- Email: [email protected]
Longitudinal changes in myelin water fraction in two MS patients with active disease
Journal of the Neurological Sciences 276 (2009) 49–53
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Longitudinal changes in myelin water fraction in two MS patients with active disease I.M. Vavasour a,⁎, C. Laule a, D.K.B. Li a, J. Oger b, G.R.W. Moore c, A. Traboulsee b, A.L. MacKay a,d a
Department of Radiology, Canada Department of Medicine (Neurology) and Multiple Sclerosis Clinic, Canada c Department of Pathology and Laboratory Medicine, Canada d Department of Physics and Astronomy, University of British Columbia, Vancouver, Canada b
a r t i c l e
i n f o
Article history: Received 3 March 2008 Received in revised form 18 August 2008 Accepted 21 August 2008 Available online 25 September 2008 Keywords: MRI Multiple sclerosis Water content Myelin water fraction New lesions Longitudinal
a b s t r a c t Multiple sclerosis (MS) is characterised by focal areas that undergo cycles of demyelination and remyelination. Although conventional magnetic resonance imaging is very effective in localising areas of damage, these techniques are not pathology specific. A newer technique, T2 relaxation, can separate water from brain into three compartments: (1) a long T2 component (N2 s) arising from CSF, (2) an intermediate T2 component (~ 80 ms) attributed to intra- and extra-cellular water and (3) a short T2 component (~ 20 ms) assigned to water trapped in between the myelin bilayers (termed myelin water). Histological evidence shows that myelin water is a specific marker of myelination. The goal of this work was to follow changes in total water content (WC) and myelin water fraction (MWF) in evolving MS lesions over one year. Multi-echo T2 relaxation data was collected and used to measure water content and myelin water fraction from three new MS lesions in two patients. WC increased in the three large (N1 cm3) lesions at lesion appearance and remained elevated in the central core. Two lesions showed low MWF in the core suggesting demyelination upon first appearance. At later time points, one lesion showed a decrease in volume of low MWF, reflecting remyelination whereas the volume of low MWF in the other lesion core remained constant. This work provides evidence that MWF and WC can monitor demyelination and remyelination in MS. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Multiple sclerosis (MS) is an autoimmune disease of the central nervous system that is characterised by focal areas of varying degrees of inflammation and demyelination. Individual lesions, while appearing indistinguishable on conventional T2 weighted images, demonstrate considerable heterogeneity on histopathologic evaluation with varying degrees of demyelination, axonal loss, inflammation and gliosis. Histological studies show evidence that cycles of demyelination and remyelination occurs in lesions [1], however, the timescales are unknown since pathological studies provide only one snapshot of the state of a lesion. Magnetic resonance imaging (MRI), which allows for in vivo measurement of lesions over time, is helpful to investigate chronological changes in MS brain [2]. Visualising demyelination and remyelination over time would add
⁎ Corresponding author. MRI Research Group, Rm M10, Purdy Pavilion, UBC Hospital, 2221 Wesbrook Mall, Vancouver, BC, Canada, V6T 2B5. Tel.: +1 604 822 0357; fax: +1 604 827 3339. E-mail address: [email protected] (I.M. Vavasour). 0022-510X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2008.08.022
depth to the contribution of MRI to measuring the evolution of MS lesions. T2 relaxation is an MRI technique that enables separation of the tissue water signal from brain into three components: (1) a long T2 component (N2 s) arising from CSF, (2) an intermediate T2 component (~ 80 ms) attributed to intra- and extra-cellular water and (3) a short T2 component (~ 20 ms) assigned to water trapped in between the myelin bilayers [3–6]. The sum of the signals from all three components is proportional to the total water content (WC) of the tissue and the fraction of signal from the myelin water component is termed the myelin water fraction (MWF). In order to determine quantitative values for WC and MWF, acquisition of T2 relaxation was done on a single-slice otherwise off-resonance effects from other slices can affect the signal from the different water components [7]. Previous work has shown strong quantitative correlations between luxol fast blue histological staining for myelin and MWF, providing compelling evidence that the short T2 component is a marker of myelin [8]. Prior work in vivo found the short T2 component was variably decreased in MS lesions [3,9–11], and diffusely reduced in the normal appearing white matter (NAWM) when compared to healthy controls [10,12]. The goal of this work was to study the longitudinal changes in total water content (WC) and myelin water fraction (MWF), obtained from a single imaging slice, in newly appearing MS lesions. The fortuitous appearance of lesions on the slice of interest allowed observation of
50
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
WC and MWF before, during and after the formation of these new lesions.
lesions and were followed in this paper. Both patients had clinically definite MS according to the Poser criteria [13] and the presence of at least one enhancing lesion on a screening scan performed 4 weeks prior to the start of the study. They were two women aged 36 and 50 years with disease duration of 9 and 10 years, respectively. They both had relapsing MS and had been followed yearly in the UBC MS clinic since diagnosis. They had an EDSS of 2.5 and 3.0 at inclusion in the study. Both had declined disease modifying drugs. Patient 2 had received IV solumedrol 4 months prior to inclusion. They were followed for one year with scans at month 0, 2, 4, 6 and 12. Both had a clinical relapse between month 6 and month 12 scans. Patient 2 received IV methylprednisolone for this relapse 2 months before the final scan. Newly enhancing lesions that were greater than 1 cm3 and had at least one follow-up scan were studied. Patient 1 had two lesions which qualified (lesions A and B) and patient 2 had one lesion (lesion C). A diagram indicating the time of scans, lesion appearance and clinical relapses for both patients is shown in Fig. 1.
