Acute hemodynamic effects of propranolol, glycerylnitrate, and exercise in coronary patients with left ventricular dysfunction

International Journal of Cardiology, 9 (1985) 465415 Elsevier

465

IJC 00322

Acute hemodynamic effects of propranolol, glycerylnitrate, and exercise in coronary patients with left ventricular dysfunction Jargen W. Haerem, Department

Arne Westheim

and Erik FDnstelien

of Medicine, and Institute of Physiology, Vlieviil Hospital, University of Oslo, Oslo I, Norway (Received

4 April 1985; revision accepted

1 July 1985)

Haerem JW, Westheim, A, Fanstelien E. Acute hemodynamic effects of propranolol, glycerylnitrate, and exercise in coronary patients with left ventricular dysfunction. Int J Cardiol 1985; 9:465-475.

Some adverse effects of beta-blockers in heart failure are counteracted by glycerylnitrate. However, the hemodynamics in this condition after giving both drugs are not well known. We examined the drug combination in exercising coronary patients with left ventricular dysfunction. Elevated left ventricular end-diastolic pressure was a measure of dysfunction. A right-heart catheterization with three successive exercise stress tests was done in 40 patients. At repeated exercise without drugs a “warming up” phenomenon was observed, consisting of small but statistically significant reductions in pulmonary capillary wedge pressure, and heart rate. At exercise propranolol reduced heart rate, cardiac output, systemic blood pressure, left ventricular work, and increased arteriovenous oxygen difference. Glycerylnitrate reduced pulmonary capillary wedge pressure at exercise, but, contrary to the findings at rest, it did not increase heart rate or reduce cardiac output. The drug combination resulted in hemodynamics that were similar to those after propranolol alone, except for a lower pulmonary capillary wedge pressure. The drug combination allowed the patients to exercise with the benefits of the beta-blocker, but at a lower ventricular filling pressure. Thus, the potential hazard of

Correspondence

to: J.W. Haerem,

0167-5273/85/$03.30

Horten

Hospital,

3190 Horten,

0 1985 Elsevier Science Publishers

Norway.

B.V. (Biomedical

Division)

466

giving beta-blockers to patients with left ventricular dysfunction may be reduced by adding glycerylnitrate. (Key words: coronary disease; exercise; hemodynamics; 101)

glycerylnitrate;

proprano-

Introduction The synergistic effect on angina pectoris of organic nitrates and beta-adrenergic blockade [1,2] is paralleled by hemodynamic changes, which indicate a reduction of left ventricular work [3,4]. Caution has been advised in using beta-blockers in ischemic left ventricular failure [5]. Nitrates, on the other hand, can reverse asynergy [6,7] and can improve left ventricular motility [8,9]. Hearts with left ventricular failure tend to benefit most from the glycerylnitrate treatment [lo]. By combining the drugs, certain disadvantages of beta-blockade (that is, increase in left ventricular filling pressure [11,12] and in systemic vascular resistance [5,11], and decrease in cardiac output [3,4]) may be counteracted by glycerylnitrate. Glycerylnitrate reduces left ventricular filling pressure [3,11] and systemic vascular resistance [lO,ll], and in some patients with left ventricular failure cardiac output is increased by glycerylnitrate [lo]. The potentially harmful effect of beta-blockade, and the favorable effect of adding glycerylnitrate would appear to be particularly great in patients with left ventricular dysfunction. We have not been able to find published invasive studies on the combined effect of glycerylnitrate and beta-blockade in such patients, diagnosed as having left ventricular dysfunction. Therefore it was appropriate to investigate the acute hemodynamic effects of the drug combination in this category of patients. Methods

Forty patients with coronary disease and stable angina pectoris were included. They were selected after having had a left-heart catheterization with ventriculography and coronary angiography 1 or 2 days previously. Inclusion criteria were a 75% or more proximal narrowing in one or more major coronary vessels and elevated left ventricular end-diastolic pressure of 13 mm Hg or more, as a measure of left ventricular dysfunction. The left ventricular end-diastolic pressures were measured in the absence of angina, with the patients supine and resting after having received 10 mg diazepam orally and 0.5 mg atropine sulphate intravenously. Patients were not included if they had peripheral edema, pulmonary rales, arrhythmias, valvular disease or obstructive lung disease. All had refrained from taking beta-blockers during the previous week, and vasodilators including nitrates had been withheld for at least 12 hr. After having given informed consent, the patients were allotted to one of four test

