Associations Between Findings From Myelin Water Imaging and Cognitive Performance Among Individuals With Multiple Sclerosis | Demyelinating Disorders | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Correlations Between Symbol Digit Modalities Test (SDMT) Performance and Myelin Heterogeneity Index
Correlations Between Symbol Digit Modalities Test (SDMT) Performance and Myelin Heterogeneity Index

Correlations between the myelin heterogeneity index in normal-appearing white matter and SDMT scores are shown for participants with multiple sclerosis (MS) and controls in 3 regions of interest. Lines denote lines of best fit.

Figure 2.  Correlations Between Selective Reminding Test (SRT) Performance and Myelin Heterogeneity Index
Correlations Between Selective Reminding Test (SRT) Performance and Myelin Heterogeneity Index

Correlations between the myelin heterogeneity index in normal-appearing white matter and SRT scores are shown for participants with multiple sclerosis (MS) and controls in 3 regions of interest. Lines denote lines of best fit.

Figure 3.  Correlations Between Controlled Oral Word Association Test (COWAT) Performance and Myelin Heterogeneity Index
Correlations Between Controlled Oral Word Association Test (COWAT) Performance and Myelin Heterogeneity Index

Correlations between the myelin heterogeneity index in normal-appearing white matter and COWAT scores are shown for participants with multiple sclerosis (MS) and controls in 3 regions of interest. Lines denote lines of best fit.

Figure 4.  Axial Map of Myelin Water Fraction (MWF) Values, MWF Distributions in Superior Longitudinal Fasciculus (SLF), Myelin Heterogeneity Index (MHI) in SLF, and Cognitive z Scores in 3 Participants with Multiple Sclerosis
Axial Map of Myelin Water Fraction (MWF) Values, MWF Distributions in Superior Longitudinal Fasciculus (SLF), Myelin Heterogeneity Index (MHI) in SLF, and Cognitive z Scores in 3 Participants with Multiple Sclerosis

Axial maps of MWF values (top), normalized histograms of MWF values in the SLF (middle), MHI in the SLF and cognitive z scores (bottom) of 3 participants with multiple sclerosis. Patient A had a high MHI in the SLF (0.59), matching their low cognitive scores compared with controls (range of z scores, −5.1 to −3.6). Patient B had both moderate MHI in the SLF (0.28) and cognitive test scores (range of z scores, −1.6 to −0.9). Patient C had a low MHI in the SLF (0.22) and performed at and above the level of controls on the cognitive tests range of z scores (−0.08 to 0.9). The z scores were calculated using the mean and SD for each cognitive test from the control sample. ROI indicates region of interest.

