Re s e c t i o n of tumors in the proximity of eloquent

Transcription

1 J Neurosurg 114: , 2011 Intraoperative magnetic resonance imaging guided tractography with integrated monopolar subcortical functional mapping for resection of brain tumors Clinical article Su j i t S. Pr a b h u, M.D., F.R.C.S.(Ed), 1 Ja i m e Ga s c o, M.D., 1 Su d h a k a r Tu m m a l a, M.D., 2 Jef f r ey S. We i n b e rg, M.D., 1 a n d Ga n e s h Ra o, M.D. 1 Departments of 1 Neurosurgery and 2 Neuro-Oncology, University of Texas M. D. Anderson Cancer Center, Houston, Texas Object. The object of this study was to describe the utility and safety of using a single probe for combined intraoperative navigation and subcortical mapping in an intraoperative MR (imr) imaging environment during brain tumor resection. Methods. The authors retrospectively reviewed those patients who underwent resection in the imr imaging environment, as well as functional electrophysiological monitoring with continuous motor evoked potential (MEP) and direct subcortical mapping combined with diffusion tensor imaging tractography. Results. As a navigational tool the monopolar probe used was safe and accurate. Positive subcortical fiber MEPs were obtained in 10 (83%) of the 12 cases. In 10 patients in whom subcortical MEPs were recorded, the mean stimulus intensity was 10.4 ± 5.2 ma and the mean distance from the probe tip to the corticospinal tract (CST) was 7.4 ± 4.5 mm. There was a trend toward worsening neurological deficits if the distance to the CST was short, and a small minimum stimulation threshold was recorded indicating close proximity of the CST to the resection margins. Grosstotal resection (95% 100% tumor removal) was achieved in 11 cases (92%), whereas 1 patient (8%) had at least a 90% tumor resection. At the end of 3 months, 2 patients (17%) had persistent neurological deficits. Conclusions. The monopolar probe can be safely implemented in an imr imaging environment both for navigation and stimulation purposes during the resection of intrinsic brain tumors. In this study there was a trend toward worsening neurological deficits if the distance from the probe to the CST was short (< 5 mm) indicating close proximity of the resection cavity to the CST. This technology can be used in the imr imaging environment as a surgical adjunct to minimize adverse neurological outcomes. (DOI: / JNS10481) Ke y Wo r d s intraoperative subcortical stimulation monopolar probe neuronavigation diffusion tensor imaging intraoperative magnetic resonance imaging corticospinal tract Abbreviations used in this paper: CST = corticospinal tract; DTI = diffusion tensor imaging; GTR = gross-total resection; imr = intraoperative MR; MEP = motor evoked potential; ROI = region of interest. J Neurosurg / Volume 114 / March 2011 Re s e c t i o n of tumors in the proximity of eloquent cortical and subcortical (speech and sensorimotor) areas is challenging and carries a higher morbidity in the immediate postoperative period. 2,11,12 A number of technological advances, including the intraoperative use of high resolution DTI maps for functional navigation, have provided safe corridors in the brain to maximize resection of tumors. 5,19 Intraoperative MR imaging provides an increased safety margin during resection by compensating for brain shift. 6 8,18,27 Functional localization of eloquent cortical and subcortical structures using intraoperative electrophysiological mapping with DTI maps may increase safety during tumor resection. 1,10,16,22 Although bipolar probes have been used for both cortical and subcortical mapping of the brain, a number of authors have successfully used monopolar stimulation to achieve the same goals. 10,21,26 We recently demonstrated the accuracy, safety, and technical nuances of using a monopolar probe for simultaneous functional neuronavigation and tractography-integrated stimulation of subcortical white matter in glioma surgery in the imr imaging environment. 4 In this report we describe our initial experience in the resection of both primary and metastatic tumors located in close proximity to the CST using the combined integrated technology described. In addition to obtaining the electrical threshold of fiber MEP of the CST, we present our results on postoperative neurological outcome. Methods The study was approved by the institutional review board. After obtaining informed consent for each patient, 719

