ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Bartolo Zingone
Giuseppe Gatti
Elisabetta Rauber
Aniello Pappalardo
Bernardo Benussi
Lorella Dreas
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zingone, B.
Right arrow Articles by Dreas, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zingone, B.
Right arrow Articles by Dreas, L.
Related Collections
Right arrow Great vessels
Right arrow Valve disease

Ann Thorac Surg 2006;82:1709-1714
© 2006 The Society of Thoracic Surgeons

Original Articles: Cardiovascular

Surgical Management of the Atherosclerotic Ascending Aorta: Is Endoaortic Balloon Occlusion Safe?

Bartolo Zingone, MD, FETCS*, Giuseppe Gatti, MD, Elisabetta Rauber, MD, Aniello Pappalardo, MD, Bernardo Benussi, MD, Lorella Dreas, MD

Division of Cardiac Surgery, Ospedali Riuniti, Trieste, Italy

Accepted for publication May 25, 2006.

* Address correspondence to Dr Zingone, 22, vicolo Scaglioni, 34141 Trieste, Italy. (Email: bartolo.zingone{at}aots.sanita.fvg.it).


   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Comment
 References
 
BACKGROUND: Occlusion of the atherosclerotic ascending aorta by an endoaortic inflatable balloon has been proposed as an alternative to conventional cross-clamping to prevent injury to the vessel and distal embolization of debris. The safety and the effectiveness of endoaortic occlusion have not been documented in this setting.

METHODS: Endoaortic occlusion was employed in 52 of 2,172 consecutive patients. Surgeon's choice was based on preoperative identification of aortic calcifications or intraoperative epiaortic ultrasonographic scanning. Deaths and strokes were analyzed casewise and in aggregate.

RESULTS: In 10 patients (19.2%), the endoaortic balloon had to be replaced by the ordinary cross-clamp because of incomplete occlusion (n = 5), hindered exposure (n = 2), or balloon rupture (n = 3). In-hospital death occurred in 13 patients (25%), and stroke on awakening from anesthesia in 2 (3.8%). The death rate of patients treated by endoaortic occlusion was significantly higher compared with all other patients (4.2%, p < 0.0001) and with the expected estimate by European System for Cardiac Operative Risk Evaluation (10.5%, p = 0.05). By multivariable analysis, use of endoaortic occlusion was independently associated with in-hospital death (odds ratio = 5.609, 95% confidence interval: 2.684 to 11.719). Although the stroke rate was higher in the endoaortic occlusion group compared with all other patients, the difference was only possibly significant (3.8% versus 0.8%, p = 0.067).

CONCLUSIONS: In this series, the endoaortic occlusion was frequently ineffective, and was associated with a significantly higher risk of in-hospital death and a numerically higher risk of stroke.


   Introduction
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Comment
 References
 
Atherosclerotic changes in the ascending aorta may constitute a major problem during cardiac surgery. Whether in the form of eggshell calcification or of protruding atheromas, they frequently interfere with aortic cannulations and cross-clamping, access for aortic valve surgery, and proximal connections of coronary grafts. Technical problems aside, the risk of dislodging debris is cause for serious concern, and there is a growing evidence of a cause-and-effect relation between atherosclerotic aorta and stroke. Patients requiring heart valve surgery, eventually combined with coronary artery revascularization, constitute a particularly high risk subset [1].

