Effect of propranolol in patients with myocardial infarction and ventricular arrhythmia

JACC Vol 7. No I January 1986 1-8

COOPERATIVE STUDIES

Effect of Propranolol in Patients With Myocardial Infarction and Ventricular Arrhythmia LAWRENCE M. FRIEDMAN, MD, ROBERT P. BYINGTON, PHD, ROBERT 1. CAPONE, MD, FACC, CURT D. FURBERG, MD, SIDNEY GOLDSTEIN, MD, FACC, EDGAR LICHSTEIN, MD, FACC, Writing Group for the BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP*

The Beta-Blocker Heart Attack Trial was a placebo• controlled, randomized, double-blind clinical trial of the long-term administration of propranolol hydrochloride to patients who had had at least one myocardial infarc• tion. Among 3,837 patients followed up for an average of 25 months, 3,290 (85.7%) had 24 hour ambulatory electrocardiograms performed at the baseline exami• nation. Four classifications of arrhythmia were exam• ined. One of these, the presence of complex ventricular arrhythmias (at least 10 ventricular premature beats/h, or at least one pair or run of ventricular premature beats or multiform ventricular premature beats) was the subgroup of major interest. Regardless of the classifi• cation, the presence of arrhythmia identifies a group of patients with a higher risk of total mortality, coronary heart disease mortality, sudden cardiac death and in• stantaneous cardiac death.

Propranolol and other beta-receptor blocking agents have been shown in clinical trials to decrease mortality (1-3) and reduce ventricular premature beat frequency (4-6) in pa• tients who have had myocardial infarction. Prospective ep• idemiologic studies (7-11) found relations between the pres• ence of ventricular premature beats and increased mortality in such patients. One might, therefore, expect propranolol to be especially effective in patients with myocardial in• farction and ventricular premature beats. The Beta-Blocker Heart Attack Trial was a placebo-con• trolled, randomized, double-blind clinical trial of the long• term administration of propranolol in patients who had had at least one documented myocardial infarction. A total of 3,837 patients in 31 centers were assigned to two groups (1,916 patients to propranolol, 180 or 240 mg/day, and 1,921 patients to matching placebo) and followed up for an *A listing of the Writing Group for the Beta-Blocker Heart Attack Tnal Research Group is presented at the end of the text Manuscript received May 7. 1985; revised manuscnpt received July 17. 1985, accepted July 24, 1985. Address for reprints: Lawrence Fnedman, MD, National Heart, Lung. and Blood Institute, Federal Building. Room 216. 7550 Wisconsin Avenue, Bethesda, Maryland 20892. © 1986 by the Amencan College of CardIOlogy

The a priori subgroup hypothesis that sudden death would be preferentially reduced by prppranolol in pa• tients with complex ventricular arrhythmias was not supported. The relative benefit of propranolol in reduc• ing sudden death for this subgroup was 28 versus 16% for the subgroup without ventricular arrhythmia (rel• ative risk of 0.72 versus 0.84, a nonsignificant relative difference of 14%). There were similar findings for two of the three other classifications of arrhythmia and for the other response variables. Although propranolol does not appear to be of special relative benefit in patients with ventricular arrhythmia, the presence of the ar• rhythmia does identify a high-risk group. The mecha• nism by which propranolol reduces mortality is still un• clear, but is probably not solely an antiarrhythmic one. (J Am Coil CardioI1986;7:1-8)

average of 25 months. A previous publication by the Beta• Blocker Heart Attack Trial (1) reported a 26% reduction of total mortality in patients with previous myocardial infarc• tion assigned to propranolol therapy as compared with sim• ilar patients assigned to a placebo. Cause-specific mortality analyses indicated that both sudden and nonsudden coronary heart disease mortality were decreased in the propranolol group. A second paper by the Beta-Blocker Heart Attack Trial research group (12) reported on coronary event inci• dence (coronary heart disease mortality plus nonfatal myo• cardial infarction) and nonfatal events. Coronary event in• cidence was 23% lower in the propranolol-treated group and nonfatal reinfarction was 16% lower. This report considers the effect of propranolol therapy on coronary deaths in patients with ventricular arrhythmias at the time of enrollment, as determined by means of 24 hour ambulatory electrocardiograms. One of the protocol• specified subgroup hypotheses of the Beta-Blocker Heart Attack Trial was that propranolol would be particularly ef• fective in reducing sudden cardiac death (that is, death within 1 hour of onset of symptoms) in patients with complex ventricular premature beats. Other end points (total mor• tality, coronary heart disease mortality and instantaneous 0735-1097/86/$350

FRIEDMAN ET AL. BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP

JACC Vol 7, No 1 January 1986.1-8

cardiac death) and adverse effects, as well as other classi• fications of ventricular arrhythmia, are also examined.

at least one ventricular premature beat during the monitoring period; 2) patients with an average of at least 10 ventricular premature beats/h; and 3) patients with an average of at least 10 ventricular premature beats/h plus at least one pair or run of ventricular premature beats plus multiform ventricular premature beats. Statistics. In making comparisons between the two groups, the risk of having a cardiac event in the propranolol• treated group relative to the placebo-treated group (relative risk) was calculated. A relative risk of less than 1 indicates a lower observed incidence of the specified events in the propranolol-treated group. When many subgroup compar• isons are made, interpretation of statistical significance is difficult because there is an increased probability of ob• serving p values below an arbitrary significance level by chance alone. In addition, the small sample sizes in the subgroups result in a reduced statistical power for detecting true relative differences. Therefore, a test of significance was performed only for the specified end point of sudden coronary heart disease death in the Beta-Blocker Heart At• tack Trial subgroup with complex ventricular premature beats, as defined in the protocol. Sudden death was defined as death from coronary heart disease, occurring less than 1 pour after the onset of symptoms. Instantaneous death, a subset of sudden death, was defined as death occurring abruptly, without symptoms.

