| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;84:759-767
© 2007 The Society of Thoracic Surgeons
a Section of Cardiothoracic Surgery, Yale University School of Medicine, New Haven, Connecticut
b Department of Diagnostic Imaging, Yale University School of Medicine, New Haven, Connecticut
c Department of Preventive Medicine, State University of New York, Stony Brook, Stony Brook, New York
d Department of Economics, State University of New York, Stony Brook, Stony Brook, New York
Accepted for publication April 24, 2007.
* Address correspondence to Dr Elefteriades, Section of Cardiothoracic Surgery, FMB, 333 Cedar St, New Haven, CT 06510 (Email: john.elefteriades{at}yale.edu).
Presented at the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Straight DHCA at 19°C was used in 394 patients (267 males, 127 females) during a 10-year period. Mean age was 61.3 years (range, 15 to 88 years). Eighty-seven cases (22.1%) were urgent or emergencies. Thirty-eight (9.6%) were performed for descending or thoracoabdominal pathology and the rest for ascending/arch (102 hemiarch, 49 total arch). Ninety-one patients (23.1%) had dissections. The head was packed in ice. No barbiturate coma was used.
Results: DHCA lasted a mean of 31.0 minutes (range, 10 to 66 minutes). Reexploration for bleeding was required in 4.5% (18/394). Overall mortality was 6.3% (25/394). Mortality was 3.6% (11/307) for elective cases and 16% (14/87) for emergency cases. The stroke rate was 4.8% (19/394). The seizure rate was 3.1% (12/394). Forty-five patients with high professional cognitive demands (MD, PhD, attorney, etc) performed without detriment postoperatively. Among patients with DHCA exceeding 40 minutes, the stroke rate was 13.1% (8/61); a neuroradiologist’s review of brain computed tomography scans found 62.5% of these strokes (5/8) to be embolic and 37.5% (3/8) hypoperfusion related. By multivariable logistic regression, emergency operation and descending location increased morbidity and mortality.
Conclusions: Straight DHCA without adjunctive perfusion suffices as a sole means of cerebral protection. Stroke and seizure rates are low. Cognitive function, by clinical assessment, is excellent. Especially for straightforward ascending/arch reconstructions, there is little need for the added complexity of brain perfusion strategies.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Concerns about increased mortality and risk of neurologic deficit led to implementation of adjuncts that might enhance the safety of the DHCA technique. The addition of retrograde cerebral perfusion (RCP) did not improve the morbidity or mortality. It was shown that little cerebral delivery of oxygen to the brain was actually accomplished [4]. Selective antegrade cerebral perfusion (ACP) effectively delivers oxygen and protects the brain [5], but this is accomplished at a cost of greater complexity and encumbrance of the operative field.
Despite the recent trend for cardiothoracic surgeons to avoid straight DHCA, we have an extensive experience with this procedure at our institution as a sole means of brain protection, without perfusion adjuncts (Fig 1). This experience provides an ample clinical substrate for evaluation of the efficacy and safety of straight DHCA as a sole means of brain preservation in aortic surgical procedures. Our goal is to determine clinical outcome, including mortality and stroke.
|
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Patient Profile
We retrospectively reviewed 394 cases of aortic surgical procedures performed under DHCA during the last 10 years. The mean age was 61.3 years (range, 15 to 88 years), and 267 were males and 127 were females. Urgent or emergency procedures were done in 87 cases (22.1%), and 307 (77.9%) were elective. Thirty-eight cases (9.6%) were performed for descending or thoracoabdominal pathology, or both, and 356 (90.4%) were for ascending or arch pathology, or both. Ninety-one patients (23.1%) had dissections and 303 (76.9%) had aneurysms. Six patients had Marfan syndrome.
Temperature Control and Deep Hypothermic Circulatory Arrest Management
The mean bladder temperature was 19°C (range, 16° to 20°C). The head was packed in ice. No barbiturate coma was used during the operation. No electroencephalogram (EEG), sensory evoked potential (SEP), or jugular venous bulb oxygen saturation monitoring were used. No special glucose management techniques were applied. The maximum temperature gradient between perfusate and body temperature was kept below 10°C. Rewarming was taken to a temperature of 34° to 36°C.
