| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;76:471-477
© 2003 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
* Address reprint requests to Dr Fullerton, Division of Cardiothoracic Surgery, Northwestern University Feinberg School of Medicine, 201 East Huron St, Suite 10-105, Chicago, IL 60611, USA.
e-mail: dfullert{at}nmh.org
Presented at the Forty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Miami Beach, FL, Nov 7–9, 2002.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
METHODS: The Ross procedure was performed as a root replacement. Without regard for patient body surface area, the aortic annulus was plicated to 23 mm and externally buttressed with felt. Geometric mismatch of the distal autograft anastomosis was avoided by liberal use of a synthetic interposition graft, and the anastomosis was also externally buttressed with felt. An over-sized pulmonary homograft (27 to 28 mm) was routinely used to reconstruct the right ventricular outflow tract.
RESULTS: Forty-four consecutive patients (27 men and 17 women; mean age, 49 ± 9 years) were operated on between January 1997 and March 2002. Mean follow-up was 38 ± 5 months. Twenty-nine patients had aortic stenosis and 15 had aortic regurgitation. Aortic annular plication was done in 41 (93%) and an aortic interposition was used in 14 (32%). There were three hospital deaths, with no subsequent deaths. Only 1 patient required reoperation 2.5 years postoperatively from recurrent endocarditis. No patient has more that "trivial" autograft insufficiency, and the mean autograft gradient was 7 ± 3 mm Hg. No patient has significant pulmonary homograft stenosis.
CONCLUSIONS: Geometric matching of the aortic and pulmonary roots may be readily accomplished using a standardized approach to the Ross procedure. In turn, this may optimize the durability of the operation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
It has recently become appreciated that geometric matching between the aortic and pulmonary roots is necessary to achieve optimal autograft durability [1]. The pulmonary autograft may become insufficient if its leaflets are prevented from coaptation by either of two mechanisms [1]: the size of the pulmonary annulus is less than the aortic annulus or [2] the sinotubular junction of the autograft is stretched [1]. When size disparity between the pulmonary and aortic roots does exist, techniques have been recommended to correct the mismatch, and most require accurate measurement of the pulmonary autograft [2–5].
Acknowledging this geometric importance, it may nonetheless be difficult to accurately measure the appropriate diameters of the pulmonary autograft. At the same time, the reproducibility of any operation is optimized if the surgical technique may be standardized. Incorporating the techniques of several surgeons, we standardized our approach to the Ross procedure. Our protocol was to: (1) "fix" the aortic annulus at a consistent size (23 mm diameter), regardless of the patient’s body surface area (BSA; (2) prevent stretching of the sinotubular junction by liberal use of an interposition graft if necessary; and (3) routinely use an oversized pulmonary homograft (27 to 28 mm diameter) in an effort to try to minimize the development of pulmonary stenosis. Our purpose is to report our intermediate-term results with this standardized technique.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Cardiopulmonary bypass and myocardial protection
Moderate systemic hypothermia was used (bladder temperature, 25°C) during cardiopulmonary bypass using bicaval cannulation; a left ventricular vent was placed in the right superior pulmonary vein. Myocardial protection was accomplished by cold-blood cardioplegia and topical cold saline. All patients received aprotinin (Bayer Corporation, West Haven, CT). After an initial administration of antegrade cardioplegia, retrograde cardioplegia was given every 20 minutes. Patients with aortic regurgitation received only retrograde cardioplegia. The myocardial temperature was monitored in the interventricular septum and was below 10°C throughout the period of aortic occlusion. Before release of the aortic cross-clamp, a terminal "hot shot" of warm blood cardioplegia enriched with L-arginine and L-aspartate was administered in retrograde fashion.
Associated cardiac anomalies were corrected after performing the Ross procedure.
Technique of the ross procedure
All Ross procedures were performed as a root replacement. Our first objective was to consistently provide a secure proximal autograft anastomosis that would not enlarge with time and in which the pulmonary autograft annulus had a greater diameter than the aortic annulus. In this manner, we hypothesized that the valve leaflets of the autograft would not be pulled apart at the level of the annulus.
