Integrated Medical-Surgical Care in Acute Coronary Artery Disease Adv. Cardiol., vol. 15, pp. 99-110 (Karger, Basel 1975)
Management of Acute Myocardial Infarction Pathophysiological Considerations BURTON
E. SOBEL
Cardiovascular Division, Washington University School of Medicine, St. Louis, Mo.
The advent of coronary care units has decreased mortality associated with acute myocardial infarction appreciably. As many as 50 % of early hospital deaths can be prevented by prompt prophylaxis and treatment of major dysrhythmias. Nevertheless, substantial residual early and late mortality - often associated with severe impairment of ventricular function - lends impetus to development and implementation of more effective means of managing patients with acute myocardial infarction. During the past several years, experimental and clinical studies have been undertaken to determine whether impairment of ventricular function reflects the overall extent of irreversible ischemic injury (infarct size) sustained by the heart. Results of these studies indicate that acute myocardial infarction is a dynamic process evolving relatively slowly, and that effective management of the condition requires protection of the heart as well as maintenance of adequate peripheral perfusion. An overriding concern in managing patients with acute myocardial infarction in the past has been to maintain peripheral arterial pressure, often requiring administration of potent vasoactive agents. It is becoming increasingly clear that this concern must be tempered by consideration of the direct effects of these agents on the heart and the consequences of the increased ventricular afterload they may produce on jeopardized, ischemic myocardium.
In collaboration with Dr. WILLIAM SHELL and others at the University of California in San Diego, we found that infarct size is an important
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Infarct Size and Prognosis
SOBEL
100
determinant of prognosis in patients with acute myocardial infarction [1]. In a series of 155 patients studied consecutively, overall early mortality (within 1 month of onset of infarction) was 210/0. Mortality was 360/0 in patients within this group with infarct size estimated to exceed 50 creatine phosphocreatinase (CPK) gEq 1 by analysis of serial changes in serum CPK activity. In marked contrast, those patients with infarct size estimated to be less than 50 CPK gEq had an early mortality of only 30/0. These findings confirmed our earlier observations in a small group of patients in which mortality and morbidity after acute myocardial infarction during a 6-month follow-up period were found to correlate closely with infarct size estimated from analysis of serial serum CPK changes [2]. Others have observed that marked impairment of ventricular function after acute myocardial infarction, particularly cardiogenic shock leading to death, is associated with extensive infarction detected morphologically [3,4]. Thus, one important determinant of the outcome of acute myocardial infarction is the extent of irreversible ischemic injury sustained by the heart. Recognition of this relationship prompts a series of questions. 1. Can the extent of irreversible injury sustained by the heart be limited during early evolution of infarction? 2. Would protection of myocardium and favorable modification of infarct size improve morbidity and mortality? 3. How can we objectively evaluate the efficacy of interventions designed to modify infarct size favorably?
Modification of Infarct Size
1 One CPK gEq is that amount of myocardium undergoing infarction liberating the same amount of CPK into the circulation as 1 g of myocardium undergoing homogeneous necrosis.
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
In studies with Dr. EUGENE BRAUNWALD, PETER MARoKO, JOHN Ross, jr., and others at the University of California, San Diego, some of these questions were explored. Favorable modification of infarct size was demonstrated in experimental animal preparations in which the balance between myocardial oxygen supply and myocardial oxygen demand was modified during early evolution of infarction [5,6]. Agents which increased myocardial oxygen requirements led to extension of infarction. Physiological or pharmacological interventions which diminished oxygen
Management of Acute Myocardial Infarction
101
requirements of the jeopardized ischemic tissue appeared to salvage myocardium and limit the ultimate extent of infarction. Thus, it seemed likely that selected interventions could favorably influence the evolution of irreversible ischemic injury. It is well recognized that oxygen requirements of the heart depend primarily upon: (1) heart rate; (2) intramyocardial wall tension or stress, and (3) contractile state. Wall tension is directly related to developed intraventricular pressure and to ventricular volume. Thus, myocardial oxygen requirements increase when stroke volume of the depressed heart is maintained by compensatory cardiac dilatation. Oxygen requirements associated with eletcrical activation of the heart, basal myocardial metabolism or ejection of blood against a low impedence are relatively modest. In view of these considerations, one would anticipate that salvage of jeopardized ischemic myocardium would be potentiated most by diminution of ventricular volume, slowing of heart rate or reduction of the influence of agents with positive inotropic effects impinging on the ischemic tissue. On the other hand, protection of jeopardized ischemic myocardium could be achieved by improvement of the delivery of oxygen to myocardium, for example by early coronary revascularization, or by facilitation of metabolic processes capable of generating intracellular adenosine triphosphate (ATP) in the face of myocardial hypoxia. This concept underlies efforts to salvage jeopardized myocardium by administration of glucose to enhance anerobic glycolytic metabolism in jeopardized tissue. As noted in the preceding discussion [7] we have utilized estimation and prediction of infarct size by analysis of serial changes in serum CPK activity to examine the effects of several physiological and pharmacological interventions on the evolution of experimentally induced ischemic injury. Examples of interventions influencing each of the major determinants of myocardial oxygen requirements follow.
Augmentation of heart rate in unanesthetized dogs, including some with experimentally induced atrio-ventricular block, markedly influenced the extent of ischemic injury sustained by the heart after experimental coronary occlusion (fig.1) [8]. Coronary occlusion was produced by constriction of an externalized coronary arterial snare placed 3-5 days
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Heart Rate
SOBEL
102
HR;'20 HR= 95-105 (spontaneous)
'S2
•
HR=T4
HR=160
Pacemaker on
~.~--------------------------.