2

Methods Study design. Recruitment for study participants began on June 19, 1978 and ended on October 2, 1980. Men and women aged 30 to 69 years who were hospitalized with symptoms and electrocardiographic and enzymatic changes compatible with an acute myocardial infarction were can• didates for enrollment in the trial. Reasons for exclusion from the study included medical contraindications to pro• pranolol, life-threatening illness other than coronary heart disease and the use of beta-receptor blocking drugs by the patient. Informed consent was obtained from all patients in the study. They were enrolled 5 to 21 days after hospital admission, while they were still hospitalized. Follow-up visits were scheduled at 1 month, 6 weeks and 3 months after randomization, and at subsequent 3 month intervals. A detailed description of the Beta-Blocker Heart Attack Trial design and methods has been reported elsewhere (13). Holter monitoring. Ambulatory electrocardiographic monitoring was encouraged at baseline, but was not man• datory for patient enrollment. At the 6 week visit, a second 24 hour ambulatory electrocardiogram was performed on 840 randomly selected patients representing approximately 25% of the number performed at bas!:!line. The 24 hour, two channel, reel to reel Holter magnetic tapes were generated on Avionics model No. 445 recorders and processed by the CardioData system which is digital, software-based (the Worcester Polytechnic Institute algo• rithm) (14) and capable of differentiating up to 25 different electrocardiographic variables involving rate and rhythm change. As electrocardiogram complexes are noted by the computer, an operator verifies the decision, thus providing human interaction with the computer. The CardioData sys• tem analysis method has been validated (14) both for ac• curacy and precision for ventricular premature beat detec• tion. A quality control program involved the blinded analysis of several nand-counted tapes and reanalysis of selected Beta-Blocker Heart Attack Trial tapes. Arrhythmia definitions. The patient subgroup of major interest, as specified in the protocol, had complex ventric• ular premature beats defined as the presence of one or more of the following features on the 24 hour electrocardiogram recordings: multiform ventricular premature beats, an av• erage of at least 10 ventricular premature beats/h on the 24 hour recording or at least one pair or run of ventricular premature beats. Multiform ventricular premature beats were defined as beats with varying configuration, in either chan• nel, noted visually by the scanning technician and confirmed by the supervising cardiologist. Because ventricular ectopic activity is categorized by differing criteria in other reports, three other patient subgroups are also cqnsidered in this report. These are: 1) patients with

Results Baseline findings. Ambulatory electrocardiograms were performed at the baseline examination in 86.1 % (1,650 of 1916) of those randomized to propranolol and 85.4% (1,640 of 1,921) of those randomized to placebo. Examination of the baseline characteristics of those with and without am• bulatory electrocardiogram showed excellent agreement in most respects. Slightly more patients with ambulatory elec• trocardiograms had a history of angina (36.8 to 32.4%) and ST elevation on the electrocardiogram at rest (13.8 to 10.6%), whereas fewer patients with ambulatory electrocardiograms had T wave abnormalities on the electrocardiogram at rest (64.5 to 71.8%) and were using an antiarrhythmic drug in the hospital before randomization (45.2 to 50.5%). Table 1 shows baseline variables for the patients with the four classifications of ventricular arrhythmia, both treat• ment groups combined. In general, variables associated with increasing risk of mortality, such as age, history of prior myocardial infarction, presence of congestive heart failure, cardiomegaly and use of antiarrhythmic agents, diuretic agents and digitalis, were more common in patients meeting more stringent classifications of arrhythmia. For all four arrhyth• mia classifications, there was excellent baseline compara• bility between the propranolol and placebo-treated groups. Mortality. Total mortality and cause-specific mortality, by treatment assignment, are shown in Table 2 for the four classifications of ventricular arrhythmia and for the patients

FRIEDMAN ET AL BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP

JACC Vol 7. No January 1986 1--8

3

Table 1. Baseline Variables for Patients With Ventricular Arrhythmia VPBs 2" 10/h or Runs or Multlform* (n = 1.338)

VPBs 2" 10/h (n = 435)

VPBs 2" IO/h Plus Runs Plus Multifonn (n = 234)

840 55.3 112.0 72.4 759 579

835 565 112.9 72.7 76.0 54.1I

81.1 57.4 1134 73.1 76.4 52.0

83.3 58.6 114.5 74.0 769 52 I

14.4 374 9.8 74

194 398 12.0 95

27.1 42.3 14.5 11.7

37.2 46.6 15.8 137

In-hospital events (%) CHF Ventncular tachycardia Use of antiarrhythmic drug

15.1 236 46.2

17.3 25.8 48.6

20.9 22.5 53.8

21.4 244 534

Drugs bemg used at time of randomization (%) Antiarrhythmic agents Diuretic drugs Digitalis

