| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;107:860-867
© 1994 Mosby, Inc.
CARDIOPULMONARY BYPASS, |
Los Angeles, Calif.
From the Division of Cardiothoracic Surgery, UCLA School of Medicine, Los Angeles, Calif.
Received for publication July 26, 1993. Accepted for publication Sept. 16, 1993. Address for reprints: Gerald D. Buckberg, MD, UCLA Medical Center, Department of Surgery, B2-375, CHS, 10833 LeConte Ave., Los Angeles, CA 90024-1741.
Abstract
Normothermic blood cardioplegia was developed originally to be used during cardioplegic induction and reperfusion as an adjunct for enhancing metabolic reversal of biochemical alterations occurring before, during, and after total myocardial ischemia. This adjunct was introduced clinically after extensive experimental testing. By contrast, continuous normothermic blood cardioplegia without hypothermia was introduced clinically without a scientific infrastructure and has generated great interest because of its simplicity and encouraging early results, but has caused substantial confusion. This report is written to (1) clarify the role of normothermic blood cardioplegia as an adjunct to available hypothermic and antegrade and retrograde methods of myocardial protection, rather than as an alternative to them, (2) call attention to the misconception that continuous coronary perfusion avoids ischemia inasmuch as "unintentional ischemia" may occur despite continuous coronary perfusion, (3) identify theoretic and practical limitations of warm continuous retrograde blood cardioplegia exposed by testing after its clinical introduction, (4) enumerate the unanswered questions posed after clinical use of this method, and (5) focus on the self-imposed inflexibility created by adoption of adversarial positions in regard to cardioprotective strategies that impedes our ability to combine, rationally, the spectrum of approaches to myocardial protection that have evolved from the recognized limitations of individual methods. (J THORACCARDIOVASCSURG1994;107:860-7)
The objective of every cardiac operation is a technically perfect anatomic correction resulting in postoperative cardiac function that comes as close to normal as possible. Two prerequisites to accomplish this goal are adequate visualization of the operative field for surgical precision and cardioprotective strategies that (1) reverse preexistent metabolic abnormalities and (2) exclude intraoperative damage that can nullify immediate and long-term benefits of repair. Deep cardiac hypothermia was the principal method of limiting ischemic damage before introduction of cardioplegic techniques. This led to the axioms that "all is well if the heart is kept as cold as possible" and that "there is a constant battle against the clock during aortic clamping." Current data suggest that these concepts are no longer applicable.
Blood cardioplegia is responsible for this conceptual change because the heart is oxygenated when blood cardioplegia is given and aortic clamping time is no longer synonymous with ischemic time, especially when warm and cold blood cardioplegia techniques are combined.
1 An added misconception is that ischemic damage is avoided if the heart is perfused continually during extracorporeal circulation. This understandable belief has led to a false sense of security that results in "unintentional ischemia" despite continuous coronary perfusion, as shown by extensive experimental studies. These studies form the scientific infrastructure for development of myocardial management techniques used currently. The clinical introduction of continuous warm blood cardioplegia without any experimental testing has generated great interest and caused substantial confusion. This analysis tries to clarify the role of normothermic blood cardioplegia in the spectrum of available methods of myocardial protection.
