422

Immediate Versus Deferred 13-Blockade Following Thrombolytic Therapy in Patients With AcuteMyocardial Infarction Results of the Thrombolysis in Myocardial Infarction (TIMI)II-B Study Robert Roberts, MD; William J. Rogers, MD; Hiltrud S. Mueller, MD; Costas T. Lambrew, MD; Daniel J. Diver, MD; Hugh C. Smith, MD; James T. Willerson, MD; Genell L. Knatterud, PhD; Sandra Forman, MA; Eugene Passamani, MD; Barry L. Zaret, MD; Frans J.T. Wackers, MD; and Eugene Braunwald, MD; for the TIMI Investigators

In the Thrombolysis in Myocardial Infarction (TIMI) Phase II trial, patients received intravenous recombinant tissue-type plasminogen activator (rt-PA) and were randomized to either a conservative or an invasive strategy. Within this study, the effects of immediate versus therapy, a group deferred ,8-blocker therapy were also assessed in patients eligible for of 1,434 patients of which 720 were randomized to the immediate intravenous group and 714 to the deferred group. In the immediate intravenous group, within 2 hours of initiating rt-PA metoprolol was given (5 mg intravenously at 2-minute intervals over 6 minutes, for a total intravenous dose of 15 mg, followed by 50 mg orally every 12 hours in the first 24 hours and 100 mg orally every 12 hours thereafter). The patients assigned to the deferred group received metoprolol, 50 mg orally twice on day 6, followed by 100 mg orally twice a day thereafter. The therapy was tolerated well in both groups and the primary end point, resting global ejection fraction at hospital discharge, averaged 50.5% and was virtually identical in the two groups. The regional ventricular function was also similar in the two groups. Overall, there was no difference in mortality between the immediate intravenous and deferred groups, but in the subgroup defined as low risk there were no deaths at 6 weeks among those receiving immediate (-blocker therapy in contrast to seven deaths among those in whom (3-blocker therapy was deferred. These findings for a secondary end point in a subgroup were not considered sufficient to warrant a recommendation regarding clinical use. There was a lower incidence of reinfarction (2.7% versus 5.1%,p=0.02) and recurrent chest pain (18.8% versus at 6 days in the immediate intravenous group. Thus, in appropriate postinfarction patients, ,l-blockers are safe when given early after thrombolytic therapy and are associated with decreased myocardial ischemia and reinfarction in the first week but offer no benefit over late administration in improving ventricular function or reducing mortality. (Circulation

,B-blocker

24.1%,p<0.02)

1991;83:422-437)

early in the therapy administered nhrombolytic course of acute infarction is as-

sociated with limitation of infarct size,12 improvement in ventricular function,3,4 and reduction in mortality.1,5-7 However, the marked benefit of

successful lysis may be somewhat attenuated by reocclusion, rethrombosis, and reinfarction.8 A varietyof approaches - pharmacological and mechanical have been suggested as possible means to reduce these complications and thereby to reduce mortality

From the Maryland Medical Research Institute, Baltimore, Md. Supported under research contracts from the National Heart, Institute, National Institutes of Health. Lung, and Blood Address for correspondence: Robert Roberts, MD, Cardiology Section, Baylor College of Medicine, The Methodist Hospital,

6535 Fannin -MS F-905, Houston, TX 77030. Address for reprints: Genell Knatterud, PhD, TIMI Coordinating Center, Maryland Medical Research Institute, Inc., 600 Wyndhurst Avenue, Baltimore, MD 21210. Received March 28, 1990; revision accepted September 18, 1990.

T

myocardial

Roberts et al Immediate Versus Delayed }3-Blockade

and morbidity. The principal objective of the Thrombolysis in Myocardial Infarction Phase II (TIMI II) trial was to determine whether these sequelae would be minimized by routine coronary arteriography and percutaneous transluminal coronary angioplasty See p 695 (PTCA) (when the anatomy is suitable) 18-48 hours after the start of thrombolytic therapy compared with no angiography or PTCA unless the patient developed symptoms or evidence of uncontrolled spontaneous or inducible myocardial ischemia.9 Within this study, the effects of immediate versus deferred ,8-blocker therapy were assessed in patients eligible to be randomized to receive either immediate pB-blocker therapy or deferred p8-blocker therapy. ,3-Blocker therapy was selected for study for several reasons. Selected ,3-blockers administered commencing days or weeks after myocardial infarction to patients at risk for future clinical events have prolonged survival.10"1' Also, ,-blockers given intravenously in the early hours following the onset of acute myocardial infarction have been shown to reduce early mortality.12 ,8-Blockers are used widely for the treatment of angina and hypertension; thus, some patients are already receiving p-blockade at the time they develop their myocardial infarction. However, the role of (3-blockers in patients recovering from myocardial infarction who have received thrombolytic therapy has not yet been assessed, and (8-blockers given with reperfusion therapy to animals following coronary occlusion appear to be beneficial in reducing infarct size and improving cardiac function.1314 On the other hand, since reperfusion is associated with transient impairment of regional contractility,'3 the possibility that ,B-blockade may have a deleterious effect should be determined. Prior to the thrombolytic era, (3-blockers were recommended in patients surviving myocardial infarction, primarily for long-term benefit.5"16 However, following thrombolysis, unlike the recognized, conventional vulnerable period of 6-12 weeks following myocardial infarction, the highly vulnerable period may start earlier-within hours to days of perfusion,8 and thus, the potential for immediate benefit exists. Accordingly, the TIMI 1I-B study was designed to assess prospectively whether ,p-blockers are beneficial if given immediately with the thrombolytic agent as opposed to being delayed for 6-8 days. The overall design of the TIMI II trial and the results of invasive versus conservative strategies have been published.9 In this paper, the results of immediate and delayed metoprolol therapy on regional and global left ventricular function, recurrent myocardial ischemia, reinfarction, and mortality are presented. Methods Patient Selection The TIMI II trial compared an invasive strategy (routine angiography and PTCA, where indicated, at 18-48 hours) and a conservative strategy (angiogra-

423

phy for possible angioplasty only in the event of recurrent ischemic symptoms or an ischemic response to predischarge submaximal exercise study) in 3,534 patients with acute myocardial infarction treated with recombinant tissue-type plasminogen activator (rt-PA), heparin, and aspirin.9 Patients were .75 years of age and were treated c4 hours after the onset of chest pain. The diagnosis of acute myocardial infarction was based on chest pain suggestive of acute myocardial ischemia lasting .30 minutes and ST segment elevation of 0.1 mV in two electrically contiguous leads. Exclusion criteria included any of the following: history of a cerebrovascular event, blood pressure of .180 mm Hg systolic or > 110 mm Hg diastolic, bleeding disorder, surgery within the previous 2 weeks, recent prolonged cardiopulmonary resuscitation, PTCA or severe trauma within 6 months, previous coronary artery bypass surgery, prosthetic heart valve replacement, left bundle branch block, dilated cardiomyopathy, or other serious illness. In the clinical centers participating in the TIMI 11-B study, patients were evaluated to determine whether they were eligible to be randomized to receive immediate or deferred pB-blocker therapy. Patients were excluded from randomization if they had one or more of the following: an implanted pacemaker; a resting ventricular rate consistently <55 beats/min; a systolic blood pressure consistently <100 mm Hg; moist rales that did not clear with cough extending above the lower one third of the lung fields or pulmonary edema with consistent chest radiographic findings; advanced first-degree or more advanced heart block; asthma by history; wheezing on physical examination, or chronic obstructive lung disease requiring chronic therapy with corticosteroids or 8,2-stimulants; or p-blocker, verapamil, or diltiazem therapy on admission. Of the 2,948 patients enrolled in the TIMI 1I-B clinical centers, 1,434 (49%) were eligible for the p-blocker study and were randomized to invasive or conservative strategy. The remaining 1,514 patients were excluded because of one or more of the following: p3-blocker, verapamil, or diltiazem therapy on admission (45.4%), resting ventricular rate consistently <55 beats/min (27.1%), systolic blood pressure consistently <100 mm Hg (21.3%), rales involving one third or more of the lung fields or pulmonary edema on chest radiogram (16.4%), significant firstdegree (PR interval of >0.26 second), second-degree or third-degree heart block (12.4%), asthma by history, wheezing on examination, or chronic obstructive pulmonary disease requiring chronic therapy with corticosteroids or ,82-stimulants (11.0%), or an implanted pacemaker (0.3%). The patients enrolled in the seven clinical centers participating in a second substudy, TIMI II-A, were not eligible for the p-blocker study. These patients were randomized to one of three treatment groups: immediate (.2 hours) catheterization and PTCA, delayed (18-48 hours) catheterization and PTCA, or no catheteriza-

