Ann Thorac Surg 2008;86:95-101. doi:10.1016/j.athoracsur.2008.03.037
© 2008 The Society of Thoracic Surgeons
Original Articles: Adult Cardiac
Combining Classic Surgery With Descending Stent Grafting for Acute DeBakey Type I Dissection
Heinz Jakob, MD, PhDa,*,
Konstantinos Tsagakis, MDa,
Paschalis Tossios, MDa,
Parwis Massoudy, MD, PhDa,
Matthias Thielmann, MDa,
Thomas Buck, MD, PhDb,
Holger Eggebrecht, MD, PhDb,
Markus Kamler, MD, PhDa
a Department of Thoracic and Cardiovascular Surgery, West-German Heart Center Essen, University Hospital Essen, Essen, Germany
b Department of Cardiology, West-German Heart Center Essen, University Hospital Essen, Essen, Germany
Accepted for publication March 20, 2008.
* Address correspondence to Dr Jakob, Klinik für Thorax- und Kardiovaskuläre Chirurgie, Westdeutsches Herzzentrum Essen, Hufelandstr 55, Essen, D-45121, Germany (Email: heinz.jakob{at}uk-essen.de).
Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
| Dr Jakob discloses that he has a financial relationship with Jotec GmbH.
|
 |
Abstract
|
|---|
Background: To possibly prevent late complications after classic type A aortic dissection repair, the radical concept of ascending/arch replacement with simultaneous antegrade descending stent grafting using a hybrid prosthesis was applied and compared with conventional repair leaving the distal false lumen untreated.
Methods: Between January 2001 and October 2007, of 71 consecutive patients with acute type A aortic dissection (AAAD), 45 had DeBakey type I dissection and underwent emergency surgery within 24 hours after onset of symptoms. These patients were separated into group 1 (n = 23) undergoing conventional surgery, and group 2 (n = 22) undergoing combined repair with antegrade stent grafting.
Results: Patients were comparable for baseline characteristics, but more group 2 patients had severely compromised hemodynamics (p = 0.05) and cerebral malperfusion at arrival (p < 0.01). Intraoperative and postoperative characteristics were similar, with an overall hospital mortality of 16% (5 [22%] versus 2 [9%], group 1 versus group 2; p = 0.22). At a mean follow-up time of 48 months for group 1 versus 23 months for group 2 (p < 0.01), late mortality did not differ between groups (p = 0.38) and was mainly related to additional surgical procedures and persisting neurologic sequelae and not to the aortic pathology. Persisting distal false lumen patency was observed in 89% of group 1 versus 10% of group 2 patients (p < 0.01).
Conclusions: This hybrid approach to patients with type I acute aortic dissection is technically feasible without increasing the operative risk and offers the chance of persistent occlusion of the persistent graft distal false lumen.
Reported hospital mortality of surgical treatment of acute type A aortic dissection (AAAD) ranges between 15% and 30%, not very much changing over the last years [1–3]. In contrast to the medical treatment of such patients resulting in 58% mortality, the surgical results compare favorably, but do not satisfy [4].
Patient selection with optional deferral of operation might be one way to improve on surgical results, but does not meet our ethical intention to save our patients' lives [5]. In addition, patients surviving the classic surgical procedure consisting of ascending aortic replacement with or without arch replacement are facing a substantial risk of late complications such as aneurysm formation of the continuously perfused false lumen, rupture, malperfusion sequelae, and often require redo surgery [6–8]. Reoperation after previous AAAD surgery represents another surgical challenge with a complex situation either requiring a major surgical approach consisting of a one-stage operation with bilateral clamshell incision or a two-stage strategy, both associated with major surgical trauma and a substantial perioperative morbidity and mortality rate [9, 10].
Credit has to be given to Kato and others [11, 12], who developed a surgical strategy consisting of the combination of the radical surgical approach to replace the ascending aorta and the arch with antegrade descending aortic stent grafting not only in the chronic, but also in the acute situation already in the mid 1990s of the last century.
Over a 6-year period, we were confronted with a total of 17 cases of chronic problems such as abdominal malperfusion due to persisting false lumen collapse, and impending rupture of a false lumen aneurysm, all as a result of ascending aortic replacement for AAAD. This prompted a change in our philosophy, starting a program of the hybrid midline approach in June 2001 for chronic cases, and extension of this approach to AAAD in December 2002 [13].
