Ann Thorac Surg 2003;75:1175-1180
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
The Carbomedics Top Hat supraannular aortic valve: a multicenter study
A. Marc Gillinov, MDa*,
Eugene H. Blackstone, MDa,b,
Joan M. Alster, MSb,
Joseph M. Craver, MDc,
William A. Baumgartner, MDd,
Scot A. Brewster, MDe,
Leonard H. Kleinman, MDf,
Nicholas G. Smedira, MDa
a Department of Thoracic and Cardiovascular Surgery, Cleveland, Ohio USA
b Department of Biostatistics and Epidemiology, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
c Department of Cardiac Surgery, Emory University, Atlanta, Georgia, USA
d Department of Cardiac Surgery, The Johns Hopkins Hospital, Baltimore, Maryland, USA
e La Jolla Cardiovascular and Thoracic Surgeons, La Jolla, California, USAUSA
f Cardiovascular Surgery Associates of Milwaukee, Milwaukee, Wisconsin, USA
Accepted for publication June 26, 2002.
* Address reprint requests to Dr Gillinov, Department of Thoracic and Cardiovascular Surgery, Desk 25, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH44195, USA
e-mail: gillinom{at}ccf.org
 |
Abstract
|
|---|
BACKGROUND: Introduced in 1993, the Carbomedics Top Hat (Sulzer, Carbomedics, Austin, TX) valve is a bileaflet mechanical aortic prosthesis designed to be placed in a supraannular position. Five institutions pooled their clinical experiences to evaluate early outcome in patients with this prosthesis.
METHODS: From 1994 to 2000, 639 patients underwent aortic valve replacement with Top Hat (Sulzer Carbomedics) valves at 5 institutions. Mean age was 60 ± 13 years. In this heterogeneous population, 28% of patients had previous cardiac operations and 64% had concomitant procedures, including procedures involving more than 1 heart valve in 32%. Implanted prostheses sizes included the 19 mm (15%), 21 mm (37%), 23 mm (33%), 25 mm (13%), and 27 mm (2%). Mean follow-up was 2.0 ± 1.5 years, and there were 1,206 patient-years of follow-up available for analysis.
RESULTS: Thirty-day mortality was 5.3%. Five-year survival was 74%. Risk factors for death included older age (p = 0.01), decreased ejection fraction (p = 0.007), and increased New York Heart Association functional class (p = 0.003). Five-year freedoms from thromboembolism and hemorrhage were 90% and 85%, respectively. Five-year freedoms from explant and endocarditis were both 99%. There were no structural valve failures.
CONCLUSIONS: The Top Hat valve outcomes have been similar to those of the standard Carbomedics intraannular prostheses. The unique design of the Top Hat valve, with all its components in the aortic sinuses, has particular advantages in the small aortic root, in settings where leaflet entrapment may occur, and in multiple valve replacement.
|
Introduction
|
|---|
| Drs Craver and Smedira disclose that they have a financial relationship with Sulzer Medica Carbomedics.
|
Introduced in 1993, the Carbomedics Top Hat (Sulzer Carbomedics, Austin, TX) aortic prosthesis is a bileaflet mechanical valve designed for implantation in the supraannular position. This design positions the valve components within the sinuses of Valsalva. Recent data document the performance of the Carbomedics CPHV standard intraannular aortic valve (Sulzer Carbomedics, Austin, TX) [1–6]; however, there is little data addressing the Top Hat (Sulzer Carbomedics) valve. This multiinstitutional study was undertaken to determine early results obtained with the Top Hat aortic prosthesis.
|
Patients and methods
|
|---|
Patients
We identified 639 consecutive patients from five institutions who underwent aortic valve replacement with Top Hat prostheses (Table 1)
between September 1994 and September 2000. All patients receiving the Top Hat valve were included in the study to reflect real world use of this device in clinical practice.
Mean age of this heterogeneous group was 60 ± 13 years, and most patients were in the New York Heart Association functional classes II and III (Table 2).
Among 182 patients with a history of cardiac operations, specific previous cardiac surgical procedures included aortic valve replacement in 29%, mitral valve replacement in 30%, and coronary artery bypass grafting in 35% (Table 3).
Thirty-six percent of the 639 patients had isolated aortic valve replacements (including 173 patients undergoing primary isolated valve replacements), but the majority of patients required additional cardiac procedures (Table 4).
Fifty-two percent (n = 332) of implanted valves were labeled either size 19 mm or 21 mm, indicating a possible preference for use of these prostheses in patients with small aortic root (Table 5).
