Combined Functional Magnetic Resonance Imaging and Magnetoencephalography for Enhanced Specificity of Presurgical Mapping: A Comparison with the Gold Standard.

Functional mapping is a vital component of the surgical treatment of brain tumors involving critical language areas. Cortical stimulation, although highly invasive and limited to the time of surgery, is the current gold standard for functional mapping. Functional magnetic resonance imaging (fMRI) is a non-invasive imaging technique suitable for pre-surgical mapping. Current protocols for mapping language by fMRI are optimized for sensitivity instead of specificity, to emphasize sparing of functional zones. A lack of specificity implies an increased number of false positive activations, which would lead to overly conservative resection boundaries, and negatively affect survival rates. Combined fMRI and magnetoencephalography (MEG) has the potential to increase the specificity of pre-surgical mapping, and enhance the potential standard of care for brain tumor resection. Twenty patients with cortical tumors in the putative language regions will be imaged using MRI, fMRI, and MEG. Each patient will complete a functional task battery including object naming, sentence comprehension, and semantic decision-making. The MEG data will be reconstructed using the beamforming approach, to create a spatially extended source description similar to functional MRI. The fMRI and MEG will be controlled for reproducibility by analyzing split-half datasets. The patients will all receive awake craniotomy, at which point cortical stimulation will be performed. All functional mapping data will be brought into the anatomical MRI space, to compare the non-invasive mapping techniques to the gold standard. It is hypothesized that the specificity achievable by combined fMRI and MEG will be greater than by either technique alone.