Reflection & Reaction IMAGES and FAST-MAG: magnesium for acute ischaemic stroke

2 Reasons for the failure of these therapies, such as extended time to administration, stroke subtype hetero- geneity, and compromised perfusion inhibiting drug delivery to ischaemic tissue, have been reviewed by a roundtable of academic and industry experts who have produced recommendations for advancing the design and development of treatments for acute stroke. 3-5 the treatment was shown to be safe and feasible. Field- initiated infusions were given substantially sooner than in historic controls (26 vs 139 mins). In this new approach to early stroke treatment, the physician-investigator obtained consent via cell phone while the paramedics went about their field-care duties uninterrupted. The FAST-MAG trialists have been awarded funding by the US National Institutes of Health to do a phase III trial of magnesium admin- istration in the field for acute stroke. The adage that "time is brain" will be tested as field administration of this drug can be accomplished within 1-2 h, or less, of stroke onset. Previous neuroprotective studies have not been able to consistently achieve such a short time interval as system delays have impeded early administration of acute stroke therapy. The FAST-MAG trialists will need to heed the lessons learned from the IMAGES trial: median time to treatment was 7 h with only 3% of participants receiving active therapy within 3 h, which may explain the loss of effectiveness of magnesium therapy. This time-delay factor should not be a barrier for the FAST-MAG invest- igators. There was an unexpected slight excess mortality in the magnesium treatment group in IMAGES. Was this a chance event or is there a hidden toxicity of magnesium in patients with acute stroke that we have yet to define? Careful monitoring of mortality and blood pressure lowering effects of magnesium will be important in the FAST-MAG trial. The surprise finding of efficacy in non-cortical strokes in IMAGES raises the possibility that the effect of magnesium could be targeted preferentially to white matter. Have the FAST-MAG trialists considered this contingency and, if so, will the FAST- MAG trial be statistically powered to include enough non-cortical strokes to study this effect? Or are we merely being misled again by the uncertainties of subgroup analysis and play of chance? The white-matter component of ischaemic stroke has generally been overlooked and has not been a primary target of neuroprotective therapy. Are we closer to finding a safe and effective neuroprotectant for acute ischaemic stroke? The IMAGES investigators are to be congratulated for doing a rigorous trial in a difficult disorder. We believe that in our quest to uncover a neuroprotective therapy that is effective for acute ischaemic stroke, early administration of the therapy will be critical. We also believe that the FAST-MAG investigators are on the right path. We will soon learn if such early administration of magnesium is both safe and effective.