ldentification of the primary motor area by three-dimensional reconstruction of the corticospinal tract using diffusion tensor imaging＊＊★

Guojun Deng, Luo Li, Guimei Chen, Erming Zeng, Xiangzuo Xiao, Meihua Li, Tao Hong,Donghai Li

1Department of Neurosurgery, First Affiliated Hospital, Nanchang University, Nanchang 330006, Jiangxi Province, China

2Department of Radiology, First Affiliated Hospital, Nanchang University, Nanchang 330006, Jiangxi Province, China

lNTRODUCTlON

It is important to identify the central cortex during tumor resection. Many functional brain mapping techniques have been used to identify the primary motor areas. Cortical stimulation has been the “gold standard” method for identification of the central sulcus[1-2]. Blood oxygen level-dependent functional MRI (fMRI)is the most widely used noninvasive technique to identify the locations of cortical primary motor areas (PMA) preoperatively[3-4]. The site of activation on fMRI has been shown to correlate well with the site of the PMA identified by intraoperative cortical stimulation[5-6]. Therefore,fMRI has been increasingly used by neurosurgeons in surgical procedures[3,7-8].

However, patients with cortical function impairment, such as those with hemiplegia, can rarely achieve hand clenching, a typical fMRI task for central sulcus identification, and the method is also of limited use in uncooperative children[9]. Therefore, it is important to develop a new method for identifying the PMA quickly in these patients. Conventional MRI T1- and T2-weighted images provide high contrast between gray and white matter, but they do not provide any information about the paths followed by fibers. By contrast, diffusion tensor imaging (DTI) provides information about white matter structure in vivo and has been used to show the relationships among the major white matter tracts in patients with brain tumors[10-12].

Diffusion tensor tractography, a method based on DTI, allows us to visualize the major white matter tracts, including the corticospinal tract(CST), in three dimensions. The present study identified the PMA using corticospinal tractography and confirmed this method’s reliability by fMRI.

RESULTS

Quantitative analysis of participants

A total of 20 patients were included in the final analysis.

Patient data

The clinical characteristics of the patients are summarized in Table 1.

Table 1 Clinical information of patients

ldentification of the primary motor area by DTl

Figure 1 Primary motor areas identified by functional MRI (fMRI) activation signals and corticospinal tract (CST)fiber tracking.

All patients completed the fMRI motor activation task successfully. Significant functional activation areas were observed in the PMAs of all 20 subjects. fMRI activation signals were distributed mainly in the contralateral central sulcus around the omega-shaped hand knob(Figure 1A). Three-dimensional reconstructions showed that PMA activations occurred mainly in the middle portion of the precentral gyrus. The CST consistently propagated from the pons and cerebral peduncle to the suspected PMA location (Figure 1B). PMA was located and CST fiber tracking results were combined with fMRI activation signals within three-dimensional anatomic illustrations (Figure 1C). Partial fibers associated with hand movement were located within fMRI activation signals and the other fibers were located in the medial portion of the precentral gyrus on the coronal view(Figure 1D). There was a good correlation between CST fiber tracking results and fMRI activation signals in terms of their abilities to identify the PMA (supplementary Figures 1-3 online).

DlSCUSSlON

DTI, an MRI technique, measures the three-dimensional Brownian motion of water molecules and evaluates the integrity of white matter tracts by virtue of its ability to visualize water diffusion characteristics in the living human brain. Diffusion tensor tractography visualizes the major white matter tracts in three dimensions by setting regions of interest (ROIs) through which fibers pass. The results of fiber tractography of major white matter tracts are consistent with those of postmortem anatomic studies[13]and fiber tracking experiments have led to a considerable understanding of neuroanatomy[14]. Brain atlases are useful for teaching and clinical studies, and DTI with fiber tractography can be used to map white matter tracts. It is important to identify the motor cortex before surgery, especially when a tumor lies close to the motor cortex area. This study used corticospinal tractography to identify the location of the PMA in 20 patients with deep-seated brain tumors. fMRI activation signals were distributed mainly in the contralateral central sulcus around the omega-shaped hand knob.

The good correlation between the results obtained using these methods suggest that DTI is a useful brain mapping technique for motor cortex identification.

