Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2223010311
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Guo, A. C.
Right arrow Articles by Provenzale, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Guo, A. C.
Right arrow Articles by Provenzale, J. M.
(Radiology 2002;222:729-736.)
© RSNA, 2002

Neuroradiology

Multiple Sclerosis: Diffusion Tensor MR Imaging for Evaluation of Normal-appearing White Matter1

Alexander C. Guo, MD, James R. MacFall, PhD and James M. Provenzale, MD

1 From the Department of Radiology, Duke University Medical Center, Box 3808, Rm 1533, Erwin Rd, Durham, NC 27710-3808. Received January 12, 2001; revision requested March 5; final revision received August 30; accepted September 20. Address correspondence to J.M.P. (e-mail: prove001@mc.duke.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To determine whether the normal-appearing white matter (NAWM) regions surrounding and remote from multiple sclerosis (MS) plaques have abnormal diffusional anisotropy and to compare anisotropy maps with apparent diffusion coefficient (ADC) maps for sensitivity in the detection of white matter (WM) abnormalities.

MATERIALS AND METHODS: Conventional and diffusion tensor magnetic resonance (MR) imaging examinations were performed in 26 patients with MS and in 26 age-matched control subjects. Fractional anisotropy (FA) and ADC maps were generated and coregistered with T2-weighted MR images. Uniform regions of interest were placed on plaques, periplaque white matter (PWM) regions, NAWM regions in the contralateral side of the brain, and WM regions in control subjects to obtain FA and ADC values, which were compared across the WM regions.

RESULTS: The mean FA was 0.280 for plaques, 0.383 for PWM, 0.493 for NAWM, and 0.537 for control subject WM. The mean ADC was 1.025 x 10-3 mm2/sec for plaques, 0.786 x 10-3 mm2/sec for PWM, 0.739 x 10-3 mm2/sec for NAWM, and 0.726 x 10-3 mm2/sec for control subject WM. Significant differences in anisotropy and ADC values were observed among all WM regions (P < .001 for all comparisons, except ADC in NAWM vs control subject WM [P = .018]).

CONCLUSION: The anisotropy and ADC values were abnormal in all WM regions in the patients with MS and were worse in the periplaque regions than in the distant regions. Diffusion tensor MR imaging may be more accurate than T2-weighted MR imaging for assessment of disease burden.

© RSNA, 2002

Index terms: Brain, diseases, 10.871 • Brain, MR, 10.121411, 10.121416, 10.12144 • Diffusion tensor • Magnetic resonance (MR), diffusion study, 10.121411, 10.121413, 10.121416, 10.12144 • Sclerosis, multiple, 10.871


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Magnetic resonance (MR) imaging has come to have an increasingly important role in the diagnosis and management of multiple sclerosis (MS) because it has been shown to increase diagnostic specificity and to augment the assessment of treatment effects in this clinically and pathologically heterogeneous disease (1,2). However, the findings at conventional MR imaging, which is usually performed as variations of T2-weighted MR imaging techniques such as spin echo, fast spin echo, or fluid-attenuated inversion recovery, do not correlate well with the specific pathologic or clinical course (1,3). In particular, recent evidence from histologic and MR spectroscopic studies has shown that the disease processes of MS involve not only the plaques seen on conventional MR images but also the white matter regions that appear to be normal on such studies (1,4,5). Quantification of disease burden by means of imaging is an important surrogate marker for assessment of treatment response in trials of therapeutic agents (1). Therefore, the possibility of disease extension beyond the boundaries of plaques, as seen on conventional MR images, has important ramifications in disease assessment. An imaging examination that depicts greater disease burden than that seen at conventional MR imaging would provide valuable additional information for both drug trials and patient treatment.

Previous investigators (4,6–10) have suggested that trace-weighted (ie, isotropic) diffusion MR imaging may enable differentiation of the underlying disease processes in MS and more specific distinction of clinical subgroups than conventional MR imaging. However, trace-weighted diffusion MR imaging does not provide sufficient information to determine the directionality (ie, anisotropy) of diffusion. Diffusion tensor MR imaging is a more recently developed diffusion-weighted MR imaging technique that enables assessment of not only the magnitude of diffusion but also the directionality of water diffusion in tissue.

The ability to evaluate the directionality of diffusion is important because the organization of myelinated white matter is highly directional. Myelin and the cell membrane are barriers to diffusion across myelinated white matter fibers but not to diffusion along these fibers (11). Therefore, a high degree of diffusion anisotropy is expected in normal myelinated white matter fibers and has been observed in histologic and imaging studies (11–13).

In white matter disease states in which myelination or axonal integrity is disrupted, diffusion anisotropy can be expected to decrease. Decreases in diffusion anisotropy have been shown to occur in disease processes that affect myelin or axonal integrity, such as MS, cerebral infarction, amyotrophic lateral sclerosis, progressive multifocal leukoencephalopathy, and Krabbe disease (4,6,13–16). Because diffusion tensor MR imaging is a physiologically specific imaging technique that is sensitive to myelination and axonal integrity, it is potentially more sensitive for the detection of abnormal white matter in MS than T2- and trace-weighted diffusion MR imaging.

The purposes of this study were (a) to determine whether the normal-appearing white matter (NAWM) regions surrounding and remote from MS plaques have abnormal diffusional anisotropy and (b) to compare anisotropy maps with apparent diffusion coefficient (ADC) maps for sensitivity in the detection of white matter abnormalities.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Patients
Forty-one consecutive patients clinically suspected of having MS were imaged for 6 months. The imaging data on 26 of these patients (nine male, 17 female; mean age, 39.5 years; age range, 17–55 years) were included in the study. The inclusion criteria were (a) characteristic radiologic findings of MS and (b) having undergone both diffusion tensor MR imaging and conventional MR imaging. Eleven of the 41 patients were excluded because they did not undergo diffusion tensor MR imaging. Four of the 41 patients were excluded because they did not have characteristic radiologic findings of MS. Of the 26 patients included in the study, 24 had the relapsing-remitting form of disease, and two had the secondary progressive form. Five of the 24 patients with relapsing-remitting MS had worsening symptoms at the time of imaging, which were the impetus for their imaging examinations. Three of the 24 patients underwent imaging as part of their initial examination. The remaining 16 patients with relapsing-remitting MS underwent imaging as part of their routine care and had stable symptoms. Imaging was performed in the two patients with secondary-progressive MS because of rapid progression of their symptoms.

