Abstract
Background: Based on the revised AHA/ACC guidelines, it might be necessary to take into consideration the average life expectancy of Japanese people and revise the Japanese guidelines accordingly. Accordingly, we performed the present study to compare the long-term outcome in patients aged 65 years or older who underwent prosthetic valve replacement at our hospital using mechanical valves or biological valves.
Methods: We have performed valve replacement in 416 patients aged 65 years or older (mechanical: 157; biological: 244).
Results: There was no significant difference between the mechanical and biological valve for the actuarial survival rate. As for the valve-related complication free rate, in the mechanical valve group, the rates were significantly higher for all patients, aortic valve replacement (AVR) patients, and mitral valve replacement (MVR) patients.
Conclusions: Following revision of the AHA/ACC guidelines for selection of prosthetic valves, it is necessary to investigate whether patients aged 60–70 represent the gray zone for selecting valves as in US and European guidelines, or whether a higher age is more appropriate in view of the longer average life expectancy in Japan. Accordingly, further evaluation of the long-term outcome for mechanical and biological valves in Japanese patients is needed to obtain evidence for preparation of original Japanese guidelines on prosthetic valve selection.
Keywords: prosthetic valve, mechanical valve, biological valve, valve replacement
Introduction
Thanks to the stable results of prosthetic valve replacement, the number of procedures performed in elderly patients has been increasing along with improving outcomes.1,2) Good long-term outcomes have been reported for patients with prosthetic valves because of improved antithrombogenicity and establishment of appropriate anticoagulant therapy for mechanical valves, as well as better durability and new treatment for preventing calcification of biological valves.3) Therefore, correct selection of prosthetic valves is important. The 2006 ACC/AHA guidelines recommend biological valves for aortic valve replacement in patients aged 65 years or older without risk factors for thromboembolism (Class IIa).4) However, the 2013 STS guidelines only provide a Class IIb recommendation for use of biological valves in patients less than 65 years old.5) In 2014, the AHA/ACC guidelines with regard to selection of prosthetic valves were extensively revised,6) with the main changes being as follows for Class I recommendations: “1) A mechanical prosthesis is reasonable for aortic valve replacement (AVR) or mitral valve replacement (MVR) in patients less than 60 years of age who do not have a contraindication to anticoagulation,” “2) A bioprosthesis is reasonable in patients more than 70 years of age,” and “3) Either a bioprosthetic or mechanical valve is reasonable in patients between 60 and 70 years of age.”
The average life expectancy of Japanese people is 3–5 years longer than for Western populations, being 80.2 years for men and 86.6 years for women according to the 2013 abridged life table data. These values are higher than in the past and Japanese women have the world’s longest average life expectancy, showing a difference of 6.4 years compared with men. In Japan, the average life expectancy at the age of 60 years is 23.1 years for men vs. 28.5 years for women, while it is 19.1 years vs. 24.0 years at age 65 and 15.3 years vs. 19.6 years at age 70. Thus, while it might be reasonable to select biological valves for patients aged 65 years or older in Western countries, whether this choice is appropriate in Japan seems questionable.
Based on the revised AHA/ACC guidelines,6) it might be necessary to take into consideration the average life expectancy of Japanese people and revise the Japanese guidelines accordingly. Accordingly, we performed the present study to compare the long-term outcome in patients aged 65 years or older who underwent prosthetic valve replacement at our hospital using mechanical valves or biological valves.
Patients and Methods
Patient population
Between 1996 and 2012, 772 patients underwent valve replacement surgery using mechanical or biological valve at the Nihon University Itabashi Hospital, of them, we have performed valve replacement in 416 patients aged 65 years or older (mechanical valve: 157 cases; biological valve: 244 cases), and examined their long-term follow-up outcomes.
Background and implanted valve data of the patients are shown in Tables 1 and 2. In present study, the average follow-up period was 6.2 ± 4.3 years (maximum: 18.5 years) in mechanical valve and 4.3 ± 3.6 years (maximum: 13.7 years) in biological valve. A follow-up rate was 97.4% in mechanical valve and 98.1% in biological valve (Tables 1 and 2).
Table 1.