Surgical technique
Although the Top Hat valve sits in a supraannular position, suture techniques varied. The most common technique consisted of horizontal mattress sutures with pledgets on the ventricular aspect (n = 431; 67%). Other techniques included mattress sutures without pledgets (n = 115; 18%) and figure eight sutures (n = 61; 10%). In 32 patients (5%), suture techniques were unspecified. The aortotomy was closed with a pericardial patch in 66 cases (10%). Four patients (1%) required unplanned bypass of the right coronary artery.
Follow-up
Follow-up information was obtained by direct contact with patients or their physicians, or both. In cases of late complications, physician or hospital records, or both, were obtained. All reported morbid events were scrutinized carefully and classified uniformly using standard definitions [7]. When active follow-up information could not be obtained from the patient or physician, survival was ascertained from the United States Social Security Death Index. Follow-up for survival was 98% complete within the calendar year 2000, and active follow-up for valve-related complications was 92% complete within the calendar year 2000. For survival, mean follow-up was 2.0 ± 1.5 years, and there were 1,206 patient-years of follow-up available for analysis. For morbid events, mean follow-up was 1.8 ± 1.5 years, and total follow-up was 1,143 patient-years. Fifty percent of patients were followed for more than 1.7 years, 25% for more than 2.9 years, 10% for more than 4 years, and 5% for more than 5.2 years. We considered time-related estimates reliable to 4 to 5 years.
End points
Standard guidelines were used for definitions of morbidity and mortality [8]. Early event refers to any event occurring within 30 days of an operation. Late event refers to occurrences at all subsequent times.
Data analysis
Descriptive
Descriptive statistics include mean and standard deviation for continuous variables and frequencies and percentages for categorical variables. Selected frequencies are accompanied by asymmetric 68% confidence limits corresponding to ±1 standard error.
Time-related events
Nonparametric estimates of time free from various events were obtained using Kaplan–Meier estimates [9]. For mortality, thromboembolism, and hemorrhage, a parametric method was used to resolve the number of phases of instantaneous risk (hazard function) and to estimate their shaping parameters [10]. Although the hazard function for time-related events was not constant, linearized rates were also compiled [8] for consistency with published literature and reporting guidelines. For events occurring after 30 days, some of which were recurrent, the linearized rate was estimated as the total number of events divided by total patient-years of follow-up after 30 days. Confidence limits for these estimates were estimated under the assumption that rates were normally distributed, although for exactly zero events, we calculated an upper confidence limit using the binomial formula [11].
Multivariate analysis
Multivariate analysis was undertaken for all-cause mortality, including hospital deaths and time to earliest occurrence of thromboembolism or hemorrhage.
Screening of variables possibly related to the events, both initially and during the analysis, used stratified the Kaplan–Meier curves. The potential risk factors (variables) were organized for analysis as shown in Table 6.
Continuous and ordinal variables were assessed univariably by decile risk analysis to suggest transformations of scale to incorporate into multivariable analysis that would ensure the relation of these variables to outcome was well calibrated with respect to model assumptions. For sporadic missing values (see Table 2), noninformative imputation of the mean was used. Bootstrap bagging was the primary method for variable selection [12, 13].
|
Results
|
|---|
Survival
Thirty-day mortality was 5.3%, and 1-year, 3-year, and 5-year survivals were 91%, 83%, and 74%, respectively (Fig 1).
Risk factors for death included older age, worse left ventricular function, advanced symptoms, and previous or concomitant tricuspid valve procedure (Table 7).
Factors not associated with higher risk of mortality included previous cardiac operations, concomitant or previous mitral valve replacements, prosthetic size, and patient size relative to prosthetic size.
View larger version (13K):
[in this window]
[in a new window]
|
Fig 1. Survival after aortic valve replacement. Each circle represents a death, positioned on the vertical axis according to the nonparametric Kaplan–Meier estimator. Vertical bars represent asymmetric 68% confidence limits, equivalent to 1 standard error. The number of patients still traced is shown periodically. The solid line, enclosed within dashed 68% confidence limits, is the parametric estimate of survival.
|
|
Thromboembolism and hemorrhage
Five-year freedoms from thromboembolism and hemorrhage were 90% (Fig 2A)
and 85% (Fig 2B, Table 8),
respectively. The events exhibited an early phase of high risk, followed by a slowly rising lower risk after about 2 weeks postoperatively (Fig 3).
Risk of hemorrhage (hazard function) was more than twice that of thromboembolism in the first years, which slowly approached similar rates by 5 years.