2.8
x
(09g)
E ~ 2.0
(45g)
"2Ul
'"
~
:
§::J
E
3
04
O~-'----'-----r----.-----r----r----'-----r--
o
300 600 900 1,200 1,500 1,800 Time after coronary artery Occlusion, min
2,100
2,4.00
previously. Heart rate was augmented at selected intervals after coronary occlusion by ventricular pacing, or administration of isoproterenol or atropine. Left atrial pressure was monitored in selected animals from each group to exclude the possibility that changes in infarct size reflected hemodynamic deterioration. Infarct size and the rate of evolution of infarction were quantified by analysis of serial changes in serum CPK activity in samples obtained hourly. Augmentation of heart rate 14 h after coronary occlusion by ventricular pacing or administration of isoproterenol or atropine led to extension of infarction averaging 40, 72 and 40 Ofo of initial infarct size, respectively. Even in animals with atrio-ventricular block (in which resting heart rate was 60 beats/min) modest increases in heart rate to 90 beats/min led to marked increases in infarct size. Surprisingly, augmentation of heart rate as late as 72 h after coronary occlusion led to extension of infarction. Thus, the results in this experimental study indicate that acceleration of heart rate, presumably leading to further imbalance between myocardial oxygen supply and demand in jeopardized ischemic myocardium, led to marked extension of infarction, even when heart rate
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Fig. 1. Augmentation of infarct size in association with acceleration of heart rate (HR) in a conscious dog with coronary occlusion. Cumulative creatine phosphocreatinase (CPK) released, an index of infarct size, was measured by analysis of serial serum CPK changes and is plotted on the ordinate. As can be seen, when heart rate was accelerated by ventricular pacing cumulative CPK released increased.
Management of Acute Myocardial Infarction
103
was accelerated from an initially low resting rate or when it was accelerated relatively late after coronary occlusion. Inferentially then, bradycardia or prevention of tachycardia might well protect ischemic myocardium in hemodynamically stable patients with evolving infarction.
Ventricular Wall Tension In collaboration with Dr. DELARIA, Dr. BERNSTEIN and others [9] we examined the effect of intra-aortic balloon counter-pUlsation on experimental animals with coronary occlusion. This intervention may decrease oxygen requirements by diminution of ventricular afterload, increase coronary flow by augmenting diastolic perfusion pressure, or both. In a group of 38 dogs with coronary occlusion, the severity of ischemic injury was assessed 15 min after occlusion with the use of epicardial ST segment electrocardiographic maps. At selected intervals after coronary occlusion myocardial CPK, myocardial ATP, and morphology were assessed in biopsies of the heart corresponding to the sites from which initial ST segment recordings had been obtained. Alternate dogs were treated with intra-aortic balloon counter-pulsation. In general, depletion of ATP, loss of myocardial CPK activity and histologic evidence of necrosis occur sequentially in ischemic myocardium undergoing infarction. Results in our investigation indicated that intra-aortic balloon counter-pulsation delayed the evolution of ischemic injury. This was manifested by decline in the rate of disappearance of myocardial ATP and a delay in the appearance of histologic evidence of necrosis in ischemic zones. On the other hand, based on analysis of myocardial CPK activity in the same preparations, it appears that intra-aortic balloon counter-pulsation did not preclude the development of gross biochemical deterioration of ischemic myocardium. Nevertheless, a procedure designed to reduce myocardial oxygen requirements and enhance coronary perfusion delayed the evolution of irreversible injury in dogs subjected to coronary occlusion.
Presumably, the oxygen requirements of ischemic myocardium would be reduced by inhibition of fJ-adrenergic stimulation and its effect on contractile state. Accordingly, we examined the effect of administration of pro-
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Contractile State
SOBEL
104
pranolol to conscious dogs with experimental coronary occlusion. Infarct size was predicted 5 h after coronary occlusion by projection of serum CPK values on the basis of early serial serum CPK changes. At this time propranolol was administered in a dose sufficient to maintain heart rate below 85 beats/min [10]. Observed infarct size was estimated from serial serum CPK changes both before and after administration of propranolol and verified by analysis of myocardial CPK activity measured in myocardial extracts obtained 24 h after coronary occlusion. Mean left atrial pressure did not increase in any of the animals studied. In 7 of 11 dogs studied, propranolol decreased infarct size by an average of 460/0. Despite early favorable trends in the 4 other animals, they did not exhibit any net reduction of observed compared to predicted infarct size. These results indicate that in a majority of experimental animals with coronary occlusion, reduction of myocardial oxygen requirements by ,8-adrenergic blockade favorably modified the evolution of ischemic injury sustained by the heart.
Coronary revascularization has received considerable attention as a means by which to reduce symptoms and perhaps improve survival in patients with chronic coronary artery disease. Recently, it has also been employed in the setting of acute myocardial infarction. A priori, the most effective means to improve myocardial energy production is to enhance delivery of oxygenated blood to ischemic tissue. In collaboration with Dr. GINKS, Dr. MARoKo, Dr. COVELL, Dr. Ross and others we examined the effects of coronary artery reperfusion in experimental animals with coronary occlusion [11,12]. Infarct size was compared to the zone of hypoperfusion assessed with the use of epicardial ST segment maps and with calibrated photographs beginning 15 min after occlusion. Ultimate infarct size was assessed morphologically and by analysis of myocardial CPK activity. In animals subjected to coronary reperfusion 3 h after coronary occlusion, marked preservation of CPK activity in initially hypoperfused zones was present 1 week later. Histologic evaluation of biopsies from the same sites confirmed the survival of tissue in these regions. Local ventricular function improved within 1 h of reperfusion manifested by reversal of paradoxical movement of the left ventricular wall. However, early reperfusion appeared to be a mixed blessing. In 16 conscious dogs, in which reperfusion was initiated 5 h after coronary occlusion by deflation
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Facilitation of Myocardial Energy Production
Management of Acute Myocardial Infarction 300 ..,
105
0 = pred Icted mil = observed
'0 2u
~0.200 "o
t
CI>
-~ N
"iii
E 100 n=7
Salvage
Extension
of an externalized coronary cuff, two disparate effects were observed (fig. 2) [13]. In all animals, CPK activity was measured hourly in serial serum samples. Infarct size was predicted from projected serum CPK values based on projected curves obtained from the initial 5-hour values [14]. In control animals projected CPK values correlated closely with those observed during the ensuing 48 h. Thus, observed and predicted infarct sizes also correlated closely. In 9 of the 16 dogs subjected to reperfusion subsequent CPK values deviated markedly below those projected, reflecting salvage of an average of 42 Ofo of myocardium which would otherwise have undergone infarction. However, in 7 of the 16 dogs subjected to reperfusion 5 h after occlusion, observed infarct size exceeded predicted infarct size by more than 100 Ofo. The extension of infarction produced by reperfusion in these animals was verified by direct analysis of myocardium and was found to be associated with extensive hemorrhage into the heart. These results indicate that reperfusion is not without hazard during
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Fig. 2. Effects of reperfusion on infarct size in conscious dogs. In control animals subjected to coronary occlusion predicted and observed infarct size correlated closely. In conrast, when coronary reperfusion was initiated 5 h after occlusion, observed infarct size deviated markedly from predicted. In one group of animals reperfusion resulted in salvage of myocardium; in the other, it led to massive extension of infarction verified at autopsy and associated with hemorrhage into myocardium.
