OBJECTIVETo compare the relation between the preoperative functional magnetic resonance imaging (fMRI) with blood oxygen level dependent (BOLD) technique and intraoperative motor evoked potential (MEP) monitoring for cortical mapping of primary motor cortex in patients with tumors near the central area. And to determine whether non-invasive preoperative fMRI can provide results equivalent to those achieved with the invasive neurosurgical "gold standard".

METHODSA prospective study of 16 patients with various pathological tumors of the central area was conducted. Preoperative fMRI scans using the BOLD contrast technique in each patient were performed. An activation scan was achieved by using a motor task paradigm, which consisted of simple flexion-extension finger movements and finger-to-thumb touching in a repeating pattern. The anatomical structure was delineated by the T(1)-weighted three-dimensional fast spoiled gradient recalled sequence (3D/FSPGR) immediately afterward. The BOLD images were overlaid on the T(1)-weighted 3D/FSPGR images, and then co-registered to the neuronavigation system. The fMRI activations were documented by using a neuronavigation system in sequence, and compared to standardized intraoperative MEP monitoring, which included direct cortical electrical stimulation (DCES) or transcranial cortical electrical stimulation (TCES) or their combination. The compound muscle action potentials of forearm flexor and hand muscle responses were recorded during either TCES or DCES. Two techniques were compared to determine the accuracy for cortical mapping of primary motor areas with fMRI.

RESULTSOverall, the intraoperative MEP monitoring showed good correlation with fMRI activation in 92.3% of cases. The coincidence rate, however, was 100.0% between TCES and fMRI, and 66.7% between DCES and fMRI respectively. There was no statistically difference between two cortical mapping techniques, chi-square test of paired comparison of enumeration data, P < 0.01.

CONCLUSIONBOLD fMRI was a high sensitive and reliable technique to locate the position of the primary motor areas and their spatial relation with adjacent tumor, especially for the presurgical planning in patients with central area brain tumor.