2 S. S. Prabhu et al. we performed surgery in the BrainSUITE (BrainLAB) using intraoperative electrophysiological monitoring. Each patient underwent initial MR imaging for registration purposes that included a DTI sequence in the high-field 1.5-T MR imaging scanner (Siemens AG Medical Solutions). Using the iplan Cranial software, segmentation of the tumor and the CST was performed and the information was transferred to the intraoperative neuronavigation system. As in our previous report, we integrated the Medtronic Xomed Prass probe, a 1-mm ball tip monopolar probe, with the VectorVision neuronavigation software (BrainLAB) in our BrainSUITE using the Star- Link interface instrument calibration matrix (Fig. 1). The Prass probe was then used as an anode during subcortical stimulation and provided simultaneous navigation of the cortical and subcortical structures. Following the dural opening, identification of the motor and sensory cortex was conducted by phase reversals obtained by stimulating the contralateral median nerve. Continuous MEPs were recorded using an 8-contact strip electrode placed on the suspected motor area. Grid contact with the best motor response was used as an anode. The grid was then sutured to the dura for the duration of the operation for continuous recording. A magnetic-safe needle placed in the exposed subcutaneous tissue was used as a cathode. Contralateral muscle responses were monitored by placing magnetic-safe single-use needle electrodes in the forearm, hand, leg, and foot muscles in each case. Motor responses from the forearm and hand were obtained using the following parameters: train of 5, 500 Hz, pulse duration µsec, with current amplitudes ranging from 6 to 20 ma. Stimulation was performed manually every 1 to 3 minutes. During the subcortical dissection of the tumor, the monopolar probe was used to directly stimulate the fiber tracts. Once a fiber MEP was obtained, the current intensity was reduced to the minimum stimulation threshold that would elicit the fiber MEP. We used 20 ma as the limit to activate the subcortical CST. An intraoperative scan was obtained when the surgeon believed it was necessary to assess the extent of resection. This scan (including a new DTI sequence) was registered to the intraoperative navigation system. Subcortical stimulation was again performed with the newly registered scans along with the newly reconstructed CST. The distance (in mm) between the tip of the monopolar probe and the CST was obtained intraoperatively using the screenshot function provided by the software. 4 All patients underwent routine postoperative MR imaging the day following surgery to assess the extent of resection. Functional outcomes, including the Karnofsky Performance Scale score and neurological function, were assessed immediately postoperatively and at 1- and 3-month follow-up. Fiber Tracking With iplan Software The DTI functionality in iplan software 2.6 processes the Digital Imaging and Communications in Medicine (DICOM) information to compute the diffusion directions in MR images. The iplan software preprocesses the image sets to improve the image quality by providing Fig. 1. Photograph of the 1-mm ball tip monopolar probe with the star link and the 3 reflection spheres attached to the handle. After registration the probe is used both for navigation and direct subcortical stimulation. smoothing, motion, and eddy current correction. Once the import is complete the software automatically adds it to the anatomical image sets. The iplan software recognizes the image sets based on the coordinate system and temporarily fuses these image sets. The fusion is complete after user confirmation of the nondiffusion-weighted (B0) and fractional anisotropy (FA) image sets with the anatomical image set (T1, T2, FLAIR, and others). After fusion, the fiber-tracking dialog allows one to track fiber structures in a defined ROI based on diffusion-weighted MR images. Because the direction of water diffusion along potential white matter fibers is calculated for the entire data volume, the iplan software traces the direction of fibers in a selected ROI and displays the corresponding directional color maps. To visualize fiber structures in the treatment plan we define the initial seed ROI in the selected image set using the DTI ROI. The tracked fibers represent the local diffusion passing through the chosen ROI. As the fibers are tracked, the software displays the fibers in different colors according to the diffusion direction. The left-right oriented fibers are displayed in red, the anteriorposterior oriented fibers are displayed in green, and the head-foot oriented fibers are displayed in blue. The fiber tracking result is based on calculations from diffusion tensor images. The iplan software therefore displays the relative representation of local anisotropy, which is related to fiber structures in brain white matter. When the fiber tracking is complete, a 3D model of the fiber bundles is created for navigation. This model is created on an anatomical image set because it offers higher resolution. Object Creation and Segmentation With iplan Software The object-creation planning task helps highlight structures of interest in an image set. This allows us to clearly mark the position of the tumor or outline anatomical structures for better orientation in the image set. This can be completed manually or with automatic segmentation using anatomical structures in the brain (such as the cortex, ventricles, and thalamus). The link between the patient s data set and the atlas data are not a simple transformation matrix that recognizes the differences in shape and size of the anatomical structures. Rather, the atlas set needs to be adapted in such a way that the similarity of the data sets increases. The result is a deformation vector field, which defines a unique mapping from the atlas into the patient data. Each voxel of an outlined atlas object can 720 J Neurosurg / Volume 114 / March 2011

