Spatial Profiling of the Corticospinal Tract in Amyotrophic Lateral Sclerosis Using Diffusion Tensor Imaging

John C.T. Wong, BSc Luis Concha, MD Christian Beaulieu, PhD Wendy Johnston, MD Peter S. Allen, PhD Sanjay Kalra, MD

ABSTRACT Background and Purpose: Diffusion tensor imaging (DTI) was used as a noninvasive method to evaluate the anatomy of the corticospinal tract (CST) and the pattern of its degeneration in amyotrophic lateral sclerosis (ALS). Methods. Fourteen patients with ALS and 15 healthy controls underwent DTI. Parameters reflecting coherence of diffusion (fractional anisotropy, FA), bulk diffusion (apparent diffusion coefficient, ADC), and directionality of diffusion (eigenvalues) parallel to (λ  ) or perpendicular to (λ ⊥ ) fiber tracts were measured along the intracranial course of the CST. Results: FA and λ  increased, and ADC and λ ⊥ decreased progressively from the corona radiata to the cerebral peduncle in all subjects. The most abnormal finding in patients with ALS was reduced FA in the cerebral peduncle contralateral to the side of the body with the most severe upper motor neuron signs. λ  was increased in the corona radiata. Internal capsule FA correlated positively with symptom duration, and cerebral peduncle ADC positively with the Ashworth spasticity score. Conclusion: There is a spatial dependency of diffusion parameters along the CST in healthy individuals. Evidence of intracranial CST degeneration in ALS was found with distinct diffusion changes in the rostral and caudal regions.

Received August 18, 2006, and in revised form November 13, 2006. Accepted for publication November 17, 2006. From the Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada ( JCTW); Division of Neurology, Department of Medicine, University of Alberta, Edmonton, Alberta, Canada (WJ, SK); and Department of Biomedical Engineering, University of Alberta, Edmonton, Alberta, Canada (LC, CB, PSA). Address correspondence to Sanjay Kalra, MD, 2E3.18 Walter C Mackenzie Health Sciences Centre, 8440-112 Street, Edmonton, Alberta, T6G 2B7, Canada. E-mail: [email protected].

234

Key words: imaging.

Amyotrophic lateral sclerosis, diffusion tensor

Wong JCT, Concha L, Beaulieu C, Johnston W, Allen PS, Kalra S. Spatial profiling of the corticospinal tract in amyotrophic lateral sclerosis using diffusion tensor imaging J Neuroimaging 2007;17:234-240. DOI: 10.1111/j.1552-6569.2007.00100.x

Introduction The hallmark of amyotrophic lateral sclerosis (ALS), a progressive neurodegenerative condition of unknown etiology, is the combination of physical exam findings reflecting lower motor neuron (LMN) and upper motor neuron (UMN) degeneration. LMN dysfunction can be evaluated by electrodiagnostic techniques, including electromyography and motor unit number estimation. However, the extent and nature of UMN involvement is difficult to characterize clinically due to limitations of the neurological examination or the presence of severe simultaneous LMN signs. Indeed, pathological evidence of corticospinal tract (CST) degeneration has been demonstrated in patients who are lacking UMN signs.1 A sensitive marker of UMN damage is required to gain further insight into the pathogenesis of the disease, to allow for improved disease detection and monitoring, and to aid in the evaluation of therapeutic agents. Magnetic resonance imaging (MRI) unfortunately affords poor sensitivity and specificity for degeneration2 and thus remains a tool to rule out other disorders. Studies continue to explore the potential of other promising techniques to provide a biomarker of cerebral degeneration. Abnormalities in cerebral neurochemistry and electrophysiological properties have consistently been demonstrated using magnetic resonance spectroscopy and transcranial magnetic stimulation, respectively.3-5