424

Circulation Vol 83, No 2, February 1991

tion and PTCA (conservative strategy). The results for TIMI 11-A have been reported.17,18 Administration of Metoprolol All patients received rt-PA intravenously, and eligible patients were assigned to receive either immediate intravenous /3-blocker therapy or delayed oral ,B-blocker therapy, which was started on day 6. In the immediate intravenous group, metoprolol was given as soon as possible after initiating rt-PA. Three intravenous injections of 5 mg metoprolol were given at 2-minute intervals, for a total intravenous dose of 15 mg. Heart rate, rhythm, and systolic blood pressure were monitored, with readings obtained at a minimum of 1-minute intervals during the intravenous administration of the drug and at 1-2-minute intervals for 15 minutes thereafter. Therapy with metoprolol was stopped if any of the following occurred: lengthening of the PR interval beyond 0.26 seconds, second- or third-degree atrioventricular block, or wheezing or rales extending one third or more of the distance up the lung fields or pulmonary congestion on chest radiogram. Therapy with metoprolol was temporarily withheld for a transient decrease of heart rate to <45 beats/min or systolic blood pressure to <90 mm Hg and was restarted at half the usual dose (2.5 mg) if the patient exhibited an increase in heart rate to >49 beats/min or systolic blood pressure to >95 mm Hg within 10 minutes. Patients who did not meet these hemodynamic requirements within 10 minutes but who had heart rates of >45 beats/min and systolic blood pressures of >95 mm Hg 6 hours later were started on oral metoprolol at 25 mg every 12 hours. Fifteen minutes after the administration of the third intravenous dose of metoprolol, patients were given 50 mg orally every 12 hours during the first 24 hours and 100 mg orally every 12 hours thereafter. In patients assigned to the delayed oral /3-blocker group, on day 6 (at least 48-72 hours before the radionuclide ventriculogram), all patients were given 50 mg metoprolol twice daily for 1 day and then 100 mg metoprolol twice a day for as long as they remained within the study and tolerated this agent clinically. Concomitant Therapy All patients received rt-PA <4 hours after the onset of symptoms. The total dose of rt-PA used during the first 6 months of recruitment for TIMI II was 150 mg administered intravenously over 6 hours. Because of an unacceptable incidence of intracranial hemorrhage, the dose was subsequently reduced to 100 mg in the remaining patients.919 The regimen consisted of a 6 mg initial bolus dose followed by 54 mg in the first hour, 20 mg in the second hour, and 5 mg in each of the next 4 hours. Patients received lidocaine in the form of a bolus injection of 1-1.5 mg/kg followed by an infusion of 2-4 mg/min for a minimum of 24 hours. Heparin was initiated at the same time as rt-PA, with a 5,000-unit bolus injection followed by a continuous infusion at the rate of 1,000

units/hr; the dose was adjusted to maintain an activated partial thromboplastin time of 1.5-2 times control values. Aspirin at 80 mg/day was initiated on day 1 in the first 93 patients and on day 2 in the remainder. The aspirin dose was increased to 325 mg/day in all patients on day 6, when intravenous heparin was replaced by subcutaneous heparin (10,000 units twice a day) until hospital discharge. Radionuclide Angiography Equilibrium radionuclide angiography was performed prior to hospital discharge and again at 6 weeks. The procedure was performed according to a standardized protocol in all participating centers.20 The Radionuclide Core Laboratory identified each patient only by code name and number and was unaware of any clinical information. Global left ventricular ejection fraction was estimated from the left anterior oblique projection in the standard fashion

(end-diastolic counts-end-diastolic counts-.÷end-systolic counts). Regional left ventricular ejection fraction was estimated, as previously described,20 from the left anterior oblique projection by dividing the left ventricular end-diastolic region of interest into five segments: basal septal, apical septal, inferoapical, inferolateral, and posterolateral. The radionuclide ventriculogram was performed at rest and during exercise in the supine position on a bicycle ergometer to a heart rate of 120 beats/min for a maximum workload of 67 W (400 kg-mimin). Coronary arteriography was recommended in patients assigned to the conservative strategy who had either recurring ischemia or a positive exercise study, as defined below. The need for subsequent PTCA or coronary artery bypass grafting (CABG) was determined on the basis of specific anatomic findings and clinical considerations in each patient. In the invasive strategy arm, patients underwent contrast ventriculography routinely and global and regional function were analyzed, as previously outlined.21 A maximal supine bicycle exercise test with electrocardiogram and radionuclide ventriculography was performed in all patients at 6 weeks, and the development of angina or the appearance of ST segment depression or a decrease in the ejection fraction of >5% was considered indicative of a positive exercise test. End Points The primary end point selected for comparing the effects of immediate and deferred /3-blocker therapy was global left ventricular ejection fraction as measured by resting radionuclide ventriculography performed prior to hospital discharge. The effects of /3-blocker therapy were assessed by pooling the results for the invasive and conservative strategies and for each strategy separately. Power calculations indicated that with a study goal of 340 patients in each of the four treatment groups (two strategies within each of the two /3-blocker groups), the study had a power of at least 80% to detect differences of 3 units in mean resting ejection fraction among the treatment groups.

Roberts et al Immediate Versus Delayed ,8-Blockade Thus, there was adequate power even if the effects of immediate 83-blocker therapy were different in the two strategy groups. Prespecified secondary end points included resting left ventricular ejection fraction at 6 weeks, left ventricular ejection fraction on exercise both at hospital discharge and at 6 weeks, and regional left ventricular ejection fraction in the infarcted area at rest and on exercise prior to hospital discharge and at 6 weeks. Patients were also monitored for clinical events including myocardial ischemia. Recurrent chest pain in the hospital refers to chest pain that occurred with or without myocardial infarction as reported to the nurse during the hospital stay. Severe ischemic event in patients after discharge refers to chest pain occurring with or without myocardial infarction for which the patient was admitted to the hospital. Myocardial infarction was diagnosed when the patient had prolonged chest pain with characteristic electrocardiographic changes and/or confirmatory cardiac enzymes. The Mortality and Morbidity Classification Committee, comprising cardiologists who were not TIMI clinical center investigators, classified reported events without knowledge of treatment assignment according to criteria established before the classification of these events began. The clinical end points of total mortality and reinfarction were evaluated alone and in combination for the two ,B-blocker groups and for each of the four treatment groups at 6 days, 6 weeks, and 1 year. Death or reinfarction by 6 weeks was the primary end point for the assessment of the two strategies in TIMI II. It was recognized that with 680 patients in each p-blocker group (obtained by pooling patients assigned to the invasive and conservative strategies in each p-blocker group) there was sufficient power to detect only very large treatment differences for the clinical end points.

Statistical Considerations Patients were randomly assigned to treatment following a factorial design to provide equal numbers of patients in the four treatment groups, that is 1) immediate intravenous p-blockade with metoprolol and invasive strategy, 2) immediate intravenous p3-blockade with metoprolol and conservative strategy, 3) deferred oral metoprolol and invasive strategy, and 4) deferred oral metoprolol and conservative strategy. The results for all patients in each p-blocker group (pooling the invasive and conservative strategies) are presented as well as the results for each of the four treatment groups. The treatments were also compared in subgroups of patients defined by baseline characteristics. At the time of randomization, patients were categorized according to risk, with not-low-risk patients defined as those having one or more of the following: history of myocardial infarction, ST segment elevation of anterior electrocardiographic leads, rales extending upward to include more than one third of the lung fields, hypotension (systolic blood pressure of < 100 mm Hg)

425

and sinus tachycardia (atrial rate of > 100 beats/min), atrial fibrillation or flutter, age of .70 years, pulmonary edema, or cardiogenic shock. None of these features were present in the low-risk patients. Patients were also classified into two subgroups by the time from the onset of symptoms to the initiation of rt-PA treatment as early-treatment (.2 hours after the onset of symptoms) and not-early-treatment (2-4 hours after the onset of symptoms) patients. One analysis of the primary end point was based only on the results for patients who had the hospital discharge radionuclide study performed. A second analysis was performed to estimate the effect of data missing because of deaths. In this analysis, patients who died before the predischarge radionuclide study could be obtained were assigned an ejection fraction of 0. The statistical comparisons of ejection fraction for the defined groups were based on two-sample tests of means. For the primary end point, p<0.05 (for a two-sided test) was considered significant. The two ,p-blocker groups were also compared with respect to other end points based on the radionuclide studies prior to hospital discharge and at 6 weeks and with respect to clinical events. To adjust for multiple testing, p<0.01 (for a two-sided test) was specified for these comparisons. All patients were included in the group to which they were randomized originally whether or not immediate p-blocker therapy was given and whether or not PTCA was actually performed within the time specified for that treatment group. The results presented in this report are based on all data processed in the TIMI Coordinating Center as of December 1989. All analyses were performed using SAS.22 The Breslow-Day test for homogeneity of odds ratios was used to evaluate whether the effects of immediate ,-blocker therapy differed in the two strategies or whether immediate p-blocker therapy could be considered as having a common effect in both strategies. For the major clinical events, Kaplan-Meier life table curves for the immediate and deferred p-blocker groups were compared using the log rank test.23 The immediate and deferred p-blocker groups were also compared in subgroups of patients defined by baseline characteristics, and the immediate versus deferred }-blocker difference in the incidence of the end point in the first subgroup was also compared with that difference in the second (complementary) subgroup by a test of interaction. The Breslow-Day test for homogeneity of odds ratios was used to evaluate whether the effect of immediate ,-blocker therapy differed in the two subgroups or whether immediate ,-blocker therapy could be considered as having a common effect (with random variability) in both subgroups. Because of multiple tests for these comparisons, p<0.001 is considered to be necessary for the difference to be significant. Results Baseline Characteristics Of the 1,434 patients enrolled in the ,-blocker substudy, 720 were randomized to the immediate

426

Circulation Vol 83, No 2, February 1991

TABLE 1. Baseline Characteristics and Initiation of 3-Blocker Therapy Immediate p-blocker

Deferred,-blocker

(n=720) Characteristic Age (years) Race (white) Sex (male) Not-low-risk Age >70 years Prior MI Anterior MI Rales .1/3 of lung field Hypotension and sinus tachycardia Atrial fibrillation or flutter Pulmonary edema Cardiogenic shock Other risk factors History of angina History of CHF History of hypertension History of diabetes mellitus Hours from onset to rt-PA initiation Intravenous p-blocker Hours from onset to ,8-blocker initiation Hours from rt-PA initiation to 8-blocker initiation MI, myocardial infarction; CHF, congestive

(n=714) %

No.

614 620 461 61 47 403 3

85.3 86.1 64.0 8.5 6.5 56.0 0.4

626 603 462 66 63 400 2

87.7 84.4 64.7 9.2 8.8 56.0 0.3

p 0.54 0.18 0.38 0.79 0.61 0.10 0.98 0.99

9 12 3 0

1.3 1.7 0.4 0.0

7 11 1 1

1.0 1.5 0.1 0.1

0.80 0.85 0.62 0.50

313 6 216 77

43.5 0.8 30.0 10.7

313 10 219 82

43.8 1.4 30.7 11.5

0.89 0.33 0.78 0.63

No.