A total of 26 age-matched control subjects (10 men, 16 women; mean age, 40 years; age range, 23–55 years) underwent both conventional MR imaging and diffusion tensor MR imaging. Five of these control subjects were healthy male volunteers (mean age, 34 years) for another study who gave their informed consent for diffusion-weighted MR imaging under the auspices of our institutional review board. These subjects also gave permission for the use of their imaging data in this study.

The remaining 21 control subjects were patients whose imaging data were chosen from those on all patients who underwent diffusion tensor MR imaging during the same 6-month period as the patients with MS and who were selected on the basis of (a) normal conventional MR examination results, (b) a low level of suspicion for cerebral white matter disease, and (c) being closely matched in age with the patients with MS. Patients with established disease affecting the central nervous system, systemic disease, or malignancy at follow-up were excluded. Clinical indications for imaging in the 21 control patients were headache (n = 11), facial numbness (n = 3), arm numbness (n = 2), memory loss (n = 1), amenorrhea (n = 1), family history of aneurysm (n = 1), optic anomaly (n = 1), and Sjögren syndrome (n = 1).

All imaging studies performed in the patients with MS, including diffusion tensor MR imaging, were a part of our routine clinical imaging protocol, so informed consents from these patients were not necessary. Permission to perform diffusion tensor MR imaging in a manner similar to conventional diffusion-weighted MR imaging and to incorporate this examination into our routine clinical protocol was obtained from our institutional review board prior to the study.

MR Imaging Data Acquisition and Analysis
MR Imaging was performed in all subjects with a 1.5-T clinical unit (Signa; GE Medical Systems, Milwaukee, Wis) by using a standard head coil. The diffusion tensor MR imaging protocol consisted of a single-shot spin-echo echo-planar sequence with 12,000/107 (repetition time msec/echo time msec), a 2,200-msec inversion time, and one signal acquired. Diffusion-sensitizing gradient encoding was applied on separate images in six directions by using a diffusion-weighted factor b of 1,000 sec/mm2, and one image was acquired without use of a diffusion gradient (b = 0 sec/mm2). The gradient directions were chosen by using the technique described by Basser (17). Thus, a total of seven diffusion-weighted images were obtained for each image section, and images through the entire brain were obtained. The section thickness was 5 mm, and the intersection gap was 2.5 mm. The field of view was 40 x 20 cm, and the matrix size was 128 x 64. The imaging time for the diffusion tensor MR sequence was approximately 2 minutes.

In addition to diffusion tensor MR sequences, conventional MR imaging sequences were performed and were tailored according to the clinical presentation; however, all imaging included a transverse T2-weighted MR sequence. The T2-weighted MR imaging parameters were as follows: 2,800/100, 22-cm2 field of view, 256 (frequency direction) x 192 (phase direction) matrix size, 5-mm section thickness, 2.5-mm intersection gap, and two signals acquired.

The raw diffusion tensor data were transferred to an independent workstation (Advantage Windows; GE Medical Systems) and processed with a computer software program (FUNCTOOL; GE Medical Systems) and proprietary software (Department of Radiology, Duke University Medical Center, Durham, NC). In the typical MR imaging unit x, y, z coordinate system, the diffusion tensor in each voxel, which is expressed as a 3 x 3 matrix, D, has only six independent diagonal elements owing to the symmetry of the conventional model of anisotropic diffusion: Dxx, Dyy, Dzz, Dxy, Dxz, and Dyz. These elements have diffusion values that are expressed in millimeters squared per second, and the average ADC is calculated by using the following equation: ADC = trace(D)/3.

For each of the six images acquired for each section with diffusion gradients applied, the diffusion sensitization effect of the imposed gradients is expressed as a 3 x 3 b matrix (bk), where k equals 1,2, ... ,6, with elements

, where, for example, for simple rectangular gradient pulses of strength Gi and Gj imposed on the i and j axes (where i and j can be one of x, y, or z), bij = {gamma}2GiGj x {delta}2[({Delta} - {delta})/3]. In practice, the b matrix element calculation takes into account all of the appropriate imaging gradients and shapes. Thus, the elements of D were calculated for each voxel on the basis of the method described by Basser (17) and are based on the equation ln{[A(bk)]/[A(b = 0)]} = {Sigma}i{Sigma}j

Dij, where i and j equal 1,2,3 summing over the elements of the b and D matrices, A(bk) is the observed voxel intensity (ie, scalar) for the images acquired with bk matrices, A(b = 0) is the voxel intensity for an image acquired when b is the zero matrix (ie, with no diffusion gradient), and Dij is the corresponding component of the D matrix.

Since there were six values of A(b) that corresponded to the six gradient directions, six equations were used to solve for the six nonzero elements to be calculated for D. Once the elements of the diffusion tensor were obtained in each voxel, the three eigenvalues of the diffusion tensor (E1, E2, and E3) were obtained by using an eigenvalue calculation of the D matrix, with the resulting eigenvalues ordered, or sorted, in decreasing magnitude. The ADC was then calculated from the eigenvalues for each voxel as follows: ADC = (E1 + E2 + E3)/3 = trace (D)/3. An ADC map for which the voxel value was the local ADC appropriately scaled for display also was constructed.

For the index of anisotropy, we chose to use fractional anisotropy (FA). FA was used because it is rotationally invariant, provides excellent gray matter–to–white matter contrast, and has a high contrast-to-noise ratio (12,15,17). It is also the most widely used index of anisotropy described in the recently published literature that can facilitate comparison with data from other investigators. FA represents the anisotropic portion of total diffusion and is calculated in each voxel from the ADC and the eigenvalues as follows: FA = (3/2)1/2{[(E1 - d)2 + (E2 - d)2 + (E3 - d)2]1/2/

1/2}, where d equals the ADC (15,17). Note that when the tissue is isotropic in diffusivity, then E1 equals E2, which equals E3, which equals ADC, and, thus, FA equals 0. When the tissue is extremely anisotropic—for example, if E2 and E3 are much smaller than E1, then FA equals 1. Thus, FA values range from 0 to 1, where 0 represents isotropic diffusion and 1 represents extremely anisotropic diffusion. The FA value is unitless because it represents a ratio of diffusion coefficients. The calculations for FA were performed for each voxel and displayed as an anisotropy map, which was appropriately scaled for display.