C 2007 by the American Society of Neuroimaging Copyright ◦

Diffusion tensor imaging (DTI) is an MRI technique that has emerged as a tool to visualize the organization and integrity of white matter bundles. This is made possible by collecting images with diffusion sensitizing gradients such that the diffusion properties of water molecules can be characterized in vivo. The pattern of diffusion is dependent on water’s anatomical localization. In nerve fibers, the net motion of water is greater parallel to the longitudinal axis of the axons than perpendicular to them.6 This preferential directionality of diffusion is attributed to physical restrictions imposed by axonal membranes and their myelin sheaths.7 The degree of this directionality in three spatial dimensions can be quantified in each voxel by the index of fractional anisotropy (FA). The minimum FA of zero corresponds to no directional dependence of diffusion (ie, isotropic diffusion, water displacement is random in all directions), whereas the maximum FA of one indicates diffusion is anisotropic (ie, diffusion is highly biased in one direction compared to the other two). The trace apparent diffusion coefficient (ADC) provides no directional information on water’s diffusion, rather is simply a measure of the magnitude of bulk diffusion. The principal eigenvalues  1 ,  2 , and  3 reflect the magnitude of diffusion (directional apparent diffusion coefficient) along the fiber tracts ( 1 ) or perpendicular to them ( 2 ,  3 ). Applying region of interest (ROI) analysis, Ellis et al first demonstrated a significant reduction of FA and an increase in mean diffusivity in the internal capsule of patients with ALS.8 Subsequent studies have expanded to study FA and ADC in the CST at the motor cortex, corona radiata, cerebral peduncles, pons, and pyramids, sometimes with inconsistent results. Discrepancies reported between studies may in part be attributed to the high variability of CNS architecture along the CST9,10 and the variable selection between studies of the regions studied within the anatomical structure of interest. To circumvent this, groups have encompassed the CST with ROIs drawn on contiguous axial slices; in effect, analyzing the CST in a three-dimensional manner.10,11 A downfall of this technique is that depending on slice thickness and the extent of the CST one wishes to study, this method can be laborious, time consuming, and prone to user-dependent errors. The objective of this study was to evaluate the integrity of the CST in ALS using a reproducible ROI-based DTI approach. We used an alternative method of studying the CST in three dimensions by encompassing it in a coronal plane. This would permit capturing a large portion of the tract in a more convenient manner than drawing ROIs on multiple axial slices. In contrast to most previous studies,

we also measured eigenvalues to better understand diffusion characteristics. Our hypothesis was that patients with ALS would show water diffusion abnormalities along the CST, providing an indirect marker of degeneration. Methods Subjects and Clinical Evaluation Patients (n = 14) were recruited from the ALS Clinic at the University of Alberta and were required to have less than 5 years of symptoms and meet El Escorial criteria for “probable” or “definite” ALS.12 According to these criteria, all subjects would have examination findings reflecting both LMN and UMN dysfunction. Healthy age-matched controls (n = 15) were free of neurological or psychiatric disease. All subjects gave informed consent and the study was approved by the Human Research Ethics Board. Subject details are given in Table 1. Patients were administered the ALS Functional Rating Scale (ALSFRS) to assess general disability. Clinical signs of UMN involvement were quantified on each side with the modified Ashworth spasticity scale for the upper and lower limb, and finger and foot tapping speeds (number of taps in 10 seconds averaged over two attempts).13 Image Acquisition MR imaging was performed on a Siemens Sonata 1.5 Tesla scanner. T2-weighted images 5-mm thick were acquired in sagittal (26 contiguous slices, repetition time [TR] 6860 msec, echo time [TE] 112 msec, matrix 256 × 216, field of view [FOV] 260 × 220 mm), axial (25 contiguous slices, TR 6950 msec, TE 113 msec) and coronal (25 contiguous slices, TR 7480 msec, TE 113 msec) planes. Diffusion tensor images were acquired using spinecho echo planar imaging in coronal orientation (20 contiguous slices, 5-mm thick, TR 3200 msec, TE 88 msec, matrix 128 × 128 zero-filled to 256 × 256, 75% phase Table 1.

Clinical Characteristics of Patients and Healthy Control Subjects

N M:F Age (years) Symptom duration (months) Limb:Bulbar Onset ALS functional rating scale (range 0-40)

ALS

Control

14 5:9 53 ± 14 22 ± 12 (18, 10-43) 12:2 30 ± 6 (31, 15-37)

15 8:7 54 ± 12

Values are mean ± standard deviation (median, range).