461

TABLE 1 Design of the study. Group

R, I

(10 min)

, E,

, R,

(7 min)

I

II III IV

(15 min) GN (8’) P P No drugs given

,

E2

1

(7 min) P

R,

R.9

(20 min)

(20 min)

,

E,

,

(7 min)

GN (15’) GN (15’)

R, to R, are symbols for measurements made at rest. E, to Es symbolise measurements at exercise. The duration in minutes of each period of rest or exercise is indicated. R,, R,, R, and R, started after 5, 10, 10, and 17 min of rest, respectively. E,, E,, and E, started after 3 min of exercise. GN = glycerylnitrate administered after so many minutes of rest as indicated. P = propranolol administered at the start of the resting period.

groups (Groups I to IV) by drawing a number from a pool of four blocks, each containing 10 numbers. The study protocol was approved by the local ethical committee. Design of the Study

The patients had three successive exercise tests preceded by resting periods (Table 1). Hemodynamic measurements were carried out during each of the resting periods, to check the stability of the pressure system and the’ reproducibility of the cardiac output measurements. The measurements at rest started 5 min after the catheters had been correctly positioned (R,), after 10 min of rest in the second resting period (R,), and after 10 min (R3) and 17 min (R4) of rest in the third resting period. During the three stress tests (E,, E,, and E3) the measurements started after 3 min of exercise. Glycerylnitrate, 0.5 mg sublingually, was given 8 min after the start of the second resting period in Group I and after 15 min of the third resting period in Groups I and II. Propranolol chloride, 0.15 mg per kg body weight, was administered intravenously in 5 min, starting immediately after the second exercise in Group I, and after the first exercise test in Groups II and III. The “warming up” phenomenon experienced in exercising coronary patients [13,14] may interfere with the evaluation of drug effects. In order to observe the magnitude of the warming up effect in the present study, no drugs were given to the patients in Group IV. Right-Heart Catheterization

Procedure

The patients were fasting, non-sedated and supine. Catheters were introduced percutaneously under local anesthesia, using a modified Seldinger technique. Ante-

468

cubital or femoral veins were used. A 7F Swan Ganz triple-lumen catheter was advanced to the pulmonary artery, and a Cordis Coumand catheter 7F was wedged in the periphery of the right or the left pulmonary artery. A 1.2 mm short teflon cannula was inserted percutaneously in the right or left brachial artery. Pressures were obtained by means of two Nycotron gauge transducers with zero reference point at 10 cm above the mattress where the patients were supine. The following pressures were measured and recorded with a Siemens Elema Mingograph 81: the right atria1 mean pressure (RAP), the pulmonary arterial pressure (PAP), the wedged pulmonary capillary mean pressure (PCWP), and the brachial arterial pressure (BAP). Mean pressures were obtained by electronic damping. Diastolic pulmonary artery pressures were used throughout in patients with unstable wedged pulmonary capillary mean pressure recordings. The diastolic pulmonary arterial pressure was obtained by calculating the mean of 10 successive diastolic readings. One singe lead of the ECG was recorded together with the pressures. Blood oxygen saturation was measured with an American Opticals Corporation reflection oxymeter. Cardiac output (CO) was determined with the thermodilution technique, using an Edwards Laboratories computer, model 9510. Ten ml of iced 5% glucose were used for the injections. The median value of 3 or 4 measurements was calculated. Heart rate was determined from the ECG tracings. All measurements and calculations were done several months after the catherizations without knowing which test group the individual patients belonged to. Exercise

The day before the catherization the patients performed a bicycle ergometer test to determine a light work load that could be carried out for an infinite time at the point when angina started. They exercised in the supine position at 60 cycles per minute on an electrically braked bicycle. The work loads ranged from 10 to 40 W. Angina was graded by the patients on the following scale: 0 = none, 1 = minimal, 2 = mild, and 3 = moderate angina. At “moderate” angina the patients would normally have taken glycerylnitrate but not necessarily stopped their activities. For the individual patient the work load was unchanged throughout the 3 exercise tests. Each of the 3 stress tests lasted for 7 min, and they were carried out after resting periods of 10, 15, and 20 min respectively. Calculations and Statistical Methods