54.8

55.2

...

3.3

0.62

2.6

2.6 651

%

33

...

4.1

...

...

0.7 1.6 heart failure; rt-PA, recombinant tissue-type plasminogen activator.

intravenous group and 714 to the deferred group (Table 1). The numbers of patients assigned to invasive and conservative strategies in the immediate intravenous group were 366 and 354 and those in the deferred group were 365 and 349, respectively. The patients randomized to the two pl-blocker groups were similar with respect to age, race, sex distribution, and risk status. Intravenous rt-PA was initiated at a mean of 2.6 hours after the onset of chest pain and when the chest pain was still ongoing in 87.4% of the patients. Most (64.4%) patients were categorized as not-low-risk, primarily because of the presence of anterior myocardial infarction (56.0%), age .70 years (8.9%), or prior myocardial infarction (7.7%). Adherence to Assigned Therapy Of the 720 patients assigned to immediate p8-blocker therapy, intravenous metoprolol was initiated in 651 (90.4%) at a mean of 42 minutes after the initiation of rt-PA. Sixty-nine patients eligible and assigned to immediate p-blocker therapy did not receive it because of the development of a contraindication between randomization and the onset of treatment for one or more of the following reasons: hypotension, low heart rate, or bradycardia (59.4%); heart failure, bundle branch block, or heart block (17.4%), or other (23.2%). Of the 651 patients in the

immediate ,-blocker group who received intravenous metoprolol, 87.6% (570) received the full 15 mg dose, whereas 5.4% (35) received 1-5 mg, 4.9% (32) received 6-10 mg, 1.4% (nine) received 11-14 mg, and 0.6% (four) received > 15 mg. Reasons for administering <15 mg of intravenous metoprolol included one or more of the following: systolic blood pressure of <95 mm Hg (8.8%, 57), heart rate of <45 beats/ min (5.4%, 35), second- (0.3%, two) or third- (0.5%, three) degree heart block, and other (2.8%, 18). Subsequent p-blocker therapy was given orally during hospitalization in 94.7% (682) of the 720 patients assigned to the immediate p-blocker group. Of the 682 patients receiving oral therapy, 82.6% (563) had therapy initiated on the same day as rt-PA (day 0), 16.4% (112) on days 1-7, and 1.0% (seven) after day 7. Patients were not started on oral p-blockers for one or more of the following reasons: died (1.1%, eight of 720), systolic blood pressure of <90 mm Hg (3.1%, 22 of 720), heart rate of <45 beats/ min (1.7%, 12 of 720), moist rales or pulmonary edema (1.3%, nine of 720), second- or third-degree heart block (0.6%, four of 720), PR interval of >0.24 second (0.1%, one of 720), or miscellaneous (1.0%, seven of 720). In the deferred group, oral p-blocker therapy was initiated in 81.5% (582) of the 714 patients, early in

Roberts et al Immediate Versus Delayed 13-Blockade TABLE 2. Clinical Events During First 24 Hours After Initiation of rt-PA Immediate p-blocker

(n=720) No. Event Chest pain in relation to rt-PA infusion 635 Present before rt-PA 150 Present after rt-PA 141 Decreased at conclusion 707 Decreased or no pain During rt-PA infusion 238 Dramatic relief of chest pain Dramatic worsening of chest 108 pain Rapid normalization of ST 235 segments Exacerbation of ST segment 64 abnormalities Appearance of new arrhythmias or conduction disturbances 398 Resolution of arrhythmias or conduction disturbances 100 Decreased blood pressure or clinical signs of reduced 239 perfusion Increased blood pressure or clinical signs of increased 74 perfusion rt-PA, recombinant tissue-type plasminogen activator.

22.9%, as scheduled in 73.2%, and late in 3.8%. Patients in the deferred group were not started on oral /3-blockers for one or more of the following reasons: died (2.5%, 18 of 714), moist rales or pulmonary edema (4.6%, 33 of 714), systolic blood pressure of <90 mm Hg (2.8%, 20 of 714), second- or third-degree heart block (0.7%, five of 714), PR interval of >0.24 second (0.6%, four of 714), or miscellaneous (9.5%, 68 of 714). Therapy was given intravenously to 33 patients (4.6%), primarily on the request of the physician for relief of pain. Effect of Therapy on Clinical Variables The systolic blood pressure in the immediate /B-blocker group decreased during the first hour on the average by 9.8% compared with 7.1% in the deferred /3-blocker group (p<0.01). The average diastolic blood pressures did not differ significantly between the two groups. The mean heart rate decreased by 6% in the immediate group compared with an increase in the deferred group (p<0.001) throughout the initial 3 hours of rt-PA treatment. The occurrence of accelerated idioventricular rhythm during rt-PA infusion was more common in the immediate group than in the deferred group (36.5% versus 31.1%, p=0.03). Otherwise, neither the appearance nor the resolution of dysrhythmias differed significantly between the two /3-blocker arms. Although chest pain was present in 87.4% of the 1,434 patients enrolled in the /3-blocker substudy when rt-PA was initiated and in only 22.0% upon

S

Deferred }-blocker (n=714) No. N

427

p

88.2 20.8 19.6 98.2

619 165 149 691

86.7 23.1 20.9 96.8

0.43 0.29 0.41 0.09

33.1

229

32.1

0.70

15.0

127

17.8

0.15

32.6

248

34.7

0.40

8.9

66

9.2

0.81

55.3

406

56.9

0.54

13.9

117

16.4

0.19

33.2

203

28.4

0.05

10.3

73

10.2

0.97

completion of rt-PA infusion, the proportion of patients having diminished chest pain or marked relief of chest pain did not differ between the immediate and deferred /3-blocker groups (Table 2). Furthermore, there were no differences between groups with respect to the frequency of rapid normalization or exacerbation of ST segment abnormalities. Hemorrhagic Events The dose of rt-PA was 100 mg in 87.6% (631) of the 720 patients in the immediate /3-blocker group and in 86.4% (617) of the 714 in the deferred /3-blocker group; in the remaining patients the dose was 150 mg (Table 3). Hemorrhagic complications were more common with the higher dose of rt-PA, but they occurred with similar frequency in the two ,3-blocker groups with the exception that intracranial hemorrhage was less frequent among patients in the immediate /3-blocker group. Among the patients receiving 150 mg rt-PA, two assigned to immediate /3-blocker therapy sustained intracranial hemorrhage .6 days after entry compared with five in the deferred /3-blocker group (difference not significant). Among patients receiving 100 mg rt-PA, none assigned to immediate /3-blocker therapy sustained intracranial hemorrhage s6 days after entry compared with five in the deferred /3-blocker group, a trend suggestive of a protective effect but not significant by study criteria (p=0.03). There was no difference in the plasma levels of plasminogen, fibrin degradation products, or fibrinogen between patients

428

Circulation Vol 83, No 2, February 1991

TABLE 3. Clinic-Reported Intracranial Hemorrhages Within First 6 Days After Entry

Immediate f-blocker (N=720) Deferred f-blocker (N=714) rt-PA % % No. n No p n dose 2 97 5.2 2.2 5 0.45 150 mg 89 0.03 5 0.8 631 0 0.0 617 100 mg All neurologic events reported by Thrombolysis in Myocardial Infarction clinics were reviewed by two neurologists unaware of treatment assignment. Of two clinic-reported intracranial hemorrhages in 150 mg rt-PA immediate fl-blocker subgroup, both were classified as intracerebral hemorrhages by neurologists. Of five clinic-reported intracranial hemorrhages in 150 mg rt-PA deferred fl-blocker subgroup, four were classified as intracerebral hemorrhages and one as subdural hematoma by neurologists. Of five clinic-reported intracranial hemorrhages in 100 mg rt-PA deferred l3-blocker subgroup, neurologists classified three as intracerebral hemorrhages, one as cerebral infarction with hemorrhagic conversion, and one as subdural hematoma. rt-PA, recombinant tissue-type plasminogen activator.

assigned to the immediate and deferred /-blocker groups (Table 4). Primary End Point Data on the primary end point of the p-blocker substudy was available in 83.8% (603) of the 720 patients in the immediate f8-blocker group and in 83.5% (596) of the 714 patients in the deferred ,8-blocker group (Table 5). Global left ventricular ejection fraction at hospital discharge averaged 50.5% among the 1,199 patients in whom it was measured and was similar in the two 8-blocker groups. Separate analyses of the subgroups in the invasive and conservative strategies showed no difference in ejection fraction between patients assigned to immediate and delayed p-blocker therapy (Table 5). Reasons for missing radionuclide ejection fraction data included the following: medical instability (30 patients), death .21 days after study entry (44 patients), patient/physician refusal (33 patients), technically unsatisfactory study (100 patients), and other (28 patients). In a second analysis, patients dying without radionuclide left ventricular ejection fraction measurement were assumed to have an ejection fraction of 0. With this imputation, an estiTABLE 4. Coagulation Core Laboratofy Measurements

Immediate fl-blocker (n=720) No. Mean Plasminogen (%) Before 278 50 minutes 340 5 hours 363 8 hours 342 Fibrin degradation products Before 375 374 50 minutes 5 hours 400 8 hours 366 Fibrinogen (mg/dl) Before 379 379 50 minutes 5 hours 407 373 8 hours

Deferred fl-blocker (n=714) No.