3 Intraoperative MR imaging guided tractography therefore be transferred into the patient data by moving it to the position defined by that mapping. Diffusion Tensor Imaging in the BrainSuite Diffusion tensor imaging data are acquired at 1.5-T using a Siemens Espree scanner (Siemens Medical Solutions) and an 8-channel phased array head coil (Noras MR imaging Products). For full brain coverage, 44 to 46 3-mm sections were acquired using a spin-echo echo planar imaging sequence with parallel imaging (acceleration factor of 2, TE = 100 ms, TR = 6700 msec, 5 averages, 20 diffusion directions, b values of approximately 0 and 750 sec/mm 2, 1260 Hz/pixel bandwidth, matrix, and a 240-mm FOV). Scan time for full brain coverage using these parameters is approximately 12 minutes. For more limited brain coverage, approximately 28 sections are acquired with a TR of 4300 msec, 4 averages, and all other parameters the same as listed above. Scan time for the limited brain coverage is approximately 6 minutes. Anesthesia Technique For induction of anesthesia, 0.5 µg/kg sufentanil, 2 mg/kg propofol, 2 mg midazolam, 20 mg pepcid, and 0.6 mg/kg rocuronium (a very short-acting muscle relaxant) are initially used. All patients receive a scalp block with a total of approximately 30 ml of 0.5% ropivacaine with epinephrine in a 1:200,000 ratio to the greater/lesser occipital, auricular-temporal, zygomatic-temporal, V1, and supratrochlear nerves, respectively. Maintenance of anesthesia is conducted with 50 µg/kg/min propofol, 0.05 µg/kg/min remifentanil, approximately 0.5% 0.6% isoflurane, and 50% oxygen/50% air. No additional muscle relaxants were used during surgery. Homeostasis was thoroughly maintained throughout surgery; a body temperature of 36 C was ascertained by routine use of an air warmer system (Bair-Hugger, Augustine Medical, Inc.). Results Twelve patients (8 men and 4 women, mean age 50 years, range years) with symptomatic primary and metastatic brain tumors located adjacent to the CST were included in the study. Subcortical Recording and Navigation In all patients the monopolar probe was registered with good accuracy (tip deviation < 1 mm) and was used as the navigation probe during tumor resection. Using a single probe for both navigation and stimulation enabled more accurate placements and recordings in the resection cavity. Subcortical recordings of the CST were obtained in 10 (83%) of the 12 patients. The range of currents used for stimulation was 2 ma to 20 ma. In 10 patients in whom subcortical MEPs were recorded, mean stimulus intensity was 10.4 ± 5.2 ma. In these patients continuous MEPs from the grid were also recorded throughout the resection with a mean stimulation intensity of 14.1 ± 3.7 ma. The distance from the probe tip to the CST in the 12 patients after tumor resections ranged from 1 19 mm. In the 10 patients in whom subcortical MEPs were recorded the mean distance was 7.4 ± 4.5 mm. J Neurosurg / Volume 114 / March 2011 Patients in Whom Subcortical Stimulation was not Recorded In 2 patients (Cases 4 and 5) subcortical responses were not obtained even at a maximal intensity of 20 ma. In Case 4, the patient underwent GTR of a metastatic tumor with no postoperative deficits. The patient in Case 5 underwent GTR of a recurrent glioblastoma resulting in a supplemental motor area syndrome postoperatively that recovered completely 2 weeks later. Although 1 patient developed a supplemental motor area syndrome, in both the cases the CSTs were at least 19 mm from the resection cavity. In both of these cases continuous MEPs were recorded during the entire tumor resection, indicating intact subcortical CSTs. As the distance between the probe tip and the CST increased, larger currents were required to elicit subcortical MEPs (Table 1). Continuous MEP Recordings It is our practice to record continuous MEPs that also monitor cortical and subcortical connections during tumor resection. We successfully recorded this in all 12 patients (100%) during the resection; the mean stimulus intensity was 14.2 ± 3.6 ma. Postoperative Analysis of Extent of Resection and Relationship to the CSTs Of the 12 patients, 5 (42%) had metastatic brain tumors and 7 (58%) had gliomas. Gross-total resection (95% 100%) was achieved in 11 cases (92%), whereas 1 patient (8%) underwent an STR (85% 94%). The