Among alternative maneuvers devised to handle the atherosclerotic aorta, an endoluminal occluding balloon catheter has been proposed to substitute for the ordinary cross-clamp, on the assumption that it would be less traumatic and therefore less prone to cause systemic embolism [2, 3]. There are almost no data reported, however, as to the effectiveness and safety of endoaortic occlusion in the setting of a severely atherosclerotic aorta. As a part of our ongoing studies on the intraoperative assessment and management of difficult aortas, we sought to retrospectively assess the outcomes associated with the use of endoaortic occlusion.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Comment
 References
 
The standard Crafoord aortic cross-clamp was replaced with an endoaortic Foley or Pruitt (LeMaitre Vascular, Burlington, Massachusetts) occluding balloon catheter owing to the extent of aortic atherosclerosis in 52 (2.4%) of 2,171 patients operated upon in the period January 1, 2000, through August 31, 2004. Cases with acute aortic dissection, descending aortic aneurysm or rupture, combined cardiac and carotid procedures, and mediastinal procedures without cardiopulmonary bypass were excluded from this study (n = 64). Two patients with documented cerebral hemorrhage due to preoperative embolizing endocarditis or intraoperative uncontrollable hypertension were also excluded. A retrospective data review with individual consent waived was approved by the Ethics Committee.

Patients were selected for endoaortic balloon occlusion on the basis of a preoperative diagnosis of porcelain aorta or intraoperative findings at epiaortic ultrasonographic scanning. Any hint of calcific ascending atherosclerosis by plain chest X-ray film or cineangiography led to the performance of chest computed tomographic scan without contrast medium. Epiaortic ultrasonographic scanning replaced aortic palpation by January 2001, and has been used quite extensively ever since.

Epiaortic scan findings were coded on a 4-level scale of increasing severity and input separately for each of 12 ascending aortic segments in our clinical database. A summary score for epiaortic ultrasonographic scanning was computed (with normality = 12, and a theoretical maximum of 48), and a dichotomous variable was created for abnormal aorta (ie, at least 1 segment ≥ grade 3, where "≥ grade 3" stands for plaques ≥ 4 mm thick, or protruding ulcerated atheromas of any size).

In both the general population and the endoluminal occlusion group, cardiopulmonary bypass was performed with open circuit membrane oxygenation and filtered pericardial suction, allowing the temperature to drift unless deep hypothermia and circulatory arrest were in order. Cold blood cardioplegia was intermittently administered antegrade and retrograde during a single cross-clamp period, and this was also first choice approach for isolated coronary revascularization. While coronary artery grafting on the beating heart, with or without cardiopulmonary bypass assistance, was generally adopted to avoid aortic manipulations of atherosclerotic aortas, surgeons sometimes felt uncomfortable with that and resorted to endoaortic occlusion in the few patients included in this report.

Clinical, surgical, and outcome data were prospectively entered into both a clinical and an independent research database that were merged for this study. Events of interest, namely in-hospital death and early stroke, were retrospectively investigated in correlation with the use of endoaortic occlusion. Death occurring at any time during the hospital stay was categorized as early death. Stroke was defined as a new focal neurologic deficit or coma appearing during the awakening time from anesthesia and at least partially evident more than 24 hours after its onset. For patients staying in the intensive care unit longer than 2 days, the single, main clinical "reason" for it was selected from a prespecified list and prospectively entered into the database.

Categorical variables were compared by two-tailed Pearson's {chi}2 or Fisher's exact test as appropriate. Continuous variables were averaged by group and compared by the Mann-Whitney U test. Multivariable analysis was performed entering variables univariately associated with hospital death in a number of logistic regression models. A logistic estimate of the risk for early mortality was obtained by European System for Cardiac Operative Risk Evaluation (EuroSCORE) [4] and included as a further adjusting variable. Interactions of interest were then added. The SPSS 10.1 software (SPSS, Chicago, Illinois) was used throughout.


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Comment
 References
 
A Foley catheter was used for endoaortic occlusion in 30 patients and a Pruitt catheter in the remaining 22. As no difference could be found in regard to clinical characteristics and outcomes, both catheter types were studied in aggregate. Ten catheters (6 Foleys, 4 Pruitt) had to be replaced with a standard cross-clamp for incomplete occlusion (5), hindered exposure (2), or balloon rupture (3).