NORMOTHERMIC AND HYPOTHERMIC BLOOD CARDIOPLEGIA
The emergence of warm blood cardioplegia as a cardioprotective strategy stems from the recognition that (1) normothermic cardioplegic arrest results in myocardial oxygen consumption that is about 10% of that of the beating working heart, or about 1 ml/100 gm/min, 2,3 (2) hypothermia adds only a small additional decrement, 3,4 and (3) specific warm blood cardioplegic formulations during induction and reperfusion enhance metabolic reversal of biochemical alterations that occur before, during, and after aortic clamping, when applied with intermittent cold blood cardioplegia. 5 The demonstrable advantages of normothermic blood cardioplegic modifications in routine elective operations are unestablished, 6 whereas the salutary effects in high-risk patients and after prolonged aortic clamping are clear. 7,8 The benefits of hypothermia include, in addition to a small decrement in myocardial oxygen consumption, retarding recurrence of electromechanical activity when the cardioplegic solution is washed away by noncoronary collateral blood flow and slowing the rate and force of the processes leading to myocardial damage 4 when coronary flow must be interrupted to produce "intentional ischemia" to create the bloodless field needed for technical precision. 9 Deep hypothermia may, however, cause adverse effects, including edema caused by inactivation of the sodium/potassium adenosine triphosphate exchange that regulates cell volume, 10 alteration of platelets and leukocytes, 11 reduction of membrane stability, 12 and impairment of calcium flux. 13 Preoccupation with reducing myocardial temperature below 15° C makes little physiologic sense because oxygen uptake in the arrested heart is negligible at less than 20° C (less than 0.3 ml/100 gm/min). 4 These limitations of hypothermia notwithstanding, experimental study of 4° C blood cardioplegia shows that up to 4 hours of safe aortic clamping can be achieved in normal hearts with unobstructed arteries, provided sufficient doses are administered intermittently to counteract noncoronary collateral washout. 14
Concern over the adverse consequences of systemic and cardiac hypothermia and intraoperative ischemia, coupled with experimental and clinically proven benefits of warm blood cardioplegic induction and reperfusion, led the surgical team at the University of Toronto (where hypothermia was introduced by Bigelow, Lindsay, and Greenwood 15 in 1950) to suggest warm heart surgery without hypothermia but with cardioplegia: the patient and heart are kept at 37° C, and cardioplegic flow is delivered continuously when feasible. 16 They reasoned that (1) electromechanical arrest lowered oxygen demands to sufficiently low levels that hypothermia was superfluous and (2) use of continuous warm blood cardioplegia would "close the loop" and allow "perfect protection" by avoiding ischemia completely. This technique is attractive because of its simplicity and the implied suggestion that continuous perfusion avoids ischemia. Early clinical results with either antegrade 16 or retrograde 17 cardioplegia were encouraging, but were not superior to results using techniques with an extensive experimental basis, and subsequent data indicated unforeseen problems (for example, increased cerebral complications, 18 right ventricular dysfunction, and ischemia caused by inadequate flow 19).
Lichtenstein, Abel, and Salerno 20 suggested that warm cardioplegic flow could be interrupted intermittently, and precondition the heart, on the basis of studies of regional ischemia that showed brief intermittent normothermic ischemia reduces myocardial stunning and limits necrosis after subsequent prolonged regional ischemia. 21 Conversely, brief prebypass ischemia before prolonged aortic clamping with cold cardioplegia worsens rather than improves recovery. 22 The objective of cardioprotective approaches during cardiac operations is complete avoidance of myocardial stunning and necrosis, and current data 23 support the use of hypothermia to lower metabolism during intervals of aortic clamping when normothermic perfusion is stopped.
Myocardial stunning was described in 1982 as temporary functional depression without necrosis after brief (15 minute) normothermic regional ischemia. 24 The reversible low-output syndrome after inadequate myocardial protection is the cardiac surgical counterpart of this phenomenon and has been responsible for the extensive studies of methods of myocardial protection over the past 20 years. It seems ironic that the same normothermic ischemia that gave rise to the term myocardial stunning is used to justify stopping coronary flow because blood obscures the operative field and is responsible also for the development of cardiac hypothermia to limit ischemic damage.
WARM BLOOD CARDIOPLEGIA: STARTING POINTS, END POINTS, AND MEDIAN LETHAL DOSE (LD50)
Cardioplegic research has not, in general, followed established procedures for drug testing. The LD50 concept is used routinely in pharmacologic studies whereby the starting point is an intervention in a model that kills 50% of live organisms; its effectiveness or end point is compared with this starting point. Consequently, an intervention is (1) ineffective if it does not change the starting point, (2) toxic if less than 50% viability results, and (3) defined as effective by how much more than 50% viability it produces. The starting point of most studies of myocardial protection is the normal heart so that the LD50 approach has no relevance to them, inasmuch as any intervention that fails to maintain biochemical and mechanical integrity must be considered ineffective. Consequently, the normal heart model is useful only to test the safety of interventions such as multidose cold blood cardioplegia that allows normal biochemical and mechanical function to recover completely after 4 hours of aortic clamping 14; the starting point is the same as the end point.
