Five-Year Follow-up After Heart Valve Replacement With the CarboMedics Bileaflet Prosthesis Suzanne M. Ro¨dler, MD, Anton Moritz, MD, Wolfgang Schreiner, MD, Adelheid End, MD, Peter Dubsky, MD, and Ernst Wolner, MD Departments of Cardiology and Cardio-Thoracic Surgery, University of Vienna, Vienna, Austria

Background. The CarboMedics valve is a relatively new, low-profile, bileaflet, mechanical prosthesis. The results of a prospective follow-up study after valve replacement with this prosthesis in a university hospital are presented. Methods. We implanted 640 CarboMedics prostheses in 583 patients in the aortic (n 5 359), mitral (n 5 167), or aortic and mitral positions (double valve replacement; n 5 57). Patient ages ranged from 11 to 81 years (mean age, 58 6 12.3 years). Results. Overall hospital mortality was 9.0%; however, when high-risk urgent cases were removed from the calculation, the operative mortality fell to 4.5%. Followup was 98% complete, comprising 2,027 patient-years for

a mean follow-up of 44 months (range, 6 to 72 months). Actuarial freedom from complications (linearized rates in parentheses) was as follows: late mortality, 85% 6 2.0% (2.3%/patient-year); thromboembolism, 92% 6 1.1% (1.6%/patient-year); anticoagulation-related hemorrhage, 87% 6 1.2% (2.8%/patient-year); prosthetic valve endocarditis, 98% 6 0.5% (0.1%/patient-year); and overall valverelated morbidity and mortality, 76% 6 2.1% (4.3%/ patient-year). Conclusions. The CarboMedics valve shows a low rate of valve-related complications comparable with other new mechanical heart valve prostheses. (Ann Thorac Surg 1997;63:1018 –25) © 1997 by The Society of Thoracic Surgeons

T

the severity of risk associated with age, left ventricular dysfunction, ischemic heart disease, or prior valve replacement, all patients operated on are included in this report. Within the cohort, 359 patients underwent isolated aortic valve replacement, 167 underwent isolated mitral valve replacement, and 57 underwent doublevalve (aortic and mitral) replacement. In this study 160 patients (27.4.%) had a previous cardiac operation; 176 patients (30%) had additional, associated procedures, including coronary artery bypass grafting (n 5 122), de Vega tricuspid annuloplasty (n 5 11), mitral reconstruction (n 5 9), thrombectomy (n 5 6), aortic annulus enlargement (n 5 4), ventricular aneurysmectomy (n 5 2), and other procedures (n 5 22). Patient data are shown in Table 1. Causes of valve diseases are presented in Figure 1. In patients without a clear history of rheumatic or other valve disease, the cause was classified as unknown. This unknown group comprises primarily individuals with calcified, degenerated valves. The distribution of valve sizes implanted in both the aortic and mitral position is presented in Figure 2.

he CarboMedics prosthetic heart valve (CarboMedics Inc, Austin, TX) is a relatively new, low-profile, bileaflet prosthesis in which the circular configuration and valve integrity are maintained by a titanium stiffening band. This design allows rotation of the valve within the sewing ring for optimal orientation of the prosthesis during implantation, and it protects against deformity with resultant leaflet binding or escape [1– 4]. The parts of the prostheses exposed to blood are made of pyrolitic carbon, and the sewing ring is carbon-coated to reduce thrombogenicity and tissue ingrowth. Additionally, its design is intended to improve the flow characteristics and reduction of turbulence. The moving leaflets are impregnated with tungsten to increase radiopacity. Clinical implantation of the CarboMedics prosthesis was started in 1986 on the European continent. This report details our experience with this valve from 1988 to 1994.

Patients and Methods This study was designed to prospectively follow up and report valve-related morbidity and mortality in 583 patients who had 640 CarboMedics valves implanted between 1988 and 1994, using guidelines established by Edmunds and associates [5]. There were 272 female patients and 311 male patients, whose ages ranged from 11 to 81 years (mean age, 58 6 12.3 years). Regardless of Accepted for publication Oct 25, 1996. Address reprint requests to Dr Wolner, Department of Cardiology, University of Vienna, Wahringergirtel 18 -20 1 090 Vienna, Austria.

© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

Operative Technique and Patient Management Standard cardiopulmonary bypass technique was employed. Until 1990, myocardial protection was achieved with moderate systemic hypothermia (28°C) and multiple-dose St. Thomas’ crystalloid cardioplegia. Since 1990, the combination of blood cardioplegia and topical cooling with iced saline solution has been our method of choice. Prostheses were inserted using 2-0 Ti-Cron pledget-armed mattress sutures. Postoperatively, the pa0003-4975/97/$17.00 PII S0003-4975(97)00174-4

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

1019

Table 1. Patient Characteristics Variable Patients (n) Mean age 6 SD (y)a Male/female Atrial fibrillation Systemic embolib Operative mortality Presence of CAD Single-vessel disease Double-vessel disease Triple-vessel disease Previous cardiac operation Closed valvulotomy Open valvulotomy Previous valve Others Additional procedures CABG Tricuspid valve repair Mitral valve repair Thrombectomy Annulus enlargement Aneurysmectomy Others

AVR

MVR

DVR

All

359 (61.6%) 58 6 12.3 (14 – 80) 1.6:1 55 (15.3%) 12 (3.3%) 33 (9.1%) 100 (27.8%) 36 (10%) 33 (9.2%) 31 (8.6%) 76 (27.8%) 22 (6.1%) 6 (1.7%) 26 (7.2%) 22 (6.1%) 111 (30.9%) 87 (24.2%) 1 (0.3%) 9 (2.5%) ... 1 (0.3%) ... 13 (3.6%)

167 (28.6%) 58 6 11.5 (11– 81) 1:1 106 (63.4%) 19 (11.3%) 15 (8.9%) 35 (20.9%) 14 (8.4%) 11 (6.6%) 10 (6%) 64 (38.3%) 18 (10.7%) 7 (4.1%) 30 (17.9%) 9 (5.3%) 51 (30.5%) 26 (15.6%) 8 (4.8%) ... 5 (3%) 3 (1.8%) 1 (0.5%) 8 (4.8%)

57 (9.8%) 57 6 6.9 (11–72) 1:1.2 36 (63.1%) 6 (10.5%) 5 (8.7%) 9 (15.7%) 2 (3.5%) 1 (1.8%) 6 (10.5%) 20 (35%) 7 (12.2%) 2 (3.5%) 8 (1.4%) 3 (5.2%) 14 (24.5%) 9 (15.8%) 2 (3.5%) ... 1 (1.7%) ... 1 (1.7%) 1 (1.7%)

583 58 6 12.3 (11– 82) 1.2:1 197 (33.8%) 37 (6.3%) 53 (9%) 144 (24.6%) 52 (8.9%) 45 (7.7%) 47 (8.0%) 160 (27.4%) 47 (8.0%) 15 (2.5%) 64 (10.9%) 34 (5.8%) 176 (30.1%) 122 (20.9%) 11 (1.9%) 9 (1.5%) 6 (1%) 4 (0.7%) 2 (0.4%) 22 (3.7%)

a

Range is in parentheses.

b

Percentage of patients with a preoperative history of systemic embolism is presented as systemic emboli.

AVR 5 aortic valve replacement; CABG 5 coronary artery bypass grafting; replacement; MVR 5 mitral valve replacement; SD 5 standard deviation.

tients were given 5,000 units of subcutaneous heparin three times a day; permanent anticoagulation with phenprocoumon was started on the fourth postoperative day. The target level of anticoagulation was a Thrombotest of 5% to 15%, which equals an international normalized ratio of 2.1 to 4.8 [6].

CAD 5 coronary artery disease;

DVR 5 double-valve

a detailed written questionnaire, which was signed by the patient’s primary physician. After the first year, they were invited to our outpatient department, where physical examination, laboratory tests, electrocardiography, and echocardiography were done. Patients with compli-

Follow-up All patients undergoing valve operations were contacted 6 months postoperatively and every year thereafter with

Fig 1. Causes of valve disease. Sixteen patients (2.7%) had been reoperated on because of paravalvular leakage or endocarditis of mechanical prostheses; 48 patients (8.2%) had previous bioprostheses.