Table.  Clinical and Demographic Characteristics
Clinical and Demographic Characteristics
1.
Kawachi  I, Lassmann  H.  Neurodegeneration in multiple sclerosis and neuromyelitis optica.   J Neurol Neurosurg Psychiatry. 2017;88(2):137-145. doi:10.1136/jnnp-2016-313300PubMedGoogle ScholarCrossref
2.
Wallin  MT, Culpepper  WJ, Nichols  E, Lancet  ZBT; GBD 2016 Multiple Sclerosis Collaborators.  Global, regional, and national burden of multiple sclerosis 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016.   Lancet Neurol. 2019;18(3):269-285. doi:10.1016/S1474-4422(18)30443-5PubMedGoogle ScholarCrossref
3.
Chiaravalloti  ND, DeLuca  J.  Cognitive impairment in multiple sclerosis.   Lancet Neurol. 2008;7(12):1139-1151. doi:10.1016/S1474-4422(08)70259-XPubMedGoogle ScholarCrossref
4.
Costa  SL, Genova  HM, DeLuca  J, Chiaravalloti  ND.  Information processing speed in multiple sclerosis: past, present, and future.   Mult Scler. 2017;23(6):772-789. doi:10.1177/1352458516645869PubMedGoogle ScholarCrossref
5.
Ruet  A, Deloire  M, Hamel  D, Ouallet  J-C, Petry  K, Brochet  B.  Cognitive impairment, health-related quality of life and vocational status at early stages of multiple sclerosis: a 7-year longitudinal study.   J Neurol. 2013;260(3):776-784. doi:10.1007/s00415-012-6705-1PubMedGoogle ScholarCrossref
6.
Rao  SM, Leo  GJ, Ellington  L, Nauertz  T, Bernardin  L, Unverzagt  F.  Cognitive dysfunction in multiple sclerosis. II. impact on employment and social functioning.   Neurology. 1991;41(5):692-696. doi:10.1212/WNL.41.5.692PubMedGoogle ScholarCrossref
7.
Morrow  SA, Classen  S, Monahan  M,  et al.  On-road assessment of fitness-to-drive in persons with MS with cognitive impairment: a prospective study.   Mult Scler. 2018;24(11):1499-1506. doi:10.1177/1352458517723991PubMedGoogle ScholarCrossref
8.
Strober  L, Chiaravalloti  N, Moore  N, DeLuca  J.  Unemployment in multiple sclerosis (MS): utility of the MS Functional Composite and cognitive testing.   Mult Scler. 2014;20(1):112-115. doi:10.1177/1352458513488235PubMedGoogle ScholarCrossref
9.
Reich  DS, Lucchinetti  CF, Calabresi  PA.  Multiple sclerosis.   N Engl J Med. 2018;378(2):169-180. doi:10.1056/NEJMra1401483PubMedGoogle ScholarCrossref
10.
Popescu  BFG, Pirko  I, Lucchinetti  CF.  Pathology of multiple sclerosis: where do we stand?   Continuum (Minneap Minn). 2013;19(4):901-921. doi:10.1212/01.CON.0000433291.23091.65.Google Scholar
11.
Thompson  AJ, Banwell  BL, Barkhof  F,  et al.  Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria.   Lancet Neurol. 2018;17(2):162-173. doi:10.1016/S1474-4422(17)30470-2PubMedGoogle ScholarCrossref
12.
Barkhof  F.  The clinico-radiological paradox in multiple sclerosis revisited.   Curr Opin Neurol. 2002;15(3):239-245. doi:10.1097/00019052-200206000-00003PubMedGoogle ScholarCrossref
13.
Mollison  D, Sellar  R, Bastin  M,  et al.  The clinico-radiological paradox of cognitive function and MRI burden of white matter lesions in people with multiple sclerosis: a systematic review and meta-analysis.   PLoS One. 2017;12(5):e0177727. doi:10.1371/journal.pone.0177727PubMedGoogle Scholar
14.
Vavasour  IM, Huijskens  SC, Li  DK,  et al.  Global loss of myelin water over 5 years in multiple sclerosis normal-appearing white matter.   Mult Scler. 2018;24(12):1557-1568. doi:10.1177/1352458517723717PubMedGoogle ScholarCrossref
15.
Jäkel  S, Agirre  E, Mendanha Falcão  A,  et al.  Altered human oligodendrocyte heterogeneity in multiple sclerosis.   Nature. 2019;566(7745):543-547. doi:10.1038/s41586-019-0903-2PubMedGoogle ScholarCrossref
16.
Yeung  MSY, Djelloul  M, Steiner  E,  et al.  Dynamics of oligodendrocyte generation in multiple sclerosis.   Nature. 2019;566(7745):538-542. doi:10.1038/s41586-018-0842-3PubMedGoogle ScholarCrossref
17.
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):673-677. doi:10.1002/mrm.1910310614PubMedGoogle ScholarCrossref
18.
Laule  C, Leung  E, Lis  DKB,  et al.  Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology.   Mult Scler. 2006;12(6):747-753. doi:10.1177/1352458506070928PubMedGoogle ScholarCrossref
19.
Laule  C, Kozlowski  P, Leung  E, Li  DKB, Mackay  AL, Moore  GRW.  Myelin water imaging of multiple sclerosis at 7 T: correlations with histopathology.   Neuroimage. 2008;40(4):1575-1580. doi:10.1016/j.neuroimage.2007.12.008PubMedGoogle ScholarCrossref
20.
McCreary  CR, Bjarnason  TA, Skihar  V, Mitchell  JR, Yong  VW, Dunn  JF.  Multiexponential T2 and magnetization transfer MRI of demyelination and remyelination in murine spinal cord.   Neuroimage. 2009;45(4):1173-1182. doi:10.1016/j.neuroimage.2008.12.071PubMedGoogle ScholarCrossref
21.
Rao  SM, Leo  GJ, Bernardin  L, Unverzagt  F.  Cognitive dysfunction in multiple sclerosis. I. frequency, patterns, and prediction.   Neurology. 1991;41(5):685-691. doi:10.1212/WNL.41.5.685PubMedGoogle ScholarCrossref
22.
Benedict  RHB, Fischer  JS, Archibald  CJ,  et al.  Minimal neuropsychological assessment of MS patients: a consensus approach.   Clin Neuropsychol. 2002;16(3):381-397. doi:10.1076/clin.16.3.381.13859PubMedGoogle ScholarCrossref
23.
Zhang  X, Zhang  F, Huang  D,  et al.  Contribution of gray and white matter abnormalities to cognitive impairment in multiple sclerosis.   Int J Mol Sci. 2016;18(1):46-13. doi:10.3390/ijms18010046PubMedGoogle ScholarCrossref
24.
Kurtzke  JF.  Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS).   Neurology. 1983;33(11):1444-1452. doi:10.1212/WNL.33.11.1444PubMedGoogle ScholarCrossref
25.
Prasloski  T, Rauscher  A, MacKay  AL,  et al.  Rapid whole cerebrum myelin water imaging using a 3D GRASE sequence.   Neuroimage. 2012;63(1):533-539. doi:10.1016/j.neuroimage.2012.06.064PubMedGoogle ScholarCrossref
26.
Prasloski  T, Mädler  B, Xiang  Q-S, MacKay  A, Jones  C.  Applications of stimulated echo correction to multicomponent T2 analysis.   Magn Reson Med. 2012;67(6):1803-1814. doi:10.1002/mrm.23157PubMedGoogle ScholarCrossref
27.
Jenkinson  M, Bannister  P, Brady  M, Smith  S.  Improved optimization for the robust and accurate linear registration and motion correction of brain images.   Neuroimage. 2002;17(2):825-841. doi:10.1006/nimg.2002.1132PubMedGoogle ScholarCrossref
28.
Smith  SM, Jenkinson  M, Woolrich  MW,  et al.  Advances in functional and structural MR image analysis and implementation as FSL.   Neuroimage. 2004;23(1)(suppl):S208-S219. doi:10.1016/j.neuroimage.2004.07.051PubMedGoogle ScholarCrossref
29.
Smith  SM.  Fast robust automated brain extraction.   Hum Brain Mapp. 2002;17(3):143-155. doi:10.1002/hbm.10062PubMedGoogle ScholarCrossref
30.
Zhang  Y, Brady  M, Smith  S.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm.   IEEE Trans Med Imaging. 2001;20(1):45-57. doi:10.1109/42.906424PubMedGoogle ScholarCrossref
31.
Mori  S, Wakana  S, van Zijl  PCM, Nagae-Poetscher  LM.  MRI Atlas of Human White Matter. Elsevier Science. 1st ed. 2005.
32.
Kolind  S, Matthews  L, Johansen-Berg  H,  et al.  Myelin water imaging reflects clinical variability in multiple sclerosis.   Neuroimage. 2012;60(1):263-270. doi:10.1016/j.neuroimage.2011.11.070PubMedGoogle ScholarCrossref
33.
Thomsen  C, Henriksen  O, Ring  P.  In vivo measurement of water self diffusion in the human brain by magnetic resonance imaging.   Acta Radiol. 1987;28(3):353-361. doi:10.1177/028418518702800324PubMedGoogle ScholarCrossref
34.
Inglese  M, Bester  M.  Diffusion imaging in multiple sclerosis: research and clinical implications.   NMR Biomed. 2010;23(7):865-872. doi:10.1002/nbm.1515PubMedGoogle ScholarCrossref
35.
Vavasour  IM, Laule  C, Li  DKB, Traboulsee  AL, MacKay  AL.  Is the magnetization transfer ratio a marker for myelin in multiple sclerosis?   J Magn Reson Imaging. 2011;33(3):713-718. doi:10.1002/jmri.22441PubMedGoogle ScholarCrossref
36.
Dineen  RA, Vilisaar  J, Hlinka  J,  et al.  Disconnection as a mechanism for cognitive dysfunction in multiple sclerosis.   Brain. 2009;132(1):239-249. doi:10.1093/brain/awn275PubMedGoogle ScholarCrossref
37.
Preziosa  P, Rocca  MA, Pagani  E,  et al; MAGNIMS Study Group.  Structural MRI correlates of cognitive impairment in patients with multiple sclerosis: a multicenter study.   Hum Brain Mapp. 2016;37(4):1627-1644. doi:10.1002/hbm.23125PubMedGoogle ScholarCrossref
38.
Ranjeva  J-P, Audoin  B, Au Duong  MV,  et al.  Local tissue damage assessed with statistical mapping analysis of brain magnetization transfer ratio: relationship with functional status of patients in the earliest stage of multiple sclerosis.   AJNR Am J Neuroradiol. 2005;26(1):119-127.PubMedGoogle Scholar
39.
Pokryszko-Dragan  A, Banaszek  A, Nowakowska-Kotas  M,  et al.  Diffusion tensor imaging findings in the multiple sclerosis patients and their relationships to various aspects of disability.   J Neurol Sci. 2018;391:127-133. doi:10.1016/j.jns.2018.06.007PubMedGoogle ScholarCrossref
40.
Bethune  A, Tipu  V, Sled  JG,  et al.  Diffusion tensor imaging and cognitive speed in children with multiple sclerosis.   J Neurol Sci. 2011;309(1-2):68-74. doi:10.1016/j.jns.2011.07.019PubMedGoogle ScholarCrossref
41.
Roosendaal  SD, Geurts  JJG, Vrenken  H,  et al.  Regional DTI differences in multiple sclerosis patients.   Neuroimage. 2009;44(4):1397-1403. doi:10.1016/j.neuroimage.2008.10.026PubMedGoogle ScholarCrossref
42.
Hill  RA.  Do short-term changes in white matter structure indicate learning-induced myelin plasticity?   J Neurosci. 2013;33(50):19393-19395. doi:10.1523/JNEUROSCI.4122-13.2013PubMedGoogle ScholarCrossref
43.
Beaulieu  C.  The basis of anisotropic water diffusion in the nervous system: a technical review.   NMR Biomed. 2002;15(7-8):435-455. doi:10.1002/nbm.782PubMedGoogle ScholarCrossref
44.
Jones  DK, Knösche  TR, Turner  R.  White matter integrity, fiber count, and other fallacies: the do’s and don’ts of diffusion MRI.   Neuroimage. 2013;73:239-254. doi:10.1016/j.neuroimage.2012.06.081PubMedGoogle ScholarCrossref
45.
Mädler  B, Drabycz  SA, Kolind  SH, Whittall  KP, MacKay  AL.  Is diffusion anisotropy an accurate monitor of myelination? correlation of multicomponent T2 relaxation and diffusion tensor anisotropy in human brain.   Magn Reson Imaging. 2008;26(7):874-888. doi:10.1016/j.mri.2008.01.047PubMedGoogle ScholarCrossref
46.
Groeschel  S, Hagberg  GE, Schultz  T,  et al.  Assessing white matter microstructure in brain regions with different myelin architecture using MRI.   PLoS ONE. 2016. 11(11):e0167274. doi:10.1371/journal.pone.0167274.Google Scholar
47.
MacKay  AL, Laule  C.  Magnetic resonance of myelin water: an in vivo marker for myelin.   Brain Plast. 2016;2(1):71-91. doi:10.3233/BPL-160033PubMedGoogle ScholarCrossref
48.
Kremer  D, Göttle  P, Flores-Rivera  J, Hartung  H-P, Küry  P.  Remyelination in multiple sclerosis: from concept to clinical trials.   Curr Opin Neurol. 2019;32(3):378-384. doi:10.1097/WCO.0000000000000692PubMedGoogle ScholarCrossref
49.
Meyers  SM, Vavasour  IM, Mädler  B,  et al.  Multicenter measurements of myelin water fraction and geometric mean T2: intra- and intersite reproducibility.   J Magn Reson Imaging. 2013;38(6):1445-1453. doi:10.1002/jmri.24106PubMedGoogle ScholarCrossref
50.
Lee  LE, Ljungberg  E, Shin  D,  et al.  Inter-vendor reproducibility of myelin water imaging using a 3D gradient and spin echo sequence.   Front Neurosci. 2018;12:854. doi:10.3389/fnins.2018.00854PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Neurology
    September 29, 2020