It was our observation that the annulus of the pulmonary autograft is consistently greater that 23 mm in diameter. Further, our experience with aortic allografts suggested that a 23-mm diameter aortic allograft provides excellent hemodynamics in patients regardless of BSA. This observation has also been suggested by others [6]. We therefore concluded that the Ross procedure could be performed by routinely creating an aortic annulus with a 23-mm diameter, rather than trying to accurately match the diameters of the aortic and pulmonary autograft annuli. This would achieve appropriate hemodynamics and an aortic annulus slightly smaller than that of the pulmonary autograft.
Using a prosthetic valve sizer, the aortic annulus was sized to a 23 sizer (St. Jude Medical, St. Paul, MN) in all patients by plicating each commissure with a horizontal mattress stitch of 3-0 monofilament suture (USSC, Norwalk, CT) (Fig 1). In patients with a biscupid aortic valve, the raphe of the incomplete leaflet was plicated as well. The proximal autograft anastomosis was performed using interrupted 4-0 suture (USSC). Each needle of the double-armed 4-0 sutures securing the proximal autograft anastomosis was brought through a 1-cm-wide strip of surgical Teflon felt (Impra, Tempe, AZ), which externally encircled the anastomosis. Thereby, the proximal autograft anastomosis was externally reinforced for both hemostasis and to further prevent enlargement of the aortic annulus.
|
When reconstructing the right ventricular outflow tract, an effort was made to over-size the pulmonary homograft by routinely using a 27- to 28-mm cryopreserved pulmonary homograft (CroLife, Inc, Marrietta, GA). Homografts were not specifically type-matched for the patient’s blood type.
After completion of the procedure and separation from cardiopulmonary bypass, the functions of the pulmonary autograft and pulmonary homograft were assessed using intraoperative transesophageal echocardiography by a cardiologist. Follow-up was achieved by personal contact with the patients of the attending cardiologist. Transthoracic echocardiograms were performed at 2 and 6 months postoperatively and annually thereafter. Data are expressed as means ± SEM
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
All patients were symptomatic from aortic valve disease with signs and symptoms of heart failure. Fifteen patients (34%) had aortic regurgitation and 29 patients (66%) had aortic stenosis. Fifteen patients (34%) had biscuspid aortic valves and 8 (18%) had aortic annulus dysplasia.
Twenty patients (45%) were in New York Heart Association (NYHA) class II, 21 patients (48%) were in NYHA class III, and 3 patients (7%) were in NYHA class IV. All 3 patients in NYHA class IV were preoperatively hospitalized in the intensive care unit, receiving intravenous medications to control heart failure and pulmonary edema. Two patients had active aortic valve endocarditis, both with Streptococcal species, and underwent urgent operations. The remainder underwent elective operations.
In addition to preoperative transthoracic or transesophageal echocardiography, all patients underwent preoperative cardiac catheterization with left and right heart catheterization, left ventriculography, and coronary angiography. Particular notation was made on the preoperative coronary angiogram of the size and location of the first septal perforator branch of the left anterior coronary artery in order to preserve it from harm’s way during the operation. Even patients with a very large septal perforator successfully underwent the procedure. No patient had more than single-vessel coronary disease, and 5 had pulmonary hypertension (mean pulmonary artery pressures > 30 mm Hg; range, 32 to 38 mm Hg). The mean preoperative left ventricular ejection fraction (LVEF) (left ventriculogram) was 42% ± 6% (range, 25% to 55%). Eight patients, all with severe aortic regurgitation, had significantly diminished preoperative LVEF between 25% and 35%.
After resection of the aortic valve leaflets, the aortic annulus was downsized if it was greater than 23-mm diameter. The largest annulus measured approximately 33-mm diameter before downsizing. The aortic annulus was plicated in 41 patients (93%) in addition to the external felt buttress; 3 patients already had an annular diameter measuring 23 mm and the annulus was simply buttressed with Teflon felt. An interposition graft was placed in 14 patients (32%); 8 had true aneurysms of the ascending aorta.