Uniform ovoid regions of interest (ROIs) were placed by using the FUNCTOOL software with the T2-weighted images. All ROIs used in this study were manually placed by one neuroradiologist (A.C.G.) who was blinded to the patients’ identities. The ROIs were drawn in every plaque that was large enough to accommodate a ROI that was at least 78 mm2 without apparent volume averaging of the surrounding brain tissue. The mean size of all the ROIs was 88 mm2 ± 10 (SD), which is equivalent to eight to 10 voxels. The data reported herein are the result of averaging FA or ADC values in voxels within the ROI; thus, the means of eight to 10 FA or ADC values were compared in the statistical analysis. Although the probability distributions of the ADC or FA values may not be normal, the mean values of voxels within the ROIs should be closer to normal for the purposes of statistical analysis.

Diffusion tensor MR images were acquired in the same section planes as the T2-weighted images. We coregistered the FA and ADC maps derived from the diffusion tensor MR imaging data to the T2-weighted images on a section-by-section basis by performing a combination of translation, rotation, and size matching with use of the FUNCTOOL and proprietary softwares. The ROIs were automatically transferred to the coregistered FA and ADC maps by the FUNCTOOL software. FA and ADC maps and T2-weighted images were qualitatively inspected by the neuroradiologist (A.C.G.) for adequacy of image registration and to compare the sizes of the plaques on the FA maps with the sizes of the plaques on the T2-weighted images. The sizes of the plaques on the FA maps were recorded as larger, smaller, or the same compared with the sizes on the T2-weighted images.

Uniform ROIs that were the same size as those drawn in plaques but more elongated were also drawn in periplaque white matter (PWM) regions. For purposes of ROI placement, PWM was defined as the white matter that was closest to and surrounding the plaque and that did not have abnormal signal intensity on coregistered spin-echo and echo-planar T2-weighted images. Each ROI in PWM was placed at a 90° angle from the adjacent ROI to form a box surrounding the plaque. When a PWM ROI overlapped cerebrospinal fluid or gray matter, it was discarded, and 19 ROIs were excluded on this basis. The FA and ADC values for the ROIs surrounding each plaque were averaged before recording.

ROIs were also drawn in the NAWM remote to the plaque. These ROIs were the same in size and shape as those drawn in the plaques. The NAWM region paired with each plaque was within the same structure in the contralateral hemisphere. If abnormal signal intensity was seen within the initial paired structure, a second comparable white matter structure in the contralateral hemisphere was measured. For example, if the plaque involved a portion of the genu of the corpus callosum in the left hemisphere and the portion of the genu in the right hemisphere also was involved, then a portion of the splenium in the right hemisphere was chosen as the NAWM. Placing the ROI within a NAWM region that closely matched the location of the plaque was necessary because marked intrinsic variations in anisotropy exist from one white matter structure to another (18). Such intrinsic variations in white matter anisotropy could have easily masked or exaggerated any differences in anisotropy that were due to histologic changes.

ROIs were also drawn in white matter regions in the age-matched control subjects. These ROIs also were the same size and shape as those drawn in the plaques (Fig 1). They were first placed on T2-weighted images and then transferred to coregistered FA and ADC maps. To minimize the effects of intrinsic variations in white matter anisotropy, the ROIs again were drawn in locations that closely matched the locations of the ROIs in the plaques.



View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained in a 48-year-old control subject shows placement of ROI in white matter adjacent to left frontal horn. This ROI matches the plaque location (ROI 1) in Figure 2e.

 
Before comparing FA and ADC values between the patients and control subjects, we compared the FA and ADC values in the five healthy control subjects with these values in the other control subjects (n = 21) by drawing ROIs in standardized compact and loose white matter tracts. The compact white matter tracts included the genu and splenium of the corpus callosum and the posterior limb of the internal capsule. The loose white matter tracts included the central frontal and parietal white matter and the centrum semiovale. ROIs were drawn in each structure in both hemispheres to produce six ROIs in the compact white matter regions and six ROIs in the loose white matter regions. FA and ADC values were obtained, and a two-sample Student t test was performed to assess for differences in FA and ADC values between the five healthy control subjects and the 21 other control subjects.

A total of 87 plaques, 329 PWM regions, and 87 NAWM regions were assessed in the 26 study patients. A total of 87 white matter regions were assessed in the control subjects. Comparisons of FA values were then made between (a) plaques and PWM, (b) plaques and NAWM, (c) plaques and corresponding regions in control subjects, (d) PWM and NAWM, (e) PWM and corresponding regions in control subjects, and (f) NAWM and corresponding regions in control subjects. Comparisons of ADCs between the same white matter regions also were made.

All comparisons were made by performing paired Student t tests, with the assumption of unequal variance. In all cases, a P value of less than .05 was considered to indicate a significant difference. In addition, differences in FA values and in ADCs were calculated as percentages to show the magnitude of differences between the white matter regions. The percentage difference between two white matter regions was calculated as follows: (a) Percentage reduction in FA = [(FA1 - FA2)/FA1] x 100 and (b) percentage increase in ADC = [(ADC1 - ADC2)/ADC2] x 100, where FA1 is the larger of the two FA values and ADC1 is the larger of the two ADCs. The mean FA values of each white matter region also were correlated with the mean ADCs in the region by calculating Pearson correlation coefficients.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Visual Inspection
At qualitative visual inspection, the area of abnormal anisotropy in white matter on the FA maps was larger than the size of the corresponding plaque on coregistered T2-weighted images of 63 of 87 plaques (Fig 2a–2c). In 17 of 87 plaques, the corresponding areas on the FA maps and T2-weighted images were judged to be approximately the same size, and in seven of 87 plaques, the area of abnormality was judged to be smaller on the FA maps than on the T2-weighted images.



View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. MR images obtained in a 47-year-old woman with MS depict the ROIs used in the study and enable comparison of the plaques seen on T2- and intermediate-weighted images with the corresponding areas of abnormality on anisotropy maps. (a) Transverse T2-weighted image (2,800/100, two signals acquired) shows plaque (arrow) adjacent and anterior to left frontal horn. (b) On the transverse intermediate-weighted image (2,800/30, two signals acquired) obtained at the same level as a, the plaque (arrow) is better delineated because of better contrast between the plaque and the adjacent cerebrospinal fluid. (c) Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained at the same level as a and b shows area of reduced anisotropy (arrows) in white matter adjacent to left frontal horn. This area of reduced anisotropy corresponds to the plaque seen in b; however, this abnormal area is larger than the plaque seen in b. (d) Same intermediate-weighted image as in b, but with placement of ROIs on the plaque (region 1), matched contralateral NAWM region (region 5), and three PWM regions (regions 2, 3, and 4). A fourth PWM region was discarded because it included cerebrospinal fluid in the left frontal horn. In this study, ROIs were placed on T2-weighted images before they were transferred to coregistered anisotropy and ADC maps. However, the intermediate-weighted image is shown here for illustration because of the greater conspicuity of this MS plaque on intermediate-weighted images. (e) Same anisotropy map as in c, but with placement of ROIs on the same white matter structures as those in d: ROI 1 on plaque; ROI 5 on corresponding contralateral NAWM structure; and ROIs 2, 3, and 4 on PWM regions.

