| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:467-473
© 2002 The Society of Thoracic Surgeons
a Siyami Ersek Thoracic and Cardiovascular Surgery Center, Istanbul, Turkey
Accepted for publication July 17, 2001.
* Address reprint requests to Dr Kaplan, 67. Ada Kardelen 4-4, D: 11 Atasehir – 81120, Istanbul, Turkey
e-mail: mehmetkaplan{at}superonline.com
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. In our institution, between January 1980 and December 2000, 129 tricuspid valve replacements were performed in 122 patients (14.7%). Bioprosthetic valves were used in 32 patients, whereas 97 patients had mechanical valve implantation. Twenty-two percent of replacements were done on men. Mean age was 35.27 ± 11.56 years. In all patients, initially an annuloplasty technique was tried. Tricuspid valve replacement was performed when annuloplasty was not sufficient. In most of the cases, tricuspid valve interventions were done under cardiopulmonary bypass and on a beating heart.
Results. Early mortality was 24.5%. Patients were followed for 2 to 228 months. Seven patients underwent reoperation because of tricuspid valve dysfunction (7.6%). Nine patients died during the follow-up period. Late mortality was 9.7%. Actuarial estimates of survival in 20 years of follow-up for all tricuspid prosthetic valves, mechanical valves, and bioprosthetic valves were 65.1% ± 9.3%, 68.3% ± 10.6%, and 54.8% ± 12.1%, respectively. For the bioprosthetic valve group, freedom from structural valve degeneration was 90% ± 5.5%; for the mechanical valve group, freedom from deterioration, endocarditis, and leakage was 97.8% ± 4.2%, and freedom from thromboembolism was 92.6% ± 6.9%.
Conclusions. We found that there was no statistically significant difference between the two groups in terms of early mortality, re-replacement, and midterm mortality (p > 0.05). Nevertheless, we recommend low profile modern bileaflet mechanical valves for prosthetic replacement of the tricuspid valve, due to their favorable hemodynamic characteristics and durability.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Selection of the suitable prosthesis is still debatable. Although modern aortic and mitral bileaflet mechanical valves provide excellent hemodynamic results, according to the literature, tricuspid prostheses are not completely satisfactory [1, 2].
In this study we present the results of 129 valve replacements in 122 patients who underwent operation in our institution, compare them with current literature data, and discuss whether bioprosthetic or mechanical valves should be selected.
|
Material and methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
Ninety-five percent of patients had atrial fibrillation. New York Heart Association status of 16% of patients was class II, 52% were class III, and 32% were class IV. We had no class I patient. Preoperatively, 80% of the patients had tricuspid insufficiency and 15% of them had tricuspid stenosis.
Etiologies were as follows: rheumatic (93.44%, 114 patients), congenital (4.92%, 6 patients including 4 Ebstein anomaly, 1 tetralogy of Fallot, and 1 ostium primum type atrial septal defect), and native tricuspid valve endocarditis (1.64%, 2 patients) (Table 1).
Fifty-nine patients underwent a previous cardiac surgical intervention. These operations were closed mitral valvotomy (14 patients), repair of ventricular septal defect (2), repair of atrial septal defect (2), repair of ostium primum type atrial septal defect plus tricuspid De Vega annuloplasty (1), total correction of tetralogy of Fallot (1), repair of Ebstein anomaly (2), mitral valve replacement (22), mitral and aortic valve replacement with tricuspid valve repair (9), and mitral, aortic, and tricuspid valve replacement (6 patients) (Table 1).
Four patients had Ebstein anomaly and 2 of them were redo cases. One of these redo patients had previously undergone Duran ring annuloplasty and this was followed by valve replacement with mechanical prosthesis. This patient then developed noncoronary sinus Valsalva aneurysm and was reoperated. A second patient had two previous bioprosthetic valve implantations and replacement with mechanical valve prosthesis was performed. Bioprosthesis and mechanical valve implantation was performed for the remaining 2 patients, respectively. The first two redo patients died, but the third and fourth patients are still alive.