A crucial parameter for fiber tracking is the definition of the appropriate ROI. Two ROIs were set in the brainstem for CST fiber tracking in this study: one at the level of the pons and the other at the level of the cerebral peduncle.

As a classic anatomical landmark, the brainstem is easily recognized in color fractional anisotropy maps and easy to operate on; therefore, setting ROIs within the brainstem ensured the validity of the CST fiber tracking findings. Some investigators have selected the posterior limb of the internal capsule or the precentral gyrus as an ROI[15-18]. However, it proved difficult to set ROIs in the internal capsule or precentral gyrus in this study. First,these areas are often displaced by the mass effects of cerebral tumors, which would render such ROI placements unreliable. Second, it is difficult to distinguish between the CST and somatosensory tracts in color fractional anisotropy maps. Fiber tracking experiments are sensitive to noise, partial volume effects, and ROI size and location[18-20]. Kamada et al[9]reconstructed the CST by placing a single ROI at the cerebral peduncle to locate the PMA. However, because there was only one ROI, and because of the size of the ROI used, these authors’ CST fiber tracking results maybe indicate“crossing” and “kissing” fibers, or even incorrect fibers.

The CST is a tight fiber bundle for portions of its traverse.

However, there are several portions where neighboring fibers could interfere with its course[21]. Fiber tracking optimizes the localization of white matter tracts in the brain by minimizing the influence of individual variation in ROI placement[15,22]and by increasing the reliability of fiber tracking results from studies that use multiple(at least two) ROIs[23]. In this study, the first ROI was placed at the superior level of the ventral pons. The fiber tracts that were traced included the CST, corticopontine tract, and other “crossing” or “kissing” fibers. By adding a second ROI at the level of the cerebral peduncle, and retrieving only fibers that penetrated both ROIs,trajectories outside the two ROIs were removed, thus decreasing the effect of the size of the first ROI (Figure 2,supplementary Figure 4 online).

Figure 2 Fiber tracking results obtained using two regions of interest (ROIs).

Fractional anisotropy threshold values are an important parameter for accurate fiber tracking[24-25]. Tracking results vary according to the fractional anisotropy threshold[26]. A high fractional anisotropy threshold may lead to failure of fiber tracking, while a low fractional anisotropy threshold may introduce many erroneous tracts. This can cause significant problems in fiber tracking experiments, with fibers being shown to terminate prematurely due to crossing the chosen minimal fractional anisotropy threshold. In this study, the CST fiber tracking findings in individuals with brain tumors showed CST discontinuity when the fractional anisotropy termination criterion was set at 0.3. However,the CST could be visualized when the fractional anisotropy was set at 0.2. Thus, brain tumors may change the diffusion characteristics of the CST and decrease anisotropy, leading to a loss of integrity in the appearance of the fiber tracking. Using a fractional anisotropy threshold of 0.2, which is considered the optimal threshold for fiber tracking of the CST[24,26], the CST was reconstructed successfully and the PMA was identified in all subjects. There was a good correlation between CST fiber tracking results and fMRI activation signals in terms of their abilities to identify the PMA.

fMRI is based on the principle of blood oxygen level-dependent changes. The activation signals in the present study were distributed in the central sulcus around the omega-shaped hand knob. The precentral gyrus is located in front of the activation signal[27-28].

Diffusion tensor tractography of the CST not only enables the PMA to be identified precisely, but also shows its subcomponents (including fibers traveling to the hand, foot and other areas). Our findings showed that hand fibers run to the lateral part of the precentral gyrus,consistent with the fMRI activation signal, but other fibers,possibly foot fibers, run to the medial part of the precentral gyrus. The difference between fMRI and CST findings may result from our functional task, which consisted only of hand movements. The results from this study indicate that DTI can be used to correctly identify the PMA. Being able to identify the PMA by CST is of great importance for paralyzed patients and uncooperative child patients.

While DTI-based tractography is considered to be a promising technique for displaying white matter fibers, it has many limitations. First, technical factors such as poor spatial resolution and the low signal-to-noise ratio of the acquired image lead to poor DTI and fiber tracking.

Second, fiber tracking is a user-defined process, and the results are dependent on the size and location of the seed ROIs, as well as the fractional anisotropy threshold.

