Correlating magnetic resonance imaging markers of axonal injury and demyelination in motor impairment secondary to stroke and multiple sclerosis

Correlating magnetic resonance imaging markers of axonal injury and demyelination in motor impairment secondary to stroke and multiple sclerosis

Magnetic Resonance Imaging 18 (2000) 369 –378 Correlating magnetic resonance imaging markers of axonal injury and demyelination in motor impairment s...

156KB Sizes 0 Downloads 17 Views

Magnetic Resonance Imaging 18 (2000) 369 –378

Correlating magnetic resonance imaging markers of axonal injury and demyelination in motor impairment secondary to stroke and multiple sclerosis Sarah T. Pendleburya,b,*, Martin A. Leeb, Andrew M. Blamireb, Peter Stylesb, Paul M. Matthewsa a

Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB), John Radcliffe Hospital, Oxford, OX3 9DU, UK MRC Biochemical and Clinical Magnetic Resonance Spectroscopy Unit, Department of Biochemistry, South Parks Road, Oxford, OX1 3QU and the John Radcliffe Hospital, Oxford, OX3 9DU, UK

b

Received 10 November 1999; accepted 28 December 1999

Abstract The primary pathological mechanisms in stroke and multiple sclerosis (MS) are very different but in both diseases, impairment may arise from a final common pathway of axonal damage. We aimed to examine the relationship between motor impairment, magnetisation transfer ratio (MTR) (an index of demyelination), and N-acetyl aspartate (NAA) loss (an index of axonal injury) localised to the descending motor pathways in stroke and MS. Twelve patients between 1 and 10 months after first ischaemic stroke causing a motor deficit and 12 patients with stable MS with asymmetric motor deficit were examined. T2-weighted imaging of the brain together with MTR and proton (voxel 1.5 ⫻ 2 ⫻ 2 cm3) MRS localised to the posterior limb of the internal capsule were performed and correlated to a composite motor deficit score. MTR and NAA in the internal capsule were reduced in both stroke and MS patients compared to controls. NAA loss correlated with motor deficit score in both stroke and MS ( p ⬍ 0.001 and p ⫽ 0.04, respectively). Correlations were seen between MTR and motor deficit ( p ⬍ 0.001) MTR and NAA loss ( p ⬍ 0.001) in stroke patients but not in MS patients. Axonal injury in the descending motor tracts would appear to be an important determinant of motor impairment in both stroke and MS. In stroke, MTR measures of demyelination are closely related to axonal damage and thus also correlate with motor deficit. However in MS, MTR measures of demyelination do not correlate with NAA loss or motor deficit suggesting that demyelination and gliosis may occur independently of axonal damage and are less closely linked with functional impairment. © 2000 Elsevier Science Inc. All rights reserved. Keywords: Stroke; Multiple sclerosis; axonal injury; demyelination

1. Introduction Stroke and multiple sclerosis (MS) are common conditions leading to long-term impairment and disability. Although much is known about the pathology of both conditions from histological studies, assessment of the nature and extent of central nervous system damage in vivo remains difficult. Conventional magnetic resonance imaging (MRI) has enabled accurate visualisation of lesions in both stroke and MS but gives limited information about pathology. In MS, lesions are very heterogeneous comprising varying

* Corresponding author. Tel.: ⫹18-65-222738; fax: ⫹18-65-222717. E-mail address: [email protected] (S.T. Pendlebury).

amounts of oedema, demyelination, axonal loss, and gliosis. Conventional imaging is relatively poor at distinguishing between these different pathological states [1]. In stroke, lesion volume correlates poorly with clinical outcome [2], owing in part to the importance of lesion location but also because of the inability of conventional imaging to quantify the amount of neuronal or axonal damage within an area of imaging abnormality. Recently, it has become clear that other magnetic resonance techniques may provide further information about pathological changes in the brain. Magnetic resonance spectroscopy enables in vivo measurement of N-acetyl aspartate (NAA), a compound believed to be present almost exclusively in neurons [3]. Reduced levels of this compound have been shown within the infarct after stroke [4,5], and within

0730-725X/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S 0 7 3 0 - 7 2 5 X ( 0 0 ) 0 0 1 1 5 - 6