 


View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. MR images obtained in a 47-year-old woman with MS depict the ROIs used in the study and enable comparison of the plaques seen on T2- and intermediate-weighted images with the corresponding areas of abnormality on anisotropy maps. (a) Transverse T2-weighted image (2,800/100, two signals acquired) shows plaque (arrow) adjacent and anterior to left frontal horn. (b) On the transverse intermediate-weighted image (2,800/30, two signals acquired) obtained at the same level as a, the plaque (arrow) is better delineated because of better contrast between the plaque and the adjacent cerebrospinal fluid. (c) Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained at the same level as a and b shows area of reduced anisotropy (arrows) in white matter adjacent to left frontal horn. This area of reduced anisotropy corresponds to the plaque seen in b; however, this abnormal area is larger than the plaque seen in b. (d) Same intermediate-weighted image as in b, but with placement of ROIs on the plaque (region 1), matched contralateral NAWM region (region 5), and three PWM regions (regions 2, 3, and 4). A fourth PWM region was discarded because it included cerebrospinal fluid in the left frontal horn. In this study, ROIs were placed on T2-weighted images before they were transferred to coregistered anisotropy and ADC maps. However, the intermediate-weighted image is shown here for illustration because of the greater conspicuity of this MS plaque on intermediate-weighted images. (e) Same anisotropy map as in c, but with placement of ROIs on the same white matter structures as those in d: ROI 1 on plaque; ROI 5 on corresponding contralateral NAWM structure; and ROIs 2, 3, and 4 on PWM regions.

 


View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c. MR images obtained in a 47-year-old woman with MS depict the ROIs used in the study and enable comparison of the plaques seen on T2- and intermediate-weighted images with the corresponding areas of abnormality on anisotropy maps. (a) Transverse T2-weighted image (2,800/100, two signals acquired) shows plaque (arrow) adjacent and anterior to left frontal horn. (b) On the transverse intermediate-weighted image (2,800/30, two signals acquired) obtained at the same level as a, the plaque (arrow) is better delineated because of better contrast between the plaque and the adjacent cerebrospinal fluid. (c) Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained at the same level as a and b shows area of reduced anisotropy (arrows) in white matter adjacent to left frontal horn. This area of reduced anisotropy corresponds to the plaque seen in b; however, this abnormal area is larger than the plaque seen in b. (d) Same intermediate-weighted image as in b, but with placement of ROIs on the plaque (region 1), matched contralateral NAWM region (region 5), and three PWM regions (regions 2, 3, and 4). A fourth PWM region was discarded because it included cerebrospinal fluid in the left frontal horn. In this study, ROIs were placed on T2-weighted images before they were transferred to coregistered anisotropy and ADC maps. However, the intermediate-weighted image is shown here for illustration because of the greater conspicuity of this MS plaque on intermediate-weighted images. (e) Same anisotropy map as in c, but with placement of ROIs on the same white matter structures as those in d: ROI 1 on plaque; ROI 5 on corresponding contralateral NAWM structure; and ROIs 2, 3, and 4 on PWM regions.

 


View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d. MR images obtained in a 47-year-old woman with MS depict the ROIs used in the study and enable comparison of the plaques seen on T2- and intermediate-weighted images with the corresponding areas of abnormality on anisotropy maps. (a) Transverse T2-weighted image (2,800/100, two signals acquired) shows plaque (arrow) adjacent and anterior to left frontal horn. (b) On the transverse intermediate-weighted image (2,800/30, two signals acquired) obtained at the same level as a, the plaque (arrow) is better delineated because of better contrast between the plaque and the adjacent cerebrospinal fluid. (c) Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained at the same level as a and b shows area of reduced anisotropy (arrows) in white matter adjacent to left frontal horn. This area of reduced anisotropy corresponds to the plaque seen in b; however, this abnormal area is larger than the plaque seen in b. (d) Same intermediate-weighted image as in b, but with placement of ROIs on the plaque (region 1), matched contralateral NAWM region (region 5), and three PWM regions (regions 2, 3, and 4). A fourth PWM region was discarded because it included cerebrospinal fluid in the left frontal horn. In this study, ROIs were placed on T2-weighted images before they were transferred to coregistered anisotropy and ADC maps. However, the intermediate-weighted image is shown here for illustration because of the greater conspicuity of this MS plaque on intermediate-weighted images. (e) Same anisotropy map as in c, but with placement of ROIs on the same white matter structures as those in d: ROI 1 on plaque; ROI 5 on corresponding contralateral NAWM structure; and ROIs 2, 3, and 4 on PWM regions.

 


View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2e. MR images obtained in a 47-year-old woman with MS depict the ROIs used in the study and enable comparison of the plaques seen on T2- and intermediate-weighted images with the corresponding areas of abnormality on anisotropy maps. (a) Transverse T2-weighted image (2,800/100, two signals acquired) shows plaque (arrow) adjacent and anterior to left frontal horn. (b) On the transverse intermediate-weighted image (2,800/30, two signals acquired) obtained at the same level as a, the plaque (arrow) is better delineated because of better contrast between the plaque and the adjacent cerebrospinal fluid. (c) Transverse echo-planar spin-echo diffusion tensor MR imaging-derived anisotropy map (12,000/107, 2,200-msec inversion time, one signal acquired, b = 1,000 sec/mm2 x six directions and b = 0) obtained at the same level as a and b shows area of reduced anisotropy (arrows) in white matter adjacent to left frontal horn. This area of reduced anisotropy corresponds to the plaque seen in b; however, this abnormal area is larger than the plaque seen in b. (d) Same intermediate-weighted image as in b, but with placement of ROIs on the plaque (region 1), matched contralateral NAWM region (region 5), and three PWM regions (regions 2, 3, and 4). A fourth PWM region was discarded because it included cerebrospinal fluid in the left frontal horn. In this study, ROIs were placed on T2-weighted images before they were transferred to coregistered anisotropy and ADC maps. However, the intermediate-weighted image is shown here for illustration because of the greater conspicuity of this MS plaque on intermediate-weighted images. (e) Same anisotropy map as in c, but with placement of ROIs on the same white matter structures as those in d: ROI 1 on plaque; ROI 5 on corresponding contralateral NAWM structure; and ROIs 2, 3, and 4 on PWM regions.