Wong et al: Spatial Profiling Using DTI in ALS

235

partial Fourier, FOV 220 × 220 mm, six diffusion gradient directions, b = 1000 sec/mm2 , eight averages, scan time 3:04 minutes). Coronal images were angulated to lie parallel to the CST as identified on sagittal images. Angulation was refined by intersection of a slice through the cerebral peduncles and the hyperintense signal of the CST in the posterior limb of the internal capsule as visualized on axial images. Diffusion Tensor Image Data Processing DTI images were processed on a PC running DTIstudio ( Johns Hopkins University, Baltimore, MD). Quantitative diffusion parameter maps were created, including fractional anisotropy (FA), mean apparent diffusion coefficient (ADC), parallel diffusivity (  =  1 ), and perpendicular diffusivity ( ⊥ = [ 2 +  3 ]/2), as well as a color map encoding the principal diffusivity of the tracts (ie, the eigenvector associated with   ). Diffusion Tensor Imaging Analysis The coronal slice containing the greatest volume of the CST as viewed on both the color-coded and FA maps was selected for ROI placement. The right and left CST of each subject were individually traced by a single operator ( JCTW) blinded to diagnosis. The rostral and caudal borders of this ROI were set at the superior margin of the superior longitudinal fascicle and at the transition of the cerebral peduncles to the pons, respectively. The lateral borders were defined by the extent of the CST itself. Each CST was further segmented into three subregions: corona radiata (rostral margin to inferior border of lateral ventricle), posterior limb of the internal capsule (inferior border of lateral ventricle to superior edge of red nucleus), and cerebral peduncles (superior edge of red nucleus to caudal margin) (Fig 1). Two different analyses were performed. First, FA, ADC,   , and  ⊥ were plotted at 1-mm intervals along the course of the entire CST within the ROI. To do this, the distance between the rostral and caudal borders of the ROI was normalized amongst all subjects. The mean and standard deviation of a diffusion parameter was then calculated at discrete contiguous cross sections of the CST along this distance. Second, mean FA, ADC,   , and  ⊥ were calculated for each of the three anatomical subregions and for the entire CST. Statistical Analysis Diffusion findings were averaged between left and right hemispheres in controls. This was compared to the CST of ALS patients ipsilateral and contralateral to the most severe UMN findings on examination. To determine if a diffusion parameter was different between these groups,

236

Journal of Neuroimaging Vol 17 No 3 July 2007

Fig 1. A region of interest (ROI) encompasses the right CST from the corona radiata to the cerebral peduncle on a coronal DTI-derived color-coded map. DTI indices were examined along the rostro-caudal course of the CST at 1-mm intervals (Fig 2) and within the labeled anatomical subregions. The colors red, green, and blue represent fibers running in left–right, anterior–posterior, and superior–inferior directions, respectively.

a multivariate analysis of variance with age as a covariate (MANCOVA) was used for each diffusion parameter with dependent variables representing each region (corona radiata, posterior limb of internal capsule, cerebral peduncle, and entire CST). Significant results indicating an abnormal parameter within a region were followed by ANCOVA and post-hoc analyses with the Tukey test to ascertain group effect. Pearson (r) and Spearman rank (R) correlation coefficients were computed to evaluate relationships with clinical measures. Correlations with age, symptom duration, and ALSFRS were performed with a diffusion parameter averaged between sides, since these clinical measures do not have lateral bias. Unilateral correlations were done with diffusion parameters and the contralateral Ashworth score and tapping speeds. Statistical significance was accepted for a two-tailed P < .05. Results Patients and healthy controls did not differ with respect to age or sex (Table 1). The overall pattern of variation in DTI parameters along the rostro-caudal extent of the CST from the corona radiata to the cerebral peduncles was similar between patients and controls (Fig 2). Significant deviations were, however, present at specific regions within the CST (Table 2).

Fig 2.

Spatial variation of diffusion indices along the corticospinal tract. The mean and one standard deviation is plotted along the course of the corticospinal tract from the corona radiata to the cerebral peduncles (see Fig 1). “ALS, Most affected” and “ALS, Least affected” refer to the CST contralateral and ipsilateral to the side with the most severe UMN signs, respectively.