370

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

the acute and chronic plaques of MS [6,7]. NAA has also been shown to be decreased within the normal appearing white matter (NAWM) in MS [8,9], suggesting damage to neurons remote from areas of focal inflammation. Changes in the levels of NAA within the NAWM have been shown to correlate with progression of disability [10,11], leading to the “Axonal Hypothesis” that axonal loss and damage are responsible for persistent functional deficit in MS. In stroke, axons and neurons are damaged by ischaemia leading to anterograde and retrograde changes that cause secondary damage in remote brain regions. Thus, axonal damage can occur throughout the motor pathways following a motor stroke. Recently, we have shown that there is a relationship between axonal injury (as shown by NAA loss) and impairment within the motor system in both stroke [12] and MS [13]. The same cohort of patients underwent further MR studies, data from which are reported here. Magnetisation transfer (MT) imaging is a technique that has been used to evaluate macromolecular integrity in diseases of the central nervous system. In conventional MRI, the hydrogen nuclei of mobile water (“free protons”) are the main source of signal; hydrogen nuclei contained within relatively immobile macromolecules such as membrane lipids (“bound protons”) do not give rise to measurable signal. However, direct interactions between the free and bound protons results in an exchange of energy between the two populations. In MT imaging, the bound protons are excited (saturated) by an off-resonance pulse and this magnetisation energy is transferred to the free protons. The difference in image signal obtained with and without pre-saturation, the magnetisation transfer ratio (MTR), is determined by the quantity of non-mobile proton containing macromolecules. The exact significance of a reduction in magnetisation transfer ratio is not known but in white matter, the MTR is thought to be highly dependent on the amount of myelin present [14 –18]. Decreases in the magnetisation ratio have been observed in both stroke and multiple sclerosis (MS) consistent with loss of myelin and cell membranes. In stroke, the MTR is initially normal within the lesion for about 1 week and then decreases over the next 2 months [19]. This is consistent with cell membrane and myelin destruction secondary to cell death within the lesion. In MS, both acute and chronic inflammatory lesions show a reduction in MTR consistent with demyelination [14,20]. Smaller MTR decreases have also been observed from NAWM in MS [21,22] and it has been proposed that this may occur in part as a result of Wallerian degeneration [23]. In the current study, we wished to compare and contrast the relationship between motor impairment and magnetic resonance measures of axonal injury and demyelination, localised to the descending motor pathways, in patients with ischaemic stroke and MS. Our hypothesis was that myelin loss would occur purely as a result of axonal injury in stroke and hence that MTR decrease and axonal injury would be closely related. However, in MS, in which there is gliosis

and demyelination in the absence of axonal damage, we hypothesized that a relationship between axonal injury and MTR would be less likely. Thus, given the known correlation between axonal damage and motor impairment, we proposed that there would be a correlation between MTR and motor deficit in stroke but not in MS.

2. Subjects and methods 2.1. Patients and controls Twelve patients (7 males, 5 females) who had suffered a stroke, as defined by the WHO criteria [24] up to 10 months prior to entry into the study and resulting in a motor deficit, were recruited from general practitioners and from the local stroke unit. Patients with hemorrhagic or brain stem stroke, history of prior stroke, other co-existent neurological disease or cognitive impairment were excluded from the study. Twelve MS patients (6 males, 6 females) with persistent asymmetric motor impairment (near normal function on one side of the body) and no history of relapse within the last three months were recruited from neurology outpatient clinics. Patients with significant cerebellar disability and joint position sense loss were excluded in order to maximize the sensitivity of the motor impairment tests to descending motor tract damage. Eight age-matched healthy controls were used to define control values for the MS group. The study was approved by the local ethics committee and informed consent was obtained prior to the study. 2.2. Magnetic resonance imaging and spectroscopy Magnetic resonance imaging and spectroscopy was performed using a 2T whole body magnet interfaced with a Bruker Avance spectrometer (Bruker Medical, Ettlingen, Germany). Care was taken to standardize head positioning across subjects by placing the subject’s head in a foam head localiser with the orbito-metal line positioned perpendicularly to the long axis of the magnet. A forehead strap and side padding were used to immobilize the head. All images and spectra were obtained using a quadrature birdcage coil tuned to 85.2 MHz. A sagittal scout image was performed to confirm correct subject head alignment followed by axial fast spin echo T2-weighted imaging (30 contiguous slices, TR ⫽ 3100 ms, TE ⫽ 82 ms, slice thickness ⫽ 5 mm with nominal in-plane resolution of 1 mm, field-of-view ⫽ 25.6 cm and averages ⫽ 2). Proton spectra were acquired from a 1.5 ⫻ 2 ⫻ 2 cm3 region of interest (ROI), which was positioned visually on screen using the T2 axial images and was centered on the posterior limb of the internal capsule at the level of the third ventricle. Volume selection was performed using a point resolved spectroscopy sequence (PRESS) [25]. ROI acquisition parameters were TE ⫽ 90 ms, TR ⫽ 1500 ms, data points ⫽ 2048, spectral width ⫽ 2500 Hz, and acquisi-

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

tions ⫽ 256. Water suppression was achieved by means of a CHESS sequence [26]. A non-water suppressed spectrum was also collected with 16 averages with no offset frequency from the same ROI. The ROI dimensions were selected to include the whole of the posterior limb of the internal capsule with the minimum of partial volume effects. To avoid significant chemical shift displacement of the signal of interest (NAA), an offset frequency of ⫺228 Hz relative to the water frequency was applied to all three pulses of the PRESS sequence. Spectral analysis was performed with the operator blinded to the patient’s clinical details and side of motor deficit. Four Hertz of exponential line broadening was applied prior to Fourier transformation. Automatic line fitting and integration was done with the software package 1D WIN-NMR (Bruker Franzen Analytik GmbH, Bremen, Germany). The apparent NAA concentration was calculated relative to the water concentration for each internal capsule using the ratio of the areas under the NAA and water peaks (NAA/H2O). Although changes in the creative and choline peaks were seen in some of the patients, these were not analyzed in this study. Magnetisation transfer imaging was performed over five slices centered on the axial T2 image used for placement of the spectroscopy ROI. Gradient echo images were acquired with the following parameters: TR ⫽ 816 ms, TE ⫽ 13 ms, 5-mm thick slices with an interslice separation of 2.5 mm. Pre-saturation of the macromolecular matrix was preformed with 8 ⫻ 16 ms Gaussian pulses with a 1500 Hz offset and 1 ms interpulse delay. A set of reference images was obtained with the same parameters but without the pre-saturation pulses. The MTR image was generated by subtracting the pre-saturation from the reference scan and then dividing by the reference scan. This produced an image whose intensity was proportional to the signal difference ratio before and after presaturation. The mean MTR was calculated for the tissue contained within the spectroscopy ROI, which comprised the posterior limb of the internal capsule together with a small amount of basal ganglia and thalamus. 2.3. Calculation of lesion volume within the ROI In those patients in whom stroke or MS lesion was seen to involve the region enclosed by the ROI, lesion area was measured using a manually defined thresholding technique (MEDx software, Sensor Systems, Baltimore, MD, USA) and used to calculate the percentage of the ROI volume occupied by lesion. 2.4. Clinical assessment Clinical assessment for each patient group was carried out at the time of the MRS/MRI examination (SP and ML for stroke and MS patients, respectively). Specific measures of motor function obtained from the patients were the motricity index [27], the nine hole peg test [28], grip strength [29] as measured by a modified strain gauge and