Patient’s characteristics
| Mechanical | Biological | |
|---|---|---|
| Number | 158 | 258 |
| Age (years) | 69.7± 3.7 | 76.3± 4.8 |
| 70 years< | 72 (46%) | 238 (92%) |
| 80 years< | 2 (1%) | 64 (25%) |
| 70∼74 years | 57 (47%) | 65 (53%) |
| Male:female | 84:74 | 129:129 |
| AVR | 99 | 177 |
| Aortic enlargement | 7 (4%) | 12 (5%) |
| MVR | 41 | 57 |
| DVR | 18 | 24 |
| Concomitant surgery | ||
| CABG | 28 (18%) | 64 (25%) |
| Maze | 27 (17%) | 29 (11%) |
| Mitral annuloplasty | 6 (4%) | 9 (3%) |
| Tricuspid annuloplasty | 4 (3%) | 1 (0.4%) |
| TAA repair | 18 (12%) | 10 (4%) |
| LV repair | 1 (1%) | 2 (1%) |
| Morrow | 3 (2%) | 8 (3%) |
| PFO closure | 0 | 1 (0.4%) |
| Re-operation | 3 (%) | 5 (%) |
AVR: aortic valve replacement; MVR: mitral valve replacement; DVR: double valve replacement; CABG: coronary arterial bypass grafting; TAA: thoracic aortic aneurysm; LV: left ventricular; PFO: patent foramen ovale
Table 2.
Type and size of prosthetic valve
| Aortic valve | Mechanical | Biological |
|---|---|---|
| Number | 117 | 201 |
| Valve type | ATS: 64 | CEP: 174 |
| St. Jude Medical: 51 | Mosaic: 25 | |
| CarboMedics: 1 | Freestyle: 2 | |
| MIRA: 1 | ||
| Size (mm) | 21.4± 2.7 | 21.0± 2.0 |
| 19 mm> | 25 (21%) | 58 (29%) |
| 16 mm | 2 | – |
| 17 mm | 6 | – |
| 18 mm | 10 | – |
| 19 mm | 15 | 74 |
| 20 mm | 4 | – |
| 21 mm | 34 | 72 |
| 22 mm | 4 | – |
| 23 mm | 20 | 40 |
| 24 mm | 3 | – |
| 25 mm | 16 | 12 |
| 27 mm | 6 | 3 |
| Mitral valve | Mechanical | Biological |
|---|---|---|
| Number | 59 | 81 |
| Valve type | ATS: 34 | CEP: 74 |
| St. Jude Medical: 25 | Mosaic: 7 | |
| Size (mm) | 28.3± 1.9 | 27.6± 2.0 |
| 25 mm | 6 | 19 |
| 27 mm | 18 | 29 |
| 29 mm | 22 | 22 |
| 31 mm | 13 | 11 |
CEP: Carpenter-Edwards perimount
Selection criteria for prosthetic valves
Selection of prosthetic valves was decided after thorough discussion with each patient based on the following standards: mechanical valves were recommended for patients aged less than 70 years, biological valves were recommended for patients aged 75 years or older, and either type was recommended for patients aged between 70 and 75 years depending on their characteristics (level of activity, living with family, atrial fibrillation, with/without aortic annulus enlargement, etc.). If patients chose a mechanical valve, they were required to undergo postoperative follow-up once a month by a cardiologist or cardiac surgeon familiar with warfarin therapy (if their INR deviated from the target range, they were required to attend outpatients once a week until the values returned to the target range).
Follow-up
Postoperative anticoagulant therapy or antiplatelet therapy comprised warfarin, aspirin (100 mg/day) or dipyridamole (200 mg/day) or ticlopidine hydrochloride (200 mg/day). The international normalized ratio (INR) was maintained at 1.6–2.0 in AVR patients, 1.8–2.5 in MVR patients, and 2.0–2.5 in double valve replacement (DVR) patients, patients with atrial fibrillation, and patients with a left atrial diameter of 50 mm or more. Postoperatively, patients were seen once a month as outpatients. Postoperative event evaluations were based on the valvulopathy postoperative evaluation guidelines of the American Association for Thoracic Surgery, the European Association for Thoracic Surgery, and the Society of Thoracic Surgeons.7) Major adverse cardiovascular and cerebrovascular events (MACCE) were defined as all-cause death, postoperative cardiac failure, myocardial infarction, repeat revascularization, hospitalization for heart failure, and cerebrovascular accidents.