Mean

p

280 328 359 340

102.2 67.4 57.2 57.6

0.13 0.23 0.76 0.76

152.0 297.0 253.0

402 395 413 381

13.8 200.0 277.0 246.0

0.26 0.18 0.64 0.86

294.3 244.8 183.7 179.4

402 399 415 387

305.3 248.0 195.6 188.3

0.05 0.46 0.02 0.11

104.6 69.0 56.8 57.3

(,ug/mg) 20.5

mate of ejection fraction was available in 86.8% (1,244) of the 1,434 study patients, but there still was no difference in ejection fraction between the two fl-blocker groups. Furthermore, the baseline exercise ejection fraction, the difference between the peak and baseline exercise ejection fractions, and the infarcted zone ejection fraction did not differ between the p-blocker arms either at hospital discharge or at 6 weeks (Table 5). In the invasive strategy arm, assessment of left ventricular function was also available from contrast ventriculography, with global ejection fraction in the subgroup assigned to immediate fl-blocker therapy averaging 48.2, which was similar to the average of 47.2 in the subgroup randomized to delayed p-blocker therapy. The regional function determined by contrast ventriculography was also similar between the two subgroups, with the infarcted segment of maximal hypokinesis averaging -2.54 in the immediate p-blocker subgroup and -2.63 (p=0.25) in the deferred fl-blocker subgroup.

Secondary End Points There was no difference between the immediate and deferred fl-blocker groups at 1 year for the secondary end points of death and reinfarction, alone or in combination (Tables 6 and 7). Reinfarction was less common at 6 days and 6 weeks in the patients assigned to receive immediate fl-blocker therapy, but there was no difference at 1 year (Tables 6 and 7). In the immediate fl-blocker group the incidence of reinfarction increased from 4.5% at 6 weeks to 8.6% at 1 year, which is similar to the incidence of 9.6% observed at 1 year in the deferred fl-blocker group. The 91% increase in the incidence of reinfarction from 6 weeks to 1 year (4.5% to 8.6%) in the immediate fl-blocker group is significantly greater than the 32% increase (7.3% to 9.6%) in the deferred fl-blocker group over the same interval. There was also a lower incidence of recurrent chest pain within the first 6 days in the immediate group than in the deferred group. In the low-risk subgroup, there were no deaths at 6 weeks among the patients randomized to immediate l3-blocker therapy in contrast to seven deaths among those assigned to deferred fl-blocker therapy (Table 8). However, survival at 6 weeks was similar in the two fl-blocker arms among the patients classified as not-low-risk (Table

Roberts et al Immediate Versus Delayed P-Blockade

429

TABLE 5. Resting and Exercise Radionuclide Left Ventricular Ejection Fraction at Discharge and at 6-Week Follow-up

Immediate p-blocker (n=720) No. Mean (%)

Deferred ,3-blocker (n=714) Mean (%) No.

p

At discharge 603 51.0 Resting EF 296 51.6 Invasive strategy 307 50.3 Conservative strategy 535 50.7 Baseline exercise EF 535 54.3 Peak exercise EF 535 3.6 Peak EF-baseline EF Infarcted zone segmental EF (at rest) 596 43.6 Computed§ 596 48.7 Adjusted J1 At follow-up 542 50.4 Resting EF 473 50.5 Baseline exercise EF 471 53.6 Peak exercise EF Peak EF-baseline EF 471 3.2 Infarcted zone segmental EF (at rest) 539 44.2 Computed§ 539 50.8 Adjusted 1 EF, ejection fraction. *p for 8-blocker effect. tp for strategy effect. jp for (-blockerxstrategy interaction. §Based on average of regional wall motion in infarct-related coronary artery. 1 Adjusted for distribution of infarct-related coronary artery in each (3-blocker group.

8). At 1 year (Table 8) the effect of immediate 13-blocker therapy was maintained in the low-risk subgroup, as was the lack of a difference between immediate and delayed therapy in the not-low-risk subgroup. The factors contributing to death during hospitalization in the two 3-blocker groups were similar. Among the 22 patients in the immediate 13-blocker group dying during the initial hospitalization, contributing causes included one or more of the following: pump failure (eight patients), cardiac rupture (seven patients), arrhythmia (five patients), PTCA complication (two patients), CABG complication (one patient), hemorrhagic complication (two patients), and other (three patients). Among the 25 patients in the deferred P-blocker group who died during the initial hospitalization, contributing causes included one or more of the following: pump failure (eight patients), cardiac rupture (six patients), arrhythmia (nine patients), CABG complication (two patients), hemorrhagic complication (five patients), and other (three patients). Factors contributing to death during hospitalization differed between the not-low-risk and low-risk subgroups. Among the 40 deaths in the not-low-risk subgroup, pump failure and ventricular rupture were the major causes. In the seven deaths in the low-risk subgroup (all occurring in the deferred 13-blocker arm) the major cause was arrhythmias.

596 307 289 513 513 513

50.1 50.3 49.8 50.3 54.1 3.8

0.22* 0.21t 0.53t 0.60 0.81 0.66

590 590

43.8 49.1

0.86 0.83

548 497 497 497

50.8 51.2 54.3 3.1

0.56 0.38 0.50 0.77

544 544

44.6 51.4

0.74 0.68

The frequency of a severe ischemic event was the same in both 13-blocker groups. Conversely, there was a nonsignificant trend toward a higher incidence of new-onset congestive heart failure during the initial hospitalization in the immediate 13-blocker group (15.3%) compared with the deferred 1-blocker group (12.2%) (p=0.10). Otherwise, the two groups had similar frequencies of major events and procedures during the first 21 days. Patients treated .2 hours after the onset of symptoms with rt-PA who were assigned to the immediate 13-blocker group exhibited a reduction in mortality and reinfarction at 6 weeks compared with such patients in the deferred 13-blocker group (Table 9). However, this difference was not evident at 1 year (Table 9). In patients treated >2 hours after onset, the frequency of mortality and reinfarction were similar regardless of whether they were assigned to the immediate or deferred 13-blocker group. Exercise Testing Bicycle exercise tests were performed at hospital discharge in 83.5% (1,198) and at 6 weeks in 79.9% (1,146) of the 1,434 study patients. Clinical, electrocardiographic, and radionuclide ventriculographic indexes of myocardial ischemia were present individually and collectively with equal frequencies in the immediate and deferred P-blocker groups. Discussion The primary objective of this study was to compare the effect of immediate intravenous versus delayed

430

Circulation Vol 83, No 2, February 1991

TABLE 6. Effect of Immediate Versus Deferred Therapy With /-Blocker on Secondary End Points by Life Table Estimates Immediate (-blocker Deferred p-blocker (n=720) (n=714) End point No. No. % % Within 6 days of entry Death 17 2.4 17 2.4 Death or reinfarction 34 4.7 50 7.0 Fatal or nonfatal reinfarction 19 2.7 36 5.1 Nonfatal reinfarction 17 2.4 33 4.7 Fatal reinfarction 2 3 0.3 0.4 Recurrent chest pain 134 18.8 170 24.1

Within 6 weeks of entry Death Death or reinfarction Fatal or nonfatal reinfarction Nonfatal reinfarction Fatal reinfarction Severe ischemic event Within 1 year of entry Death Death or reinfarction Fatal or nonfatal reinfarction Nonfatal reinfarction Fatal reinfarction Severe ischemic event

3.6 7.2 4.5 4.0 0.7 13.0

25 69 51 46 7 102

3.5 9.7 7.3 6.6 1.0 14.5

0.91 0.10 0.03 0.03 0.55 0.40

34 84 60 54 9 170

4.8 11.8 8.6 7.8 1.3 24.4

35 93 67 61 8 170

5.0 13.1 9.6 8.8 1.2 24.5

0.87 0.40 0.44 0.43 0.82 0.85

left ventricular ejection fraction after exercise, as well as mortality, reinfarction, and recurrent ischemic events. The results indicate that global resting left ventricular ejection fraction prior to hospital discharge did not differ between the immediate and deferred pl-blocker groups. Further, the response to peak exercise was similar in the two regimens. Regional function assessed by radionuclide and, in the invasive strategy arm, by contrast ventriculography showed no difference between the immediate and

TABLE 7. Subgroup Analyses of Secondary End Points by Strategy Invasive strategy Immediate Deferred

(n=366)

End point Within 6 days of entry Death Death or reinfarction Recurrent chest pain Within 6 weeks of entry Death Death or reinfarction Severe ischemic event Within 1 year of entry Death Death or reinfarction Severe ischemic event

0.98 0.07 0.02 0.02 0.65 0.02

26 52 32 28 5 92

oral ,X-blockers in patients undergoing thrombolytic therapy. All patients received rt-PA <4 (mean 2.6) hours after the onset of symptoms with one half of the patients receiving fl-blockers c2 hours after the initiation of thrombolytic therapy; in the remaining one half fl-blockers were deferred until days 6-8. The primary end point of the study was the global resting left ventricular ejection fraction measured just prior to hospital discharge, with secondary end points being resting regional ventricular function and

,(-blocker

p

Conservative strategy Immediate Deferred

(3-blocker

(3-blocker

(n-365)

(n=354)

(3-blocker

(n=349)

p (subgroup

No.

%

No.

%

p

No.

%

No.

%

p

effects)

13 23 67

3.6 6.3 18.3

8 30 94

2.2 8.2 25.8

0.27 0.31 0.02

4 11 67

1.1 3.1 18.9

9 20 76

2.6 5.7 21.8

0.15 0.09 0.35

0.07 0.46 0.32

18 30 40

4.9 8.2 10.9

14 39 49

3.8 10.7 13.4

0.47 0.25 0.30

8 22 52

2.3 6.2 14.7

11 30 53

3.2 8.6 15.2

0.46 0.23 0.85

0.31 0.88 0.53

20 41 75

5.5 11.2 20.5

16 49 83

4.4 13.4 22.7

0.50 0.36 0.46

14 43 95

4.0 12.1 26.8

19 44 87

5.4 12.6 24.9

0.35 0.85 0.56

0.25 0.61 0.35

Roberts et al Immediate Versus Delayed ,B-Blockade TABLE 8. Subgroup Analyses of Secondary End Points by Risk Category Low risk Deferred Immediate p-blocker p-blocker

End point Within 6 days of entry Death Death or reinfarction Recurrent chest pain Within 6 weeks of entry Death Death or reinfarction Severe ischemic event Within 1 year of entry Death Death or reinfarction Severe ischemic event

(n=252)

431

Not low risk

Immediate p-blocker (n=461) No. %

Deferred p-blocker (n=462) No. %

(n=259) No. %

No.

