EXPERIMENTAL STUDIES

Myocardial Infarction in the Dog: Effects of Intravenous Propranolol

KIM FOX, MB, MRCP ELIZABETH WELMAN, PhD ANDREW SELWYN, MB, MRCP

London, England

The effects of propranolol on myocardial perfusion and metabolism during acute myocardial infarction were studied in 18 mongrel dogs. A reversible snare was placed on the left anterior descending coronary artery; regional myocardial perfusion was continuously measured using the short-lived isotope krypton-8lm, and myocardial metabolism was assessed using the epicardial electrocardiogram and measuremenj of release of creatine kinase activity from the affected segment of myocardium. Six dogs with no arterial occlusion acted as “sham operated” dogs; six others in which the snare was occluded acted as a control group and a third group of six were given propranolol, 0.5 mg/kg, 30 minut~es after coronary occlusion. All variables were recorded before and for 5 hours after coronary occlu- sion. Dogs treated with propranolol showed a significant improvement in regional myocardial perfusion to the affected segment, decreased loss of electrically active myocardium at the end of each experiment for any given degree of early S-T segment elevation and a delay in the local re- lease of creatine kinase activity compared with that in the control dogs. These results suggest that propranolol exerts a beneficial effect on the progress of ischemlc myocardiai damage when given shortly after the onset of infarction.

From the Cardiovascular Unit, Royal Postgraduate Medical School, Hammersmith Hospital, London, England. This study was supported in part by the British Heart Foundation and the Wellcome Trust. London, England. Manuscript received July 31, 1979: revised manuscript received November 19, 1979, accepted November 20, 1979.

Address for reprints: Kim l%x, MB, MRCP, Cardiovascular Unit, Hammersmith Hospital, London W12 OHS, England.

The place of the beta receptor antagonists in the treatment of angina pectoris is now firmly established .Q The effects of these drugs on heart rate, blood pressure and myocardial contractility are well known,3,4 and it is likely that they will have beneficial effects on myocardial perfusion and metabolism during acute regional myocardial ischemia. However, only since the introduction of the short-lived isotopes has it been possible to observe rapid changes that may occur in regional myocardial perfu- sion.5 In this study krypton Urn, a short-lived (half-life 13 seconds), inert and freely diffusing isotope, was used to study alterations in regional myocardial perfusion in the dog for 5 hours after occlusion of the left anterior descending coronary artery. In addition, myocardial metabolism was assessed using the epicardial electrocardiogram and measurement of increases of creatine kinase activity in coronary venous blood. The effects on these variables of intravenous propranolol, given 30 minutes after coronary occlusion, were measured. The possible role of this drug in the protection of the ischemic myocardium is discussed.

Methods Experimental preparation: Eighteen mongrel dogs weighing 42 to 53 kg were

anesthetized with intravenous sodium thiopentone (Pentothala, 12 mg/kg body weight). Respiration was maintained by means of a cuffed endotracheal tube and a mechanical ventilator. Anesthesia was maintained by the intermittent intravenous injection of pentabarbitone sodium (Sagattalm, 2 mg/kg). Care was taken with each administration to ensure that the cornea1 reflex was not abol- ished and that the heart rate and aortic blood pressure changed by less than 5 percent.

A left thoracotomy was performed and the heart supported in a pericardial cradle. A reversible snare was placed around the left anterior descending coronary artery or one of its major branches. A no. 7 French cardiac catheter was intro- duced through a left femoral thoracotomy and positioned in the thoracic aorta; another catheter was introduced through a femoral venotomy into the inferior vena cava. Arterial blood pressure, in millimeters of mercury, was measured

April 1980 The American Journal of CARDIOLOGY Volume 45 769

PROPRANOLOL AND MYOCARDIAL INFARCTION-FOX ET AL.

continuously with a Statham P23 Db transducer and recorded on a multichannel instrument (Hewlett-Packard model 7788A).