The mean aortic occlusion time for all patients was 215 ± 15 minutes. All patients received blood products. The median length of hospital stay was 6 days, with a range of 4 to 30 days.
Associated procedures
Eight patients (18%) had an ascending aortic aneurysm ranging in diameter (preoperative computed tomography scan) from 55 mm to 80 mm in diameter. In each case, the aneurysm was resected and replaced with an interposition graft sized to the diameter of the pulmonary autograft. Five patients (11%) underwent single coronary artery bypass grafting, 2 with use of the left internal mammary artery grafted to the left anterior descending coronary and 3 with saphenous vein grafts to other vessels. Three patients (7%) underwent patch closure of a secundum atrial septal defect, and 1 patient underwent closure of patent ductus arteriosus.
Postoperative complications
There were no perioperative myocardial infarctions, and no infections. No patient required a permanent pacemaker. Six patients (14%), all with a preoperative LVEF below 35%, required inotropic support (milrinone and epinepherine) after cardiopulmonary bypass. One of these patients required an intraaortic balloon pump for 24 hours postoperatively, and made an uneventful recovery thereafter.
Two patients had more than transient postoperative atrial fibrillation. One of these patients (63-year-old male) sustained a small stroke when going in and out of atrial fibrillation; his left upper extremity paresis completely resolved. The second patient (61 year-old-male) ultimately underwent a catheter-based ablation of his atrial fibrillation.
One patient (59-year-old male) required a pericardial window performed 6 weeks postoperatively to evacuate a large pericardial effusion. One patient (43-year-old female) was returned to the operating room on the first postoperative day to evacuate a new right hemothorax; no bleeding site was identified and she was discharged to home on the fifth postoperative day. One patient (58-year-old female) required tracheostomy for prolonged ventilatory support. This patient ultimately died.
There were three hospital deaths. The first occurred in a 52-year-old male from cardiac tamponade after removal of his temporary pacing wires on postoperative day 5. He was brain dead after successful cardiac resuscitation. The second death was in the 58-year-old female noted above. Thirty days postoperatively and 2 weeks after tracheostomy, she died of exsanquination when a chest tube eroded into an intercostal artery. The third death was a 43-year-old male who was ambulatory on his first postoperative day after an uneventful operation but died of micturition syncope on the second postoperative day. Autopsies were performed on all 3 patients. Each patient had severe left ventricular hypertrophy. In no case did the pathologist identify a problem with the heart or surgical repair. There was specifically no evidence of myocardial infarction, suture line dehiscence, or aortic dissection.
Early autograft and homograft function
All patients were studied with intraoperative transesophageal echocardiography. Immediately after the procedure, 34 patients (77%) had no autograft regurgitation and 10 patients (23%) had "trivial" autograft regurgitation. Acknowledging the limitations of transesophageal echocardiography to accurately assess the pulmonary valve, no patient had immediate stenosis or regurgitation of the pulmonary homograft.
Follow-up
There have been no late deaths (Fig 2).
All patients are in NYHA class I. Four patients who initially had no autograft regurgitation have progressed to "trivial" regurgitation. No patient has had a mean gradient across the autograft greater than 10 mm Hg (Fig 3). The diameter of the autograft annulus has been stable in follow-up in all patients. Echocardiocardiographic assessment of the diameter of the distal autograft anastomosis and the sinuses of Valsalva are not available in all patients. But in no patient was either noted to be dilated in follow-up.
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Although the results of the present report are encouraging, its limitation is the lack of long-term follow-up. Follow-up of the Ross procedure must consider both the pulmonary autograft and the pulmonary homograft. Although autograft insufficiency is most commonly found early after the operation, it may develop in longer-term follow-up [9–12]. Further, recent reports have cautioned that pulmonary homograft stenosis may not become apparent until several years after the procedure [13, 14]. Nonetheless, the present series suggests that a standardized approach may be easily used to effectively avoid geometric mismatching with the Ross procedure. In turn, this may contribute to reliable durability after the operation.