We cardiac surgeons rarely get the chance to operate on normal hearts, so our clinical starting point conforms more closely to the LD50 model in pharmacologic studies. Experimental study of the energy- and substrate-depleted heart model has been useful to develop strategies intended to metabolically resuscitate the heart, because cold cardioplegia confers no metabolic benefit other than offsetting further damage. 5 The concepts of warm induction and reperfusion of blood cardioplegia developed from such models, whereby extensive testing of various cardioplegic modifications resulted in a regimen that restored function to ischemically damaged hearts, 5,25-29 with the rationale for individual factors described previously. 1,30 Normothermia is only one element in this regimen and is included to optimize the rate of metabolic recovery, which is retarded by hypothermia. 31
The objective of adding normothermic blood cardioplegia to cold intermittent blood cardioplegia is to use a cardioprotective strategy in the impaired myocardium that acts in concert with mechanical repair to restore near normal biochemical and mechanical function (that is, a normal starting point). Conversely, cold cardioplegic techniques alone can only prevent further damage so that total reliance is placed on the mechanical benefits of operative correction to improve cardiac performance. For example, the operative mortality rate for the surgical treatment of cardiogenic shock approximates 50% with conventional hypothermic cardioprotective techniques, because left ventricular power failure progresses unabatedly despite revascularization. 32,33 In contrast, use of warm blood cardioplegic induction and reperfusion with intermittent cold blood cardioplegia lessens the duration of postoperative circulatory support and improves mortality. 34
CONTINUOUS CORONARY PERFUSION AND "UNINTENTIONAL ISCHEMIA"
Efforts to avoid intentional ischemia by continuous coronary perfusion are not new, but it has been impossible to provide homogeneous flow distribution by continuous perfusion techniques. Every prior attempt has been accompanied by the creation of "unintentional ischemia" because of the artificial conditions imposed to facilitate technical repair (for example, beating empty heart, ventricular fibrillation, arrest).
"Unintentional ischemia" occurs when there is subendocardial underperfusion in the beating heart that is either hypertrophied or contains obstructed coronary arteries. 35,36 Sapsford, Blackstone, and Kirklin 37 reported that continuous perfusion of the hypertrophied beating empty heart during aortic valve replacement produced the same damage as hypothermic ischemia without cardioplegia. The "unintentional ischemia" produced by continuous coronary perfusion was most evident with ventricular fibrillation during aortic valve replacement. The classic observations of Taber, Norales, and Fine 38 and Najafi, Henson, and Dye 39 document subendocardial hemorrhagic necrosis, which was shown subsequently to be caused by an imbalance between energy supply and demand owing to redistribution of flow away from fibrillating subendocardial muscle where energy demands are high. 2,35 Myocardial contracture occurred sometimes after defibrillation, producing the same end point of "stone heart" that may follow normothermic ischemia, 40 in which the mechanism of injury is an imbalance between the absence of supply caused by "intentional ischemia" in an arrested heart with low oxygen demands. Prolonged normothermic ischemia 40 is no longer used because the advantages of ideal operating conditions are nullified by severe ischemic injury. Established methods of warm and cold cardioplegic myocardial management have essentially avoided the iatrogenic complication of massive hemorrhagic subendocardial necrosis, so that current generations of cardiac surgeons are spared the discouraging experience of this extreme example of inadequate myocardial protection.
These lessons of the past concerning "unintentional ischemia" have relevance today with the use of warm continuous retrograde blood cardioplegia. Coronary sinus retroperfusion results in systematic underperfusion of the right ventricle, which receives less than 20% of flow delivered to the left ventricle. 41 Furthermore, inhomogeneous left ventricular perfusion is more pronounced with retrograde versus antegrade cardioplegia. 42 The combined antegrade/retrograde methods are complementary in ensuring more homogeneous perfusion of both ventricles. 41
Retrograde cardioplegic perfusion drains preferentially via thebesian veins so that it is readily undestandable why (1) there is scant drainage from the coronary ostia during aortic valve replacement and (2) the right ventricle becomes cold despite limited nutritive flow, 41 as venovenous connections in the right ventricular free wall and septum allow cooling by conductance. Cold intermittent retrograde cardioplegia reduces the supply/demand imbalance during isolated coronary sinus retroperfusion and provides good, but inconsistent right ventricular recovery; this limitation is overcome by adding intermittent antegrade perfusion. 19 The potential for impaired right ventricular performance may be accentuated with right ventricular hypertrophy or if coronary artery obstruction limits antegrade perfusion; right ventricular hypothermia is a valuable adjunct under these conditions. The nonhomogeneous distribution of nutritive (that is, capillary) left ventricular retrograde flow 42 makes inhomogeneous perfusion more likely in hypertrophied hearts, in which optimum flow rates are not yet established. Consequently, inadequate continuous warm blood cardioplegic perfusion might cause the same "unintentional ischemia" or "stone heart" as occurred with ventricular fibrillation or ischemia without cardioplegia. Postoperative univentricular or biventricular failure or death after continuous warm retrograde blood cardioplegia reflects a problem that hypothermia and antegrade cardioplegia might avoid, and which is caused directly by an inflexible approach based on the misconception that "all is well if the heart is perfused continually."