Neurologic Evaluation
The patients were grossly evaluated neurologically after awakening from anesthesia and every day of their hospital stay. Each patient with a recognized neurologic deficit underwent neurologic consultation, including a cranial computed tomographic scan (CT), to rule out stroke or bleeding. All stroke cases were clinically and radiologically documented, and CT-negative temporary neurologic disturbances were not classified as stroke events.
Mortality Definition
We included in our mortality group all patients that died within 30 days after the operation or before being discharged from the hospital.
Duration of Deep Hypothermic Circulatory Arrest
The DHCA duration time was recorded for each patient. Because previous studies suggest 40 minutes as a safety cutoff point for the technique, we evaluated separately all the cases that required more than 39 minutes of DHCA.
Surgical Techniques
DHCA was used as the sole means of cerebral protection during this time period. Carbon dioxide flooding of the field was used in all cases. Femoral cannulation was used for arterial perfusion unless there was known arteriosclerotic disease of the descending aorta from preoperative studies or by intraoperative transesophageal echocardiography; in which case, axillary cannulation or direct cannulation of the aneurysm or distal aorta were used.
The extent of resection was determined by the extent of disease, with the goal of excising all severely dilated aortic segments. If the dilated segment was confined to the ascending aorta, only this portion was resected, with an open distal anastomosis being performed. If the dilatation tapered off in the region of the aortic arch, a hemiarch resection was performed. If the aortic arch itself was severely dilated, a formal arch replacement was performed. For formal arch replacement, we preferred anastomosis of the head vessels as a pedicle to the main aortic graft. We often chose to anastomose only the innominate and left carotid arteries during DHCA, leaving the anastomosis of the left subclavian artery to be performed with a separate sidearm graft after resuming circulation (during rewarming) or even after terminating cardiopulmonary bypass.
With this technique, the distal (descending aortic) anastomosis could be performed in about 20 minutes and the arch anastomosis in about 20 minutes, leading to a total DHCA time of approximately 40 minutes (Fig 2). Associated procedures included aortic valve replacement in 131 patients, coronary artery bypass in 31, and aortic valve replacement and coronary artery bypass in 10.
|
Computed Tomography Scan Review
The CT scans of all patients who sustained a stroke were reviewed by an expert neuroradiologist. CT scans or magnetic resonance imaging (MRI) of the brain were interpreted to determine whether the cerebral event was due to embolism or to nonembolic hypoperfusion/anoxia. Sharply demarcated, multifocal hypodensities, both infratentorial and supratentorial, were considered characteristic of embolic infarctions (Fig 3). Less demarcated hypodensities in the watershed zones between major vascular territories were considered characteristic of infarctions secondary to nonembolic anoxia (Fig 4).
|
|
2 tests. The potential risk factors that were included in the statistical evaluation were age, DHCA time exceeding 39 minutes, urgency of operation, male gender, location of the pathology (ascending versus descending), and type of pathology (aneurysm versus dissection).
Follow-Up
All discharged patients were reevaluated in the office within 8 weeks postoperatively and screened by our staff for any neurologic sequelae or impaired mental or physical functioning.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Reexploration for bleeding was required in 18 (4.5%) of the 394 patients. Most had other concomitant medical conditions that may affect bleeding diathesis, such as chronic renal failure in 4 patients, hepatitis B or C in 4, or use of anticoagulant or antiplatelet medications in 5. The mean cardiothoracic intensive care unit (ICU) stay (excluding outliers) was 3.0 days (median, 2 days), and the mean postoperative hospitalization stay was 7.9 days (median, 7 days).
Among the ascending and arch group, 199 patients had ascending aortic replacement alone, 102 had ascending aortic replacement and a hemiarch procedure, and 49 had a total arch replacement.