Since DeBakey type II aortic dissection probably is cured by ascending aortic replacement with or without hemiarch replacement, treatment of the descending aorta is not indicated, thus obviating the use of a descending stent graft. The purpose of this study is to compare the outcome of the classic surgical treatment of acute DeBakey Type I aortic dissection with the more radical approach of combining the conventional surgical approach with hybrid stent grafting of the descending aorta, in regard to development of thrombosis within the thoracic false lumen, and morbidity and mortality at midterm follow-up.
 |
Patients and Methods
|
|---|
From January 2001 to October 2007, a total of 71 consecutive patients underwent surgery for AAAD at the Department of Thoracic and Cardiovascular Surgery, West German Heart Center, University Hospital of Essen. Twenty-six patients with DeBakey type II dissection were excluded from the study, leaving 45 patients with acute DeBakey type I dissection for investigation. The patients were separated into group 1, the historic group (control group, n = 23), undergoing conventional surgery, and group 2 (study group, n = 22) undergoing combined classic surgical repair with antegrade stent grafting of the descending aorta, the so-called hybrid approach. In group 2 patients, the indication for full arch replacement was a primary entry tear in the arch in 5 patients, a reentry site in 8, full circular dissection in 8, and arch aneurysm in 1 patient.
The use of either the Talent (Medtronic Inc, Minneapolis, MN) stent graft prosthesis (n = 4) or the E-vita Open (Jotec GmbH, Hechingen, Germany) hybrid graft (n = 18) for descending aorta endografting was approved by the Ethics Committee of the University Hospital of Essen. According to the specific emergency situation of the involved patients, the chairman of the Ethics Committee waived the need for patient consent for this study.
Baseline patient characteristics are given in Table 1. All patients underwent transesophageal echocardiography (TEE) upon arrival at our hospital, being awaited by a cardiologist, an anesthesiologist, and a cardiac surgeon in the operation theater to save time and to shorten the interval between symptoms and surgical treatment. From April 2004 on, the same team received the patient in our hybrid room (fully equipped catheterization laboratory/operative suite). In case of suspicion of coronary artery disease, a coronary angiography was immediately performed on the catheter laboratory operating table followed by surgery without time loss. Concomitant diseases as well as preoperative complicating factors are listed in Table 1. In case of abdominal malperfusion, angiography and a stenting/fenestration procedure performed by the cardiologist preceded surgery.
Choice of Endograft
A modified version of the Talent abdominal stent graft without distal bare springs was used until December 2004 for antegrade application. Owing to shortcomings like stiffness of the delivery system, high friction forces during positioning and proximal bare springs tending to penetrate the cephalad native aortic arch island, its use was abandoned and replaced by a novel hybrid prosthesis, E-vita Open, a hybrid polyester stent graft prosthesis, combining a 15-cm long nitinol stented portion with an integrated, unstretched 7-cm long nonstented and nonpreclotted portion for aortic arch/ascending aortic replacement [14]. In all but the cardiopulmonary resuscitation cases, a stiff backup Meier wire was placed into the true lumen of the proximal descending aorta through the femoral artery by a cardiologist using online TEE control or fluoroscopy to secure later landing of the stent graft in the true lumen low in the descending thoracic aorta.
Surgery
All patients but the ones under cardiopulmonary resuscitation were operated upon with 3 arterial pressure lines (2 radials, 1 femoral) in place for malperfusion control. Right axillary artery cannulation was the arterial cannulation site of choice in 23 patients. In case of computed tomography–proven brachiocephalic trunc dissection, in cardiopulmonary resuscitation cases or extremely unstable patients, direct ascending aortic cannulation was performed in 22 patients, more recently after primary venous exsanguination and direct sight cannulation of the true lumen, as reported previously [15]. Core cooling to 25°C bladder temperature was begun. Aortic cross-clamping was performed early with the administration of cold, crystalloid cardioplegia (1,500 to 2,000 cc Custodiol solution [Dr Franz Köhler Chemie GmbH, Alsbach-Hähnlein, Germany]) into the coronary ostia after transecting the ascending aorta. Then proximal repair was started while cooling continued. After reaching the target temperature, a brief period of hypothermic circulatory arrest was initiated with removal of the cross-clamp and inspection of the arch. According to the principles of removing all reachable entry—or reentry—sites, the aortic arch was resected, and the proximal descending aorta was transected just beneath the often encountered reentry tear at the origin of the left subclavian artery. Then selective antegrade cerebral perfusion was started through the right axillary artery cannula after cross-clamping the proximal brachiocephalic trunk, and additional cannulation of the left common carotid artery was performed for bilateral perfusion, while the left subclavian artery was blocked proximally using a 6F Fogarty catheter. In case of primary direct aortic cannulation, the brachiocephalic trunk was cannulated separately. Perfusate temperature chosen was 18°C, at a flow rate of approximately 10 cc · kg–1 · min–1 to maintain a line pressure between 40 and 60 mm Hg.