Statistical analysis
The actuarial survival rate and postoperative complication-free rate were evaluated using the Kaplan-Meier method, and differences between groups assessed using the log-rank test. Numerical values were expressed as mean ± standard deviation. The level of significance was taken as p <0.05.
Results
Early outcome
There were 16 deaths within 30 days postoperatively (3.8%) (four in mechanical valve (2.5%), 12 in biological valve (4.7%)). There were 17 deaths within 60 and 90 days postoperatively (4.1%) (four in mechanical valve (2.5%), 13 in biological valve (5.0%)). Hospital death was 19 cases (4.6%) (four in mechanical valve (2.5%), 15 in biological valve (5.8%). In the mechanical valve group, there were four hospital deaths after AVR (4%), while there were none after MVR or DVR. In the biological valve group, there were seven hospital deaths after AVR (4%), four deaths after MVR (7%), and four deaths after DVR (16.7%). The cause of death was low cardiac output syndrome (LOS) in two cases, arrhythmia and bleeding in one case each in mechanical valve, and left ventricular rupture in four cases, LOS in three cases, mediastinitis and pneumonia in two cases each, acute heart failure, acute myocardial infarction, arrhythmia, hepatic failure in one case each in biological valve.
Late outcome and MACCE
Late death occurred in 28 patients with mechanical valves and 34 patients with biological valves. In the mechanical valve group, the cause of death was heart failure in eight patients, malignancy in six patients, sudden death and pneumonia in two patients each, and arrhythmia, aortic dissection, prosthetic valve endocarditis (PVE), cerebral infarction, cerebral hemorrhage, sepsis, acute exacerbation of chronic renal failure, senility, mental disease, and heat stroke in one patient each. In the biological valve group, the cause of death was malignancy in nine patients, cardiac failure in seven patients, cerebral infarction in six patients, sudden death and pneumonia in three patients each, sepsis in two patients, and acute myocardial infarction, PVE, gastrointestinal bleeding, and acute exacerbation of chronic renal failure in one patient each.
The actuarial survival rate, including early mortality, was 84.3% ± 3.0% at 5 years, 77.1% ± 4.5% at 10 years, and 67.1% ± 6.7% for 13 years, 62.3% ± 7.8% for 15 years in mechanical valve, 79.7% ± 2.9% at 5 years, 71.6% ± 4.3% at 10 years, and 67.8% ± 5.5% at 13 years in biological valve, following all case (p = 0.387); 82.4% ± 4.0% at 5 years, 70.5% ± 6.7% at 10 years, 64.6% ± 8.3% at 13 years, and 57.4% ± 10.0% at 15 years in mechanical valve, 82.7% ± 3.3% at 5 years, 72.7% ± 6.0 % at 10 years, and 66.7% ± 8.0% at 13 years in biological valve, following AVR (p = 0.753); 87.0% ± 5.5% at 5 and 10 years, and 63.4% ± 14.9% at 13 and 15 years in mechanical valve, 78.4% ± 6.0% at 5 years, 73.8% ± 7.2% at 10 and 13 year in biological valve, following MVR (p = 0.282); and 88.2% ± 7.8% at 5, 10, 13, 15 years in mechanical valve, 61.9% ± 12.4% at 5 years, 53.0% ± 13.4% at 10 and 13 years in biological valve, following DVR (p = 0.099) (Fig. 1).
Fig. 1.
Actuarial survival rate in AVR, MVR, and DVR.
AVR: aortic valve replacement; MVR: mitral valve replacement; DVR: double valve replacement
MACCE occurred in 45 patients from the mechanical valve group (early death in four patients, late death in 28 patients, cardiac failure in nine patients, cerebral infarction in two patients, and paravalvular leakage (PVL) and atrioventricular block (AVB) requiring pacemaker implantation in one patient each). In addition, MACCE occurred in 74 patients from the biological valve group (early death in 15 patients, late death in 33 patients, cardiac failure in 11 patients, cerebral hemorrhage in four patients, AVB requiring pacemaker implantation in thre patients, PVL, structural valve dysfunction (SVD), and cerebral infarction in two patients each, and ischemic heart disease and left ventricular rupture in one patient each).