    Associations Between Findings From Myelin Water Imaging and Cognitive Performance Among Individuals With Multiple Sclerosis

    Author Affiliations
    • 1Department of Medicine (Neurology), The University of British Columbia, Vancouver, British Columbia, Canada
    • 2Department of Radiology, The University of British Columbia, Vancouver, British Columbia, Canada
    • 3Department of Pathology & Laboratory Medicine, The University of British Columbia, Vancouver, British Columbia, Canada
    • 4Department of Physics & Astronomy, The University of British Columbia, Vancouver, British Columbia, Canada
    • 5International Collaboration on Repair Discoveries, The University of British Columbia, Vancouver, British Columbia, Canada
    • 6Department of Psychiatry, The University of British Columbia, Vancouver, British Columbia, Canada
    • 7Department of Clinical Neurological Sciences, Western University, London, Ontario, Canada
    • 8Department of Neurology, Georgetown University Hospital, Washington, DC
    • 9Washington Neuropsychology Research Group LLC, Fairfax, Virginia
    • 10Department of Pediatrics, The University of British Columbia, Vancouver, British Columbia, Canada
    JAMA Netw Open. 2020;3(9):e2014220. doi:10.1001/jamanetworkopen.2020.14220
    Key Points

    Question  Is myelin damage in normal-appearing white matter associated with cognitive impairment in participants with multiple sclerosis (MS)?

    Findings  In this cross-sectional study of 73 participants with MS and 22 age-, sex-, and education-matched healthy controls, significant associations were observed in participants with MS between a quantitative neuroimaging measure of myelin and performance on cognitive tests validated for MS. No significant associations were found between myelin measures and cognitive performance in controls.

    Meaning  These findings suggest that myelin damage that is completely invisible on standard clinical images but can be measured using myelin water imaging is involved in MS-related cognitive impairment.

    Abstract

    Importance  Cognitive impairment is a debilitating symptom of multiple sclerosis (MS) that affects up to 70% of patients. An improved understanding of the underlying pathology of MS-related cognitive impairment would provide considerable benefit to patients and clinicians.

    Objective  To determine whether there is an association between myelin damage in tissue that appears completely normal on standard clinical imaging, but can be detected by myelin water imaging (MWI), with cognitive performance in MS.

    Design, Setting, and Participants  In this cross-sectional study, participants with MS and controls underwent cognitive testing and magnetic resonance imaging (MRI) from August 23, 2017, to February 20, 2019. Participants were recruited through the University of British Columbia Hospital MS clinic and via online recruitment advertisements on local health authority websites. Cognitive testing was performed in the MS clinic, and MRI was performed at the adjacent academic research neuroimaging center. Seventy-three participants with clinically definite MS fulfilling the 2017 revised McDonald criteria for diagnosis and 22 age-, sex-, and education-matched healthy volunteers without neurological disease were included in the study. Data analysis was performed from March to November 2019.

    Exposures  MWI was performed at 3 T with a 48-echo, 3-dimensional, gradient and spin-echo (GRASE) sequence. Cognitive testing was performed with assessments drawn from cognitive batteries validated for use in MS.

    Main Outcomes and Measures  The association between myelin water measures, a measurement of the T2 relaxation signal from water in the myelin bilayers providing a specific marker for myelin, and cognitive test scores was assessed using Pearson correlation. Three white matter regions of interest—the cingulum, superior longitudinal fasciculus (SLF), and corpus callosum—were selected a priori according to their known involvement in MS-related cognitive impairment.

    Results  For the 95 total participants, the mean (SD) age was 49.33 (11.44) years. The mean (SD) age was 50.2 (10.7) years for the 73 participants with MS and 46.4 (13.5) for the 22 controls. Forty-eight participants with MS (66%) and 14 controls (64%) were women. The mean (SD) years of education were 14.7 (2.2) for patients and 15.8 (2.5) years for controls. In MS, significant associations were observed between myelin water measures and scores on the Symbol Digit Modalities Test (SLF, r = −0.490; 95% CI, −0.697 to −0.284; P < .001; corpus callosum, r = −0.471; 95% CI, −0.680 to −0.262; P < .001; and cingulum, r = −0.419; 95% CI, −0.634 to −0.205; P < .001), Selective Reminding Test (SLF, r = −0.444; 95% CI, −0.660 to −0.217; P < .001; corpus callosum, r = −0.411; 95% CI, −0.630 to −0.181; P = .001; and cingulum, r = −0.361; 95% CI, −0.602 to −0.130; P = .003), and Controlled Oral Word Association Test (SLF, r = −0.317; 95% CI, −0.549 to −0.078; P = .01; and cingulum, r = −0.335; 95% CI, −0.658 to −0.113; P = .006). No significant associations were found in controls.

    Conclusions and Relevance  This study used MWI to demonstrate that otherwise normal-appearing brain tissue is diffusely damaged in MS, and the findings suggest that myelin water measures are associated with cognitive performance. MWI offers an in vivo biomarker feasible for use in clinical trials investigating cognition, providing a means for monitoring changes in myelination and its association with symptom worsening or improvement.

    Introduction

    Multiple sclerosis (MS) is an inflammatory, neurodegenerative disease of the central nervous system1 that affects more than 2 million people globally, rendering it the most prevalent chronic neuroinflammatory disease of the central nervous system worldwide.2 Cognitive impairment is a common symptom in MS that presents in up to 70% of patients.3 Cognitive symptoms in MS typically manifest as deficits in attention, memory, and processing speed,3 with processing speed being most frequently affected.4

    MS-related cognitive impairment has a substantial impact on quality of life,5 including the ability to perform tasks of daily living,6 fitness to drive,7 and social functioning.6 It is also a major contributor to unemployment in patients with MS.5,8 Undoubtedly, cognitive impairment presents a major burden to those living with MS, and an improved understanding of its underlying pathology would be of great benefit to patients and clinicians.