Krypton-81m cardiac scintigrams: A modified Paulin catheter was introduced through a right femoral arteriotomy and seated in the aortic sinuses. Krypton-81m was contin- uously eluted in 5 percent dextrose in water from a rubid- ium-81 generator and infused into the aortic sinuses of each dog as a constant infusion at a rate of 5 to 10 ml/min (Wat- son-Marlow MHRE 200).5,6

rejected. Epicardial electrograms recorded during the 20 minute control period and in the remote areas throughout the experiments showed no significant change in S-T segments (more than f2 mV change from the isoelectric), and no sig- nificant change in R wave amplitude (more than f4 percent variation).

Each dog was placed in the left lateral position at the be- ginning of each experiment, and the camera positioned above the thoracotomy site, 4 to 6 cm from the surface of the heart. The left anterior descending coronary artery was occluded for 30 seconds, and the camera was angled so that the regional defect in krypton-81m activity was on the edge of the image of the heart. This procedure helped to ensure that the counts from the posterior aspect of the heart did not interfere with analysis of the regional count rate in the ischemic zone.

Epicardial electrograms were recorded at 5 minute inter- uals before and after occlusion of the left anterior descending coronary artery. Recordings were made at 20 minute intervals for the remainder of each experiment. The effects of respira- tion were taken into consideration by measuring the mean S-T segment elevation and R wave amplitude from five electro- cardiographic complexes with each observation.

Quantitative high spatial resolution images of total and regional myocardial counts per minute of krypton-81m were recorded at 30 second intervals throughout each experiment. The gamma camera was linked to a digital computer (Del- tron-Nova 1220), and these images were recorded on a mag- netic disc. The resolution of the camera and collimator for krypton-81m was 7 mm, using a line spread function test. The digital computer was programmed to recall and display images recorded on an oscilloscopic screen within a 64 by 64 matrix of squares. Corrections were made for the decay of rubid- ium-81. The background activity was estimated by the si- multaneous recording of counts per minute from an area around the image in the field of the gamma camera, that was the same size as the heart image. Counts per minute of kryp- ton-81m were analyzed at the end of each experiment by constructing areas of interest on the visual display unit with an electronic light pen. The computer calculated the activity recorded in each area of interest for each 30 second interval of the experiment. These areas enclosed (1) the aortic root, (2) the area directly affected by the left anterior descending coronary arterial snare, and (3) the rest of the myocardial image (remote area).

Creatine kinase activity: In each dog, Portex tubing (outer diameter 0.63 mm) was introduced into a coronary vein lying parallel to the left anterior descending coronary artery. The tubing was advanced until the end was distal to the cor- onary snare and within a coronary vein draining the area supplied by the snared artery. One milliliter blood samples were drawn every 30 minutes from the aortic and coronary venous catheters simultaneously. The tubing was flushed with heparinized dextrose after sampling, and care was taken to remove the dextrose before each sample was taken. The blood samples were stored in heparinized tubes on ice for up to 3 hours before centrifuging. The plasma was separated by centrifuging at 2,500 g for 10 minutes. Plasma samples were kept overnight at 4OC before assaying for creatine kinase ac- tivity. Creatine kinase activity was determined kinetically at 25°C in a Cecil CE 275 spectrophotometer linked to a Ser- voscribe recorder. The reaction mixture was prepared from Boehringer creatine kinase-activated ultra violet test systems. Samples with activity greater than 250 m.units.ml-l were diluted with heat-deactivated dog plasma to reduce the acti- vation effect of dilution.gJo

Experimental groups: The 18 dogs were separated into three groups:

Epicardial electrograms: The input impedance of the recorder amplifier was 50 ma (Hewlett; Packard 7788A, multichannel), and the frequency response of the whole sys- tem was less than 3 db from 0.05 to 120 hertz. The tracings obtained were reproducible throughout the 5 hour experi- ments. Epicardial electrograms were recorded through a single saline-soaked cotton electrode held over the myocardium to include the area affected by the snared left anterior de- scending coronary artery.6 The calibration was 1 mm-l mV. Two electrode positions were in remote areas and six were positioned within the area supplied by the snared coronary artery.