Epiaortic scan data were available for 1,060 patients, and 361 of them were found to have at least 1 segment of abnormal aorta, as defined earlier (Table 1). Of the 52 patients with endoaortic occlusion 44 had scan data available, and all of them had at least 1 segment of abnormal aorta. Figure 1 depicts the proportion of endoaortic occlusion cases by the number of abnormal aortic segments.


View this table:
[in this window]
[in a new window]
  		Table 1. Distribution of Abnormal Findings (at Least 1 Segment ≥ Grade 3) Across Ascending Aortic Segments in 1,060 Patients With Scan Data Available a
 

Figure 1
View larger version (18K):
[in this window]
[in a new window]
  		Fig 1. The use of endoaortic occlusion is shown as a proportion of cases sorted by the extension of aortic abnormalities, based on the 361 patients with at least one abnormal aortic segment by epiaortic ultrasonographic scanning. Numbers at risk are in italics. (Open portion of bars = percent who died.)

 
Additional maneuvers were altered in conjunction with endoaortic occlusion. In 29 patients, cannulation for arterial return was moved away from the usual site in the distal ascending aorta to either different sites in the aorta (n = 16), or to the right axillary (n = 12) or femoral artery (n = 1). Proximal anastomoses of vein or free arterial grafts were fashioned to an in situ internal thoracic artery graft in 12 patients. On several occasions, cardioplegia was only administered retrograde owing to lack of aortic access. Circulation was briefly set down to trickle flow when an ordinary clamp was needed to replace an ineffective balloon. Finally, in 1 patient, the ascending aorta was replaced under deep hypothermic circulatory arrest after temporary endoluminal occlusion.

Of the 52 patients with endoaortic occlusion, 13 died before hospital discharge (25%), and 2 more survived an early stroke (3.8%). Temporary and definitive endoaortic occlusion were aggregated for lack of difference in the occurrence of events (12 events among patients with endoaortic occlusion alone, and 3 after endoaortic and cross-clamping in sequence). Table 2 shows the rates of events according to surgical procedure. There was a trend toward a greater risk in the valve procedures, although small numbers precluded significance (p = 0.08 for in-hospital death between isolated coronary artery bypass graft and valves).


View this table:
[in this window]
[in a new window]
  		Table 2. Hospital Death and Early Stroke Rates
 
The rates of in-hospital death and early stroke were significantly higher in association with endoaortic occlusion compared with all other patients, although only probably so for early strokes (Table 2). Indeed, patients treated by endoaortic occlusion had a higher risk profile because of older age (74.5 ± 5.5 versus 67.1 ± 9.4 years, p < 0.0001), greater prevalence of serum creatinine 2.3 mg/dL or greater (5.8% versus 1.6%, p = 0.05), history of neoplastic disease (9.6% versus 3.5%, p = 0.04), extracardiac arteriopathy (46.2% versus 17.4%, p < 0.001), and of carotid artery disease in particular (25% versus 8.2%, p < 0.0001). Among procedural variables, duration of cardiopulmonary bypass was longer with endoaortic occlusion (123.2 ± 50.0 versus 105.5 ± 54.1 minutes, p = 0.02). By definition, the mean epiaortic ultrasonographic scanning score was significantly greater in patients treated by endoaortic occlusion (29.8 ± 5.6 versus 17.7 ± 6.7, p < 0.0001). In keeping, the expected death rate as estimated by EuroSCORE was higher, although not significantly so (10.5% versus 6.6%, p = 0.16) in the endoaortic occlusion group compared with all other patients.

The higher risk profile only partly explained the difference in death rates, however. The estimated death rate in the endoaortic occlusion group remained significantly higher than in all other patients even after risk-adjusting by EuroSCORE (Table 3). In fact, compared with expected estimates by EuroSCORE, observed death rates were significantly higher in the endoaortic occlusion group, while they were highly significantly lower in the remaining patients (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Observed Versus Expected Death Rates
 
Postmortem examination was performed in 10 of the 13 patients who died. The main reason for intensive care unit stay longer than 2 days and the attribution of death as formulated at the time of occurrence are listed in Table 4. Among patients who died, multiple organ failure as a reason for prolonged intensive care unit stay was more frequent among those with endoaortic occlusion compared with all others (53.8 versus 23.3, p = 0.04).