NORMOTHERMIC CONTINUOUS BLOOD CARDIOPLEGIA: THEORY, FACTS, AND QUESTIONS
The inhomogeneity of warm retrograde cardioplegic distribution to the left and right ventricles contradicts the theory that warm cardioplegia resuscitates the heart continuously. 43 Furthermore, the process of resuscitation requires more than normothermia to optimize metabolic rate, and prolonged perfusion of specialized solutions intended for limited application during induction and reperfusion may have adverse systemic effects (for example, hypocalcemia, alkalosis, hypotension caused by vasodilation) that must be defined and counteracted.
Despite the satisfying early clinical findings, 16,44 subsequent experimental data have exposed some other limitations of warm continuous cardioplegia as compared with cold continuous or intermittent cold antegrade/retrograde blood cardioplegic delivery in jeopardized myocardium. 45,46 Ischemic damage in jeopardized muscle is minimized by continuous cold versus warm retrograde blood cardioplegic perfusion, probably because insufficient retrograde capillary perfusion is compensated by hypothermic lowering of oxygen demands. 46 Both clinical and experimental studies document that arterial and coronary sinus perfusion are delivered to different vascular beds. 42,47 Furthermore, intermittent cold antegrade/retrograde blood cardioplegia provides superior results to those of continuous antegrade or retrograde warm cardioplegia under experimental conditions when warm cardioplegic flow is interrupted intermittently, as must be done clinically to optimize visualization during construction of distal anastomoses. 23 The cumbersome operating conditions caused by blood obscuring the operative field can be solved partially by techniques that improve visualization, but it is not established whether continuous perfusion is necessary. These observations emphasize that both warm and cold cardioplegic techniques, as well as antegrade and retrograde methods of delivering cardioplegia, may be useful and complementary adjuncts that should be integrated into the cardiac surgeon's armamentarium.
Additional questions raised by missing data concerning warm heart surgical techniques include (1) What cardioplegic flow rates are needed to ensure all areas receive sufficient flow to meet metabolic needs, especially in patients with right and/or left ventricular hypertrophy? (2) How long can flow be interrupted before ischemic damage occurs in normal or jeopardized muscle, and how can these changes be overcome with restoration of flow? (3) Is the ideal composition of the cardioplegic solution different from that used for cold cardioplegia? (4) What is the optimal blood/cardioplegic ratio for minimizing hemodilution, yet maintaining the benefits of warm reperfusion when flow must be interrupted? (5) Will production of hyperkalemia and hyperglycemia with continuous cardioplegia require specialized techniques and expensive devices such as hemoconcentrators or dialysis machines to restore a normal electrolyte and metabolic profile? (6) Does normothermic whole body perfusion lead to increased bleeding as a result of higher systemic flow rates? 48 (7) Is the recently reported increased prevalence of cerebral complications 18 an acceptable occurrence, in view of the unproven benefits of warm techniques over those used when mild to moderate systemic hypothermia is used? (8) Are these cerebral complications related to decreased perfusion pressure that can be overcome by pulsatile perfusion or vasoconstrictor administration, or are they caused by early and late hypercoagulability? (9) Will more fatal perfusion accidents occur because of the limited time available for the perfusionist to stop cardiopulmonary bypass and correct problems before irreversible ischemic cerebral damage occurs?