Fractional Anisotropy FA increased progressively from the corona radiata to the cerebral peduncles in patients and controls. It was increased in ALS for most of the extent of the corona radiata on the least-affected side (Fig 2). The plots cross over rostral to the mid-PLIC with FA in patients lower in the remaining extent of the CST to the cerebral peduncles. Quantitative regional analysis (Table 2) revealed significantly reduced FA at the cerebral peduncle contralateral to the most severe UMN signs and a statistical trend to a reduction ipsilateral to the most severe UMN signs.

evations of ADC at the corona radiata and PLIC of the patients did not reach statistical significance (Table 2). Parallel and Perpendicular Diffusivity Parallel diffusivity (  ) for both patient and control groups exhibited a continuous rise from the corona radiata to the cerebral peduncles. It was significantly increased in the corona radiata of patients (Table 2). Perpendicular diffusivity ( ⊥ ) decreased from the rostral to caudal region of the CST in patients and controls. It appeared to be elevated on the most-affected side in ALS patients; however, this failed to reach statistical significance. Clinical Correlations

Apparent Diffusion Coefficient In both patients and controls, ADC decreased from the corona radiata to the midcerebral peduncles before exhibiting a small rise at the caudal cerebral peduncles. El-

Of the diffusion parameters averaged over the left and right side, FA in the internal capsule correlated with symptom duration (r = 0.56, P = .036). Diffusion parameters did not correlate with age or ALSFRS.

Wong et al: Spatial Profiling Using DTI in ALS

237

FA: fractional anisotropy; ADC: apparent diffusion coefficient (×10−3 mm2 /sec);   : parallel diffusivity (×10−3 mm2 /sec);  ⊥ = perpendicular diffusivity (×10−3 mm2 /sec). ∗ indicates values different from controls with P < .05. There was a statistical trend for reduced FA in the cerebral peduncle in ALS patients ipsilateral to the most severe UMN signs (P < .10) compared to controls. For statistical analysis, each diffusion parameter was tested using multivariate analysis of covariance with Tukey’s posthoc pairwise analysis when required.

0.61 ± 0.03 0.76 ± 0.04 1.38 ± 0.06 0.44 ± 0.04 0.59 ± 0.04 0.78 ± 0.04 1.40 ± 0.06 0.46 ± 0.05 0.60 ± 0.04 0.77 ± 0.04 1.40 ± 0.04 0.45 ± 0.05

0.75 ± 0.03 0.74 ± 0.03 1.55 ± 0.08 0.33 ± 0.03 0.72 ± 0.04∗ 0.74 ± 0.05 1.51 ± 0.09 0.36 ± 0.05 0.72 ± 0.03 0.75 ± 0.04 1.51 ± 0.07 0.36 ± 0.05

0.65 ± 0.03 0.75 ± 0.04 1.40 ± 0.07 0.42 ± 0.04 0.64 ± 0.04 0.77 ± 0.05 1.42 ± 0.06 0.44 ± 0.06 0.66 ± 0.04 0.75 ± 0.04 1.42 ± 0.04 0.42 ± 0.05

0.46 ± 0.04 0.78 ± 0.04 1.20 ± 0.07 0.57 ± 0.04 0.47 ± 0.06 0.81 ± 0.05 1.27 ± 0.06∗ 0.59 ± 0.06 0.48 ± 0.05 0.81 ± 0.04 1.27 ± 0.05∗ 0.58 ± 0.06

Corona radiata Internal capsule Cerebral peduncle Entire CST

⊥  ADC  ADC Region

FA

Control (n = 15)

⊥

FA

ADC



⊥

FA

Ipsilateral to Most Severe UMN Signs Contralateral to Most Severe UMN Signs

ALS (n = 14)

Quantitative DTI Analysis by Regions of the Corticospinal Tract in Healthy Controls (n = 15) and ALS Patients (n = 14) Table 2. 238