Fig 2. Distribution of valve sizes implanted in both the aortic and mitral position. Most frequently used sizes were 27 mm and 29 mm in the mitral and 21 mm and 23 mm in the aortic position.

1020

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

cations were examined in our outpatient department. Information was available for 572 of the 583 patients, a 98% complete follow-up. The cumulative patient follow-up was 2,027 patient-years, with a mean observation time of 44 months and a range of 6 to 72 months.

Statistical Analysis Actuarial estimates of morbid events were calculated by the Kaplan-Meier survival analysis method (product limit). Confidence limits of 95% for the survival function were calculated according to the Greenwood formula [7], that is, the incidence of complications for 12 months was determined as the conditional probability of an event. The survival rates of an age- and sex-matched normal population were calculated from Austrian life tables ¨ sterreichisches Statistisches Zentralamt 1989). The (O Wilcoxon log rank test for equality of survival curves was used to compare the groups.

Results Operative Mortality Operative mortality and cause of death are summarized in Table 2. Fifty-three patients died within 30 days after the operation or during the same hospital stay, resulting in an operative mortality of 9%. As previously noted, this group of patients contained many in the high-risk category. For example, of these 53 patients 24 (45%) underwent additional cardiac procedures and 18 (34%) had had previous cardiac procedures. When unusually high risk, urgent patients were separated out (for example, patients taken to the operating room in uncontrolled hypotension, those with respiratory insufficiency unable to be weaned from the respirator, those experiencing acute myocardial infarction with ruptured papillary muscle, and patients undergoing repeat cardiac operations with an ejection fraction less than 0.20 in whom additional procedures had to be performed in addition to the valve replacement), the operative mortality fell to 4.5%. Sixteen patients died of low cardiac output syndrome, 13 sustained lethal myocardial infarctions, 11 died of sepsis, and in 9 bleeding (including three ventricular ruptures) was the major cause of death. Four patients died of uncontrollable arrhythmias.

Table 2. Operative Mortalitya Variable

AVR

MVR

Patients 33 (9.1%) 15 (8.9%) Low cardiac output 9 (27.2%) 7 (46.6%) Myocardial infarction 8 (24.3%) 3 (20%) Sepsis 8 (24.3%) 3 (20%) Hemorrhage 7 (21.2%) 1 (6.7%) Arrhythmia 1 (3%) 1 (6.7%) a

DVR

All

5 (8.7%) ... 2 (40%) ... 1 (20%) 2 (40%)

53 (9%) 16 (30.2%) 13 (24.5%) 11 (20.8%) 9 (16.9%) 4 (7.6%)

Operative mortality 5 death less than 30 days after operation.

AVR 5 aortic valve replacement; MVR 5 mitral valve replacement.

DVR 5 double-valve replacement;

Table 3. Late Deaths Cause Heart failure Sudden/unknown Arrhythmia Myocardial infarction Noncardiac Valve embolism Cerebral hemorrhage Other cardiac All

Non–Valve-Related Valve-Related All 11 ... 2 1 16 ... ... ... 30

... 16 1 2 ... 4 9 2 34

11 16 3 3 16 4 9 2 64

Late Deaths Sixty-four (10.9%) patients died during the follow-up period, most of progressive heart failure, suddenly of unknown causes, or of noncardiac causes (Table 3). Five years after operation, survival was 85% 6 2% excluding operative mortality and 81% 6 2.2% including operative mortality (Fig 3). The survival rate of an age- and sexmatched population according to Austrian life-tables was estimated to be 96% after 60 months (Fig 4). The linearized rate of late deaths after operation was 2.3%/patientyear: 2.3% for aortic valve replacement, 1.4% for mitral valve replacement, and 2.6% for double-valve replacement (Table 4; Fig 5). Survival according to valve site, excluding operative mortality, was 87% 6 1.2% for aortic valve replacement, 89% 6 1.2% for mitral valve replacement, and 82% 6 2.7% for double-valve replacement. Thirty-four late deaths were valve-related (1.7%/patientyear.)

Structural Valve Failure and Dysfunction There was no instance of structural valve failure.

Nonstructural Valve Dysfunction Twelve patients exhibited evidence of nonstructural valve dysfunction (0.7%/patient-year), leaving 98% 6 0.1% event-free. Among these, 2 patients sustained obstruction of the valve, 1 due to massive tissue overgrowth and 1 due to pannus formation.