106
early evolution of myocardial infarction. It appears that a critical time exists before which reperfusion protects the ischemic myocardium by improving the delivery of oxygen and substrate, thereby reducing ultimate infarct size. On the other hand, when reperfusion is carried out after the vasculature itself has lost functional integrity, it results in myocardial hemorrhage with extension of infarction. Myocardial energy production theoretically can be enhanced by augmentation of anerobic metabolism. With this in mind, we administered glucose and insulin to anoxic perfused guinea pig hearts for prolonged intervals [15]. During early anoxia, administration of supplemental substrate increased peak left ventricular pressure and dP/dt more than 2-fold. Biochemical criteria, such as mitochondrial oxidative phosphorylation and maintenance of myocardial CPK activity, indicated that anoxic hearts perfused with supplemental glucose for as long as 10 h exhibited less deterioration than their counterparts perfused with glucose at physiological concentrations. Nonmetabolizable sugars, at equimolar concentrations, were not effective. In dogs subjected to coronary occlusion systemic administration of glucose-insulin-potassium appears to reduce the severity and extent of ischemic injury. In studies performed with Dr. MARoKo, Dr. LIBBY, Dr. BRAUNWALD and others, we found that this intervention protected ischemic myocardium [5]. The severity of ischemia after coronary occlusion in selected regions of myocardium was assessed by epicardial ST segment electrocardiographic recordings. In control animals the extent of CPK depletion 24 h later from the same sites bore a consistent relationship to the magnitude of initial ST segment elevation. Furthermore, initial ST segment elevation presaged histologic evidence of necrosis 24 h later in the same sites. In animals treated with glucose-insulin-potassium beginning 30 min after experimental coronary occlusion, protection of myocardium was demonstrated. Thus, with respect to initial ST segment elevation (before administration of glucose) CPK depletion and histologic evidence of necrosis 24 h later were much less marked. One might argue that administration of glucose systemically is unlikely to favorably influence ischemic myocardium because the limitation of blood flow would preclude delivery of the supplemental substrate. However, in additional experiments we have demonstrated marked increases in myocardial ATP in ischemic myocardium of the open-chest dog after administration of glucose systemically (fig. 3) and have found that radioactively labeled glucose does reach intracellular sites in ischemic tissue.
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
SOBEL
Management of Acute Myocardial Infarction
107
100 90
80 70
60 ;;'!.
50
c
o
~ 40 Ci
'"
"0
a. 30 ~
~
'g"
:>.
20
::;: 10
r-.i
~i
g>i
a.i
m
W ~ 00 00 1W Time after coronary occlusion, min
Fig. 3. Salutary effects on repletion of myocardial adenosine triphosphate (ATP) of administration of glucose (G) systemically to dogs with coronary occlusion. In the 3 animals illustrated myocardial ATP content was assayed in ischemic zones with comparable ST segment elevation in epicardial electrograms at selected intervals after coronary occlusion. As can be seen, ATP depletion (plotted on the ordinate based on control values from the same animals) increased as a function of time after coronary occlusion. However, administration of glucose intravenously led to replenishment of ATP in ischemic zones 20 min later.
These findings indicate that under selected experimental conditions, ischemic myocardium can be salvaged by improvement of myocardial energy production.
Selected physiological and pharmacological interventions, implemented several hours after the onset of myocardial infarction, appear to modify favorably the evolution of ischemic injury in experimental animals. Since infarct size is an important determinant of prognosis, it follows that protection of ischemic myocardium in patients might reduce morbidity and
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Clinical Implications
SOBEL
108
mortality from acute myocardial infarction. Preliminary clinical findings suggest that favorable modification of infarct size is possible. Betaadrenergic blockade [16], reduction of ventricular afterload with intraaortic balloon counter-pulsation [17], or pharmacologically [18], appears to benefit ischemic myocardium, reflected by electrophysiological criteria. With the use of serial serum CPK analyses in patients with acute myocardial infarction we have found that reduction of ventricular afterload by administration of trimethaphan appears to protect myocardium. In a preliminary study, 14 hypertensive patients with acute myocardial infarction (blood pressure greater than 150/90 mm Hg) were treated after infarct size had been predicted by analysis of serum CPK changes during the first 7 h after apparent onset of infarction. Intra-arterial blood pressure, pulmonary artery wedge pressure, cardiac output (determined by dye dilution) and ejection fraction (determined by radioisotope angiocardiography) were measured serially. In 30 consecutive control patients and 10 hypertensive controls with infarction, observed and predicted infarct size correlated closely (r = 0.93). In contrast, hypertensive patients treated with trimethaphan consistently exhibited a reduction of observed compared to predicted infarct size (average difference 31 ± 5 % standard error [SED. Cardiac output, ejection fraction and circumferential fiber shortening velocity increased by average of 12, 37 and 400/0, respectively, although heart rate did not change appreciably. Pulmonary artery wedge pressure declined by an average of 20 0/0. Mortality within 30 days after infarction was less than that in controls with corresponding infarct size. These preliminary results suggest that cautious reduction of ventricular afterload in patients with acute myocardial infarction may decrease the extent of irreversible ischemic injury sustained by the heart by improving the balance between myocardial oxygen supply and demand. Although it is too early to tell whether the apparent favorable modification of infarct size will be reflected by a meaningful reduction of mortality, the early mortality figures in this small sample are encouraging.