    MS is characterized by demyelination,9 with the radiological hallmark being focal areas of myelin loss, referred to as lesions.10 Lesions are visible on conventional T1-weighted and T2-weighted contrast-enhanced magnetic resonance imaging (MRI) and are the mainstay of MS diagnosis and disease monitoring.11 However, the association of focal lesion burden with physical and cognitive disability is limited; this is known as the clinicoradiological paradox.12,13 One possible cause of this paradox may be because MS pathology extends beyond lesions that are visible on conventional MRI.14 Postmortem histopathology studies suggest that normal-appearing white matter (NAWM)—areas that appear normal on standard imaging—is diffusely demyelinated in MS.15,16 Investigating the contribution of demyelination within NAWM to clinical outcomes, such as cognition, requires advanced imaging techniques that are quantitative, sensitive, and biologically specific to MS pathology.

    Quantitative characterization of myelin in vivo is feasible using myelin water imaging (MWI). MWI separates the MRI signal into contributions from the distinct water pools within a voxel according to the MRI property known as T2 relaxation time. In central nervous system tissue, these water pools generally correspond to (1) a long relaxation time component that arises from cerebrospinal fluid (T2 ≈ 2 seconds), (2) an intermediate component stemming from intracellular and extracellular water (T2 ≈ 60-80 ms), and (3) a short component stemming from water trapped between the myelin bilayers (T2 ≈ 20 ms).17 The fraction of MR signal arising from the myelin water divided by the total water signal is the myelin water fraction (MWF). The MWF has been histologically validated as a specific marker for myelin using human tissue18,19 and animal models of myelin damage.20 At present, MWI is the most direct means of assessing alterations in myelin noninvasively.

    Here, we used MWI to investigate the role of myelin damage in NAWM and cognitive function in MS. MWF results are typically described by the mean value within a region of interest (ROI) or by the heterogeneity (variance). To increase sensitivity to disease-associated changes, we characterized the entire MWF distribution by combining both measures as the coefficient of variation (SD / mean), termed the myelin heterogeneity index (MHI), with an increased MHI indicating more myelin damage. To assess cognitive deficits, we used measures from widely used and validated cognitive batteries for MS.21,22 Processing speed and attention were measured using the oral version of the Symbol Digit Modalities Test (SDMT), verbal memory with the Selective Reminding Test (SRT), word retrieval with the Controlled Oral Word Association Test (COWAT), and visuospatial memory with the Brief Visuospatial Memory Test Revised (BVMT-R). We selected 3 white matter (WM) ROIs—the cingulum, superior longitudinal fasciculus (SLF), and corpus callosum—a priori on the basis of their known involvement in MS-related cognitive impairment.23 We hypothesized that an increased MHI would be associated with worse cognitive performance in MS.

    Methods
    Participants

    This study was approved by the University of British Columbia Clinical Research Ethics Board. All participants provided written informed consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

    Participants were recruited through the University of British Columbia Hospital MS clinic and via online recruitment advertisements on local health authority websites. Study appointments took place from August 23, 2017, to February 20, 2019. Seventy-three participants with clinically definite MS fulfilling the 2017 revised McDonald criteria for diagnosis11 (38 with relapsing-remitting MS, 12 with primary progressive MS, and 23 with secondary progressive MS) and 22 age-, sex- and, education-matched healthy volunteers without neurological disease were included in the study. All MS phenotypes were recruited to capture varying levels of cognitive disability and MWF values.

    Clinical and Neuropsychological Assessments

    To characterize overall disability, participants were examined with the Kurtzke Expanded Disability Status Scale.24 To investigate whether our MWI findings were specific to cognition rather than a proxy for physical disability, participants with MS performed the timed 25-foot walk (T25-FW) as a measure of lower limb function and the 9-hole peg test (9-HPT) as a measure of upper limb function. Participants performed a battery of neuropsychological assessments validated for use in MS.

    The oral version of the SDMT21 was used as a measure of processing speed. One control participant performed the written version with instructions provided by a translator as they were non–English speaking. This test contains a reference key with the numbers 1 through 9 each corresponding to different geometric symbols. The answer key contains only symbols to which the participant must match the corresponding number according to the key. The subject responds orally with the digit associated with the symbol as quickly as possible. The test is scored by tallying the total number of correct responses achieved in 90 seconds.

    The SRT21 was used to assess verbal memory. The participant is read aloud a list of 12 words that they are asked to repeat back immediately. After they have repeated all words they can remember, the participant is read back only the words they have missed and asked to repeat all 12 words again. This procedure is repeated for 6 rounds. The participant is again asked to repeat all 12 words subsequent to a delay during which they perform other cognitive tests in the battery. The SRT is scored by tallying the total correct recalled words.

    Word retrieval was assessed using the COWAT.21,22 The participant is given a letter of the alphabet (eg, “F”) and asked to list as many words as they can produce beginning with that letter in 1 minute’s time. The version of the test used in this study included 3-letter prompting categories and an animal category, for which the participant names as many animals as they can recall. The COWAT is scored by tallying the total number of permissible answers; proper names (eg, cities) and variations of the same word (eg, runs, running, ran) are not permitted.

    Visuospatial memory was evaluated with the BVMT-R.22 The participant is presented with a display of 6 geometric figures for 10 seconds and then asked to reproduce the display by drawing it exactly as it was seen on a blank page. The participant is given 3 opportunities to view and reproduce the display. In the present study, the total recall t-score was used, which is the sum of all valid items generated across learning trials 1 through 3, corrected for age. Lower scores indicate worse performance on all tests.

    All participants completed the SDMT at a minimum. However, because of the time constraints of the research appointment, not all participants completed every test. The SRT was completed by 66 patients and 17 controls, 65 patients and 16 controls completed the COWAT, and 63 patients and 15 controls completed the BVMT-R.

    MRI Data Acquisition

    MRI scans were conducted on a 3-T scanner (Achieva; Philips Healthcare). Sequences included a 3-dimensional (3D), T1-weighted anatomical scan (whole-brain 3D magnetization-prepared rapid gradient-echo [MPRAGE]; repetition time, 3000 ms; inversion time, 1072 ms; 1 × 1 × 1 mm voxel; 160 slices) for registration and segmentation of WM ROIs and a 48-echo, 3D gradient and spin-echo (GRASE) T2 relaxation sequence with an echo-planar imaging factor of 3 (repetition time, 1073 ms; echo spacing, 8 ms; 20 slices acquired at 1 × 2 × 5 mm reconstructed to 40 slices at 1 × 1 × 2.5 mm) for MWF determination.25 A spin-echo, proton density–weighted, T2-weighted scan (repetition time, 2900 ms; echo time, 8.4/80 ms; 0.94 × 0.94 × 3 mm; 54 slices) was also acquired for lesion identification.

    MR Image Registration and Analysis

    Voxelwise signal decay curves obtained from the T2 relaxation (GRASE) sequence were modeled by multiple exponential components, with no a priori assumptions about the number of contributing exponentials. Analysis used a regularized, nonnegative, least-squares algorithm with the extended phase graph algorithm and flip angle estimation to correct for stimulated echoes using both in-house software and Matlab version R2013b (MathWorks).26 Voxelwise MWF maps were computed as the ratio of the area under the T2 distribution with times of less than 40 ms to the total area under the distribution.25

    MWF maps were aligned with the anatomical images for each individual by linearly coregistering the 3D T1-weighted images to the first echo of the GRASE scan using FLIRT (with 9 df),27 which is a linear registration tool and part of the Oxford Centre for Functional MRI of the Brain Software Library (FSL version 5.0.2).28 Non–brain parenchyma signal was removed using an automated approach with Brain Extraction Tool, which is part of FSL.29 WM masks were generated from the 3D T1-weighted image using the automated brain segmentation algorithm FAST,30 which is part of FSL, followed by in-plane erosion using a 2-dimensional kernel and 3 × 3 × 1 box centered on the target voxel to eliminate gray matter and cerebrospinal fluid voxels. The John Hopkins University tract atlas31 in the MNI (standard template) space was used to segment 3 WM ROIs (cingulum, SLF, and corpus callosum selected a priori because of their known involvement in MS-related cognitive impairment) using FSL.23 The 3D T1-weighted image was nonlinearly warped to MNI space using FNIRT (a nonlinear registration tool), and the ROIs were then transformed onto the MWF map in native space. Each ROI mask was multiplied by the participant’s global WM mask to eliminate gray matter, cerebrospinal fluid, and most of the lesioned tissue, then manually checked and edited as needed. Lesion masks produced by a neurologist were subtracted from the ROI masks to eliminate any remaining lesioned tissue. The MHI was computed for the cingulum, SLF, and corpus callosum NAWM for each individual by dividing the SD of MWF values by the mean MWF.

