Brain metabolism relies on synchronized interaction between neurons and astrocytes [1]. Anaesthetics modulate basal neuronal activity, and therefore the so-called neurometabolic coupling, by acting on different molecular targets, such as γ-aminobutyric acid type A (GABA_A) receptors, in the case of α-chloralose [2] or thiopental [3] anaesthesia. The aim of this study was to characterize neuronal and astrocytic pathways of energy metabolism in the rat cortex under thiopental anesthesia in vivo by direct ¹³C MRS.Sprague Dawley rats (n=7, 267±23 g) under thiopental (50 mg/kg bolus followed by a continuous intravenous infusion rate of 90 mg/kg/h) anesthesia were stereotaxically fixed and positioned in a homebuilt holder. MRS experiments were performed on a 14.1 T/26 cm horizontal bore magnet with a homebuilt coil consisting of a ¹H quadrature surface coil combined to a ¹³C linearly polarized surface coil. After FAST(EST)MAP shimming, localized ¹H and ¹³C spectra (during infusion of [1,6-¹³C]glucose [1]) were acquired from a 94 μL volume in the cortex with STEAM and semi-adiabatic distortionless enhancement by polarization transfer (DEPT) combined with 3D-ISIS for ¹H localization [4], respectively. LCModel was used for analysis of both ¹H and ¹³C spectra [5]. The scaling of ¹³C fractional enrichment (FE) curves was based on MRS of brain extracts [1]. Data was fitted to a two-compartment model and variance of parameters was determined by Monte-Carlo analyses [6].The administration of [1,6-^₁₃C]glucose under thiopental anaesthesia increased plasma lactate concentration by 104% compared to experiments under α-chloralose anaesthesia [1,7]. After 4 hours of [1,6-¹³C]glucose infusion, plasma and brain lactate FE was increased by 36% and 33%, respectively, in the thiopental compared to α-chloralose anaesthesia. However, at the end of the experiment, FE of glutamate and glutamine C4, C3 and C2, and aspartate C3 and C2 was similar under both types of anaesthesia, as measured ex vivo from the brain extracts. Glucose transport analysis resulted in an apparent Michaelis constant (K_t) of 4.5±8.1 mM, in an apparent maximum transport (T_max) of 2.6±0.6 µmol/g/min and in a cerebral metabolic rate (CMR_glc) of 0.26±0.09 µmol/g/min. Under thiopental anaesthesia, the rate of glutamatergic neurotransmission i.e. the glutamate-glutamine cycle (V^_NT) was 0.06±0.01 µmol/g/min and glutamine synthetase (V_GS) was 0.12±0.02 µmol/g/min. The flux through the tricarboxylic acid cycle in astrocytes (V_TCA^g) and neurons (V_TCAⁿ) were 0.27±0.03 µmol/g/min and 0.52±0.02 µmol/g/min, respectively, resulting in a total cerebral metabolic rate of glucose (CMR_glc(ox)) of 0.42±0.02 µmol/g/min.Thiopental data from ¹³C MRS in vivo indicate glycolysis reduction as well as decreased CMR_glc(ox), although still substantial as indicated by similar amino acids FE at the end of the experiment. The mismatch between glycolytic flux and global mitochondrial oxidative metabolism, depicted by CMR_glc(ox), suggests that other substrates are substantially oxidized, probably lactate that was increased two-fold in the plasma of rats anaesthetised with thiopental, relative to experiments under α-chloralose anaesthesia [1,7].Preliminary metabolic flux analysis of brain metabolism under thiopental anaesthesia assessed in vivo using ¹³C MRS suggests that neuron-glia interactions are similar to those under α-chloralose anaesthesia. However, the reduction of glycolysis under thiopental anaesthesia implies a substantial utilisation of substrates other than glucose. Since lactate concentration was high, it may be the carbon source that complements glucose in sustaining cortical oxidative metabolism under thiopental anaesthesia.