Group 1 (control group): In six dogs, quantitative kryp- ton-81m cardiac scintigrams, the epicardial electrocardiogram, blood pressure and arterial and coronary venous blood sam- ples were monitored during a 20 minute control period. The left anterior descending coronary artery was occluded and all observations were continued for 5 hours.

Group 2 (propranolol-treated group): In six dogs, the ob- servations were recorded as in group 1 during a 20 minute control period. The left anterior descending coronary artery was occluded, and all observations were continued for 5 hours. Thirty minutes after coronary arterial narrowing, propranolol, 0.5 mg/kg, was given intravenously over 15 minutes.. This dose was selected because it abolished the inotropic and chrono- tropic effects of an intravenous injection of isoprenaline (0.3 pg/kg) and this alteration was thought to represent significant adrenoreceptor blockade.

S-T segment elevation was measured above the T-P seg- ment 20 ms after the end of the QRS complex.7 The amplitude of the R waves was measured in millimeters to the nearest 0.5 mm, using the same isoelectric line. The R wave amplitude at each site was calculated throughout the experiments, and the difference between each R wave at the start of each experi- ment and at 5 hours was calculated (AR). Small Q waves were seen in epicardial leads over the interventricular septum. The development of Q waves was recorded as the difference (in millimeters) 5 hours after severe coronary arterial narrowing (AQ). All electrocardiographic recordings showing a Q to in- trinsic deflection interval greater than 40 msec or a QRS du- ration greater than 65 ms were excluded from the study.7,8

Group 3 (sham operated group): In six dogs, the coronary arteries were not occluded. All observations were recorded as in group 1 for 5 hours. The krypton-81m scintigrams, epicar- dial electrocardiogram and blood pressure showed no abnor- malities and were stable.

Analysis of data: The changes in the regional and total myocardial activity of krypton-81m and plasma creatine ki- naae activity during the experiments were analyzed by analysis of variance. The electrocardiographic signs were analyzed using linear regression and an f test of slopes and a Wilcoxon rank sum teat to compare the slopes.

Results

If the R waves in the remote epicardial positions or the limb Heart rate and blood pressure: Arterial pH and leads showed progressive changes in axis, the recordings were partial pressures of oxygen and carbon dioxide all varied

770 April 1990 The American Journal of CARDIOLOGY Volume 45

by less than 10 percent throughout the experiments. The mean aortic pressure was 92.3 f 6.3 mm Hg (mean f standard deviation) and mean heart rate was 138 f 24.5 beats/min at the beginning of the experiments. These varied by less than f3 percent during the 20 minute control, periods in all three groups and throughout the 5 hour experiments in the control group (group 1) and the sham operated group (group 3). In group 2, the six dogs given propranolol(0.5 mg/kg).30 minutes after occlusion of the left anterior descending coronary artery, the heart rate and systolic blood pres- sure varied by less than 3 percent before occlusion and then decreased from 144 f 23.3 to 96 f 13.2 beats/min and from 93.1 f 9.0 to 85.7 f 7.4 mm Hg, respectively, after the intravenous injection of the drug. These vari- ables returned to control levels after 2.8 f 0.5 hours.

Krypton-8lm myocardial activity: The measured background activity did not exceed 2.5 percent of the total myocardial activity of krypton-81m in any ex- periment. The activity :in the area enclosing the aortic root did not vary by more than f3 percent throughout all the experiments. The initial value of counts per minute of krypton-81m in each region has been treated as 100 percent and changes have been calculated as time-activity graphs.