 
Comparison of FA in Various White Matter Regions
The lowest FA values were measured in plaques (mean FA, 0.280 ± 0.065), followed by, in ascending order, PWM (mean FA, 0.383 ± 0.048), NAWM (mean FA, 0.493 ± 0.056), and white matter regions in the control subjects (mean FA, 0.537 ± 0.064). Examples of ROIs that were placed to measure FA in plaques, PWM, NAWM, and white matter in control subjects are shown (Fig 2d, 2e). Significant differences in FA were noted between the plaques and the white matter in control subjects (P < .001), between the PWM and the white matter in control subjects (P < .001), and between the NAWM and the white matter in control subjects (P < .001). Significant differences were also found between plaques and PWM (P < .001), between plaques and NAWM (P < .001), and between PWM and NAWM (P < .001). There was no significant difference between the FA of the white matter in the healthy control subjects and the FA of the white matter in the other control subjects (P = .07 for compact white matter, P = .25 for loose white matter).

Comparison of ADC in Various White Matter Regions
The highest ADCs were measured in plaques (mean ADC, 1.025 x 10-3 mm2/sec ± 0.207), followed by, in descending order, PWM (mean ADC, 0.786 x 10-3 mm2/sec ± 0.046), NAWM (mean ADC, 0.739 x 10-3 mm2/sec ± 0.040), and white matter regions in control subjects (mean ADC, 0.726 x 10-3 mm2/sec ± 0.043). Significant differences in ADCs were noted between plaques and the white matter regions in control subjects (P < .001) and between PWM and the white matter regions in control subjects (P < .001). The differences in ADC between the NAWM and the white matter regions in control subjects were significant (P = .018) but not as significant as the differences in ADC between the other white matter regions. Significant differences in ADC were also noted between plaques and PWM (P < .001), between plaques and NAWM (P < .001), and between PWM and NAWM (P < .001). There was no significant difference in the ADC between the white matter regions in the healthy control subjects and those in the other control subjects (P = .37 for compact white matter, P = .84 for loose white matter).

Anisotropy Differences Compared with ADC Differences
Although significant differences in both FA and ADC values were found across all comparisons, the differences in FA were generally greater than the differences in ADC in terms of percentages. When plaques were compared with the white matter in control subjects, a 57% reduction in FA and a 38% reduction in ADC were observed. When PWM was compared with the white matter in control subjects, a 31% reduction in FA and an 8% increase in ADC were observed. When NAWM was compared with the white matter in control subjects, an 8% reduction in FA and a 2% increase in ADC were observed. When plaques were compared with PWM, a 27% reduction in FA and a 30% increase in ADC were observed. When plaques were compared with NAWM, a 49% reduction in FA and a 37% increase in ADC were observed. When PWM was compared with NAWM, a 22% reduction in FA and a 6% increase in ADC were observed. Increases in ADC correlated well with decreases in FA (Pearson r = -0.906).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
MS generally has been thought of as a demyelinating disease only. However, more recent histologic studies (5,19–22) have revealed increasing evidence that axonal transection or disruption also has an important role in MS, especially in terms of correlation with long-term disability. Multiple histologic analyses and MR spectroscopic studies focusing on axonal damage as an important marker for MS disease processes have provided evidence that NAWM is abnormal in MS. Several investigator groups (19–21,23,24) have observed significant decreases in the N-acetylaspartate level and N-acetylaspartate-to-creatine ratio, which are markers for neuronal or axonal damage, in the NAWM of patients with MS. The investigators in one study (24) specifically noted a decreased N-acetylaspartate-to-creatine ratio in the NAWM adjacent to plaques (the PWM in our study), which occurred to a greater extent than it did in the white matter regions remote from plaques (the NAWM in our study). Axonal loss and axonal transection in NAWM also have been noted in histologic studies (25–27). One recent report noted axonal damage in the periplaque regions surrounding acutely demyelinating plaques (26). Other investigators (3,5,24) have found both myelin breakdown products and transected axons in the periphery of acute and reactivated chronic plaques.

Evaluation of Normal-appearing PWM
The anisotropy measurements in our study correlated well with evidence from histologic and MR spectroscopic studies that the disease process in MS extends beyond the borders of plaques that are visible on T2-weighted images, as outlined earlier. Anisotropy in the PWM was significantly lower compared with that seen in the NAWM and with the white matter regions in the control subjects. In addition, anisotropy in the PWM was significantly higher than that in the plaques, and this suggests that damage to the white matter at the periphery of plaques is not as severe as that within the plaque.

The apparent extension of the disease process in MS beyond the borders of plaques that are visible on T2-weighted images was also suggested by the findings at visual inspection of the FA maps alone, because the area of abnormal anisotropy was usually larger than the plaque size on the corresponding T2-weighted images. One previous group of investigators (28) also noted that the area of abnormal anisotropy often is larger than the area of plaque on the T2-weighted MR image. However, these investigators did not verify this observation in a quantitative manner, as was done in this study.

The reasons for the apparent extension of the disease process in MS beyond the borders of visible plaques are not yet clear. It has been hypothesized, however, that the decreased N-acetylaspartate level and N-acetylaspartate-to-creatine ratio seen in white matter regions outside of plaques at MR spectroscopy may be due to wallerian degeneration (23). The results of a recent MR imaging study (29) showed the development of lesions that were characteristic of wallerian degeneration along the cortical spinal tracts in patients with MS. Recent histologic study (30) results have suggested that both wallerian degeneration and retrograde degeneration of the cell body occur in MS. An alternative explanation for the presence of axonal injury and demyelination adjacent to plaques is suggested by their natural history. During their evolution, MS plaques enlarge and regress in a concentric manner around a perivenular focus (3). This pattern of expansion and regression occurs in both acute and reactivated chronic plaques (3). As plaques regress, they may leave behind a surrounding area of damaged white matter that appears to have normal signal intensity at T2-weighted MR imaging. Myelin breakdown products and transected axons have been found at the periphery of both active and reactivated chronic plaques at histologic analyses (3,5,24,26).

