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


Myocardial infarction: a cumulative meta-analysis

ACP J Club. 1992 Nov-Dec;117:74. doi:10.7326/ACPJC-1992-117-3-074

Source Citation

Lau J, Antman EM, Jimenez-Silva J, et al. Cumulative meta-analysis of therapeutic trials for myocardial infarction. N Engl J Med. Jul



To evaluate by cumulative meta-analysis the effectiveness of 15 treatments and preventive measures for acute myocardial infarction.

Data sources

All published clinical trials of treatment and secondary prevention were assembled using MEDLINE and the references of published articles.

Study selection

Only randomized controlled trials of the treatment of myocardial infarction in which the outcome was all-cause or cardiovascular mortality were included. At least 1 meta-analysis of treatments selected for review had to have been previously published.

Data extraction

The meta-analysis of each treatment effect was updated after each new trial was published. The Mantel-Haenszel fixed-effects method was used to calculate odds ratios.

Main results

Reductions in mortality after myocardial infarction occurred with intravenous magnesium (odds ratio [OR], 0.44; 95% CI, 0.27 to 0.71), intravenous vasodilators (OR, 0.57; CI, 0.41 to 0.79), intravenous thrombolytic agents (OR, 0.75; CI, 0.71 to 0.79), aspirin (OR, 0.77; CI, 0.70 to 0.84), anticoagulants (OR, 0.78; CI, 0.65 to 0.92), and oral and intravenous β-blockers (OR, 0.88; CI, 0.80 to 0.98). Trends in favor of placebo were observed for calcium-channel blockers (OR, 1.12; CI, 0.92 to 1.37) and prophylactic lidocaine (OR, 1.15; CI, 0.86 to 1.55). Reductions in long-term mortality associated with secondary prevention after myocardial infarction occurred with β-blockers (OR, 0.81; CI, 0.73 to 0.89), anticoagulants (OR, 0.78; CI, 0.67 to 0.90), cholesterol-lowering measures (diet, drugs, and ileal-bypass surgery) (OR, 0.86; CI, 0.79 to 0.94), rehabilitation programs (OR, 0.80; CI, 0.67 to 0.95), and antiplatelet agents (OR, 0.90; CI, 0.82 to 1.00). An adverse effect was observed for class I antiarrhythmic drugs (OR, 1.28; CI, 1.02 to 1.61), and no benefit was shown for calcium-channel blockers (OR, 1.01; CI, 0.90 to 1.12).


Intravenous magnesium, vasodilators and thrombolytic agents, and aspirin, anticoagulants, and oral and intravenous β-blockers are effective in reducing all-cause mortality in the treatment of acute myocardial infarction. Anticoagulants, antiplatelet agents, rehabilitation programs, β-blockers, and cholesterol-lowering measures are effective in reducing long-term mortality produced by secondary prevention. Class I antiarrhythmic drugs have an adverse effect.

Source of funding: Agency for Health Care Policy and Research.

For article reprint: Dr. T.C. Chalmers, Technology Assessment Group, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115.


Few topics in medicine have been the subject of as many clinical trials as has acute myocardial infarction. Consequently, the efficacy of several treatments and preventive measures has been clearly documented. The widespread dissemination of this information has, however, been difficult, and many patients have not received the potential benefits. The data summarized by Lau and colleagues have been previously published, but never as efficiently as in this article. The major contribution of this paper, besides summarizing the data for the clinician, is to show the potential ability of cumulative meta-analysis (defined as the performance of an updated meta-analysis every time a new trial is published) to more quickly determine clinical efficacy, plan future trials, and make clinical therapeutic recommendations. Many possible methodologic and clinical issues complicate this new meta-analytic technique, however, and they are reviewed in this article.

Deciding which of these interventions to use for an individual patient requires clinical skill and a clear understanding of the indications, contraindications, and side effects. Aspirin can be given immediately to almost all patients in the emergency room, but only a minority qualify for thrombolytic therapy or β-blocking agents. Heparin can be given to most patients, but may not be necessary if streptokinase or anisoylated plasminogen streptokinase activator complex is administered. Magnesium and nitroglycerin appear promising, but the results of 2 large international trials that should be published next year will clarify the treatment benefit in relation to other effective treatments and test the predictive ability of meta-analysis with relatively small sample sizes. Lidocaine and calcium antagonists should not be routinely administered.

Aspirin and anticoagulants decrease subsequent risk, but it is not clear whether the effects are additive. Cholesterol-lowering agents and β-blockers are worthwhile and so are smoking cessation and blood pressure control. The empiric use of class I antiarrhythmic agents to suppress premature ventricular complexes is contraindicated.

Eric R. Bates, MD
The University of Michigan Medical Center Ann Arbor, Michigan