371

leg extension power measured using a leg extensor rig dynamometer. A composite motor deficit score was generated by taking the mean of the percentage performance of the affected arm and leg relative to the unaffected limbs for each of the scores for motricity index, grip strength, nine hole peg test time and leg extensor power with equal weight given to each test. This was performed in an attempt to avoid floor and ceiling effects present in individual tests and provide a more accurate measure of overall motor impairment. Using this method, a complete hemiparesis with no function in the arm or leg gave a motor deficit score of 100. This composite motor score has not in itself been validated but all the tests from which it has been derived have been validated and are widely used. Hand preference was assessed using the Salmaso hand preference index [30]. 2.5. Data analysis Mean internal capsule NAA and mean internal capsule MTR were calculated for the hemisphere ipsilateral and contralateral to the motor deficit in both stroke and MS patients. For the stroke patients, the values taken from the internal capsule ipsilateral to the motor deficit (which was assumed to be normal) were taken as control values. For the MS patients, since the internal capsule ipsilateral to the motor deficit in the patients could not be assumed to be normal, control values were taken from the internal capsules of the age-matched control subjects. The percentage reduction in NAA was calculated by taking the difference between the internal capsule NAA level from each hemisphere and expressing this as a percentage of the higher NAA level for both stroke and MS patients. A similar calculation was used to determine the percentage reduction in MTR. For the MS patients, the percentage reductions in MTR in the internal capsule of the hemisphere contralateral to the motor deficit was also calculated relative to the mean MTR values obtained from the control subjects. Comparisons were made between percentage NAA reduction, percentage MTR reduction, and contralateral motor deficit score using correlational analysis. Left-sided motor deficit and NAA and MTR reductions in the right internal capsule were assigned negative values in data plots, except where reductions were seen in the internal capsule ipsilateral to the motor deficit, when the sign was reversed. Similar correlational analysis was performed for the MS data using the reductions in contralateral hemisphere MTR expressed relative to the control subjects mean MTR. Throughout the text, hemispheres are defined as “contralateral” and “ipsilateral” relative to the side of motor deficit. 2.6. Statistics Null hypotheses were tested using the Mann–Whitney U test for non parametric data and the ANOVA test for parametric data. Correlations were tested using Spearman’s rank test.

372

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

Table 1a Clinical data for the stroke patients Stroke Stroke location/ patient Age/Sex Handedness Hemisphere

Time since stroke, months

1 2 3 4 5 6 7 8 9 10 11 12

10 1 2.5 2 1 1.5 2 1 1 7 8 10

77/M 77/F 75/M 75/F 70/F 77/M 77/M 70/M 77/M 31/F 77/M 61/F

R R R R R R L R R R R L

multiple lacunes striatocapsular/R multiple lacunes frontoparietal/R corona radiata/L corona radiata/parietal/L frontoparietal/R frontoparietal/L temperoparietal/L striatocapsular/R multiple lacunes striatocapsular/frontoparietal/L

3. Results Clinical details for the stroke and MS patients are shown in Tables 1a and 1b. Data from the MS control subjects are shown in Table 2. The stroke patients were significantly older than the MS patients (70.3 ⫾ 13.2 years vs. 48.8 ⫾ 8.7 years, mean ⫾ standard deviation, p ⬍ 0.001). There was no significant difference between the age of the MS patients and their controls. Tables 3a and 3b show the MR data and functional assessment scores for the stroke and MS patients. The MS control subjects had higher internal capsule MTR as compared to the stroke control values (36.2 ⫾ 1.8 vs. 30.9 ⫾ 4.6, p ⫽ 0.004; MS controls and stroke patient ipsilateral capsule values respectively) (Fig. 1). Similarly, MS control subjects had higher internal capsule NAA values as compared to stroke control values (0.18 ⫾ 0.01 vs. 0.17 ⫾ 0.02. p ⫽ 0.002; MS controls and stroke patient ipsilateral capsule values respectively) (Fig. 2). MTR data were not available from one stroke patient and one MS patient, owing to motion artefact on the MT images. The MTR from the internal capsule was significantly re-