In the six dogs that received no drugs (group 1, control group), the total and regional activity of kryp- ton-81m varied by less than f5 percent during the 20 minute control period. After occlusion of the left ante- rior descending coronary artery there was a decrease in regional myocardial perfusion in the area of myocar- dium supplied by the occluded vessel. This was seen as

PROPRANOLOL AND MYOCARDIAL INFARCTION-FOX ET AL,

a defect in the images (Fig. 1) and a sudden loss of ac- tivity in the ischemic area of interest (Fig. 2). Both the images and the time-activity graphs showed a sponta- neous improvement over the 5 hour monitoring period (Fig. 1 and 2) (probability [p] <O.Ol).

In the six dogs that received propranolo10.5 mg/kg (group 21, the total and regional activity of krypton-81m also varied by less than f5 percent over the 20 minute control period, and there was a sudden regional decrease in activity seen in both the images and time activity graphs after occlusion of the left anterior descending coronary artery (Fig. 2). After the injection of pro- pranolol there was a 28 percent increase in the myo- cardial activity of krypton-81m over the 5 hour study period in the areas of the heart directly affected by the left anterior descending coronary arterial snare (p <O.Ol) (Fig. 2). The improvement in regional myocar- dial perfusion to the ischemic area was significantly different. from results in the control group (p <O.Ol).

The six dogs in group 3 (sham operated group) showed no significant variation (less than f5 percent) in the total or regional myocardial activity of krypton- 81m throughout these experiments.

Epicardial electrographic changes: During the control periods in all dogs and throughout the experi- ments in the sham operated dogs (group 3) there were no significant S-T segment changes or alterations in R wave amplitude. Epicardial S-T segment elevation was noted within 45 seconds of coronary occlusion in the dogs in groups 1 and 2. This elevation increased to a maximum of 56 f 12.4 mV at 21 f 2.4 minutes, and in both the control and propranolol-treated dogs it de-

FIGURE 1. Krypton scintigrams in a control dog show a typical ex- ample of A, uniform distribution of perfusion during the control period; B. regional ischemia 1 minute after left anterior descending coronary artery narrowing; C and D, the natural history showing spontane- ous diminution in regional ischemia at 45 minutes and- 5 hours iafter coronary artery narrowing, re- spectively. Intravenous propranolol given 30 mintues after coronary artery narrowing produced a greater improvement in regional ischemia than that seen in the control dogs.

April 1980 The American Journal of CARDIOLOGY Volume 45 771

PROPRANOLOL AND MYOCARDIAL INFARCTION-FOX ET AL.

160 r occlusbn

20 I

PropraMlol 05nq kg-'

0 I I I I I I 1 2 3 4 5 5-5

TIME hxr?.l

FIGURE 2. Percent changes in the regional activity of krypton-81m from the area of myocardium directly affected by the coronary arterial snare (lower graphs) and the rest of the myocardium (upper graphs). The stable control period, the onset of ischemia, (occlusion). the results over 5 hours for the control group (dotted lines), and the effects of intrave- nous propranolol (solid lines) are shown. The differences between the propranolol-treated and control groups were significant (p < 0.01). s’mKr = krypton-81m; SD = standard deviation.

creased significantly during the 5 hours that followed (p <O.Ol). However, in the latter group the epicardial S-T segment followed the same natural history as in the control dogs until the intervention, after which it de- creased more rapidly. The analysis of variance showed a p value of <O.Ol at 1 and 5 hours.

An increase in epicardial R wave amplitude was

noted within 45 seconds of narrowing of the left anterior descending coronary artery and reached a peak of 46 f 10.3 mV at 18.7 f 3.6 minutes. All the dogs in the control and propranolol-treated groups showed a progressive loss of R wave amplitude and appearance of Q waves during the 5 hours at those epicardial sites within the area affected by the coronary arterial snare.

