The safety of thrombolytic therapy of acute myocardial infarction could be improved if a method were developed to dissolve fresh occlusive coronary thrombus without simultaneously dissolving hemostatic thrombi outside the coronary arteries. This study is based on the assumption that, in a patient with evolving acute myocardial infarction, hemostatic thrombi are likely to be older than the thrombus responsible for occlusion of the coronary artery. It explored whether the relative rates of lysis of fresh and old thrombi could be influenced by the rapidity of recombinant tissue-type plasminogen activator (rt-PA) administration. In each of 17 dogs, two 1 h and two 24 h old thrombi were produced by inserting copper coils into both jugular and both femoral veins. After 24 h and 1 h, respectively, the coils with the thrombi were removed, weighed and inserted into the adjacent carotid and femoral arteries. A 1 mg/kg body weight dose of rt-PA was given either over 180 or over 30 min. The coils were removed and weights of the residual thrombi determined at the end of the 180 min infusion (Group I), at the end of the 30 min infusion (Group IIA) and 45 min after the 30 min infusion (Group IIB). The 24 h old thrombi were lysed significantly less than the 1 h old thrombi in all three experimental groups: 53.9 +/- 4.8% (mean +/- SE) versus 86.1 +/- 2.5% in Group I (p less than 0.001), 16.6 +/- 3.5% versus 65.2 +/- 6.0% in Group IIA (p less than 0.001) and 21.6 +/- 5.4% versus 91.7 +/- 1.7% in Group IIB (p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)