%

p

0 8 50

0.0 3.1 19.3

6 13 53

2.4 5.2 21.0

0.01 0.24 0.63

17 26 84

3.7 5.6 18.2

11 37 117

0 14 38

0.0 5.4 14.7

7 17 31

2.8 6.8 12.3

0.007 0.53 0.43

26 38 54

5.6 8.2 11.7

1 27 67

0.4 10.4 25.9

9 25 54

3.6 9.9 21.4

0.009 0.85 0.24

33 57 103

7.2 12.4 22.3

p (subgroup

p

effects)

2.4 8.0 25.3

0.25 0.15 0.009

0.006 0.76 0.25

18 52 71

3.9 11.3 15.4

0.21 0.12 0.10

0.003 0.80 0.11

26 68 116

5.6 14.7 25.1

0.34 0.30 0.32

0.006 0.46 0.12

days, the functional status of the infarcted region correlated with long-term survival. Metoprolol given to patients with a first myocardial infarction immediately after acute angioplasty was associated with improved function of the infarcted region measured 9 days after coronary recanalization.25 These studies suggest that any significant improvements in ventricular function would have been detected, and thus our lack of a difference in ventricular function between patients assigned to the immediate and delayed P-blocker groups suggests that early p-blocker therapy is not associated with any important improvements in ventricular function. Animal experiments suggest that immediate /3-blockade can augment the salvage of infarcting myocardium accomplished by coronary reperfusion. In anesthetized dogs, timolol administered 15 minutes into a 3-hour period of

deferred ,3-blocker groups. Regional ejection fraction, adjusted for the infarct-related artery, did not show significant differences either at rest or at peak exercise prior to hospital discharge, and similar results were observed at 6 weeks. These results indicate that although it is safe to administer metoprolol to appropriately selected patients with acute myocardial infarction immediately following thrombolytic therapy, the lack of a benefit on ventricular function by immediate /3-blocker administration requires further discussion. Sheehan et a121 recently demonstrated that the functional changes in the infarcted zone early after myocardial infarction have important prognostic implications. Following thrombolytic therapy infarct zone function declined in all patients, but it declined least in those with sustained reperfusion.24 After 3

TABLE 9. Subgroup Analyses of Secondary End Points by Time To Treatment With rt-PA >2 hours after onset .2 hours after onset Deferred Immediate Deferred Immediate

p-blocker (n= 185)

,-blocker (n= 190) No. %

% End point Within 6 days of entry 5 2 1.1 Death 19 7 3.8 Death or reinfarction 59 30 16.2 Recurrent chest pain WVithin 6 weeks of entry 8 3 1.6 Death 26 10 5.4 Death or reinfarction 34 27 14.6 Severe ischemic event Within 1 year of entry 10 7 3.8 Death 30 12.4 23 Death or reinfarction 53 51 27.6 Severe ischemic event rt-PA, recombinant tissue-type plasminogen activator. No.

p-blocker

(n=535)

8-blocker

(n=524) %

p (subgroup p

effects)

2.3 5.9 21.2

0.60 0.53 0.48

0.22 0.10 0.01

17 43 68

3.2 8.2 13.0

0.37 0.83 0.68

0.08 0.03 0.62

25 63 117

4.8 12.0 22.3

0.84 0.75 0.97

0.48 0.54 0.97

p

No.

%

No.

2.6 10.0 31.0

0.27 0.02 0.001

15 27 104

2.8 5.0 19.4

12 31 111

4.2 13.7 17.9

0.14 0.007 0.39

23 42 65

4.3 7.8 12.2

5.3 15.8 27.9

0.49 0.35 0.94

27 61 119

5.0 11.4 22.2

432

Circulation Vol 83, No 2, February 1991

coronary occlusion followed by 3 hours of reperfusion resulted in a decrease in infarct size.26 The short-acting /3-blocker esmolol used in a similar design prevented the early functional deterioration of the infarcted region observed in control dogs after reperfusion alone.14 Again, infarct size was reduced in the esmolol-treated group. In experiments with conscious dogs, the effect of P-blockade was less convincing.13 Propranolol was administered 90 minutes after coronary occlusion and reperfusion, established after 2 and 4 hours, maintained for 24 hours and 1 month. Although propranolol diminished the early deterioration of contractile function of the infarcted region, there was no long-term functional improvement and only a slight decrease in infarct size. The combined administration of propranolol and the calcium blocker diltiazem, however, reduced infarct size and preserved regional ventricular function.13 Van de Werf et a127 simulated in closed-chest dogs the clinical situation of reperfusion therapy. These workers administered metoprolol immediately after thrombotic coronary occlusion and induced reperfusion with rt-PA 30 minutes later. At 24 hours, metoprolol-treated dogs showed a significant recovery of systolic function of the infarcted area in contrast to dogs receiving rt-PA alone. After 1 week, however, regional and global systolic function were identical in both treatment groups. This suggests by inference that /3-blockers had no significant effect on infarct size, which is similar to the results of /3-blocker trials in the prethrombolytic therapy era.15,28 In the TIMI 11-B study, /3-blockade was induced by the administration of metoprolol. Throughout the initial 3 hours of treatment with intravenous metoprolol, the heart rate was approximately 6% lower than that in the group that received /3-blocker after 6 days; during the first hour systolic blood pressure decreased by 9.8% compared with 7.1% in the delayed ,B-blocker group. Metoprolol, however, did not seem to have an impact on left ventricular function. It is conceivable that the lack of a change in the regional and global ejection fractions was related to the timing of the radionuclide ventriculography. Early improvement of the infarcted region by immediate /3-blockade might no longer be detectable in studies performed prior to hospital discharge. However, this is unlikely to be the reason since in the patients assigned to the invasive strategy who underwent cardiac catheterization 18-48 hours after randomization, infarct zone function was similar in the two /3-blocker groups. The lack of improvement in the global ejection fraction might have been related to opposing changes, contractility increasing over time in the infarcted region while decreasing in the noninfarcted zone.29 On the other hand, it is conceivable that immediate /3-blocker treatment did not change infarct zone function sufficiently to be detected by radionuclide ventriculography since patients with cardiac failure (who would be expected to have severely depressed ejection fractions and the

greatest potential for improvement) were excluded from the /3-blocker study. Using the same methodology in the TIMI I and TIMI II studies, however, infarct zone function tended to be better in patients treated with rt-PA than in those treated with streptokinase.20,30 Therefore, we conclude that immediate ,/-blocker therapy is unlikely to have a substantial benefit on ventricular function. The secondary end point total mortality was also unaffected by the timing of administration of /3-blockade since at 6 weeks mortality was 3.5% in the deferred ,3-blocker group and 3.6% in the immediate ,3-blocker group; at 1 year mortality was 5.0% and 4.8%, respectively. In the low-risk subgroup, the data indicate a reduced mortality among patients randomized to immediate /8-blocker therapy. At 6 weeks, there were no deaths among the patients randomized to immediate /3-blocker therapy compared with seven deaths among those assigned to deferred /3-blocker treatment; this difference was still evident at 1 year but was not significant by study criteria. These results, however, should be interpreted with caution since they represent a subgroup, albeit a prespecified subgroup, of the study patients; because of the small number of patients the results may reflect random chance rather than a therapeutic effect of /3-blockade. Thus, we do not recommend the use of early /3-blockers for prevention of death in this low-risk subgroup. In the other prespecified subgroup (i.e., patients receiving rt-PA treatment <2 hours after the onset of symptoms), the incidence of reinfarction and death was compared between patients assigned to the immediate and deferred /3-blocker groups. The incidence of the combined end points of reinfarction and death was less among patients assigned to immediate /3-blocker therapy, but for only one interval, namely, 6 weeks (5.4% versus 13.7%, p<0.007). Considering that this is a subgroup analysis of a secondary end point, these data, while suggestive, were not considered sufficient to warrant a recommendation regarding clinical use of /3-blockers for prevention of death or reinfarction even when given c2 hours after the onset of symptoms. The incidence of reinfarction and recurrent chest pain in the immediate /3-blocker group at 6 days was less than that in the deferred /-blocker group (p=0.02). These results are similar to those reported in previous /8-blocker trials that were not associated with thrombolytic therapy.10-12,31-33 Thus, it would appear reasonable to recommend early /3-blockade for prevention of ischemia and reinfarction during the first week following thrombolytic therapy. This difference, however, was not maintained between the two groups at 1 year. While these results suggest that the beneficial effect of /3-blockade was transient, we cannot from this study make meaningful observations as to why this occurred. The suggestion of the ISIS-1 investigators12 that immediate /3-blockade may decrease the incidence of cardiac rupture could not be supported by the TIMI II /B-blocker study. Twelve patients sustained fatal