Several authors have reported early autograft failures, particularly in situations in which the autograft was sewn into an enlarged aortic annulus [1–4]. In these series, modification of the surgical technique was associated with improved surgical outcomes.
These modifications have typically incorporated an aortic annuloplasty to downsize an enlarged aortic annulus to more closely match the diameter of the pulmonary autograft [2, 5]. Some authors have likewise recommended external reinforcement of the proximal suture line with surgical felt or Dacron material to buttress it against dilatation [4]. Collectively, the experience of these authors emphasizes the importance of geometric matching of the aortic and pulmonary roots in achieving a durable result.
Geometric mismatch of the aortic and pulmonary roots may lead to autograft insufficiency for several reasons. The absolute diameter of the aortic and pulmonary annuli may differ significantly, especially in patients with aortic valve insufficiency or aortic annulus dysplasia [2]. Approximately 55% of the circumference of the aortic annulus is supported by fibrous tissue, but the pulmonary annulus is completely supported by muscle of the right ventricular outflow tract [15]. Hence, if the muscular pulmonary annulus is sewn to an oversized aortic annulus, the pulmonary annulus may dilate when used as an autograft. In fact, Hokken and colleagues have reported significant increases in pulmonary autograft annular diameter over time when measured by echocardiography in patients after the Ross procedure [16]. To correct for such annular size disparity, techniques have been advocated that either downsize the aortic annulus to more closely match the diameter of the pulmonary annulus or provide an external buttress to the proximal autograft anastomosis [1–3].
In the present series, an effort was made to incorporate the principles of all of these surgical techniques into a standardized approach. From a practical standpoint, however, it may be difficult to accurately measure the pulmonary annular diameter and assure that it is unchanged by implantation as an autograft. Further, we sought to standardize our surgical technique for this operation in an effort to achieve reproducible results. We hypothesized that creation of a standard diameter of the aortic annulus could be done more easily and reliably than attempting to accurately match the annular diameter of the aorta with that of the pulmonary autograft. We observed that the diameter of the proximal pulmonary autograft is consistently at least 23 mm, which could be assured by taking a little extra muscle around its perimeter from the right ventricular outflow tract. This observation was in concert with our experience and that of others [6] suggesting that a 23-mm aortic allograft provides excellent hemodynamics in patients regardless of BSA. We therefore proceeded under the assumption that a 23-mm aortic annular diameter would achieve appropriate hemodynamics and simultaneously provide an aortic annulus slightly smaller than that of the pulmonary autograft. In this way, the valve leaflets of the autograft would not be distracted at the level of the annulus. With this technique, no patient has autograft stenosis or regurgitation. Given the excellent flow dynamics inherent to the pulmonary autograft, the results of the present series suggests that a 23-mm aortic annular diameter appears to be sufficient over a broad range of BSAs.
The diameter of the sinotubular junction of the autograft is the second geometric consideration. If it is stretched open, the leaflets of the valve will be pulled apart and the valve made insufficient [1, 17, 18]. An aortoplasty may be performed at the distal autograft anastomosis in an effort to match the annular diameters of the aorta and autograft. Nonetheless, the pulmonary autograft may dilate under systemic loading conditions [9, 17]. Whereas dilation of the sinuses of valsava alone may not produce valvular insufficiency, dilation of the sinotubular junction may [18, 19]. Some have suggested this is more likely to occur in patients with a bicuspid aortic valve [17]; however, this has not been confirmed [20].
