et al. The lifetime risk of stroke: estimates from the Framingham Study Stroke 2006;37:345-350.[Abstract/Free Full Text]
- Aupart MR, Mirza A, Meurisse YA, Sirinelli AL, Neville PH, Marchand MA. Perimount pericardial bioprosthesis for aortic calcified stenosis: 18-year experience with 1133 patients J Heart Valve Dis 2006;15:768-775.[Medline]
- Ruel M, Masters RG, Rubens FD, et al. Late incidence and determinants of stroke after aortic and mitral valve replacement Ann Thorac Surg 2004;78:77-83.[Abstract/Free Full Text]
- Graf T, Reul H, Dietz W, Wilmes R, Rau G. Cavitation of mechanical heart valves under physiologic conditions J Heart Valve Dis 1992;1:131-141.[Medline]
- Graf T, Reul H, Detlefs C, Wilmes R, Rau G. Causes and formation of cavitation in mechanical heart valves J Heart Valve Dis 1994;3(Suppl):S49-S64.[Medline]
- Ennker J, Dalladaku F, Rosendahl U, Ennker IC, Mauser M, Florath I. The stentless freestyle bioprosthesis: impact of age over 80 years on quality of life, perioperative, and mid-term outcome J Card Surg 2006;21:379-385.[Medline]
- Dalal PM, Mishra NK, Battacherjee M, Bhat P. Antithrombotic agents in cerebral ischemia J Assoc Physicians India 2006;54:555-561.[Medline]
- Shireman TI, Mahnken JD, Howard PA, Kreskowik TF, Hou Q, Ellerbeck EF. Development of a contemporary bleeding risk model for elderly warfarin recipients Chest 2006;130:1390-1399.[Medline]
- DiMarco JP, Flaker G, Waldo AL, et al. Factors affecting bleeding risk during anticoagulant therapy in patients with atrial fibrillation: observations from the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study Am Heart J 2005;149:650-656.[Medline]
- Andersen KK, Olsen TS. Reduced poststroke mortality in patients with stroke and atrial fibrillation treated with anticoagulants: results from a Danish quality-control registry of 22,179 patients with ischemic stroke Stroke 2007;38:259-263.[Abstract/Free Full Text]
- Kakar P, Lip GY. Atrial fibrillation and stroke prevention Expert Rev Neurother 2006;6:1523-1530.[Medline]
- Ngaage DL, Schaff HV, Mullany CJ, et al. Does preoperative atrial fibrillation influence early and late outcomes of coronary artery bypass grafting? J Thorac Cardiovasc Surg 2007;133:182-189.[Abstract/Free Full Text]
- Gulbins H, Kreuzer E, Reichard B. Homografts: a review Expert Rev Cardiovasc Ther 2003;1:533-539.[Medline]
- O'Brien MF, Harrocks S, Stafford EG, et al. The homograft aortic valve: a 29-year, 99.3% follow-up of 1,022 valve replacements J Heart Valve Dis 2001;10:334-344.[Medline]
- Kappetein AP, Puvimanasinghe JP, Takkenberg JJ, Steyerberg EW, Bogers AJ. Predicted patient outcome after aortic valve replacement with Medtronic Stentless Freestyle bioprostheses J Heart Valve Dis 2007;16:423-428.[Medline]
- Deleuze PH, Fromes Y, Khoury W, Maribas P, Lemaire S, Bical OM. Eight-year results of Freestyle stentless bioprosthesis in the aortic position: a single-center study of 500 patients J Heart Valve Dis 2006;15:247-252.[Medline]
Top
Abstract
Introduction