Thromboembolism Thromboembolic events were observed in 32 patients, leaving 92% 6 1.1% event-free (Fig 6). (The linearized rate is 1.6%/patient year; see Table 4.) Event-free rates were 94% 6 1.4% (1.4%/patient-year) for aortic valve replacement, 89% 6 2.3% (2.1%/patient-year) for mitral valve replacement, and 95% 6 2.0% (1.1%/patient-year) for double-valve replacement. Four embolic episodes were lethal (two myocardial infarctions and two cerebral infarctions, one of the latter in relation to prosthetic valve endocarditis). Seven embolic episodes resulted in permanent neurologic deficits.

Valve Thrombosis Late valve thrombosis developed in 2 patients with mitral prostheses. In both cases the organized thrombus was

Ann Thorac Surg 1997;63:1018 –25

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

1021

Fig 3. Survival with and without operative mortality after 5 years of follow-up. 3.1%/pt/yr refers to the linearized incidence of occurrence.

adherent to fibrous tissue overgrowth and the prothrombin times were suboptimal. Each was successfully reoperated on.

were event-free after 60 months (see Fig 6). Nine events were lethal.

Anticoagulant-Related Hemorrhage

A hemodynamically significant paravalvular leak developed in 7 patients (0.34%/patient-year), leaving 97% 6 0.7% event-free. Five patients underwent reoperation, without operative mortality.

Sixty-two events of anticoagulation-related hemorrhage were observed in 56 patients with a linearized incidence of 2.8%/patient-year (see Table 4). Of these, 87% 6 1.2%

Paravalvular Leak

Fig 4. Survival curves. Upper curve shows the survival probability of age- and sex-matched Austrian population. Lower curve shows survival 5 years after heart operation. 2.3%/pt/yr refers to the linearized incidence of occurrence.

1022

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

Table 4. Morbidity and Mortalitya AVR

MVR

DVR

All

Event

Event-Free

Incidence

Event-Free

Incidence

Event-Free

Incidence

Event-Free

Incidence

Late deaths Thromboembolism Anticoagulant-related hemorrhage Paravalvular leak Prosthetic valve endocarditis Valve-related mortality Permanent valverelated impairment Survival including perioperative mortality Valve-related morbidity and mortality

87% 6 1.2% 94% 6 1.4% 87% 6 1.6%

2.3% 1.4% 1.9%

89% 6 1.2% 89% 6 2.3% 87% 6 1.6%

1.4% 2.1% 1.6%

82% 6 2.7% 95% 6 2.0% 86% 6 1.1%

2.6% 1.1% 2.1%

85% 6 2% 92% 6 1.1% 87% 6 1.2%

2.3% 1.6% 2.8%

... ...

... ...

... ...

... ...

... ...

... ...

97% 6 0.7% 98% 6 0.5%

0.34% 0.1%

...

...

...

...

...

...

91% 6 1.5%

1.79%

...

...

...

...

...

...

97% 6 0.9%

0.5%

...

...

...

...

...

...

81% 6 2%

4.5%

76% 6 2.1%

4.3%

a

75% 6 1%

4.7%

77% 6 2%

3.6%

73% 6 1%

5.3%

Actuarial rates are presented with 95% confidence limits.

AVR 5 aortic valve replacement;

DVR 5 double-valve replacement;

MVR 5 mitral valve replacement.

Endocarditis

Functional Improvement

Prosthetic valve endocarditis developed in 7 patients, leaving 98% 6 0.5% free from this event after 60 months. One patient died of this complication, and 3 underwent reoperation.

New York Heart Association functional class was improved at 1 year postoperatively and at 4 years postoperatively (Fig 7).

Hemolysis

Paravalvular leak was the indication for reoperation five times, prosthetic valve endocarditis the indication three times, and valve thrombosis two times (Table 6). Another

Hemolysis that did not produce clinically significant anemia was found in all patients (Table 5).

Reoperation

Fig 5. Survival according to valve site. 2.6%/pt/yr refers to the linearized incidence of occurrence.