In group 1 patients, a total arch, hemiarch, or isolated ascending aortic replacement was done in 13, 6, and 4 patients, respectively. In group 2 patients, the ascending aorta and aortic arch were resected and the previously placed guidewire was picked up, and sizing of the true lumen of the descending aorta was performed with specially designed highly bendable nitinol obturators (Fehling Instruments GmbH&Co.KG, Karlstein, Germany) to measure the real diameter of the true lumen and to avoid oversizing. Then the endograft was placed over the stiff guidewire and deployed in the descending aorta. After retraction of the introducer the unfolded stent graft ended around 1 cm below the proximal end of the cut off descending aortic stump. The elephant-trunklike inverted nonstented polyester tube was then pulled back approximately 5 mm for proximal suturing of the graft to the proximal descending aortic stump using a 3–0 polypropylene continuous sutureline stabilized externally with a Teflon (Impra Inc, subsidiary of L. R. Bard, Tempe, Arizona) felt before full withdrawal for creation of the cephalad island anastomosis and the ascending aortic anastomosis (Fig 1). Owing to its water permeability, the free polyester prosthesis was preclotted with a thin layer of fibrin glue (1.5 cc to 2 cc) for water tightness.

View larger version (112K):
[in this window]
[in a new window]
|
Fig 1. (A) Insertion of the hybrid prosthesis into the descending aorta. (B) Distal anastomosis between the integrated polyester prosthesis and descending aorta. (C) Unfolding of the attached polyestergraft for arch island anastomosis. (D) Angioscopic view through arch prosthesis demonstrating direct continuity between arch prosthesis and descending stent graft.
|
|
In case of dissection of the head vessels, direct implantation of the vessels after shortening took place. Five minutes before the expected finishing of the anastomosis, the brain perfusate was gradually rewarmed to 25°C. Full body reperfusion was restarted either by using the axillary arterial cannula or direct cannulation of the arch prosthesis, after deairing of the arch vessels with the patient's head down 20 degrees. While rewarming, proximal repair was completed if necessary. Finally, an end-to-end anastomosis between the ascending aortic graft and the arch prosthesis was performed. Additional procedures such as coronary revascularization were done during rewarming (Table 2). After reaching normothermia, extracorporeal circulation was discontinued under TEE control. While protamine was administered, the thrombosing process of the false lumen around the descending aortic stent graft was observed and documented with TEE.
Follow-Up Protocol
All patients underwent intraoperative TEE control after termination of extracorporeal circulation. In case of suspicion of cardiac or abdominal malperfusion, immediate angiographic control on the operating table took place.
Between postoperative days 6 and 10, all surviving patients underwent computed tomography angiography, as well as after 3, 6, and 12 months, and annually thereafter. At these time intervals, patients were also seen in our outpatient clinic.
Data Analysis
Data were collected prospectively supported by our database for thoracic aortic surgery and studied retrospectively by team members. Follow-up was complete in all patients by end of October 2007. The SPSS 14.0 package (SPSS, Chicago, Illinois) was used for analysis. Univariate analysis was performed with Fisher's exact test for categorical and Student's t test for continuous variables.
 |
Results
|
|---|
Preoperative status analysis showed group 2 patients more severely compromised than group 1 patients in regard to hemodynamic depression (12 versus 6, p = 0.05) and cerebral malperfusion (12 versus 2, p < 0.01; Table 1). No difference in intraoperative variables could be found between groups (Table 3). In-hospital mortality was 7 of 45 patients (16%): 5 of 23 patients (22%) in group 1 versus 2 of 22 (9%) in group 2 (p = 0.22). Postoperative complications like stroke (2 versus 2, p > 0.05), pulmonary insufficiency with prolonged assisted ventilation longer than 72 hours (6 versus 10, p > 0.05), and low-output syndrome (3 versus 2, p > 0.05), did not differ between groups (Table 4).