    Statistical Analysis

    All statistical procedures were performed using SPSS Statistics for Mac software version 25.0 (IBM Corp) from March to November 2019. Assumptions of normality were tested with the Shapiro-Wilk test for normality. The assumption of homogeneity of variance was tested with the Levene test of equality of variance. If the assumptions for a parametric test were violated, we proceeded with the appropriate nonparametric test. The Welch t test was used to determine whether there was a significant difference in age and years of education between groups. A χ2 test was used to determine whether the groups were matched for sex. Associations between MHI and performance on each cognitive test were explored with Pearson correlation. Although all P values less than .05 are reported, significance thresholds were set with the Bonferroni correction for the 3 brain regions assessed, with each cognitive domain treated separately (P < .016). All tests were 2-sided.

    Results
    Participant Characteristics

    There were 95 total participants with a mean (SD) age of 49.33 (11.44) years. Of the 73 participants with MS, the mean (SD) age was 50.2 (10.7) years (range, 26-65 years), 48 (66%) were women, and they had a mean (SD) of 14.7 (2.2) years of education (range, 12-22 years). Participants with MS had a median Kurtzke Expanded Disability Status Scale of 3.5 (range, 1.0-8.5) and median disease duration of 12 (range, 0.3-48) years. Controls had a mean (SD) age of 46.4 (13.5) years (range, 27-65 years), 14 (64%) were women, and they had a mean (SD) of 15.8 (2.2) years of education (range, 12-22 years). Participants with MS and controls did not differ significantly in age, sex, or education. The clinical and demographic characteristics of participants with MS and controls are shown in the Table.

    Symbol Digit Modalities Test

    MS patients had a mean SDMT score of 56 (range, 7-88). The mean SDMT score for controls was 62 (range, 41-79). Figure 1 illustrates the correlations between the MHI in NAWM and SDMT scores. In patients with MS, an increased MHI in the SLF (r = −0.490; 95% CI, −0.697 to −0.284; P < .001), corpus callosum (r = −0.471; 95% CI, −0.680 to −0.262; P < .001), and cingulum (r = −0.419; 95% CI: −0.634 to −0.205; P < .001) was associated with worse performance on the SDMT. In controls, MHI was not associated with SDMT performance in any ROI.

    Selective Reminding Test

    The mean SRT score was 55 (range, 23-83) for participants with MS and 62 (range, 46-78) for controls. Increased MHI in the SLF (r = −0.444; 95% CI, −0.660 to −0.217; P < .001), corpus callosum (r = −0.411; 95% CI, −0.630 to −0.181; P = .001), and cingulum (r = −0.361; 95% CI, −0.602 to −0.130; P = .003) was significantly correlated with worse SRT performance in participants with MS. MHI was not associated with SRT performance in controls (Figure 2).

    Controlled Oral Word Association Test

    Participants with MS had a mean score of 60 (range, 19-103) on the COWAT, whereas controls had a mean score of 69 (range, 55-88). Worse COWAT scores were correlated with increased MHI in NAWM of participants with MS in the SLF (r = −0.317; 95% CI, −0.549 to −0.078; P = .01) and cingulum (r = −0.335; 95% CI, −0.658 to −0.113; P = .006); however, the corpus callosum (r = −0.294; 95% CI, −0.535 to −0.053; P = .02) did not pass Bonferroni correction for multiple comparisons. No significant correlations were observed between MHI and COWAT scores in controls (Figure 3).

    Brief Visuospatial Memory Test–Revised

    Participants with MS had a mean BVMT-R score of 49 (range, 1-71), whereas controls had a mean score of 56 (range, 25-68). Worse BVMT-R scores were correlated with increased MHI in NAWM of participants with MS in the SLF (r = −0.257; 95% CI, −0.582 to −0.011; P = .04), corpus callosum (r = −0.250; 95% CI, −0.505 to −0.002; P = .048), and cingulum (r = −0.266; 95% CI, −0.515 to −0.019; P = .04); however, these associations did not reach significance after Bonferroni correction. No significant correlations were observed between MHI and BVMT-R scores in controls (eFigure in the Supplement).

    MHI and Cognitive Performance

    To illustrate the association between MHI and cognitive performance, Figure 4 depicts MWF maps and the distribution of MWF values within the SLF for high, moderate, and low MHI, and the associated range of cognitive z scores in 3 people with MS. The SLF was selected for this example as it most clearly exhibited an association with cognitive test scores in MS of the 3 ROIs. A range of cognitive z scores obtained from the 4 tests for each of the 3 patients are displayed as they are more informative than raw cognitive scores for this illustrative purpose. The z scores were calculated using the mean and SD for each cognitive test from our control sample. Patient A had a high MHI (0.59) in the SLF, matching their low cognitive scores compared with controls (z = −3.6 to −5.1). Patient B had both moderate MHI (0.28) in the SLF and cognitive test scores (z = −0.9 to −1.6). Patient C had a low MHI (0.22) in the SLF and performed at and above the level of controls on the cognitive tests (z = −0.08 to 0.9).

    Physical Disability (T25-FW and 9-HPT)

    MHI was not associated with lower limb disability, as measured by the T25-FW, in corpus callosum (r = −0.018; P = .90), SLF (r = −0.007; P > .99), or the cingulum (r = 0.068; P = .60). In addition, MHI in the corpus callosum (r = 0.07; P = .60), SLF (r = 0.226; P = .08), and cingulum (r = 0.184; P = .20) was not associated with the 9-HPT.

    Discussion

    Increased MHI in the MS cohort indicative of diffuse myelin abnormalities was found in NAWM of tracts that are associated with cognition. Furthermore, the MHI abnormalities correlated with cognitive deficits that are common in MS. Recent histopathological studies15,16 suggest that myelin in NAWM is affected in MS, and previous MWI reports32 suggested that decreased mean MWF and increased MWF heterogeneity (increased MWF variance in the brain) in NAWM is associated with worse physical disability. The MHI incorporates both MWF mean and variance; a decrease in MWF mean or an increase in MWF heterogeneity would result in an increased MHI. Thus, the MHI is sensitive to gross changes in MWF that shift the entire distribution, as well as smaller changes that broaden the distribution more subtly yet are clinically relevant.