Manual setting of ROIs is also subjective. The last limitation is that crossing fibers and kissing fibers will affect the accuracy of fiber tracking experiments. With the development of new technology, the application of high magnetic field equipment, and the improvement of fiber tracking methods, these problems will gradually be resolved.

SUBJECTS AND METHODS

Design

A neuroimaging, observational study.

Time and setting

The study was performed at the First Affiliated Hospital of Nanchang University in China from April 2008 to August 2009.

Subjects

Twenty deep-seated brain tumor patients with motor cortex or motor fiber involvement, including 9 males and 11 females, aged 13-63 years (37.5 ± 14.0 years), were included in this study. Extrinsic brain tumor patients were excluded. Lesions included astrocytoma (n = 9),oligodendroglioma (n = 3), glioblastoma (n = 4),medulloblastoma (n = 1), ependymoma (n = 1),hemangioblastoma (n = 1), and inflammatory granulation(n = 1). All patients were conscious and able to complete the fMRI motor activation task as a prerequisite for participation in this study.

Methods

Data acquisition

All imaging studies were performed on a 3.0T Siemens magnetic resonance system (Siemens, Erlangen,Germany). A 12-channel head coil was used to measure signal intensity. For anatomical reference, we used a T1-weighted three-dimensional magnetization-prepared rapid acquisition gradient-echo sequence with a repetition time (TR) of 1 900 ms, echo time (TE) of 2.26 ms, matrix of 192 × 256, field of view (FOV) of 230 mm × 230 mm and slice thickness of 1.0 mm.

Acquisition time was 4 minutes and 20 seconds. For functional imaging, a T2-weighted gradient-echo echo-planar imaging sequence (TR/TE, 3 000 ms/30 ms;matrix, 128 × 128; FOV, 192 mm × 192 mm; slice thickness, 3.0 mm, 36 contiguous sections) was used.

The motor activation task consisted of hand clenching movements on the affected side[29]. Subjects were instructed on the task and were allowed to practice briefly before imaging. Each series comprised six cycles of task performance (30 seconds) and rest (30 seconds),with a 3-second interval between task performance and rest. Acquisition time was 3 minutes and 16 seconds. For DTI, we used a single-shot diffusion-weighted spin-echo echo-planar imaging sequence (TR/TE, 3 000 ms/93 ms;matrix, 128 × 128; FOV, 230 mm × 230 mm, slice thickness, 5.0 mm). A total of 20 slices with no intersectional gap and an isotropic voxel size of 1.8 mm ×1.8 mm × 5.0 mm were measured. The b value was 1 000 s/mm2in 20 noncollinear directions. Acquisition time was 9 minutes and 46 seconds.

Data analysis

就當前鄉村實際來說，全國各地區發展是不均衡的，即便是同一地區，平原、山區、丘陵等不同區域在區位、交通、資源、功能等方面也存在差異，其發展路徑和進度必然無法同步。因此，按照鄉村振興戰略目標任務要求，在具體的實踐路徑上，應注意整體推進與重點突破的結合。

All fMRI and DTI data were processed and analyzed using software provided with the Siemens work station(Leonardo Workstation, Syngo 2008A, Erlangen,Germany). fMRI data were analyzed using a blood oxygen level-dependent fMRI software package. The first two data points from every phase were ignored automatically by the system. The significance of differences between active data and baseline data were automatically calculated by the Student’s t-test, with an absolute value of t ＞ 4.0. Differential maps were overlaid on the high-resolution T1-weighted structural images for neuroanatomical correlation of activation. The distribution characteristics of activation signals were analyzed.

The diffusion weighted images were smoothed with a three-dimensional Gaussian filter and image resolution was enhanced using interpolation methods and parallel processing. Fractional anisotropy maps were generated automatically. In the DTI color maps, red, green, and blue colors were assigned to right-left, anterior-posterior, and superior-inferior orientations, respectively.

Fiber tracking was performed with fiber assignment by the continuous tractography method[15,30-32]. Tracking stops at predefined thresholds of fractional anisotropy and turning angle to limit the detection of spurious fibers.