In 116 patients (123 replacements), a midsternal incision was used, whereas in 6 redo cases, a right thoracotomy was preferred to avoid retrosternal adhesions. Operations were performed under standard cardiopulmonary bypass and mild systemic hypothermia (28°C to 32°C). Cold intermittent antegrade crystalloid or blood cardioplegia and topical myocardial cooling with iced saline slush were used. Annuloplasty technique was initially tried for all patients. To evaluate the repair, an aortic cardioplegia cannula (Stockert Malleable Coronary Perfusion Cannula, P616 to 40, Stockert Instrumente GmbH, Munich, Germany) was attached to the tip of an external cooling line and the right ventricle was filled with saline by inserting this line into the right ventricle. Meanwhile, the pulmonary artery was digitally clamped and the amount of insufficiency was assessed by direct inspection. In case of a large insufficiency (4+), replacement was performed. If the right atrial pressure at the time of weaning from cardiopulmonary bypass was greater than 10 mm Hg, jet flow was checked by digital exploration of the right auricle. If jet flow reached 3 cm or more above the valve, replacement was performed. But if the jet flow was less, a large insufficiency was suspected. Then cardiopulmonary bypass was restarted, the right atrium was opened, and if leaflets did not form a dome shape and if the diameter of annulus was 2.5 times the surgeon’s finger, then replacement was done. Replacement was also performed in patients who had repair and revealed an insufficiency within acceptable limits during intraoperative evaluations but who could not be weaned from cardiopulmonary bypass.
Surgical exploration of the tricuspid valve revealed that the annular dilatation with tricuspid insufficiency was the predominant lesion in 80% of the patients. Fifteen percent of them had a commissural fusion.
We initially performed repair in all patients. Time necessary for repair is not too long and this procedure may eliminate the risk of high morbidity and mortality due to TVR.
In appropriate patients and when possible, tricuspid valvular interventions were performed under cardiopulmonary bypass without cross-clamping and on a beating heart. Nine percent of TVR (12) were performed with cross-clamping, whereas 91% (117) were done on a beating heart. We performed most operations on a beating heart, because the functional classes of these patients were III and IV, they were mostly redo operations, and most of the patients needed concomitant interventions. Therefore, we did not want to extend the duration of cross-clamping for concomitant interventions and we did not want to use cross-clamping in isolated TVR operations.
Prosthetic valves were implanted by using pledgeted interrupted horizontal mattress sutures. We used pledgeted sutures, because the pledget has a supportive function below the suture. This protects the annulus from the cutting effect of the suture material and prevents paravalvular leakage. Our routine clinical practice in valve replacement is to use sutures with pledget.
One hundred twenty-nine prosthetic tricuspid valves were implanted in 122 patients. Ninety-seven of them had implantation of mechanical prostheses. Mechanical prostheses were as follows: 28 Carbomedics (Sulzer Carbomedics Inc, Austin, TX), 21 Medtronic (Medtronic Inc, Minneapolis, MN), 28 St. Jude (St. Jude Medical Inc, Minneapolis, MN), 9 Sorin Monoleaflet (Sorin Biomedica, Saluggia, Italy), 8 Bjork-Shiley (Sorin Biomedica, Saluggia, Italy), 2 Hall-caster (Medical Inc, Minneapolis, MN), and 1 Omniscience (Medical Inc, Minneapolis, MN) (Table 2). The following bioprostheses were used in 32 cases: 19 Biocor porcine (Biocor Industriae Pesquisahtda, Belo Horizonte, MG, Brazil), 7 Wessex Medical (Holmbush Estate, West Sussex, England) porcine bioprosthesis, 3 Hancock (Medtronic Inc, Irvine, CA), 2 Carpentier-Edwards (Baxter-Edwards Inc, Irvine, CA), and 1 Ionescu-Shiley bovine pericardial xenograft (Shiley Inc, Irvine, CA) (Table 3).
|
|
Sodium warfarin was used for anticoagulation; it was taken for life in patients with mechanical valve implantation, whereas with bioprostheses it was taken for only 3 months. However, in patients with bioprosthetic valves who also have atrial fibrillation or a right atrial diameter of greater than 5.5 cm, oral anticoagulation was taken lifelong. International normalized ratio was kept between 2.5 to 3.5 in TVR cases.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Causes of high early mortality are as follows: 59 patients (48.36%) were redo cases; all early mortality cases were in functional class III and IV; concomitant intervention rate was 61.47% (75 patients); some had ascites or high bilirubin levels. Concomitant interventions of early mortality were mitral valve replacement in 14 patients (7 of them were redo), aortic valve replacement in 1 (redo), and aortic valve replacement + mitral valve replacement in 9 patients (3 of them were redo). The remaining 6 patients who were early mortality cases had isolated TVR. Early mortality was significantly related to extended cardiopulmonary bypass duration, advanced deterioration of functional capacity, presence of a previous cardiac operation, and presence of a concomitant intervention (p < 0.05). Complete heart block developed in 6 patients (4.9%) and they had permanent pacemaker implantation. Postoperatively 19%, 48%, 28%, and 5% of cases were New York Heart Association functional classes I, II, III, and IV, respectively.