The MACCE free rate was 77.8% ± 3.5% at 5 years, 69.1% ± 4.9% at 10 years, 57.9% ± 6.6% at 13 years, and 46.1% ± 9.5% at 15 years in mechanical valve, 71.2% ± 3.3% at 5 years, 56.0% ± 5.1% at 10 years, and 46.1% ± 6.7% at 13 years in biological valve, following all case (p = 0.086); 78.0% ± 4.4% at 5 years, 66.0% ± 6.8% at 10 years, 57.4% ± 8.2% at 13 years, and 51.0% ± 9.5% at 15 years in mechanical valve, 73.5% ± 3.9% at 5 years, 63.2% ± 5.8% at 10 years, and 52.7% ± 8.3% at 13 years, following AVR (p = 0.639); 74.3% ± 7.1% at 5 years, 69.3% ± 8.2% at 10 years, and 50.5% ± 12.9% at 13 and 15 years, 71.7% ± 6.7% at 5 years, 44.1% ± 10.9% at 10 years, and 29.4% ± 14.0% at 13 years in biological valve, following MVR (p = 0.129); and 87.5% ± 8.3% at 5, 10, and 15 years in mechanical valve, 52.7% ± 12.7% at 5 years, and 44.0% ± 13.3% at 10 and 13 years in biological valve, following DVR (p = 0.037). In the mechanical valve group, the rate was significantly higher after DVR (Fig. 2).
Fig. 2.
MACCE free rate in AVR, MVR, and DVR.
AVR: aortic valve replacement; MVR: mitral valve replacement; DVR: double valve replacement
Valve-related complications
Valve-related complications occurred in nine patients from the mechanical valve group (cerebral infarction in three patients, sudden death in two patients, and cerebral hemorrhage, gastrointestinal bleeding, PVE, and PVL in one patient each), while such complications occurred in 28 patients from the biological valve group (cerebral infarction in nine patients, left ventricular rupture in five patients, cerebral hemorrhage in four patients, sudden death in three patients, gastrointestinal bleeding, SVD, and PVL in two patients each, and PVE in one patient).
The valve-related complication free rate was 96.6% ± 1.5% at 5 years, 94.5% ± 2.5% at 10 years, and 77.9% ± 12.6% for 15 years in mechanical valve, 88.3% ± 2.4% at 5 years, 80.9% ± 4.8% at 10 years, and 71.4% ± 7.7% for 13 years in biological valve, following all case (p = 0.004); 97.8% ± 1.5% at 5, 10, and 15 years in mechanical valve, 92.1% ± 2.5% at 5 years, 87.3% ± 5.3% at 10 years, and 80.0% ± 8.5% for 13 years in biological valve, following AVR (p = 0.04); 100% at 5 years, 93.8% ± 6.1% at 10 years, and 82.0% ± 12.2% at 15 years in mechanical valve, 80.2% ± 6.2% at 5 years, 66.2% ± 10.5% at 10 years in biological valve, following MVR (p = 0.004); and 81.6% ± 9.6% at 5, 10, and 15 years in mechanical valve, 78.6% ± 10.3% at 5, 10, and 13 years in biological valve, following DVR (p = 0.955). In the mechanical valve group, the rates were significantly higher for all patients, AVR patients, and MVR patients (Fig. 3).
Fig. 3.
Valve related complication free rate in AVR, MVR, and DVR.
Discussion
The Japanese guidelines for selecting prosthetic valves were revised in 2012 (www.jcirc.or.jp/guideline/pdf/JCS2012_ookita_h.pdf). Class I recommendations regarding the indications for biological valves are patients aged 65 years or older who require AVR and who do not have risk factors for thromboembolism. Class IIa recommendations are patients aged 70 years or older who require MVR and who do not have risk factors for thromboembolism, or patients in sinus rhythm who chose a biological valve after detailed discussion of the risks of anticoagulant therapy and the risk of future reoperation, even if younger than 65 years. These guidelines are in accordance with those of the United States and Europe before 2012. There have been few reports about the long-term results in Japanese patients with prosthetic valves, which mean that the Japanese guidelines were largely devised on the basis of literature from the United States and Europe.