View this table:
[in this window]
[in a new window]
  		Table 4. Analysis of Deaths After Endoaortic Occlusion
 
Patients who died differed from survivors in respect to several variables (Table 5). By multivariable analysis, however, the use of endoaortic occlusion balloon catheter turned out to be the most powerful predictor of death. The analysis was repeated in the subset of patients who had undergone epiaortic ultrasonographic scanning, so that the extent of aortic disease could also be taken into account. Even so, the use of the balloon catheter was confirmed as the most powerful predictor of death, although the interaction of balloon occlusion with aortic valve replacement showed a slightly higher odds ratio when forced into the model before other procedural variables.


View this table:
[in this window]
[in a new window]
  		Table 5. Univariable and Multivariable Predictors of In-Hospital Death
 

   Comment
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
References
 
A number of alternatives may replace conventional aortic access when an atherosclerotic ascending aorta is encountered during cardiac surgery. Solutions for patients requiring isolated coronary artery bypass graft are conceptually quite simple when off-pump procedures can be performed exclusively by arterial pedicled and composite grafts [5, 6]. If needed, circulatory assistance can be provided through axillary artery cannulation, while branching free venous or arterial grafts to an in situ internal thoracic artery can help skipping the aorta [7]. Also, using a healthy island of aorta for a single graft connection to which subsequently branch all other needed grafts only requires a few minutes of trickle circulatory flow under mild hypothermia [8]. As for open-heart surgery, the mitral valve can be approached without aortic cross-clamping in the fibrillating heart, eventually facilitated by low-flow hypothermic cardiopulmonary bypass [9, 10]. More serious problems arise with surgery of the aortic valve, when avoidance of cross-clamping requires establishing circulatory arrest under deep hypothermia. A variety of options are then available, from placing a clamp after inspecting the aorta from inside, eventually after a localized endarterectomy, to going ahead with aortic valve replacement or with ascending aortic replacement preliminary to the valve procedure [11-14].

Occluding the aorta endoluminally with a balloon catheter is a further option associated, in the mind of its proponents, with a degree of trauma to the aortic wall perceived as being considerably less than with an ordinary cross-clamp [2, 3]. Pertinent to the fact that a consistent proportion of bad aortas are discovered only intraoperatively, thus preventing the possibility of scheduling those cases for the most experienced surgeons, the idea that endoluminal aortic occlusion implies a minor and limited derangement to the ordinary flow of operative steps is perhaps what makes endoluminal occlusion particularly appealing for the less experienced surgeons.

Despite the growing prevalence of aortic atherosclerosis in an aging surgical population, endoaortic occlusion has only been the subject of a few case reports [2, 12, 15, 16], and of a single series including 22 patients undergoing aortic endarterectomy after temporary endoaortic occlusion [17]. In the latter series, 1 patient died and 1 had transient hemiparesis, whereas among 33 additional patients in whom aortic endarterectomy was only considered, 4 suffered perioperative neurologic injury and 3 died of it [17]. As the main focus of the study was on aortic endarterectomy, it was not clearly stated whether endoaortic occlusion had been retained as part of the protocol in the second group of patients. It remained for Moro and coworkers [18] to voice the limitations of this approach in a further case report.

At variance with the generally favorable assessment emerging from these studies, the use of endoaortic occlusion has been associated with disappointing results in our hands. From the technical point of view, the occlusion has been frequently incomplete, and the exposure of the aortic root hindered by the leak around the balloon or its protrusion through the aortotomy. Enhancing occlusion by further inflating the balloon increased the protrusion, while deflating the balloon increased the leak, so that it had to be eventually exchanged with an ordinary aortic cross-clamp placed under vision in about 1 every 5 patients. All in all, the balloon constituted an encumbering presence requiring time and attention without necessarily ensuring a good service in return.

