In the rat middle cerebral artery occlusion (MCAO) model of embolic stroke, calcium antagonists, as well as competitive and noncompetitive N-methyl-D-aspartate (NMDA) antagonists have been studied extensively by various groups. However, there are divergent conclusions concerning their efficiencies. These discrepancies may be due to the different experimental conditions and/or rat strains used. In this study we have assessed in five commonly used rat strains, under identical experimental conditions, the ability of one representative of each of these drug classes to reduce the infarct size in this model. Drugs or corresponding vehicle were injected immediately after permanent unilateral MCAO and the infarct volume was determined 48 hr later by magnetic resonance imaging. In vehicle-treated controls, the total infarct volume was strain-dependent (range, 70-200 microliters), primarily due to variations in the cortex. The extent of total infarct volume reduction (up to 65%) elicited by the three drugs also was strain-dependent, the main effect and variation being observed in the cortex (range, 0-83% reduction). No difference in efficiency was seen between the competitive and noncompetitive NMDA antagonist, whereas the calcium antagonist showed better results in all five strains. The potential to reduce the infarct size in the rat MCAO model with calcium or NMDA antagonists depends upon the strain used. The exact reasons and mechanisms are presently not understood. Such knowledge would certainly add to the confidence in this model and would allow more accurate predictions concerning the clinical efficacy of drugs in stroke trials.