Overall and subgroup mortality rates are summarized in Table 1. The overall mortality rate was 6.3% (25/394): 3.6% (11/307) for elective cases (6 ascending and arch, 5 descending and thoracoabdominal) and 16.1% (14/87) for emergency operations (2 ascending and arch, 12 descending and thoracoabdominal). Mortality was high for descending and thoracoabdominal aneurysms operated on under DHCA because these generally were emergency cases and represented patients with ruptured descending and thoracoabdominal aneurysms extending too proximally into the aortic arch to be clamped directly, thus requiring DHCA. Among the patients with ascending and arch procedures, elective mortality was 2.0%, urgent/emergency mortality was 3.2%, and total mortality was 2.2%.
|
Multivariable logistic regression was used to analyze the impact of risk factors for complications and death. This revealed that emergency operation and descending location of the pathology were associated with significantly higher odds of complications (death, stroke, seizures, dialysis) and mortality (Table 2ab).
Multivariable logistic regression analysis for risk factors for stroke found that elective surgery and surgery on the ascending/arch portions of the aorta were protective against stroke, with an odds ratio (OR) of 0.32 (p = 0.05) for elective operation, and an OR of 0.23 (p = 0.02) for ascending/arch location. Although there was a trend for increased OR of stroke in patients with a DHCA time exceeding 40 minutes, this was not statistically significant.
|
|
After a standard postoperative recovery period, 44 patients with high professional cognitive demands (MD, PhD, attorney, etc) returned to work and performed their professional activities without any impairment perceived by themselves or others (Table 3).
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
We believe the data in this report argue strongly that straight DHCA suffices in almost all cases. We found low hospital mortality in ascending and arch operations (2.0% in elective cases), a low stroke rate (1.9% in elective cases), and a low rate of seizures (3.1%) with straight DHCA applied in a relatively large number of 394 cases. We are not aware of consistently better results reported with antegrade cerebral perfusion or retrograde perfusion [9–28]. Our data demonstrate strongly the safety of straight DHCA, especially in cases where the duration of DHCA is less than 40 minutes. Many excellent series of cases using retrograde and antegrade cerebral perfusion with good results have been reported (Table 4) [9–29]. Although it is difficult to compare series directly because of different patient profiles, aneurysm extent, acuities, and complexities, it is important to note that mortality and stroke rates in those series are not generally superior to those realized in the present series of patients operated with straight DHCA. We believe this further demonstrates that adjunctive perfusion techniques are largely unnecessary.
|
We believe that most aortic cases in which DHCA is used can safely be accomplished within 40 minutes. A hemiarch replacement suffices for almost all of the combined ascending and arch procedures because the pathology usually involves an ascending aneurysm extending into the aortic arch, where the aortic diameter usually tapers toward normal. Also, for emergency operations on acute type A aortic dissections, we believe there is little need for direct cerebral perfusion because speed and simplicity are of the essence in that setting. Ascending and hemiarch replacement suffices for most acute type A dissections [30].
Even for total arch replacement, we favor straight DHCA at our institution. We find that the two anastomoses—the elephant trunk circumferential suture line at the descending aorta and the Carrel patch of the head vessels—can safely be accomplished during this time. We facilitate this process by anastomosing only the innominate and left carotid arteries on a single patch, as described in the "Methods" section (see Fig 5). This makes for a smaller, easier, more accessible anastomosis. We reimplant the left subclavian artery after completion of the operation and termination of cardiopulmonary bypass, usually with a separate, small-caliber Dacron (DuPont, Wilmington, DE) graft.
|
We especially wish to emphasize that straight DHCA should suffice handily for operations on acute ascending aortic dissections in which one single distal anastomosis can be readily accomplished well within the safe time range. We believe there is no need in this urgent setting, often in the middle of the night, for the added complexity of the retrograde or selective cerebral perfusion methods.
We were especially reassured by the findings in four dozen patients with high cognitive and dexterity demands undergoing DHCA, including doctors, lawyers, and practitioners of the fine arts, among others. In all cases, these individuals returned to their professions without impediment noted by themselves or reported by others. We find the excellent performance level of these individuals especially reassuring in terms of the ability of DHCA to preserve fine cognitive and motor functions of the highest levels.