Journal of Neuroimaging Vol 17 No 3 July 2007

For unilateral analyses, the Ashworth score correlated with the ADC of the contralateral cerebral peduncle (R = 0.46, P = .009) and whole CST (R = 0.36, P = .039). A correlation of the Ashworth score with  ⊥ approached statistical significance in the internal capsule (R = 0.41, P = .065) and the cerebral peduncle (R = 0.48, P = .053). Correlations were not detected with unilateral FA or   , nor with tapping speed. Discussion The observed trends of the various diffusion parameters along the length of the corticospinal tract in all subjects (controls and patients) can be explained with present knowledge of the organization of the CST. Starting rostrally, the corona radiata is a region of converging fibers with the corticofugally oriented fibers of the CST intermingled with U-fibers and fibers of association tracts that are directed in oblique planes to it in both left–right and anterior–posterior directions. Thus, a net incoherence of directionality of diffusion is reflected by a low FA accompanied by similar   and  ⊥ . Descending caudally, the CST becomes progressively more densely packed and homogenous, free of alternately arranged bundles. This is reflected by increasing FA and   , and declining  ⊥ and ADC. The increase of ADC in the cerebral peduncle is perhaps due to inclusion of CST fibers that are beginning to disperse as they enter the pons. In the ALS patients, the most abnormal finding was reduced FA at the cerebral peduncles, indicative of CST degeneration at this level. The 8% increase in  ⊥ , although not statistically significant, is likely responsible for the reduced diffusion anisotropy given that   was intact. This pattern of reduced FA and increased  ⊥ is in agreement with prior studies of the cerebral peduncle14 and internal capsule.15 Several histopathological findings in ALS, including neuronal degeneration, inclusion bodies, cytoskeletal structural abnormalities, and astrocytic gliosis, may each serve as candidates responsible for such observed changes in diffusion parameters. However, the main determinants of anisotropic diffusion are the tight packing of axons and axonal membranes,7 structures that are notably deranged in ALS. Axonal degeneration would increase permeability transversely across the CST, permitting water diffusion with greater ease and increasing  ⊥ . Myelin has a modulating effect on anisotropy16,17 and thus demyelination, which is present to a variable extent in ALS,18 could contribute to some of the observed loss of anisotropy. The correlation of FA in the posterior limb of the internal capsule with symptom duration further supports the potential of these indices as biomarkers, perhaps of

different pathologic processes. The lack of a correlation with the ALSFRS is not unexpected given that many of the functions surveyed with this disability scale are significantly dependent on strength,19 which in turn is dependent heavily on LMN integrity.13 A positive correlation between FA and symptom duration would not seem intuitive; however, this may be reflective of subjects with more aggressive disease who seek medical attention sooner because of rapidly progressing symptoms. Such patients would have advanced disease with low FA and shorter symptom duration. Indeed, shorter symptom duration is predictive of reduced survival.20 Inspection of the high resolution plots (Fig 2) would suggest that ADC may be abnormally increased in ALS, as others have inconsistently found.8,10,11,21,22 That this may have pathological relevance is supported by a correlation of ADC at the cerebral peduncle with the Ashworth spasticity score. Parallel diffusivity and FA were increased in the corona radiata. At areas where coherence of fiber orientation is high, such as the cerebral peduncles, alterations in diffusion indices are attributed to changes in tissue structure. In contrast, at regions where intersecting fibers coexist, such as the corona radiata, structural and “architectural” features have competing influences.9 Subsequent to CST degeneration at the corona radiata in ALS, which is supported by abnormal axonal staining present subjacent to the motor cortex23 and by myelin pallor that may be detected quite rostrally in the CST,18 the superior longitudinal fasciculus and the corpus callosum would become the remnant fiber tracts. This would result in greater congruence in fiber orientation, which would be reflected by an increase in FA. Although our study failed to show the FA change as statistically significant, this may in part be due to the low anisotropy of the subcortical white matter such that FA changes are less noticeable or detectable than, for example, at the cerebral peduncles. This would also suggest that at regions where multiple fiber tracts coexist, the sensitivity of FA as a surrogate marker is less compared to a region like the cerebral peduncles. The distal DTI changes are in keeping with the histological observations that support the presence of a “dying back” axonopathy,24 wherein the usual DTI findings due to chronic degeneration are decreased FA and increased  ⊥ .9,25,26 The magnitude of diffusion abnormalities was relatively small in these patients. This may reflect the clinical and pathological heterogeneity of ALS patients as a group; however, it questions the ability of DTI to function as a diagnostic tool. In contrast to DTI studies of ALS to date, we chose to encompass the entire segment of the CST from the