2. Materials and methods
2.2. MRI studies
2.1. Patient selection
Images were acquired using a 1.5 Tesla MR scanner (GE Medical Systems, Milwaukee, USA, version 5.7). First, proton density (PD) and T2-weighted scans (TR 2500 ms, TE 30/90 ms, 22 contiguous slices, matrix 256 × 192) were obtained using a dual echo conventional spin echo sequence. T2 relaxation was then measured using a single-slice
Fig. 1. Diagram of the timeline for scanning, lesion appearance and clinical relapses in both patients. Below the dashed line is data for patient 1 and above the dashed line is data for patient 2. Arrows indicate the time of scans, letters indicate the time of first appearance of new lesions A, B and C, and an X indicates the time of a clinical relapse with a subscript s indicating treatment with steroids.
Seven subjects with multiple sclerosis volunteered for this study and signed an informed written consent approved by the Clinical Research Ethics Board of our institution. Of these subjects, two had large (N1 cm3)
Fig. 2. Images for large lesion A at pre-lesion (1st row), lesion first appearance (2nd row), 2 months post-lesion (3rd row), 4 months post-lesion (4th row) and 10 months post-lesion (5th row). Proton-density (left column), T1-weighted post-Gd-DTPA (middle column) and myelin map (right column) images are shown. The box indicates the location of the lesion of interest.
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
32-echo relaxation sequence (TE 10 ms, TR 3000 ms, 4 averages, matrix 256 × 128, bandwidth 32 kHz) [3]. Next, T1 relaxation was measured using a single-slice fast gradient echo inversion recovery prepared sequence (TE 8 ms, 1 NEX, 15 TIs ranging from 0.05–2.5 s). The slice of interest for the T2 and T1 relaxation measurements was chosen to include the enhancing lesion observed on the screening scan. Finally, a post-contrast T1-weighted spin echo scan (TR 550 ms, TE 8 ms, 22 slices, matrix 256 × 192) was collected 5 min after the injection of gadolinium-DTPA (0.2 mL/kg). The field of view for all exams was 22 cm and the slice thickness was 5 mm. Exact patient repositioning was attempted by first performing a true midline sagittal localiser and then manually positioning the patient so that the angle of a line drawn through the base of the genu and splenium of the corpus callosum was within one degree of the baseline scan. Four NiCl–agarose water standards were placed within the T2 relaxation slice. 2.3. Calculation of MR measurements In-house software was created for registration of the single-slice T2 and T1 relaxation data whereby the matching PD slice and the first image from the relaxation data were enlarged using bilinear interpolation to 4 times original size, the mean sum of squares difference between the images was taken and minimised by single pixel shifts to the right or left, up or down and also by fractional rotations. By enlarging the images, single pixel shifts translated into fractional pixel shifts on the original image size, thereby increasing the accuracy of the registration. Regions of interest (ROIs) were drawn on the first echo of the 32-echo experiment around the core and periphery of large (N1 cm) hyperintense lesions as well as in adjacent NAWM and mapped onto registered images from all months. For each ROI, the T1, MWF and WC were calculated. The T1 relaxation data was fit to a mono-exponential function. T2 relaxation distributions were calculated from the 32-echo experiment using a regularised non-negative least-squares (NNLS) algorithm [14]. The total WC was defined as the total area under the T2 distribution normalised to the temperature corrected water standards and corrected for T1 relaxation. Myelin water maps were generated by
51
displaying the MWF (area from 10–40 ms divided by the entire area of the T2 distribution) at each pixel in the image [3]. 3. Results Seven patients were recruited with MRI active disease as seen by the presence of a gadolinium enhancing lesion. These lesions determined the slice of interest. By happenstance, the three lesions in the two subjects examined in this study appeared in the same slice as the original enhancing lesion and allowed for water content and myelin water measurements before and after the lesions appeared. Clinically, the two patients were also active with patient 1 having a relapse at month 8 (affecting the spinal cord) and patient 2 having a relapse 2 months prior to screening and another two at months 7 and 8 (both clinically localised to the brainstem) into the study (Fig. 1). All relapses in patient two were treated with IV steroids. Figs. 2–4 show the evolution for each of the lesions on PD (related to WC), Gadolinium T1 and myelin water images. An increase in water content compared to baseline was seen in all lesions studied. Lesion A (Fig. 2) showed a virtual absence of myelin water in the very centre of the lesion at first appearance with the periphery showing a reduction in MWF. At later months, the MWF of the periphery recovered to normal values whereas the core still had low MWF although its size had decreased (Table 1). Later still, a wide area of decreased MWF, related to the appearance of a new enhancing lesion adjacent to lesion A, filled the region. Lesion B (Fig. 3) showed a slight decrease in MWF over the entire lesion area at first appearance which recovered within 6 months (Table 1). Lesion C (Fig. 4) also showed a relatively homogeneous decrease in MWF over the entire lesion area at first appearance. However, at 6 month follow-up, the MWF at the core of the lesion decreased further whereas the periphery recovered to prelesion values (Table 1). 4. Discussion The two patients serially studied here had active disease but nothing that departed from the usual course of MS: i.e. more new MRI
Fig. 3. Images for large lesion B at pre-lesion (top row), lesion first appearance (middle row) and 6 months post-lesion (bottom row). Proton-density (left column), T1-weighted postGd-DTPA (middle column) and myelin map (right column) images are shown. The box indicates the location of the lesion of interest.