A fractional anisotropy threshold of 0.2 and a turning angle threshold of 60° were chosen. The two ROIs approach was used to reconstruct the CST exploiting existing anatomical knowledge of tract trajectories. Two ROIs were set within the brainstem for CST fiber tracking:one at the level of the pons and the other at the level of the cerebral peduncle (Figure 3). To include only the region of the CST, the typical ROI size was set between 5 and 10 voxels. Fiber tracts that passed through both ROIs were designated as the final tract of interest.

After image registration, the CST fiber tracking results and fMRI activation signals were merged with three-dimensional anatomic MRI results. The consistency of identifying the PMA by CST and fMRI was analyzed.

Figure 3 Regions of interest (ROIs) for corticospinal tract(CST) tracking. C: Caudate nucleus; P: putamen; Th:thalamus; VP: ventral posterolateral nucleus.

Author contributions:Donghai Li designed the study and was responsible for the experimental procedures and results. All fMRI and DTI scans were performed by Guimei Chen. Data were collected by Tao Hong, Meihua Li, Erming Zeng and Luo Li. Data analysis was completed by Guojun Deng and Xiangzuo Xiao. The manuscript was drafted by Guojun Deng.

Funding:This study was supported by the Science and Research Project of Jiangxi Provincial Department of Science and Technology, No. 07-1012; a grant from the Jiangxi Provincial Department of Education, No. GJJ08116.

Ethical approval:Informed consents were obtained from all subjects. The study was approved by the Ethics Committee of Nanchang University in China.

Supplementary information:Supplementary data associated with this article can be found, in the online version, by visiting www.nrronline.org, and entering Vol. 6, No. 14, 2011 after selecting the “NRR Current Issue” button on the page.

[1]Kombos T, Suess O, Funk T, et al. Intra-operative mapping of the motor cortex during surgery in and around the motor cortex. Acta Neurochir (Wien). 2000;142(3):263-268.

[2]Towle VL, Khorasani L, Uftring S, et al. Noninvasive identification of human central sulcus: a comparison of gyral morphology,functional MRI, dipole localization, and direct cortical mapping.Neuroimage. 2003;19(3):684-697.

[3]Sunaert S, Yousry TA. Clinical applications of functional magnetic resonance imaging. Neuroimaging Clin N Am. 2001;11(2):221-236, viii.

[4]Haughton VM, Turski PA, Meyerand B, et al. The clinical applications of functional MR imaging. Neuroimaging Clin N Am.1999;9(2):285-293.

[5]Roberts TP, Ferrari P, Perry D, et al. Presurgical mapping with magnetic source imaging: comparisons with intraoperative findings. Brain Tumor Pathol. 2000;17(2):57-64.

[6]Roessler K, Donat M, Lanzenberger R, et al. Evaluation of preoperative high magnetic field motor functional MRI (3 Tesla) in glioma patients by navigated electrocortical stimulation and postoperative outcome. J Neurol Neurosurg Psychiatry. 2005; 76(8):1152-1157.

[7]Overduin SA, Servos P. Distributed digit somatotopy in primary somatosensory cortex. Neuroimage. 2004;23(2):462-472.

[8]Blatow M, Nennig E, Durst A, et al. fMRI reflects functional connectivity of human somatosensory cortex. Neuroimage. 2007;37(3):927-936.

[9]Kamada K, Sawamura Y, Takeuchi F, et al. Functional identification of the primary motor area by corticospinal tractography. Neurosurgery. 2005;56(1 Suppl):98-109.

[10]Ulmer JL, Salvan CV, Mueller WM, et al. The role of diffusion tensor imaging in establishing the proximity of tumor borders to functional brain systems: implications for preoperative risk assessments and postoperative outcomes. Technol Cancer Res Treat. 2004;3(6):567-576.

[11]Nimsky C, Ganslandt O, Hastreiter P, et al. Preoperative and intraoperative diffusion tensor imaging-based fiber tracking in glioma surgery. Neurosurgery. 2005;56(1):130-138.

[12]Nimsky C, Ganslandt O, Merhof D, et al. Intraoperative visualization of the pyramidal tract by diffusion-tensor-imagingbased fiber tracking. Neuroimage. 2006;30(4):1219-1229.

[13]Wakana S, Jiang H, Nagae-Poetscher LM, et al. Fiber tract-based atlas of human white matter anatomy. Radiology. 2004;230(1):77-87.

