Defining the risk to benefit ratio for individual patients remains impossible

Treating acute stroke is one of medicine's most difficult challenges as this millennium ends. It remains handicapped by deep rooted ignorance, fatalism, nihilism, and scepticism. This contrasts with the well accepted management of acute myocardial infarction, although the treatment of “heart attacks,” as for “brain attacks,” relies more on general and intensive care, including thrombolysis, than on drugs that specifically protect myocardial cells.

The most encouraging prospects for treating acute stroke include protecting ischaemic brain tissue (neuroprotection) and recanalising the occluded artery (thrombolysis).2 Interest in thrombolysis in acute stroke has been emphasised recently by several randomised trials.3 Nearly 3000 patients with acute stroke were studied in more than 10 trials. Meta-analysis of these data shows that intravenous thrombolysis was associated with a 20% reduction in the odds ratio for poor outcome (death or severe disability) but also with a 44% increase in the odds ratio of death—mainly due to a more than threefold increase in brain haemorrhage.4

The only positive trial of thrombolysis is that organised by the National Institute of Neurological Disorders and Stroke,5 which included 624 patients randomised to receive either intravenous placebo or alteplase (0.9 mg/kg to a maximum of 90 mg) given as a 10% bolus followed by an infusion lasting 60 minutes within three hours of onset of ischaemic stroke. There was no significant difference in neurological …