Ischaemic stroke is a leading cause of death and long-lasting disability. Several neuroprotective drugs have been developed that have the potential to limit ischaemic brain damage and improve outcome for patients. While promising results with these drugs have been achieved in animal stroke models, all Phase III trials conducted so far indicate that these drugs have failed to live up to their promise. Despite the limits of animal models, which cannot mimic the clinical situation, the disappointing results of neuroprotective trials might largely be due to methodological problems, Future trials with neuroprotective drugs should be performed in stroke (care) units, after sufficient information - regarding therapeutic time window, dosage, duration of therapy and safety has been gathered from pilot studies, and a better selection of target patients has been made. Much of this information can now be obtained by techniques that visualize the penumbra, such as combined diffusion-weighted and perfusion MRI,Consideration should also be given to clinical trials with well-designed combinations of treatments.