With technological assistance, approaches to acute stroke trials may be improved. Dose selection, eligibility criteria, control of confounding factors and endpoints can each be optimised. Adaptive randomisation techniques and Bayesian approaches can assist not only dose selection, but also balance of prognostic factors between treatment groups. Magnetic resonance imaging can facilitate enrichment of the trial population with potential responders. Utilisation of existing trial databases can enhance statistical approaches to outcome assessment. Confounding imbalance may also be limited by controlling or adjusting for concomitant conditions and their management. Recent trials have not yet identified a neuroprotectant, but have greatly assisted the prospects of success. Copyright 2001 S. Karger AG, Basel