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

1023

Fig 6. Event-free and yearly incidence of thromboembolic and anticoagulant-related hemorrhage. 1.6%/pt/yr refers to the linearized incidence of occurrence.

ported by de Luca and associates [10] and Subotic and colleagues [11] in 1990. This report details our experience from 1988 to 1994. The relatively high perioperative mortality (9%) and low 5-year survival rates in our study reflect the inclusion of the previously described high-risk patients, in addition to those in the older age groups and those with severe ischemic heart disease, advanced left ventricular dysfunction, serious preoperative ventricular arrhythmias, and previous cardiac operation. In a similar, high-risk group of patients, Czer and associates [8] reported an operative mortality of 8.3%, and Hammermeister and colleagues [12] reported an operative mortality of 11%. The leading cause of hospital mortality in our study, as in other series, was perioperative low output syndrome and myocardial infarction [8, 12, 13]. Patients with previous cardiac operations constitute an increasing percentage of all valve operations performed at this institution. In an earlier report describing our experience with the Edwards-Duromedics prosthesis, most reoperations were performed for prosthetic

valve had to be changed because of pannus formation causing abnormal leaflet motion, and 1 patient had to undergo subsequent coronary artery bypass grafting. All 12 patients survived reoperation; 1 patient died 2 years after reoperation for noncardiac reasons.

Comment Although many advances have been made in the design, construction, and materials used in the production of prosthetic heart valves, prosthesis-related complications are currently reported in up to 50% of heart valve recipients 10 years after the operation [8]. As a result, surgical treatment of valvular heart disease remains a field of constant clinical and experimental research. Because valve performance is contingent on many patientrelated factors, the continuous evaluation of quality of life in mechanical valve recipients is important [7, 9]. The CarboMedics cardiac prosthesis was introduced into clinical use in 1986. Preliminary results were re-

Table 5. Intravascular Hemolysis in Relation to Valve Positiona AVR Variable Hb (g/dL) LDH (IU/L) Hbf (mg/dL) Hpt (mg/dL) a

MVR

DVR

All

Preop

1 Year

Preop

1 Year

Preop

1 Year

Preop

1 Year

13.9 6 2.1 232 6 2.5 3.3 6 2.5 79 6 27

14 6 7.4 263 6 3.0 3.6 6 3.5 55 6 17

13.8 6 2.2 261 6 113 4.8 6 6.1 68 6 19

13.6 6 2.3 316 6 115 4 6 3.1 54 6 20.4

13.5 6 7.2 263 6 98 3.9 6 2 66 6 26

13.3 6 2.6 383 6 180 4.9 6 4.2 44.8 6 13

13.8 6 2.1 244 6 117 3.8 6 0.1 69.4 6 23

13.8 6 2.1 289 6 110 3.9 6 3.5 54 6 15

Samples were taken preoperatively and at 1-year follow-up.

AVR 5 aortic valve replacement; DVR 5 double-valve replacement; Hb 5 hemoglobin (normal range, 12 to 16 mg/dl); Hbf 5 plasma hemoglobin (normal range, 0 to 4 mg/dL); Hpt 5 haptoglobin (normal range, 40 to 250 mg/dL); LDH 5 lactic dehydrogenase (normal range, 80 to 240 IU/L); MVR 5 mitral valve replacement.

1024

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

Fig 7. New York Heart Association (NYHA) functional class status before and 1 and 4 years after operation. There was no significant change in the profile of functional class after the initial improvement. The vertical axis depicts the percentage of patients in each functional class during the specific time periods.

dysfunction. In that report, we reported a reoperation rate of 15% [14]. The incidence of cardiac reoperations in the present series was 27% (see Table 1). The most frequent valve-related complication was hemorrhage, which occurred at a rate of 2.8%/patientyear. Nine events (15%) were fatal, and these deaths accounted for 26% of all valve-related deaths (see Table 3). Thus, warfarin-related hemorrhage has been the major contributor to valve-related morbidity and mortality in our series. Recognizing that anticoagulation regimens based on current available prothrombin assays may lead to excessive anticoagulation, we are considering lowering the upper limit of our target prothrombin time ratio. Spontaneous fluctuations in the prothrombin time are a well-described occurrence in patients maintained on stable dosages of warfarin [2, 15]. Unfortunately, these fluctuations may cause significant shortfalls or excesses in anticoagulation that, in turn, increase the risk of

