Identification of the sensorimotor cortex with functional MRI: Frequency and actual contribution in a neurosurgical context

BACKGROUND AND PURPOSE: We assessed the actual frequency of motor functional MRI (fMRI) in a neurosurgical environment and estimated the extent to which it aided surgeons' identifications of the sensorimotor cortex. METHODS: During five consecutive years, an fMRI protocol aimed at generating a selective activation of the hand cortical area was prescribed to 147 patients showing a centrally located space-occupying lesion, which represents 6.7% of all assisted surgical candidates showing an intracranial mass. Three senior neurosurgeons indicated the position of the sensorimotor cortex on two different anatomical displays, reporting confidence ratings for each decision. RESULTS: The sensorimotor cortex could not be identified in 16.5% of cases using conventional anatomical MRI, and in 15% of cases using 3-dimensional reconstructions. In an additional 12.5% of cases, the neurosurgeons were not confident when they correctly identified the sensorimotor cortex. The tumor distorting effect on central region anatomy significantly contributed to sensorimotor cortex misidentification. fMRI, by contrast, showed a selective activation indicating the position of the sensorimotor cortex in all but 4% of cases. CONCLUSIONS: In our neurosurgical environment, fMRI was prescribed to a selected group of surgical candidates showing a centrally located brain lesion. Compared to conventional anatomical imaging, fMRI does appear to improve the identification of sensorimotor cortex. © 2008 by the American Society of Neuroimaging.