    Previous quantitative MRI studies using diffusion tensor imaging (DTI) or magnetization transfer imaging (MTI) have reported associations between abnormalities in NAWM and cognitive impairment in MS in the same brain regions as our study. DTI studies use fractional anisotropy (FA) as a general measure of microstructural WM tissue integrity. In WM, diffusion of water is restricted by microstructural components such as myelin. This causes the diffusion to be parallel, rather than perpendicular, to the direction of the axonal fibers, and it yields a high FA value in healthy WM. When microstructural damage to WM occurs, restrictions on the movement of water molecules are reduced, and diffusion becomes more isotropic. This results in a reduction in the FA.33 DTI also provides the mean diffusivity (MD) metric (equal to the magnitude of diffusion) with increased MD thought to represent microstructural abnormalities in WM.34 MTI studies typically report the magnetization transfer ratio, a semiquantitative metric that estimates the exchange of magnetization between nonaqueous tissue and water and is proposed to decrease with myelin loss.35

    In the current study, we found a significant correlation between increased MHI in NAWM in the SLF, corpus callosum, and cingulum, and slower processing speed performance as measured by the SDMT. Similarly, decreased FA and magnetization transfer ratio,as well as increased MD, have been associated with worse processing speed in the SLF,36-38 corpus callosum,36-41 and cingulum NAWM.36,37 Likewise, we observed an association between increased MHI in NAWM in the SLF, corpus callosum, and cingulum, and decreased verbal memory (worse SRT scores). Associations between decreased FA in NAWM in these same 3 brain regions and impaired verbal memory have also been reported.36 In contrast to a previous study36 that failed to find significant associations between FA in NAWM and COWAT scores, we found significant correlations between MHI in NAWM and performance on the COWAT in participants with MS. This may be because MHI is a more specific measure of myelin damage than mean FA values. We noted the correlation between increased MHI in NAWM in SLF, corpus callosum, and cingulum, and decreased visuospatial memory as measured by the BVMT-R was P < 0.05, but it did not pass multiple testing correction. However, it is worth noting that increased FA and decreased MD in these brain regions and a significant association with decreased visuospatial memory has been observed in studies36,37 that did not use the conservative Bonferroni correction. It is possible that investigating visual pathway ROIs would lead to more significant correlations.

    It was observed that there was a subset of patients with particularly low cognitive scores and high MHI values. Not surprisingly, these tended to be participants with progressive MS.3 This finding emphasizes the importance of including both relapsing-remitting and progressive phenotypes, with a broad spectrum of cognitive ability and severity of myelin damage representative of the population with MS as a whole, to comprehensively characterize the association between cognitive ability and severity of myelin damage in correlation studies.

    To determine whether the observed associations between MHI values and cognitive performance were specific to cognition rather than a proxy for overall disability, we investigated the association between MHI in the 3 WM tracts we selected and performance on the T25-FW and the 9-HPT. We found no association between MHI in the corpus callosum, SLF, and cingulum and upper and lower limb disability. Therefore, we believe our findings are specific to cognitive function.

    Although our results are consistent with previous DTI and MTI studies, MWI offers greater biological interpretability. DTI measures reflect a large number of biological changes that occur in MS, and caution is warranted when interpreting diffusion anisotropy changes as myelin changes.42 DTI measures are influenced by biological factors other than myelin, such as the directionality of fiber bundles, tortuosity, fiber crossings, and fiber orientation and coherence.42-46

    Similarly, MTI estimates of macromolecular-bound water include but are not limited to myelin, and they are heavily influenced by inflammation and edema,35 which are often present in MS. In contrast, MWI has been validated with both human histological18,19,47 and animal models20 as a specific measure of myelin. We acknowledge that the denominator in the MWF is total water; therefore, increases in edema and inflammation can influence changes in MWF. However, if the MWF decreases in MS reported in the literature were due to only edema rather than myelin loss, dilution of the MWF from edema would require such significant swelling in the brain that it would result in a lethal increase in intracranial pressure.14 The use of MWI in the current study provides evidence that cognitive symptoms in MS are associated, at least in part, with myelin abnormalities in NAWM. This is of major clinical importance as it not only provides insight into the underlying pathology contributing to cognitive symptoms in MS, but also offers a noninvasive, tissue-specific biomarker for monitoring treatment efficacy, particularly for therapies geared toward remyelination.48

    Limitations

    The study was limited to 1 hospital site, using a single scanner, which might restrict the generalizability of the results. However, the findings further support the use of MWI, and specifically MHI, to quantify MS-related demyelination and the association with cognitive impairment. This would be feasible for multicenter studies because MWI has excellent intersite49 and intervendor50 reproducibility.

    Conclusions

    This study implements a myelin-specific imaging technique to demonstrate that otherwise normal-appearing brain tissue is diffusely damaged in participants wtih MS. We have also found that these changes are significantly associated with disease-related cognitive symptoms. These findings contribute to a better understanding of the underlying pathology involved in MS-related cognitive impairment; myelin damage in NAWM is likely playing a role. The MHI metric offers an in vivo marker feasible for use in clinical trials investigating cognitive symptoms in MS, for which a reliable, quantitative biomarker is sorely needed.

    Back to top
    Article Information

    Accepted for Publication: June 1, 2020.

    Published: September 29, 2020. doi:10.1001/jamanetworkopen.2020.14220

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Abel S et al. JAMA Network Open.

    Corresponding Authors: Shawna Abel, MSc (shawna.galley@gmail.com), and Shannon H. Kolind, PhD (shannon.kolind@ubc.ca), Department of Medicine (Neurology), The University of British Columbia, 2221 Wesbrook Mall, M10 Purdy Pavilion, Vancouver, BC V6T 2B5, Canada.

    Author Contributions: Ms Abel had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Abel, Vavasour, Laule, Dvorak, Kuan, Wilken, Li, Kolind.

    Acquisition, analysis, or interpretation of data: Abel, Vavasour, Lee, Johnson, Ristow, Ackermans, Chan, Cross, Laule, Schabas, Hernández-Torres, Tam, Morrow, Wilken, Rauscher, Bhan, Sayao, Devonshire, Li, Carruthers, Traboulsee, Kolind.

    Drafting of the manuscript: Abel, Vavasour, Laule, Schabas, Hernández-Torres, Morrow, Rauscher, Carruthers.

    Critical revision of the manuscript for important intellectual content: Abel, Vavasour, Lee, Johnson, Ristow, Ackermans, Chan, Cross, Laule, Dvorak, Schabas, Tam, Kuan, Wilken, Rauscher, Bhan, Sayao, Devonshire, Li, Traboulsee, Kolind.

    Statistical analysis: Abel.

    Obtained funding: Kolind.

    Administrative, technical, or material support: Abel, Vavasour, Lee, Johnson, Cross, Laule, Dvorak, Schabas, Tam, Kuan, Morrow, Wilken, Rauscher, Devonshire, Traboulsee, Kolind.

    Supervision: Vavasour, Morrow, Sayao, Traboulsee, Kolind.

    Conflict of Interest Disclosures: Dr Cross reported receiving personal fees from Sanofi Genzyme outside the submitted work. Dr Schabas reported receiving personal fees from Biogen, Teva, Novartis, and Genzyme outside the submitted work. Dr Sayao reported receiving speaking honoraria from Biogen and Merck/Serono and participating in advisory boards from Biogen, Merck/Serono, Teva, Roche, and Novartis. Dr Li reported receiving grants from Consortium of MS Centers and personal fees from Vertex Pharmaceuticals, Sanofi Genzyme, Celgene, Biogen, and Academy of Health Care Learning outside the submitted work. Dr Carruthers reported receiving grants and personal fees from Teva Innovation Canada, Roche Canada, and Biogen; personal fees and study site investigator from Novartis, study site investigator from MedImmune, personal fees and study site investigator from EMD Serono outside the submitted work. Dr Traboulsee reported receiving grants and personal fees from Roche and Sanofi Genzyme, personal fees and nonfinancial support from Consortium of MS Centers, and personal fees from Biogen, Novartis, and Teva outside the submitted work. Dr Kolind reported receiving grants from Genzyme and F. Hoffmann–La Roche and personal fees from Novartis outside the submitted work. No other disclosures were reported.