Aging of the population in Japan is the most rapid worldwide. In view of the Japanese tendency to periodically attend hospital outpatient departments and to adhere to medication, our findings suggested that it may be reasonable to increase the age at which mechanical valves are selected. According to the guidelines for selection of prosthetic valves released by the Japanese Circulation Society, biological valves are recommended for AVR in patients aged 65 years or older. However, the present study showed that patients aged 65 years or older undergoing AVR with mechanical valves had a lower incidence of valve-related complications than patients receiving biological valves with complications occurring in only two patients, which suggests that our criteria for prosthetic valve selection are appropriate. Japanese have a long lifespan, particularly women, and the average life expectancy of a Japanese woman aged 65 is 24.0 years. Because there is no biological valve for which durability of 20 years or longer has been demonstrated, it is questionable whether the current Japanese guidelines are appropriate. However, our study was conducted at a single hospital and patients in whom shorter survival was expected may have received biological valves compared with those given mechanical valves. Accordingly, more long-term outcome studies of prosthetic valve replacement in Japanese patients are needed.
Long-term comparative studies of mechanical and biological valves performed in the USA and Europe have identified reoperation due to SVD as a problem for biological valves, while bleeding complications are problematic with mechanical valves.8,9) A randomized study of mechanical and biological valves performed in 310 AVR patients aged 55 to 70 years demonstrated that bleeding complications could be prevented by maintaining the INR within a low range of 2.0–2.5.10) A comparative study of AVR patients aged 65–80 years from the STS database revealed that the survival rate of those given biological valves was higher from 9 years onwards, the reoperation rate was significantly higher in patients with biological valves, and the number of patients aged 65–69 with biological valves was three times larger than that of patients aged 75–80 years.11)
There have been a few reports regarding long-term results over 10 years in Japanese patients. Nishida et al. compared AVR with mechanical or biological valves and reported no difference in freedom from valve-related death at 20 years between the two valve types in patients aged 60 years or older, while there was a significant difference in patients younger than 60 years (91.1% ± 2.4% for mechanical valves vs. 61.2% ± 19.0% for biological valves). In addition, freedom from valve-related morbidity at 20 years in patients aged 70 years or older was better with biological valves (51.2% ± 11.0% for mechanical valves vs. 57.5% ± 14.0% for biological valves), while mechanical valves were superior in patients younger than 60 years (73.3% ± 3.9% for mechanical valves vs. 11.1% ± 7.3% for biological valves). They concluded that selection of mechanical valves for patients aged less than 60 years and biological valves for patients aged 70 years or older was appropriate.12) Minakata et al. studied the long-term outcome of implanting the CEP valve in the aortic position, and reported that the rate of freedom from reoperation due to SVD was 99.5% at 5 years, 96.7% at 10 years, and 97.5% at 15 years. In patients aged 65 years or older, the rate was 94.4% at 10 years, while it was a significantly lower 47.2% in patients aged less than 65 years, indicating that more favorable results were obtained with the CEP valve in older patients.13) In the 15-year outcome study of the ATS valve, a mechanical valve used at our hospital, the rate of freedom from thromboembolism was 98.0% ± 1.4% at 10 years vs. 98.0% ± 1.4% at 15 years in AVR patients, 98.0% ± 1.4% at 10 years vs. 98.0% ± 1.4% at 15 years in MVR patients, and 93.3% ± 4.6% at 10 years vs. 93.3% ± 4.6% at 15 years in DVR patients. In addition, the rate of freedom from bleeding was 98.0% ± 1.4% at 10 years vs. 98.0% ± 1.4% at 15 years in AVR patients, 100% at 10 years vs. 75.0% ± 2.2% at 15 years in MVR patients, and 100% at both 10 and 15 years in DVR patients, which were favorable results. The reason for the low incidence of bleeding is that the INR was controlled by considering the patient’s underlying diseases state and low INRs were maintained (1.6–2.0 for AVR patients in sinus rhythm, 1.8–2.5 for MVR patients, and 2.0–2.5 for DVR patients, patients with atrial fibrillation, or patients with a left atrial diameter ≥50 mm). It was concluded that strict follow-up after prosthetic valve replacement is most important for reducing valve-related complications.14)
Since the quality of medical care and patient compliance with medication are higher in Japan than in the USA or Europe, following revision of the AHA/ACC guidelines for selection of prosthetic valves, it is necessary to investigate whether patients aged 60–70 represent the gray zone for selecting valves as in US and European guidelines, or whether a higher age is more appropriate in view of the longer average life expectancy in Japan. Accordingly, further evaluation of the long-term outcome for mechanical and biological valves in Japanese patients is needed to obtain evidence for preparation of original Japanese guidelines on prosthetic valve selection. The results of this study may contribute to the development of such guidelines, but further studies in a larger number of subjects with a longer observation period are needed in the future.