Table 6. Reoperation Reason Paravalvular leak Thrombosis Endocarditis CABG Pannus Noncardiac events Total

Patients

Survivors

Deaths

5 2 3 1 1 1

5 2 3 1 1 0

0 0 0 0 0 1

13

12

1

CABG 5 coronary artery bypass grafting.

adverse clinical events such as thrombosis, embolism, or hemorrhage. Butchart and associates [15] have shown the safety of decreasing the warfarin intensity with resultant significantly lower bleeding complications, and minimal, if any, increase in thrombotic problems. They reported 5-year embolism-free rates of 92%, 84%, and 83% for aortic, mitral, and double valves. Although warfarin is considered to be the most effective drug for reducing intravascular thrombogenesis, the primary thrombogenic mechanism starts from the adhesion and aggregation of platelets with endothelium mediated by the von Willebrand factor and fibrinogen [16]. Hayashi and colleagues [17], reporting on their experience with the St. Jude prosthesis, noted an increased survival, lower incidence of cerebrovascular stroke, and better quality of life using a combined warfarin plus antiplatelet therapy. The lower incidence of hemorrhage in this series was probably due to a less intensive warfarin treatment regimen (a Thrombotest level of 10% to 20%, which is equivalent to a prothrombin time international normalized ratio of 1.8 to 2.8) and the use of an antiplatelet agent. Thromboembolism is one of the major risk factors for patients with prosthetic heart valves, and therefore, it is one of the most specific measures of valve performance. The linearized rate of thromboembolic complications in our series was 1.6%/patient-year or 92% 6 1.1% eventfree. This incidence can be favorably compared with 2.0%/patient-year reported by Czer and colleagues for the St. Jude valve [8] and 2.3% 6 0.3%/patient-year by Keenan and associates for the Medtronic Hall valve [18]. Our more successful experience was probably due to the improved hemodynamic design that resulted in a better flow-velocity profile, less turbulence, and a reduction of stagnant areas, all of which are known to affect the thrombogenicity of the prosthesis. In our study, the 2 patients with valve thromboses had atrial fibrillation, left atrial enlargement, and poor anticoagulant compliance. Both patients were reoperated on successfully. The fact that the orifice (with the valve leaflets) can be rotated in situ was found to be a distinct advantage; potential interference between the valve leaflets and surrounding tissue could thus be avoided in a technically simple manner. Data published in the available literature [4] have now shown that there are no secondary axial changes and that there will be no secondary interference with the surrounding tissue. Endocarditis, although a significant determinant of fatal valve failure and reoperation, did not significantly influence the likelihood of thromboembolism, valve failure, or late survival. As for durability, no structural deterioration such as stress fracture or leaflet escape or binding occurred. As we previously reported [19], echocardiographic examinations of the aortic valves showed mean peak velocity of 2.6 6 0.4 m/s, and calculated instantaneous peak pressure gradient ranged from 11 to 58 mm Hg (mean, 28.1 6 10.3 mm Hg). These results presented earlier compare favorably with those of other mechanical valves of similar design [20, 21]. When comparing patients with different valve sizes, peak velocity and pressure gradients

¨ DLER ET AL RO FIVE-YEAR FOLLOW-UP OF CARBOMEDICS VALVE

Ann Thorac Surg 1997;63:1018 –25

are inversely related to annulus diameter. As in all small aortic roots, the pressure gradient increased with diminishing root size. In patients with these small aortic roots, consideration should be given to the use of the supraannular valve to obtain the greatest flow area for the smallest outside diameter. We have previously reported on our favorable experience using the CarboMedics Top Hat valve in these cases [22]. In conclusion, at 5 years of careful follow-up, the CarboMedics bileaflet prosthesis demonstrates no structural deterioration, a low incidence of complications, and good hemodynamic performance. Additionally, it has the features of rotatability and protection against deformity of the valve housing, thus avoiding leaflet escape and binding of the leaflets. The continued use of this valve is recommended.

10.

11.

12.

13.

14. 15.

