Comparative localization of auditory comprehension by using functional magnetic resonance imaging and cortical stimulation.

OBJECT The authors previously described a functional magnetic resonance (fMR) imaging task for the localization of auditory comprehension in which focal activation of posterior temporal and inferior frontal regions of the left hemisphere was reliably demonstrated. Because this study was conducted in neurologically normal volunteers, it was not possible to determine whether the activated regions were critical to the performance of language tasks; that is, whether the fMR imaging activations provided a valid measure of language processing. A direct comparison of fMR imaging language activation with cortical stimulation must be completed before it can be used with confidence in presurgical planning, and this comparison is performed in the present study. METHODS The authors report on a series of 33 consecutive patients who underwent dominant hemisphere resection and in whom fMR imaging mapping of auditory comprehension was performed at the Yale neurosurgical program. In 23 of the 33 patients fMR imaging activation was consistent with the typical results obtained in normal participants in the earlier study. In 16 of these 23 patients language mapping was performed using either intra- or extraoperative cortical stimulation. Cortical stimulation failed to localize language areas in two of the 16 patients. Electrical stimulation that was performed in proximity to the fMR image activations interfered with auditory comprehension, object naming, or speech production in 12 of the remaining 14 patients. Five of the 10 cases in which evocation of reliable fMR imaging activation failed were attributable to technical problems and/or patient head movement. CONCLUSIONS Cortical stimulation results and fMR imaging findings were consistent in all but two patients. However, the spatial extent of the activation produced by fMR imaging and the spatial extent of stimulation-induced language disruption that was caused by direct cortical stimulation did not always correspond. Problems in defining the extent of activation by both methods are discussed.