Of greater concern was the significant association of endoaortic occlusion with unfavorable events. Although patients treated in this manner exhibited a higher risk profile, the death rate was disproportionately higher than expected estimates. While it may be questionable that such an excess death rate be attributed to embolic events caused by the balloon, it is worth emphasizing that nonfatal strokes and the finding of multiple visceral infarcts at postmortem strongly supports embolism as a relevant phenomenon in one third of the patients with adverse events following endoaortic occlusion. Furthermore, early postoperative multiple organ failure more frequently preceded death in patients treated by endoaortic occlusion compared with all others, although, once again, the lack of more detailed autopsy data prevents a better insight as to the link with the potential microembolic load.

There are several limitations to this study that should be considered while interpreting its results. First, although clinical and outcome data were collected prospectively, there was no specific set of criteria dictating which patients would be treated by endoaortic occlusion. Basically, endoaortic occlusion acted a residual option when all others would not satisfactorily apply. In fact, it was reserved to highly selected cases as shown by the extension of their aortic changes and the limited proportion of cases with abnormal aorta so treated. Second, adverse outcomes were undoubtedly related to the aortic pathology no less than to the method of aortic occlusion, so much so that extending our conclusions to different clinical settings would be totally unwarranted. Third, neither transcranial Doppler studies during surgery nor specifically oriented postmortem investigations were available to demonstrate an embolic mechanism behind negative outcomes. Fourth, although three senior surgeons with a common background were involved in 88% of patients treated by endoluminal occlusion in this series, the weight of experience could have led other surgeons to different solutions—and outcomes. No-clamp replacement of the atherosclerotic aorta, for instance, was performed successfully in 5 patients in the most recent part of this series, and has become our first choice thereafter. With this option at hand, the selection for endoluminal occlusion may change—and so might its results.

Thus, despite the quite compelling statistical association, this study cannot prove that the use of endoaortic occlusion of atherosclerotic ascending aorta is causally related to perioperative systemic embolism and death. While its role as a cause of adverse events remains uncertain, however, it seemed quite evident to us that endoaortic occlusion is not part of the solution either. Putting together the disagreeable association with unfavorable outcomes and the less than desirable effectiveness, and while awaiting better evidence, we have virtually discontinued the use of endoaortic occlusion in our practice.