References 1. Copeland JG III, Sethi GK, North American team of clinical investigators for the CarboMedics prosthetic heart valve. Four year experience with the CarboMedics valve: the North American experience. Ann Thorac Surg 1994;58:603– 8. 2. Von Segesser LK, Schwaller A, Amann FW, et al. First late results of aortic valve replacement with CarboMedics bileaflet valve. Helv Chir Acta 1993;60:457– 60. 3. Burnett C. The clinical report. CarboMedics 1995;7:1– 8. 4. Chambers JB, Deverall PB. A study of the stability of the orientation in the CarboMedics bileaflet heart valve prosthesis. Eur J Cardiothorac Surg 1992;6:442–5. 5. Edmunds LH Jr, Clark RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1988;46:257–9. 6. Altmann R, Rouvier J, Gurfinkel E, et al. Comparison of two levels of anticoagulant therapy in patients with substitute heart valves. J Thorac Cardiovasc Surg 1991;101:427–31. 7. The lifetable procedure. In: SAS users guide to statistics. Version 5. Cary, NC: SAS Institute Inc, 1985:529–57. 8. Czer LSC, Chaux A, Matloff JM, et al. Ten-year experience with the St. Jude valve for primary valve replacement. J Thorac Cardiovasc Surg 1990;100:44–55. 9. Mitchell RS, Miller DC, Stinson EB, et al. Significant patient-

16. 17.

18. 19.

20. 21. 22.

1025

related determinants of prosthetic valve performance. J Thorac Cardiovasc Surg 1986;91:807–17. De Luca L, Vitale N, Giannolo B, Cafrella G, Piazza L, Cotrufo M. Mid-term follow-up after heart valve replacement with CarboMedics bileaflet prostheses. J Thorac Cardiovasc Surg 1993;106:1159– 65. Subotic D, Petrovic P. Boskovic D, et al. Clinical and functional evaluation of the CarboMedics prosthetic heart valve in mitral position: preliminary report. J Cardiovasc Surg 1990;31:509–11. Hammermeister KE, Sethi GK, Henderson WG, et al. A comparison of outcomes in men 11 years after heart valve replacement with a mechanical valve or bioprosthesis. N Engl J Med 1993;328:1289–96. Lund O, Knudsen MA, Pilegaard HK, Magnussen K, Nielsen TT. Long-term performance of Starr-Edwards Silastic ball valves and St. Jude Medical bileaflet valves. Eur Heart J 1990; 11:108–19. Moritz A, Klepetko W, Ro¨dler S, et al. Six-year follow-up after heart valve replacement with the Edwards-Duromedics bileaflet prosthesis. Eur J Cardiothorac Surg 1993;7:88–90. Butchart EG, Lewis PA, Grunkemeier GL, Kulatilake N, Breckinridge IM. Low risk of thrombosis and serious embolic events desite low-intensity anticoagulation. Experience with 1,004 Medtronic-Hall valves. Circulation 1988;78:166–77. Ruggeri ZM, Zimmemann TS. Von Willebrand factor and von Willebrand disease. Blood 1987;70:895–904. Hayashi J-I, Nakazawa S, Oguma F, Miyamura H, Eguchi S. Combined warfarin and antiplatelet therapy after St. Jude Medical valve replacement for mitral valve disease. J Am Coll Cardiol 1994;23:672–7. Keenan RJ, Armitage JM, Trento A, et al. Clinical experience with Medtronic Hall valve prosthesis. Ann Thorac Surg 1990; 50:758– 62. Ren JF, Chandrasekaran K, Mintz GS, et al. Effect of depressed left ventricular function on hemodynamics of normal St. Jude Medical prosthesis in aortic valve position. Am J Cardiol 1989;65:1004 –9. Globits S, Ro¨dler S, Mayr H, et al. Doppler sonographic evaluation of the CarboMedics bileaflet valve prosthesis: one-year experience. J Card Surg 1992;7:9–16. Panidis IP, Ross J, Mintz GS. Normal and abnormal prosthetic valve function as assessed by Doppler echocardiography. J Am Coll Cardiol 1986;8:317–26. Ro¨dler S, Moritz A, Wutte M, Hoda R, Wolner E. The CarboMedics “top hat” supraannular prosthesis in the small aortic root. J Card Surg 1995;10:198 –204.