Current issues of ACP Journal Club are published in Annals of Internal Medicine


Early metoprolol for myocardial infarction did not reduce mortality

ACP J Club. 1992 Jan-Feb;116:2. doi:10.7326/ACPJC-1992-116-1-002

Source Citation

The MIAMI Trial Research Group. Long-term prognosis after early intervention with metoprolol in suspected acute myocardial infarction: experiences from the MIAMI Trial. J Intern Med. 1991 Sep;230: 233-8.



To assess the effectiveness of early intervention with metoprolol in reducing mortality at 1 year after suspected myocardial infarction.


Randomized, placebo-controlled trial.


104 hospitals in 17 countries.


Patients had suspected acute myocardial infarction defined by indicative electrocardiographic (ECG) signs and chest pain for ≥ 15 min, with onset within the previous 24 h; myocardial infarction was diagnosed by these criteria and by elevated enzyme levels. Patients were age < 75 years, were not taking β-blockers or calcium-channel blockers, had a heart rate of > 65 beats/min and systolic blood pressure > 105 mm Hg, had not been resuscitated before hospitalization, and had no other serious disease. Of 26 439 potentially eligible patients, 19 585 met exclusion criteria, 1063 refused, and 5778 were randomized.


2877 patients were randomized to treatment with intravenous metoprolol (three 5-mg doses), followed by 200 mg/d, orally, for 15 days. 2901 patients were assigned to placebo administered similarly.

Thereafter, all patients were given metoprolol for 1 year, unless contraindicated.

Main outcome measures

Mortality in the first 15 days and during 1 year, when morbidity was also assessed.

Main results

In the first 3 days after randomization, 70.5% of the metoprolol group compared with 72.5% of the placebo group were diagnosed with definite infarction. 22% of each group had no sign of infarction. Mortality information was unavailable for 307 patients (5%) at 1 year. Mortality at 15 days and 1 year for the metoprolol group was 4.3% and 10.6% (289 deaths), respectively, compared with 4.9% and 10.7% (295 deaths) for the placebo group (P > 0.2 {95% CI -0.02% to 0.02%}*). The groups did not differ in 1-year mortality when they were stratified into high-risk and low-risk groups by ≥ 3 of the following: age > 60 years, abnormal ECG at entry; history of myocardial infarction, angina pectoris, congestive heart failure, hypertension, or diabetes; and previous long or short term treatment with diuretics or cardiac glycosides. The metoprolol and placebo groups did not differ with respect to symptoms of angina pectoris (33% vs 32%), bypass surgery (6%), return to work (57% vs 58%), and requirements for medications (82% vs 81%) including /f1 β-blockers, nitrates, calcium-antagonists, diuretics, and digitalis.


Metoprolol for suspected acute myocardial infarction, given early and continued for 15 days, did not affect 1-year mortality or morbidity rates.

Source of funding: Astra Pharmaceutical.

Address for article reprint: Dr. J. Herlitz, Division of Cardiology, Department of Medicine I, Sahlgrenska Hospital, S-413 45 Gothenburg, Sweden.

*Numbers calculated from data in article.


β-blockers have become a cornerstone in the treatment of acute myocardial infarction over the last decade. In the same period, however, many other interventions have been used, including aspirin, parenteral nitroglycerine, various thrombolytic agents, angioplasty, and bypass surgery. Patients are rarely treated with a single intervention and results can be difficult to interpret, as evidenced by the current controversies regarding thrombolytic therapy.

This study is based on patients enrolled from 1982 to 1984. Then, as now, a substantial number of patients could not be treated with β-blockade because of contraindications. Studies from the early 1980s showed that a risk reduction in mortality of approximately 25% resulted from therapy with β-blockers when begun within the first 25 days after myocardial infarction. The MIAMI trial addresses the question of immediate compared with delayed administration.

Beginning therapy at admission or after 15 days made no statistically significant difference in patient mortality. There was, however, a tendency toward greater benefit in "high risk" patients. This tendency may be related to better early myocardial "salvage" but was not sustained in comparisons of the 15-day to 1-year period in both groups. A meta-analysis of studies of early compared with delayed β-blockade indicates a small benefit (13% reduction) in short-term mortality but no difference in longer-term mortality (1).

Nevertheless, these studies do show a reduction in nonfatal reinfarction and nonfatal cardiac arrests. Furthermore, there are modest benefits of early β-blocker therapy after use of recombinant tissue-type plasminogen activator (2), although, again, total mortality is not reduced. β-blocker therapy should certainly be initiated during hospitalization after myocardial infarction for all patients without contraindications. Although the evidence is not compelling, it appears that the benefits are increased if treatment is begun sooner rather than later.

Paul Wangenheim, MD
Consultants in CardiologyMillburn, New Jersey, USA

Paul Wangenheim, MD
Consultants in Cardiology
Millburn, New Jersey, USA


1. Yusuf S, Sleight P, Held P, McMahon S. Circulation. 1990;82(Suppl 2):II-117-34.

2. Roberts R, Rogers WJ, Mueller HS, et al. Circulation. 1991;83:422-37.