Again from a practical standpoint, it may be difficult to accurately match the size of the aorta and pulmonary artery at the distal autograft anastomosis. We therefore routinely placed a synthetic interposition graft between the aorta and autograft if the aortic diameter was greater than approximately 30 mm. The anastomosis was created as close to the sinotubular junction as possible and fixed with external surgical felt reinforcement. In the present series, this approach permitted application of the Ross procedure even in patients with large aneurysms of the ascending aorta. Acknowledging the length of follow-up of the present series, progressive autograft dilation in patients with a bicuspid aortic valve has not been seen.
The performance of the cryopreserved pulmonary homograft must be followed as well. Some [13, 14], but not all [21], investigators have reported stenosis of the pulmonary homograft with longitudinal follow-up. Use of homografts has been associated with a humoral immune response (anti-HLA antibodies) after the Ross procedure, but this response was not been conclusively linked to pulmonary homograft dysfunction [22]. Because human valves do not appear to have ABO antigens [23], it seems unlikely that ABO blood type incompatibility is an overriding consideration. In the present report, no attempt was made to use blood type–specific homografts. Recent data suggest that multiple factors may combine to contribute to late pulmonary homograft stenosis, including smaller homograft size [13]. In the present series, we chose to oversize the homograft and routinely placed a 27- or 28-mm pulmonary homograft. Although no patients have developed significant pulmonary homograft stenosis, longer follow-up is necessary.
In summary, the Ross procedure may be utilized with excellent results in adults. Nonetheless, it is a complex operation. To assure competence of the pulmonary autograft, geometric mismatching of the pulmonary and aortic roots must be avoided. We conclude that reproducible geometric matching of the aortic and pulmonary roots may be readily accomplished by use of a standardized approach to the operation. ([8])
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Second, your manuscript and your data were extremely complete and helpful to us, and will be to us as we look at things. I have a few questions.
First, how did you select your patients for a Ross versus other options since you clearly must have done more aortic valve operations in your institution during this time?
Second, why did you not choose to make the distal autograft anastomosis smaller than, say, the 30 mm you arbitrarily chose?
And third, in view of what some say about a bicuspid aortic valve and large aortas and their walls, have you given more pause to applying the Ross in this subset?
I would like to thank you for this fine addition to our understanding of the pulmonary autograft and its application. Thanks.
DR FULLERTON: Thank you, Dr Calhoon, for your remarks. To try to address your questions, I think that as our experience accrued during this period of time, our application of the technique became more focused on patients of a younger age group. In our own institution, having looked at these data recently, about 80% of all the aortic valve surgery that we do now is in people above the age of 70 years. Since we initiated this series, I think the data that have become available regarding the durability of porcine valves as well as bovine pericardial valves are really quite excellent, particularly in that age group, and that has become our preferred valve replacement device in those patients. Patients who are below the age of 60 years now are considered candidates for the Ross procedure in our own institution.
We chose to use the technique we did at the distal anastomosis somewhat arbitrarily, in all honesty. It seemed that if we measured an aortic diameter of about 30 mm before creating the anastomosis, it was possible to downsize that a bit during the technical application of the anastomosis. So we did not feel it was necessary to go to greater lengths to decrease the size of that anastomosis, and we are optimistic that buttressing it with felt, as others have recommended, would secure it.
We are certainly aware of the data that have emerged regarding patients with bicuspid aortic valves, and we need to follow our patients longer to assure that we do not have a problem. But having looked at this, in the 15 patients who did have a preoperative bicuspid aortic valve, we have not found much dilatation in the aortic root beyond any of the other groups. So we will just have to see.
DR DAVIS C. DRINKWATER (Nashville, TN): I enjoyed your talk very much, Dave, and congratulations on excellent results. I wanted to say that we also participate in your enthusiasm for selective usage of the Ross Procedure in less than 60-year-old patients, who for example wish to avoid anticoagulation. Several technical points that we may differ in is our use of autologous glutaraldehyde-treated pericardium for anastomotic buttressing. This is particularly useful in the case of endocarditis and makes this a great operation in our view for endocarditis patients who are hemodynamically stable. Did the use of felt pledgets in the case of the recurrent endocarditis, which you reported, have any role to play?