Development of improved techniques for assessing infarct size in intact experimental animals and in patients has made it possible to evaluate the effectiveness of selected interventions more objectively. It is becoming increasingly apparent that infarct size is an important determinant of prognosis. Experimentally, infarct size can be modified favorably under specified conditions by interventions designed to
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Summary
Management of Acute Myocardial Infarction
109
reduce myocardial oxygen requirements or increase myocardial energy supply during the early evolution of infarction. Preliminary clinical applications of these concepts have been encouraging. The time seems to be right for carefully controlled clinical investigations to determine whether techniques designed to protect ischemic myocardium will be effective in improving survival and reducing morbidity from acute myocardial infarction. On the basis of available evidence, it seems clear that appropriate management of patients with this disorder should include careful attention to the impact of any therapeutic modality on the heart itself. It does not suffice to treat patients with p-adrenergic agonists simply because they may elevate systemic arterial blood pressure. Hemodynamics can not be the sole criteria in guiding the physician charged with the responsibility for patient management. Rather, he must consider the potential effects of any contemplated therapeutic modality on myocardial viability as well. The ultimate place of interventions designed to protect ischemic myocardium in the therapeutic armamentarium remains to be determined. However, judicious therapy should be designed to minimize myocardial oxygen requirements and maximize myocardial energy supply as well as to prevent critical impairment of peripheral perfusion.
1 2 3 4 5 6
7 8 9
10
SHELL, W. E. and SOBEL, B. E.: Infarct size index. An effective predictor of prognosis after myocardial infarction. Circulation 48: suppl. IV, p.39 (1973). SOBEL, B. E.; BRESNAHAN, G. F.; SHELL, W. E., and YODER, R. D.: Estimation of infarct size in man and its relation to prognosis. Circulation 46: 640 (1972). PAGE, D. L.; CAULFIELD, J. B.; KASTOR, J. A., and DESANCTIS, R. W.: Myocardial changes associated with cardiogenic shock. New Engl. J. Med. 285: 133 (1971). SHELL, W. E.; LAVELLE, J. F.; COVELL, J. W., and SOBEL, B. E.: Early estimation of myocardial damage in conscious dogs and patients with evolving acute myocardial infarction. J. clin. Invest. 52: 2579 (1973). MAROKO, P. R.; KrnKSHUS, J. K.; SOBEL, B. E.; WATANABE, T.; COVELL, J. W.; Ross, J., jr., and BRAUNWALD, E.: Factors influencing infarct size following experimental coronary artery occlusion. Circulation 43: 67 (1971). MAROKO, P. R.; LmBY, P.; SOBEL, B. E.; BLOOR, C. M.; SYBERS, H. D.; SHELL, W. E.; COVELL, J. W., and BRAUNWALD, E.: Effect of glucose-insulin-potassium infusion on myocardial infarction following experimental coronary artery occlusion. Circulation 45: 1160 (1972). SOBEL, B. E.: Quantification of myocardial ischemic injury; in VOGEL Adv. Cardiol., vol. 15, pp. 86-98 (Karger, Basel 1975). SHELL, W. E. and SOBEL, B. E.: Deleterious effects of increased heart rate on infarct size in the conscious dog. Amer. J. Cardiol. 31: 474 (1973). DELARIA, G. A; JOHANSEN, K. H.; SOBEL, B. E.; SYBERS, H. D., and BERNSTEIN, E. F.: Delayed evolution of myocardial ischemic injury after intra-aortic balloon counterpulsation. Circulation 49 and 50 (Suppl. II): 242 (1974). SHELL, W. E. and SOBEL, B. E.: Changes in infarct size following administration of propranolol in the conscious dog. Amer. J. Cardiol. 31: 157 (1973).
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
References
SOBEL
12
13
14
15 16
17
18
MAROKO, P. R.; LmBY, P.; OINKS, W. R.; BLOOR, C. M.; SHELL, W. E.; SOBEL, B. E., and Ross, J., jr.: Coronary artery reperfusion. I. Early effects on local myocardial function and the extent of myocardial necrosis. J. clin. Invest. 51: 2710 (1972). OINKS, W. R.; SYBERS, H. D.; MAROKO, P. R.; COVELL, J. W.; SOBEL, B. E., and Ross, J., jr.: Coronary artery reperfusion. II. Reduction of myocardial infarct size at one week after the coronary occlusion. J. clin. Invest. 51: 2717 (1972). BRESNAHAN, O. F.; ROBERTS, R.; SHELL, W. E.; Ross, J., jr., and SOBEL, B. E.: Deleterious effects due to hemorrhage after myocardial reperfusion. Amer. J. Cardiol. 33: 82 (1974). SHELL, W. E.; LAVELLE, J. F.; COVELL, J. W., and SOBEL, B. E.: Early estimation of myocardial damage in conscious dogs and patients with evolving acute myocardial infarction. J. clin. Invest. 52: 2579 (1973). HENRY, P. D.; SOBEL, B. E., and BRAUNWALD, E.: Protection of hypoxic guinea pig hearts with glucose and insulin. Amer. J. Physiol. 226: 309 (1974). PELIDES, L. J.; REID, D. W.; THOMAS, M., and SHILLINGFORD, J. P.: Inhibition by beta-blockade of the ST segment elevation after acute myocardial infarction in man. Cardiovasc. Res. 6: 295 (1972). MAROKO, P. R.; BERNSTEIN, E. F.; LIBBY, P.; DELARIA, O. A; COVELL, J. W.; Ross, J., jr., and BRAUNWALD, E.: The effects of intra-aortic balloon counterpulsation on the severity of myocardial ischemic injury following acute coronary occlusion. Counterpulsation and myocardial injury. Circulation 45: 1150 (1972). SHELL, W. E.; EHSANI, A A, and SOBEL, B. E.: Reduction of infarct size in hypertensive patients with acute myocardial infarction. J. clin. Invest. 52: 76a (1973).
