BACKGROUND: The precise mechanism by which sevoflurane exerts its effects in the human brain remains unknown. In the present study, we quantified the effects of sevoflurane on regional cerebral glucose metabolism (rGMR) in the human brain measured with positron emission tomography. METHODS: Eight volunteers underwent two dynamic 18F-fluorodeoxyglucose positron emission tomography (PET) scans. One scan assessed conscious-baseline metabolism and the other scan assessed metabolism during 1 minimum alveolar concentration (MAC) sevoflurane anaesthesia. Cardiovascular and respiratory parameters were monitored and bispectral index responses were registered. Statistical parametric maps and conventional regions of interest analysis were used to determine rGMR differences. RESULTS: All subjects were unconsciousness at 1.0 MAC sevoflurane. Cardiovascular and respiratory parameters were constant over time. In the awake state, rGMR ranged from 0.24 to 0.35 mumol/g/min in the selected regions. Compared with the conscious state, total GMR decreased 56% in sevoflurane anaesthesia. In white and grey matter, GMR was averaged 42% and 58% of normal, respectively. Sevoflurane reduced the absolute rGMR in all selected areas by 48-71% of the baseline (P< or = 0.01), with the most significant reductions in the lingual gyrus (71%), occipital lobe in general (68%) and thalamus (63%). No increases in rGMR were observed. CONCLUSIONS: Sevoflurane caused a global whole-brain metabolic reduction of GMR in all regions of the human brain, with the most marked metabolic suppression in the lingual gyrus, thalamus and occipital lobe.