For the calculation of left ventricular work (LVW), systemic vascular resistance (SVR), and arteriovenous oxygen difference (AVO, diff) the following formulae were used: LVW = (systolic BAP-PCWP) X CO X 0.136 mg m/min SVR = mean BAP X 80,420 dyn X set X crne5 AVO, diff = (arterial oxygen saturation-venous oxygen ml/100 ml.

saturation)

X

Hb

X

1.34

469 TABLE

2

Clinical

and laboratory

Group

I II III IV

Number and sex

lF, 9M lF, 9M 2F, 8M 10 M

* Mean values

data for the 40 patients Age (years) *

59.Ort2.2 55.5k2.5 56.Ok2.1 58.6k2.4

Number

in the 4 test groups.

previous myocardial infarction

of patients

digitalis or diuretics

with

Relative heart volume

Number of patients stenosis in

with

(ml) *

one artery

two arteries

three arteries

6 7 4 I

3 5 3 5

505*43 492 f 28 457k36 419 + 23

0 1 1 0

2 1 1 0

8 8 8 10

*ISEM.

The left ventricular ejection fraction (LVEF) had been calculated previously from biplane projections of the ventriculography films. In patients with large left ventricular aneurysms, the left ventricular ejection fraction was not calculated. Abnormal motility of the left ventricular wall was noted as local (region showing hypokinesia, akinesia or dyskinesia) or global (LVEF < 55% with or without local abnormality). Relative heart volume was the ratio between the heart volume, calculated from frontal and side projections on chest X-ray films, and the body surface as obtained from conventional tables. The distance from the X-ray tube to the film was 2 m. Non-parametric two-tailed tests were used in the statistical analysis. The Wilcoxon test for pair differences was applied in most comparisons, but the sign test or the Wilcoxon signed-ranks test were used also. P values of 5% or less were considered to be significant.

Results Four women and 36 men were randomized to the 4 test groups. Tables 2 and 3 show some clinical and laboratory data for the 40 patients. Twenty-eight of the 40 patients had local or global abnormal motility of the left ventricular wall at rest. In the remaining 12 patients the ventricular dysfunction was shown by an elevated left ventricular end-diastolic pressure only. The reproducibility of the hemodynamic measurements at rest is shown in Table 4. Effect of Repeated Exercise

The 10 patients exercised 3 times successively with a rest between each test and without receiving glycerylnitrate or propranolol. Hemodynamics changed from rest to exercise, as can be seen by comparing Table 4 (rest) with Table 5 (exercise). Eight patients experienced angina pectoris during 1, 2, or all of the 3 stress tests. The angina was minimal or mild at all tests except for one test, during which one patient had moderate angina. It was felt to be milder during the second exercise than the first one (P C 0.04; sign test).

!

!60 ! 20 75 75 65 53 36 57

I

LVEF

Group I

18 30 32 24 20 20 18 28 20 20 23&2

LVEDP 245 88 120 163 115 130 135 150 210 135 149+47

SBAP

!

47 ! 48

!

76 44 ! 45 67 ! 54 ! 83

LVEF

Group II LVEDP 20 18 30 17 20 17 14 15 24 16 19+2

105 175 155 140 146 164 122 175 135 165 148 + 23

SBAP 66 72 ! 66 67 ! 39 83 44 ! 59 ! 45 ! 54

LVEF

Group III LVEDP 20 18 24 16 16 15 32 16 15 16 19+2

115 105 208 103 138 153 153 147 149 179 145 + 33

SBAP

51

65 70

1

!

!60

! 45 ! 39

! I

LVEF

Group IV LVEDP 20 26 28 18 26 14 24 14 16 16 20+2

150 140 180 170 110 135 135 145 150 128 144+21

SBAP

LVEF= left ventricular ejection fraction (W); LVEDP = left ventricular end-diastolic pressure (mm Hg); SBAP = systolic brachial artery pressure (initial measurement at rest, mm Hg); ! localizedhypokinesia, akinesia, or dyskinesia of the left ventricular wall. In some hearts with abnormal motility of the left ventricular wall LVEF was not determined.

1 2 3 4 5 6 7 8 9 10 Mean k SEM

Patient

Individual laboratory data for the 40 patients in the 4 test groups.

