Objective: Language functional magnetic resonance imaging (fMRI) has been used extensively in the past decade for both clinical and research purposes. Its integration in the preoperative imaging assessment of brain lesions involving eloquent areas is progressively more diffused in neurosurgical practice. Nevertheless, the reliability of language fMRI is unclear. To understand the reliability of preoperative language fMRI in patients operated on for brain tumors, the surgical studies that compared language fMRI with direct cortical stimulation (DCS) were reviewed.
Methods: Articles comparing language fMRI with DCS of language areas were reviewed with attention to the lesion pathology, the magnetic field, the language tasks used pre- and intraoperatively, and the validation modalities adopted to establish the reliability of language fMRI. We tried to explore the effectiveness of language fMRI in gliomas.
Results: Nine language brain mapping studies compared the findings of fMRI with those of DCS. The studies are not homogeneous for tumor types, magnetic fields, pre- and intraoperative language tasks, intraoperative matching criteria, and results. Sensitivity and specificity were calculated in 5 studies (respectively ranging from 59% to 100% and from 0% to 97%).
Conclusion: The contradictory results of these studies do not allow consideration of language fMRI as an alternative tool to DCS in brain lesions located in language areas, especially in gliomas because of the pattern of growth of these tumors. However, language fMRI conducted with high magnet fields is a promising brain mapping tool that must be validated by DCS in methodological robust studies.