Acute myocardial infarction is triggered by coronary artery occlusion that may be recanalized by thrombolytic therapy with a success rate of up to 75% only. The resistance of coronary artery occlusion to thrombolysis may either be due to obstruction of the lumen by a nonthrombotic mechanism or by intrinsic resistance of thrombus to dissolution. Coronary arterial thrombi are composed of platelet-rich and erythrocyte-rich material in variable proportions. To evaluate the relative sensitivity of these thrombus components to thrombolysis, we have used two femoral arterial thrombosis models in the rabbit, consisting of erythrocyte-rich clot produced by injecting whole blood and thrombin in an isolated segment and of platelet-rich thrombus spontaneously formed on an everted (inside out) femoral arterial segment. Intravenous infusion of recombinant tissue-type plasminogen activator (rt-PA) at a rate of 30 micrograms/kg/min consistently reperfused arteries occluded with erythrocyte-rich clot (six of six animals compared with zero of six placebo-treated animals, p = 0.002), whereas infusion of 30 or 100 micrograms/kg/min was significantly less efficient for reperfusion of everted segments occluded with platelet-rich material (only four of 12 animals, p = 0.01). Intra-arterial infusion proximal to the occlusion, at a rate of 20 micrograms/kg/min reperfused six of seven rabbits with erythrocyte-rich clots but only one of seven rabbits with occluded everted segments (p = 0.03). A dose of 100 micrograms/kg/min was necessary to reperfuse platelet-rich occlusions in five of six rabbits.(ABSTRACT TRUNCATED AT 250 WORDS)