Ann Thorac Surg 2002;74:665-670
© 2002 The Society of Thoracic Surgeons
Original article: cardiovascular
Results of valve replacement with omniscience mechanical prostheses
Matthew S. Edwards, MDa,
Gregory B. Russell, MSb,
Angela F. Edwards, MDc,
John W. Hammon, Jr, MDa,
A. Robert Cordell, MDa,
Neal D. Kon, MDa*
a Department of Cardiothoracic Surgery, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA
b Department of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA
c Department of Anesthesiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA
Accepted for publication April 21, 2002.
* Address reprint requests to Dr Kon, Department of Cardiothoracic Surgery, Wake Forest University School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157, USA
e-mail: nkon{at}wfubmc.edu
 |
Abstract
|
|---|
Background. The Omniscience mechanical valve has been the subject of multiple clinical investigations with variable results, including reports of high complication and reoperation rates.
Methods. Records of all patients who received Omniscience valves were reviewed, and follow-up interviews were conducted to determine the incidence of valve-related morbidity, mortality, and functional results. Incidence of complications was expressed as events per 100 patient-years follow-up. Survival and freedom from valve-related complications and mortality were calculated using a product limit method.
Results. Between 1984 and 1988, 192 patients received 213 Omniscience valves [93 mitral (M), 79 aortic (A), and 20 multiple (D) valve replacements]. Perioperative mortality was 9%. The incidence of major valve-related morbidity was as follows: thrombosis, 1.30 M, 0.17 A, 0.72 D; endocarditis, 0.48 M, 0.18 A, 0 D; hemorrhagic, 4.67 M, 2.84 A, 5.00 D; embolic, 2.90 M, 2.27 A, 1.57 D; nonstructural dysfunction, 1.66 M, 1.08 A, 2.27 D; reoperation, 4.02 M, 1.99 A, 6.48 D. All explanted valves (n = 43) were examined, and 40% (n = 17) were found to have limited disc excursion in the absence of thrombus. Freedom from valve-related morbidity, mortality, or reoperation at 10 years was 22% for mitral, 39% for aortic, and 17% for multivalve replacements. At follow-up, only 73% of patients were New York Heart Association class I or II. Five- and 10-year estimated survivals were 72% and 55% for M, 80% and 51% for A, and 65% and 50% for D replacements.
Conclusions. Use of the Omniscience valve provided poor functional improvement and a significant incidence of valve-related complications, including the need for reoperation.
|
Introduction
|
|---|
Replacement of diseased cardiac valves with mechanical prostheses has become an accepted and effective treatment for end-stage valvular disorders. Furthermore, it has become apparent that clinical results vary with the prosthesis used, as the ideal replacement valve has yet to be designed. The Omniscience mechanical cardiac valve (Medical CV Inc, Inner Grove Heights, MN) is a low-profile, central flow prosthesis using a single tilting concave-convex disc of pyrolytic carbon in a single-piece titanium housing within a polytetrafluoroethylene sewing ring [1, 2]. There have been two versions of this prosthesis, with the most current version being available since 1982. Both versions have been the subject of multiple clinical investigations with variable results. Early reports claimed high rates of thrombosis and reoperation [3–5], whereas later reports claimed much improved results [6–11]. It was postulated that the poor results of the earlier series may have been related to remediated structural issues of the earlier version of the valve, suboptimal valve orientation at operation, or improper valve sizing [9]. Only one of these reports contained follow-up exceeding 7 years [11], reporting excellent long-term results with the Omniscience prosthesis in both aortic and mitral positions. The Omniscience valve was the prosthesis of choice at our institution from 1984 to 1988, and this report summarizes our clinical experience with this valve.
|
Patients and methods
|
|---|
Between February 1985 and July 1988, 192 patients received 213 Omniscience prosthetic valves. All valves were of the second-generation, post-1982 model. Operations were performed with cardiopulmonary bypass using cold blood cardioplegia and topical cooling for myocardial protection. Each valve was sized and positioned to ensure full, free, unimpeded motion of the occluder disc. Interrupted, horizontal mattress sutures utilizing felt pledgets were used in all cases either above, within, or below the valve annulus according to the discretion of the operating surgeon. All patients were given oral warfarin anticoagulation on postoperative day two (unless oral intake was contraindicated) and maintained on chronic anticoagulation by their primary care physicians. Recommended anticoagulation levels were an international normalized ratio of 2.5 to 3.0 for aortic replacements and 3.0 to 3.5 for mitral and multivalve replacements.
