Functional magnetic resonance imaging is more reliable than somatosensory evoked potential or mapping for the detection of the primary motor cortex in proximity to a tumor.

The accurate localization of the primary motor cortex (M1) is critical for the preservation of motor function during resection of brain tumors in and around the M1. The goal of the present study was to determine which technique provided the most accurate localization of M1. The accuracy of preoperative functional magnetic resonance imaging (fMRI), intraoperative somatosensory evoked potential (SEP) and cortical mapping for the localization of M1 was determined in 17 patients with brain tumors in and around the M1. Because localization of the M1 is typically symmetrical in the cerebral hemispheres, the M1 on the affected side was localized by determination of the M1 location on the unaffected side using fMRI with patient hand clenching. The location of M1 was successfully determined by SEP in 5 of 11 cases. In the remainder of cases, the sulcus at which phase reversal occurred during SEP was shifted 1 or 2 gyri rostral to the central sulcus. The location of M1 was successfully determined by brain mapping in 9 of 15 cases. In the remainder of cases, stimulation failed to elicit a motor response. Finally, the location of M1 was successfully determined by fMRI in 16 of 17 cases. These data indicate that fMRI was more reliable than SEP or brain mapping for the detection of M1 in proximity to a tumor. Copyright (c) 2007 S. Karger AG, Basel.