View larger version (14K):
[in this window]
[in a new window]
|
Fig 2. Thromboembolism and hemorrhage after aortic valve replacement. Format is as in Figure 1, except vertical axis has been expanded. (A) Freedom from thromboembolism. (B) Freedom from hemorrhage.
|
|
Other morbidity
Three episodes of valve thrombosis occurred, two managed with thrombolysis and one by surgical thrombectomy. Five-year freedom from valve thrombosis was 99%. There was no early prosthetic valve endocarditis, and 5-year freedom from this event was 99%. Three patients required valve explant, 2 for prosthetic valve endocarditis and 1 for periprosthetic leakage. Five-year freedom from explant was 99%. No instances of structural valve failure were observed. Other valve-related complications are listed in Table 6. One patient underwent reoperation for repair of periprosthetic leakage.
|
Comment
|
|---|
The Carbomedics Top Hat valve is a bileaflet prosthesis with pyrolytic carbon leaflets. Although the valve mechanism is identical to that of the Carbomedics standard intraannular valve, the Top Hat valve is designed for supraannular implantation. This unique design positions all valve components in the aortic sinuses. This feature may facilitate aortic valve replacement in patients with a small aortic root and affords a particular advantage when combined aortic and mitral valve replacements are necessary [14]. This design is also advantageous when there is potential for bileaflet entrapment by annular or subannular components of the left ventricular outflow tract.
There is little clinical data in the literature concerning the Top Hat Valve. Thus the purpose of this report was to examine early outcomes with this prosthesis.
Survival and freedoms from mortality and valve-related morbidity were within the range of those reported in other series examining bileaflet aortic valves [1–7, 14–17]. Although the 74%, 5-year survival is lower than in some series [3, 5, 6], it is similar to that reported by Khan and colleagues [15] and Craver [3]. Dissimilar patient characteristics and inclusion of both previous and concomitant operations vary among reports. This makes direct comparisons of survival problematic, as does omission of early mortality and morbidity in some series. In the current series, risk factors for death were consistent with those identified previously [2, 5].
Thromboembolism and anticoagulant-related hemorrhage were the most common sources of valve-related morbidity. This finding is consistent with patients who have received mechanical valves [1–7, 14, 15]. Identification and analyses of these events are problematic. Identification by follow-up likely underestimates the true number of adverse events. Furthermore, anticoagulant management is an important determinant of the frequency of thromboembolic and hemorrhagic events [16, 17]. In this study, hemorrhage was more frequent than thromboembolism. This implies that the intensity of anticoagulants appropriate for this prosthesis should be lower than that which was used. Further study, perhaps in a prospective fashion, is necessary to examine the issue of anticoagulant-related hemorrhage with this prosthesis.
Several specific issues concerning the Top Hat valve deserve comment. First, Top Hat valves used in this series were of small labeled size. This was surprising, because supraannular placement is intended to facilitate use of larger prostheses [14]. It is possible that surgeons selected smaller valves for patients with a small aortic annulus; however, precise selection criteria and size of the aortic annulus were not available in most cases. Second, the valve sits in the aortic sinuses and there is concern that this may complicate aortotomy closure. Although patch closure of the aorta was performed routinely by one surgeon, aortic root augmentation otherwise was rare. Finally, 4 patients required unplanned bypass of the right coronary artery. This may have been caused by obstruction of the coronary artery by the prosthesis or damage to the coronary ostium at aortotomy closure. As with any aortic prosthesis, it is necessary to remain cognizant of the coronary artery positions during valve implantation.
Limitations
This was a clinical study of a heterogeneous population from several institutions that involved multiple surgeons. However, this is also a strength because it provides a large patient cohort for analysis and permits reporting of outcomes associated with this valve in standard clinical practice. Follow-up concerning morbid events was available in only 92% of patients, although available with respect to death in 98%. As noted previously, collecting morbidity data by follow-up methodology likely results in underreporting; this may be minimized by the short time frame of this study and the diligence with which we confirmed and classified each morbid event. Late echocardiographic data documenting transvalvar gradients were unavailable in the majority of patients.
This study represents early follow-up of patients receiving the Top Hat prosthesis. Long-term results are not yet available. It is an important shortcoming that the international normal ratio was unavailable in the majority of patients; interpretation of thromboembolic and hemorrhagic events requires knowledge of the international normal ratio. Furthermore, there was no uniform strategy for anticoagulation, because these patients were managed by many different physicians at several institutions. Although the high incidence of bleeding relative to thromboembolism suggests that anticoagulation management could be improved in these patients, a prospective study directed specifically at this issue is necessary to formulate appropriate recommendations. Finally, the size of the ventriculoaortic junction was not determined in the majority of patients. This precludes definitive comment concerning the extent to which this valve was used in patients with a small aortic root.
