Failure to adequately anticoagulate the blood of patients receiving recombinant tissue plasminogen activator (TPA) leads to greater rates of rethrombosis. In a multicentered, randomized trial in 51 patients we compared the ability to achieve and maintain therapeutic anticoagulation by use of computer‐assisted heparin therapy or the GUSTO (Global Utilization of Streptokinase and TPA for Occluded Coronary Arteries) heparin nomogram guidelines in patients with myocardial infarction treated with recombinant TPA. Heparin therapy was initiated with either computer‐generated starting doses or GUSTO guideline starting doses. Activated partial thromboplastin times were measured every 6 to 8 hours for the first 24 hours. The therapeutic range used in this trial was 1.5 to 2.5 times the patient's baseline activated partial thromboplastin time (APTT). Ninety‐four percent of the APTT ratios in the computer group were equal to or greater than 1.5 in the first 24 hours compared with 78% in the GUSTO group (p < 0.009). No significant difference in bleeding was found (7.7% for GUSTO; 4.2% for computer). Incremental time‐dependent changes in heparin dose were found (day 1, 1110 ± 243 units/hr, APTT ratio = 2.5 ± 1.4; day 3, 1380 ± 374 units/hr, APTT ratio, 1.9 ± 0.4). Computer‐assisted heparin therapy TPA results in superior anticoagulation accuracy compared with the GUSTO guidelines. In addition, the pharmacodynamic response to heparin changes in the 2 to 3 days after administration of TPA, leading to greater heparin requirements.