Stroke is the third cause of death in industrialised countries and a major cause of disability. It results either from an intracerebral haemorrhage or more commonly from an ischaemia, a lack in blood supply to (part of) the brain. In the centre of an ischaemic lesion, the ischaemic core, the neurons die rapidly. The core is surrounded by the ischae mic penumbra, an area where neurons have suffered but can potentially be rescued. In human stroke, the only proven way to rescue these neurons is by intravenous administration of a clot-lysing agent, recombinant tissue plasminogen activator (rTPA) within 3 hours of the onset of symptoms.Such a narrow time-window and the po tential complications of the treatment (sympto matic bleeding) in not carefully selected patients result in a very small percentage of patients which receive and benefit from this treatment. Experi ments in animal models have allowed the study of complex mechanisms occurring in the penumbra, including excitotoxicity, inflammation and apop tosis but further efforts still need to be made in this area to obtain a pharmacological tool which would improve the outcome after stroke by signifi cantly reducing neuronal death in the penumbra zone. From a clinical point of view, in the case of a stroke, the key procedure at present state is to ensure a very rapid hospitalisation and treatment by an experienced stroke team. The following article, summarising a public conference at the 2002 Brain Awareness Week, gives a short over view of this field.