Glucose consumption is severely depressed in the ischemic core, whereas it is maintained or even increased in penumbral regions during ischemia. Conversely, glucose utilization is severely reduced early after reperfusion in spite that glucose and oxygen are available. Experimental studies suggest that glucose hypometabolism might be an early predictor of brain infarction. However, the relationship between early glucose hypometabolism with later development of infarction remains to be further studied in the same subjects. Here, glucose consumption was assessed in vivo by positron emission tomography (PET) with (18)F-fluorodeoxyglucose ((18)F-FDG) in a rat model of ischemia/reperfusion. Perfusion was evaluated by PET with (13)NH(3) during and after 2-hour (h) middle cerebral artery occlusion, and (18)F-FDG was given after 2h of reperfusion. Brain infarction was evaluated at 24h. Mitochondrial oxygen consumption was examined ex vivo using a biochemical method. Cortical (18)F-FDG uptake was reduced by 45% and 25% in the ischemic core and periphery, respectively. However, substantial alteration of mitochondrial respiration was not apparent until 24h, suggesting that mitochondria retained the ability to consume oxygen early after reperfusion. These results show reduced glucose use at early reperfusion in regions that will later develop infarction and, to a lesser extent, in adjacent regions. Depressed glucose metabolism in the ischemic core might be attributable to reduced metabolic requirement due to irreversible cellular injury. However, reduced glucose metabolism in peripheral regions suggests either an impairment of glycolysis or reduced glucose demand. Thus, our study supports that glycolytic depression early after reperfusion is not always related to subsequent development of infarction.