Patient records were examined, and follow-up interviews with each patient, a primary relative, or the primary care physician were performed to determine the incidence of major valve-related morbidity and mortality, survival, and functional results. The interval encompassing data collection was 2 months. Complications were classified according to the Society of Thoracic Surgeons-American Association for Thoracic Surgery guidelines for reporting morbidity and mortality of operated valves [12]. Valves found to have limited excursion of the occluder disc at reoperation (for any reason), in the absence of gross thrombus or infection, were so classified. Tests of association were performed using 
2 tests for categorical measures and the log-rank approximation of the 2 in testing for differences among strata in survival data. Complication rates were expressed using composite rates (events per 100 patient-years) with confidence intervals (CI) calculated using an exponential distribution and the resulting variance estimates. The distributions of time until complication or death were portrayed with Kaplan-Meier curves. For overall survival analyses the date of death or last known follow-up was used. For valve-related morbidity and mortality analyses, the earliest date to event or date of death was used for calculations. Patients surviving without complication were censored at the last known date of follow-up or death.
|
Results
|
|---|
Patient demographics
Table 1
contains complete demographics for each group. Ninety-four mitral (M), 79 aortic (A), and 20 multivalve (D) replacements were performed in 193 operations on 192 patients. Thirty-day and long-term follow-up were 100% and 90% complete, respectively. Eighteen patients were lost to follow-up after 30 days. One patient underwent two separate mitral valve replacements; only the first was considered in analyses. Total patient follow-up was 1,339 patient-years with a median follow-up of 114 months. Concomitant cardiac procedures were performed in 66 patients (Table 2)
and prior valvular operations had been performed in 47 patients (Table 3).
In the multivalve group there were 18 mitral and aortic, 1 mitral, aortic, and tricuspid, and 1 aortic and tricuspid replacements.
Early and late mortality
Early or operative mortality was defined as all deaths occurring within 30 days of operation, regardless of cause, or within the same hospitalization, regardless of interval. Overall early mortality was 9% (M 9%, A 9%, D 14%). Valve-related deaths were classified according to Society of Thoracic Surgeons-American Association for Thoracic Surgery guidelines as all deaths occurring secondary to valve-related complications (eg, endocarditis, anticoagulant-related hemorrhage) and all sudden, unexplained deaths. There were 38 complication-related deaths (19 M, 17 A, 2 D) and 13 sudden deaths (3 M and 10 A) for a total of 51 valve-related deaths. Estimated survival curves are depicted in Figure 1.
Five- and 10-year estimated survival rates were 72% and 55% for mitral valve patients, 80% and 51% for aortic valve patients, and 65% and 50% for multivalve patients.
Functional results
New York Heart Association (NYHA) classes were assigned to all patients (n = 63) alive at follow-up. These data are summarized according to operative group in Table 4.
Overall, 73% of patients were NYHA class I or II at follow-up. No significant differences were observed between postoperative NYHA class III or IV patients and NYHA class I or II patients in terms of preoperative NYHA functional status, implant size, surgeon, or preoperative ventricular performance.
Valve-related complications
Thrombosis
Thrombus occurred within the Omniscience valve in 10 patients (8 M, 1 A, 1 D) in the absence of infection. Five patients were inadequately anticoagulated (prothrombin time < 18.0 s or international normalized ratio < 2.5) at the time of their thrombosis, and 1 patient suffering a fatal thrombotic event had unknown anticoagulation status. Nine patients underwent reoperation with valve explantation and replacement; 6 survived. Composite rates for thrombosis were M, 1.30 (95% CI, 0.40 to 2.20); A, 0.17 (95% CI, 0 to 0.52); and D, 0.72 (95% CI, 0 to 2.14).
Endocarditis
Infections involving the Omniscience valve were exceedingly rare in our experience. Endocarditis occurred in 4 patients (3 M and 1 A). One patient underwent reoperation with valve explant and a fatal outcome. Overall, mortality was 75%. The lone survivor was treated with antibiotics alone. Composite rates for endocarditis were M, 0.48 (95% CI, 0 to 1.03); A, 0.18 (95% CI, 0 to 0.52); and D, 0.
