Patients with intracranial haemorrhage may sometimes require antithrombotic drugs or be inadvertently given antithrombotic therapy. We systematically reviewed all published trials comparing any antithrombotic agent with control among patients with any form of intracranial haemorrhage. We extracted data on deaths, recurrent intracranial haemorrhage and functional outcome. There were 9 randomised trials of 5 different antithrombotic agents versus control in patients with subarachnoid haemorrhage (6 trials, n = 1,224) or with acute intracerebral haemorrhage (3 trials, n = 819). The overall odds ratio (OR) for death among patients with any intracranial haemorrhage given an antiplatelet agent (8 trials, 1,997 patients) was 0.85 (95% confidence interval, Cl, 0.63-1.15), and for recurrent intracranial haemorrhage it was 1.00 (95% Cl 0.73-1.37). The corresponding ORs for patients with intraparenchymal cerebral haemorrhage were 0.96 (0.62-1.5) and 1.02 (0.5-1.8), respectively, but 65% of these patients received only a few doses of antithrombotic treatment. The overall OR for death in patients with any intraparenchymal cerebral haemorrhage given heparin compared with control (3 trials, 819 patients, subcutaneous heparin) was 0.96 (95% Cl 0.38-2.40), and for recurrent intracranial haemorrhage it was 2.00 (95% Cl 0.86-4.70). There were no reliable data on the effects of antithrombotic agents on functional outcome. These scant data do not support reliable conclusions about the safety or otherwise of antithrombotic agents in patients with acute intracranial haemorrhage. Antithrombotic agents should be avoided where possible in patients with acute intracerebral haemorrhage. Copyright (C) 2002 S. Karger AG, Basel.