Roberts et al Immediate Versus Delayed /3-Blockade

ventricular rupture in the immediate ,3-blocker group, 11 in the not-low-risk subgroup. Four ruptures occurred .48 hours after initiation of treatment, three more during the subsequent hospital phase, and five between hospital discharge and 1 year. In the deferred 13-blocker group, six fatal ventricular ruptures were observed, four in the not-low-risk subgroup. Three ruptures occurred during the initial 2 days of hospitalization and three during the subsequent hospital phase. The diagnosis of cardiac rupture was confirmed by autopsy in six instances. In the remaining patients, the diagnosis was made clinically. Complications of rt-PA and metoprolol therapy were prospectively considered as secondary end points in the TIMI IL-B study. The results suggest that the distribution of intracranial hemorrhages differed between the two 13-blocker groups. There were two intracranial hemorrhages in the immediate and 10 in the deferred 13-blocker group. In both groups, hemorrhages occurred more frequently among patients receiving 150 mg rather than 100 mg rt-PA. Although these data are of interest, they also need to be interpreted with caution. Because of multiple secondary end points in the TIMI IL-B study, the probability value of 0.02 does not indicate statistical significance. The results could represent a cluster of rare events in a relatively small population. During the pilot phase of the TIMI II study,19 five (1.5%) of 326 patients treated with 150 mg rt-PA developed intracranial hemorrhage. In contrast, intracranial hemorrhage occurred in only two (0.5%) of 386 patients in the TAMI I study34 also treated with 150 mg rt-PA and an adjunctive regimen comparable to that of the TIMI II pilot study. A similar variability was observed in two European cooperative studies.35,36 Although patients in each of these studies received 100 mg rt-PA, they were treated by the same centers and physicians using the same inclusion/ exclusion criteria. Only one (0.3%) of 367 patients suffered an intracranial hemorrhage in the study assessing the value of emergency angioplasty35 in comparison with five (1.4%) of 355 treated patients in the placebo-controlled study.36 Alternatively, metoprolol itself could have a protective effect and prevent intracranial hemorrhage. Regions of the brain with decreased vascular reserve because of atherosclerosis, thrombosis, and emboli are at risk for bleeding by lysis of thrombus, a systemic lytic state, impaired platelet function, and/or sudden changes in cerebral perfusion pressure. 13-Blockade might have an impact on the risk of intracranial hemorrhage by altering some of these potential triggers. During the initial 3 hours of treatment, systolic blood pressure was approximately 10% lower in the immediate than in the deferred 13-blocker group. Immediate 13-blocker therapy did not alter rt-PA kinetics. Plasma rt-PA levels, available in approximately 50% of the patients, did not differ between the two 13-blocker groups. As expected, there were no differences in the plasma levels of fibrinogen and fibrin degradation products in the

433

two 13-blocker groups. In previous studies, it has been shown that metoprolol does not alter ADP-induced platelet aggregation in patients with severe coronary artery disease37 or acute myocardial infarction38 or in normal volunteers.39 In contrast to propranolol, metoprolol did not significantly alter platelet aggregation, platelet cyclic AMP content, and fibrinolytic activity in hypertensive patients.40 The 1,1-selective blocker did not change endogenous thromboxane formation in the platelets of normal volunteers41,42 or alter the binding of prostaglandin D2 to rabbit platelets.43 Metoprolol might exert a protective effect on the cerebral circulation by altering the response to sudden bursts of catecholamine release during evolving myocardial infarction by dampening sudden increases in blood pressure, decreasing cardiac contractility, and decreasing the force of ventricular ejection. The data of the TIMI II-B study indicate that intravenous metoprolol is safe to give to patients with evolving myocardial infarction undergoing reperfusion therapy with rt-PA. Immediate metoprolol therapy did not appear to improve global or regional left ventricular function measured at hospital discharge but was associated with decreased rates of myocardial ischemia and reinfarction over the same interval, and for this benefit 13-blockade is recommended in association with thrombolytic therapy. Total mortality was not altered by immediate 1-blocker therapy, although there were fewer deaths in the low-risk subgroup. The data suggest that intracranial hemorrhage occurs less frequently in patients receiving immediate ,3-blocker therapy than in patients in whom such therapy is deferred. However, this must be confirmed in a larger randomized population.

Appendix 1. Investigators and Centers Participating in the TIMI Phase II Trial Study Chairman Eugene Braunwald, MD, Harvard University, Boston, Mass.

Coordinating Center Maryland Medical Research Institute, Baltimore, Md. Principal Investigator: G. L. Knatterud, PhD. Coinvestigators: M.L. Terrin, MD, MPH; S. Forman, MA; D.T. Harris; R. Ross, MSc; P.C. Wilkins, BS; M. Bryant, PhD; P.L. Canner, PhD; M. Carroll; J. Depkin, BS; J. Dotson; C. Fiery; M. Johnson; C. Kelly; P. Noble, BS; B. Thompson, PhD; W.R. Bell, MD; and L. Scherlis, MD.

Radiographic Core Laboratory University of Washington, Seattle, Wash. Principal Investigator: Harold T. Dodge, MD. Coinvestigators: B.G. Brown, MD, PhD; J.W. Kennedy, MD; F.H. Sheehan, MD; B. Bisson; and E. Bolson.

434

Circulation Vol 83, No 2, February 1991

Radionuclide Core Laboratory Yale University, New Haven, Conn. Principal Investigator: B. Zaret, MD. Coinvestigators: F. Wackers, MD; D.S. Kayden, MD; K. Davis, TTNM; and R. Green, RTNM.

Coagulation Core Laboratory University of Vermont, Burlington, Vt. Principal Investigator: K. Mann, PhD. Coinvestigators: D. Stump, MD; D. Collen, MD; E. Bovill, MD; and R. Tracy, PhD.

Electrocardiographic Core Laboratory for Qualifying Electrocardiograms George Washington University, Washington, D.C. Principal Investigator: A.M. Ross, MD. Coinvestigators: G.B. Bren, MD, and A.G. Wasserman, MD. Electrocardiographic Core Laboratory for Exercise Electrocardiograms St. Louis University, St. Louis, Mo. Principal Investigator: B.R. Chaitman, MD. Coinvestigators: R.D. Wiens, MD; L. Shaw, MS; M. Haueisen, BS; and L.T. Younis, MD, PhD. National Heart, Lung, and Blood Institute Program Office National Institutes of Health, Bethesda, Md. Principal Investigator: E.R. Passamani, MD. Coinvestigators: T.L. Robertson, MD; G. Lan, PhD; R. Solomon, MHS; and G. Sopko, MD.

Pathology Core Laboratory National Institutes of Health, the Clinical Center, Bethesda, Md. Principal Investigator: W.C. Roberts, MD. Coinvestigator: J. Kalan, MD. Percutaneous Transluminal Coronary Angioplasty

Quality-Control Laboratory Brown University, Providence, R.I. Principal Investigator: D.O. Williams, MD. Coinvestigators: R. Riley, MD; H. White, MD; B. Sharaf, MD; F. Fedele, MD; E. Thomas, MD; T. Drew, MD; J. Joelson, MD; and D. Hardink, RN. Drug Distribution Center Cooperative Studies Program, Veterans Administration Medical Research Service, Albuquerque, N.M. Principal Investigator: C. Colling, RPh, MS. Coinvestigators: C. Haakenson, RPh, MS, and M. Sather, RPh, MS. Clinical Centers Albert Einstein College of Medicine, Montefiore Medical Center, New York, N.Y. Principal Investigator: H.S. Mueller, MD. Coinvestigators: M.A. Greenberg, MD; R. Grose, MD; G. Gordon, MD; J.D. Goldfischer, MD; M. Bensman, MD; J. Cooper, MD; B. Ventura, RN; K. Hemingway, RN; M. Stein, RN; P.

Michaud-Edelstein, RN; L. Henson, RN; J. Durkin, RN; and P. Murphy, RN. Baylor College of Medicine, Houston, Tex. Principal Investigator: R. Roberts, MD. Coinvestigators: P. Nelson, RN; S. Minor, MD; C. Pratt, MD; A. Raizner, MD; W.L. Winters, MD; M.S. Verani, MD; J.M. Lewis, MD; J. Heibig, MD; N. Kleiman, MD; and M.K. VanderMolen, RN. Baystate Medical Center, Springfield, Mass. Principal Investigator: M.J. Schweiger, MD. Coinvestigators: R.E. Gianelly, MD; T. Marantz, MD; M. Porway, MD; E. Brickman, RN; F. Blank, RN; and J. Mitchell, RN. Boston University Medical Center, Boston, Mass. Principal Investigator: T.J. Ryan, MD. Coinvestigators: C.S. Apstein, MD; J.B. Cadigan III, MD; D.P. Faxon, MD; A.K. Jacobs, MD; M.A. Kellett Jr., MD; B.J. Polansky, MD; N.A. Ruocco, MD; T.A. Sanborn, MD; T. Varricchione, RRT; D.A. Weiner, MD; N. Battinelli, RN; and B. Hankin, RN. Bridgeport Hospital, Bridgeport, Conn. Principal Investigator: J.D. Babb, MD. Coinvestigators: Z.A. Adefuin, MD; M. Driesman, MD; J. Meizlish, MD; and D. Yasick, RN. Brown University, Providence, R.I. Principal Investigator: D.O. Williams, MD. Coinvestigators: T.M. Drew, MD; R.S. Riley, MD; H.J. White, MD; D. Shefcyk, MD; J. Joelson, MD; E. Thomas, MD; B. Sharaf, MD; F. Fedele, MD; M. Nathanson, MD; G. McKendall, MD; D. Becker, MD; D.L. Hardink, RN; M. Macedo, RN; G. Weeks, MD; R. Mich, MD; and E. Berger, MD. Columbia University, New York, N.Y. Principal Investigator: E. Powers, MD. Coinvestigators: A. Berke, MD; L. Johnson, MD; A.B. Nichols, MD; D.S. Reison, MD; A. Schwartz, MD; R. Watson, MD; E. Escala, RN; H.S. Wasserman, MD; and M. Apfelbaum, MD. Cornell Medical Center, New York, N.Y. Principal Investigators: J.S. Borer, MD, and T.L. Schreiber, MD. Coinvestigators: D.H. Miller, MD; J.W. Moses, MD; I. Tamari, MD; B. Charash, MD; B. Gerling, MD; D.A. Silvasi, RN; and A. McNulty, RN. George Washington University, Washington, D.C. Principal Investigator: A.M. Ross, MD. Coinvestigators: G.B. Bren, MD; R.I. Katz, MD; R.H. Leiboff, MD; P.J. Varghese, MD; A.G. Wasserman, MD; M. Magee, RN; G. Cavallo, RN; and J. Mendelson, RN. Harvard University, Boston, Mass. Principal Investigator: D.S. Baim, MD. Coinvestigators: D. Diver, MD; S. Herson, MD; J.E. Markis, MD; R.G. McKay, MD; B. Lorell, MD; C. (Brewer) Senerchia, RN, MS; G.A. Carey, RN; and J. Schweiger, RN. Maine Medical Center, Portland, Me. Principal Investigator: C.T. Lambrew, MD. Coinvestigators: W.D. Alpern, MD; R.A. Anderson, MD; D.J. Cutler, MD; J.P. Driscoll, MD; M. Kellett, MD; J.C. Love, MD; P.R. Minton, MD; R.L. Morse, MD; P.K. Shaw, MD; P.W. Sweeney, MD; S. Vermilya, RN; P. Birmingham, RN; and N. McIntire, RN.