52
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
Fig. 4. Images for large lesion C at pre-lesion (top row), lesion first appearance (middle row) and 6 months post-lesion (bottom row). Proton-density (left column), T1-weighted postGd-DTPA and myelin map (right column) image are shown. The box indicates the location of the lesion of interest.
lesions than relapses and a relatively poor correlation between clinical relapses and MRI activity. They fit, however, with the loose relationship between clinical and MRI events by which younger patients with one relapse per year tend to have more enhancing lesions than older patients with less active disease. It is also generally accepted that, when patients are screened for the presence of gadolinium enhancement, they tend to have more active MRI scans. This study is the first to monitor both MWF and WC in the evolution of newly enhancing MS lesions, thereby allowing us to comment on the time course of probable myelin changes in the natural history of the lesion. The myelin water fraction (MWF) is that fraction of the entire T2 water distribution accounted for by short T2 relaxation times (less than 40 ms) and its value may be reduced either by loss of myelin water content (presumably from demyelination) or an increase in the total water. MWF could potentially be affected by edema from inflammation that causes separation of myelinated axons, an increase in extra-cellular water, and a dilution effect on the MWF resulting in a lower MWF without necessarily any accompanying loss of myelin. It is therefore necessary to determine the WC at the same time as the MWF. A recent study dealing with peripheral nerve, suggested that fragments of myelin also contribute to the myelin water signal [15]. This myelin debris have been shown to remain in place locally for weeks to months before removal [16]. Also, if the myelin sheath were to become leaky such that water in the sheath exchanged with intra- and extra-cellular water in a few 10s of ms or less, then the myelin water signal would be indistinguishable from the intra/extra-cellular water signal. We have so far seen no evidence supporting the latter interpretation.
The three lesions examined show different evolution of MWF with time. Lesion A showed a dark central core on the myelin map indicating evidence of severe demyelination present at the lesion's first appearance. This dark central core was slightly smaller than the bright central core observed on the proton density images but similar in size to the area of Gad enhancement. Using a brain water model developed by Laule et al. [9], the decrease in MWF at the core of the lesion could not be entirely accounted for by dilution resulting from the increase in WC and was highly suggestive of demyelination (Table 1). The periphery of lesion A showed decreases in MWF that were not as severe as those observed in the central core, and which could be wholly accounted for by dilution resulting from the observed WC increases in the periphery. This observation has been previously noted in the histopathology of active MS lesions where a focal region of myelin loss is seen surrounded by inflammatory cells, mainly macrophages [1] and edema. On follow-up, the myelin water dark central core decreased in size. This is consistent with a newly acquired myelin water signal, e.g. remyelination and resolution of edema, on the time scale of two months or less. The periphery of the lesion was no longer bright on PD, presumably due to resolution of inflammation and edema. At first appearance, lesion B showed a decrease in MWF which fully recovered at later months. While this could be interpreted as remyelination, using the brain water model [9] to explain potential changes on MWF from changes in total water, the observed decrease in MWF for lesion B can be fully accounted for by the increase in WC, indicating that mostly inflammation/edema occurred at lesion formation which resolved at later months (Table 1).
Table 1 Measured and predicted myelin water fraction (assuming only dilution from increased WC) for the three large MS lesions using the brain water model [9] Lesion
B C-periphery C-core A-periphery A-core a
Month of 1st appearance
Pre-lesion WC
MWF
WC
MWF
Predicted (dilution)
WC
MWF
Predicted (dilution)
6 6 6 2 2
0.80 0.76 0.76 0.76 0.76
0.06 0.08 0.08 0.07 0.07
0.89 0.84 0.84 0.84 0.89
0.05 0.06 0.06 0.06 0.02a
0.03 0.05 0.05 0.04 0.03
0.85 0.76 0.80 0.76 0.81
0.06 0.06 0.05a 0.07 0.02a
0.04 0.06 0.06 0.07 0.05
Lesion 1st appearance
Post-lesion
MWF is lower than predicted value from simple dilution and therefore is indicative of demyelination.