Evaluation of NAWM
The anisotropy values in the NAWM were significantly decreased compared with those in the corresponding white matter regions in the control group. This finding indicates that diffusion tensor MR imaging is more sensitive than conventional MR imaging for assessment of white matter integrity in MS. This finding also indicates that even those regions that are not surrounding plaques have abnormal white matter pathways in MS, which is in agreement with the histologic and MR spectroscopic study findings discussed earlier (19–27). Evidence of a decreased magnetization transfer ratio in NAWM that is suggestive of demyelination has also been reported in magnetization transfer imaging studies (31,32).

The progressive decrease in anisotropy observed from the NAWM to the PWM to the plaque suggests a gradient of progressively more extensive white matter injury. Although we did not directly measure the white matter region immediately adjacent to the PWM regions, it is reasonable to assume that the anisotropy in this region is similar to that seen in the distant NAWM. The apparent gradient of progressively more extensive white matter injury is supported by other investigators’ (32) observations that a progressive decrease in the magnetization transfer ratio, which correlates with demyelinating white matter in MS, is seen as one traverses a line extending from the remote NAWM toward the plaque.

Evaluation of ADC in Various White Matter Regions
The ADCs measured in the PWM were significantly higher than those measured in the white matter regions of the control subjects. ADCs were highest in the plaques, followed by the PWM, NAWM, and white matter regions in the control subjects. A strong inverse correlation was seen between the mean FA values and the mean ADCs of the four white matter regions studied (ie, plaques, PWM, NAWM, white matter in control subjects; Pearson r = -0.906). The close inverse correlation between decreased anisotropy and increased ADC suggests that the disease processes in MS result in an overall increase in water diffusivity, as well as a decrease in diffusion anisotropy, probably as a result of a breakdown of diffusion barriers.

Previous investigators (7–10) also have noted elevated ADCs in both MS plaque and NAWM, with higher ADCs observed in the plaques than in the NAWM. However, these investigators did not assess the PWM. The investigators in these studies observed the ADCs in plaques to be in the range of 0.92–1.59 x 10-3 mm2/sec and the ADCs in NAWM to be in the range of 0.69–0.79 x 10-3 mm2/sec (7–10). Our mean ADC measurements of 1.025 x 10-3 mm2/sec for plaques and 0.739 x 10-3 mm2/sec for NAWM are within these ranges.

Although we noted a strong correlation between decreased anisotropy and increased ADC, the measured differences in FA values were greater than the measured differences in ADCs in terms of percentages. This finding was consistent across all comparisons between plaques, PWM, NAWM, and white matter in control subjects, with the exception of the comparison between plaque and PWM, in which the percentage difference in FA and percentage difference in ADC were approximately equal. The generally greater differences measured in FA relative to the differences measured in ADC may indicate a greater sensitivity of diffusion anisotropy measurements compared with isotropic diffusivity measurements for the detection of diseased white matter in MS. This greater sensitivity may exist because anisotropy measurements have greater specificity for abnormality detection. However, this possibility would have to be verified by correlation with histologic data, which was beyond the scope of this study.

Comparisons with Previous Anisotropy Measurement Studies
To our knowledge, only a few previously published studies (4,6,28) have involved the assessment of anisotropy in MS with diffusion tensor MR imaging. The imaging techniques used in our study and in the other studies were somewhat similar, and FA was the anisotropy index used in all studies. In general, the measurements in the four studies correlated fairly well with each other. The mean FA in NAWM in the patients with MS was 0.493, which is within the range of 0.40 observed by Tievsky et al (6), of 0.44 observed by Bammer et al (28), and of 0.56 observed by Werring et al (4). The mean FA measured in plaques in the patients with MS was 0.280, which is similar to the mean FA values in subacute and chronic plaques reported by Tievsky et al (6), 0.278 and 0.289, respectively, and within the range of 0.23–0.43 for FA values in nonacute (ie, non–homogeneously enhancing) plaques reported by Bammer et al (28). The mean FA in plaques in our study, however, was lower than the mean FA of 0.50 reported by Werring et al (4) in a small series.

One potential limitation in our study was the poor identification and measurement of small plaques due to the relatively poor spatial resolution (approximately 3.1 x 3.1 x 5.0-mm voxel size) and low signal-to-noise ratio of diffusion tensor images. The large difference in spatial resolution between diffusion tensor MR images and T2-weighted MR images limited the precise coregistration of these images, and the limited coregistration limited the accuracy of ROI placement and visual assessment. To minimize the effect of differences in spatial resolution, only plaques that were large enough to be clearly visualized on the anisotropy maps and to fully contain a 78-mm2 ROI (ie, eight voxels) were included in the study. Excluding the plaques smaller than 78 mm2 may have introduced a selection bias, because these plaques may be biologically different from larger plaques. In future studies, improvements in sequences and hardware may facilitate improved spatial resolution and signal-to-noise ratios to a degree that enables measurements in smaller plaques and thus improves the coregistration of images.

Another potential limitation of our study was the use of a combination of patients and healthy volunteers as the control group. The inclusion of patients who did not have MS was deemed justifiable because these patients’ MR imaging findings were judged to be normal during both routine clinical review and assessment for the purposes of this study. These patients also had indications for imaging that were considered to have very low pretest probability for central nervous system abnormality in general and for demyelinating disease in particular. In addition, as a measure of internal control, the mean white matter anisotropy and ADC values for the control patients were compared with those for the healthy control subjects, and no significant differences were found.

As mentioned earlier, the statistical distribution of ADC and FA measurements was not necessarily normal owing to nonlinearities in the calculation of D, even though the MR imaging noise in the data approximated a normal distribution. Unpublished simulations of the distributions of the ROI data indicate that at the signal-to-noise ratios typical for these data (average of approximately 30 for the T2-weighted images with b value of 0), the distributions of FA and ADC are symmetric and close to normal and develop noticeable departures from this pseudonormality only at a three-times-worse signal-to-noise ratio for FA and at a 10-times-worse signal-to-noise ratio for ADC. Thus, at the signal-to-noise ratios observed in this study, we believe that the results interpreted with the assumption of normality are valid, although the exact P values could be slightly different from the ones reported.

The sensitivity and specificity of diffusion tensor MR imaging, as compared with those of other advanced MR imaging techniques, are not yet known. Therefore, in patients with MS, it will be worthwhile to directly compare the diffusion tensor MR imaging findings with the results of examinations such as magnetization transfer imaging and MR spectroscopy to determine whether more specific characterization of plaques can be achieved by using the combined data from multiple techniques.