CONTINUITY, COMPARISONS, AND CONFUSION
The "ideal" cardioplegic method is one that can be integrated into the operation so that (1) work can begin as soon as the aorta is clamped and the heart quieted, (2) the work can proceed without interruption, and (3) cardiopulmonary bypass can be discontinued promptly after unclamping with near normal myocardial function. Disruption of the continuity of the operation to give cardioplegia or use field-clearing devices (suckers, blowers, vessel occluders, traction tapes, irrigation, and so on) and to add extra personnel has been, until now, essential to limit intraoperative damage. Continuous coronary perfusion requires such field-clearing maneuvers, and they are "needed" (1) under the misconception that continuous coronary perfusion ensures adequate blood cardioplegic delivery and (2) because comparisons in experimental studies with inherent design limitations suggest some superiority of continuous warm blood cardioplegia over intermittent blood cardioplegic methods. 49 The outcome of these studies is determined, of course, by their experimental design and the confusion created by different outcomes can be clarified only by analysis of the responsible pathophysiologic principles.
The basic objectives of warm blood cardioplegia are that (1) normothermia optimizes aerobic metabolism and (2) resuscitation occurs over time, so that duration of infusion is more important than dose. Understanding the concept of delivery over time versus dose will avoid confusion over interpreting results of studies that compare warm and cold cardioplegic methods. For example, delivery of initial and/or terminal warm blood cardioplegia by dose (for example, 10 ml/kg) at a fixed pressure may provide only a 1- to 2-minute infusion to an ischemic heart with maximally dilated coronary arteries and deprive the myocardium of its full benefits. Consequently, continuous warm blood cardioplegia may prove more beneficial than subsequent interspersed cold infusions, despite the disadvantage of a bloody operative field requiring field-clearing devices. Similarly, restricting delivery of a specific formulation of a normothermic cardioplegic reperfusate to 5 minutes after regional ischemia (for example, percutaneous transluminal coronary angioplasty closure) when the superiority of its prolonged infusion (for example, 20 minutes) is established 50 will limit its effectiveness. Under these conditions, continuous infusion of formulations previously untested in damaged hearts 51 may be beneficial because time-related benefits of metabolic repair were allowed to become evident, and results are related less to blood cardioplegic composition.
Confusion over the issue of dose versus duration was compounded unintentionally by our study of "overdose" reperfusion normothermic blood cardioplegia 52 in which application of concepts developed in regional ischemic studies (that is, 20 minutes of segmental perfusion at 50 mm Hg pressure) to the global ischemic model caused myocardial stunning, because a 3500 ml dose was given to 25 kg dogs. Global depression by overdose cardioplegia offset the metabolic benefits of warm reperfusion, and this shortcoming was overcome by limiting the total dose and duration. 52 Conversely, delay in regional recovery did not alter global myocardial performance in regional studies, 50 inasmuch as this segment was nonfunctional before treatment and any functional recovery was beneficial. These experimental findings form the basis of our clinical use of brief warm induction and reperfusion in high-risk patients and prolonged regional reperfusion after acute coronary occlusion 34,50,53 or after postoperative intractable ventricular fibrillation. 54,55
Unfortunately, precise monitoring of "how much is enough" is unavailable, so that we have applied the experimentally determined end point of the return of myocardial oxygen consumption to basal levels (that is, 1 ml/100 gm/min) in the arrested decompressed heart. 56 The clinical indicator is when the coronary effluent turns red, signifying minimal oxygen uptake, but this may be misleading because it reflects only the drainage of the region perfused by that delivery route. Reversal of flow (that is, from antegrade to retrograde) results often in increased oxygen and glucose uptake and lactate washout and indicates that different vascular beds are perfused by delivering cardioplegic solution in either direction. 47 It is hoped that clinical monitoring methods will be developed for precise determination of the dose and duration of normothermic blood cardioplegia to optimize its benefits so that intermittent cold ischemia can be used to maximum advantage for surgical precision.
CARDIOPROTECTIVE ADVERSARIAL POSITIONS: AN IMAGINARY DIVIDING LINE
Cardiac surgeons have assumed, for uncertain reasons, adversarial positions in regard to cardioprotective strategies: ventricular fibrillation versus ischemia; crystalloid versus blood cardioplegia; warm versus cold cardioplegia; antegrade versus retrograde delivery; continuous versus intermittent cardioplegia. This has led to the creation of an imaginary dividing line that places them in the awkward position of (1) aligning themselves on one side of the line or another to avoid the appearance of indecision and incomplete allegiance and (2) becoming prisoners of self-imposed inflexibility that may limit clinical adoption of more comprehensive and effective cardioprotective strategies.