Although we believe that antegrade cerebral perfusion is a proven technique that does protect the brain very well, it has its own inherent problems. These include the added complexity and cluttering of the operative field, potential vessel injury, potential embolization of atherosclerotic material, potential exacerbation of great vessel dissection, the issue of which vessels to perfuse (innominate only versus innominate and left carotid), the issue of the correct rate of flow, the issue of potential inhomogeneous perfusion of various zones of the brain, the potential for induction of cerebral edema, and the issue of clamping or not clamping the left subclavian artery during perfusion. All of these issues are obviated by straight DHCA. Ours is not a comparative study, however, and the favorable clinical results achieved in this series might have been possible with any of the techniques of circulation management.
Issues of brain preservation in DHCA are complex and difficult to study. Accordingly, we wish to recognize certain weaknesses of this clinical study. First, we have focused on fixed stroke. Ergin and colleagues [31] have emphasized the importance of transient neurologic deficits after operations performed with DHCA. We note disturbances of this sort even after conventional cardiac surgical procedures, without DHCA of any type. We did not evaluate these temporary deficits in our study, focusing on permanent strokes. Neurologic results were determined by gross neurologic assessment by the clinical staff performed postoperatively; a neurologist specialist and routine preoperative and postoperative specialized assessment would have provided more supportive information.
Second, there is no proven method to identify inadequately preserved tissue on head CT scans after DHCA. We do believe that we can identify obvious emboli and exonerate preservation technique in those cases. For lack of specific detection methods, we presume that inadequately protected brain tissue will have a radiographic appearance like that of ischemic strokes.
Third, our observations of the individuals with high cognitive demand are based solely on clinical level of function. Although the importance of brain protection applies to all patients, subtle neurocognitive defects may become more apparent in such individuals compared with those who perform more repetitive tasks in their daily life. Detailed neuropsychometric studies of this interesting subgroup of our patient population are underway.
Fourth, this series cannot prove that no patient needs perfusion adjuncts. Especially with very long duration procedures, the safe limit of DHCA may be exceeded. In a very small fraction of cases, we prepare for antegrade cerebral perfusion in case the duration of DHCA should run too long; however, we have not needed to implement this technique. We believe that the patients who need cerebral perfusion adjuncts are few and far between.
In conclusion, we wish to argue not against antegrade cerebral perfusion but, rather, in favor of straight DHCA. We believe that the clinical experience presented in this investigation adds to the evidence that DHCA alone, without perfusion adjuncts, is a sufficient and efficient method for cerebral protection during aortic arch procedures. We suggest that surgeons not feel compelled to use antegrade cerebral perfusion, especially in cases of acute type A aortic dissection, and that those cases can be accommodated by rapid hemiarch anastomosis at the distal end of the resection.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
I have two questions. Are these selected patients, and if so, what is the basis of the selection? The second question is, what do you usually do if the procedure requires longer protection time than initially sought? The incidence of stroke is about three times higher in those requiring longer than 40 minutes of circulatory protection time.
DR ELEFTERIADES: We are very much aware of your important work, Dr Kazui, and thank you for your questions. These are unselected patients. This is everyone who had those aortic operations. I wish to say a word about how this series originated. Straight deep hypothermic arrest has been our policy for a long time at Yale; it yielded satisfactory results and we continued that way. The topic of optimal brain protection is so controversial that we thought we should examine in detail our sizable experience with straight deep hypothermic arrest. We looked back with an open mind, and we found that the experience was good, even on detailed retrospective inspection.
I do think that there are patients who need some type of cerebral perfusion, and we often prepare for that in the operating room if we think that the procedure may exceed the safe time limit. We will have our cannulas ready and we will be prepared to institute antegrade cerebral perfusion. We, frankly, haven’t needed to do that. Most of these operations come to completion at about 40 to 45 minutes, and we feel comfortable in that zone. When we start to approach more than 45 minutes or so, certainly up to 60, I think that there may be an important role for some type of perfusion adjunct. I think that those cases are unusual. Those cases that require such a long hypothermic arrest interval are in and of themselves complex and may have a higher rate of cerebral events regardless of perfusion technique.