In each dog, the S-T segment elevation in millivolts recorded from each epicardial site at 15 minutes was related to the final loss of R waves (ARmV) plus the appearance of Q waves (AQ mV) as assessed with linear

TIME (hwm)

FIGURE 4. Changes in plasma creatine kinase activity in the aorta and coronary vein (within the ischemic area) show that the arteriovenous difference was positive up to 90 minutes in the control group, but thereafter the venous creatine kinase activity increased and exceeded the level in the aorta. In dogs receiving propranoioi, creatine kinase activity in the coronary venous samples did not exceed that in the aorta until 180 minutes (p < 0.01). The mean and 1 standard deviation of the creatine kinase activity are shown.

0 5 IQ

EPICARDtALb? +hQ,m"Iat5HO"RS

FIGURE 3. A median slope and interquartile range for S-T elevation and R and Q wave voltage in the control group (six dogs) and in the pro- pranolol-treated group (six dogs). Wilcoxon rank sum tests showed that in the propranoiol-treated dogs, less myocardium was lost (AR + AQ at 5 hours) than in the control group for any single measure of ischemia (S-T elevation at 15 minutes) (p < 0.01).

regression analysis. All 12 dogs in both the control and propranolol-treated groups showed correlation coeffi- cient (r) values of >0.9 and a p value of <O.Ol. The slopes of the six regression lines in group 1 were signif- icantly different (f test p <O.Ol) and were used to cal- culate a median slope and interquartile range (Fig. 3). The electrocardiographic signs in the six dogs given propranolol were treated in the same way and showed the same r and p values; a median slope and interquar- tile range were calculated (Fig. 3). A Wilcoxon rank sum test showed that the slopes of the two groups were sig- nificantly separated and different. This showed that for any given degree of S-T segment elevation the control dogs showed more QRS changes than the dogs treated with propranolol.

Creatine kinase activity: The plasma activity of creatine kinase in the aortic samples increased throughout the experiments. However, although in the control dogs the plasma activity of creatine kinase in the coronary venous samples exceeded that in the aorta at 90 f 15 minutes, in the dogs receiving propranolol the coronary venous activity did not exceed the aortic level until 180 f 21 minutes (p <O.Ol) (Fig. 4). In the sham operated dogs (group 3) the coronary venous plasma activity of creatine kinase was consistently below that of the aorta throughout the 5 hour experiment.

Discussion In this study we evaluated the effects of propranolol

given shortly after coronary arterial occlusion. In con- trast with previous work, we used the short-lived isotope krypton-Urn, which allows assessment of moment to moment changes in myocardial perfusion. In addition, myocardial metabolism was assessed using the epicar- dial electrocardiogram and measurement of the local release of creatine kinase activity.

Krypton-8lm activity as a measure of myocardial blood flow: Previous work1’J2 in dogs showed that, provided the alterations in heart rate and blood pressure are within the limits of those seen in this study, a con-

772 April 1980 The American Journal of CARDIOLOGY Volume 45

stant infusion of krypton-8lm will result in a stable arterial concentration and mixing of the indicator being delivered to the coronary circulation. Temporary changes of less than 5 percent in heart rate and blood pressure were noted with coronary occlusion. The ab- sence of consistent or significant changes in these he- modynamic variables is ascribed to the fact that only small or at most moderate-sized infarcts were produced. This inert and freely diffusing radionuclide will accu- mulate in the myocardial water space resulting in an equilibrium of activity within the heart. If the aortic sinus infusion is const.ant, this dynamic equilibrium permits the measurement of regional myocardial ac- tivity of krypton-81m to investigate changes in regional myocardial perfusion.6J2 As blood flow increases, the effects of washout of krypton-81m on the relation be- tween regional counts per minute and blood flow be- come more important.