17.9 17.9 139

20.6 200 16.1

306 24.4 20.5

31 2 286 25.6

LocatIOn of BHAT MI (%) Anterior Anterior/mfenor Infenor Nontransmural Non-BHAT MI

27.1 100 31.7 227 85

273 12.3 30.8 21 7 79

28.5 14.5 278 21.1 8.0

30.8 15.4 252 20.1 8.5

Abnonnalitles at rest ECG (%) Q-QS waves ST depre5slon ST elevation T wave abnonnality Ventncular conduction defect

674 280 13.5 65.8 93

66.9 3\.1 12.9 68.9 10.3

656 306 11.8 664 11.5

64.9 35.1 132 70.2 12.3

Cardiomegaly (%)

36.0

38.3

40.2

46.2

93

96

9.7

9.5

VPB~

Vanable Male (%) Mean age (yr) Mean SBP (mm Hg) Mean DBP (mm Hg) Mean HR (beats/mm) Current smoker (%) Medical hl~tory (0/() Prior MI Angma pectons CHF Takmg beta-blocker before MI

Median time from MI to enrollment (days)

(n

> 0

= 2.750)

*Protocol definition. BHAT = Beta-Blocker Heart Attack Trial; CHF = congestive heart failure; DBP = diastolic blood pressure; ECG = electrocardiogram; HR = heart rate; MI = myocardial infarctIOn; Non-BHAT MI = c1imcal myocardial mfarction not meeting Beta-Blocker Heart Attack Trial definitIOn; SBP = systolic blood pressure; VPB = ventricular premature beat

without arrhythmia. In general, the placebo-treated patients having more stringent classifications of arrhythmia had higher event rates. Total mortality for patients with increasing grades of arrhythmias went from 10.9 to 15.2 to 19.9 to 19.6%, respectively. Sudden death, which accounts for about half of total mortality, increased from 5. 1 to 7.7 to 9.0 to 10.7 % , respectively. The other outcomes showed similar trends. Regardless of the classification of arrhythmia, total and cause-specific mortality were higher among those patients

with arrhythmia than among those without. Thus, for the patients with at least one ventricular premature beat, total mortality was 7.9 and 10.9% in the propranolol- and pla• cebo-treated groups, respectively. Among patients with no recorded ventricular premature beats, the mortality rate of the propranolol-treated group was 3.2% versus 5.0% for the placebo-treated group. Comparisons for other end point!> and classifications of arrhythmia show the same relations. In all except one case, the frequencies of events in the

JACC VoL 7, No 1 January 1986.1-8

FRIEDMAN ET AL. BETA-BLOCKER HEART ATIACK TRIAL RESEARCH GROUP

4

Table 2. Outcome by Classification of Arrhythmia According to Treatment Group Classification of Arrhythmia VPB > 0 Present Propranolol Placebo Absent Propranolol Placebo VPB

2:

2:

2:

(%)

(%)

Relative Risk

1,370 1,380

7.9J 10.9

0.73

6.8J 9.6

0.71

280 260

3.2J 5.0

0.64

2.9J 3.8

0.74

IO.4J

0.69

9.1 13.5

0.75

4.1 5.3

12.9J 19.9

0.65

10.7 17.1...1

6.2J

0.73

Sudden Death (%)

4.0J 5.1

Relative Risk

0.79

Instantaneous Death (%)

2.6J 3.7

0.46

Relative Risk

0.71

0.70

672 666

15.2

978 974

J

0.67

5.5J 7.7

0.72

J

0.77

2.IJ

0.84

0.63

7.6J 9.0

0.84

5.8J 7.6

0.77

0.73

2.9J 4.0

0.72

1.8J 2.7

0.67

0.87

1O.7J 10.7

0.99

9.8J 8.9

1.10

2.9J

0.70

2.6

3.6J 5.7

0.63

0.88

IO/h

Present Propranolol Placebo Absent Propranolol Placebo VPB

n

CHD Mortality

Relative Risk

IO/h or runs or multiform VPB*

Present Propranolol Placebo Absent Propranolol Placebo VPB

Total Mortality

224

211 1,426 1,429

8.5

IO/h plus runs plus multiform VPB

Present Propranolol Placebo Absent Propranolol Placebo

122 112

18.9J 19.6

0.96

1,528

6.2J 9.2

0.67

1,528

*Protocol definition. CHD

=

coronary heart disease; VPB

15.6J 17.9

0.67

=

4.2

0.60

ventricular premature beat.