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
Lesion C exhibited an overall decrease in MWF at first appearance, but unlike lesion B, lesion C recovered only in the periphery with the centre showing further decreases in MWF (Table 1). Applying the brain water model [9], the overall decrease in MWF at first appearance can be accounted for by the increase in WC and edema. However, upon 6 month follow-up, the MWF at the centre of lesion C could no longer be explained by edema alone, suggesting the persistence of demyelination. In previous longitudinal studies of MS subjects, both magnetization transfer imaging [17] and magnetic resonance spectroscopy [18] found changes in NAWM before the appearance of a new lesion. The magnetization transfer ratio (MTR) showed a reduction in areas of NAWM which subsequently developed lesions compared to NAWM which did not. After the new lesion appearance, MTR was found to decrease further in some lesions, remain constant in others or return to near pre-lesion values [17]. In acute lesions, a transient reduction in N-acetyl aspartate concentration was observed at lesion first appearance. Also in acute lesions, the concentration of choline increased either at the time of lesion appearance or in NAWM which developed into a new lesion. In three out of four areas of NAWM which displayed strong lipid peaks, new lesions developed [18]. No pre-lesion changes were observed with MWF and WC in the three lesions investigated in this paper. Furthermore, the increase in choline and lipids within new lesions also supports our observation of demyelination at the time of lesion appearance. 5. Concluding remarks This was the first study to follow the evolution of water content and myelin water fraction in new MS lesions. Three new lesions that appear similar on T2-weighted imaging demonstrated a spectrum of findings, including changes in WC and MWF, identifying edema and/or demyelination at lesion appearance and on follow-up, recovery from edema, edema progressing to demyelination and recovery from demyelination possibly due to remyelination. The results indicate that the determination of MWF and WC is a very useful means of following edema, demyelination and remyelination in the evolution of the visible MS lesion. Identifying subgroups of lesions based on their ability to remyelinate may help segregate MS patients for clinical trials and evaluate therapies for specific studies aimed at improving recovery from CNS injury. Acknowledgements The authors wish to thank the MS subjects and the technologists at UBC Hospital. We thank Dr. Donald Paty for his help and advice in
53
earlier versions of this manuscript. We would like to acknowledge the support of the Multiple Sclerosis Society of Canada. References [1] Moore GRW. Neuropathology and pathophysiology of the multiple sclerosis lesion. In: Paty D, Ebers GC, editors. Multiple sclerosis. Philadelphia: F.A. Davis; 1998. [2] Willoughby EW, Grochowski E, Li DK, Oger J, Kastrukoff LF, Paty DW. Serial magnetic resonance scanning in multiple sclerosis: a second prospective study in relapsing patients. Ann Neurol 1989;25(1):43–9. [3] MacKay A, Whittall K, Adler J, Li D, Paty D, Graeb D. In vivo visualization of myelin water in brain by magnetic resonance. Magn Reson Med 1994;31(6):673–7. [4] Whittall KP, MacKay AL, Graeb DA, Nugent RA, Li DK, Paty DW. In vivo measurement of T2 distributions and water contents in normal human brain. Magn Reson Med 1997;37(1):34–43. [5] Menon RS, Allen PS. Application of continuous relaxation time distributions to the fitting of data from model systems and excised tissue. Magn Reson Med 1991;20 (2):214–27. [6] Vasilescu V, Katona E, Simplaceanu V, Demco D. Water compartments in the myelinated nerve. III. Pulsed NMR results. Experientia 1978;34(11):1443–4. [7] Vavasour IM, Whittall KP, Li DK, MacKay AL. Different magnetization transfer effects exhibited by the short and long T(2) components in human brain. Magn Reson Med 2000;44(6):860–6. [8] Laule C, Leung E, Lis DK, Traboulsee AL, Paty DW, MacKay AL, et al. Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology. Mult Scler 2006;12(6):747–53. [9] Laule C, Vavasour IM, Moore GRW, Oger J, Li DKB, Paty DW, et al. Water content and myelin water fraction in multiple sclerosis: a T2 relaxation study. J Neurol 2004;251(3):284–93. [10] Vavasour IM, Whittall KP, MacKay AL, Li DK, Vorobeychik G, Paty DW. A comparison between magnetization transfer ratios and myelin water percentages in normals and multiple sclerosis patients. Magn Reson Med 1998;40(5):763–8. [11] Tozer DJ, Davies GR, Altmann DR, Miller DH, Tofts PS. Correlation of apparent myelin measures obtained in multiple sclerosis patients and controls from magnetization transfer and multicompartmental T2 analysis. Magn Reson Med 2006;53(6):1415–22. [12] Oh J, Han ET, Lee MC, Nelson SJ, Pelletier D. Multislice brain myelin water fractions at 3T in multiple sclerosis. J Neuroimaging 2007;17(2):156–63. [13] Poser CM, Paty DW, Scheinberg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983;13(3):227–31. [14] Whittall KP, MacKay AL. Quantitative interpretation of NMR relaxation data. J Magn Reson 1989;84:64–71. [15] Webb S, Munro CA, Midha R, Stanisz GJ. Is multicomponent T2 a good measure of myelin content in peripheral nerve? Magn Reson Med 2003;49(4):638–45. [16] Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis.[see comment]. N Engl J Med 1998;338(5):278–85. [17] Filippi M, Rocca MA, Martino G, Horsfield MA, Comi G. Magnetization transfer changes in the normal appearing white matter precede the appearance of enhancing lesions in patients with multiple sclerosis. Ann Neurol 1998;43 (6):809–14. [18] Narayana PA, Doyle TJ, Lai D, Wolinsky JS. Serial proton magnetic resonance spectroscopic imaging, contrast-enhanced magnetic resonance imaging, and quantitative lesion volumetry in multiple sclerosis. Ann Neurol 1998;43(1):56–71.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Longitudinal changes in myelin water fraction in two MS patients with active disease I.M. Vavasour a,⁎, C. Laule a, D.K.B. Li a, J. Oger b, G.R.W. Moore c, A. Traboulsee b, A.L. MacKay a,d a