TABLE 3

471 TABLE 4 Reproducibility of hemodynamic measurements at rest in the 10 untreated patients * Measurement at

R, R2

R, R,

PCWP

co

HR

SBAP

SVR

9+1 9*1 9+1 10&l

6.2kO.2 5.9*0.4 6.1 f 0.3 6.1 f 0.3

71*3 74*3 72*3 73*3

144+6 145+7 148*7 149*8

1333+ 1510* 1485 f 1454*

84 144 145 117

AVO, diff

LVW

3.8kO.2 3.6kO.2 3.8kO.2 3.7 f 0.2

11.3*0.5 10.7kO.6 11.2kO.5 11.5 *0.7

* Mean values f SEM. PCWP = pulmonary capillary wedge pressure (mm Hg); CO = cardiac output (l/min); HR = heart rate (beats/mm); SBAP = systolic bra&al artery pressure (mm Hg); SVR = systemic vascular resistance (dyn X set X cm- 5); AVO, diff = arteriovenous oxygen difference (ml/100 ml); LVW = left ventricular work (mg m/mitt).

The wedged pulmonary capillary mean pressure decreased and a small reduction in heart rate occurred from the first to the third exercise (Table 5). No other significant changes were observed. Effect of Glycerylnitrate

At rest heart rate increased after glycerylnitrate from 72 + 7 to 82 & 10 (beats/ min, mean f SEM, R, versus R,, P < 0.01). The corresponding cardiac output fell from 5.9 k 1.7 to 5.4 f 1.1 (l/min, mean + SEM, P > 0.10, not significant (NS)). At exercise wedged pulmonary capillary mean pressure fell after glycerylnitrate from 20 &-3 to 10 f 2 (mm Hg, mean + SEM, P -c 0.01). This was a greater reduction in wedged pulmonary capillary mean pressure than that obtained by repeated exercise without glycerylnitrate (comparison: the change in individual patients in mm Hg from E, to E, in Group I versus the corresponding change from E, to E, in Group IV; P > 0.05, NS, signed ranks test). Small reductions in heart rate, systolic bra&al arterial pressure, systemic vascular resistance, and arteriovenous oxygen difference were not statistically significant. This was the case for a small increase in left ventricular work as well. The cardiac output remained unchanged. Effect of Propranolol

The wedged pulmonary capillary mean pressure at rest increased after the administration of propranolol in Group III, from 9 + 2 mm Hg (mean f SEM) in

TABLE 5 Effect of repeated exercise in 10 untreated patients. *. Measurement at

PCWP

co

HR

SBAP

SVR

AVO, diff

LVW

E,

22*3 20*3 17*2 **

8.4kO.4 8.7kO.4 8.4kO.4

92rt4 91*4 89*4 **

168k.9 166&8 168*8

1156*109 1083*102 1127* 94

6.2kO.3 6.1f0.5 6.4kO.5

16.6k1.2 17.0* 1.0 17.7kl.O

E2

Es

* Mean values f SEM; abbreviations as in Table 4. ** Et versus Es, P < 0.05.

472

TABLE 6 Effect of propranolol

(P) Group

Measurement

CO

HR

9.1*0.4 1.5 kO.4 **

92k4 J9k3

at

E, (without P) E, (with P) * Mean values

II, E, versus Es) *.

f SEM; abbreviations

SBAP lJ5&9 160f7

**

as in Table 4. **

**

AVO, diff

LVW

7.0*0.6 7.9kO.7

18.1 f 1.6 13.8k4.4 **

P < 0.05.

R, to 12 + 2 mm in R, (P < 0.05), to 13 & 2 mm in R, (R, versus R,, P < 0.05), and to 12 + 2 mm Hg in R, (R, versus R,, P < 0.05). Smaller increases in Groups I and II did not attain statistical significance. At exercise angina occurred less frequently or was milder after propranolol (Group III, E, versus E,, P < 0.05, sign test). It was unchanged at repeated exercise after propranolol (Group III, E, versus E3). The hemodynamic changes during exercise after propranolol were similar in both Groups II and III. Cardiac output, heart rate, systemic brachial arterial pressure, and left ventricular work were reduced, and arteriovenous oxygen difference increased (in Group III; P < 0.05). No significant changes in wedged pulmonary capillary mean pressure or systemic vascular resistance were observed after propranolol. The results for Group II are shown in Table 6. Effect of Glycerylnitrate

