Sixty-six patients presenting with their first evolving transmural acute myocardial infarction (AMI) were randomized to receive either streptokinase (n = 41) or placebo therapies (n = 25) within 6 hours of the onset of chest pain. These patients then underwent supine rest, exercise and after-nitroglycerin radionuclide angiography 3 weeks after AMI. Nuclear magnetic resonance (NMR) imaging was performed at 3 weeks as a more direct estimate of AMI size. Although peak creatine kinase values were comparably elevated between groups (2,367 +/- 1,486 IU/liter for streptokinase vs 2,637 +/- 1,305 IU/liter for placebo), there was a significant reduction in NMR-measured AMI size in the streptokinase group (3 +/- 2% of left ventricular volume vs 10 +/- 4% in the placebo group, p less than 0.05). This occurred despite comparable resting (54 +/- 11 vs 47 +/- 10% and exercise (53 +/- 12 vs 49 +/- 11%) global ejection fractions. However, following nitroglycerin, there was an improvement in global ejection fraction in the streptokinase-treated group that was not observed with placebo (61 +/- 13 vs 48 +/- 10%, p less than 0.05). A similar pattern was also observed with regional functional analysis. Thus, streptokinase therapy leads to a significant reduction in NMR-measured AMI size and to a greater degree of reversible left ventricular dysfunction.