Department of Radiology, Canada Department of Medicine (Neurology) and Multiple Sclerosis Clinic, Canada c Department of Pathology and Laboratory Medicine, Canada d Department of Physics and Astronomy, University of British Columbia, Vancouver, Canada b
a r t i c l e
i n f o
Article history: Received 3 March 2008 Received in revised form 18 August 2008 Accepted 21 August 2008 Available online 25 September 2008 Keywords: MRI Multiple sclerosis Water content Myelin water fraction New lesions Longitudinal
a b s t r a c t Multiple sclerosis (MS) is characterised by focal areas that undergo cycles of demyelination and remyelination. Although conventional magnetic resonance imaging is very effective in localising areas of damage, these techniques are not pathology specific. A newer technique, T2 relaxation, can separate water from brain into three compartments: (1) a long T2 component (N2 s) arising from CSF, (2) an intermediate T2 component (~ 80 ms) attributed to intra- and extra-cellular water and (3) a short T2 component (~ 20 ms) assigned to water trapped in between the myelin bilayers (termed myelin water). Histological evidence shows that myelin water is a specific marker of myelination. The goal of this work was to follow changes in total water content (WC) and myelin water fraction (MWF) in evolving MS lesions over one year. Multi-echo T2 relaxation data was collected and used to measure water content and myelin water fraction from three new MS lesions in two patients. WC increased in the three large (N1 cm3) lesions at lesion appearance and remained elevated in the central core. Two lesions showed low MWF in the core suggesting demyelination upon first appearance. At later time points, one lesion showed a decrease in volume of low MWF, reflecting remyelination whereas the volume of low MWF in the other lesion core remained constant. This work provides evidence that MWF and WC can monitor demyelination and remyelination in MS. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Multiple sclerosis (MS) is an autoimmune disease of the central nervous system that is characterised by focal areas of varying degrees of inflammation and demyelination. Individual lesions, while appearing indistinguishable on conventional T2 weighted images, demonstrate considerable heterogeneity on histopathologic evaluation with varying degrees of demyelination, axonal loss, inflammation and gliosis. Histological studies show evidence that cycles of demyelination and remyelination occurs in lesions [1], however, the timescales are unknown since pathological studies provide only one snapshot of the state of a lesion. Magnetic resonance imaging (MRI), which allows for in vivo measurement of lesions over time, is helpful to investigate chronological changes in MS brain [2]. Visualising demyelination and remyelination over time would add
⁎ Corresponding author. MRI Research Group, Rm M10, Purdy Pavilion, UBC Hospital, 2221 Wesbrook Mall, Vancouver, BC, Canada, V6T 2B5. Tel.: +1 604 822 0357; fax: +1 604 827 3339. E-mail address: [email protected] (I.M. Vavasour). 0022-510X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2008.08.022
depth to the contribution of MRI to measuring the evolution of MS lesions. T2 relaxation is an MRI technique that enables separation of the tissue water signal from brain into three components: (1) a long T2 component (N2 s) arising from CSF, (2) an intermediate T2 component (~ 80 ms) attributed to intra- and extra-cellular water and (3) a short T2 component (~ 20 ms) assigned to water trapped in between the myelin bilayers [3–6]. The sum of the signals from all three components is proportional to the total water content (WC) of the tissue and the fraction of signal from the myelin water component is termed the myelin water fraction (MWF). In order to determine quantitative values for WC and MWF, acquisition of T2 relaxation was done on a single-slice otherwise off-resonance effects from other slices can affect the signal from the different water components [7]. Previous work has shown strong quantitative correlations between luxol fast blue histological staining for myelin and MWF, providing compelling evidence that the short T2 component is a marker of myelin [8]. Prior work in vivo found the short T2 component was variably decreased in MS lesions [3,9–11], and diffusely reduced in the normal appearing white matter (NAWM) when compared to healthy controls [10,12]. The goal of this work was to study the longitudinal changes in total water content (WC) and myelin water fraction (MWF), obtained from a single imaging slice, in newly appearing MS lesions. The fortuitous appearance of lesions on the slice of interest allowed observation of
50
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
WC and MWF before, during and after the formation of these new lesions.
lesions and were followed in this paper. Both patients had clinically definite MS according to the Poser criteria [13] and the presence of at least one enhancing lesion on a screening scan performed 4 weeks prior to the start of the study. They were two women aged 36 and 50 years with disease duration of 9 and 10 years, respectively. They both had relapsing MS and had been followed yearly in the UBC MS clinic since diagnosis. They had an EDSS of 2.5 and 3.0 at inclusion in the study. Both had declined disease modifying drugs. Patient 2 had received IV solumedrol 4 months prior to inclusion. They were followed for one year with scans at month 0, 2, 4, 6 and 12. Both had a clinical relapse between month 6 and month 12 scans. Patient 2 received IV methylprednisolone for this relapse 2 months before the final scan. Newly enhancing lesions that were greater than 1 cm3 and had at least one follow-up scan were studied. Patient 1 had two lesions which qualified (lesions A and B) and patient 2 had one lesion (lesion C). A diagram indicating the time of scans, lesion appearance and clinical relapses for both patients is shown in Fig. 1.