Nine patients died during the follow-up period. Late mortality was 9.7%. Causes of late mortality were infective endocarditis (1 patient), right heart failure (3), arrhythmia (1), valvular thrombosis (2), pulmonary embolism (1), and sudden death with unknown cause (1 patient).
Patients who developed pulmonary embolism had thrombolytic treatment and anticoagulation. However, 2 patients died of pulmonary embolism (one in the early and one in the late postoperative period) due to the severity of the clinical presentation and failure to respond to the medication.
Only 1 patient had endocarditis during the postoperative period with the major findings of endocarditis. Postoperative fever was present in 7 patients (5.73%) and these were not included in the endocarditis group, but they were treated for endocarditis; blood samples were obtained for culture and prophylaxis for endocarditis was initiated.
Seven patients underwent re-replacement of the prosthetic tricuspid valve. Three had mechanical valve thrombosis after 17 days, 1 year, and 7 years, respectively. One patient was operated for paravalvular leakage after 1 year. One had tricuspid valve endocarditis at the fourth year and was operated. Two patients were operated for degeneration of the bioprosthesis after 6 and 7 years (at age 42 and 39 years, respectively) and mechanical valve implantation was performed (Table 1).
In the patients with TVR who were rehospitalized due to dyspnea, if the central venous pressure was more than 15 mm Hg, transthoracic echocardiography was performed for the diagnosis of valvular thrombosis. Five were diagnosed with valvular thrombosis. Thrombolytic treatment was performed in 2 of them and it was successful. It was too late for thrombolytic treatment in the remaining 3 patients. Irregular use of oral anticoagulants had led to thrombosis. Preoperative functional impairment of these patients was advanced (class IV) and 80% of them were redo cases.
Mean survival time according to Kaplan-Meier test (Fig 1) was 159.9 ± 19 months for mechanical valves, whereas it was 85.7 ± 12.1 months for bioprostheses. Actuarial estimates of survival in the 20-year follow-up were 65.12% ± 9.39%, 68.37% ± 10.67%, and 54.84% ± 12.13% for all tricuspid prosthetic valves, mechanical valves, and bioprosthetic valves, respectively. For mechanical valve group, freedom from thromboembolism (Fig 2) was 92.6% ± 6.9% and freedom from deterioration, endocarditis, and leakage (Fig 3) was 97.83% ± 4.2%; for the bioprosthetic valve group, freedom from structural valve degeneration (Fig 4) was 90% ± 5.5%.
|
|
|
|
2 test (p > 0.05). Although there seems to be a difference in favor of mechanical valves, particularly in terms of early mortality, this difference is not statistically significant (Table 4).
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
The prevalence of TVR in the series by Ratnatunga [1], Spampinato [3], and McGrath [4] and their colleagues was 1.9%, 1.8%, and 1.7%, respectively. According to McGrath and colleagues [4], tricuspid valve operations constitute 5.7% of all valvular interventions. Sixty-seven percent of tricuspid operations are valve repairs and the remaining 33% are valve replacements. Re-replacement rate in TVR operations is 0.75%. In our series, TVR rate among all tricuspid valvular operations is 14.7% and re-replacement rate in TVR operations is 7.6%. In the Mayo clinic series, prognosis was poor and mortality was high in patients with a mitral or aortic prosthesis who also had tricuspid insufficiency [5]. Sixty percent of our patients had mitral valve replacement and 20% of them had aortic valve replacement as a concomitant procedure.
Our literature search related to TVR revealed that early mortality rate, reoperation rate, and 5- and 10-year survival rate were 13% to 30%, 9% to 31%, 50% to 70%, and 40%, respectively [6–10]. Kawachi and associates [11] reported that in multivalvular replacement cases, which also included TVR, operative mortality was high and long-term results were unfavorable. Our early mortality rate was 24.5%. However, high perioperative mortality may decrease statistical accuracy and it may be difficult to differentiate whether mortality is caused by cardiac failure or tricuspid valve failure [5].
A high mortality rate with TVR within first 30 days indicates that patients with tricuspid valve replacement are generally surgical candidates with a poor prognosis and this high rate is significantly related to the following factors: age more than 50 years, mitral valve replacement as a concomitant procedure, triple valve replacement, functional status, presence of hepatic dysfunction, high mean pulmonary artery pressure, gender, and duration of cardiopulmonary bypass [1, 5].