In conclusion, we found the diffusion anisotropy and ADC values to be abnormal in all white matter regions assessed in our study population of patients with MS. The degree of anisotropy decrease in PWM was less than that in plaques but more than the anisotropy decrease in NAWM; these findings suggest that there is a gradient of white matter abnormality extending in an outward direction from the plaque. Our study results support the hypothesis that the extent of white matter disease in MS may differ substantially from that seen at conventional T2-weighted MR imaging. Because T2-weighted images are routinely used in drug trials to assess disease burden and treatment response, our study findings suggest that diffusion tensor MR imaging may be more sensitive for monitoring the effects of therapy. Our study findings also indicate that anisotropy measurements are potentially more sensitive compared with ADC measurements for the detection of white matter abnormality in MS.


   FOOTNOTES
 
Abbreviations: ADC = apparent diffusion coefficient, FA = fractional anisotropy, NAWM = normal-appearing white matter, PWM = periplaque white matter, ROI = region of interest

Author contributions: Guarantors of integrity of entire study, A.C.G., J.M.P.; study concepts and design, A.C.G., J.M.P.; literature research, A.C.G.; clinical studies, A.C.G.; data acquisition, A.C.G.; data analysis/interpretation, A.C.G., J.M.P.; statistical analysis, J.R.M., A.C.G.; manuscript preparation, A.C.G., J.M.P.; manuscript definition of intellectual content, editing, and revision/review, J.R.M., A.C.G., J.M.P.; manuscript final version approval, A.C.G., J.M.P.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Miller DH, Grossman RI, Reingold SC, McFarland HF. The role of magnetic resonance techniques in understanding and managing multiple sclerosis. Brain 1998; 121:3-24.[Abstract/Free Full Text]
  2. Lucchinetti CF, Bruck W, Rodriguez M, Lassmann H. Distinct patterns of multiple sclerosis pathology indicates heterogeneity in pathogenesis. Brain Pathol 1996; 6:259-274.[Medline]
  3. McDonald WI, Miller DH, Barnes D. The pathological evolution of multiple sclerosis. Neuropathol Appl Neurobiol 1992; 18:319-334.[Medline]
  4. Werring DJ, Clark CA, Barker GJ, Thompson AJ, Miller DH. Diffusion tensor imaging of lesions and normal-appearing white matter in multiple sclerosis. Neurology 1999; 52:1626-1632.[Abstract/Free Full Text]
  5. Waxman SG. Demyelinating diseases: new pathological insights, new therapeutic targets (editorial). N Engl J Med 1998; 338:323-325.[Free Full Text]
  6. Tievsky AL, Ptak T, Farkas J. Investigation of apparent diffusion coefficient and diffusion tensor anisotropy in acute and chronic multiple sclerosis lesions. AJNR Am J Neuroradiol 1999; 20:1491-1499.[Abstract/Free Full Text]
  7. Horsfield MA, Lai M, Webb SL, et al. Apparent diffusion coefficients in benign and secondary progressive multiple sclerosis by nuclear magnetic resonance. Magn Reson Med 1996; 36:393-400.[Medline]
  8. Castriota-Scanderbeg A, Tomaiuolo F, Sabatini U, Nocentini U, Grasso MG, Caltagirone C. Demyelinating plaques in relapsing-remitting and secondary progressive multiple sclerosis: assessment with diffusion MR imaging. AJNR Am J Neuroradiol 2000; 21:862-868.[Abstract/Free Full Text]
  9. Tsuchiya K, Hachiya J, Maehara T. Diffusion-weighted MR imaging in multiple sclerosis: comparison with contrast-enhanced study. Eur J Radiol 1999; 31:165-169.[CrossRef][Medline]
  10. Christiansen P, Gideon P, Thomsen C, Stubgaard M, Henrikson O, Larsson HB. Increased water self-diffusion in chronic plaques and in apparently normal white matter in patients with multiple sclerosis. Acta Neurol Scand 1993; 87:195-199.[Medline]
  11. Beaulieu C, Allen PS. Determinants of anisotropic water diffusion in nerves. Magn Reson Med 1994; 31:394-400.[Medline]
  12. Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Di Chiro G. Diffusion tensor MR imaging of the human brain. Radiology 1996; 201:637-648.[Abstract/Free Full Text]
  13. Hajnal JV, Doran M, Hall AS, et al. MR imaging of anisotropically restricted diffusion of water in the central nervous system: technical, anatomic, and pathologic considerations. J Comput Assist Tomogr 1991; 15:1-18.[Medline]
  14. Ulug AM, Moore DF, Bojko AS, Zimmerman RD. Clinical use of diffusion-tensor imaging for diseases causing neuronal and axonal damage. AJNR Am J Neuroradiol 1999; 20:1044-1048.[Abstract/Free Full Text]
  15. Sorensen AG, Wu O, Copen WA, et al. Human acute cerebral ischemia: detection of changes in water diffusion anisotropy by using MR imaging. Radiology 1999; 212:785-792.[Abstract/Free Full Text]
  16. Guo AC, Petrella JR, Kurtzberg J, Provenzale JM. Evaluation of white matter anisotropy in Krabbe disease using diffusion tensor MR imaging: initial experience. Radiology 2001; 218:809-815.[Abstract/Free Full Text]
  17. Basser PJ. Inferring microstructural features and the physiological state of tissues from diffusion-weighted images. NMR Biomed 1995; 8:333-344.[Medline]
  18. Shimony JS, McKinstry RC, Akbudak E, et al. Quantitative diffusion-tensor anisotropy brain MR imaging: normative human data and anatomic analysis. Radiology 1999; 212:770-784.[Abstract/Free Full Text]
  19. Fu L, Matthews PM, De Stefano N, et al. Imaging axonal damage of normal appearing white matter in multiple sclerosis. Brain 1998; 121:103-113.[Abstract/Free Full Text]
  20. Davie CA, Barker GJ, Thompson AJ, Tofts PS, McDonald WI, Miller DH. 1H magnetic resonance spectroscopy of chronic white matter lesions and normal appearing white matter in multiple sclerosis. J Neurol Neurosurg Psychiatry 1997; 63:736-742.[Abstract/Free Full Text]