2. Materials and methods
2.2. MRI studies
2.1. Patient selection
Images were acquired using a 1.5 Tesla MR scanner (GE Medical Systems, Milwaukee, USA, version 5.7). First, proton density (PD) and T2-weighted scans (TR 2500 ms, TE 30/90 ms, 22 contiguous slices, matrix 256 × 192) were obtained using a dual echo conventional spin echo sequence. T2 relaxation was then measured using a single-slice
Fig. 1. Diagram of the timeline for scanning, lesion appearance and clinical relapses in both patients. Below the dashed line is data for patient 1 and above the dashed line is data for patient 2. Arrows indicate the time of scans, letters indicate the time of first appearance of new lesions A, B and C, and an X indicates the time of a clinical relapse with a subscript s indicating treatment with steroids.
Seven subjects with multiple sclerosis volunteered for this study and signed an informed written consent approved by the Clinical Research Ethics Board of our institution. Of these subjects, two had large (N1 cm3)
Fig. 2. Images for large lesion A at pre-lesion (1st row), lesion first appearance (2nd row), 2 months post-lesion (3rd row), 4 months post-lesion (4th row) and 10 months post-lesion (5th row). Proton-density (left column), T1-weighted post-Gd-DTPA (middle column) and myelin map (right column) images are shown. The box indicates the location of the lesion of interest.
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
32-echo relaxation sequence (TE 10 ms, TR 3000 ms, 4 averages, matrix 256 × 128, bandwidth 32 kHz) [3]. Next, T1 relaxation was measured using a single-slice fast gradient echo inversion recovery prepared sequence (TE 8 ms, 1 NEX, 15 TIs ranging from 0.05–2.5 s). The slice of interest for the T2 and T1 relaxation measurements was chosen to include the enhancing lesion observed on the screening scan. Finally, a post-contrast T1-weighted spin echo scan (TR 550 ms, TE 8 ms, 22 slices, matrix 256 × 192) was collected 5 min after the injection of gadolinium-DTPA (0.2 mL/kg). The field of view for all exams was 22 cm and the slice thickness was 5 mm. Exact patient repositioning was attempted by first performing a true midline sagittal localiser and then manually positioning the patient so that the angle of a line drawn through the base of the genu and splenium of the corpus callosum was within one degree of the baseline scan. Four NiCl–agarose water standards were placed within the T2 relaxation slice. 2.3. Calculation of MR measurements In-house software was created for registration of the single-slice T2 and T1 relaxation data whereby the matching PD slice and the first image from the relaxation data were enlarged using bilinear interpolation to 4 times original size, the mean sum of squares difference between the images was taken and minimised by single pixel shifts to the right or left, up or down and also by fractional rotations. By enlarging the images, single pixel shifts translated into fractional pixel shifts on the original image size, thereby increasing the accuracy of the registration. Regions of interest (ROIs) were drawn on the first echo of the 32-echo experiment around the core and periphery of large (N1 cm) hyperintense lesions as well as in adjacent NAWM and mapped onto registered images from all months. For each ROI, the T1, MWF and WC were calculated. The T1 relaxation data was fit to a mono-exponential function. T2 relaxation distributions were calculated from the 32-echo experiment using a regularised non-negative least-squares (NNLS) algorithm [14]. The total WC was defined as the total area under the T2 distribution normalised to the temperature corrected water standards and corrected for T1 relaxation. Myelin water maps were generated by
51
displaying the MWF (area from 10–40 ms divided by the entire area of the T2 distribution) at each pixel in the image [3]. 3. Results Seven patients were recruited with MRI active disease as seen by the presence of a gadolinium enhancing lesion. These lesions determined the slice of interest. By happenstance, the three lesions in the two subjects examined in this study appeared in the same slice as the original enhancing lesion and allowed for water content and myelin water measurements before and after the lesions appeared. Clinically, the two patients were also active with patient 1 having a relapse at month 8 (affecting the spinal cord) and patient 2 having a relapse 2 months prior to screening and another two at months 7 and 8 (both clinically localised to the brainstem) into the study (Fig. 1). All relapses in patient two were treated with IV steroids. Figs. 2–4 show the evolution for each of the lesions on PD (related to WC), Gadolinium T1 and myelin water images. An increase in water content compared to baseline was seen in all lesions studied. Lesion A (Fig. 2) showed a virtual absence of myelin water in the very centre of the lesion at first appearance with the periphery showing a reduction in MWF. At later months, the MWF of the periphery recovered to normal values whereas the core still had low MWF although its size had decreased (Table 1). Later still, a wide area of decreased MWF, related to the appearance of a new enhancing lesion adjacent to lesion A, filled the region. Lesion B (Fig. 3) showed a slight decrease in MWF over the entire lesion area at first appearance which recovered within 6 months (Table 1). Lesion C (Fig. 4) also showed a relatively homogeneous decrease in MWF over the entire lesion area at first appearance. However, at 6 month follow-up, the MWF at the core of the lesion decreased further whereas the periphery recovered to prelesion values (Table 1). 4. Discussion The two patients serially studied here had active disease but nothing that departed from the usual course of MS: i.e. more new MRI
Fig. 3. Images for large lesion B at pre-lesion (top row), lesion first appearance (middle row) and 6 months post-lesion (bottom row). Proton-density (left column), T1-weighted postGd-DTPA (middle column) and myelin map (right column) images are shown. The box indicates the location of the lesion of interest.