extent of resection was assessed with the use of intraoperative ultrasonography scans and imr imaging. All patients in this study underwent 1 additional imr image with the exception of Case 11 who underwent 2 imr images to document the extent of resection. The distance between the monopolar probe and the CSTs was measured after completing the tumor resections using the intraoperative scans (Table 1). Neurological Function Following Surgery New and/or worsening neurological deficits were observed in 7 (58%) of the 12 patients. In 2 of these patients (17%), a persistent neurological deficit was noted at 3 months. In 2 of these patients (Cases 6 and 12) the minimum stimulation threshold was 5 ma, indicating close proximity of the CST to the resection cavity. These patients had mild hemiparesis with a Karnofsky Performance Scale score of 70 or higher at 3-month follow-up. One patient (Case 2) who did not experience any immediate postoperative neurological deficit died of systemic disease progression before the 1-month follow-up. Preoperative neurological deficit was present in 2 (17%) of the 12 patients. Both of these patients worsened immediately following surgery. At 1 month, Case 6 still had a persistent deficit while Case 9 had no deficits. In Case 6 the distance from the monopolar probe to the CST was 4 mm with a minimum stimulation threshold of 3 ma, indicating close proximity to the CST, while in Case 9 this distance and threshold were 11 mm and 12 ma, respectively. 721

4 S. S. Prabhu et al. TABLE 1: Demographics, neurological outcomes, and intraoperative data in 12 patients undergoing simultaneous navigation and stimulation using imr imaging* Case No. Age (yrs), Sex Tumor Location Diagnosis Neurological Deficits Preop Postop 1-month Extent of Resection MST of CMEPs (ma) MST of Fiber MEPs (ma) Distance from Probe Tip to CST (mm) 1 50, M rt parietal glioblastoma no no no GTR , F lt frontal metastasis no no deceased GTR , M lt frontal metastasis no deficit no GTR , M rt parietal metastasis no no no GTR , M rt frontal glioblastoma no deficit no GTR , F rt frontal metastasis deficit deficit deficit GTR , M rt frontal glioblastoma no deficit no STR , F lt parietal glioblastoma no no no GTR , F lt frontal metastasis deficit deficit no GTR , M rt temporal glioblastoma no no no GTR , M lt frontal Grade II oligoden- no deficit deficit GTR droglioma 12 75, M rt parietal glioblastoma no deficit deficit GTR * CMEP = continuous MEPs; MST = minimum stimulation threshold. Shift of CSTs in Relationship to Resection Cavity In all 12 patients, we recorded the shift of the CST (range 0 18 mm, mean 5.4 ± 5.4 mm) from the margins of the resection cavity based on the preoperative and imr images. Illustrative Cases The combined implementation of functional tractography and use of imr imaging is illustrated in the following 2 examples. Case 11 This 40-year-old man had a WHO Grade II oligodendroglioma located in the right superior frontal gyrus in the supplemental motor area, extending posteriorly to the central sulcus (Fig. 2A). Following the initial resection the first intraoperative scan showed residual tumor adjacent to the CST at the posterior margin of the resection cavity (Fig. 2B). Fiber MEPs recorded with the monopolar probe at the posterior resection margin were 20 ma, and the CST was measured at 20 mm from the probe tip. Additional resection was then performed until the fiber MEPs were recorded at 5 ma with the probe tip just adjacent to the CST (recorded as 1 mm; Fig. 2C E). The patient underwent GTR of the tumor, with resulting mild left-sided weakness at 1 month. The CST on the 2 intraoperative scans (Fig. 2B and C) appears less stretched and more compact compared with the preoperative scan (Fig. 2A), which has to be taken into account during subsequent resections as the CST appears closer to the posterior resection cavity. The intraoperative navigation and subcortical fiber MEPs gave the surgeon a better understanding of the proximity of this CST to the resection cavity. Case 8 This 42-year-old woman had a deep-seated, mostly nonenhancing glioblastoma under the left parietal cortex primarily involving the white matter (Fig. 3A). In the anterior and deep portions of the resection cavity, where the tumor was very infiltrative and nonenhancing, fiber MEPs of 14 ma for the upper extremity and 18 ma for the lower extremity were obtained. The resection was stopped at this point when the imr image revealed this deep portion of the tumor to be in close proximity (approximately 12 mm) from the CST (Fig. 3B). At this point, at least 90% of the tumor was resected and no additional resection was completed because of the infiltrative nature of the tumor and possible risk of injury to the CST. The patient did not experience any postoperative neurological deficits. Discussion To define the subcortical fibers (CST) from the primary motor cortex as well as monitor their function intraoperatively, MEPs elicited by direct electrical stimulation remain the most reliable index. Traditional methods used to stimulate the cortical and subcortical tracts include a bipolar probe (Ojemann bipolar probe) or stimulation via a cortical grid sutured to the dura, which then functions as a monopolar anode. 