Prior studies,12 however, have shown a significant relation between the myocardial flow measured using an electromagnetic flow probe and that calculated using the calibrated activity of krypton-81m. The gamma camera represents the heart in tw6 dimensions and the images may be affected by heart size and shape. These effects can be partially overcome by observing the region of interest on the edge of the image of myocardial ac- tivity. This practice minimizes interference from krypton-81m counts per minute from the opposite side of the heart. In its present form the technique cannot separate endocardial and epicardial counts per minute of krypton-81m. The measurements obtained are a mean estimate of transmural flow events in the area of interest. Single photon tomography may be of help in the future. The present technique does not depend for its value on absolute measures of flow; rather, the steady state approach gives faithful moment to moment changes in directional alterations in flow. The high count rates localized to the myocardium, the negligible background and the single 190 kev gamma ray provides excellent resolution of events. No attempt was made to interpret counts per minute in the border zone between the ischemic and normal myocardium. The use of krypton-81m provides an advance over existing meth- ods of determining myocardial perfusion. In particular, radioactive microspheres allow only single measure- ments to be made, although previous work13 has shown good agreement between the two techniques.

Effect of propranolol on perfusion of ischemic myocardium: Propranolol given 30 minutes after the occlusion of the left anterior descending coronary artery caused a significant increase in regional myocardial perfusion to the affected segment of myocardium. Theoretically at least, blockade of the beta receptors may be expected to produce systemic and coronary vasoconstriction.14 Experimental animal studies con- firmed that, although this did occur in the normal heart,15J6 in the presence of myocardial ischemia pro- pranolol did not reduce flow17 and indeed caused a significant increase in the ratio of endocardial to epi- cardial flow in both ischemic and nonischemic myo- cardium,18 a factor that will favor the function of the ischemic myocardium. Vatner and colleagues,1gT20 using

PROPRANOLOL AND MYOCARDIAL INFARCTION-FOX ET AL.

short-term studies and evaluating myocardial perfusion in conscious dogs, found that propranolol depressed function slightly but improved flow to the ischemic myocardium. These studies, like our own, used modest doses of propranolol. In our studies the ability to ob- serve changes in myocardial perfusion is demonstrated. It is possible that the improvement in perfusion noted during myocardial ischemia resulted from blocking of excessive sympathetic activity, a physiologic response that may detrimentally affect regional blood flow to the ischemic area by increasing perfusion to the normal areas. Further, it has been postulatedlg that propranolol may improve myocardial perfusion consequent to im- proved collateral flow to the endocardium. This change would not be evident in studies using a flow probe but was detected with use of radioactive microspheres and xenon clearance.17J8

In our study propranolol caused a gradual increase in myocardial perfusion to the ischemic area instead of the rapid increase that one would expect if the alter- ations in flow followed a propranolol-induced decrease in heart rate with prqlongation of diastole. The rise in myocardial perfusion only became manifest as heart rate and blood pressure were beginning to increase. We suggest that propranolol improved myocardial perfusion by reducing heart work and thereby myocardial oxygen consumption in both ischemic and nonischemic areas. Consequently less adenosine triphosphate will be con- sumed and more high energy phosphate stores will be available to preserve the cell. These events will lead to a decrease in capillary endothelial edema and myocar- dial swelling, which normally impair collateral flow. These changes would not be expected to occur suddenly but would have a slow gradual effect on the natural history of the ischemic process. However, Peter et a1.21 were unable to demonstrate an improvement in myo- cardial blood flow using radiolabeled microspheres given before and 24 hours after coronary ligation. Apart from different methodology their studies differed from ours in that relatively enormous doses of intravenous propranolol were given (5 mg/kg). These effects of propranolol may well be dose-related, and at larger doses the effects on myocardial perfusion could be ex- pected to be different.

Effect of propranolol on electrocardiographic signs of ischemia: Considerable objections have been raised to the use of the S-T segment in precordial and epicardial electrocardiograms for evaluating the effe&s of interventions on infarct size.22 Nevertheless, Reimer and his colleagues23 recently found a close relation be- tween the early manifestations of ischemia (S-T seg- ment elevation) and the later signs of cell death (R wave loss and Q wave gain). In their study, dogs given intra- venous propranolol manifested less R wave loss and Q wave gain for any given degree of early S-T segment elevation than did the control dogs. Their qualitative approach, in which each dog was used as its own control, provides supportive evidence for the beneficial effects that intravenous propranolol may theoretically pro- vide.