Roberts et al Immediate Versus Delayed f-Blockade

Mayo Foundation, Rochester, Minn. Principal Investigator: J.H. Chesebro, MD. Coinvestigators: B.J. Gersh, MD; F.A. Miller, MD; M.B. Mock, MD; H.C. Smith, MD; R. Frye, MD; D.L. Hayes, MD; I. Clements, MD; W.K. Freeman, MD; J.A. Rumberger, MD; R. Gibbons, MD; R. Nishimura, MD; R. Rodeheffer, MD; R. Click, MD; J. Oh, MD; L. Sinak, MD; D. Klees, LPN; L. Meyers, LPN; R. Vlietstra, MD; J. Bresnahan, MD; D. Holmes Jr., MD; and G. Reeder, MD. New York Medical College, Valhalla, N.Y. Principal Investigator: M.V. Herman, MD. Coinvestigators: M.B. Weiss, MD; M. Cohen, MD; J. Levy, MD; M. Feld, MD; R. Grief, MD; J.H. Stein, MD; R. Wallach, MD; A.M. Kanakaraj, RN; V. Rosal-Greif, RN; and Y. Sait, PA. New York University, New York, N.Y. Principal Investigator: F. Feit, MD. Coinvestigators: J.N. Slater, MD; A. Simon, RN; J. Breed, RN; M.S. Nachamie, MD; W.J. Cole, MD; I.C. Schulman, MD; M.J. Rey, MD; M. Attubato, MD; and S. Shapiro, RN. North Shore University, Manhasset, N.Y. Principal Investigator: J. Morrison, MD. Coinvestigators: V. Padmanabhan, MD; P. Reiser, MD; L. Ong, MD; S. Green, MD; A. Tortolani, MD; M.L. Andresen, RN; T. Imhof, RN; L. Genovese, RN; and M. Ward, RN. Northwestern University, Evanston, Ill. Principal Investigator: R. Davison, MD. Coinvestigators: T. McDonough, MD; B. Kramer, MD; S. Meyers, MD; P. Niemyski, RN; M. Parker, RN; K. Kaplan, MD; D. Fintel, MD; M. Salinger, MD; D.C. Hueter, MD; G. Wilner, MD; C. Berkowitz, MD; and K. Duun, RN. St. Louis University, St. Louis, Mo. Principal Investigator: B.R. Chaitman, MD. Coinvestigators: M.G. Vandormael, MD; M.J. Kern, MD; W.P. Hamilton, MD; J.G. Dwyer, MD; T. Thornton, RN; J. Anthony, RN; K. Galan, RN; M. Major, RN; and G. Huber, RN. University of Alabama, Birmingham, Ala. Principal Investigator: W.J. Rogers, MD. Coinvestigators: J.G. Arciniegas, MD; W.A. Baxley, MD; R.C. Bourge, MD; T.M. Bulle, MD; T.B. Cooper, MD; L.S. Dean, MD; R. Hess, DO; W.A.H. MacLean, MD; S.E. Papaietro, MD; C. Saenz, MD; A.W.H. Stanley, MD; M.T. Simpson, MD; K. Bynum, RN; T. Eubanks, RN; and L. Maske, RN. University of Massachusetts, Worcester, Mass. Principal Investigator: J. Gore, MD. Coinvestigators: J.S. Alpert, MD; J.R. Benotti, MD; J. Leppo, MD; I.S. Ockene, MD; J.F. Rippe, MD; B.H. Weiner, MD; J. Dalen, MD; J.M.J. Gaca, MD; S.P. Ball, RN; J. Corrao, RN; and C. Mahan, RN. University of Minnesota, Minneapolis, Minn. Principal Investigator: M. Hodges, MD. Coinvestigators: W.T. Hession, MD; S.W. Sharkey, MD; D. Wysham, MD; I. Goldenberg, MD; A. Adicoff, MD; R. Brandenburg Jr., MD; G.L. Gobel, MD; L. Nordstrom, MD; R. Van Tassel, MD; C. White, MD; R. Wilson, MD; A. Ettinger, RN; L. Palmquist, RN; C. Siebold, RN; N. Carruthers, RN; and C. Farmer, RN.

435

University of Texas, Dallas, Tex. Principal Investigator: J. Willerson, MD. Coinvestigators: L.D. Hillis, MD; G.J. Dehmer, MD; D.L. Brown, MD; M. Winniford, MD; B.G. Firth, MD; M.M. Carry, MD; B. Toates, RN; S. Cochran, RN; P. Surratt, RN; and J. Moore, RN, MSN. Washington University, St. Louis, Mo. Principal Investigator: P.A. Ludbrook, MD. Coinvestigators: A.J. Tiefenbrunn, MD; N.A. Riciotti, RN, MSN; A.S. Jaffe, MD; and B.E. Sobel, MD. William Beaumont Hospital, Royal Oak, Mich. Principal Investigator: R. Ramos, MD. Coinvestigators: G. Timmis, MD; V. Gangadharan, MD; S. Gordon, MD; C. Tollis, RN; and E. Worden, RN. Yale University, New Haven, Conn. Principal Investigator: L.S. Cohen, MD. Coinvestigators: C.K. Francis, MD; J. Alexander, MD; D. Copen, MD; M. Cleman, MD; H. Cabin, MD; M. Remetz, MD; L. Decklebaum, MD; J. Gerard-Amatruda, RN; D. Penn, RN; A. Miller, RN; and C. Piselli, RN. TIMI Phase II Committees Operations Committee: Chairman: E. Braunwald, MD. Members: G. Knatterud, PhD; E. Passamani, MD; T. Robertson, MD; and R. Sdlomon, MHS. Executive Committee: Chairman: E. Braunwald, MD. Members: B. Chaitman, MD; J. Chesebro, MD; H. Dodge, MD; G. Knatterud, PhD; K. Mann, PhD; H. Mueller, MD; E. Passamani, MD; R. Roberts, MD; W. Rogers, MD; B. Sobel, MD; D. Stump, MD; D. Williams, MD; and B. Zaret, MD. Hemorrhagic Event Review Committee: Chairman: J. Chesebro, MD. Members: A. Berke, MD; E. Bovill, MD; F. Feit, MD; J. Gore, MD; L.D. Hillis, MD; C. Lambrew, MD; R. Leiboff, MD; J. Markis, MD; L. Offen, MD; C. Pratt, MD; S. Sharkey, MD; G. Sopko, MD; and M. Terrin, MD. Mortality and Morbidity Classification Committee: Chairman: M. Weisfeldt, MD. Members: W. Baker, MD; M. Cowley, MD; K. Kent, MD; E. Lichstein, MD; T. Robertson, MD; L. Scherlis, MD; and M. Terrin, MD. Safety and Data Monitoring Committee: Chairman: F. Klocke, MD. Members: J. Bailar, MD; R. Conti, MD; D. DeMets, PhD; V. Fuster, MD; T. Killip, MD; H. Roberts, MD; and L. Walters, PhD. ExOfficio Members: E. Braunwald, MD; G. Knatterud, PhD; E. Passamani, MD; and T. Robertson, MD. Steering Committee: The members of the Steering Committee are the Study Chairman and the Principal Investigators from the TIMI Clinical Centers, the Core Laboratories, the Coordinating Center, and the National Heart, Lung, and Blood Institute Program Office.

Acknowledgments We deeply appreciate the assistance of Donna L. Brown and Debora H. Fuhr in the preparation of this manuscript.