15,17,23,25 The usefulness and safety of subcortical stimulation, especially in the resection of low-grade gliomas, is well described by a number of authors. 2,3,11 Some studies combined either bipolar or monopolar cortical stimulation (with a surface grid) with neuronavigation-assisted DTI tractography and analyzed the distances between stimulus points and CST using preoperative MR imaging. 1,9,10,14,16,22,28 Berman et al., 1 using bipolar subcortical stimulation in 9 patients, were successful in identifying the CST in all patients at distances of 8.7 ± 3.1 mm from the CST and reported an immediate postoperative neurological deficit in 3 (33%) of 9 cases. Mikuni et al., 16 using both cortical and subcortical grid stimulation, reported positive MEPs in 18 (90%) of 20 patients if the distance to the CST was 10 mm. How- 722 J Neurosurg / Volume 114 / March 2011

5 Intraoperative MR imaging guided tractography Fig. 2. Case 11. Intraoperative images obtained during resection of a right frontal WHO Grade II oligodendroglioma. A: Preoperative axial FLAIR sequence showing the fibers of the CST stretched out at the back of the tumor. B: Following initial resection, the first intraoperative axial FLAIR sequence shows residual tumor. At this point the measured distance between the probe tip and the CST was 20 mm and fiber MEPs were obtained at 20 ma. C: The patient underwent additional resection of the residual tumor and the second intraoperative scan shows the CST in close proximity to the posterior resection margin. D and E: Screen shot of sagittal (D) and axial (E) images during intraoperative navigation after the residual tumor was resected. These images show the segmented tumor cavity in green and purple and the segmented CST in light brown adjacent to the resection cavity. The measured distance between the tip of the probe (yellow line) and the CST was approximately 1 mm. While the imr image accounts for the shift, the short distance between the probe and the CST (1 mm) along with the small amplitude of current (5 ma) to generate a fiber MEP alerted the surgeon to the proximity of the CST. ever, if this distance was 1 3 cm, MEPs were recorded in only 3 (15%) of 20 patients. The immediate postoperative neurological morbidity reported in this study was 5% (2 of 40 patients). Ozawa et al.22 used neuronavigation based on intraoperative diffusion-weighted imaging combined with bipolar subcortical mapping for the resection of tumors in the deep white matter in 7 patients. In this study intraoperative imaging corrected for brain shift and diffusion-weighted imaging was used to avoid underestimation of the tract size. Positive motor evoked potentials were detected in 5 of 7 patients (8 stimulations) and the distance from the stimulation site to the depicted bundle ranged from 0 mm to 4.7 mm (mean 1.4 ± 2.1 mm). These investigators obtained negative responses in 2 patients when the distance was more than 5 mm to the CST. Because different stimulus conditions were used in these reports, each result was slightly different. J Neurosurg / Volume 114 / March 2011 Kamada et al.10 reported the use of a monopolar probe for stimulation and showed that the shorter the probe distances from the CST the smaller the minimum stimulation threshold required, but they did not correlate this to neurological outcome. In their study no fiber MEPs were observed at the maximum stimulus intensity in 22 (55%) of 40 cases throughout resection, although cortical MEPs (mean stimulus intensity 15.9 ± 2.31 ma) were continuously observed in these patients. They also showed that in cases with no fiber MEPs, the mean distance between the resection cavity and the CST was ± 3.8 mm, and concluded that the major factor in the failed fiber stimulation was the distance between the CST and the stimulus points. Except for the study by Ozawa et al.,22 which was conducted in an imr imaging environment, the other studies relied on preoperative and postoperative imag723