4. Discussion

Table 1b Clinical data for the MS patients MS patient

Age/Sex

Handedness

Disease typea

Disease duration, years

13 14 15 16 17 18 19 20 21 22 23 24

46/F 41/F 40/F 55/F 40/M 66/M 63/M 50/M 45/M 48/F 41/M 51/F

R R R R R R R R R L R R

RR RR SP RR SP SP SP SP RR RR SP RR

10 15 10 12 16 42 6 15 8 21 23 7

a

duced in the stroke affected hemisphere compared to the control hemisphere of the stroke patients (25.2 ⫾ 6.6 vs. 30.9 ⫾ 4.6, p ⫽ 0.04; for stroke and control values, respectively) (Fig. 1). In the MS patients, there was no significant difference between MTR values from the ipsilateral and contralateral hemispheres but both were reduced compared to control subjects (29.7 ⫾ 4.6 and 29.9 ⫾ 4.5 vs. 36.2 ⫾ 1.8, p ⫽ 0.001, p ⫽ 0.03; MS contralateral and ipsilateral vs. control respectively) (Fig. 1). Internal capsule NAA was significantly lower in the stroke affected hemisphere as compared to the control hemisphere of the stroke patients (0.093 ⫾ 0.064 vs. 0.158 ⫾ 0.017, p ⫽ 0.01; stroke and control values, respectively) (Fig. 2). In the MS patients, NAA was significantly reduced in the contralateral as compared to the ipsilateral hemisphere (0.16 ⫾ 0.01 vs. 0.18 ⫾ 0.02, p ⫽ 0.03) and in the contralateral but not the ipsilateral hemisphere compared to controls (0.16 ⫾ 0.01 vs. 0.18 ⫾ 0.01, p ⫽ 0.003 and 0.18 ⫾ 0.02 vs. 0.18 ⫾ 0.01 for MS contralateral vs. control and MS ipsilateral vs. control, respectively) (Fig. 2). Internal capsule NAA reduction showed a correlation with motor deficit for both stroke and MS patients ( p ⬍ 0.001 and p ⫽ 0.04; stroke and MS patients, respectively) (Fig. 3). MTR reduction was correlated both with motor deficit (Fig. 4) (r ⫽ 0.63, p ⫽ 0.03) and NAA reduction (Fig. 5) (r ⫽ 0.74, p ⫽ 0.006) in the stroke patients. For the MS patients, MTR reduction was not related to motor deficit or NAA reduction whether reductions were defined relative to the ipsilateral internal capsule values or to the control mean values. In some patients, there was lesion in the ROI. In the stroke patients, there was a correlation between the percentage of the ROI occupied by lesion and both the MTR reduction and the percentage NAA loss ( p ⫽ 0.01 and p ⫽ 0.003, respectively) (Figs. 6a and 6b). In the MS patients, only 6/24 voxels contained lesions and no significant correlation was seen between the amount of lesion within the voxel and either MTR or NAA reduction.

RR, relapsing and remitting; SP, secondary progressive.

We have examined the relationship between indexes of myelin loss and axonal injury localised to the descending motor pathways in patients with asymmetric motor deficit secondary to stroke or chronic MS. In both stroke and MS, there was a correlation between axonal injury in the internal capsule and motor deficit. Reduction in MTR from the internal capsule was seen in both stroke and MS patients indicating myelin loss. However, in the stroke patients, there was a correlation between MTR, axonal injury, and motor deficit whereas no such correlation was seen in the MS patients. This suggests that demyelination is closely linked to axonal loss in stroke but not in MS where demyelination may occur in the absence of axonal injury.

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

373

Table 2 Control subjects data Subject

Age/Sex

Handed ness

Left internal capsule NAA/H20a

Right internal capsule NAA/H20a

Left internal capsule MTR

Right internal capsule MTR

1 2 3 4 5 6 7 8

29/M 41/F 40/F 55/F 40/M 66/M 32/M 50/M

R R R R R R R R

0.18 0.16 0.17 0.18 0.20 0.18 0.17 0.17

0.17 0.17 0.18 0.19 0.19 0.17 0.18 0.16

37.0 39.0 36.0 35.0 37.3 36.6 32.4 35.8

37.0 40.0 36.3 34.0 36.0 36.0 34.5 35.8

a

Not corrected for saturation effects.

The similar relationship between NAA reduction and motor impairment in both stroke and chronic MS suggests that axonal injury is the final common pathway that mediates functional deficit in the two diseases. The correlation between NAA reduction and motor impairment was considerably stronger in stroke. This may have been because the range of motor deficit was smaller in the MS patients owing to the difficulty of finding patients with strongly lateralised impairment. Also, in the MS patients, spinal cord disease may have contributed to the motor impairment measured. However, it is interesting to note that the relationship between NAA loss from the internal capsule and motor impairment was similar in both stroke and MS suggesting that the extent of axonal damage is the common determinant of the magnitude of motor impairment in both diseases despite the difference in pathological mechanisms. A close relationship was observed between MR measures of myelin loss and axonal injury in the stroke patients. In stroke, there is neuronal and axonal death at the site of ischaemia followed by anterograde loss of axons downstream from the site of the injury (Wallerian degeneration), a process that may take months to complete in mammalian brain [31]. Retrograde changes may also occur [32]. Thus, the close relationship that we observed between NAA loss and MTR decrease in stroke (Fig. 4) is unsurprising since axonal injury and demyelination would be expected to occur together. However, it is of note that there was considerable variation in the amount of MTR decrease for a given magnitude of NAA reduction: for example, two patients had a complete loss of NAA but the MTR decreases were 60% and 27%. Further, there appeared to be a closer relationship between motor impairment and NAA loss (Fig. 3) than between motor impairment and MTR decrease (Fig. 5). This indicates that for a given amount of axonal injury, there was a variable amount of myelin loss observed in the stroke patients. The relationship between indexes of axonal injury and myelin loss in stroke may be examined further by considering the proportion of the ROI occupied by lesion. There was a correlation between the percentage of the ROI volume occupied by stroke and both the percentage reduction in MTR and the percentage reduction in NAA (Fig. 6). How-