Anticoagulant-related hemorrhage
Hemorrhage secondary to systemic anticoagulation was recognized as a complication only if hospital admission, operation, or transfusion was required or if permanent disability or death resulted. There were 45 episodes of anticoagulant-related hemorrhage (25 M, 14 A, 6 D). Of these, 17 episodes occurred in the setting of excessive anticoagulation (defined as prothrombin time > 22.0 s or international normalized ratio > 3.5). Composite rates of anticoagulant-related hemorrhage were M, 4.67 (95% CI, 2.84 to 6.50); A, 2.84 (95% CI, 1.35 to 4.33); and D, 5.00 (95% CI, 1.00 to 9.00).
Thromboembolic
All cases of transient ischemic attack and all cases of systemic embolism leading to permanent disability, operation, or death (in the absence of infection or macroscopic thrombosis) were recognized as thromboembolic complications. Thrombosis of the valve without systemic embolism was not included. There were 30 patient events (16 M, 12 A, 2 D) as defined above with 9 occurring in the setting of inadequate anticoagulation. Composite rates for thromboembolic events were M, 2.90 (95% CI, 1.48 to 4.31); A, 2.27 (95% CI, 0.99 to 3.56); and D, 1.57 (95% CI, 0 to 3.74).
Nonstructural dysfunction
Nonstructural problems resulting in functional stenosis, insufficiency, or hemolysis were reported. There were 19 recognized events (10 M, 6 A, 3 D) in the absence of endocarditis or valve thrombosis. Specific recognized causes were as follows: 13 perivalvular leaks, 2 cases of hemolysis, 1 chronic dissection, and 3 cases of tissue ingrowth. Reoperation with valve explantation and replacement was required in 15 patients. Composite rates of nonstructural dysfunction were M, 1.66 (95% CI, 0.63 to 2.68); A, 1.08 (95% CI, 0.22 to 1.95); and D, 2.27 (95% CI, 0 to 4.84).
Limited disc excursion
Limited excursion of the occluder disc in vivo, in the absence of any gross thrombus or infection, was found in 17 of 43 patients at reoperation. All reoperations were performed for progressive clinical symptoms of cardiac failure in the presence of imaging studies that suggested valve dysfunction with secondary functional stenosis or regurgitation. In all cases the occluder had a maximal opening angle of less than 50 degrees in vivo with no other identifiable abnormality such as thrombus, tissue ingrowth, or suboptimal orientation. Valves were examined in vivo and ex vivo after removal from the annulus and were found to stick at angles of less than 50 degrees. After thoroughly cleaning the prosthesis with saline solution there was no apparent limitation of valve opening. Subsequently, two of the explanted valves were then evaluated in vitro by an independent scientific firm revealing that the structural wear characteristics of each valve indicated incomplete valve opening throughout the in vivo period of function.
We have classified these events as limited disc excursion in our database. Composite rates of limited excursion were M, 1.50 (95% CI, 0.52 to 2.49); A, 0.70 (95% CI, 0.01 to 1.39); and D, 3.07 (95% CI, 0.06 to 6.08).
Reoperation/replacement
All cases of reoperation to repair, alter, or replace Omniscience valves were reported. There were 43 cases requiring reoperation (24 M, 11 A, 8 D); specific reasons are summarized in Table 5. Composite rates of reoperation were M, 4.02 (95% CI, 2.38 to 5.67); A, 1.99 (95% CI, 0.81 to 3.17); and D, 6.48 (95% CI, 1.99 to 10.97). Figure 2
depicts the 10-year freedom from reoperation. There were no significant differences in the rates of reoperation among operating surgeons.
Freedom from complications
Freedom from all major valve-related morbidity and mortality were calculated using a product limit method. Figure 3
depicts the 10-year freedom from all valve-related morbidity and mortality. Estimated freedoms were as follows: M, 22%; A, 39%; and D, 17%.
|
Comment
|
|---|
Clinical results after cardiac valve replacement are determined by patient survival, symptomatic improvement, and freedom from valve-related complications and death. The Omniscience prosthetic valve was designed to minimize transvalvular energy loss in an attempt to improve hemodynamics and clinical outcomes [2]. Early series with Omniscience valves reported high incidences of valve thrombosis and reoperation [3–5]. Later series reported excellent clinical results and low complication rates [6–9, 11]. Among these, only one study, by Teijeira [11], contained long-term follow-up exceeding 10 years. In that report, Teijeira observed excellent functional results and freedoms from valve-related morbidity and mortality in a large patient group. With the exception of a trend toward better preoperative functional status (NYHA class) in our study population, the patients and clinical techniques studied were extremely similar. Our long-term results, however, were much less favorable.