corona radiata to the cerebral peduncles as one ROI positioned in coronal orientation, thus allowing analysis of DTI parameters with high rostro-caudal spatial resolution using a scan with short acquisition time. This method is practically advantageous compared to positioning and analyzing multiple axial or coronal ROIs, which can be time-consuming during both acquisition and analysis. Studying the entire CST would also conceptually address the discrepancies between studies that are at least in part due to the use of different landmarks to define the extent of small anatomical regions of interest. Also different from most previous studies was our analysis of the left and right CSTs separately, as opposed to averaging them to yield single value for each individual. This would be a more sensitive method since patients can present with asymmetric findings. These factors, in addition to differences in patient characteristics, may account for discrepancies in results at the various levels of the CST studied among various groups.8,10,11,21,22 Our study has limitations in common with all ROIbased techniques, namely the potential of user-dependent errors and the inherent challenge of complete inclusion of the anatomical structure of interest with exclusion of extraneous fibers. We addressed the latter by careful angulation of the coronal acquisition and selection of the coronal image where the majority of the CST existed, as reflected by the strongest intensity (and homogeneity) on color-coded and FA maps. Use of a single coronal slice limited our study to the CST between the corona radiata and the cerebral peduncles; rostral to this it is dispersed widely and caudal to the peduncles it deflects posteriorly in the pons. ROI analysis on multiple coronal or axial images would be required to study these regions. Relatively thick (5 mm) images were acquired to minimize MR acquisition time and be inclusive of as much of the CST within a single image plane. This could have resulted in partial volume averaging of non-CST tissue. The use of tractography to segment out the CST specifically may be beneficial in this regard,27 though it too is associated with technical challenges. In summary, intracranial CST degeneration was demonstrated noninvasively using DTI. Although the small magnitude of change in ALS patients questions the ability of DTI to be a diagnostic tool, it has significant potential to shed light on pathophysiology. Further work is required to determine the reproducibility of DTI measures and to what extent biological vs. methodological issues are responsible for the variability in results between DTI studies to date. A better understanding of the in vivo pathogenesis of ALS may be possible by correlative studies with complementary imaging23 and neurophysiological8,28 modalities.

Wong et al: Spatial Profiling Using DTI in ALS

239

Funding: This study was supported by the University Hospital Foundation and the MSI Foundation of Alberta. Dr. Beaulieu is supported by a salary award from the Alberta Heritage Foundation for Medical Research and L. Concha by PROMEP. MRI infrastructure provided by the Canada Foundation for Innovation, Alberta Science and Research Authority, and the University Hospital Foundation. Fiber tracking software kindly provided by Drs. Hangyi Jiang and Susumu Mori (National Institutes of Health grant P41 RR15241–01).

14.

15.

16.

References 1. Ince PG, Evans J, Knopp M, et al. Corticospinal tract degeneration in the progressive muscular atrophy variant of ALS. Neurology 2003;60:1252-1258. 2. Hecht MJ, Fellner F, Fellner C, Hilz MJ, Heuss D, Neundorfer B. MRI-FLAIR images of the head show corticospinal tract alterations in ALS patients more frequently than T2-, T1- and proton-density-weighted images. J Neurol Sci 2001;186:37-44. 3. Pioro EP, Antel JP, Cashman NR, Arnold DL. Detection of cortical neuron loss in motor neuron disease by proton magnetic resonance spectroscopic imaging in vivo. Neurology 1994;44:1933-1938. 4. Zanette G, Tamburin S, Manganotti P, Refatti N, Forgione A, Rizzuto N. Changes in motor cortex inhibition over time in patients with amyotrophic lateral sclerosis. J Neurol 2002;249:1723-1728. 5. Kaufmann P, Pullman SL, Shungu DC, et al. Objective tests for upper motor neuron involvement in amyotrophic lateral sclerosis (ALS). Neurology 2004;62:1753-1757. 6. Moseley ME, Cohen Y, Kucharczyk J, et al. Diffusionweighted MR imaging of anisotropic water diffusion in cat central nervous system. Radiology 1990;176:439-445. 7. Beaulieu C. The basis of anisotropic water diffusion in the nervous system—a technical review. NMR Biomed 2002;15:435-455. 8. Ellis CM, Simmons A, Jones DK, et al. Diffusion tensor MRI assesses corticospinal tract damage in ALS. Neurology 1999;53:1051-1058. 9. Pierpaoli C, Barnett A, Pajevic S, et al. Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. Neuroimage 2001;13:11741185. 10. Hong YH, Lee KW, Sung JJ, Chang KH, Song IC. Diffusion tensor MRI as a diagnostic tool of upper motor neuron involvement in amyotrophic lateral sclerosis. J Neurol Sci 2004;227:73-78. 11. Toosy AT, Werring DJ, Orrell RW, et al. Diffusion tensor imaging detects corticospinal tract involvement at multiple levels in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 2003;74:1250-1257. 12. Brooks BR. El Escorial World Federation of Neurology criteria for the diagnosis of amyotrophic lateral sclerosis. J Neurol Sci 1994;124(suppl):96-107. 13. Kent-Braun JA, Walker CH, Weiner MW, Miller RG. Functional significance of upper and lower motor neuron