ever, whilst the percentage MTR reduction was consistently slightly less than the percentage ROI stroke volume, the NAA loss was more variable with patients showing greater NAA losses than would be expected from the percentage of stroke within the ROI. This was particularly the case for patients with cortical as opposed to striatocapsular strokes presumably as a result of Wallerian degeneration. One might have expected the relationship between the percentage of the ROI volume occupied by stroke and measures of axonal injury and myelin loss to have been more similar indicating that loss of myelin was occurring in step with axonal injury. However, histological studies of Wallerian degeneration show that the breakdown of axoplasm is complete by 72 h but myelin removal takes several months [33]. Thus, within the time scale of our study in which most patients were examined within 2.5 months of stroke, Wallerian degeneration may not have advanced sufficiently to cause much myelin loss despite the fact that significant axonal injury had occurred. It is interesting to note that even within a stroke lesion, there appears to be a time lag between loss of NAA and MTR decrease. Large NAA losses have been shown to occur within the first 3 days [5,34] whereas MTR decreases are very small during the first week and then become progressively larger over the next 2 months [19]. It should be noted that 2 of the 3 patients with striatocapsular strokes were examined at least 7 months after stroke whereas the patients with cortical strokes were examined earlier at 1–2.5 months after stroke and this may have contributed to the discrepancy between the magnitude of NAA reduction and the magnitude of the MTR reduction observed between the cortical and striatocapsular strokes. No clear relationship was observed between NAA loss and MTR reduction or between motor deficit and MTR reduction in the MS patients. MS is characterised by recurrent multifocal inflammatory lesions and demyelination. It is known that neurological function can recover fully despite persistent demyelination and that demyelination with relative axonal sparing and gliosis, which may increase MTR, occur. These factors make a close relationship between MTR and axonal loss and MTR and motor impairment less likely in MS patients than in patients after stroke. There have been only two other studies to these authors’

374

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

Table 3a MR data and motor impairment scores for the stroke patients

Stroke patient

Left internal capsule NAA/H20a

Right internal capsule NAA/H20a

Left internal capsule MTR

Right internal capsule MTR

Composite motor deficit score/ Affected side

% ROI occupied by stroke Left/Right

1 2 3 4 5 6 7 8 9 10 11 12

0.14 0.17 0.18 0.15 0.17 0.09 0.14 0.00 0.14 0.17 0.15 0.00

0.12 0.10 0.18 0.04 0.16 0.15 0.07 0.18 0.16 0.04 0.15 0.16

26.7 28.3 35.1 35.3 29.7 – 24.7 26.7 26.3 35.0 27.6 11.7

27.8 26.0 37.3 30.8 33.3 – 19.3 36.3 25.3 17.7 27.1 29.3

14/R 60/L 42/R 100/L 34/R 71/R 93/L 93/L 44/R 74/L 31/L 77/R

3/10 0/42 0/0 0/3 0/0 9/0 14/0 38/0 0/0 0/52 4/2 70/0

a

Not corrected for saturation effects. –, denotes motion artefact making the data uninterpretable.

knowledge looking at the relationship between measures of axonal injury and demyelination in MS and these have concentrated on changes within the lesions visible on imaging. The first by Kimura et al. [35] found a correlation between MTR and NAA decrease suggesting that axonal damage was linked to levels of demyelination within plaques. However, Hiehle et al. [36] did not find such a relationship. There appears to be a correlation between disability and both average lesion MTR [37] and MTR decrease in the NAWM [22] in MS. The MTR has been shown to be reduced to a greater extent in lesions than within NAWM consistent with greater levels of demyelination in lesions [36]. We did not observe a correlation between the amount of lesion within the ROI and MTR decrease or NAA loss in the MS patients. This is unsurprising in view of the fact that only 6 out of the 24 voxels contained lesion and the mean

ROI lesion volume was approximately 1%. Thus, MTR and NAA decreases were occurring in NAWM possibly reflecting the effects of longstanding Wallerian degeneration [21]. However, the absence of a correlation between MTR decrease and axonal injury suggests that demyelination, gliosis and microscopic inflammatory changes also occur in the NAWM. Thus, although patients with very severe MS will have lower MTR in lesions and within the NAWM compared to mildly affected patients, it is unlikely that within a group of patients with similar levels of disease there would be a clear correlation between axonal injury, demyelination and impairment. There was a significant difference in MTR and NAA in the internal capsule in the MS control subjects as compared to the unaffected internal capsule of the stroke patients. This may have been because the stroke patients were significantly older than the MS control subjects. Age related re-