Author's address: BURTON E. SOBEL, M.D., Cardiovascular Division, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110 (USA)
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
11
110
Management of Acute Myocardial Infarction Pathophysiological Considerations BURTON
E. SOBEL
Cardiovascular Division, Washington University School of Medicine, St. Louis, Mo.
The advent of coronary care units has decreased mortality associated with acute myocardial infarction appreciably. As many as 50 % of early hospital deaths can be prevented by prompt prophylaxis and treatment of major dysrhythmias. Nevertheless, substantial residual early and late mortality - often associated with severe impairment of ventricular function - lends impetus to development and implementation of more effective means of managing patients with acute myocardial infarction. During the past several years, experimental and clinical studies have been undertaken to determine whether impairment of ventricular function reflects the overall extent of irreversible ischemic injury (infarct size) sustained by the heart. Results of these studies indicate that acute myocardial infarction is a dynamic process evolving relatively slowly, and that effective management of the condition requires protection of the heart as well as maintenance of adequate peripheral perfusion. An overriding concern in managing patients with acute myocardial infarction in the past has been to maintain peripheral arterial pressure, often requiring administration of potent vasoactive agents. It is becoming increasingly clear that this concern must be tempered by consideration of the direct effects of these agents on the heart and the consequences of the increased ventricular afterload they may produce on jeopardized, ischemic myocardium.
In collaboration with Dr. WILLIAM SHELL and others at the University of California in San Diego, we found that infarct size is an important
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Infarct Size and Prognosis
SOBEL
100
determinant of prognosis in patients with acute myocardial infarction [1]. In a series of 155 patients studied consecutively, overall early mortality (within 1 month of onset of infarction) was 210/0. Mortality was 360/0 in patients within this group with infarct size estimated to exceed 50 creatine phosphocreatinase (CPK) gEq 1 by analysis of serial changes in serum CPK activity. In marked contrast, those patients with infarct size estimated to be less than 50 CPK gEq had an early mortality of only 30/0. These findings confirmed our earlier observations in a small group of patients in which mortality and morbidity after acute myocardial infarction during a 6-month follow-up period were found to correlate closely with infarct size estimated from analysis of serial serum CPK changes [2]. Others have observed that marked impairment of ventricular function after acute myocardial infarction, particularly cardiogenic shock leading to death, is associated with extensive infarction detected morphologically [3,4]. Thus, one important determinant of the outcome of acute myocardial infarction is the extent of irreversible ischemic injury sustained by the heart. Recognition of this relationship prompts a series of questions. 1. Can the extent of irreversible injury sustained by the heart be limited during early evolution of infarction? 2. Would protection of myocardium and favorable modification of infarct size improve morbidity and mortality? 3. How can we objectively evaluate the efficacy of interventions designed to modify infarct size favorably?
Modification of Infarct Size
1 One CPK gEq is that amount of myocardium undergoing infarction liberating the same amount of CPK into the circulation as 1 g of myocardium undergoing homogeneous necrosis.
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
In studies with Dr. EUGENE BRAUNWALD, PETER MARoKO, JOHN Ross, jr., and others at the University of California, San Diego, some of these questions were explored. Favorable modification of infarct size was demonstrated in experimental animal preparations in which the balance between myocardial oxygen supply and myocardial oxygen demand was modified during early evolution of infarction [5,6]. Agents which increased myocardial oxygen requirements led to extension of infarction. Physiological or pharmacological interventions which diminished oxygen
Management of Acute Myocardial Infarction
101
requirements of the jeopardized ischemic tissue appeared to salvage myocardium and limit the ultimate extent of infarction. Thus, it seemed likely that selected interventions could favorably influence the evolution of irreversible ischemic injury. It is well recognized that oxygen requirements of the heart depend primarily upon: (1) heart rate; (2) intramyocardial wall tension or stress, and (3) contractile state. Wall tension is directly related to developed intraventricular pressure and to ventricular volume. Thus, myocardial oxygen requirements increase when stroke volume of the depressed heart is maintained by compensatory cardiac dilatation. Oxygen requirements associated with eletcrical activation of the heart, basal myocardial metabolism or ejection of blood against a low impedence are relatively modest. In view of these considerations, one would anticipate that salvage of jeopardized ischemic myocardium would be potentiated most by diminution of ventricular volume, slowing of heart rate or reduction of the influence of agents with positive inotropic effects impinging on the ischemic tissue. On the other hand, protection of jeopardized ischemic myocardium could be achieved by improvement of the delivery of oxygen to myocardium, for example by early coronary revascularization, or by facilitation of metabolic processes capable of generating intracellular adenosine triphosphate (ATP) in the face of myocardial hypoxia. This concept underlies efforts to salvage jeopardized myocardium by administration of glucose to enhance anerobic glycolytic metabolism in jeopardized tissue. As noted in the preceding discussion [7] we have utilized estimation and prediction of infarct size by analysis of serial changes in serum CPK activity to examine the effects of several physiological and pharmacological interventions on the evolution of experimentally induced ischemic injury. Examples of interventions influencing each of the major determinants of myocardial oxygen requirements follow.
Augmentation of heart rate in unanesthetized dogs, including some with experimentally induced atrio-ventricular block, markedly influenced the extent of ischemic injury sustained by the heart after experimental coronary occlusion (fig.1) [8]. Coronary occlusion was produced by constriction of an externalized coronary arterial snare placed 3-5 days
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Heart Rate
SOBEL
102
HR;'20 HR= 95-105 (spontaneous)
'S2
•
HR=T4
HR=160
Pacemaker on
~.~--------------------------.