6 S. S. Prabhu et al. Fig. 3. Case 8. Preoperative (A) and intraoperative (B) Gd-enhanced T1-weighted axial MR images. A: Image showing minimal enhancement of a deep-seated right parietal glioblastoma outlined in black. B: Intraoperative image obtained after resection shows the greater part of the tumor resected with residual tumor outlined in black. The posterior part of the CST was approximately 12 mm from the resection cavity, and fiber MEPs were obtained at 14 ma in the upper extremities and 18 ma in the lower extremities, respectively. Because of the infiltrative nature of this residual tumor and its proximity to the CST, further resection was not undertaken. ing to measure distances between the fiber MEPs and the CST and hence minimize measurement errors due to brain shift. To compensate for some of the effects of brain shift, Yamaguchi et al. 28 used a rigid bipolar NY Tract Finder II, which was modified for navigation-assisted detection of motor tracts in cerebral white matter in 8 patients. Yamaguchi and colleagues concluded that 5 mm from the point of MEP positivity in the resection cavity to the CST was a safe distance when no neurological deterioration was noted. The scale on the needle electrodes indicated the approximate distance from the tumor cavity wall to the pyramidal tract. All of these techniques described serve a common goal to maximize the safe resection of both primary and metastatic tumors in close proximity to eloquent brain. Any type of functional neuronavigation can lose its reliability due to brain shift during surgery as a consequence of tumor resection, CSF loss, surgical retraction, or gravity. Nimsky et al. 18 reported on 64 case experiences and found that shift of the cortical surface ranged from 23.8 to 0 mm (mean 8.4 ± 5.6 mm), shift of the deep tumor margin ranged from 7.9 to 30.9 mm (mean 4.4 ± 6.8 mm), and midline shift was between 4.0 and 5.9 mm (mean 0.1 ± 1.7 mm). They concluded that considering these large brain deviations, navigation based on preoperative image cannot be relied upon any longer when tumor resection is approaching pyramidal tracts. In this report the use of the imr imaging during resections was invaluable as it showed a shift in the CST in near-real time in most of the patients (range 0 18 mm, mean 5.4 ± 5.4 mm). It is important to recognize this intraoperatively as the previously compressed CST fibers move closer to the resection cavity as noted in Figs. 2 and 4, and fiber MEPs help in confirming this proximity, thus minimizing or preventing neurological sequelae. Tracking algorithms for DTI currently undergoing implementation cannot clearly visualize and differentiate a pyramidal tract being displaced or distorted from tissue edema (cytotoxic or vasogenic), tumor infiltration, or tumor destruction. 13,24 Hence, real-time changes in the morphology of the CST following tumor decompression and/or resection (with imr imaging) combined with fiber MEP recordings provide invaluable information for surgical planning and can afford additional safety during resection. In at least 2 patients (Cases 8 and 11; 17%), the combined implementation of functional tractography and use of imr imaging influenced intraoperative surgical decision making, which ultimately resulted in very satisfactory neurological outcomes for these patients. In a recent technical note we demonstrated the combined use of imr imaging and functional tractography using a single monopolar probe for both navigation and subcortical stimulation. 4 In the present study we were able to navigate and stimulate the subcortical fibers (CSTs) using the monopolar probe with excellent accuracy. Using a single probe we believe that the intraoperative orientation always remains in view of the surgeon especially during deep subcortical stimulations as the surgical cavity collapses. Although we have not recorded the stimulation times during surgery, we believe this technique should reduce the overall time of surgery as the navigation probe is not replaced by a stimulation probe during the resection. As a rule of thumb we add an extra hour or two to the operating time for cases performed in the BrainSUITE compared with cases performed outside the BrainSUITE because of the time involved with planning the surgery and obtaining scans. In this study the mean duration of surgery was 8.2 ± 1.3 hours, which included patient scanning and surgery times. A recent study from our institution showed a significant increase in median extent of tumor resection (from 76% to 96%) for glioma surgery with intraoperative high-field (1.5-T) MR imag- 724 J Neurosurg / Volume 114 / March 2011

7 Intraoperative MR imaging guided tractography Fig. 4. Sagittal imr images obtained before (A) and after (B) resection of a recurrent glioblastoma adjacent to the motor cortex. Note the significant edema at the posterior aspect of the mass prior to resection (A) with the posterior margin of the tumor 19.4 mm from the CST. Following resection (B) the CST moved forward toward the resection cavity and appeared less compressed and was located 11 mm from the resection cavity. ing.8 We have also shown the feasibility of intraoperative monitoring, which is used in a regular operating room, in combination with the BrainSUITE (imr imaging) technology, with no additional risks to the patient. We were successful in obtaining fiber MEPs in 83% and continuous CMEPs in 100% of the patients throughout the resection. In 2 (Cases 4 and 5) of 12 patients (17%) we were unable to obtain fiber MEPs and