52
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
Fig. 4. Images for large lesion C at pre-lesion (top row), lesion first appearance (middle row) and 6 months post-lesion (bottom row). Proton-density (left column), T1-weighted postGd-DTPA and myelin map (right column) image are shown. The box indicates the location of the lesion of interest.
lesions than relapses and a relatively poor correlation between clinical relapses and MRI activity. They fit, however, with the loose relationship between clinical and MRI events by which younger patients with one relapse per year tend to have more enhancing lesions than older patients with less active disease. It is also generally accepted that, when patients are screened for the presence of gadolinium enhancement, they tend to have more active MRI scans. This study is the first to monitor both MWF and WC in the evolution of newly enhancing MS lesions, thereby allowing us to comment on the time course of probable myelin changes in the natural history of the lesion. The myelin water fraction (MWF) is that fraction of the entire T2 water distribution accounted for by short T2 relaxation times (less than 40 ms) and its value may be reduced either by loss of myelin water content (presumably from demyelination) or an increase in the total water. MWF could potentially be affected by edema from inflammation that causes separation of myelinated axons, an increase in extra-cellular water, and a dilution effect on the MWF resulting in a lower MWF without necessarily any accompanying loss of myelin. It is therefore necessary to determine the WC at the same time as the MWF. A recent study dealing with peripheral nerve, suggested that fragments of myelin also contribute to the myelin water signal [15]. This myelin debris have been shown to remain in place locally for weeks to months before removal [16]. Also, if the myelin sheath were to become leaky such that water in the sheath exchanged with intra- and extra-cellular water in a few 10s of ms or less, then the myelin water signal would be indistinguishable from the intra/extra-cellular water signal. We have so far seen no evidence supporting the latter interpretation.
The three lesions examined show different evolution of MWF with time. Lesion A showed a dark central core on the myelin map indicating evidence of severe demyelination present at the lesion's first appearance. This dark central core was slightly smaller than the bright central core observed on the proton density images but similar in size to the area of Gad enhancement. Using a brain water model developed by Laule et al. [9], the decrease in MWF at the core of the lesion could not be entirely accounted for by dilution resulting from the increase in WC and was highly suggestive of demyelination (Table 1). The periphery of lesion A showed decreases in MWF that were not as severe as those observed in the central core, and which could be wholly accounted for by dilution resulting from the observed WC increases in the periphery. This observation has been previously noted in the histopathology of active MS lesions where a focal region of myelin loss is seen surrounded by inflammatory cells, mainly macrophages [1] and edema. On follow-up, the myelin water dark central core decreased in size. This is consistent with a newly acquired myelin water signal, e.g. remyelination and resolution of edema, on the time scale of two months or less. The periphery of the lesion was no longer bright on PD, presumably due to resolution of inflammation and edema. At first appearance, lesion B showed a decrease in MWF which fully recovered at later months. While this could be interpreted as remyelination, using the brain water model [9] to explain potential changes on MWF from changes in total water, the observed decrease in MWF for lesion B can be fully accounted for by the increase in WC, indicating that mostly inflammation/edema occurred at lesion formation which resolved at later months (Table 1).
Table 1 Measured and predicted myelin water fraction (assuming only dilution from increased WC) for the three large MS lesions using the brain water model [9] Lesion
B C-periphery C-core A-periphery A-core a
Month of 1st appearance
Pre-lesion WC
MWF
WC
MWF
Predicted (dilution)
WC
MWF
Predicted (dilution)
6 6 6 2 2
0.80 0.76 0.76 0.76 0.76
0.06 0.08 0.08 0.07 0.07
0.89 0.84 0.84 0.84 0.89
0.05 0.06 0.06 0.06 0.02a
0.03 0.05 0.05 0.04 0.03
0.85 0.76 0.80 0.76 0.81
0.06 0.06 0.05a 0.07 0.02a
0.04 0.06 0.06 0.07 0.05
Lesion 1st appearance
Post-lesion
MWF is lower than predicted value from simple dilution and therefore is indicative of demyelination.