Effect of propranolol on creatine kinase release: Attempts to compare the total release of creatine kinase

April 1980 The American Journal of CARDIOLOGY Volume 45 773

PROPRANOLOL AND MYOCARDIAL INFARCTION-FOX ET AL.

activity from one dog to another will not provide meaningful results because of the large random varia- tion in the size of the ischemic segment. Changes in the coronary venous activity of creatine kinase will occur postoperatively and for this reason we performed si- multaneous measurement of creatine kinase activity from the aorta and local coronary vein. This procedure permitted study of release of creatine kinase activity in the period after coronary occlusion. Lower levels of creatine kinase activity were found in the dogs treated with propranolol. Certainly it is possible that the lesser values of creatine kinase in the venous blood after ad- ministration of propranolol are simply due to dilution by increased collateral flow; consequently we made no attempt at quantitation on the basis of creatine kinase activity levels. Any measure of the total creatine kinase activity released would require measurement of venous blood flow in the ischemic myocardial segment. How- ever, several studies24,25 using isolated perfused hearts have shown that acceleration or delayed release of cre- atine kinase activity is associated with changes in the time course of myocardial injury. Our study revealed

marked differences in the time for the release of creatine kinase activity that are unlikely to be due simply to a change in coronary collateral venous blood flow. In ad- dition, the delayed release in the propranolol-treated dogs correlated with an improvement of blood flow and electrocardiographic signs.

Clinical implications: In our study open chest anesthetized dogs were studied. The effects of pro- pranolol in the conscious closed chest model may well be different.lg The myocardial depressant effects of the anesthesia26 and the recent surgical procedure could intensify the depressant effects of the beta adrenergic blocking agent. Propranolol in this study and in others using the open chest anesthetized dog27928 was associ- ated with a marked decrease in heart rate; in contrast, workers1g*2g using a closed chest model found more modest reductions in heart rate. Therefore, our results cannot be directly extrapolated to human beings but do indicate that further work is necessary to determine if similar beneficial effects can be achieved in patients and if these same effects will occur when propranolol is given 2,4 or 8 hours after myocardial infarction.

References

1. Gillam PMS, Prfchard BNC. Propranolol in the therapy of angina

2. pectoris. Am J Cardiol 1966; 18:366-g. Elllott WC, Stone JN. Beta adrenergic blocking agents for the treatment of angina pectoris. Prog Cardiovasc Dis 1969; 12: 83-98.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