    Funding/Support: This work was supported by grant F16-04023 from the Multiple Sclerosis Society of Canada. Dr Kolind also reported receiving salary support from the Michael Smith Foundation for Health Research and the Milan and Maureen Ilich Foundation during the conduct of this study.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Meeting Presentation: This paper was presented as a poster at the 24th Congress of the European Committee for the Treatment and Research in Multiple Sclerosis (ECTRIMS); September 11, 2019; Stockholm, Sweden.

    Additional Contributions: Devann Paterson, an undergraduate volunteer, provided assistance with data scoring and entry. The Statistical Opportunity for Students program from the University of British Columbia Statistics and Data Science Group provided consultation. We thank all the study participants. Aside from $40 to refund travel, parking, and a meal, no additional compensation was provided.

    References
    1.
    Kawachi  I, Lassmann  H.  Neurodegeneration in multiple sclerosis and neuromyelitis optica.   J Neurol Neurosurg Psychiatry. 2017;88(2):137-145. doi:10.1136/jnnp-2016-313300PubMedGoogle ScholarCrossref
    2.
    Wallin  MT, Culpepper  WJ, Nichols  E, Lancet  ZBT; GBD 2016 Multiple Sclerosis Collaborators.  Global, regional, and national burden of multiple sclerosis 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016.   Lancet Neurol. 2019;18(3):269-285. doi:10.1016/S1474-4422(18)30443-5PubMedGoogle ScholarCrossref
    3.
    Chiaravalloti  ND, DeLuca  J.  Cognitive impairment in multiple sclerosis.   Lancet Neurol. 2008;7(12):1139-1151. doi:10.1016/S1474-4422(08)70259-XPubMedGoogle ScholarCrossref
    4.
    Costa  SL, Genova  HM, DeLuca  J, Chiaravalloti  ND.  Information processing speed in multiple sclerosis: past, present, and future.   Mult Scler. 2017;23(6):772-789. doi:10.1177/1352458516645869PubMedGoogle ScholarCrossref
    5.
    Ruet  A, Deloire  M, Hamel  D, Ouallet  J-C, Petry  K, Brochet  B.  Cognitive impairment, health-related quality of life and vocational status at early stages of multiple sclerosis: a 7-year longitudinal study.   J Neurol. 2013;260(3):776-784. doi:10.1007/s00415-012-6705-1PubMedGoogle ScholarCrossref
    6.
    Rao  SM, Leo  GJ, Ellington  L, Nauertz  T, Bernardin  L, Unverzagt  F.  Cognitive dysfunction in multiple sclerosis. II. impact on employment and social functioning.   Neurology. 1991;41(5):692-696. doi:10.1212/WNL.41.5.692PubMedGoogle ScholarCrossref
    7.
    Morrow  SA, Classen  S, Monahan  M,  et al.  On-road assessment of fitness-to-drive in persons with MS with cognitive impairment: a prospective study.   Mult Scler. 2018;24(11):1499-1506. doi:10.1177/1352458517723991PubMedGoogle ScholarCrossref
    8.
    Strober  L, Chiaravalloti  N, Moore  N, DeLuca  J.  Unemployment in multiple sclerosis (MS): utility of the MS Functional Composite and cognitive testing.   Mult Scler. 2014;20(1):112-115. doi:10.1177/1352458513488235PubMedGoogle ScholarCrossref
    9.
    Reich  DS, Lucchinetti  CF, Calabresi  PA.  Multiple sclerosis.   N Engl J Med. 2018;378(2):169-180. doi:10.1056/NEJMra1401483PubMedGoogle ScholarCrossref
    10.
    Popescu  BFG, Pirko  I, Lucchinetti  CF.  Pathology of multiple sclerosis: where do we stand?   Continuum (Minneap Minn). 2013;19(4):901-921. doi:10.1212/01.CON.0000433291.23091.65.Google Scholar
    11.
    Thompson  AJ, Banwell  BL, Barkhof  F,  et al.  Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria.   Lancet Neurol. 2018;17(2):162-173. doi:10.1016/S1474-4422(17)30470-2PubMedGoogle ScholarCrossref
    12.
    Barkhof  F.  The clinico-radiological paradox in multiple sclerosis revisited.   Curr Opin Neurol. 2002;15(3):239-245. doi:10.1097/00019052-200206000-00003PubMedGoogle ScholarCrossref
    13.
    Mollison  D, Sellar  R, Bastin  M,  et al.  The clinico-radiological paradox of cognitive function and MRI burden of white matter lesions in people with multiple sclerosis: a systematic review and meta-analysis.   PLoS One. 2017;12(5):e0177727. doi:10.1371/journal.pone.0177727PubMedGoogle Scholar
    14.
    Vavasour  IM, Huijskens  SC, Li  DK,  et al.  Global loss of myelin water over 5 years in multiple sclerosis normal-appearing white matter.   Mult Scler. 2018;24(12):1557-1568. doi:10.1177/1352458517723717PubMedGoogle ScholarCrossref
    15.
    Jäkel  S, Agirre  E, Mendanha Falcão  A,  et al.  Altered human oligodendrocyte heterogeneity in multiple sclerosis.   Nature. 2019;566(7745):543-547. doi:10.1038/s41586-019-0903-2PubMedGoogle ScholarCrossref
    16.
    Yeung  MSY, Djelloul  M, Steiner  E,  et al.  Dynamics of oligodendrocyte generation in multiple sclerosis.   Nature. 2019;566(7745):538-542. doi:10.1038/s41586-018-0842-3PubMedGoogle ScholarCrossref
    17.
    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):673-677. doi:10.1002/mrm.1910310614PubMedGoogle ScholarCrossref
    18.
    Laule  C, Leung  E, Lis  DKB,  et al.  Myelin water imaging in multiple sclerosis: quantitative correlations with histopathology.   Mult Scler. 2006;12(6):747-753. doi:10.1177/1352458506070928PubMedGoogle ScholarCrossref
    19.
    Laule  C, Kozlowski  P, Leung  E, Li  DKB, Mackay  AL, Moore  GRW.  Myelin water imaging of multiple sclerosis at 7 T: correlations with histopathology.   Neuroimage. 2008;40(4):1575-1580. doi:10.1016/j.neuroimage.2007.12.008PubMedGoogle ScholarCrossref
    20.
    McCreary  CR, Bjarnason  TA, Skihar  V, Mitchell  JR, Yong  VW, Dunn  JF.  Multiexponential T2 and magnetization transfer MRI of demyelination and remyelination in murine spinal cord.   Neuroimage. 2009;45(4):1173-1182. doi:10.1016/j.neuroimage.2008.12.071PubMedGoogle ScholarCrossref
    21.
    Rao  SM, Leo  GJ, Bernardin  L, Unverzagt  F.  Cognitive dysfunction in multiple sclerosis. I. frequency, patterns, and prediction.   Neurology. 1991;41(5):685-691. doi:10.1212/WNL.41.5.685PubMedGoogle ScholarCrossref
    22.
    Benedict  RHB, Fischer  JS, Archibald  CJ,  et al.  Minimal neuropsychological assessment of MS patients: a consensus approach.   Clin Neuropsychol. 2002;16(3):381-397. doi:10.1076/clin.16.3.381.13859PubMedGoogle ScholarCrossref
    23.
    Zhang  X, Zhang  F, Huang  D,  et al.  Contribution of gray and white matter abnormalities to cognitive impairment in multiple sclerosis.   Int J Mol Sci. 2016;18(1):46-13. doi:10.3390/ijms18010046PubMedGoogle ScholarCrossref