DR JOHN E. CONNOLLY (Irvine, CA): I enjoyed this paper very much and I would like to congratulate the authors on their excellent results, and I support their recommendation that brain perfusion tends to unnecessarily complicate the procedure.
I would like to ask how long they cooled before circulatory arrest? I suspect their excellent results indicate that they have cooled longer than other investigators who have not had such good results. Also, where did they measure the temperature of their patients at the time of arrest? We have found in dogs that the brain temperature lags well behind all of the other methods that one can measure temperature in the body. Neurosurgeons used to use venous jugular monitoring, but they have switched to direct temperature monitoring through a very small hole in the skull, and they leave these monitoring needles in, often for days, to monitor head trauma and for other indications in patients on anticoagulation. Even in infants and children, they have found no bleeding or clot complications from direct brain cerebral oxygen pressure and temperature monitoring (LICOXimc). And we found that by monitoring the brain temperature in dogs undergoing deep hypothermia arrest, that it takes 35 minutes of cooling to reach a brain temperature of 24°C, which allowed us 30 minutes of safe circulatory arrest. If we cooled to a brain temperature of 20°C, which took us an hour to reach a brain temperature of 20, we had an hour of safety. The dogs employed in these studies had been preconditioned by our psychologists for retainment of memory to show after recovery from deep hypothermia that they had suffered no supratentorial neurological changes.
In conclusion, I would like to ask the authors, do they think that direct brain temperature monitoring might be the way to go now that it is safe? I think the technique is excellent, and the whole problem is to be sure that the brain is at a safe temperature at time of circulatory arrest, and in dogs, we have been able to determine what those brain temperatures are for certain periods of time. Thank you again for a very interesting presentation.
DR ELEFTERIADES: Thank you, Dr Connolly, for sharing that important experience with us. We cool for about a half an hour to 35 minutes. We wait for the Foley temperature to be 18°. In terms of where we monitor, we do monitor the Foley temperature. We are just now assessing a technology that measures the brain temperature through the eye, but we have no experience with that yet. We are interested also, as you are, in measuring brain temperature directly. We do apply ice packs to the head at the inception of the operation. A lot of times the anesthesiologist thinks he can wait to apply the ice until deep hypothermic arrest has started. Really, you lose valuable cooling time by doing that.
Thank you, sir, for your comments.
DR CONNOLLY: I suspect those few neurological complications of yours may have been secondary to too short a time cooling.
DR RANDALL B. GRIEPP (New York, NY): I have just two areas of disagreement. The first, of course, is including attorneys among those that require high cortical functioning for their professional activities (laughter).
The second, though, is the message that hypothermic circulatory arrest is safe for periods of greater than 30 minutes even at very low temperatures. For many years we thought—based on clinical impression—that circulatory arrest was a great method for brain protection for up to an hour, but when we looked at a contemporary series of our own patients who had between 40 and 80 minutes of absence of normal brain perfusion, we could see a distinct difference between antegrade cerebral perfusion and simple DHCA, as assessed by the incidence of postoperative temporary neurological dysfunction. We subsequently confirmed that later on using cognitive testing. Consequently, I worry when an authority like you, John, suggests that other surgeons, perhaps less meticulous in technique, and possibly without the facility to move easily to cerebral perfusion if the HCA interval becomes worrisome, will push the envelope at the expense of their patients’ brains. If you have quantitative cognitive testing which suggests that periods of HCA exceeding 30 minutes are safe, I would be happy to hear it, but at present I am more than a little worried about recommending longer periods of circulatory arrest.
DR ELEFTERIADES: Thank you, Dr Griepp. Your contributions in this area are simply extraordinary, and every subtopic that I investigated in preparation for today was dominated by your research and your papers and chapters.