Owing to the sequential development of the hybrid approach out of our classic repair technique, group 1 served as a historic control group. Thus, follow-up time differs significantly between groups (48 versus 23 months, p < 0.01). Intermediate-term mortality, 2 versus 4 (p = 0.38), did not differ significantly between groups and was related to bowel ischemia 7 months after aortic surgery, immediately after redo surgery for severe mitral and tricuspid valve insufficiency in 1 patient, and multiorgan failure after sigmoid resection for sigmoid cancer 4 months after primary surgery, all in group 1. In group 2, 2 patients had severe, persisting neurologic deficit leading to death in a rehabilitation center 3 and 4 months after surgery, 1 patient coming in already with stroke. One 77-year-old patient died of aspiration pneumonia 13 months after surgery. A 42-year-old women with persisting dissection in abdominal aorta succumbed suddenly, and surgery proven, to complete mesenteric ischemia in a rehabilitation center 2 months after surgery while recovering from neurologic deficit. Angio-computed tomography follow-up scans demonstrated patency of the false lumen in 89% of patients in group 1, necessitating stenting of the descending aorta in 4 patients owing to symptomatic true lumen collapse (n = 3) and pseudoaneurysm formation in 1 patient (Table 5). In group 2, all false lumina of the thoracic aorta around the stent graft were occluded but for 1 patient with a Talent stent graft who had minor proximal endoleak not necessitating secondary intervention to date. One patient with an E-vita Open stent graft had a secondary distal endoleak after 16 months, which had to overstented; and another patient required extension of the E-vita stent graft into a previously placed thoracoabdominal uncovered stent for endoleak. According to patient height and length of the descending aorta, no stent graft extended beyond level T9 in any patient. No cases of postoperative paraplegia were encountered. Two patients of group 1 had to be reoperated on for new mitral and tricuspid valve insufficiency, and abdominal malperfusion syndrome with persisting arch and descending aortic dissection, respectively. One patient in group 2 had severe aortic regurgitation 9 months after primary repair due to noncoronary cusp necrosis, necessitating aortic root replacement.
 |
Comment
|
|---|
The overall results of surgical treatment of AAAD are not satisfying. Owing to the dynamic course of the disease with a 1% to 3% death rate per hour mostly in the prehospital phase [16] and the multifactorial nature of its complications before and during surgery, no evidence-based data are available with guidelines of best treatment. Besides the existing observational literature, our own (anecdotal) experience, and the ethical demand of the surgeon, "to produce a life patient" are the keystone guidelines in treating this difficult patient group [17]. In addition, leaving the false lumen in the aortic arch or descending aorta untreated causes secondary problems in the long-term follow-up in a substantial subset of patients.
Being confronted with several patients having complex thoracic aortic disease after previous aortic surgery for AAAD, we as well as others changed our surgical philosophy, integrating the reports of Japanese authors with improvised stent grafting of the descending aorta through a midline approach even in acute type A aortic dissection with our own experience with thoracic stent grafting primarily for complex thoracic aortic disease [18–20]. To get the patient operated on earlier in this "vicious cycle," in 2001, we began to await the patient with acute aortic syndrome with a team consisting of cardiologist, anesthesist, and cardiac surgeon for quick diagnosis and treatment, and from 2004 on, in a hybrid room, to shorten the interval from arrival to surgery. This practice resulted in a consecutive group of emergency patients, who all were operated upon within 24 hours after onset of symptoms, not denying the surgical option even in case of ongoing cardiopulmonary resuscitation. Comparing the severity of the preoperative clinical status and the operative procedures performed with the conventionally treated historic (control) group, there were more patients with severe hemodynamic compromise and cerebral malperfusion in the hybrid group. Taking this into consideration, hospital mortality improved in the hybrid group although not yet reaching statistical significance. This trend toward better outcome reflects both an advanced surgical experience as well as the safety of the procedure itself, although being more complex (Fig 2).

