The authors report a positron emission tomography (PET) study on humans with parallel exploration of the dose-dependent effects of an intravenous (propofol) and a volatile (sevoflurane) anesthetic agent on regional cerebral blood flow (rCBF) using quantitative and relative (Statistical Parametric Mapping [SPM]) analysis. Using H(2)(15)O, rCBF was assessed in 16 healthy (American Society of Anesthesiologists [ASA] physical status I) volunteers awake and at three escalating drug concentrations: 1, 1.5, and 2 MAC/EC(50), or specifically, at either 2, 3, and 4% end-tidal sevoflurane (n = 8), or 6, 9, and 12 microg/ml plasma concentration of propofol (n = 8). Rocuronium was used for muscle relaxation. Both drugs decreased the bispectral index and blood pressure dose-dependently. Comparison between adjacent levels showed that sevoflurane initially (0 vs. 1 MAC) reduced absolute rCBF by 36-53% in all areas, then (1 vs. 1.5 MAC) increased rCBF in the frontal cortex, thalamus, and cerebellum (7-16%), and finally (1.5 vs. 2 MAC) caused a dual effect with a 23% frontal reduction and a 38% cerebellar increase. In the propofol group, flow was also initially reduced by 62-70%, with minor further effects. In the SPM analysis of the "awake to 1 MAC/EC(50)" step, both anesthetic agents reduced relative rCBF in the cuneus, precuneus, posterior limbic system, and the thalamus or midbrain; additionally, propofol reduced relative rCBF in the parietal and frontal cortices. Both anesthetic agents caused a global reduction of rCBF (propofol > sevoflurane) at the 1 MAC/EC(50) level. The effect was maintained at higher propofol concentrations, whereas 2 MAC sevoflurane caused noticeable flow redistribution. Despite the marked global changes, SPM analysis enabled detailed localization of regions with the greatest relative decreases.