Conclusions
Despite its limitations, this report has value in documenting results of a new valve prosthesis. The Top Hat valve has outcomes similar to those of the standard Carbomedics intraannular prosthesis. The unique design of the Top Hat valve, with all components in the aortic sinuses, is particularly advantageous with small aortic roots, in settings where leaflet entrapment may occur, and with multiple valve replacements.
|
Acknowledgments
|
|---|
This study was funded in part by Sulzer Medica Carbomedics. We thank Colleen Laffey, Diane Alejo, and Barbara Dobbs for their assistance in compiling the data and follow-up, and Tess Knerik for her editorial assistance.
|
References
|
|---|
- Copeland J.G. The CarboMedics prosthetic heart valve: a second generation bileaflet prosthesis. Sem Thorac Cardiovasc Surg 1996;8:237-241.[Medline]
- Li H.H., Hahn J., Urbanski P., Torka M., Grunkemeier G.L., Hacker R.W. Intermediate-term results with 1,019 Carbomedics aortic valves. Ann Thorac Surg 2001;71:1181-1188.[Abstract/Free Full Text]
- Craver J. CarboMedics prosthetic heart valve. Eur J Cardiothorac Surg 1999;15:S3-11.
- Fiane A.E., Geiran O.R., Svennevig J.L. Up to eight years’ follow-up of 997 patients receiving the CarboMedics prosthetic heart valve. Ann Thorac Surg 1998;66:443-448.[Abstract/Free Full Text]
- Fiane A.E., Saatvedt K., Svennevig J.L., Geiran O., Nordstrand K., Frøysaker T. The CarboMedics valve: midterm follow-up with analysis of risk factors. Ann Thorac Surg 1995;60:1053-1058.[Abstract/Free Full Text]
- Bernal J.M., Rabasa J.M., Gutierrez-Garcia F., Morales C., Nistal F., Revuelta J.M. The CarboMedics valve: experience with 1049 implants. Ann Thorac Surg 1998;65:137-143.[Abstract/Free Full Text]
- Dalrymple-Hay M.J., Pearce R.K., Dawkins S., et al. Mid-term results with 1,503 CarboMedics mechanical valve implants. J Heart Valve Dis 2000;9:389-395.[Medline]
- Edmunds L.H., 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. J Thorac Cardiovasc Surg 1996;112:708-711.[Free Full Text]
- Kaplan E.L., Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-481.
- Blackstone E.H., Naftel D.C., Turner M.E., Jr The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assoc 1986;81:615-624.
- Grunkemeier G.L., Anderson W.N., Jr Clinical evaluation and analysis of heart valve substitutes. J Heart Valve Dis 1998;7:163-169.[Medline]
- Breiman L. Bagging predictors. Machine Learning 1996;26:123-140.
- Blackstone E.H. Breaking down barriers: helpful breakthrough statistical methods you need to understand better. J Thorac Cardiovasc Surg 2001;122:430-439.[Free Full Text]
- Lundblad R., Hagen O.M., Smith G., Kvernebo K. The CarboMedics supraanular Top Hat valve improves prosthesis size in the aortic root. J Heart Valve Dis 2001;10:196-201.[Medline]
- Khan S., Chaux A., Matloff J., et al. The St. Jude Medical valve. Experience with 1000 cases. J Thorac Cardiovasc Surg 1994;108:1010-1020.[Abstract/Free Full Text]
- Stein P.D., Alpert J.S., Horstkotte D., Turpie A.G. Antithrombotic therapy in patients with mechanical and biological prosthetic heart valves. Chest 1998;114:S602-610.[Abstract/Free Full Text]
- Horstkotte D., Schulte H.D., Bircks W., Strauer B.E. Lower intensity anticoagulation therapy results in lower complication rates with the St. Jude Medical prosthesis. J Thorac Cardiovasc Surg 1994;107:1136-1145.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
J. M. Bernal, J. Lorca, D. Prieto-Salceda, I. Pulitani, A. Ponton, I. Garcia, and J. M. Revuelta
Performance at 10 years of the CarboMedics 'Top-Hat' valve. Postclamping time is a predictor of mortality
Eur. J. Cardiothorac. Surg.,
February 1, 2006;
29(2):
144 - 149.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Chambers, J. Roxburgh, C. Blauth, J. O'Riordan, F. Hodson, and H. Rimington
A randomized comparison of the MCRI On-X and CarboMedics Top Hat bileaflet mechanical replacement aortic valves: Early postoperative hemodynamic function and clinical events
J. Thorac. Cardiovasc. Surg.,
September 1, 2005;
130(3):
759 - 764.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