2.8
x
(09g)
E ~ 2.0
(45g)
"2Ul
'"
~
:
§::J
E
3
04
O~-'----'-----r----.-----r----r----'-----r--
o
300 600 900 1,200 1,500 1,800 Time after coronary artery Occlusion, min
2,100
2,4.00
previously. Heart rate was augmented at selected intervals after coronary occlusion by ventricular pacing, or administration of isoproterenol or atropine. Left atrial pressure was monitored in selected animals from each group to exclude the possibility that changes in infarct size reflected hemodynamic deterioration. Infarct size and the rate of evolution of infarction were quantified by analysis of serial changes in serum CPK activity in samples obtained hourly. Augmentation of heart rate 14 h after coronary occlusion by ventricular pacing or administration of isoproterenol or atropine led to extension of infarction averaging 40, 72 and 40 Ofo of initial infarct size, respectively. Even in animals with atrio-ventricular block (in which resting heart rate was 60 beats/min) modest increases in heart rate to 90 beats/min led to marked increases in infarct size. Surprisingly, augmentation of heart rate as late as 72 h after coronary occlusion led to extension of infarction. Thus, the results in this experimental study indicate that acceleration of heart rate, presumably leading to further imbalance between myocardial oxygen supply and demand in jeopardized ischemic myocardium, led to marked extension of infarction, even when heart rate
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Fig. 1. Augmentation of infarct size in association with acceleration of heart rate (HR) in a conscious dog with coronary occlusion. Cumulative creatine phosphocreatinase (CPK) released, an index of infarct size, was measured by analysis of serial serum CPK changes and is plotted on the ordinate. As can be seen, when heart rate was accelerated by ventricular pacing cumulative CPK released increased.
Management of Acute Myocardial Infarction
103
was accelerated from an initially low resting rate or when it was accelerated relatively late after coronary occlusion. Inferentially then, bradycardia or prevention of tachycardia might well protect ischemic myocardium in hemodynamically stable patients with evolving infarction.
Ventricular Wall Tension In collaboration with Dr. DELARIA, Dr. BERNSTEIN and others [9] we examined the effect of intra-aortic balloon counter-pUlsation on experimental animals with coronary occlusion. This intervention may decrease oxygen requirements by diminution of ventricular afterload, increase coronary flow by augmenting diastolic perfusion pressure, or both. In a group of 38 dogs with coronary occlusion, the severity of ischemic injury was assessed 15 min after occlusion with the use of epicardial ST segment electrocardiographic maps. At selected intervals after coronary occlusion myocardial CPK, myocardial ATP, and morphology were assessed in biopsies of the heart corresponding to the sites from which initial ST segment recordings had been obtained. Alternate dogs were treated with intra-aortic balloon counter-pulsation. In general, depletion of ATP, loss of myocardial CPK activity and histologic evidence of necrosis occur sequentially in ischemic myocardium undergoing infarction. Results in our investigation indicated that intra-aortic balloon counter-pulsation delayed the evolution of ischemic injury. This was manifested by decline in the rate of disappearance of myocardial ATP and a delay in the appearance of histologic evidence of necrosis in ischemic zones. On the other hand, based on analysis of myocardial CPK activity in the same preparations, it appears that intra-aortic balloon counter-pulsation did not preclude the development of gross biochemical deterioration of ischemic myocardium. Nevertheless, a procedure designed to reduce myocardial oxygen requirements and enhance coronary perfusion delayed the evolution of irreversible injury in dogs subjected to coronary occlusion.
Presumably, the oxygen requirements of ischemic myocardium would be reduced by inhibition of fJ-adrenergic stimulation and its effect on contractile state. Accordingly, we examined the effect of administration of pro-
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Contractile State
SOBEL
104
pranolol to conscious dogs with experimental coronary occlusion. Infarct size was predicted 5 h after coronary occlusion by projection of serum CPK values on the basis of early serial serum CPK changes. At this time propranolol was administered in a dose sufficient to maintain heart rate below 85 beats/min [10]. Observed infarct size was estimated from serial serum CPK changes both before and after administration of propranolol and verified by analysis of myocardial CPK activity measured in myocardial extracts obtained 24 h after coronary occlusion. Mean left atrial pressure did not increase in any of the animals studied. In 7 of 11 dogs studied, propranolol decreased infarct size by an average of 460/0. Despite early favorable trends in the 4 other animals, they did not exhibit any net reduction of observed compared to predicted infarct size. These results indicate that in a majority of experimental animals with coronary occlusion, reduction of myocardial oxygen requirements by ,8-adrenergic blockade favorably modified the evolution of ischemic injury sustained by the heart.
Coronary revascularization has received considerable attention as a means by which to reduce symptoms and perhaps improve survival in patients with chronic coronary artery disease. Recently, it has also been employed in the setting of acute myocardial infarction. A priori, the most effective means to improve myocardial energy production is to enhance delivery of oxygenated blood to ischemic tissue. In collaboration with Dr. GINKS, Dr. MARoKo, Dr. COVELL, Dr. Ross and others we examined the effects of coronary artery reperfusion in experimental animals with coronary occlusion [11,12]. Infarct size was compared to the zone of hypoperfusion assessed with the use of epicardial ST segment maps and with calibrated photographs beginning 15 min after occlusion. Ultimate infarct size was assessed morphologically and by analysis of myocardial CPK activity. In animals subjected to coronary reperfusion 3 h after coronary occlusion, marked preservation of CPK activity in initially hypoperfused zones was present 1 week later. Histologic evaluation of biopsies from the same sites confirmed the survival of tissue in these regions. Local ventricular function improved within 1 h of reperfusion manifested by reversal of paradoxical movement of the left ventricular wall. However, early reperfusion appeared to be a mixed blessing. In 16 conscious dogs, in which reperfusion was initiated 5 h after coronary occlusion by deflation
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Facilitation of Myocardial Energy Production
Management of Acute Myocardial Infarction 300 ..,
105
0 = pred Icted mil = observed
'0 2u
~0.200 "o
t
CI>
-~ N
"iii
E 100 n=7
Salvage
Extension
of an externalized coronary cuff, two disparate effects were observed (fig. 2) [13]. In all animals, CPK activity was measured hourly in serial serum samples. Infarct size was predicted from projected serum CPK values based on projected curves obtained from the initial 5-hour values [14]. In control animals projected CPK values correlated closely with those observed during the ensuing 48 h. Thus, observed and predicted infarct sizes also correlated closely. In 9 of the 16 dogs subjected to reperfusion subsequent CPK values deviated markedly below those projected, reflecting salvage of an average of 42 Ofo of myocardium which would otherwise have undergone infarction. However, in 7 of the 16 dogs subjected to reperfusion 5 h after occlusion, observed infarct size exceeded predicted infarct size by more than 100 Ofo. The extension of infarction produced by reperfusion in these animals was verified by direct analysis of myocardium and was found to be associated with extensive hemorrhage into the heart. These results indicate that reperfusion is not without hazard during
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Fig. 2. Effects of reperfusion on infarct size in conscious dogs. In control animals subjected to coronary occlusion predicted and observed infarct size correlated closely. In conrast, when coronary reperfusion was initiated 5 h after occlusion, observed infarct size deviated markedly from predicted. In one group of animals reperfusion resulted in salvage of myocardium; in the other, it led to massive extension of infarction verified at autopsy and associated with hemorrhage into myocardium.