propranolol-treated group were less than those in the com• parable placebo-treated group (relative risk < 1). Table 2 also compares the relative risks (propranolol versus placebo) of the various outcomes in the four defined subgroups with the relative risks in the complementary groups. In many of the cases, the relative risks in the patients with arrhythmias were similar to those in the patients without arrhythmias. The major exception is the group meeting the most stringent classification of arrhythmia. Although there were only a few patients with this classification, the relative risks were uniformly greater than those in the patients with• out this arrhythmia. The a priori subgroup hypothesis that sudden death would be preferentially reduced in patients with complex ventric• ular arrhythmias was not supported. The relative risk of sudden death for this subgroup was 0.72 versus 0.84 for the subgroup without ventricular arrhythmia, a 14% relative difference. The value of the Mantel-Haenszel chi-square

analysis of homogeneity was 0.20 (l df), indicating that there is no statistical evidence of a difference between these results (p > 0.05). Using the other classifications of arrhythmia, no preferential effect of propranolol on sudden death was observed. Concomitant therapy and compliance. The percent• ages of patients withdrawn from the study medication be• cause of adverse effects or with complaints at any time during the study are similar among the different classifi• cations of arrhythmia. There is no evidence that those pa• tients with arrhythmia who were receiving propranolol had a greater frequency of complaints or adverse effects than patients without arrhythmia. The propranolol-placebo com• parisons are similar to those for the study group as a whole (1).

Table 3 shows the frequency of use of nonstudy drugs during the trial. Antiarrhythmic drugs were prescribed more often for the placebo-treated than for the propranolol-treated

JACC Vol 7, No. I January 1986.1-8

5

FRIEDMAN ET AL. BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP

Table 3. Percent of Patients Taking Nonstudy Medication at any Time During the Trial for the Four Classifications of Arrhythmia VPBs 2: IO/h or Runs or Multiform*

VPBs > 0

Medication

Propranolol (n = 1,370)

Diuretic drugs DIgitalis AntiarrhythmIc agents Nonstudy beta-blockers

41 7 28.5 21.0 17.5

*Protocol definition. VPBs

=

(n

Placebo = 1,380) 44.6 28.2 27.3 22.8

VPBs

2:

IO/h

VPBs 2: IO/h Plus Runs Plus Multiform

Propranolol (n = 672)

Placebo (n = 666)

Propranolol (n = 224)

Placebo (n = 211)

Propranolol (n = 122)

Placebo (n = 112)

45.7 33.5 25.0 17.0

48.7 32.1 33.0 22.1

52.7 41 I 38.4 15.2

50.7 36.5 48.3 20.4

62.3 48.4 45.1 14.8

54.5 38.4 52.7 14.3

ventricular premature beats.

patients for all four classifications. Except for the patients meeting the most stringent classification of arrhythmia, non• study beta-receptor blockers were also used more frequently in the placebo-treated group. There was also a trend toward increasing use of diuretic drugs, digitalis and antiarrhythmic agents with the more stringent classifications. Patient compliance to medication at the last completed study visit is shown in Table 4. Compliance in the patients receiving propranolol decreased as the severity of arrhyth• mia increased. This is primarily because of an increase in the number of patients not taking any study medication and a decrease in the number receiving the full protocol dose.

Discussion Propranolol and ventricular arrhythmia. Previous re• ports (15-17) of trials of beta-blockers claimed a reduction in sudden cardiac death, though the definitions of sudden death in those studies included mortality within 24 hours of the onset of symptoms. In addition, a decrease in ambient ectopic rhythm after administration of beta-blockers had been reported (18,19). Ventricular arrhythmia had been shown (8,10,20) to be an independent risk factor for sudden death and total mortality. Therefore, at the initiation of the Beta• Blocker Heart Attack Trial, it was postulated that much of any observed benefit would occur in those patients who had complex ventricular arrhythmias. Analyses of data from the patients participating in the

Beta-Blocker Heart Attack Trial who had both baseline and 6 week ambulatory electrocardiograms have confirmed the antiarrhythmic properties of propranolol (4). Among pa• tients with ventricular arrhythmia (defined as ventricular premature beats 2: 10th and pairs or runs) at baseline, the prevalence of arrhythmia at 6 weeks was 53% in the pro• pranolol-treated group as compared with 73% in the pla• cebo-treated group. Among patients with no baseline ar• rhythmia, the prevalence of arrhythmia at 6 weeks was 12% in the propranolol-treated group and 21 % in the placebo• treated group. In the latter group, the overall prevalence of ventricular premature beats was higher at 6 weeks than at baseline. This increase was blunted in the propranolol-treated group (4). Mortality and effect of propranolol. The data pre• sented here do not support the hypothesis that reduction in mortality due to propranolol is primarily restricted to those patients with ventricular arrhythmia at baseline. When the protocol-specified classification of arrhythmia is examined, the relative benefit of propranolol is similar, regardless of whether the patient had the arrhythmia. This holds for all• cause mortality, as well as coronary heart disease mortality and sudden cardiac death. These results are similar for two of the other three arrhythmia classifications (ventricular pre• mature beats> and 2: lO/h). The presence of complex ventricular premature beats did identify a group of patients at approximately twice the risk of death as those without complex ventricular premature beats, though similar relative benefit from propranolol was

°

Table 4. Percent of Patients Taking Study Medication at the Last Completed Study Visit for the Four Subgroups of Patients VPBs

VPBs 2: IO/h or Runs or Multiform*

>0

VPBs

2:

IO/h

VPBs 2: 10/h Plus Runs Plus Multiform

Compliance Measure

Propranolol

Placebo

Propranolol

Placebo

Propranolol

Placebo

Propranolol

Placebo

No study medication Any study medication Full dose On nonstudy beta-blocker Any beta-blocker

25.7 74.3 61.0 8.7 83.0

244 75.6 694 12.0 120

262 738 60.9 8.6 82.4

26.7 733 66.6 132 13.2

31 8 68.2 52.3 7.3 75.5

25.7 74.3 66.4 II 8 11.8

34.2 65.8 49.1 5.8 716

25.0 75.0 65.7 6.5 6.5

*Protocol definition. All abbreviation, as in Table 2

6

rRIEDMAN ET AL BETA-BLOCKER HEAR [ AfT ACK [RIAL

RE~EARCH

JACC Vol 7. No I

GRUUP

observed. Therefore, for the ~ame number of patients treated, the high-risk group will have a greater absolute benelit from the drug (that is, more lives saved per number treated). Interestingly, little or no benefit from propranolol is found in those patients with the classification of arrhythmia re• quiring the most stringent criteria (ventricular premature beats ~ I plus pairs or runs plus multiform ventricular premature beats). This contrasts with the benefit noted in those without this arrhythmia. This might be a chance ob• servation, enhanced by the relatively small number of pa• tients and events in the arrhythmia subgroup. Another ex• planation is that the requirement for having frequent ventricular premature beats and pairs or run~ and multiform ventricular premature beats identifies patients not only at high risk of mortality, but also at high risk of being harmed by taking beta-blockers. Beta-blockers, as well as other antiar• rhythmic receptors, have at times been shown to be proar• rhythmic (21). Thus, any benefit from the drug might be offset by toxicity. The use of diuretic drugs, digitali~ and antiarrhythmic agent~ was highest in this patient subgroup. These drugs may have played a role in minimizing the potential benefit from propranolol. It is also true that com• pliance to study medication was not as good in this subgroup of patients, possibly impairing the ability to detect benefit. if any, from propranolol.

°

Postulated mechanisms of propranolol's beneficial ef• fect. The observation that patients without arrhythmia ben• efited relatively as much from propranolol as those with arrhythmia, regardless of the c1a~sification. implies that there are mechanisms of action other than the reduction in ven• tricular ectopic rhythm. Propranolol has been shown (22.23) to increase the threshold for ventricular fibrillation. Thl~ effect may be distinct from its observed effect on ventricular premature beats. There appears to be an association, how• ever. between the presence of ventricular premature beat~ and the risk of fibrillation and sudden death (10.20,24,25). although it is uncertain whether the association i~ greater than with nonsudden cardiac death (9, II ,26). If the antl• fibriIlatory action were primary. one might have expected more benefit in patients with arrhythmia. The benefit from propranolol might relate to an anti-ischemic effect, either directly (27) or through suppression of ventricular arrhyth• mia (28). Propranolol may also operate by blocking beta2receptors, and thus, arrhythmia mediated through cate• cholamine-induced hypokalemia (29). Whether the negative inotropic effect of propranolol or its bradycardic and hy• potensive actions affected mortality is unknown. As previously reported (30). using baseline clinical cri• teria (exclusive of ambulatory electrocardiogram) ascer• tained in the acute phase of the myocardial infarction, the patients participating in the Beta-Blocker Heart Attack Trial were retrospectively divided into four subgroups: those with electrical problems, those with mechanical or pump prob-

Janual) 1986

J-

8

with both and those with neither. The three with electncal or mechanical disturbance~, or both. appeared to benefit from propranolol therapy, showing reduced mortality. Limited benefit from thc drug was found in the subgroup without such problems. Thus, patients who experienced electrical disturbances (ventricular fibrillation, ventricular tachycardia. atrioventricular [A Vj block or atrial fibnllation) before enrollment in the Beta-Blocker Heart Attack Trial ~eemed to benefit more from propranolol than tho~e without electrical disturbances. This differs from the finding that patient~ with ventricular arrhythmias on an am• bulatory electrocardiogram recorded an average of 2 weeks after hospital admission did no better while taking propran• olol. relative to placebo. than patients without such arrhyth• mia~. The reason for this di~crepancy may be that the clinical as~eSf>ment relied on finding~ early in the course of the infarction. while the ambulatory electrocardiogram was per• tormed in the convalescent phase. It has been reported that arrhythmia~. particularly ventricular fibrillation, occurring in the first few days of a myocardial infarction have little or no independent association with subsequent arrhythmias or mortality 01-35). although thi~ finding has been ques• tioned (25.36). In addition. the difference in the Beta-Blocker Heart Attack Trial rei>ults may reHect the fact that the am• bulatory electrocardiogram record!> episodes in one 24 hour period, wherea~ clinical findings occur over several days. Summary. The hypothe5is that propranolol would be of speCial benefit in patients with prior myocardial infarction who have ventncular anhythmia was not supported. The presence of arrhythmia on an ambulatory electrocardiogram, however, can Identify high-risk patienb. Because the per• cent mortality i~ higher in ;,uch patlcnts. the !>imilar relative benefit lead~ to a greater absolute benefit in those receiving propranolol. With more lives ~aved per number treated. The mechanism of action by which propranolol reduces mortality i~ probably not solely an antiarrhythmic one. lem~, tho~e

