Presurgical motor and somatosensory cortex mapping with functional magnetic resonance imaging and positron emission tomography

Object. Accurate identification of eloquent cortex is important to ensure that resective surgery in the region surrounding the central sulcus is performed with minimum risk of permanent neurological deficit. Functional localization has traditionally been accomplished using intraoperative cortical stimulation (ICS). However, this technique suffers from several disadvantages that make the development and validation of noninvasive methods desirable. Functional localization accomplished by activation studies in which positron emission tomography (PET) scanning and the tracer [O-15]H2O have been used has been shown to correlate well with the results of ICS. Another noninvasive method for functional localization is functional magnetic resonance (fMR) imaging. We compared the locations of activation peaks obtained in individual patients using fMR and [O-15]H2O PET imaging. Methods. Twenty-six combined PET activation-fMR imaging studies were performed in 11 patients who were admitted for evaluation before undergoing surgery in the region surrounding the central sulcus. The PET scans were Obtained using bolus injections of the cerebral blood flow tracer [O-15]H2O (10 mCi). Multislice T-2*-weighted gradient-echo echoplanar images were acquired using a 1.5-tesla MR imaging system. Activation maps were aligned with anatomical MR images and transformed into stereotactic space, after which the locations of activation peaks obtained using both modalities were compared. The average distance between activation peaks obtained using fMR imaging and those obtained using PET imaging was 7.9 +/- 4.8 mm (p > 0.05), with 96% of the peaks being located on either the same or adjacent sulci and gyri. Overlapping of voxels activated by each modality occurred in 92% of the studies. Functional MR imaging failed to activate the primary sensorimotor cortex in one study and produced results that were ambiguous in the clinical setting in three cases. Conclusions. Overall, fMR imaging produced activation that correlated well with that obtained using PET scanning. Discrepancies between the sites of activation identified using these two techniques may reflect differences in their physiological bases.