OBJECTIVE To investigate whether plaque rupture and thrombosis have a role in silent ischaemia as well as in unstable angina. DESIGN Prospective analysis of the results of haemostatic diagnostic tests at the moment of developing silent ischaemia at rest. SETTING Coronary care unit. PATIENTS 22 patients with acute myocardial infarction, 12 patients with symptomatic angina (unstable angina), and 10 normal volunteers (control group). INTERVENTIONS Continuous cardiac monitoring detected 15 asymptomatic episodes (silent ischaemia) in 6 patients with unstable angina. Blood samples were obtained at admission and when an asymptomatic alteration was detected and 10 minutes later. MAIN OUTCOME MEASURES Comparisons of concentrations of tissue plasminogen activator, urokinase type plasminogen activator, tissue plasminogen activator inhibitor-1, cross-linked fibrin degradation products, von Willebrand factor, and thrombin-antithrombin III complexes in patients and controls at admission; same comparisons in patients with silent ischaemia at the start of an episode and 10 minutes later. RESULTS Tissue plasminogen activator concentrations were raised at admission in patients with acute myocardial infarction (mean (SD) 14.2 (6) ng/ml) and in patients with unstable angina (10.1 (2.5) ng/ml) in comparison with controls (5.1 (2.7) ng/ml, p < 0.01 and < 0.05 respectively). There was no differences between the two groups of patients, however. Similar results were observed at the start of a silent ischaemic episode (9.8 (1.9) ng/ml) and 10 minutes later (10.5 (2.9) ng/ml) compared with controls (p < 0.05). Tissue plasminogen activator inhibitor-1 concentrations were raised in patients with acute myocardial infarction (45.1 (15) ng/ml) compared with volunteers (20.6 (16) ng/ml, p < 0.01). In patients with silent ischaemia tissue plasminogen activator inhibitor-1 concentrations were slightly but not significantly increased. Concentrations of cross-linked fibrin degradation products (D dimer) increased during unstable angina (2150 (350) ng/ml) and silent ischaemia (2270 (450) ng/ml) compared with the concentrations in volunteers (340 (80) ng/ml) and patients with acute myocardial infarction (310 (120) ng/ml; p < 0.01). CONCLUSIONS The results suggest that thrombosis mediates the pathophysiological mechanisms of silent ischaemia and unstable angina.