in both of these patients the mean distances between the probe tip and the CSTs was 19 mm. Using a monopolar probe, Kamada et al.10 found a nonlinear correlation between stimulus intensity and distance by using a convergent calculation. We also show that less stimulus intensity was required as the CSTs got closer to the resection cavity (probe tip). The mean fiber MEP threshold, which records mainly in the white matter, was 10.4 ma compared with the mean continuous MEP threshold of 14.1 ma, which records both in cortical and subcortical tissues. This difference may be due to the fact that the CST and white matter in general have much lower MEP thresholds than the primary motor cortex.20 Monopolar multipulse stimulation appears to be better suited for muscle recordings. Monopolar stimulation compared with bipolar stimulation is less localized in regard to the current loop. However, knowledge of the stimulation site is precise. The monopolar Prass probe used in our study had an added advantage of easy integration into the navigation system. Published studies and this study provide evidence of the feasibility and usefulness of this technique in assessing proximity to the CST in regard to the minimal current threshold needed to elicit motor responses. No seizures were reported in this patient series, which was consistent with the fact that monopolar multipulse stimulation involves less epileptogenic effect compared with bipolar stimulation. This is the first study to correlate direct monopolar fiber MEP recordings with neurological outcome. In 4 (57%) of the 7 patients who developed a new or worsenj Neurosurg / Volume 114 / March 2011 ing neurological deficit, the probe distance was 4 mm from the CST, and the minimum stimulation threshold was 5 ma, whereas in the 5 patients who did not develop a new neurological deficit the measured distances of the probe to the CST were 6 mm and the minimum stimulation threshold was 10 ma. In this small group of patients there was a trend toward worsening neurological deficits if the probe distance was short and a small minimum stimulation threshold was recorded, indicating close proximity of the CST to the resection margins. All patients with immediate postoperative neurological deficits continued to recover with deficits persisting in 2 (17%) of 12 patients at the 3-month follow-up. Given the proximity of these tumors to the CSTs we were able to achieve a complete resection (GTR) in 11 (92%) of 12 cases. We believe that combined utilization of subcortical stimulation and intraoperative navigation will give the surgeon added confidence to achieve a more than satisfactory tumor resection in eloquent areas of the brain. Conclusions The use of the monopolar probe can be safely implemented in an imr imaging environment both as a navigation device and a stimulation probe in the resection of intrinsic brain tumors. We report a trend toward worsening neurological deficits if the distance from the probe to the CST is short (< 5 mm), indicating close proximity of the resection cavity to the CST. Implementing anatomical (DTI maps) and functional (fiber MEP) information in the imr imaging environment may alert the surgeon to minimize adverse neurological outcomes. Although these are important intraoperative advancements for the safe resection of brain tumors, there is no substitute for sound surgical judgment during tumor resection in eloquent brain areas. 725

8 S. S. Prabhu et al. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Prabhu, Rao. Ac quisition of data: Prabhu, Gasco, Tummala. Analysis and interpretation of data: Prabhu, Rao. Drafting the article: Prabhu, Rao. Critically revising the article: all authors. Reviewed final version of the manuscript and approved it for submission: all authors. Study supervision: Prabhu. References 1. Berman JI, Berger MS, Chung SW, Nagarajan SS, Henry RG: Accuracy of diffusion tensor magnetic resonance imaging tractography assessed using intraoperative subcortical stimulation mapping and magnetic source imaging. J Neurosurg 107: , Duffau H, Capelle L, Denvil D, Sichez N, Gatignol P, Taillandier L, et al: Usefulness of intraoperative electrical subcortical mapping during surgery for low-grade gliomas located within eloquent brain regions: functional results in a consecutive series of 103 patients. J Neurosurg 98: , Duffau H, Lopes M, Arthuis F, Bitar A, Sichez JP, Van Effenterre R, et al: Contribution of intraoperative electrical stimulations in surgery of low grade gliomas: a comparative study between two series without ( ) and with ( ) functional mapping in the same institution. J Neurol Neurosurg Psychiatry 76: , Gasco J, Tummala S, Mahajan NM, Weinberg JS, Prabhu SS: Simultaneous use of functional tractography, neuronavigationintegrated subcortical white matter stimulation and intraoperative magnetic resonance imaging in glioma surgery: technical note. Stereotact Funct