[14]Bürgel U, Amunts K, Hoemke L, et al. White matter fiber tracts of the human brain: three-dimensional mapping at microscopic resolution, topography and intersubject variability. Neuroimage.2006;29(4):1092-1105.

[15]Wakana S, Caprihan A, Panzenboeck MM, et al. Reproducibility of quantitative tractography methods applied to cerebral white matter. Neuroimage. 2007;36(3):630-644.

[16]Yang DS, Hong JH, Byun WM, et al. Identification of the medial lemniscus in the human brain: combined study of functional MRI and diffusion tensor tractography. Neurosci Lett. 2009;459(1):19-24.

[17]Danielian LE, Iwata NK, Thomasson DM, et al. Reliability of fiber tracking measurements in diffusion tensor imaging for longitudinal study. Neuroimage. 2010;49(2):1572-1580.

[18]Schimrigk SK, Bellenberg B, Schlüter M, et al. Diffusion tensor imaging-based fractional anisotropy quantification in the corticospinal tract of patients with amyotrophic lateral sclerosis using a probabilistic mixture model. AJNR Am J Neuroradiol. 2007;28(4):724-730.

[19]Ozturk A, Sasson AD, Farrell JA, et al. Regional differences in diffusion tensor imaging measurements: assessment of intrarater and interrater variability. AJNR Am J Neuroradiol. 2008;29(6):1124-1127.

[20]Farrell JA, Landman BA, Jones CK, et al. Effects of signal-tonoise ratio on the accuracy and reproducibility of diffusion tensor imaging-derived fractional anisotropy, mean diffusivity, and principal eigenvector measurements at 1.5 T. J Magn Reson Imaging. 2007;26(3):756-767.

[21]Holodny AI, Watts R, Korneinko VN, et al. Diffusion tensor tractography of the motor white matter tracts in man: Current controversies and future directions. Ann N Y Acad Sci. 2005;1064:88-97.

[22]Partridge SC, Mukherjee P, Berman JI, et al. Tractography-based quantitation of diffusion tensor imaging parameters in white matter tracts of preterm newborns. J Magn Reson Imaging. 2005;22(4):467-474.

[23]Huang H, Zhang J, van Zijl PC, et al. Analysis of noise effects on DTI-based tractography using the brute-force and multi-ROI approach. Magn Reson Med. 2004;52(3):559-565.

[24]Stadlbauer A, Nimsky C, Buslei R, et al. Diffusion tensor imaging and optimized fiber tracking in glioma patients: Histopathologic evaluation of tumor-invaded white matter structures. Neuroimage.2007;34(3):949-956.

[25]Son SM, Ahn YH, Sakong J, et al. Diffusion tensor imaging demonstrates focal lesions of the corticospinal tract in hemiparetic patients with cerebral palsy. Neurosci Lett. 2007;420(1):34-38.

[26]Kunimatsu A, Aoki S, Masutani Y, et al. The optimal trackability threshold of fractional anisotropy for diffusion tensor tractography of the corticospinal tract. Magn Reson Med Sci. 2004;3(1):11-17.

[27]Li DH, Gong HH, Xiao XZ, et al. Functional MRI activation of primary and secondary motor areas in healthy subjects. Neural Regen Res. 2008;3(2):138-141.

[28]Shinoura N, Suzuki Y, Yamada R, et al. Precentral knob corresponds to the primary motor and premotor area. Can J Neurol Sci. 2009;36(2):227-233.

[29]Schaechter JD, Perdue KL, Wang R. Structural damage to the corticospinal tract correlates with bilateral sensorimotor cortex reorganization in stroke patients. Neuroimage. 2008;39(3):1370-1382.

[30]Mori S, Crain BJ, Chacko VP, et al. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging.Ann Neurol. 1999;45(2):265-269.

[31]Holodny AI, Gor DM, Watts R, et al. Diffusion-tensor MR tractography of somatotopic organization of corticospinal tracts in the internal capsule: initial anatomic results in contradistinction to prior reports. Radiology. 2005;234(3):649-653.

[32]M?ller M, Frandsen J, Andersen G, et al. Dynamic changes in corticospinal tracts after stroke detected by fibretracking. J Neurol Neurosurg Psychiatry. 2007;78(6):587-592.