106
early evolution of myocardial infarction. It appears that a critical time exists before which reperfusion protects the ischemic myocardium by improving the delivery of oxygen and substrate, thereby reducing ultimate infarct size. On the other hand, when reperfusion is carried out after the vasculature itself has lost functional integrity, it results in myocardial hemorrhage with extension of infarction. Myocardial energy production theoretically can be enhanced by augmentation of anerobic metabolism. With this in mind, we administered glucose and insulin to anoxic perfused guinea pig hearts for prolonged intervals [15]. During early anoxia, administration of supplemental substrate increased peak left ventricular pressure and dP/dt more than 2-fold. Biochemical criteria, such as mitochondrial oxidative phosphorylation and maintenance of myocardial CPK activity, indicated that anoxic hearts perfused with supplemental glucose for as long as 10 h exhibited less deterioration than their counterparts perfused with glucose at physiological concentrations. Nonmetabolizable sugars, at equimolar concentrations, were not effective. In dogs subjected to coronary occlusion systemic administration of glucose-insulin-potassium appears to reduce the severity and extent of ischemic injury. In studies performed with Dr. MARoKo, Dr. LIBBY, Dr. BRAUNWALD and others, we found that this intervention protected ischemic myocardium [5]. The severity of ischemia after coronary occlusion in selected regions of myocardium was assessed by epicardial ST segment electrocardiographic recordings. In control animals the extent of CPK depletion 24 h later from the same sites bore a consistent relationship to the magnitude of initial ST segment elevation. Furthermore, initial ST segment elevation presaged histologic evidence of necrosis 24 h later in the same sites. In animals treated with glucose-insulin-potassium beginning 30 min after experimental coronary occlusion, protection of myocardium was demonstrated. Thus, with respect to initial ST segment elevation (before administration of glucose) CPK depletion and histologic evidence of necrosis 24 h later were much less marked. One might argue that administration of glucose systemically is unlikely to favorably influence ischemic myocardium because the limitation of blood flow would preclude delivery of the supplemental substrate. However, in additional experiments we have demonstrated marked increases in myocardial ATP in ischemic myocardium of the open-chest dog after administration of glucose systemically (fig. 3) and have found that radioactively labeled glucose does reach intracellular sites in ischemic tissue.
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
SOBEL
Management of Acute Myocardial Infarction
107
100 90
80 70
60 ;;'!.
50
c
o
~ 40 Ci
'"
"0
a. 30 ~
~
'g"
:>.
20
::;: 10
r-.i
~i
g>i
a.i
m
W ~ 00 00 1W Time after coronary occlusion, min
Fig. 3. Salutary effects on repletion of myocardial adenosine triphosphate (ATP) of administration of glucose (G) systemically to dogs with coronary occlusion. In the 3 animals illustrated myocardial ATP content was assayed in ischemic zones with comparable ST segment elevation in epicardial electrograms at selected intervals after coronary occlusion. As can be seen, ATP depletion (plotted on the ordinate based on control values from the same animals) increased as a function of time after coronary occlusion. However, administration of glucose intravenously led to replenishment of ATP in ischemic zones 20 min later.
These findings indicate that under selected experimental conditions, ischemic myocardium can be salvaged by improvement of myocardial energy production.
Selected physiological and pharmacological interventions, implemented several hours after the onset of myocardial infarction, appear to modify favorably the evolution of ischemic injury in experimental animals. Since infarct size is an important determinant of prognosis, it follows that protection of ischemic myocardium in patients might reduce morbidity and
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Clinical Implications
SOBEL
108
mortality from acute myocardial infarction. Preliminary clinical findings suggest that favorable modification of infarct size is possible. Betaadrenergic blockade [16], reduction of ventricular afterload with intraaortic balloon counter-pulsation [17], or pharmacologically [18], appears to benefit ischemic myocardium, reflected by electrophysiological criteria. With the use of serial serum CPK analyses in patients with acute myocardial infarction we have found that reduction of ventricular afterload by administration of trimethaphan appears to protect myocardium. In a preliminary study, 14 hypertensive patients with acute myocardial infarction (blood pressure greater than 150/90 mm Hg) were treated after infarct size had been predicted by analysis of serum CPK changes during the first 7 h after apparent onset of infarction. Intra-arterial blood pressure, pulmonary artery wedge pressure, cardiac output (determined by dye dilution) and ejection fraction (determined by radioisotope angiocardiography) were measured serially. In 30 consecutive control patients and 10 hypertensive controls with infarction, observed and predicted infarct size correlated closely (r = 0.93). In contrast, hypertensive patients treated with trimethaphan consistently exhibited a reduction of observed compared to predicted infarct size (average difference 31 ± 5 % standard error [SED. Cardiac output, ejection fraction and circumferential fiber shortening velocity increased by average of 12, 37 and 400/0, respectively, although heart rate did not change appreciably. Pulmonary artery wedge pressure declined by an average of 20 0/0. Mortality within 30 days after infarction was less than that in controls with corresponding infarct size. These preliminary results suggest that cautious reduction of ventricular afterload in patients with acute myocardial infarction may decrease the extent of irreversible ischemic injury sustained by the heart by improving the balance between myocardial oxygen supply and demand. Although it is too early to tell whether the apparent favorable modification of infarct size will be reflected by a meaningful reduction of mortality, the early mortality figures in this small sample are encouraging.