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
 References
 

  1. Wolman R, Nussmeier N, Aggarwal A, et al. Cerebral injury after cardiac surgeryIdentification of a group at extraordinary risk. Stroke 1999;30:514-522.[Abstract/Free Full Text]
  2. Erath Jr HG, Stoney WS. Balloon catheter occlusion of the ascending aorta Ann Thorac Surg 1983;35:560-561.[Abstract]
  3. Cosgrove DM. Management of the calcified aorta: an alternative method of occlusion Ann Thorac Surg 1983;36:718-719.[Abstract]
  4. Roques F, Michel P, Goldstone AR, Nashef SAM. The logistic EuroSCORE Eur Heart J 2003;24:1-2.[Free Full Text]
  5. Hangler HB, Nagele G, Danzmayr M, et al. Modification of surgical technique for ascending aortic atherosclerosis: impact on stroke reduction in coronary artery bypass grafting J Thorac Cardiovasc Surg 2003;126:391-400.[Abstract/Free Full Text]
  6. Fukui T, Takanashi S, Hosoda Y, Suehiro S. Total arterial myocardial revascularization using composite and sequential grafting with the off-pump technique Ann Thorac Surg 2005;80:579-585.[Abstract/Free Full Text]
  7. Mills NL, Everson CT. Atherosclerosis of the ascending aorta and coronary artery bypassPathology, clinical correlates, and operative management. J Thorac Cardiovasc Surg 1991;102:546-553.[Abstract]
  8. Antunes P, Ferrao de Oliveira J, Antunes M. Predictors of cerebrovascular events in patients subjected to isolated coronary surgeryThe importance of aortic cross-clamping. Eur J Cardiothorac Surg 2003;23:328-333.[Abstract/Free Full Text]
  9. Cohn LH, Peigh PS, Sell J, DiSesa VJ. Right thoracotomy, femorofemoral bypass, and deep hypothermia for re-replacement of the mitral valve Ann Thorac Surg 1989;48:69-71.[Abstract]
  10. Ogino H, Ueda Y, Morioka K, Matsubayashi K, Nomoto T. Mitral valve repair in a patient with severe porcelain aorta Ann Thorac Surg 1997;64:1179-1181.[Abstract/Free Full Text]
  11. Aranki SF, Nathan M, Shekar P, Couper G, Rizzo R, Cohn LH. Hypothermic circulatory arrest enables aortic valve replacement in patients with unclampable aorta Ann Thorac Surg 2005;80:1679-1687.[Abstract/Free Full Text]
  12. Gillinov AM, Lytle BW, Hoang V, et al. The atherosclerotic aorta at aortic valve replacement: surgical strategies J Thorac Cardiovasc Surg 2000;120:957-965.[Abstract/Free Full Text]
  13. Wareing TH, Davila-Roman VG, Daily BB, et al. Strategy for the reduction of stroke incidence in cardiac surgical patients Ann Thorac Surg 1993;55:1400-1408.[Abstract]
  14. Girardi LN, Krieger KH, Mack CA, Isom OW. No-clamp technique for valve repair or replacement in patients with a porcelain aorta Ann Thorac Surg 2005;80:1688-1692.[Abstract/Free Full Text]
  15. Black JJM, Allan A, Williams BT. The use of Foley balloon catheters in cardiac surgery Eur J Cardiothorac Surg 1993;7:617-619.[Abstract]
  16. Liddicoat JR, Doty JR, Stuart RS. Management of the atherosclerotic ascending aorta with endoaortic occlusion Ann Thorac Surg 1998;65:1133-1135.[Abstract/Free Full Text]
  17. Vogt PR, Hauser M, Schwarz U, et al. Complete thromboendarterectomy of the calcified ascending aorta and aortic arch Ann Thorac Surg 1999;67:457-461.[Abstract/Free Full Text]
  18. Moro H, Hayashi J, Hirahara H. Endoaortic occlusion for the calcified thoracic aorta Ann Thorac Surg 1999;67:1536.[Free Full Text]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
S. Goland, A. Trento, L. S.C. Czer, S. Eshaghian, K. Tolstrup, T. Z. Naqvi, M. A. De Robertis, J. Mirocha, K. Iida, and R. J. Siegel
Thoracic Aortic Arteriosclerosis in Patients With Degenerative Aortic Stenosis With and Without Coexisting Coronary Artery Disease
Ann. Thorac. Surg., January 1, 2008; 85(1): 113 - 119.
[Abstract] [Full Text] [PDF]

Home page
 Eur. J. Cardiothorac. Surg.Home page
B. Zingone, E. Rauber, G. Gatti, A. Pappalardo, B. Benussi, G. Forti, U. Tognolli, and M. Gabrielli
Diagnosis and management of severe atherosclerosis of the ascending aorta and aortic arch during cardiac surgery: focus on aortic replacement
Eur. J. Cardiothorac. Surg., June 1, 2007; 31(6): 990 - 997.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Bartolo Zingone
Giuseppe Gatti
Elisabetta Rauber
Aniello Pappalardo
Bernardo Benussi
Lorella Dreas
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zingone, B.
Right arrow Articles by Dreas, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zingone, B.
Right arrow Articles by Dreas, L.
Related Collections
Right arrow Great vessels
Right arrow Valve disease

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