To evaluate the feasibility and cost savings of hospital discharge three days after acute myocardial infarction, we screened 507 consecutive patients prospectively for clinical complications and exercise-test performance. Of 179 patients whose condition was classified as uncomplicated (no angina, heart failure, or arrhythmia 72 hours after admission), 126 underwent early exercise testing and 90 had no provocable myocardial ischemia. Eighty of these patients were randomly assigned to early (day 3) or conventional (days 7 to 10) hospital discharge. Seventy-six of them had received coronary reperfusion therapy (thrombolysis, angioplasty, or both). At six months of follow-up, there were no deaths or new ventricular aneurysms, and the early-discharge and conventional-discharge groups had similar numbers of hospital readmissions (6 and 10), reinfarctions (none and 5), and patients with angina (3 and 8). In the early-discharge group, 25 of 29 previously employed patients returned to work 40.7 +/- 21.9 days (mean +/- SD) after admission, as compared with 25 of 27 patients in the conventional-discharge group, who returned to work after a mean of 56.9 +/- 30.3 days (P = 0.054). The mean cumulative hospital and professional charges were $12,546 +/- 3,034 in the early-discharge group, as compared with $17,868 +/- 3,688 in the conventional-discharge group (P less than 0.0001). In carefully selected patients with uncomplicated myocardial infarction, hospital discharge after three days is feasible and leads to a substantial reduction in hospital charges. Before this strategy can be widely recommended, however, its safety must be confirmed in larger prospective clinical trials.