Table 3b MR data and motor impairment scores for all MS patients Stroke patient

Left internal capsule NAA/H20a

Right internal capsule NAA/H20a

Left internal capsule MTR

Right internal capsule MTR

Composite motor deficit score/ Affected side

% ROI occupied by T2 lesion Left/Right

13 14 15 16 17 18 19 20 21 22 23 24

0.15 0.16 0.14 0.18 0.19 0.15 0.15 0.15 0.21 0.16 0.12 0.16

0.17 0.17 0.15 0.16 0.18 0.20 0.14 0.16 0.16 0.19 0.18 0.21

37.7 32.9 29.2 – 27.8 28.9 28.1 30.3 27.4 25.1 27.1 38.4

35.5 32.2 27.2 – 25.1 25.9 25.2 27.2 29.3 24.8 31.0 39.3

20/R 34/L 17/L 12/L 34/L 39/R 24/R 33/L 27/L 21/R 26/R 16/R

0/0 0/8 0/9 0/0 0/0 0/0 0/0 3/0 0/0 8/13 0/2 3/3

a

Not corrected for saturation effects. –, denotes motion artefact making the data uninterpretable.

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

Fig. 1. MTR values from the internal capsule ipsilateral and contralateral to the motor deficit for MS and stroke patients and control values for the MS patients from age-matched volunteers.

ductions in MTR [38] and NAA [39,40] have been shown previously highlighting the importance of using agematched controls. It is also possible that the presence of more widespread vascular disease in some stroke patients may have caused occult damage to the internal capsule of the unaffected hemisphere. There are some possible confounds in our study owing to partial volume effects in that the ROI contained some grey matter as well as the posterior limb of the internal capsule. In the MS patients, demyelination changes and NAA losses from the white matter may have been proportionally greater than that measured owing to the dilutional effect of inclusion of some grey matter. In the stroke patients in whom there was stroke seen within the ROI, NAA loss, and MTR decrease would have occurred from the infarcted grey matter. In patients in whom the majority of measured change was secondary to Wallerian degeneration, there may have been a similar dilutional effect to that proposed for the MS

Fig. 2. NAA levels (expressed relative to the unsuppressed water peak) from the internal capsule ipsilateral and contralateral to the motor deficit for MS and stroke patients and control values for the MS patients from age-matched volunteers.

375

Fig. 3. Combined motor deficit score versus the percentage NAA reduction ((higher ⫺ lower NAA/higher NAA) ⫻ 100) in the internal capsule for stroke and MS patients. Stroke patients are represented by circular symbols and MS patients by triangular symbols. Left-sided disability and lower NAA in the right internal capsule are assigned negative values.

patients. Alternatively, the presence of retrograde thalamic changes and metabolic effects that are known to occur following cortical infarction may have caused changes in the grey matter within the ROI even in cases in which there was little lesion within the ROI. However, the presence of a similar graded relationship between NAA loss and motor deficit for stroke and MS patients would imply that the partial volume effects were not of major significance. Further confounds in the study relate to the MS data. MS is a disease which is often widespread throughout the central nervous system, which leads to a number of problems when attempting to define damage in a specific functional system. Because of this we have targeted our measurements of markers of axonal injury and myelin loss to a specific tract and have correlated this with outcome measures de-

Fig. 4. Combined motor deficit score versus percentage MTR reduction ((higher ⫺ lower MTR/higher MTR) ⫻ 100) in the internal capsule for stroke patients. MTR reductions in the internal capsule ipsilateral to the motor deficit are assigned negative values.

376

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

Fig. 5. Internal capsule percentage NAA reduction versus percentage MTR reduction for stroke patients.

signed to reflect dysfunction specifically in that tract. As mentioned earlier, it is possible that spinal cord disease may have been present in some of the patients in this study, which may have contributed to the measured motor deficit. However, spinal cord disease is often bilateral and patients were selected only if they had strongly lateralised impairment. Also, the fact that retrograde damage to axons occurs following distal damage makes it possible that axons at the level of the internal capsule would be damaged even as a result of a cord lesion. A further problem caused by the widespread nature of MS pathological change relates to the analysis of the MS data in that, unlike in the stroke patients, the ipsilateral internal capsule could not be assumed to be normal even in those patients who had no deficit in the contralateral limbs. The NAA levels measured from the ipsilateral hemisphere in the MS patients in the current study were very similar to the control subject values, indicating that significant axonal injury had not occurred in the descending motor pathways in the ipsilateral hemisphere. Hence, it was not unreasonable to calculate the NAA reduction in the contralateral capsule relative to the ipsilateral capsule values, particularly

