Sevoflurane, a volatile anesthetic with different effect on central nervous system from isoflurane, is commonly used in the human patient. Functional MRI studies in children also often use sevoflurane. However, in contrast to isoflurane, there is less evidence showing effect of the sevoflurane on blood oxygenation level dependent (BOLD) response using animal model. Here, we used 9.4 T high-field MRI system to assess the effect of sevoflurane on BOLD signals in rats. Rats were artificially ventilated to maintain the respiration and body temperature was maintained at 36.0-36.5 °C throughout the experiment. First, high resolution gradient echo echo-planar imaging (GE-EPI) images (100 x 100 x 400 μm/voxel, repetition time = 3 s, 20 averages) were obtained to compare the BOLD signal intensities in the cerebral cortex under 2.0-5.0% sevoflurane with those under 3.0-5.0 % isoflurane. The high concentration of isoflurane increased BOLD signals more intensely in large blood vessels rather than those in tissues sections including the capillary. While sevoflurane increased BOLD signals tissue sections as well as blood vessels. Second, to assess the effect of sevoflurane on hemodynamic response in the primary somatosensory cortex, the hindpaw electrical stimuli were used. The BOLD images were obtained using GE-EPI (300 x 300 x 800 μm/voxel, repetition time = 1 s) and we analyzed hemodynamic response in the upper layer or lower layer of primary somatosensory cortex. The BOLD response to electrical stimuli in the higher layer under 4.0 % sevoflurane anesthesia showed earlier return-to-baseline and the larger peak than 3.0% sevoflurane anesthesia. Furthermore, the peak in the upper layer was significantly larger than lower layer at all doses. These results indicate that sevoflurane modulates the BOLD signals with the different way from isoflurane and changes the hemodynamic response dose-dependently in the primary somatosensory cortex.