BACKGROUND: Hemodilution after focal cerebral ischemia increases cerebral blood flow to ischemic brain tissue and reduces neurologic injury. With rare exceptions, most studies have reduced hematocrit (Hct) to no less than 30%. We studied the effect of moderate hemodilution (hematocrit 27%) on cerebral infarct volume after focal cerebral ischemia in rabbits. METHODS: Twenty rabbits were divided into a control group (n = 10) and a hemodilution group (n = 10). In the control group, cerebral infarction was induced by embolization of the middle cerebral artery using an autologous blood clot without hemodilution. In the hemodilution group (n = 10), hemodilution of around hematocrit 27% was achieved by exchanging arterial blood with 10% hydroxyethyl starch 1 hour before embolization of the middle cerebral artery in the hemodilution group. Seven hours after embolization, coronal brain slices were made with 2 mm thickness at 1 cm from the frontal pole and stained with 2% 2,3,5-triphenyltetrazolium chloride. The infarct volume was quantitated by image analysis of photography of the infarcted area. RESULTS: The infarct volume of the cerebral hemisphere (25.9 +/- 8.9%), subcortex (16.3 +/- 3.1%) in the hemodilution group was significantly smaller than in the control group (34.9 +/- 8.9%, 19.3 +/- 5.1%) (P<0.05), but, in the cortex, the difference of infarct volume is not statistically significant between the control group (23.5 +/- 11.9%) and the hemodilution group (15.6 +/- 2.7%). CONCLUSIONS: These results indicate that moderate hemodilution (hematocrit 27%) reduces neurologic injury after focal cerebral ischemia.