All Omniscience valves reported in this series were carefully placed to ensure full, unimpeded motion of the occluder disc. In the vast majority of cases, implantation included supraannular suture placement for valves in the mitral position and infraannular suture placement for valves in the aortic position. Early and in-hospital mortality occurred in 9% of our patients, which is comparable to other contemporary reports [3–9]. Estimated survival at 5 and 10 years was 80% and 51% in the aortic group, 72% and 55% in the mitral group, and 65% and 50% in the multiple valve operative group. These survival results are comparable to 5- and 10-year actuarial survivals reported using other mechanical prostheses [13–19]. Functional improvement results, however, were disappointing. At follow-up, only 73% of patients were in NYHA class I or II, which compares poorly with reports of 95% or greater with other mechanical prostheses [14, 17] and with the only other long-term Omniscience experience [11].
Complication rates were also disappointing. Although the composite rates of most individual complications were within or slightly above the ranges of historic controls reported by Akins [20], the aggregate risk of complications was extremely high. Ten-year estimated freedom from valve-related morbidity and mortality was only 39% for aortic, 22% for mitral, and 17% for multivalve replacements. These rates compare poorly with 10-year estimated freedoms in excess of 70% in reports using other mechanical prostheses [14, 16, 19] and our own intrainstitutional complication rates with other prostheses (unpublished data). Rates of reoperation, in particular, were unacceptably high. Ten-year freedoms from reoperation were 76% for aortic, 61% for mitral, and 44% for multivalve replacements, compared with historic freedoms exceeding 90% [20]. Furthermore, it is important to note that follow-up interviews were performed only once, potentially allowing for patient omission of remote complications. These omissions could have artificially lowered the individual and overall rates of valve-related morbidity.
Several authors have reported diminished opening angles [5, 8] with elevated pressure gradients. Forty percent of explanted valves in this series, in the absence of any thrombus, were found to have limited excursion of the occluder disc in vivo with opening angles of less than 50 degrees. Limited excursion, per se, was never a specific indication for reoperation; rather it was an in vivo finding at the time of reoperation and valve removal for clinically progressive cardiac failure. Furthermore, all of these valves were carefully examined at the time of reoperation and no technical problems with the original valve placement were appreciated. We speculate that the diminished opening angles, and their attendant relative obstruction to flow, may have created suboptimal hemodynamics. Such impedance may have led to diminished functional improvement and may have contributed to the development of such complications as thrombosis and thromboembolism. Furthermore, the Omniscience valve by design minimizes diastolic regurgitation [2]. This lack of regurgitation limits valve washing, which may have predisposed to thrombus formation, both microscopic and macroscopic, within the valve [3, 4]. This may have been especially true for the pivot points of the occluder disc, where regurgitation is the least. In all cases of thrombosis in our series, the clot involved the pivot points for the disc.
In conclusion, use of the Omniscience mechanical prosthesis resulted in poor functional improvement and a significant number of valve-related complications including the need for reoperation. We speculate that suboptimal hemodynamics secondary to limited occluder disc opening in vivo may have been a contributory factor.