and Propranolol Combined

Versus No Treatment. Reductions in wedged pulmonary capillary mean pressure, cardiac output, heart rate, brachial arterial pressure, and left ventricular work were observed after combined treatment with glycerylnitrate and propranolol compared with untreated patients (Table 7). The arteriovenous oxygen difference increased after glycerylnitrate and propranolol in Group I, whereas the systemic vascular resistance was unchanged in both groups. Versus Glycerylnitrate. Patients receiving combined treatment with glycerylnitrate and propranolol had higher wedged pulmonary capillary mean pressures and greater TABLE 7 Effect of combined glycerylnitrate (GN) and propranolol versus E,; Group II, E, versus Es) *. Measurement

at

Group I E, (before treatment) E, (GN and P) Group 11 E, (before treatment) E, (GN and P) * Mean values

HR

(P): comparison

SBAP

PCWP

co

2Ok 3 14k2

8.1 f 0.6 97*4 6.6kO.4 ** J6*3

22 f 3 14*1**

9.1 f 0.4 92&3 175+ J.5kO.3 ** 79*3 ** 152*

f SEM; abbreviations

SVR

AVO, diff

(Group

I, E,

LVW

1289k 130 7.1 kO.4 16.3 f 3.1 12275116 8.3kO.3 ** 12.5~1.7

lJ4+22 ** 154klJ

as in Table 4. **

with no treatment

9 6**

P c 0.05.

1178~121 7.OkO.6 1164* 96 7.7kO.J

18.1&1.6 13.4+0.9

**

473

TABLE

8

Effect of combined glycerylnitrate (Group I, E, versus E,) *. Measurement

at

E, (GN) E, (GN and P) * Mean values

(GN)

and propranolol

(P); comparison

with glycerylnitrate

alone

PCWP

CO

HR

SBAP

SVR

AVO, diff

LVW

10*2 14*2**

8.1rtO.6 6.6kO.4

94*3 76*3**

168kl8 154k17

1152k138 1227k116

6.8kO.3 8.3f0.3**

17.4 f 2.4 12.5k1.7

f SEM; abbreviations

as in Table 4. ** P < 0.05.

arteriovenous oxygen differences than patients receiving glycerylnitrate alone (Table 8). The heart rate was lower with the combined treatment. Reduction in cardiac output, systolic brachial arterial pressure, and left ventricular work and an increase in systemic vascular resistance were not statistically significant. Versus Propranolol. The patients in Group II has lower wedged pulmonary capillary mean pressures when receiving combined treatment with glycerylnitrate and propranolol than propranolol alone (E, 14 + 2 versus E, 22 f 2 mm Hg, mean + SEM, P < 0.01). No differences were observed in cardiac output, heart rate, systolic brachial arterial pressure, left ventricular work, or arteriovenous oxygen difference whether the patients were receiving combined treatment or only propranolol.

Discussion The patients in this study had left ventricular dysfunction without overt heart failure, and the majority of the patients had left ventricular ejection fraction of > 40% (36 of the 40 patients). The study demonstrates that the acute combined administration of glycerylnitrate and propranolol favorably altered the hemodynamics at exercise in these patients. Previous studies have shown a similar effect of the drug combination in coronary patients [3], but patients with impaired left ventricular function have not been investigated invasively, even if they represent a risk group regarding beta-adrenergic blockade [5]. After glycerylnitrate administration the wedged pulmonary capillary mean pressure fell, while other hemodynamic changes were small and statistically insignificant. The reduction in wedged pulmonary capillary mean pressure could partly be an effect of repeated exercise, as shown here, and partly a drug effect, since other authors attribute the unloading effect of glycerylnitrate on the left ventricular filling pressure to the drug itself [3]. Propranolol reduced cardiac output, heart rate, systolic brachial arterial pressure, and left ventricular work and increased arteriovenous oxygen difference. These findings are similar to the findings of other authors [3,5]. The wedged pulmonary capillary mean pressure at rest increased after propranolol. This has been shown by other authors and has been attributed to an increase in left ventricular wall tension [ll]. The increased wall tension would counteract the oxygen-sparing effect of heart

