This study aims to establish a mouse global cerebral ischemia model in which the physiological parameter measurements and neuronal injury evaluations are conducted in the same group of animals and to identify the effect of post-ischemic core temperature (CT) on the outcome of neuronal injury. Global ischemia was induced by 12-min bilateral common carotid artery occlusion followed by 7 days of reperfusion in C57BL/6 mice. Immediately after occlusion, mice were randomly assigned to be kept in environments of different temperatures [25 degrees C (room temperature, group 1), 33-34 degrees C for 2h (group 2), and 33-34 degrees C for 24h (group 3)] before being returned to their home cages. We found that in group 1, CT declined to approximately 32 degrees C after ischemia and then recovered at 24h post-ischemia; in group 2, CT remained at the pre-ischemia level during the first 2h, declined after the mice were returned to room temperature, and recovered at 24h post-ischemia; and in group 3, CT remained constant at the pre-ischemia level throughout the reperfusion period. The number of surviving neurons in a sector of the hippocampal CA1 region was significantly lower in all ischemic groups than in the sham controls, but the number was significantly higher in group 1 than that in groups 2 or 3 (P<0.05). We observed that CT declines initially but recovers spontaneously at 24h post-ischemia. Early post-ischemic hypothermia impacts the delayed neuronal injury, suggesting that tight temperature control immediately following ischemia is important to obtain the most reproducible neuronal damage in mouse models of cerebral global ischemia.