~ubgroupi>

Appendix ~--~-~-

-

~-

-~-~---

Beta·Blocker Heart Attack Trial Investigators Clinical Centers Baylor College of Medicine, Houston, TX. Craig M Pratt. MD. Richard R Miller. MD. James B Young. MD. Charlotte Paytun-Ross. Boston University School of Medicine, Boston, MA. Pantel S. Vo• kona,. MD. Bruce Abramo\\ Itl. MD. Lorrallle M Kilcoyne. RN. Mary T Colhns. RN. Brown Unh'ersity Affiliated Hospitals, Providence, RI. Robert J. Capone. MD. Richard S Shulman. MD. Abdul Khan. MD. Eh,abeth Curwen. RN Emor} University, Atlanta, GA. Rubert C. Schlant, MD. Damel Arcn<;berg. MD. Shirley Muw,. MT. Velma Jeffnc'i. RN. Mane Thunnond Elanston Hospital. El·anston. IL Gar) N. Wilner. MD. Cary E Berko\\ ItZ. MD. WIIII.tm B Barnhart. MD. Deborah Hunter-Goldblatt. Geisinger Medical Center. Danville, PA. Charb A Laubach. MD. John H Chapman MD. ClIld) Wllher, Mananne Kac)on

JACC Vol. 7, No.1 January 1986:1-8

Greater Baltimore Medical Center, Baltimore, MD. Thaddeus E. Prout, MD, James H. Biddison, MD, Luis F. Gonzales, MD, James C. RiceJy, MD, Roberta H. Franks. Henry Ford Hospital, Detroit, MI. Gerald M. Breneman, MD, Re• migio Garcia, MD, Tennyson Lee, MD, Sharon Baumann, Susan Housholder. Kaiser Foundation Hospital, Portland, OR. John A. Grover, MD, Barbara Brokop, RN, Merwyn Greenlick, PhD, John B. Wild, MD. Lankenau Hospital, Philadelphia, PA. William L. Holmes, PhD, Leonard S. Dreifus, MD, Dolores DodeJin, Ginny Murphy. Long Island Jewish-Hillside Medical Center, New Hyde Park, NY. Kul D. Chadda, MD, Robert H. Kramer, MD, Robert l. Hamby, MD, Janet Mascaro, RN, Jane Smith, RN. Maimonides Medical Center, Brooklyn, NY. Edgar Lichstein, MD, Gerald Hollander, MD, Amy Harrison, Nancy Zimet, RN. Medical College of South Carolina, Charleston, SC. Peter C. Gazes, MD, William H. BamwelllI, MD, Elizabeth Smith, Lola Engler. Me4ical College of Virginia, Richmond, VA. David W. Richardson, MD, Donald Romhilt. MD, A. Jarrell Raper, MD, Zubair-UI-Hassan, MD, William K. Smith, MD. Miami Heart Institute, Miami Beach, FL. Frank Canosa, MD, Jeff Raines, PhD, Thomas Hashway, Jr, MD, Garcia Garrison, RN. Montreal Heart Institute, Montreal, Quebec, Canada. Pierre A. Theroux, MD, DaVId D. Waters, MD, Doris Morissette, Marilyn Hache, RN, Marie-Andree SeqUIn. Mount Sinai Hospital, Minneapolis, MN. Phillip J. Ranheim, MD, Marvin Segal, MD, Reuben Berman, MD, Julie Levin. Northwestern University Medical School, Chicago, IL. Olga M. Haring, MD, Richard Davison, ¥D, Gustave Bermudez, MD, David Berk• son, MD, Linda Anderson. Overiook Hospital, Summit, NJ. John J. Gregory, MD, Donald Brock, MD, Joel Cannilla, MD, Martin J. Sheehy, MD, Carol Bowerbank, RN. Pacific Health Research Institute, Honolulu, HI. J. Judson Mc• Namara, MD, John J. Cogan, MD, Samuel C. Gresham, MD, Lei Honda, RN. Providence Medical Center, Portland, OR. Gordon L. Maurice, MD, FrankO. McBarron, MD, JamesH. Mackay, MD, Judy Cunningham. Rush-Presbyterian-Saint Luke's Medicial Center, Chicago, IL. James A. Schoenberger, MD, Gilberto S. Neri, Jr, MD, Tracy Remijas, Arline Wilson. Rutgers Medical School, New Brunswick, NJ. Peter T. Kuo, MD, Paul Jennings, MD, Norman Reitman, MD, John B. Kostis, MD, Evelyn Urdanoff. Salt Lake Clinic Research Foundation, Salt Lake City, UT. Allan H. Barker, MD, Kenneth A. Crockett, MD, Michael J. Preece, MD, Ernest Wilkinson, MD, Ellen Doran, Julie Orison. Ogden Research Foundation, Ogden, UT. C. Basil Williams, MD, G. Thomas Blanch, MD, Barbara Fowler. State University of New York at Buffalo, Buffalo, NY. Robert M. Kohn, MD, Phihp D. Morey, MD, Douglas L. Roberts, MD, Dennis Dubois, MD, Mary Bonora. University of California, Davis, CA. Nemat O. Borhani, MD, Thomas Evans, MD, Allen Frankel, MD, Jan Grogan, RN. University of California, San Francisco, CA. Mary Anne Wamow• icz. DO. Robert W. Peters. MD, Randolph Byrd. MD, Carolyn Some• lofski, RN, MS. University of Rochester School of Medicine, Rochester, NY. Paul N. Yu. MD, Theodore Biddle, MD, William Henion, MD, Alvani Santos. MD. Charlene Roth, RN. Julie Olijnyk, RN. University of Southem California, Los Angeles, CA. L. Julian Hay• wood, MD, Maria De Guzman, MD, Leora Jackson, RN. Veterans Administration Hospital, Little Rock, AK. Marvin L. Mur• phy, MD. Neil De Soyza, MD. Linda Treat, RN. Veterans Administration Hospital, West Roxbury, MA. Kevin M. McIntyre, MD, Alfred F. Parisi, MD, William E. Strauss, MD, G.V.R.K. Sharma, MD, Patricia WO()ds, RN. Coordinating Center University of Texas, Houston, TX. C. Morton Hawkins, SeD, J. David Curb. MD, MPH, Robert Hardy. PhD. Rose Lee Bell, PhD, Robert P. Byington, PhD. Darwin R. Laberthe, MD. PhD. Ronald B. Harrist, PhD, Louis Kern, MS, Emmit Hubble, MS.

