BACKGROUND Reocclusion after thrombolysis diminishes the benefits of early reperfusion after acute myocardial infarction (AMI). No clinical or laboratory variables have been identified as predictors for reocclusion yet. RESULTS To evaluate hemostatic variables as potential risk determinants platelet aggregation (PA, representing platelet activity), thrombin/antithrombin complexes (TAT, representing thrombin generation), and plasminogen activator inhibitor type 1 (PAI-1, representing endogenous fibrinolysis) were determined in 31 patients with AMI at 0, 1, 2. and 12 h after the start of thrombolysis as well as at hospital discharge. Reocclusion (defined as reinfarction or angiographically confirmed, clinically silent coronary reocclusion) occurred in 5 patients within 5-14 days and in 8 patients within 1 year. TAT plasma concentrations were lower in patients with reocclusion than in those without (9.9+/-5.7 vs. 22.9+/-22.2 ng/ml at 2 h, 6.5+/-3.1 vs. 1 1.2+/-6.4 ng/ml at 12 h, means+/-SD, p <0.05 each). Neither concentration nor activity of PAI-1 in plasma differed between both patient groups. However, both slope and maximum of PA (induced by 2 micromol/l ADP) were augmented in patients with reocclusion (slope: 39.4+/-1.7 vs. 32.5+/-7.4 at 2 h, p <0.001; 42.6+/-2.6 vs. 36.6+/-8.9 at 12 h, p <0.01). Results were independent of the thrombolytic agent used (alteplase or reteplase). A PA slope at 2 h higher than the average slope before thrombolysis (37.2+/-5.7) could be identified as best predictor for early (within 5-14 d, p=0.017, sensitivity 1.00, specificity 0.69) and late reocclusion (within 1 y, p=0.009, 0.88 and 0.74, respectively). CONCLUSIONS Increased PA following coronary thrombolysis appears to be associated with early and late reocclusion. This marker could be useful in identifying patients who may benefit from more aggressive antiplatelet (such as GP IIb/IIIa receptor antagonists), interventional, or both strategies.