The fMRI-BOLD effect is one of the most widely applied methods to study brain function non-invasively in health and disease. However, the complex physiological basics, as interplay of CMRO₂, CBF and CBV changes, of the BOLD effect are not fully understood. Recent studies have also implicated that in some circumstances a mismatch between BOLD brain activation and brain activation displayed e.g. by CMRGlc PET techniques exists [1]. Aim of this study was, to investigate and compare brain activation as response to a whisker stimulus using BOLD-fMRI and [¹⁵O]O₂ to study CMRO₂ in the same animals during the same anesthesia session.Ten male SD rats (weight = 393±25 g) were used in this study. Animals were intubated and after surgery the anesthesia was switched from isoflurane to α-chloralose (i. v. 40 mg/kg bolus, 30 mg/kg/h infusion) and tubocurarine as a muscle relaxant (i.v. 0.5 mg/kg bolus and 0.5 mg/kg/h infusion). Animals were artificially ventilated (60 1/min, 70% N₂, 30% O₂, the N₂ was substituted during PET scans with radioactive gas (mainly N₂ and trace amounts of ¹⁵O labeled gas). A pair of subcutaneous needle electrodes was inserted in the left whisker pad of the animals for stimulation (1.5 mA, 3 Hz). Animals were first scanned in a small animal PET using 6 [¹⁵O]O₂ scans (3 rest, 3 stim) and additional [¹⁵O]CO₂, [¹⁵O]CO scans. Each PET scan lasted 10 minutes (1 GBq/min, 300 mL/min flow). After the PET experiment, animals were moved to a 7T small animal MR. Anatomical as well as EPI imaging (GE-EPI, TR: 2000 ms, TE: 18 ms, matrix: 64x64x20, voxel size: 0.5x0.5x1.0 mm³, 300 volumes) was performed applying the same stimulus as used during PET imaging. Data processing was performed using SPM 12 and PMOD. PET images were averaged for each scan to represent a mean image proportional to CMRO₂, coregistered across animals and smoothed with a 1.5 mm 3D Gauss kernel. fMRI-BOLD images were motion corrected, coregistered to an atlas constructed from the anatomical MR scans and smoothed with 1.5 mm 3D Gauss. A two sample t-test was performed on the PET images (rest vs. stim), GLM analysis on 1st and 2nd level was used for the fMRI data.PET imaging showed significant (P<0.05) changes in CMRO₂ in the area of the cl barrel field cortex (S1BF) (T=2.23) but also additional activations in areas such as the cl Amygdala (T=2.33), the cl Thalamic area (T=1.84) and cl insula cortex (T=2.25) as well as other areas (Figure 1). BOLD-fMRI found significant activation (P<0.05) in the cl S1BF (T=3.98), vicinity of the cl Amygdala (T=4.06) and some border areas of cl Thalamus and Cortex (T=3.76). Using the same threshold (p<0.05), PET showed more activated areas within the brain compared to fMRI-BOLD, e.g. the cl ectorhinal, perihnal and entorhinal cortex (T=2.25) and further areas in the il and cl Thalamus. Furthermore a spatial offset between peak PET and BOLD-fMRI activation centers was observed, e.g. in the cl S1BF of 3.9 mm and in the il Amygdala of 1.8 mm. In general the activated areas appeared smaller in size (number of significant voxels) in [¹⁵O]O₂-PET compared to BOLD-fMRI.This study revealed, that there exists in some areas of the brain a spatial mismatch between activation patterns measured by CMRO₂ proportional [¹⁵O]O₂-PET and BOLD-fMRI. A shift between peak activation centers of the same region can be explained by the fact, that T2* weighted BOLD-fMRI is more weighted towards the venous space. However, the additional regions observed in [¹⁵O]O₂-PET, not present in BOLD-fMRI point towards a more complex mismatch between the two techniques, which could stem from the complex nature of the BOLD-fMRI response as a convolution of CMRO₂, CBF and CBV changes. BOLD-fMRI might therefore, in certain circumstances, show only a limited number of activated brain regions. Limitations of this study are sequential PET and MR scans as well as stimulus presentation and anesthesia. To our knowledge this is the first small animal study comparing [¹⁵O]O₂-PET and BOLD-fMRI during the same imaging session. This work opens up the domain for a better understanding of the BOLD effect and further refined studies using simultaneous PET/MR imaging.