  21. Leary SM, Davie CA, Parker GJ, et al. 1H magnetic resonance spectroscopy of normal appearing white matter in primary progressive multiple sclerosis. J Neurol 1999; 246:1023-1026.[CrossRef][Medline]
  22. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis. N Engl J Med 1998; 338:278-285.[Abstract/Free Full Text]
  23. Narayanan S, Fu L, Pioro E, et al. Imaging of axonal damage in multiple sclerosis: spatial distribution of magnetic resonance imaging lesions. Ann Neurol 1997; 41:385-391.[CrossRef][Medline]
  24. Arnold DL, Matthews PM, Francis GS, O’Connor J, Antel JP. Proton magnetic resonance spectroscopic imaging for metabolic characterization of demyelinating plaques. Ann Neurol 1992; 31:235-241.[CrossRef][Medline]
  25. Allen IV, McKeown SR. A histological, histochemical and biochemical study of the macroscopically normal white matter in multiple sclerosis. J Neurol Sci 1979; 41:81-91.[CrossRef][Medline]
  26. Kornek B, Storch MK, Weissert R, et al. Multiple sclerosis and chronic autoimmune encephalomyelitis: a comparative quantitative study of axonal injury in active, inactive, and remyelinated lesions. Am J Pathol 2000; 157:267-276.[Abstract/Free Full Text]
  27. Evangelou N, Esiri MM, Smith S, Palace J, Matthews PM. Quantitative pathological evidence for axonal loss in normal appearing white matter in multiple sclerosis. Ann Neurol 2000; 47:391-395.[CrossRef][Medline]
  28. Bammer R, Augustin M, Strasser-Fuchs S, et al. Magnetic resonance diffusion tensor imaging for characterizing diffuse and focal white matter abnormalities in multiple sclerosis. Magn Reson Med 2000; 44:583-591.[CrossRef][Medline]
  29. Simon JH, Kinkel RP, Jacobs L, Bub L, Simonian N. A Wallerian degeneration pattern in patients at risk for MS. Neurology 2000; 54:1155-1160.[Abstract/Free Full Text]
  30. Perry VH, Anthony DC. Axon damage and repair in multiple sclerosis. Philos Trans R Soc Lond B Biol Sci 1999; 354:1641-1647.[CrossRef][Medline]
  31. Filippi M, Campi A, Dousset V, et al. A magnetization transfer imaging study of normal-appearing white matter in multiple sclerosis. Neurology 1995; 45:478-482.[Abstract/Free Full Text]
  32. Bagley LJ, Grossman RI, Galetta SL, Sinson GP, Kotapka M, McGowan JC. Characterization of white matter lesions in multiple sclerosis and traumatic brain injury as revealed by magnetization transfer contour plots. AJNR Am J Neuroradiol 1999; 20:977-981.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 BrainHome page
R. A. Dineen, J. Vilisaar, J. Hlinka, C. M. Bradshaw, P. S. Morgan, C. S. Constantinescu, and D. P. Auer
Disconnection as a mechanism for cognitive dysfunction in multiple sclerosis
Brain, October 25, 2008; (2008) awn275v1.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
T. Welzel, A. Niethammer, U. Mende, S. Heiland, F. Wenz, J. Debus, and R. Krempien
Diffusion Tensor Imaging Screening of Radiation-Induced Changes in the White Matter after Prophylactic Cranial Irradiation of Patients with Small Cell Lung Cancer: First Results of a Prospective Study
AJNR Am. J. Neuroradiol., February 1, 2008; 29(2): 379 - 383.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurol. Neurosurg. PsychiatryHome page
M. Ito, H. Watanabe, Y. Kawai, N. Atsuta, F. Tanaka, S. Naganawa, H. Fukatsu, and G. Sobue
Usefulness of combined fractional anisotropy and apparent diffusion coefficient values for detection of involvement in multiple system atrophy
J. Neurol. Neurosurg. Psychiatry, July 1, 2007; 78(7): 722 - 728.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
H. Oouchi, K. Yamada, K. Sakai, O. Kizu, T. Kubota, H. Ito, and T. Nishimura
Diffusion Anisotropy Measurement of Brain White Matter Is Affected by Voxel Size: Underestimation Occurs in Areas with Crossing Fibers
AJNR Am. J. Neuroradiol., June 1, 2007; 28(6): 1102 - 1106.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
A.M. Saindane, M. Law, Y. Ge, G. Johnson, J.S. Babb, and R.I. Grossman
Correlation of Diffusion Tensor and Dynamic Perfusion MR Imaging Metrics in Normal-Appearing Corpus Callosum: Support for Primary Hypoperfusion in Multiple Sclerosis
AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 767 - 772.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
F.G. Garaci, V. Colangelo, A. Ludovici, F. Gaudiello, S. Marziali, D. Centonze, L. Boffa, G. Simonetti, and R. Floris
A Diffusion Longitudinal MR Imaging Study in Normal-Appearing White Matter in Untreated Relapsing-Remitting Multiple Sclerosis
AJNR Am. J. Neuroradiol., March 1, 2007; 28(3): 475 - 478.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
H. Ni, V. Kavcic, T. Zhu, S. Ekholm, and J. Zhong
Effects of number of diffusion gradient directions on derived diffusion tensor imaging indices in human brain.
AJNR Am. J. Neuroradiol., September 1, 2006; 27(8): 1776 - 1781.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
Y. Ge
Multiple Sclerosis: The Role of MR Imaging
AJNR Am. J. Neuroradiol., June 1, 2006; 27(6): 1165 - 1176.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
A Pena, H A L Green, T A Carpenter, S J Price, J D Pickard, and J H Gillard
Enhanced visualization and quantification of magnetic resonance diffusion tensor imaging using the p:q tensor decomposition
Br. J. Radiol., February 1, 2006; 79(938): 101 - 109.
[Abstract] [Full Text] [PDF]

Home page
 NeurologyHome page
M. Rovaris, A. Gass, R. Bammer, S. J. Hickman, O. Ciccarelli, D. H. Miller, and M. Filippi
Diffusion MRI in multiple sclerosis
Neurology, November 22, 2005; 65(10): 1526 - 1532.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Neuroradiol.Home page
S. Kitahara, S. Nakasu, K. Murata, K. Sho, and R. Ito
Evaluation of Treatment-Induced Cerebral White Matter Injury by Using Diffusion-Tensor MR Imaging: Initial Experience
AJNR Am. J. Neuroradiol., October 1, 2005; 26(9): 2200 - 2206.
[Abstract] [Full Text] [PDF]