Epstein SE, RobMson BF, Kahler RL, Braunwald E. Effects of beta adrenergic blockade on cardiac response to maximal and sub- maximal exercise in man. J Clin Invest 1965; 44:1745-53. Parker JO, West RO, DlGlorgl S. Hemodynamic effects of propranolol in coronary heart disease. Am J Cardiol 1968; 21: 1 l-9. Tumer JH, Selwyn AP, Jones T, Evans TR, Raphael MR, Lavender JP. Continuous imaging of regional myocardial blood flow in dogs using krypton-elm. Cardiovasc Res 1976; 10:398-412. Selwyn AP, Welman E, Fox K, Horlock P, Pratt 1, Klein M. The effects of nifedipine on acute experimental myocardial ischemia and infarction in dogs. Circ Res 1979; 44: 16-23. Yuller JF, Maroko PR, Braunwald E. Evaluation of precordial electrocardiographic mapping as a means of assessing changes in myocardial ischemic injury. Circulation 1975; 52:16-23. Samson WE, Scher AM. Mechanism of ST segment alteration during acute myocardial injury. Circ Res 1960; 8:780-e. Hearse DJ, Humphrey SM, Chain E. The effects of reoxygenation on enzyme release from the anoxic isolated perfused rat heart. Biochem Sot Trans 1973; 1:869-71. Oliver IT. A spectrophotometric method for the determination of creatine phosphokinase and myokinase. Biochem J 1965; 61: 116-22. Selwyn AP, Welman E, Pratt TA, Clark J, k&Arthur C, Lavender JP. The interpretation of thallium-201 cardiac scintigrams. Studies in experimental ischemic heart disease in dogs. Circ Res 1978; 43:287-93. Selwyn AP, Jones T, Turner JH, Clark J, Lavender P. Continuous assessment of regional myocardial perfusion in dogs using kryp- ton-elm. Circ Res 1978; 42:771-7. Selwyn AP, Fox K, Shillingford JP. Myocardial imaging with ex- tractable cations and inert tracers: The effects of flow and me- tabolism. Clin Cardiol 1978; 1:60-7. Robln E, Cowan C, Purl P, et al. A comparative study of nitro- glycerine and propranolol. Circulation 1967; 36: 175-86. Nayler WG, Mclnnes I, Swan JB, Carson U, Lowe TE. Effect of propranolol, a beta adrenergic antagonist on blood flow in the coronary arteries and other vascular fields. Am Heart J 1967; 73~207-16.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Davlda WG. Macdonald DC, Mason DFJ. Effects of pronetholol and propranolol on the coronary circulation of the dog. Br J Pharm 1969; 37:338-56. PM B, Craven P. Effect of propranolol on regional myocardiil blood flow in acute ischemia. Cardiovasc Res 1970; 4:176-g. Becker LC, Fortuin NJ, Pitt B. Effect of ischemia and antianginal drugs on the distribution of radioactive microspheres in the canine left ventricle. Circ Res 1971; 38:263-g. Vatner SF, Balg H, Manders WR, Ochs H, Pagani M. Effects of propranolol on regional myocardial function, electrograms and blood flow in conscious dogs with myocardial ischemia. J Clin In- vest 1977; 60:353-60. Vatner SF, Balg H, Mandere WT, Murray PA. Effects of a cardiac glycoside in combination with propranolol on the ischemic heart of conscious dogs. Circulation 1978; 57:568-75. Peter T, Heng MK, Slngh BN, et al. Failure of high doses of pro- pranolol to reduce experimental myocardial ischemic damage. Circulation 1978; 57:534-40. Reimer KA, Rasmussen MM, Jennings RB. Reduction by pro- pranolol of myocardial necrosis following temporary coronary artery occlusion in dogs. Circ Res 1973; 33:353-63. Holland RP, Brooks H: TO-ST segment mapping. Critical review and analysis of current concepts. Am J Cardiol 1977; 40: 1 lo- 29. Hearse DJ, Humphrey SM, Chain E. The effects of reoxygenation on enzyme release from the anoxic isolated perfused rat heart. Biochem Sot Trans 1975; 31442-6. Naylor WG, Grace A, Slade A. A protective effect of verapamil on hypoxic heart muscle. Cardiovasc Res 1976; 10:650-67. Manders WT, Vatner SF. Effects of sodium pentobarbital anes- thesia on left ventricular function and distribution of cardiac output in dogs, with particular reference to themechanism~fortachycardia. Circ Res 1976; 39:512-7. Maroko PR, Kjekshus JK, Sobel BE, et al. Factors influencing infarct size following experimental coronary artery occlusions. Circulation 1971; 43:67-82. Shatney CH, Maccarter DJ, Llllehel RC. Effects of allopurinol, propranolol and methylprednisolone on infarct size in experimental myocardial infarction. Am J Cardiol 1978; 37:572-80. Llang CS, Heed WB. The myocardial depressant effects of beta receptor blocking agents. Comparative study of DL propranolol, D propranolol and practolol in awake dogs with and without acute myocardial infarction. Circ Res 1974; 35:272-80.

774 April 1980 The American Journal of CARDIOLOGY Volume 45