Limitations
This was not a randomized study. Since biological valves might be selected for more severe patients compared with those given mechanical valves or might be chosen for patients in whom long-term survival is not expected, the possibility of selection bias cannot be ruled out.
Conflict of Interest Statement
Akira Sezai received lecture fees from Dai-ichi Sankyo Co., Ltd, and the other authors have no conflict of interest.
References
- 1).Kamiya H, Tanzeem N, Akhyari P, et al. Impact of severe postoperative complications after cardiac surgery on mortality in patients aged over 80 years. Ann Thorac Cardiovasc Surg 2014; 20: 383-9. [DOI] [PubMed] [Google Scholar]
- 2).Hosono M, Sasaki Y, Hirai H, et al. Risk factors for late valve-related mortality after aortic valve replacement in elderly patients. Ann Thorac Cardiovasc Surg 2013; 19: 368-74. [DOI] [PubMed] [Google Scholar]
- 3).Brown JM, O’Brien SM, Wu C, et al. Isolated aortic valve replacement in North America comprising 108,687 patients in 10 years: changes in risks, valve types, and outcomes in the Society of Thoracic Surgeons National Database. J Thorac Cardiovasc Surg 2009; 137: 82-90. [DOI] [PubMed] [Google Scholar]
- 4).Bonow RO, Carabello BA, Chatterjee K, 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. Circulation 2006; 114: e84-231. [DOI] [PubMed] [Google Scholar]
- 5).Svensson LG, Adams DH, Bonow RO, et al. Aortic valve and ascending aorta guidelines for management and quality measures: executive summary. Ann Thorac Surg 2013; 95: 1491-505. [DOI] [PubMed] [Google Scholar]
- 6).Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline 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. Circulation 2014; 129: e521-643. [DOI] [PubMed] [Google Scholar]
- 7).Akins CW, Miller DC, Turina MI, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. J Thorac Cardiovasc Surg 2008; 135: 732-8. [DOI] [PubMed] [Google Scholar]
- 8).Forcillo J, El Hamamsy I, Stevens LM, et al. The perimount valve in the aortic position: twenty-year experience with patients under 60 years old. Ann Thorac Surg 2014; 97: 1526-32. [DOI] [PubMed] [Google Scholar]
- 9).Prasongsukarn K, Jamieson WR, Lichtenstein SV. Performance of bioprostheses and mechanical prostheses in age group 61-70 years. J Heart Valve Dis 2005; 14: 501-8, 510-1; discussion 509. [PubMed] [Google Scholar]
- 10).Stassano P, Di Tommaso L, Monaco M, et al. Aortic valve replacement: a prospective randomized evaluation of mechanical versus biological valves in patients ages 55 to 70 years. J Am Coll Cardiol 2009; 54: 1862-8. [DOI] [PubMed] [Google Scholar]
- 11).Brennan JM, Edwards FH, Zhao Y, et al. Long-term safety and effectiveness of mechanical versus biologic aortic valve prostheses in older patients: results from the Society of Thoracic Surgeons Adult Cardiac Surgery National Database. Circulation 2013; 127: 1647-55. [DOI] [PubMed] [Google Scholar]
- 12).Nishida T, Sonoda H, Oishi Y, et al. Long-term results of aortic valve replacement with mechanical prosthesis or Carpentier-Edwards perimount bioprosthesis in Japanese patients according to age. Circ J 2014; 78: 2688-95. [DOI] [PubMed] [Google Scholar]
- 13).Minakata K, Tanaka S, Okawa Y, et al. Long-term outcome of the Carpentier-Edwards pericardial valve in the aortic position in Japanese patients. Circ J 2014; 78: 882-9. [DOI] [PubMed] [Google Scholar]
- 14).Sezai A, Hata M, Niino T, et al. Fifteen years of experience with ATS mechanical heart valve prostheses. J Thorac Cardiovasc Surg 2010; 139: 1494-500. [DOI] [PubMed] [Google Scholar]