I.M. Vavasour et al. / Journal of the Neurological Sciences 276 (2009) 49–53
Lesion C exhibited an overall decrease in MWF at first appearance, but unlike lesion B, lesion C recovered only in the periphery with the centre showing further decreases in MWF (Table 1). Applying the brain water model [9], the overall decrease in MWF at first appearance can be accounted for by the increase in WC and edema. However, upon 6 month follow-up, the MWF at the centre of lesion C could no longer be explained by edema alone, suggesting the persistence of demyelination. In previous longitudinal studies of MS subjects, both magnetization transfer imaging [17] and magnetic resonance spectroscopy [18] found changes in NAWM before the appearance of a new lesion. The magnetization transfer ratio (MTR) showed a reduction in areas of NAWM which subsequently developed lesions compared to NAWM which did not. After the new lesion appearance, MTR was found to decrease further in some lesions, remain constant in others or return to near pre-lesion values [17]. In acute lesions, a transient reduction in N-acetyl aspartate concentration was observed at lesion first appearance. Also in acute lesions, the concentration of choline increased either at the time of lesion appearance or in NAWM which developed into a new lesion. In three out of four areas of NAWM which displayed strong lipid peaks, new lesions developed [18]. No pre-lesion changes were observed with MWF and WC in the three lesions investigated in this paper. Furthermore, the increase in choline and lipids within new lesions also supports our observation of demyelination at the time of lesion appearance. 5. Concluding remarks This was the first study to follow the evolution of water content and myelin water fraction in new MS lesions. Three new lesions that appear similar on T2-weighted imaging demonstrated a spectrum of findings, including changes in WC and MWF, identifying edema and/or demyelination at lesion appearance and on follow-up, recovery from edema, edema progressing to demyelination and recovery from demyelination possibly due to remyelination. The results indicate that the determination of MWF and WC is a very useful means of following edema, demyelination and remyelination in the evolution of the visible MS lesion. Identifying subgroups of lesions based on their ability to remyelinate may help segregate MS patients for clinical trials and evaluate therapies for specific studies aimed at improving recovery from CNS injury. Acknowledgements The authors wish to thank the MS subjects and the technologists at UBC Hospital. We thank Dr. Donald Paty for his help and advice in
53
earlier versions of this manuscript. We would like to acknowledge the support of the Multiple Sclerosis Society of Canada. References [1] Moore GRW. Neuropathology and pathophysiology of the multiple sclerosis lesion. In: Paty D, Ebers GC, editors. Multiple sclerosis. Philadelphia: F.A. Davis; 1998. [2] Willoughby EW, Grochowski E, Li DK, Oger J, Kastrukoff LF, Paty DW. Serial magnetic resonance scanning in multiple sclerosis: a second prospective study in relapsing patients. Ann Neurol 1989;25(1):43–9. [3] MacKay A, Whittall K, Adler J, Li D, Paty D, Graeb D. In vivo visualization of myelin water in brain by magnetic resonance. Magn Reson Med 1994;31(6):673–7. [4] Whittall KP, MacKay AL, Graeb DA, Nugent RA, Li DK, Paty DW. In vivo measurement of T2 distributions and water contents in normal human brain. Magn Reson Med 1997;37(1):34–43. [5] Menon RS, Allen PS. Application of continuous relaxation time distributions to the fitting of data from model systems and excised tissue. Magn Reson Med 1991;20 (2):214–27. [6] Vasilescu V, Katona E, Simplaceanu V, Demco D. Water compartments in the myelinated nerve. III. Pulsed NMR results. Experientia 1978;34(11):1443–4. [7] Vavasour IM, Whittall KP, Li DK, MacKay AL. Different magnetization transfer effects exhibited by the short and long T(2) components in human brain. Magn Reson Med 2000;44(6):860–6. [8] Laule C, Leung E, Lis DK, Traboulsee AL, Paty DW, MacKay AL, et al. Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology. Mult Scler 2006;12(6):747–53. [9] Laule C, Vavasour IM, Moore GRW, Oger J, Li DKB, Paty DW, et al. Water content and myelin water fraction in multiple sclerosis: a T2 relaxation study. J Neurol 2004;251(3):284–93. [10] Vavasour IM, Whittall KP, MacKay AL, Li DK, Vorobeychik G, Paty DW. A comparison between magnetization transfer ratios and myelin water percentages in normals and multiple sclerosis patients. Magn Reson Med 1998;40(5):763–8. [11] Tozer DJ, Davies GR, Altmann DR, Miller DH, Tofts PS. Correlation of apparent myelin measures obtained in multiple sclerosis patients and controls from magnetization transfer and multicompartmental T2 analysis. Magn Reson Med 2006;53(6):1415–22. [12] Oh J, Han ET, Lee MC, Nelson SJ, Pelletier D. Multislice brain myelin water fractions at 3T in multiple sclerosis. J Neuroimaging 2007;17(2):156–63. [13] Poser CM, Paty DW, Scheinberg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983;13(3):227–31. [14] Whittall KP, MacKay AL. Quantitative interpretation of NMR relaxation data. J Magn Reson 1989;84:64–71. [15] Webb S, Munro CA, Midha R, Stanisz GJ. Is multicomponent T2 a good measure of myelin content in peripheral nerve? Magn Reson Med 2003;49(4):638–45. [16] Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis.[see comment]. N Engl J Med 1998;338(5):278–85. [17] Filippi M, Rocca MA, Martino G, Horsfield MA, Comi G. Magnetization transfer changes in the normal appearing white matter precede the appearance of enhancing lesions in patients with multiple sclerosis. Ann Neurol 1998;43 (6):809–14. [18] Narayana PA, Doyle TJ, Lai D, Wolinsky JS. Serial proton magnetic resonance spectroscopic imaging, contrast-enhanced magnetic resonance imaging, and quantitative lesion volumetry in multiple sclerosis. Ann Neurol 1998;43(1):56–71.