240

Journal of Neuroimaging Vol 17 No 3 July 2007

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

impairment in amyotrophic lateral sclerosis. Muscle Nerve 1998;21:762-768. Cosottini M, Giannelli M, Siciliano G, et al. Diffusiontensor MR imaging of corticospinal tract in amyotrophic lateral sclerosis and progressive muscular atrophy. Radiology 2005;237:258-264. Graham JM, Papadakis N, Evans J, et al. Diffusion tensor imaging for the assessment of upper motor neuron integrity in ALS. Neurology 2004;63:2111-2119. Gulani V, Webb AG, Duncan ID, Lauterbur PC. Apparent diffusion tensor measurements in myelin-deficient rat spinal cords. Magn Reson Med 2001;45:191-195. Song SK, Sun SW, Ramsbottom MJ, Chang C, Russell J, Cross AH. Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water. Neuroimage 2002;17:1429-1436. Ince PG. Neuropathology. In: Brown RHJ, Meininger V, Swash M, eds. Amyotrophic Lateral Sclerosis. London: Martin Dunitz, 2000. The ALS CNTF Treatment Study (ACTS) Phase I-II Study Group. The Amyotrophic Lateral Sclerosis Functional Rating Scale. Assessment of activities of daily living in patients with amyotrophic lateral sclerosis. The ALS CNTF treatment study (ACTS) phase I-II Study Group. Arch Neurol 1996;53:141-147. Kaufmann P, Levy G, Thompson JL, et al. The ALSFRSr predicts survival time in an ALS clinic population. Neurology 2005;64:38-43. Karlsborg M, Rosenbaum S, Wiegell M, et al. Corticospinal tract degeneration and possible pathogenesis in ALS evaluated by MR diffusion tensor imaging. Amyotroph Lateral Scler Other Motor Neuron Disord 2004;5:136-140. Yin H, Lim CC, Ma L, et al. Combined MR spectroscopic imaging and diffusion tensor MRI visualizes corticospinal tract degeneration in amyotrophic lateral sclerosis. J Neurol 2004;251:1249-1254. Smith MC. Nerve fibre degeneration in the brain in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 1960;23:269-282. Cavanagh JB. The ‘dying back’ process. A common denominator in many naturally occurring and toxic neuropathies. Arch Pathol Lab Med 1979;103:659-664. Song SK, Sun SW, Ju WK, Lin SJ, Cross AH, Neufeld AH. Diffusion tensor imaging detects and differentiates axon and myelin degeneration in mouse optic nerve after retinal ischemia. Neuroimage 2003;20:1714-1722. Concha L, Gross DW, Wheatley BM, Beaulieu C. Diffusion tensor imaging of time-dependent axonal and myelin degradation after corpus callosotomy in epilepsy patients. Neuroimage 2006;32:1090-1099. Aoki S, Iwata NK, Masutani Y, et al. Quantitative evaluation of the pyramidal tract segmented by diffusion tensor tractography: Feasibility study in patients with amyotrophic lateral sclerosis. Radiat Med 2005;23:195-199. Sach M, Winkler G, Glauche V, et al. Diffusion tensor MRI of early upper motor neuron involvement in amyotrophic lateral sclerosis. Brain 2004;127:340-350.