Total and regional cerebral glucose metabolism (CMRgl) was measured by positron emission tomography with 2-( 18 F) fluoro-2-deoxy-D-glucose ( 18 FDG) in 20 term infants with hypoxic ischemic encephalopathy (HIE) after perinatal asphyxia. All infants had signs of perinatal distress, and 15 were severely acidotic at birth. Six infants developed mild HIE, twelve moderate HIE, and two severe HIE during their first days of life. The positron emission tomographic scans were performed at 4 -24 d of age (median, 11 d). One hour before scanning, 2-3.7 MBq/kg (54 -100 Ci/kg) 18 FDG was injected i.v. No sedation was used. Quantification of CMRgl was based on a new method employing the glucose metabolism of the erythrocytes, requiring only one blood sample. In all infants, the most metabolically active brain areas were the deep subcortical parts, thalamus, basal ganglia, and sensorimotor cortex. Frontal, temporal, and parietal cortex were less metabolically active in all infants. Total CMRgl was inversely correlated with the severity of HIE (p Ͻ 0.01). Six infants with mild HIE had a mean (range) CMRgl of 55.5 (37.7-100.8) mol·min -1 ·100 g -1 , 11 with moderate HIE had 26.6 (13.0 -65.1) mol·min -1 ·100 g -1 , and two with severe HIE had 10.4 and 15.0 mol·min -1 ·100 g -1 , respectively. Five of six infants who developed cerebral palsy had a mean (range) CMRgl of 18.1 (10.2-31.4) mol·min -1 ·100 g -1 compared with 41.5 (13.0 -100.8) mol·min -1 ·100 g -1 in the infants with no neurologic sequela at 2 y. We conclude that CMRgl measured during the subacute period after perinatal asphyxia in term infants is highly correlated with the severity of HIE and short-term outcome. Birth asphyxia among term infants remains a considerable problem in perinatal medicine with high rates of mortality and handicap (1-4). The clinical grading of HIE after perinatal asphyxia has become a useful prognostic tool in the term infant (5-7). Mild HIE usually results in a normal outcome and severe HIE in either death or CP. However, moderate HIE can result in either CP or a normal outcome (8). Early prediction of future handicap is important when selecting infants for early interventions, which may be possible in the near future (9, 10).Several techniques have been used to investigate the cerebral pathology related to perinatal asphyxia. Cerebral morphology has been evaluated with computed tomography (CT) (11) and magnetic resonance imaging (MRI) (12, 13). Methods capable of detecting functional disturbances, such as phosphorus ( 31 P) magnetic resonance spectroscopy (14) and proton ( 1 H) magnetic resonance spectroscopy (15), have been used to investigate changes in brain energy metabolism during the evolution of the secondary energy failure (16).PET, measuring either cerebral blood flow (17) or CMRgl, has been used in a few studies on newborn infants. The first PET studies with FDG in newborn infants did not allow any quantification, making only relative values possible (18,19