Estimates are that fewer than one half of patients who might benefit from beta-adrenergic blockers after a myocardial infarction (MI) are prescribed these drugs. Suggested reasons for the underuse of these agents after an MI include concerns about the validity of clinical trials and uncertainty about the relative merits of the different beta blockers. Freemantle and colleagues conducted a systematic review of 82 randomized controlled trials of beta blockers for short- and long-term prevention of death and reinfarction after an MI.

Multiple databases, reference lists, reviews and other sources were used to identify relevant studies conducted since 1966. Short-term treatment (i.e., up to six weeks after MI) was examined in 51 trials and long-term treatment (i.e., from six to 48 months) was examined in 31 trials. Data were obtained from each study concerning the number of patients randomized to treatment or control status, the specific beta-blocking agent used and its dosage and route of administration, the duration of treatment, the loss to follow-up and the level of blinding sustained. The principle outcome measures were death and reinfarction. Several specialized statistical techniques were used to allow for heterogeneity in the trials and random factors.

Pooled data from the 51 trials of short-term treatment following an acute MI revealed 3,062 deaths among 29,260 patients, constituting a mortality rate of 10.5 percent during the six weeks after infarction. This mortality rate translates to a reduction in deaths of 0.4 per 100 patients; 250 patients would require treatment to prevent one death. The reduction in mortality associated with short-term treatment was not statistically significant.

The 31 trials of long-term beta-blocker therapy included 24,974 patients; 2,415 (9.7 percent) of these patients died. This translates into 1.2 deaths per 100 patients, or a requirement to treat 84 patients for one year to prevent one death.

For nonfatal reinfarctions, long-term beta-blocker therapy was calculated to result in an annual reduction of 0.9 events per 100 patients, or a requirement to treat 107 patients per year to prevent one nonfatal reinfarction.

When the data were examined according to classification of beta blockers by cardio-selectivity, a nonsignificant trend was found toward a reduction of benefit in long-term trials. Similarly, drugs with intrinsic sympathomimetic activity were found to have reduced benefit at a level approaching statistical significance.

The authors conclude that long-term use of beta blockers following MI reduces morbidity and mortality. Most of the evidence of benefit is for metoprolol, propranolol and timolol. The authors found that atenolol has been inadequately evaluated for long-term use. The pooled data indicate that benefit may be reduced when cardioselective agents or those with intrinsic sympathomimetic activity are used.