View larger version (42K):
[in this window]
[in a new window]
|
Fig 2. In-hospital mortality of 45 patients with acute DeBakey type I aortic dissection. The white bars represent the number of patients operated upon for acute type I aortic dissection. The solid (black) bars represent the number of patients who died.
|
|
During the follow-up period, the primary goal of the study, to possibly create stable thrombosis of the false lumen in the peristent graft portion of the descending aorta, was achieved in all but 1 patient early in the series with a Talent stent graft. A reduction of the false lumen diameter around the stent grafted descending aorta was observed within all patients of the hybrid group (Fig 3). Late reopening of an intercostal artery at the stent graft end could be managed by transfemoral placement of a secondary stent graft for prolongation of the surgically placed stent graft.

View larger version (149K):
[in this window]
[in a new window]
|
Fig 3. A 52-year-old man after E-vita Open stent grafting for acute type A aortic dissection with durable thrombosis/shrinking of the false lumen in the descending aorta. Computed tomography scans (A) 10 days postoperatively, (B, D) after 6 months, and (C) after 12 months.
|
|
The use of a reversed thoracic stent graft for antegrade thoracic aortic application like the Talent model early in our series demonstrated feasibility but shortcomings, prompting the development of an integrated stent graft-polyester hybrid prosthesis joining both the possibility for classic surgical descending aortic stump anastomosis with seamless transition from stent graft to arch prosthesis, thus most probably avoiding any risk of endoleakage within the arch/descending aortic zone. Technical problems associated with the permeability of the polyester prosthesis are currently managed by the application of a thin layer of fibrin glue (1.5 to 2 cc), but will probably be shortly overcome by a more tightly woven next-generation prosthesis. Our results show that this goal has been achieved for all 18 E-vita Open cases, with no periprosthetic perfusion of the false lumen besides 1 case of late reopening of an intercostal artery at the distal end of the thoracic stent graft necessitating interventional extension with a second stent graft. In contrast, 4 of 17 surviving patients in the conventionally operated on group had to undergo retrograde stent grafting of the thoracic aorta during follow-up. Early mortality remains of concern but is largely related to the primary condition the patient came in with. Two of 3 patients under cardiopulmonary resuscitation finally succumbed (1 survivor in group 1). Stroke is still a major problem associated with this surgery despite improvements in cannulation, selective antegrade cerebral perfusion, and careful deairing.
Limitations of the Study
Owing to its nature as an observational, retrospective study and to the small number of patients, limited conclusions can be drawn. It solely reflects our own efforts to improve on unsatisfactory results with emergency surgery of an unselected, heavily compromised group of type I AAAD patients.
The demonstrated approach tries to address the multifactorial scenario of acute DeBakey type I aortic dissection by focusing on different variables: shortening the interval from time of diagnosis to surgery to operate before the advent of multiorgan failure; integration of diagnostic, interventional, and surgical methods using the hybrid room benefits; removing all reachable entry and reentry sites; and splinting the true lumen in the descending thoracic aorta simultaneously to avoid late problems with false lumen complications. Although this goal has not been fully reached yet, a new way to cope with this complex surgical problem has been demonstrated. Thus, these early results could probably serve as a discussion basis.
In conclusion, we believe that the demonstrated approach could be a step forward to preserve life and to treat the descending aortic false lumen simultaneously without increasing hospital mortality, eventually resulting in a remodeling of the entire thoracic aorta. Future technical improvement with primarily impervious graft material seems to offer a promising way toward risk reduction of massive proximal blood loss and later endoleakage within the grafted aortic area. Longer follow-up will show whether late complications within the thoracoabdominal aorta will be reduced or not, leaving the option for additional interventional treatment if necessary. Lastly, a multi-institutional, prospective, randomized trial is warranted to get a definite answer on this complex procedural challenge.
 |
References
|
|---|
- Fann JI, Smith JA, Miller DC, et al. Surgical management of aortic dissection during a 30-year period Circulation 1995;92:113-121.[Abstract/Free Full Text]
- Ehrlich MP, Ergin MA, McCullough JN, et al. Results of immediate surgical treatment of all acute type A dissections Circulation 2000;102:248-252.