|
References
|
|---|
- Scotten L.N., Racca R.G., Nugent A.H., Walker D.K., Brownlee R.T. New tilting disc cardiac valve prostheses. J Thorac Cardiovasc Surg 1981;82:136-146.[Medline]
- DeWall R.A., Caffarena Raggio J.M., Dittrich H., et al. The Omni design. Evolution of a valve. J Thorac Cardiovasc Surg 1989;98:999-1007.[Abstract]
- Cortina J.M., Martinelli J., Artiz V., Fraile J., Rabago G. Comparative clinical results with Omniscience, Medtronic Hall, and Bjork-Shiley convexo-concave prostheses in mitral valve replacement. J Thorac Cardiovasc Surg 1986;91:174-183.[Abstract]
- Fananapazir L., Clarke D.B., Dark J.F., Lawson R.A., Moussalli H. Results of valve replacement with the Omniscience prosthesis. J Thorac Cardiovasc Surg 1983;86:621-625.[Abstract]
- Rabago G., Martinelli J., Fraile J., Andrade I.G., Montenegro R. Results and complications with the Omniscience prosthesis. J Thorac Cardiovasc Surg 1984;87:136-140.[Abstract]
- Kazui T., Komatsu S., Inoue N. Clinical evaluation of the Omniscience aortic disc valve prosthesis. Scand J Thorac Cardiovasc Surg 1987;21:173-178.[Medline]
- Kazui T., Yamada O., Yamagishi M., Watanabe N., Komatsu S. Aortic valve replacement with Omniscience and Omnicarbon valves. Ann Thorac Surg 1991;52:236-244.[Abstract/Free Full Text]
- Carrier M., Martineau J.P., Bonan R., Pelletier L.C. Clinical and hemodynamic assessment of the Omniscience prosthetic heart valve. J Thorac Cardiovasc Surg 1987;93:300-307.[Abstract]
- DeWall R., Pelletier L.C., Panebianco A., et al. Five-year clinical experience with the Omniscience cardiac valve. Ann Thorac Surg 1984;38:275-280.[Abstract/Free Full Text]
- Akalin H., Corapcioglu T., Ozyurda U., et al. Clinical evaluation of the Omniscience cardiac valve prosthesis. J Thorac Cardiovasc Surg 1992;103:259-266.[Abstract]
- Teijeira F.J. Long-term experience with the Omniscience cardiac valve. J Heart Valve Dis 1998;7:540-547.[Medline]
- Edmunds L.H., Clark R.E., Cohn L.H., et al. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1996;62:932-935.[Abstract/Free Full Text]
- Khan S., Chaux A., Matloff J., et al. The St. Jude Medical valve. J Thorac Cardiovasc Surg 1994;108:1010-1020.[Abstract/Free Full Text]
- Aoyagi S., Oryoji A., Nishi Y., et al. Long-term results of valve replacement with the St. Jude Medical valve. J Thorac Cardiovasc Surg 1994;108:1021-1029.[Abstract/Free Full Text]
- Ibrahim M., O’Kane H., Cleland J., et al. The St. Jude Medical prosthesis. J Thorac Cardiovasc Surg 1994;108:221-230.[Abstract/Free Full Text]
- Baudet E.M., Puel V., McBride J.T., et al. Long-term results of valve replacement with the St. Jude Medical prosthesis. J Thorac Cardiovasc Surg 1995;109:858-870.[Abstract]
- Fernandez J., Laub G.W., Adkins M.S., et al. Early and late-phase events after valve replacement with the St. Jude Medical prosthesis in 1200 patients. J Thorac Cardiovasc Surg 1994;107:394-407.[Abstract/Free Full Text]
- Nitter-Hauge S., Abdelnoor M., Svennevig J.L. Fifteen-year experience with the Medtronic-Hall valve prosthesis. Circulation 1996;94:105-108.
- Nakano K., Koyanagi H., Hashimoto A., et al. Twelve years’ experience with the St. Jude Medical valve prosthesis. Ann Thorac Surg 1994;57:697-703.[Abstract/Free Full Text]
- Akins C.W. Results with mechanical cardiac valvular prostheses. Ann Thorac Surg 1995;60:1836-1844.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
A. A. Mikhail
Unusual omniscience valve results
Ann. Thorac. Surg.,
November 1, 2003;
76(5):
1774 - 1774.
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. S. Edwards, G. B. Russell, A. F. Edwards, J. W. Hammon Jr, A. R. Cordell, and N. D. Kon
Unusual omniscience valve results: Reply
Ann. Thorac. Surg.,
November 1, 2003;
76(5):
1774 - 1774.
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y. Misawa
Pannus formation after mitral valve replacement with the omniscience prosthesis: letter 1
Ann. Thorac. Surg.,
August 1, 2003;
76(2):
658 - 658.
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Teijeira
Omniscience valve results: letter 2
Ann. Thorac. Surg.,
August 1, 2003;
76(2):
659 - 659.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
