OBJECTIVE: Evaluation of an integrated multi-function batter of preoperative fMRI tasks developed to identify cortical areas associated with tactile, motor, language, and visual functions.
METHODS: Sensitivity of each task was determined by the probability that a targeted region was activated for both healthy volunteers (n=63) and surgical patients with lesions in these critical areas (n=125). Accuracy of each task was determined by the correspondence between the fMRI maps and intraoperative electrophysiology including somatosensory evoked potentials, SSEP (n=16), direct cortical stimulation (n=9) and language mapping (n=5); and preoperative Wada test (n=13) and visual field examinations (n=6).
RESULTS: For healthy volunteers the overall sensitivity was 100% for identification of central sulcus, visual cortex, and putative Wernicke's Area, and 93% for putative Broca's Area (dominant hemisphere). For patients with tumors affecting these regions of interest, the task sensitivities were 97% for identification of central sulcus, 100% for visual cortex, 91% for putative Wernicke's Area, and 77% for putative Broca's Area. These sensitivities were enhanced by the use of multiple tasks to target related functions. Concordance of the fMRI maps and intraoperative electrophysiological resultys was observed whenever both techniques yielded maps, and Wada and visual field examinations were consistent with fMRI results.
CONCLUSION: This integrated fMRI task battery offers standardized and non-invasive preoperative maps of multiple critical functions to facilitate assessment of surgical risk, planning of surgical routes, and direction of conventional, intraoperative electrophysiological procedures. Thus, a greater range of structural and functional relationships are employed in the service of optimal outcomes for neurosurgery.