Reteplase, a K2P deletion mutant of tissue plasminogen activator, is a novel recombinant thrombolytic agent, designed primarily for bolus thrombolysis in acute myocardial infarction. To avoid early reocclusion, which had been substantial after single-bolus therapy, the concept of double-bolus thrombolysis was developed and evaluated in recently published clinical trials. In this study, the pharmacokinetic profiles of double-bolus thrombolysis with 10 + 10 or 10 + 5 units (U) of reteplase at 30 min intervals were compared to that of a single 15 U bolus and the relative impacts on the hemostatic system were investigated. By splitting the dose of 15 U of reteplase in boluses of 10 + 5 U, median plasma antigen and activity levels above 700 ng/ml and 250 aU/ml (activity units) were prolonged by 1.5 min; all other pharmacokinetic properties and the pharmacodynamic variables investigated were similar in the 15 and 10 + 5 U groups. By doubling the dose of the second bolus to 10 U, a further prolongation of 45 min and 35 min was observed; median reteplase activity peaks were distinctly higher than relative antigen peaks after the second 10 U bolus, probably indicating that residual inhibitory activity against the plasminogen activator was overcome by the high activator dose, corresponding well with the higher patency rates found for this dosage in earlier clinical studies. In the 10 + 10 U group, more distinct alterations in hemostatic variables were observed, with increased fibrinogen consumption and split-product production resulting in a more pronounced fibrinolytic state. In accordance with previous studies, we found plasma alpha- and beta-half-lives of reteplase to be 4 and 19 times longer than those of alteplase, No deaths, strokes, major bleedings or overt allergic reactions were observed. As an important limitation of our study, it has to be stated that our own clinical findings, particularly the patency rates, disharmonize with expected results and results from larger trials investigating clinical end-points. Reasons for this could have included the small sample size of our study, relatively long times to treatment and an increased degree of thrombin activation prior to thrombolysis in the groups with low patency rates.