The other technical variance that we find useful in the case of bicuspid aortic dilation is a V-plasty resection and external wrap with prolene mesh, such as reported by Robiscek some years ago. This can frequently avoid the use of synthetic graft material and more importantly shorten what is generally a lengthy operation. I wonder about your thoughts on this technique.
Finally, just a brief comment that the mortalities you cited are unique and honest, but do not relate specifically to the Ross procedure. I think, therefore, it should not reflect poorly on what is otherwise an excellent operation in selected patients. Thank you.
DR FULLERTON: Well, thank you very much. This series ended with our experience in March 2002, and since then, we have performed the Ross procedure on an additional 3 patients with endocarditis at the time of surgery. In those patients, I have used pericardium in 2 of them rather than the felt, and it is for that concern. I honestly do not know if felt played a role in this woman, who had recurrent endocarditis, but obviously it is a synthetic material and intuitively one has to think that that may augment the risk.
We do not have any experience in wrapping the autograft in an effort to avoid the interposition graft. I think that is an intriguing idea. We just have not done that yet.
DR JOHN D. OSWALT (Austin, TX): David, congratulations. I cannot pass up the opportunity to congratulate you on embracing the Ross procedure for endocarditis. As you are aware, we started doing that 13 years ago, and the thing that I find that is so encouraging about it is that it is true that the aortic homograft functions very well and still is, I would say, the valve of choice by most of the people in the world.
However, I think we are overlooking the fact that they are going to have to have a second operation. When you have had somebody that has had a destroyed root and serious pericardial inflammation secondary to their endocarditis and then do a root replacement with a homograft, there is going to be significant adhesions. Now, you know that later on you are going to have to go and do that again. The Ross procedure is the one operation that gives them the opportunity at least to possibly have a single operation.
And the recurrence rate in our series, which is over 50 patients now, shows no recurrences in the aortic position, and only one recurrence in the pulmonary homograft.
Thank you very much and congratulations.
DR FULLERTON: Thank you.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. J.M. Takkenberg, L. M.A. Klieverik, P. H. Schoof, R.-J. van Suylen, L. A. van Herwerden, P. E. Zondervan, J. W. Roos-Hesselink, M. J.C. Eijkemans, M. H. Yacoub, and A. J.J.C. Bogers The Ross Procedure: A Systematic Review and Meta-Analysis Circulation, January 20, 2009; 119(2): 222 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Brown, M. Ruzmetov, M. D. Rodefeld, and M. W. Turrentine Right Ventricular Outflow Tract Reconstruction in Ross Patients: Does the Homograft Fare Better? Ann. Thorac. Surg., November 1, 2008; 86(5): 1607 - 1612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Frigiola, M. Ranucci, C. Carlucci, A. Giamberti, R. Abella, and M. Di Donato The Ross Procedure in Adults: Long-Term Follow-Up and Echocardiographic Changes Leading to Pulmonary Autograft Reoperation Ann. Thorac. Surg., August 1, 2008; 86(2): 482 - 489. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Brown, M. Ruzmetov, M. D. Rodefeld, Y. Mahomed, and M. W. Turrentine Incidence of and Risk Factors for Pulmonary Autograft Dilation After Ross Aortic Valve Replacement Ann. Thorac. Surg., May 1, 2007; 83(5): 1781 - 1789. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Rocha-e-Silva, P. M.A. Pomerantzeff, R. T. Munhoz, D. D. Lourenco Filho, L. F. Caneo, and S. A. de Oliveira Ross Procedure and Ventricular Septal Defect Correction With Prolapsed Cusp Ann. Thorac. Surg., July 1, 2005; 80(1): 330 - 331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G. Gleason, T. E. David, J. S. Coselli, J. W. Hammon Jr, and J. E. Bavaria St. Jude Medical Toronto biologic aortic root prosthesis: Early FDA phase II IDE study results Ann. Thorac. Surg., September 1, 2004; 78(3): 786 - 793. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