as differences in magnet performance between subjects would have introduced additional error in comparing NAA reductions relative to the control mean value. Further, use of the NAA reduction expressed relative to the ipsilateral capsule values would have tended to underestimate rather than overestimate any relationship between NAA reduction and motor deficit. In contrast to NAA, the MTR values from both the ipsilateral and contralateral capsules were reduced relative to the control values in the MS patients. This in itself indicated that there was unlikely to be a relationship between demyelination and axonal injury in the current study. Thus, MTR reductions calculated relative to the ipsilateral capsule values provided a measure of capsular MTR asymmetry rather than a measure of the reduction from normal. Because of this, the MTR reduction was also calculated relative to the control mean MTR value. However, whichever method of calculating MTR reduction was used, there was no relationship to NAA loss in the internal capsule or therefore, to motor deficit. Finally, the motor deficit was calculated relative to the “unaffected” limbs making this also a measure of motor deficit asymmetry in the MS patients. Data from the MS patients “unaffected limbs” showed a mean performance of approximately 80% on the motor deficit scores compared to control values from the literature and our own studies (data not shown). This indicates that relatively minor deficits were present in the “unaffected” limbs in keeping with the fact that the NAA levels from the “unaffected” capsule were within the normal range. In conclusion, the use of MTR and proton spectroscopy, localised to a specific functional system of the brain, has allowed us to evaluate the relationship between indexes of axonal injury and demyelination and functional impairment in two diseases of the CNS with widely differing pathological characteristics. In stroke, axonal injury appears to be closely liked to myelin loss within the infarct itself. The relationship between axonal damage and myelin loss in anterograde axonal injury in stroke is less clear but our findings suggest that there may be a time lag between NAA

Fig. 6. Percentage of the ROI volume occupied by stroke versus (a) percentage NAA reduction and (b) percentage MTR reduction. Data points have been labeled with the patient’s identification number except for data points close to the origin which been left unlabeled for the sake of clarity.

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

loss and myelin breakdown. In MS, there is a reduction in MTR in the NAWM consistent with loss of myelin. However, this is less closely linked to axonal injury presumably because of the pathological heterogeneity of MS in which gliosis and demyelination with relative axonal sparing occur. In both stroke and MS, axonal injury as measured by NAA loss, appears to be the primary determinant of the magnitude of functional impairment. In future, monitoring of NAA loss may provide a useful surrogate marker of outcome when assessing treatment effect in MS and stroke.

[15]

[16]

[17]

[18]

Acknowledgments We would like to acknowledge the Medical Research Council for their support of the FMRIB centre and the MRC Biochemical and Clinical Magnetic Resonance Spectroscopy Unit. References [1] Miller DH. Magnetic resonance in monitoring the treatment of multiple sclerosis. Ann Neurol 1994;36(Suppl):S91– 4. [2] Saver JL, Johnston KC, Homer D, et al. Infarct volume as a surrogate or auxiliary outcome measure in ischemic stroke clinical trials. Stroke 1999;30:293– 8. [3] Urenjak J, Williams SR, Gadian DG, Noble M. Specific expression of N-acetylaspartate in neurons, oligodendrocyte-type-2 astrocyte progenitors, and immature oligodendrocytes in vitro. J Neurochem 1992; 59:55– 61. [4] Bruhn H, Frahm J, Gyngell ML, Merboldt KD, Hanicke W, Sauter R. Cerebral metabolism in man after acute stroke: new observations using localised proton NMR spectroscopy. Magn Reson Med 1988;9:126 –31. [5] Graham GD, Blamire AM, Rothman DL, et al. Early temporal variation of cerebral metabolites after human stroke. A proton magnetic resonance spectroscopy study. Stroke 1993;24:1891– 6. [6] Matthews PM, Francis G, Antel J, Arnold DL. Proton magnetic resonance spectroscopy for metabolic characterization of plaques in multiple sclerosis. 1991;41:1251– 6 [published erratum appears in Neurology 1991 Nov;41(11):1828]. [7] Davie CA, Hawkins CP, Barker GJ, et al. Serial proton magnetic resonance spectroscopy in acute multiple sclerosis lesions. Brain 1994;117:49 –58. [8] Fu L, Matthews PM, De SN, et al. Imaging axonal damage of normalappearing white matter in multiple sclerosis. Brain 1998;121:103–13. [9] Husted CA, Goodin DS, Hugg JW, et al. Biochemical alterations in multiple sclerosis lesions and normal-appearing white matter detected by in vivo 31P and 1H spectroscopic imaging. Ann Neurol 1994;36:157– 65. [10] De Stefano N, Matthews PM, Fu L, et al. Axonal damage correlates with disability in patients with relapsing-remitting multiple sclerosis. Results of a longitudinal magnetic resonance spectroscopy study. Brain 1998;121:1469 –77. [11] Davie CA, Barker GJ, Webb S, et al. Persistent functional deficit in multiple sclerosis and autosomal dominant cerebellar ataxia is associated with axon loss. Brain 1995;118:1583–92 [published erratum appears in Brain 1996 Aug;119(Pt 4):1415]. [12] Pendlebury ST, Blamire AM, Lee MA, Styles P, Matthews PM. Axonal injury in the internal capsule correlates with motor impairment after stroke. Stroke 1999;30:956 – 62. [13] Lee MA, Blamire AM, Pendlebury ST, et al. Axonal injury or loss in the internal capsule and motor impairment in multiple sclerosis. Arch Neurol 2000;57:65–70. [14] Dousset V, Grossman RI, Ramer KN, et al. Experimental allergic encephalomyelitis and multiple sclerosis: lesion characterization with

[19]

[20]

[21]

[22]

[23]

[24] [25] [26] [27] [28]

[29]

[30] [31]

[32] [33]

[34]

[35]