In terms of temporary neurologic dysfunction, we didn’t look at that in this study. We have a hard time sometimes separating that from the confusion that every patient can get after conventional cardiac surgery for coronary artery bypass or valve replacement. We have a detailed neuropsychological profile being constructed on all of our deep hypothermic arrest patients, including those with a high level of technical function, and we want to look for any subclinical defects, but certainly in terms of their function in their professions, they report no difficulties at all.
DR NICHOLAS T. KOUCHOUKOS (St. Louis, MO): I want to amplify on what Dr Griepp said, and I will be brief. There is clear evidence in studies from his institution, and from others, that the prevalence of temporary neurologic dysfunction increases with the duration of circulatory arrest, and that fine motor and memory deficits are also more prevalent with longer periods of circulatory arrest. You showed quite clearly that after 40 minutes of arrest, the stroke rate was substantially higher than that for an arrest interval less than that, and this is not a new finding. While some of the strokes in your series were of embolic origin, some were likely related to hypoperfusion. With the current availability of other well-tested methods for cerebral protection, namely direct antegrade cerebral perfusion and antegrade/retrograde perfusion through the right axillary artery, there is no longer any justification for extending the interval of hypothermic circulatory arrest beyond 30 to 40 minutes, with few exceptions, and I strongly discourage this practice. Thank you.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. W. Fischer, P. B. Benni, H.-M. Lin, A. Satyapriya, A. Afonso, G. Di Luozzo, R. B. Griepp, and D. L. Reich Mathematical model for describing cerebral oxygen desaturation in patients undergoing deep hypothermic circulatory arrest Br. J. Anaesth., November 20, 2009; (2009) aep335v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Shimamura, T. Kuratani, G. Matsumiya, Y. Shirakawa, M. Takeuchi, H. Takano, and Y. Sawa Hybrid endovascular aortic arch repair using branched endoprosthesis: The second-generation "branched" open stent-grafting technique J. Thorac. Cardiovasc. Surg., July 1, 2009; 138(1): 46 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Di Bartolomeo, L. Di Marco, A. Armaro, D. Marsilli, A. Leone, E. Pilato, and D. Pacini Treatment of complex disease of the thoracic aorta: the frozen elephant trunk technique with the E-vita open prosthesis Eur. J. Cardiothorac. Surg., April 1, 2009; 35(4): 671 - 676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Harrison, M. F. Swartz, and G. W. Fink Novel Treatment of an 11-cm Saphenous Vein Graft Aneurysm Ann. Thorac. Surg., April 1, 2009; 87(4): 1291 - 1292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Percy, S. Widman, J. A. Rizzo, M. Tranquilli, and J. A. Elefteriades Deep Hypothermic Circulatory Arrest in Patients With High Cognitive Needs: Full Preservation of Cognitive Abilities Ann. Thorac. Surg., January 1, 2009; 87(1): 117 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Estrera, C. C. Miller III, T.-Y. Lee, P. Shah, and H. J. Safi Ascending and Transverse Aortic Arch Repair: The Impact of Retrograde Cerebral Perfusion Circulation, September 30, 2008; 118(14_suppl_1): S160 - S166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Sundt III, T. A. Orszulak, D. J. Cook, and H. V. Schaff Improving Results of Open Arch Replacement Ann. Thorac. Surg., September 1, 2008; 86(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Morita, M. Yasaka, K. Yasumori, Y. Oishi, T. Takaseya, H. Sonoda, and T. Kawara Transcranial Doppler Study to Assess Intracranial Arterial Communication Before Aortic Arch Operation Ann. Thorac. Surg., August 1, 2008; 86(2): 448 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G.T. Augoustides Contemporary Conduct of Adult Deep Hypothermic Circulatory Arrest: Possible Roles of Retrograde Cerebral Perfusion, Anesthetic Preconditioning, and Aprotinin Ann. Thorac. Surg., August 1, 2008; 86(2): 689 - 689. [Full Text] [PDF]
|
|
|
|
|
|
J. A. Elefteriades, A. Gega, and J. A. Rizzo Reply Ann. Thorac. Surg., August 1, 2008; 86(2): 689 - 690. [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 |
