Quantitation of perfusion defect size using tomographic imaging with technetium-99m-hexakis-2-methoxy isobutyl isonitrile was performed at the time of hospital discharge in 32 patients with a first myocardial infarction who underwent successful coronary reperfusion within 8 h of the onset of chest pain. Reperfusion was accomplished with thrombolysis or primary coronary angioplasty. Radionuclide angiography was performed at discharge and 6 weeks later. There was a close correlation between perfusion defect size and values for ejection fraction and regional wall motion both at discharge (r = -0.80 and -0.75, respectively) and 6 weeks later (r = -0.81 and -0.81, respectively). There was no overall group difference in ejection fraction between the value at discharge and at 6 weeks; however, five patients had a significant increase (greater than or equal to 0.08) and six had a significant decrease (greater than or equal to 0.08) in ejection fraction. In patients with a significant increase at 6 weeks, ejection fraction was significantly lower at discharge than the value predicted from perfusion defect size (0.37 +/- 0.09 measured versus 0.47 +/- 0.13 predicted, p less than 0.05) and it improved at 6 weeks to near predicted values (0.51 +/- 0.07). In patients with a significant decrease at 6 weeks, ejection fraction was significantly higher at discharge than the value predicted from perfusion defect size (0.60 +/- 0.10 measured versus 0.50 +/- 0.10 predicted, p less than 0.05) and it decreased at 6 weeks to near predicted levels (0.51 +/- 0.09). Left ventricular ejection fraction at the time of hospital discharge is a potentially misleading index of the efficacy of reperfusion therapy for myocardial infarction. In a significant minority (34%) of patients this index does not accurately reflect perfusion defect size, apparently because of the effects of myocardial stunning and compensatory hyperkinesia.