474

rate reduction after propranolol administration [ll]. In our study wedged pulmonary capillary mean pressure at exercise was similar before and after propranolol administration. A possible explanation is that no significant increase in left ventricular wall tension or stiffness occurred at work after propranolol. Apparently, results are conflicting with regard to left ventricular function at exercise after propranolol [5,6]. Some reasons for this can be differences in patient selection and the magnitude of the work load. The combination of glycerylnitrate and propranolol resulted in a lower wedged pulmonary capillary mean pressure, but otherwise the effect of the combined drugs was similar to the effect of propranolol administration alone. This could mean a lower left ventricular wall tension with the combination of the drugs than with propranolol given separately. Generally considered, determinants of myocardial oxygen consumption are heart rate, systemic blood pressure, ventricular wall tension, and the contractile state of the myocardium. Each of the drugs contributed to reducing myocardial oxygen consumption as estimated by changes in the present measures. Glycerylnitrate reduces oxygen consumption by lowering the ventricular filling pressure, the ventricular volume, and the ventricular wall tension [17]. Propranolol reduces myocardial work and hence oxygen consumption by lowering heart rate and systolic blood pressure. The combination of the drugs allowed the patients to exercise with the benefits of the beta-blocker, but at a lower left ventricular filling pressure. Thus, the present study supports the view that the potential hazard of giving beta-blockers to patients with compromized left ventricular function is reduced by combining the beta-blocker with glycerylnitrate.

References 1 Russek HI. Propranolol and isosorbide dinitrate synergism in angina pectoris. Am J Cardiol 1968;21:44-54. 2 Baxter RH, Lennox IM. Increased exercise tolerance with nitrates in beta-blockaded patients with angina. Br Med J 1977;2:550-552. 3 Wiener L, Dwyer EM Jr, Cox JW. Hemodynamic effects of nitroglycerin, propranolol, and their combination in coronary heart disease. Circulation 1969;39:623-632. 4 Steele PP, Maddoux G, Kirch DL, Vogel RA. Effects of propranolol and nitroglycerin on left ventricular performance in patients with coronary arterial disease. Chest 1978;73:19-23. 5 Taylor SH, Silke B. Haemodynamic effects of beta-blockade in ischaemic heart failure. Lancet 1981;2:835-837. 6 Helfant RH, Pine R, Meister SG, Feldman MS, Trout RG, Banka VS. Nitroglycerin to unmask reversible asynergy. Circulation 1974;50:108-113. 7 Sniderman AD, Herscovitch P, Mat-pole D, Fallen EL. Restoration of regional wall motion by nitroglycerin therapy in patients with left ventricular asynergy. Chest 1974;66:545-548. 8 Strauer BE, Scherpe A. Ventricular function and coronary hemodynamics after intravenous nitroglycerin in coronary artery disease. Am Heart J 1978;95:210-219. 9 McEwan MP, Berman ND, March JE, Feiglin DH. McLaughlin PR. Effect of intravenous and intracoronary nitroglycerin on left ventricular wall motion and perfusion in patients with coronary artery disease. Am J Cardiol 1981;47:102-108.

475 10 Gold HK, Leinbach RC, Sanders CA. Use of sublingual nitroglycerin in congestive failure following acute myocardial infarction. Circulation 1972;46:839-845. 11 Robin E, Cowan C, Puri P, et al. A comparative study of nitroglycerin and propranolol. Circulation 1967;36:175-186. 12 Amende I, Simon R, Hood WP Jr, Lichtlen PR. The effects of the beta blocker atenolol and nitroglycerin on left ventricular function and geometry in man. Circulation 1979;60:836-849. 13 Jaffe MD, Quinn NK. Warm-up phenomenon in angina pectoris. Lancet 1980;2:934-936. 14 Thadani U, Lewis JR, Manyari D, et al. Are the clinical and hemodynamic events during exercise stress testing in invasive studies in patients with angina pectoris reproducible? Circulation 1980;61:744-750. 15 Parker JO, West RO, DiGiorgi S. Hemodynamic effects of propranolol in coronary heart disease. Am J Cardiol 1968;21:11-19. 16 Battler A, Ross J Jr, Slutsky, R, Pfisterer M, Ashburn W, Froelicher V. Improvement of exercise-induced left ventricular dysfunction with oral propranolol in patients with coronary heart disease. Am J Cardiol 1979;44:318-324. 17 Ludbrook P, Karliner JS, Kostuk W, O’Rourke RA. Effects of intravenously administered propranolol on wall motion abnormalities. Am J Cardiol 1973:31:712-717.