FRIEDMAN ET AL. BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP

7

Drug Distribution Center Public Health Service, Perry Point, MD. James Grigdesby, Sydney H. Hamet, Patricia Murphy. Electrocardiogram Reading Center UniverSity of Minnesotl\, Minneapolis, MN. Richard S. Crow, MD, Ronald Prineas, MD, Marsha McDonald. 24 Hour Ambulatory Electrocardiogram Center Anthropometrics Heart Clinic, Haddonfield, NJ. Joel Morganroth, MD, Robert Loring. Central Laboratory Bio-Science Laboratories, Van Nuys, CA. Frank Ibbott, PhD. National Heart, Lung, and Blood' Institute, Bethesda, MD. Curt D. Furberg, MD (project officer), Lawrence M. Friedman, MD, David L. DeMets, PhD, William T. Friedewald, MD, C. Eugene Harris. Steering Committee Sidn\:y Goldstein, MD (chairman). Policy and Data Monitoring Board john Naughton, MD (chairman), Paul L. Canner, PhD, Lawrence J. Cohen, MD, Max Halperin, PhD, Lois Pratt, PhD, Burton Sobel, MD, Harpld Strauss, MD, William T. Friedewald, MD (ex officio).

References I. Beta-Blocker Heart Attack Trial Research Group. A randomized trial of propranolol in patients with acute myocardial infarction. l. Mortality results. JAMA 1982;247:1707-14. 2. Norwegian Multicenter Study Group. Timolol-induced reduction in mortality and reinfarction in patients surviving acute myocardial in• farction. N Engl J Med 1981;304:801-7. 3. Hjalmarson A, ElfIIfeldt 0, Herlit~ 1. et al. Effect on mortality of metoprolol in acute myocardial infarction: a double-blind randomized trial. Lancet 1981;2:823-7 4. Lichstein E, Morganroth J, Harrist R, Hubble E. Effect of propranolol on ventricular arrhythmia: the Beta-Blocker Heart Attack Trial ex• perience. Circulation 1983;67(suppl 1):1-5-10. 5. Olsson G, Rehnqvist N. Ventricular arrhythmias during the first year after acute myocardial infarction: influence of long-term treatment with metoprolol. Circulation 1984;69: 1129-34. 6. Ryden L. Ariniego R, Amman K. et al. A double-blind trial of me• toprolol in acute myocardial infarction: effects on ventricular tachy• arrhythmias. N Engl J Med 1983;308:614-8. 7. Bigger JT Jr, FJeiss JL, Kleiger R, Miller JP, Rolnitzky LM, Mul• ticenter Post-Infarction Research Group. The relationships among ven• tricular arrhythmias, left ventricular dysfunction, and mortality in the 2 years after myocardial infarction. Circulation 1984;69:250-8. 8. Ruberman W, Weinblatt E, Goldberg 10. Frank CW, Shapiro S. Ventricular premature beats and mortality after myocardial infarction. N Engl J Med 1977;297:750-7. 9. Moss AJ, Davis HT, DeCamilla J, Bayer LW. Ventricular ectopic beats and their relation to sudden and nonsudden cardiac death after myocardial infarction. Circulation 1979;60:998-1003. 10. Schulze RA Jr, Strauss HW, Pitt B. Sudden death in the year followmg myocardial infarction: relation to ventricular premature contractions in the late hospital phase and left ventricular ejection fraction. Am J Med 1977;62:192-9. II. Mukherji J, Rude RE. Poole KE, et aI. Risk factors for SUdden death after acute myocardial infarction: two-year follow-up. Am J Cardiol 1984;54:31-6. 12. Beta-Blocker Heart Attack Trial Research Group. A randomized trial of propranolol in patients with acute myocardial infarction. II. Mor• bidity results. JAMA 1983;240:2814-9. 13. Beta-Blocker Heart Attaek Trial Research Group. Beta-Blocker Heart Attack Trial: design features. Controlled Clin Trials 1981 ;2:275-85. 14. Klem MD, Baker S, Feldman CL, Hubelbank M, J..ane B. A vahdation