436

Circulation Vol 83, No 2, February 1991

References 1. Gruppo Italiano per lo Studio della Streptochinasi nell'Infarto Miocardico (GISSI): Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986;

1:397-401 2. ISAM Study Group: A prospective trial of intravenous streptokinase in acute myocardial infarction (ISAM). N Engl J Med 1986;314:1465-1471 3. Van de Werf F, Arnold AER: Intravenous tissue plasminogen activator and size of infarct, left ventricular function, and survival in acute myocardial infarction. Br Med J 1988;197: 1374-1379 4. Serruys PW, Simoons ML, Suryapranata H, Vermeer F, Wijns W, van den Brand M, Bar F, Zwaan C, Krauss XH, Remme WJ, Res J, Verheugt FWA, van Domburg R, Lubsen J, Hugenholtz PG, for the Working Group on Thrombolytic Therapy in Acute Myocardial Infarction of the Netherlands Interuniversity Cardiology Institute: Preservation of global and regional left ventricular function after early thrombolysis in acute myocardial infarction. J Am Coll Cardiol 1986;7: 729-742 5. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group: Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988;2: 349-360 6. Wilcox GR, von der Lippe G, Olsson CB, Jensen G, Skene AM, Hampton JR: Trial of tissue plasminogen activator for mortality reduction in acute myocardial infarction: AngloScandinavian Study for Early Thrombolysis (ASSET). Lancet

1988;2:525-530 7. The AIMS Trial Study Group: Effect of intravenous APSAC on mortality after acute myocardial infarction. Lancet 1988;1: 545-549 8. Schaer DH, Ross AM, Wasserman AG: Reinfarction, recurrent angina, and reocclusion after thrombolytic therapy

(abstract). Circulation 1987;76(suppl II):II-57 9. The TIMI Study Group: Comparison of invasive and conservative strategies after treatment with intravenous tissue plasminogen activator in acute myocardial infarction: Results of the Thrombolysis in Myocardial Infarction (TIMI) Phase II

Trial. N Engl J Med 1989;320:618-627 10. Beta-Blocker Heart Attack Trial Research Group: A randomized trial of propranolol in patients with acute myocardial infarction: I. Mortality results. JAMA 1981;274:1707-1714 11. Norwegian Multicenter Study Group: Timolol-induced reduction in mortality and reinfarction in patients surviving acute myocardial infarction. N Engl J Med 1981;304:801-807 12. ISIS-1 (First International Study of Infarct Survival) Collaborative Group: Mechanisms for the early mortality reduction produced by beta-blockade started early in acute myocardial infarction: ISIS-1. Lancet 1988;1:921-923 13. Bush LR, Buja LM, Tilton G, Wathen M, Apprill P, Ashton J, Willerson JT: Effects of propranolol and diltiazem alone and in combination on the recovery of left ventricular segmental function after temporary coronary occlusion and long-term reperfusion in conscious dogs. Circulation 1985;72:413-430 14. Lange R, Kloner RA, Braunwald E: First ultra-short-acting beta-adrenergic blocking agent: Its effect on size and segmental wall dynamics of reperfused myocardial infarcts in dogs.

Am J Cardiol 1983;51:1759-1767 15. Yusuf S, Peto R, Lewis J, Collins R, Sleight P: Beta blockade during and after myocardial infarction. An overview of the

randomized trials. Prog Cardiovasc Dis 1985;27:335-371

16. The MIAMI Trial Research Group: Metoprolol in acute myocardial infarction (MIAMI): A randomised placebo-

controlled international trial. Eur Heart J 1985;6:199-226

17. The TIMI Research Group: Immediate vs delayed catheterization and angioplasty following thrombolytic therapy for acute myocardial infarction: TIMI Phase II A results. JAMA

1988;260:2849-2858 18. Rogers WJ, Baim DS, Gore JM, Brown BG, Roberts R, Williams DO, Chesebro JH, Babb JD, Sheehan FH, Wackers

FJT, Zaret BL, Robertson TL, Passamani ER, Ross R, Knatterud GL, Braunwald E, for the TIMI II-A Investigators: Comparison of immediate invasive, delayed invasive, and conservative strategies following tissue plasminogen activator: Results of the Thrombolysis in Myocardial Infarction (TIMI) Phase II-A Trial. Circulation 1990;81:1457-1476 19. TIMI Operations Committee, Braunwald E, Knatterud GL, Passamani E, Robertson TL. Solomon R: Update from the Thrombolysis in Myocardial Infarction (TIMI) Trial. J Am

Coll Cardiol 1987;10:970 20. Wackers FJT, Terrin ML, Kayden DS, Knatterud G, Forman S, Braunwald E, Zaret BL, and the TIMI Investigators: Quantitative radionuclide assessment of regional ventricular function after thrombolytic therapy for acute myocardial infarction: Results of Phase I Thrombolysis in Myocardial Infarction (TIMI) Trial. JAm Coll Cardiol 1989;13:998-1005 21. Sheehan FH, Doerr R, Schmidt WG, Bolson EL, Uebis R, von Essen R, Effert S, Dodge HT: Early recovery of left ventricular function after thrombolytic therapy for acute myocardial infarction: An important determinant of survival. J Am Coll Cardiol 1988;12:289-300 22. SAS Institute, Inc: SAS Users Guide: Statistics. Version 5 Edition. Gary, NC, SAS Institute, Inc, 1985, pp 401-431, 529-557, 795-800 23. Kaplan KL, Meier P: Nonparametric estimation from incomplete observations. JAm Stat Assoc 1958;53:457-481 24. Sheehan FH: Left ventricular function deteriorates in the high-risk period after infarction (abstract). Circulation 1988; 78(suppl lI):II-128 25. Steingart RM, Matthews R, Gambino A, Kantrowitz N, Katz S: Effects of intravenous metoprolol on global and regional left ventricular function after arterial reperfusion in acute myocardial infarction. Am J Cardiol 1989;63:767-771 26. Hammerman H, Kloner RA, Briggs LL, Braunwald E: Enhancement of salvage of reperfused myocardium by early beta-adrenergic blockade (timolol). JAm Coll Cardiol 1984;3: 1438-1443 27. Van de Werf F, Vanhaecke J, Jang IK, Flameng W, Collen D, De Geest H: Reduction in infarct size and enhanced recovery of systolic function after coronary thrombolysis with tissuetype plasminogen activator combined with beta-adrenergic blockade with metoprolol. Circulation 1987;75:830-836 28. Roberts R, Croft C, Gold HK, Hartwell TD, Jaffe AS, Muller JE, Mullin SM, Parker C, Passamani ER, Poole WK, Raabe DS, Rude RE, Stone PH, Turi ZG, Sobel BE, Willerson JT, Braunwald E, and the MILIS Study Group: Effect of propranolol on myocardial infarct size in a randomized, blinded, multicenter trial. N Engl J Med 1985;312:932-936 29. Sheehan FH, Mathey DG, Schofer J, Krebber H-J, Dodge HT: Effect of interventions in salvaging left ventricular function in acute myocardial infarction: A study of intracoronary streptokinase. Am J Cardiol 1983;52:431-438 30. Wackers FJ, Zaret BL, Terrin ML, Ross R, Bourge R, Kayden D, Morrison J, Slater J, Weiss M, Braunwald E, and TIMI Investigators: Impact of invasive and conservative thrombolytic strategies on global and regional left ventricular function: Results of the TIMI II Trial (abstract). Circulation 1989;80(suppl II):II-1244 31. ISIS-1 Collaborative Group: A randomized trial of intravenous atenolol among 16,027 cases of suspected acute myocardial infarction. Lancet 1986;2:57-66 32. Yusuf S: Interventions that potentially limit myocardial infarct size: Overview of clinical trials.Am I Cardiol 1987;60:11A-17A 33. Hjalmarson A, Herlitz H, Malek I, Ryden L, Vedin A, Waldenstrom A, Wedel H, Elmfeldt D, Holmberg S, Myberg G, Swedberg K, Waagstein F, Waldenstrom J, Wilhelmsen L, Wilhelmsson C: Effect on mortality of metoprolol in acute myocardial infarction: A double blind randomised trial. Lancet 1981;2:823-827 34. Topol EJ, Califf RM, George BS, Kereiakes DJ, Abbottsmith CW, Canela RJ, Lee KL, Pitt B, Stack RS, O'Neill WW, and the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) Study Group: A randomized trial of immediate versus delayed elective angioplasty after intravenous plasminogen

Roberts et al Immediate Versus Delayed

35.

36.

37.

38.

activator in acute myocardial infarction. N Engi J Med 1987; 317:581 Simoons ML, Arnold AER, Betriu A, de Bono DP, Col L, Dougherty FC, von Essen R, Lambertz H, Lubsen J, Meier B, Michel PL, Raynaud P, Rutsch W, Sanz GA, Schmidt W, Serruys PW, Thery C, Uebis R, Vahanian A, Van De Werf F, Willems GM, Wood D, Verstraete M: Thrombolysis with tissue plasminogen activator in acute myocardial infarction: No additional benefit of immediate percutaneous coronary angioplasty. Lancet 1988;1:197 Van de Werf F, Arnold AER, the European Cooperative Study Group for Recombinant Tissue-Type Plasminogen Activator (rt-PA): Effect of intravenous tissue plasminogen activator on infarct size, left ventricular function and survival in patients with acute myocardial infarction. Br Med J 1988;297: 1374 Willich SN, Pohjola-Sintonen S, Bhatia SJS, Shook TL, Tofier GH, Muller JE, Curtis DG, Williams GH, Stone PH: Suppression of silent ischemia by metoprolol without alteration of morning increase of platelet aggregability in patients with stable coronary artery disease. Circulation 1989;79:557-565 Vilen L, Kutti J, Swedberg K, Vedin A: ADP-induced platelet aggregation and metoprolol treatment of myocardial infarction patients: A controlled study. Acta Med Scand 1985;217: 15-20

P-Blockade

437

39. Vilen L, Kutti J, Freden K, Lundborg P, Cronberg S: The peripheral platelet count and ADP-induced platelet aggregation in response to metoprolol and propranolol as studied in young healthy male volunteers. Scand J Haematol 1983;31: 440-446 40. Winther K: The effect of beta-blockade on platelet function and fibrinolytic activity. J Cardiovasc Pharmacol 1987;10(suppl 2):594-598 41. Greer IA, Walker JJ, McLaren M, Calder AA, Forbes CD: A comparative study of the effects of adrenoceptor antagonists on platelet aggregation and thromboxane generation. Thromb Haemost 1985;54:480-484 42. Srivastava KC: Influence of some beta blockers (pindolol, atenolol, timolol and metoprolol) on aggregation and arachidonic acid metabolism in human platelets. Prostaglandins Leukotnienes Med 1987;29:79-84 43. Hanasaki K, Mizuno Y, Ikeda M, Shimonnishi M, Yoshimura S, Tomita K, Ichikawa A: Beta2-adrenergic regulation of prostaglandin D2 receptor in rabbit platelets. Biochim Biophys Acta 1987;927:261-268

KEY WORDS * metoprolol * reinfarction * left ventricular function * tissue-type plasminogen activator, recombinant