Both sevoflurane and isoflurane are used in moderate concentrations in neuroanesthesia practice. The limiting factors for using higher concentrations of inhalational anesthetics in patients undergoing neurosurgery are the agents' effects on cerebral blood flow (CBF) and cerebral blood volume (CBV). In particular, an increase in CBV, which is a key determinant of intracranial pressure, may add to the neurosurgical patient's perioperative risk. To compare the effects of a subanesthetic concentration (0.4 minimum alveolar concentration) of sevoflurane or isoflurane on regional CBF (rCBF), regional CBV (rCBV) and regional mean transit time (rMTT), contrast-enhanced magnetic resonance imaging perfusion measurements were made in spontaneously breathing human volunteers. Absolute changes in rCBF, regional CBV, and rMTT during administration of either drug in regions of interest outlined bilaterally in white and grey matter were nonparametrically (Mann-Whitney test) analyzed. Sevoflurane increased rCBF in practically all regions (absolute change, 4.44 +/- 2.87 to 61.54 +/- 2.39 mL/100g per minute) more than isoflurane did (absolute change, 12.91 +/- 2.52 to 52.67 +/- 3.32 mL/100g per minute), which decreased frontal, parietal, and white matter rCBF (absolute change, -1.12 +/- 0.59 to -14.69 +/- 3.03 mL/100g per minute). Regional CBV was higher in most regions during isoflurane administration (absolute change, 0.75 +/- 0.03 to 4.92 +/- 0.16 mL/100g) than during sevoflurane administration (absolute change, 0.05 +/- 0.14 to 3.57 +/- 0.14 mL/100g). Regional mean transit time was decreased by sevoflurane (absolute change, -0.18 +/- 0.05 to -0.60 +/- 0.04 s) but increased by isoflurane (absolute change, 0.19 +/- 0.03 to 0.69 +/- 0.04 s). In summary, regional CBV was significantly lower during sevoflurane than during isoflurane administration, although sevoflurane increased rCBF more than isoflurane, which even decreased rCBF in some regions. For sevoflurane and, even more pronouncedly, for isoflurane, the observed changes in cerebral hemodynamics cannot be explained by vasodilatation alone.