Object: Preserving motor function is a major challenge in surgery for intraaxial brain tumors. Navigation systems are unreliable in predicting the location of the corticospinal tracts (CSTs) because of brain shift and the inability of current intraoperative systems to produce reliable diffusion tensor imaging data. The authors describe their experience with elaborate neurophysiological assessment and tractography-based navigation, corrected in real time by 3D intraoperative ultrasonography (IOUS) to identify motor pathways during subcortical tumor resection.
Methods: A retrospective analysis was conducted in 55 patients undergoing resection of tumors located within or in proximity to the CSTs at the authors' institution between November 2007 and June 2009. Corticospinal tract tractography was coregistered to surgical navigation-derived images in 42 patients. Direct cortical-stimulated motor evoked potentials (dcMEPs) and subcortical-stimulated MEPs (scrtMEPs) were recorded intraoperatively to assess function and estimate the distance from the CSTs. Intraoperative ultrasonography updated the navigation imaging and estimated resection proximity to the CSTs. Preoperative clinical motor function was compared with postoperative outcome at several time points and correlated with incidences of intraoperative dcMEP alarm and low scrtMEP values.
Results: The threshold level needed to elicit scrtMEPs was plotted against the distance to the CSTs based on diffusion tensor imaging tractography after brain shift compensation with 3D IOUS, generating a trend line that demonstrated a linear order between these variables, and a relationship of 0.97 mA for every 1 mm of brain tissue distance from the CSTs. Clinically, 39 (71%) of 55 patients had no postoperative deficits, and 9 of the remaining 16 improved to baseline function within 1 month. Seven patients had varying degrees of permanent motor deficits. Subcortical stimulation was applied in 45 of the procedures. The status of 32 patients did not deteriorate postoperatively (stable or improved motor status): 27 of them (84%) displayed minimum scrtMEP thresholds > 7 mA. Six patients who experienced postoperative deterioration quickly recovered (within 5 days) and displayed minimum scrtMEP thresholds > 6.8 mA. Five of the 7 patients who had late (> 5 days postoperatively) or no recovery had minimal scrtMEP thresholds < 3 mA. An scrtMEP threshold of 3 mA was found to be the cutoff point below which irreversible disruption of CST integrity may be anticipated (sensitivity 83%, specificity 95%).
Conclusions: Combining elaborate neurophysiological assessment, tractography-based neuronavigation, and updated IOUS images provided accurate localization of the CSTs and enabled the safe resection of tumors approximating these tracts. This is the first attempt to evaluate the distance from the CSTs using the threshold of subcortical monopolar stimulation with real-time IOUS for the correction of brain shift. The linear correlation between the distance to the CSTs and the threshold of subcortical stimulation producing a motor response provides an intraoperative technique to better preserve motor function.