In almost all series, the main cause of death is right ventricular dysfunction. Preoperative severe right heart failure, advanced deterioration of functional capacity, and extended duration of cardiopulmonary bypass are usually associated with a high early mortality rate. On the other hand, most of the patients have concomitant aortic and mitral valve disease or a history of one or more previous open or closed cardiac surgical interventions. Their functional capacities are usually class III or IV [4]. Multiorgan failure and intraoperative hemorrhage are other causes of death [6]. In our series, causes of early mortality were failure to wean from cardiopulmonary bypass, right heart failure, hemorrhage, arrhythmia, pulmonary embolism, and mediastinitis. Fifty-three percent of our early mortality cases had a history of a previous cardiac operation and 47% underwent concomitant additional procedures.
Risk factors for late mortality are preoperative edema, extended duration of aortic cross-clamping, high pulmonary artery pressure, large tricuspid valve size, previous cardiac operation, age more than 55 years, and advanced functional insufficiency [7]. However, in our study right heart failure and valvular thrombosis predominate among all causes of late mortality. Sixty-seven percent of late mortality cases had undergone previous cardiac operation and 33% had concomitant additional cardiac intervention.
Valve selection for the tricuspid position is debatable. Initial mechanic prostheses (ball and cage, tilting disc) were replaced by bioprostheses. Because of the low pressure, some surgeons prefer a bioprosthesis in the tricuspid position. Van Nooten and coworkers [8] suggest that large-sized bioprostheses are preferable if a bioprosthesis is to be used. Low pressure and low stress in the right heart provide higher durability for the bioprosthesis.
Mechanical prostheses have certain disadvantages, such as the need for anticoagulation, risk of hemorrhage, risk of thromboembolism, and its effect on the right ventricular internal morphology. On the other hand, the new bileaflet mechanical valves have good hemodynamic properties, low gradient, decreased turbulence, and optimal durability [1, 8].
Bioprostheses at the tricuspid position have a lower degeneration rate than the ones located in systemic circulation. However, long-term durability of tissue valves is limited due to fibrocalcification and fatigue-related disruption [3]. The average time until bioprosthetic valve failure is reported at 7 years by Rizzoli and colleagues [5] and at 10 years by Del Campo and associates [12]. Our reoperation rate was 7.6% and the average duration to reoperation was 6.5 years [5]. Supporters of bioprosthesis prefer these prostheses for the following reasons. They do not require anticoagulation, they degenerate more slowly at the tricuspid position than at either the aortic or mitral position, their short-term durability is favorable, and the early reoperation rate is low. Contraindication to anticoagulation use, likelihood of pregnancy, older patients whose life expectancy is compatible with valve durability, socioeconomic status, lifestyle, inability to use drugs without help, limited access to healthcare facilities are reasons to prefer the use of a bioprosthesis. Patients with drug addiction and with a history of endocarditis should also have a bioprosthesis implanted [2, 3, 12].
Homografts are also recommended in the literature. McKay and colleagues [13] used unstented pericardium to encircle the pulmonary homograft in a patient with Ebstein anomaly. Katsumata and coworkers [14] recommend homografts in patients with endocarditis.
A modern bileaflet mechanical valve, St. Jude Medical (St. Jude Medical, Inc, Minneapolis, MN), is a low profile valve. It can be safely implanted without any restriction of leaflet motion caused by subvalvular muscular tissue. In this way, complete central flow is provided and turbulence is not observed. This protects the prosthesis from valvular thrombosis and tissue enlargement [15]. Mechanical prostheses should be preferred in young patients and in patients with mechanical prosthesis on the left heart [16]. Singh [10] and Kawano [17] and their colleagues recommend the St. Jude Medical valve (St. Jude Medical, Inc, Minneapolis, MN). Nakano and associates [6] reported 1 valve thrombosis in 39 patients with St. Jude Medical valve during 14 years of follow-up and stated that with appropriate anticoagulation, rate of freedom from operative complications due to valve, operative death, and sudden death was 80.5% ± 6.7%. Our rate of freedom from thromboembolism is 92.6% ± 6.9%. Ratnatunga [1], Scully [16], and Rizzoli [5] and their colleagues found that there was no statistically significant difference between the two groups. This is similar to our results.
In conclusion, according to the review of the literature and our experience, there are two basic points about tricuspid valve interventions. First and the more important issue is when and how to decide on tricuspid valve replacement. All conservative approaches should be tried. After repair, 2 to 3+ tricuspid insufficiency is acceptable and does not necessitate tricuspid valve replacement, because the size of the tricuspid annulus decreases rapidly after valvular operations of left side of the heart [18]. The second point concerns the selection of the prosthetic valve. According to the recent literature there is a trend toward new generation bileaflet mechanical valves because of their hemodynamic characteristics and durability, and some studies indicate that these valves have better long-term results in the tricuspid position; however, articles reporting no statistically significant difference between the two groups are also published. We also did not find any significant difference between the two groups in terms of early mortality, re-replacement, and midterm mortality.