Neurosurg 87: , Gasser T, Ganslandt O, Sandalcioglu E, Stolke D, Fahlbusch R, Nimsky C: Intraoperative functional MRI: implementation and preliminary experience. Neuroimage 26: , Hall WA, Liu H, Martin AJ, Truwit CL: Intraoperative magnetic resonance imaging. Top Magn Reson Imaging 11: , Hall WA, Truwit CL: Intraoperative MR-guided neurosurgery. J Magn Reson Imaging 27: , Hatiboglu MA, Weinberg JS, Suki D, Rao G, Prabhu SS, Shah K, et al: Impact of intraoperative high-field magnetic resonance imaging guidance on glioma surgery: a prospective volumetric analysis. Neurosurgery 64: , Kamada K, Todo T, Masutani Y, Aoki S, Ino K, Takano T, et al: Combined use of tractography-integrated functional neuronavigation and direct fiber stimulation. J Neurosurg 102: , Kamada K, Todo T, Ota T, Ino K, Masutani Y, Aoki S, et al: The motor-evoked potential threshold evaluated by tractography and electrical stimulation. Clinical article. J Neurosurg 111: , Keles GE, Lundin DA, Lamborn KR, Chang EF, Ojemann G, Berger MS: Intraoperative subcortical stimulation mapping for hemispherical perirolandic gliomas located within or adjacent to the descending motor pathways: evaluation of morbidity and assessment of functional outcome in 294 patients. J Neurosurg 100: , Kim SS, McCutcheon IE, Suki D, Weinberg JS, Sawaya R, Lang FF, et al: Awake craniotomy for brain tumors near eloquent cortex: correlation of intraoperative cortical mapping with neurological outcomes in 309 consecutive patients. Neurosurgery 64: , Kinoshita M, Goto T, Okita Y, Kagawa N, Kishima H, Hashimoto N, et al: Diffusion tensor-based tumor infiltration index cannot discriminate vasogenic edema from tumor-infiltrated edema. J Neurooncol 96: , Kinoshita M, Yamada K, Hashimoto N, Kato A, Izumoto S, Baba T, et al: Fiber-tracking does not accurately estimate size of fiber bundle in pathological condition: initial neurosurgical experience using neuronavigation and subcortical white matter stimulation. Neuroimage 25: , Kombos T, Süss O, Vajkoczy P: Subcortical mapping and monitoring during insular tumor surgery. Neurosurg Focus 27(4):E5, Mikuni N, Okada T, Nishida N, Taki J, Enatsu R, Ikeda A, et al: Comparison between motor evoked potential recording and fiber tracking for estimating pyramidal tracts near brain tumors. J Neurosurg 106: , Neuloh G, Pechstein U, Cedzich C, Schramm J: Motor evoked potential monitoring with supratentorial surgery. Neurosurgery 54: , Nimsky C, Ganslandt O, Hastreiter P, Wang R, Benner T, Sorensen AG, et al: Intraoperative diffusion-tensor MR imaging: shifting of white matter tracts during neurosurgical procedures initial experience. Radiology 234: , Nimsky C, Ganslandt O, Merhof D, Sorensen AG, Fahlbusch R: Intraoperative visualization of the pyramidal tract by diffusion-tensor-imaging-based fiber tracking. Neuroimage 30: , Nunez PL, Srinivasan R: Electric Fields of the Brain: The Neurophysics of EEG, ed 2. New York: Oxford University Press, Ojemann G, Ojemann J, Lettich E, Berger M: Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigation in 117 patients. J Neurosurg 71: , Ozawa N, Muragaki Y, Nakamura R, Iseki H: Identification of the pyramidal tract by neuronavigation based on intraoperative diffusion-weighted imaging combined with subcortical stimulation. Stereotact Funct Neurosurg 87:18 24, Sala F, Lanteri P: Brain surgery in motor areas: the invaluable assistance of intraoperative neurophysiological monitoring. J Neurosurg Sci 47:79 88, Stadlbauer A, Nimsky C, Gruber S, Moser E, Hammen T, Engelhorn T, et al: Changes in fiber integrity, diffusivity, and metabolism of the pyramidal tract adjacent to gliomas: a quantitative diffusion tensor fiber tracking and MR spectroscopic imaging study. AJNR Am J Neuroradiol 28: , Suess O, Suess S, Brock M, Kombos T: Intraoperative electrocortical stimulation of Brodman area 4: a 10-year analysis of 255 cases. Head Face Med 2:20, Taniguchi M, Cedzich C, Schramm J: Modification of cortical stimulation for motor evoked potentials under general anesthesia: technical description. Neurosurgery 32: , Weingarten DM, Asthagiri AR, Butman JA, Sato S, Wiggs EA, Damaska B, et al: Cortical mapping and frameless stereotactic navigation in the high-field intraoperative magnetic resonance imaging suite. Technical note. J Neurosurg 111: , Yamaguchi F, Takahashi H, Teramoto A: Navigation-assisted subcortical mapping: intraoperative motor tract detection by bipolar needle electrode in combination with neuronavigation system. J Neurooncol 93: , 2009 Manuscript submitted April 7, Accepted September 13, Please include this information when citing this paper: published online October 22, 2010; DOI: / JNS Address correspondence to: Sujit S. Prabhu, M.D., F.R.C.S.(Ed), The University of Texas M. D. Anderson Cancer Center, Department of Neurosurgery, 1515 Holcombe Boulevard, Unit 442, Houston, Texas sprabhu@mdanderson.org. 726 J Neurosurg / Volume 114 / March 2011