The introduction of warm heart surgery has been a valuable contribution because it renews focus on the important issues of normothermic blood cardioplegia and continuous coronary perfusion as cardioprotective adjuncts rather than alternatives to the methods that have been tested extensively. Recognition that aerobic metabolism can be supported by blood cardioplegic perfusion allays anxiety that there is a "battle against the clock when the aortic clamp is in place" and clarifies that "ischemic time and crossclamp time are not synonymous," provided homogeneous cardioplegic distribution is ensured either by combining the benefits of antegrade and retrograde delivery or by adding hypothermia when regional distribution is inadequate to meet energy demands. The major technical limitation of continuous blood cardioplegia is to limit visualization and potentially compromise surgical precision, but this shortcoming can be overcome by intermittent cold ischemia to slow the rate and force of cell damage that is accelerated by normothermia.
Retrograde delivery permits continuous coronary perfusion when visualization is not essential (for example, constructing proximal anastomoses with the aorta vented, placing sutures from valve anulus to valve ring or prosthesis, closure of atrium or aorta). Subsequent strategies will probably include current and future cardioprotective methods, and their acceptance will, I hope, be unimpeded by the rigidity of thinking that underlies the current confusion and retards progress in this evolving field. Incorporation of warm cardioplegia and continuous coronary perfusion into any strategy will likely expedite evolution toward an integrated approach that allows cardiac operations to be done more safely in the increasing number of high-risk patients who require surgical repair of congenital and acquired lesions.
References
This article has been cited by other articles:
|
|
 |
|
  K. A. Eagle, R. A. Guyton, R. Davidoff, G. A. Ewy, J. Fonger, T. J. Gardner, J. P. Gott, H. C. Herrmann, R. A. Marlow, W. C. Nugent, et al. ACC/AHA guidelines for coronary artery bypass graft surgery: A report of the American College of Cardiology/ American Heart Association task force on Practice Guidelines (Committee to revise the 1991 Guidelines for Coronary Artery Bypass Graft Surgery) J. Am. Coll. Cardiol., October 1, 1999; 34(4): 1262 - 1347. [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Cook Changing Temperature Management for Cardiopulmonary Bypass Anesth. Analg., June 1, 1999; 88(6): 1254 - 1254. [Full Text] [PDF]
|
|
|
|
 |
|
 T. S. Larsen, O. Irtun, T. K. Steigen, T. V. Andreasen, and D. Sorlie Myocardial Substrate Oxidation During Warm Continuous Blood Cardioplegia Ann. Thorac. Surg., September 1, 1996; 62(3): 762 - 768. [Abstract] [Full Text]
|
|
|
|
|
|
U. Mehlhorn, S. J. Allen, D. L. Adams, K. L. Davis, G. R. Gogola, and R. D. Warters Cardiac Surgical Conditions Induced by {beta}-Blockade: Effect on Myocardial Fluid Balance Ann. Thorac. Surg., July 1, 1996; 62(1): 143 - 150. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. V. Lichtenstein, C. D. Naylor, C. M. Feindel, K. Sykora, J. G. Abel, A. S. Slutsky, C. D. Mazer, G. T. Christakis, B. S. Goldman, and S. E. Fremes Intermittent Warm Blood Cardioplegia Circulation, November 1, 1995; 92(9): 341 - 346. [Abstract] [Full Text]
|
|
|
|
|
|
U. Mehlhorn, S. J. Allen, D. L. Adams, K. L. Davis, G. R. Gogola, E. R. de Vivie, and G. A. Laine Normothermic Continuous Antegrade Blood Cardioplegia Does Not Prevent Myocardial Edema and Cardiac Dysfunction Circulation, October 1, 1995; 92(7): 1940 - 1946. [Abstract] [Full Text]
|
|
|
|
|
|
W. R. Chitwood Jr and C. L. Wixon Update in Myocardial Protection Ann. Thorac. Surg., January 1, 1995; 59(1): 253 - 254. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