    24.
    Kurtzke  JF.  Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS).   Neurology. 1983;33(11):1444-1452. doi:10.1212/WNL.33.11.1444PubMedGoogle ScholarCrossref
    25.
    Prasloski  T, Rauscher  A, MacKay  AL,  et al.  Rapid whole cerebrum myelin water imaging using a 3D GRASE sequence.   Neuroimage. 2012;63(1):533-539. doi:10.1016/j.neuroimage.2012.06.064PubMedGoogle ScholarCrossref
    26.
    Prasloski  T, Mädler  B, Xiang  Q-S, MacKay  A, Jones  C.  Applications of stimulated echo correction to multicomponent T2 analysis.   Magn Reson Med. 2012;67(6):1803-1814. doi:10.1002/mrm.23157PubMedGoogle ScholarCrossref
    27.
    Jenkinson  M, Bannister  P, Brady  M, Smith  S.  Improved optimization for the robust and accurate linear registration and motion correction of brain images.   Neuroimage. 2002;17(2):825-841. doi:10.1006/nimg.2002.1132PubMedGoogle ScholarCrossref
    28.
    Smith  SM, Jenkinson  M, Woolrich  MW,  et al.  Advances in functional and structural MR image analysis and implementation as FSL.   Neuroimage. 2004;23(1)(suppl):S208-S219. doi:10.1016/j.neuroimage.2004.07.051PubMedGoogle ScholarCrossref
    29.
    Smith  SM.  Fast robust automated brain extraction.   Hum Brain Mapp. 2002;17(3):143-155. doi:10.1002/hbm.10062PubMedGoogle ScholarCrossref
    30.
    Zhang  Y, Brady  M, Smith  S.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm.   IEEE Trans Med Imaging. 2001;20(1):45-57. doi:10.1109/42.906424PubMedGoogle ScholarCrossref
    31.
    Mori  S, Wakana  S, van Zijl  PCM, Nagae-Poetscher  LM.  MRI Atlas of Human White Matter. Elsevier Science. 1st ed. 2005.
    32.
    Kolind  S, Matthews  L, Johansen-Berg  H,  et al.  Myelin water imaging reflects clinical variability in multiple sclerosis.   Neuroimage. 2012;60(1):263-270. doi:10.1016/j.neuroimage.2011.11.070PubMedGoogle ScholarCrossref
    33.
    Thomsen  C, Henriksen  O, Ring  P.  In vivo measurement of water self diffusion in the human brain by magnetic resonance imaging.   Acta Radiol. 1987;28(3):353-361. doi:10.1177/028418518702800324PubMedGoogle ScholarCrossref
    34.
    Inglese  M, Bester  M.  Diffusion imaging in multiple sclerosis: research and clinical implications.   NMR Biomed. 2010;23(7):865-872. doi:10.1002/nbm.1515PubMedGoogle ScholarCrossref
    35.
    Vavasour  IM, Laule  C, Li  DKB, Traboulsee  AL, MacKay  AL.  Is the magnetization transfer ratio a marker for myelin in multiple sclerosis?   J Magn Reson Imaging. 2011;33(3):713-718. doi:10.1002/jmri.22441PubMedGoogle ScholarCrossref
    36.
    Dineen  RA, Vilisaar  J, Hlinka  J,  et al.  Disconnection as a mechanism for cognitive dysfunction in multiple sclerosis.   Brain. 2009;132(1):239-249. doi:10.1093/brain/awn275PubMedGoogle ScholarCrossref
    37.
    Preziosa  P, Rocca  MA, Pagani  E,  et al; MAGNIMS Study Group.  Structural MRI correlates of cognitive impairment in patients with multiple sclerosis: a multicenter study.   Hum Brain Mapp. 2016;37(4):1627-1644. doi:10.1002/hbm.23125PubMedGoogle ScholarCrossref
    38.
    Ranjeva  J-P, Audoin  B, Au Duong  MV,  et al.  Local tissue damage assessed with statistical mapping analysis of brain magnetization transfer ratio: relationship with functional status of patients in the earliest stage of multiple sclerosis.   AJNR Am J Neuroradiol. 2005;26(1):119-127.PubMedGoogle Scholar
    39.
    Pokryszko-Dragan  A, Banaszek  A, Nowakowska-Kotas  M,  et al.  Diffusion tensor imaging findings in the multiple sclerosis patients and their relationships to various aspects of disability.   J Neurol Sci. 2018;391:127-133. doi:10.1016/j.jns.2018.06.007PubMedGoogle ScholarCrossref
    40.
    Bethune  A, Tipu  V, Sled  JG,  et al.  Diffusion tensor imaging and cognitive speed in children with multiple sclerosis.   J Neurol Sci. 2011;309(1-2):68-74. doi:10.1016/j.jns.2011.07.019PubMedGoogle ScholarCrossref
    41.
    Roosendaal  SD, Geurts  JJG, Vrenken  H,  et al.  Regional DTI differences in multiple sclerosis patients.   Neuroimage. 2009;44(4):1397-1403. doi:10.1016/j.neuroimage.2008.10.026PubMedGoogle ScholarCrossref
    42.
    Hill  RA.  Do short-term changes in white matter structure indicate learning-induced myelin plasticity?   J Neurosci. 2013;33(50):19393-19395. doi:10.1523/JNEUROSCI.4122-13.2013PubMedGoogle ScholarCrossref
    43.
    Beaulieu  C.  The basis of anisotropic water diffusion in the nervous system: a technical review.   NMR Biomed. 2002;15(7-8):435-455. doi:10.1002/nbm.782PubMedGoogle ScholarCrossref
    44.
    Jones  DK, Knösche  TR, Turner  R.  White matter integrity, fiber count, and other fallacies: the do’s and don’ts of diffusion MRI.   Neuroimage. 2013;73:239-254. doi:10.1016/j.neuroimage.2012.06.081PubMedGoogle ScholarCrossref
    45.
    Mädler  B, Drabycz  SA, Kolind  SH, Whittall  KP, MacKay  AL.  Is diffusion anisotropy an accurate monitor of myelination? correlation of multicomponent T2 relaxation and diffusion tensor anisotropy in human brain.   Magn Reson Imaging. 2008;26(7):874-888. doi:10.1016/j.mri.2008.01.047PubMedGoogle ScholarCrossref
    46.
    Groeschel  S, Hagberg  GE, Schultz  T,  et al.  Assessing white matter microstructure in brain regions with different myelin architecture using MRI.   PLoS ONE. 2016. 11(11):e0167274. doi:10.1371/journal.pone.0167274.Google Scholar
    47.
    MacKay  AL, Laule  C.  Magnetic resonance of myelin water: an in vivo marker for myelin.   Brain Plast. 2016;2(1):71-91. doi:10.3233/BPL-160033PubMedGoogle ScholarCrossref
    48.
    Kremer  D, Göttle  P, Flores-Rivera  J, Hartung  H-P, Küry  P.  Remyelination in multiple sclerosis: from concept to clinical trials.   Curr Opin Neurol. 2019;32(3):378-384. doi:10.1097/WCO.0000000000000692PubMedGoogle ScholarCrossref
    49.
    Meyers  SM, Vavasour  IM, Mädler  B,  et al.  Multicenter measurements of myelin water fraction and geometric mean T2: intra- and intersite reproducibility.   J Magn Reson Imaging. 2013;38(6):1445-1453. doi:10.1002/jmri.24106PubMedGoogle ScholarCrossref
    50.
    Lee  LE, Ljungberg  E, Shin  D,  et al.  Inter-vendor reproducibility of myelin water imaging using a 3D gradient and spin echo sequence.   Front Neurosci. 2018;12:854. doi:10.3389/fnins.2018.00854PubMedGoogle ScholarCrossref
    ×