Although in our study there were some differences between the mechanical and bioprosthetic valve groups in terms of patient and valve characteristics, we prefer and recommend low profile modern bileaflet mechanical valves in our recent clinical practice for prosthetic replacement of the tricuspid valve because of their favorable hemodynamic characteristics and durability.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. M. Russell, P. M. McCarthy, R. Lee, S. C. Malaisrie, and E. C. McGee Jr Jr Tricuspid valve replacement with a bioprosthetic valve MMCTS, March 28, 2008; 2008(0328): 2774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tokunaga, M. Masuda, A. Shiose, Y. Tomita, S. Morita, and R. Tominaga Long-Term Results of Isolated Tricuspid Valve Replacement Asian Cardiovasc Thorac Ann, February 1, 2008; 16(1): 25 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Shemin Tricuspid Valve Disease Card. Surg. Adult, January 1, 2008; 3(2008): 1111 - 1128. [Full Text]
|
|
|
|
 |
|
 Z. H. Iscan, K. M. Vural, I. Bahar, L. Mavioglu, and A. Saritas What to expect after tricuspid valve replacement? Long-term results Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 296 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Kunadian, K. Vijayalakshmi, S. Balasubramanian, and J. Dunning Should the tricuspid valve be replaced with a mechanical or biological valve? Interactive CardioVascular and Thoracic Surgery, August 1, 2007; 6(4): 551 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. L. Bartlett, D. L. Atkins, T. L. Burns, K. J. Engelkes, S. J. Powell, C. B. Hills, and J. H. Moller Early Outcomes of Tricuspid Valve Replacement in Young Children Circulation, January 23, 2007; 115(3): 319 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-Q. Lai, X. Meng, T. Bai, C. Zhang, Y. Luo, and Z.-G. Zhang Edge-to-Edge Tricuspid Valve Repair: An Adjuvant Technique for Residual Tricuspid Regurgitation Ann. Thorac. Surg., June 1, 2006; 81(6): 2179 - 2182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ikeda, Y. Okazaki, K. Furukawa, S. Ohtsubo, J. Yunoki, M. Natsuaki, and T. Itoh Direct imaging of bileaflet mechanical valve behavior in the tricuspid position. Eur. J. Cardiothorac. Surg., June 1, 2006; 29(6): 1014 - 1019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.-C. Chang, S.-H. Lim, G. Yi, Y. S. Hong, S. Lee, K.-J. Yoo, M. S. Kang, and B. K. Cho Long-Term Clinical Results of Tricuspid Valve Replacement Ann. Thorac. Surg., April 1, 2006; 81(4): 1317 - 1324. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Filsoufi, A. C. Anyanwu, S. P. Salzberg, T. Frankel, L. H. Cohn, and D. H. Adams Long-Term Outcomes of Tricuspid Valve Replacement in the Current Era Ann. Thorac. Surg., September 1, 2005; 80(3): 845 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. A G Solomon, R. C H Lim, P. Nand, and K. J Graham Tricuspid Valve Replacement: Bioprosthetic or Mechanical Valve? Asian Cardiovasc Thorac Ann, June 1, 2004; 12(2): 143 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Rizzoli, I. Vendramin, G. Nesseris, T. Bottio, C. Guglielmi, and L. Schiavon Biological or mechanical prostheses in tricuspid position? a meta-analysis of intra-institutional results Ann. Thorac. Surg., May 1, 2004; 77(5): 1607 - 1614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Chachques, P. G. Argyriadis, G. Fontaine, J.-L. Hebert, R. A. Frank, N. D'Attellis, J.-N.o. Fabiani, and A. F. Carpentier Right ventricular cardiomyoplasty: 10-year follow-up Ann. Thorac. Surg., May 1, 2003; 75(5): 1464 - 1468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Carrier, Y. Hebert, M. Pellerin, D. Bouchard, L. P. Perrault, R. Cartier, A. Basmajian, P. Page, and N. C. Poirier Tricuspid valve replacement: an analysis of 25 years of experience at a single center Ann. Thorac. Surg., January 1, 2003; 75(1): 47 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Shemin Tricuspid Valve Disease Card. Surg. Adult, January 1, 2003; 2(2003): 1001 - 1015. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