- Tan ME, Morshuis WJ, Dossche KM, Kelder JC, Waanders FG, Schepens MA. Long-term results after 27 years of surgical treatment of acute type a aortic dissection Ann Thorac Surg 2005;80:523-529.[Abstract/Free Full Text]
- Hagan PG, Nienaber CA, Isselbacher EM, et al. The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease JAMA 2000;283:897-903.[Abstract/Free Full Text]
- Rampoldi V, Trimarchi S, Eagle KA, et al. Simple risk models to predict surgical mortality in acute type A aortic dissection: the international registry of acute aortic dissection score Ann Thorac Surg 2007;83:55-61.[Abstract/Free Full Text]
- Kirsch M, Soustelle C, Houel R, Hillion ML, Loisance D. Risk factor analysis for proximal and distal reoperations after surgery for acute type A aortic dissection J Thorac Cardiovasc Surg 2002;123:318-325.[Abstract/Free Full Text]
- Ergin MA, Phillips RA, Galla JD, et al. Significance of distal false lumen after type A dissection repair Ann Thorac Surg 1994;57:820-824.[Abstract]
- Halstead JC, Meier M, Etz C, et al. The fate of the distal aorta after repair of acute type A aortic dissection J Thorac Cardiovasc Surg 2007;133:127-135.[Abstract/Free Full Text]
- Kouchoukos NT, Mauney MC, Masetti P, Castner CF. Single-stage repair of extensive thoracic aortic aneurysms: experience with the arch-first technique and bilateral anterior thoracotomy J Thorac Cardiovasc Surg 2004;128:669-676.[Abstract/Free Full Text]
- Estrera AL, Miller III CC, Villa MA, et al. Proximal reoperations after repaired acute type A aortic dissection Ann Thorac Surg 2007;83:1603-1608.[Abstract/Free Full Text]
- Kato M, Ohnishi K, Kaneko M, et al. New graft-implanting method for thoracic aortic aneurysm or dissection with a stented graft Circulation 1996;94:188-193.
- Uchida N, Ishihara H, Shibamura H, Kyo Y, Ozawa M. Midterm results of extensive primary repair of the thoracic aorta by means of total arch replacement with open stent graft placement for an acute type A aortic dissection J Thorac Cardiovasc Surg 2006;131:862-867.[Abstract/Free Full Text]
- Herold U, Tsagakis K, Kamler M, et al. Change of paradigms in the surgical treatment of complex thoracic aortic disease Herz 2006;31:434-442.[Medline]
- Jakob H, Tsagakis K, Leyh R, Buck T, Herold U. Development of an integrated stent graft-dacron prosthesis for intended one-stage repair in complex thoracic aortic disease Herz 2005;30:766-768.[Medline]
- Jakob H, Tsagakis K, Szabo A, Wiese I, Thielmann M, Herold U. Rapid and safe direct cannulation of the true lumen of the ascending aorta in acute type A aortic dissection J Thorac Cardiovasc Surg 2007;134:244-245.[Free Full Text]
- Khan IA, Nair CK. Clinical, diagnostic, and management perspectives of aortic dissection Chest 2002;122:311-328.[Medline]
- Dobrilovic N, Elefteriades JA. Stenting the descending aorta during repair of type A dissection: technology looking for an application? J Thorac Cardiovasc Surg 2006;131:777-778.[Free Full Text]
- Herold U, Piotrowski J, Baumgart D, Eggebrecht H, Erbel R, Jakob H. Endoluminal stent graft repair for acute and chronic type B aortic dissection and atherosclerotic aneurysm of the thoracic aorta: an interdisciplinary task Eur J Cardiothorac Surg 2002;22:891-897.[Abstract/Free Full Text]
- Eggebrecht H, Baumgart D, Herold U, et al. Interventional management of aortic dissection Herz 2002;27:539-547.[Medline]
- Karck M, Chavan A, Khaladj N, Friedrich H, Hagl C, Haverich A. The frozen elephant trunk technique for the treatment of extensive thoracic aortic aneurysms: operative results and follow-up Eur J Cardiothorac Surg 2005;28:286-290.[Abstract/Free Full Text]
Related Article
-
Invited Commentary
- Wilson Y. Szeto
Ann. Thorac. Surg. 2008 86: 101-102.
[Extract]
[Full Text]
[PDF]
This article has been cited by other articles:

|
 |

|
 |
 
S. Miyamoto
Invited Commentary
Ann. Thorac. Surg.,
January 1, 2009;
87(1):
108 - 108.
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
W. Y. Szeto
Invited Commentary
Ann. Thorac. Surg.,
July 1, 2008;
86(1):
101 - 102.
[Full Text]
[PDF]
|
 |
|