Development of improved techniques for assessing infarct size in intact experimental animals and in patients has made it possible to evaluate the effectiveness of selected interventions more objectively. It is becoming increasingly apparent that infarct size is an important determinant of prognosis. Experimentally, infarct size can be modified favorably under specified conditions by interventions designed to
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
Summary
Management of Acute Myocardial Infarction
109
reduce myocardial oxygen requirements or increase myocardial energy supply during the early evolution of infarction. Preliminary clinical applications of these concepts have been encouraging. The time seems to be right for carefully controlled clinical investigations to determine whether techniques designed to protect ischemic myocardium will be effective in improving survival and reducing morbidity from acute myocardial infarction. On the basis of available evidence, it seems clear that appropriate management of patients with this disorder should include careful attention to the impact of any therapeutic modality on the heart itself. It does not suffice to treat patients with p-adrenergic agonists simply because they may elevate systemic arterial blood pressure. Hemodynamics can not be the sole criteria in guiding the physician charged with the responsibility for patient management. Rather, he must consider the potential effects of any contemplated therapeutic modality on myocardial viability as well. The ultimate place of interventions designed to protect ischemic myocardium in the therapeutic armamentarium remains to be determined. However, judicious therapy should be designed to minimize myocardial oxygen requirements and maximize myocardial energy supply as well as to prevent critical impairment of peripheral perfusion.
1 2 3 4 5 6
7 8 9
10
SHELL, W. E. and SOBEL, B. E.: Infarct size index. An effective predictor of prognosis after myocardial infarction. Circulation 48: suppl. IV, p.39 (1973). SOBEL, B. E.; BRESNAHAN, G. F.; SHELL, W. E., and YODER, R. D.: Estimation of infarct size in man and its relation to prognosis. Circulation 46: 640 (1972). PAGE, D. L.; CAULFIELD, J. B.; KASTOR, J. A., and DESANCTIS, R. W.: Myocardial changes associated with cardiogenic shock. New Engl. J. Med. 285: 133 (1971). SHELL, W. E.; LAVELLE, J. F.; COVELL, J. W., and SOBEL, B. E.: Early estimation of myocardial damage in conscious dogs and patients with evolving acute myocardial infarction. J. clin. Invest. 52: 2579 (1973). MAROKO, P. R.; KrnKSHUS, J. K.; SOBEL, B. E.; WATANABE, T.; COVELL, J. W.; Ross, J., jr., and BRAUNWALD, E.: Factors influencing infarct size following experimental coronary artery occlusion. Circulation 43: 67 (1971). MAROKO, P. R.; LmBY, P.; SOBEL, B. E.; BLOOR, C. M.; SYBERS, H. D.; SHELL, W. E.; COVELL, J. W., and BRAUNWALD, E.: Effect of glucose-insulin-potassium infusion on myocardial infarction following experimental coronary artery occlusion. Circulation 45: 1160 (1972). SOBEL, B. E.: Quantification of myocardial ischemic injury; in VOGEL Adv. Cardiol., vol. 15, pp. 86-98 (Karger, Basel 1975). SHELL, W. E. and SOBEL, B. E.: Deleterious effects of increased heart rate on infarct size in the conscious dog. Amer. J. Cardiol. 31: 474 (1973). DELARIA, G. A; JOHANSEN, K. H.; SOBEL, B. E.; SYBERS, H. D., and BERNSTEIN, E. F.: Delayed evolution of myocardial ischemic injury after intra-aortic balloon counterpulsation. Circulation 49 and 50 (Suppl. II): 242 (1974). SHELL, W. E. and SOBEL, B. E.: Changes in infarct size following administration of propranolol in the conscious dog. Amer. J. Cardiol. 31: 157 (1973).
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
References
SOBEL
12
13
14
15 16
17
18
MAROKO, P. R.; LmBY, P.; OINKS, W. R.; BLOOR, C. M.; SHELL, W. E.; SOBEL, B. E., and Ross, J., jr.: Coronary artery reperfusion. I. Early effects on local myocardial function and the extent of myocardial necrosis. J. clin. Invest. 51: 2710 (1972). OINKS, W. R.; SYBERS, H. D.; MAROKO, P. R.; COVELL, J. W.; SOBEL, B. E., and Ross, J., jr.: Coronary artery reperfusion. II. Reduction of myocardial infarct size at one week after the coronary occlusion. J. clin. Invest. 51: 2717 (1972). BRESNAHAN, O. F.; ROBERTS, R.; SHELL, W. E.; Ross, J., jr., and SOBEL, B. E.: Deleterious effects due to hemorrhage after myocardial reperfusion. Amer. J. Cardiol. 33: 82 (1974). SHELL, W. E.; LAVELLE, J. F.; COVELL, J. W., and SOBEL, B. E.: Early estimation of myocardial damage in conscious dogs and patients with evolving acute myocardial infarction. J. clin. Invest. 52: 2579 (1973). HENRY, P. D.; SOBEL, B. E., and BRAUNWALD, E.: Protection of hypoxic guinea pig hearts with glucose and insulin. Amer. J. Physiol. 226: 309 (1974). PELIDES, L. J.; REID, D. W.; THOMAS, M., and SHILLINGFORD, J. P.: Inhibition by beta-blockade of the ST segment elevation after acute myocardial infarction in man. Cardiovasc. Res. 6: 295 (1972). MAROKO, P. R.; BERNSTEIN, E. F.; LIBBY, P.; DELARIA, O. A; COVELL, J. W.; Ross, J., jr., and BRAUNWALD, E.: The effects of intra-aortic balloon counterpulsation on the severity of myocardial ischemic injury following acute coronary occlusion. Counterpulsation and myocardial injury. Circulation 45: 1150 (1972). SHELL, W. E.; EHSANI, A A, and SOBEL, B. E.: Reduction of infarct size in hypertensive patients with acute myocardial infarction. J. clin. Invest. 52: 76a (1973).
Author's address: BURTON E. SOBEL, M.D., Cardiovascular Division, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110 (USA)
Downloaded by: Université René Descartes Paris 5 193.51.85.197 - 11/18/2017 1:10:44 PM
11
110