FRIEDMAN ET AL BETA-BLOCKER HEART ATTACK TRIAL RESEARCH GROUP

JACC Vol 7. No I January 1986.1-8

techmque for computenzed Holter tape processing systems used in drug efficacy testlOg. Comput Cardiol IEEE 1977;199-201.

farctlOn and ventncular runs 2 weeks to I year after infarction. Cir• culatIOn 1981 ;63:64-70.

15. Multicentre InternatIOnal Study. Improvement in prognosis of myo• cardial lOfarctlOn by long-term beta-adrenoreceptor blockade uslOg practolol. Br Med 1 1975;3:735-40.

27 Fnshman WH, Furberg CD, Friedewald WT. Beta-adrenergic block• ade for survivors of acute myocardial infarction. N Engl 1 Med 1984;310:830-7.

16. Wilhelmsson C, Vedin lA. Wilhelmsen L, Tibblin G, Werko L. Re• ductIOn of sudden deaths after myocardial infarction by treatment with alprenolol: preliminary results. Lancet 1974;2: 1157-60.

28. Bigger JT lr, Coromilas 1. How do beta-blockers protect after myo• cardial infarctIOn? (editorial) Ann Intern Med 1984;101:256-8.

8

17. Ahlmark G. Saetre H. Long-term treatment with beta-blockers after myocardial lOfarction. Eur 1 Clin Pharmacol 1976;10:77-83.

29. Vincent HH, Man in't Veld Al, Boomsma F, Derkx FHM, Wenting Gl, Schalekamp MADH. CardioprotectlOn by blockade of beta2-ad• renoceptors. Eur Heart 1 1983;4(suppl D):109-15.

18. Lemberg L, Castellanos A. Arcebal AG. The use of propranolol in arrhythmias complicating acute myocardial infarction. Am Heart 1 1970;80:479-87.

30. Furberg CD, Hawkil1s CM, Lichstein E. Effect of propranolol in postinfarction patients with mechanical or electrical complications. Circulation 1984;69:761-5.

19. Lemberg L, Arcebal AG, Castellanos A, Slavin D. Use of alprenolol in acute cardiac arrhythmias. Am 1 Cardiol 1972;30:77-81.

31. deSoyza N, Bennett FA, Murphy ML, Bissett lK, Kane 11. The relationship of paroxysmal ventricular tachycardia complicating the acute phase and ventricular arrhythmia during the late hospital phase of myocardial infarction to long-term survival. Am 1 Med 1978;64:377-81.

20. Kotler MN, Tabatznik B, Mower MM, Tominaga S Prognostic sig• nificance of ventricular ectopic beats with respect to sudden death in the late postlOfarctlOn period. Circulation 1973;47:959-66. 21. Podnd Pl. Can antiarrhythmic drugs cause arrhythmia? 1 Clin phar• macol 1984;24:313-9. 22. Anderson lL, Rodier liE, Green LS. Comparative effects of beta• adrenergic blocking drugs on experimental ventricular fibrillation threshold. Am 1 Cardiol 1983;51:1196-202. 23. Gang ES, Bigger 1T lr, Uhl EW. Effects of timolol and propranolol on inducible sustained ventricular tachyarrhythmias in dogs with sub• acute myocardial infarction. Am 1 Cardiol 1984;53:275-81. 24. Vismara LA, Amsterdam EA, Mason DT. Relation of ventricular arrhythmias in the late hospital phase of acute myocardial infarction to sudden death after hospital discharge. Am 1 Med 1975;59:6-12. 25. Denborough MA, Lovell RRH, Nestel Pl, Goble Al. Arrhythmias and late sudden death after myocardial infarction. Lancet 1968;1:386-8. 26. Kleiger RE, Miller lP, Thanavaro S, Province MA, Martin TF, Oliver Gc. Relationship between clinical features of acute myocardial in-

32. Morrison GW. Kumar EB, Portal RW, Aber CPo Cardiac arrhythmias 48 hours before, during, and 48 hours after discharge from hospital following acute myocardial infarction. Br Heart 1 1981;45:500-1 I. 33. Geddes 1S, Adgey AA1, Pantridge IF. Progress after recovery from ventricular fibrillation complicating ischaemic heart-disease. Lancet 1967;2:273-5. 34. Lawrie DM. Long-term survival after ventricular fibrillation compli• cating acute myocardial infarction. Lancet 1969;2: 1085-7. 35. Dewhurst NG, Hannan W1, Muir AL. Ventricular performance and prognosis after primary ventricular fibrillation complicating acute myocardial !nfarction. Eur Heart 11984;5:275-81. 36. Kitchen AH, Pocock S1. Prognosis of patients with acute myocardial infarction admitted to a coronary care unit. II. Survival after hospital discharge. Br Heart 1 1977 ;39: I 167-71.