377

magnetization transfer imaging. Radiology 1992;182:483–91. [published erratum appears in Radiology 1992;183:878] Mehta RC, Pike GB, Enzmann DR. Measure of magnetization transfer in multiple sclerosis demyelinating plaques, white matter ischemic lesions, and edema. AJNR AM J Neuroradiol 1996;17:1051–5. Thorpe JW, Barker GJ, Jones SJ, et al. Magnetisation transfer ratios and transverse magnetisation decay curves in optic neuritis: correlation with clinical findings and electrophysiology. J Neurol Neurosurg Psychiatry 1995;59:487–92. Dousset V, Brochet B, Vital A, et al. Lysolecithin-induced demyelination in primates: preliminary in vivo study with MR and magnetization transfer. AJNR Am J Neuroradiol 1995;16:225–31. Silver NC, Barker GJ, MacManus DG, Miller DH, Thorpe JW, Howard RS. Decreased magnetisation transfer ratio due to demyelination: a case of central pontine myelinolysis (letter). J Neurol Neurosurg Psychiatry 1996;61:208 –9. Hanyn H, Imon Y, Sakurai H, et al. Diffusion-weighted magnetic resonance and MY magnetization transfer imaging in the assessment of ischemic human stroke. Intern Med 1998;37:360 –5. Tomiak MM, Rosenblum JD, Prager JM, Metz CE. Magnetization transfer: a potential method to determine the age of multiple sclerosis lesions. AJNR Am J Neuroradiol 1994;15:1569 –74. Loevner LA, Grossman RI, Cohen JA, Lexa FJ, Kessler D, Kolson DL. Microscopic disease in normal-appearing white matter on conventional MR images in patients with multiple sclerosis: assessment with magnetization-transfer measurements. Radiology 1995;196:511–5. Filippi M, Campi A, Dousset V, et al. A magnetization transfer imaging study of normal-appearing white matter multiple sclerosis. Neurology 1995;45:478 – 82. Lexa FJ, Grossman RI, Rosenquist AC. Detection of early axonal degeneration in the mammalian central nervous system by magnetization transfer techniques in magnetic resonance imaging. Ann N Y Acad Sci 1993;679:336 – 40. Hatano S. Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ 1976;54:541–53. Bottomley PA. Spatial localization in NMR spectroscopy in vivo. Ann N Y Acad Sci 1987;508:333– 48. Haase A, Frahm J, Hanicke W, Matthaei D. 1H NMR chemical shift selective (CHESS) imaging. Phys Med Biol 1985;30:341– 4. Demeurisse G, Demol O, Robaye E. Motor evaluation in vascular hemiplegia. Eur Neurol 1980;19:382–9. Sharpless JW. The nine hole peg test of finger hand co-ordination for the hemiplegic patient. In: Mossman, editor. A Problem-Orientated Approach to Stroke Rehabilitation. Springfield: Charles CT Thomas, 1982. p. 420 –3. Sunderland A, Tinson D, Bradley L, Hewer RL. Arm function after stroke. An evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry 1989;52:1267–72. Salmaso D, Longoni AM. Problems in the assessment of hand preference. Cortex 1985;21:533– 49. Griffin JW, George R, Lobato C, Tyor WR, Yan LC, Glass JD. Macrophage responses and myelin clearance during Wallerian degeneration: relevance to immune-mediated demyelination. J Neuroimmunol 1992;40:153– 65. Barron KD, Mean ED, Larsen E. Ultrastructure or retrograde degeneration in thalamus of rat. J Neuropathol Exp Neurol 1973;32:218 – 44. George R, Griffin JW. Delayed macrophage responses and myelin clearance during Wallerian degeneration in the central nervous system: the dorsal radiculotomy model. Exp Neurol 1994;129:225–36. Gideon P, Henriksen O, Sperling B, et al. Early time course of N-acetylaspartate, creatine and phosphocreatine, and compounds containing choline in the brain after acute stroke. A proton magnetic resonance spectroscopy study. Stroke 1992;23:1566 –72. Kimura H, Grossman RI, Lenkinski RE, Gonzalez SF. Proton MR spectroscopy and magnetization transfer ratio in multiple sclerosis: correlative findings of active versus irreversible plaque disease. AJNR Am J Neuroradiol 1996;17:1539 – 47.

378

S.T. Pendlebury et al. / Magnetic Resonance Imaging 18 (2000) 369 –378

[36] Hiehle JFJ, Lenkinski RE, Grossman RI, et al. Correlation of spectroscopy and magnetization transfer imaging in the evaluation of demyelinating lesions and normal appearing white matter in multiple sclerosis. Magn Reson Med 1994;32:285–93. [37] Gass A, Barker GJ, Kidd D, et al. Correlation of magnetization transfer ratio with clinical disability in multiple sclerosis. Ann Neurol 1994;36:62–7. [38] Silver NC, Barker GJ, MacManus DG, Tofts PS, Miller DH. Magnetisation transfer ratio of normal brain white matter: a normative

database spanning four decades of life. J Neurol Neurosurg Psychiatry 1997;62:223– 8. [39] Charles HC, Lazeyras F, Krishnan KR, et al. Proton spectroscopy of human brain: effects of age and sex. Prog Neuropsychopharmacol Biol Psychiatry 1994;18:995–1004. [40] Christiansen P, Toft P, Larsson HB, Stubgaard M, Henriksen O. The concentration of N-acetyl aspartate, creatine ⫹ phosphocreatine, and choline in different parts of the brain in adulthood and senium. Magn Reson Imaging 1993;11:799 – 806.