Thrombolytic therapy salvages jeopardized myocardium and preserves left ventricular function. Therefore, a beneficial effect on exercise tolerance and haemodynamic adaptation to exercise can be anticipated. In the present study the results of bicycle exercise tests, performed at 10 to 14 days and at more than 1 year after infarction, were compared between patients randomized double blindly to recombinant tissue plasminogen activator (rt-PA) or placebo. At 2 weeks, the maximal heart rate (140 vs 130 beats.min-1 P = 0.017), systolic blood pressure (169 vs 161 mmHg, P = 0.018) and pressure-rate product (22907 vs 20865 beats.min-1.mmHg, P = 0.0025) were higher in rt-PA treated patients (n = 145) than in controls (n = 142). At a mean follow-up of 16 months, rt-PA patients (n = 126) performed a higher total and peak work load (733 vs 600 Watts.min, P = 0.027 and 133 vs 100 Watts, P = -0.036, respectively) than control patients (n = 122). During both exercise tests the incidence of clinical or electrocardiographic signs of ischaemia was similar. Multiple regression analysis indicated that age and end-systolic volume were predictors of peak pressure-rate product at 2 weeks (R2 = 0.11, P = 0.0001). Age, sex and left ventricular ejection fraction at discharge were independent predictors of peak work load at 2 weeks (R2 = 0.32, P = 0.0001) and the square root of peak work load at 16 months (R2 = 0.39, P = 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)