UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mechanisms of white matter in multiple sclerosis and neuromyelitis optica Manogaran, Praveena 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_september_manogaran_praveena.pdf [ 2.56MB ]
Metadata
JSON: 24-1.0167701.json
JSON-LD: 24-1.0167701-ld.json
RDF/XML (Pretty): 24-1.0167701-rdf.xml
RDF/JSON: 24-1.0167701-rdf.json
Turtle: 24-1.0167701-turtle.txt
N-Triples: 24-1.0167701-rdf-ntriples.txt
Original Record: 24-1.0167701-source.json
Full Text
24-1.0167701-fulltext.txt
Citation
24-1.0167701.ris

Full Text

MECHANISMS OF WHITE MATTER INJURY IN MULTIPLE SCLEROSIS AND NEUROMYELITIS OPTICA    by  PRAVEENA MANOGARAN  B.Sc., The University of British Columbia, 2013       A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF    MASTER OF SCIENCE    in    THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Experimental Medicine)     THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    April 2015         © Praveena Manogaran, 2015 ii  ABSTRACT  Neuromyelitis optica (NMO) and multiple sclerosis (MS) both result in acute injury (i.e. attacks or relapses) to the central nervous system with focal demyelination and axonal loss that varies in severity along a spectrum. A variety of non-invasive structural imaging and functional tools can be used to investigate mechanisms of white matter injury and secondary axonal injury in MS and NMO. These include advanced magnetic resonance imaging (MRI) measures of myelin water fraction; optical coherence tomography (OCT) for retinal nerve fibre layer thickness and total macular volume; and transcranial magnetic stimulation (TMS) to determine cortical excitability and integrity of cortical spinal pathways.  First, the relationship between a functional measure using TMS and a structural measure of myelin in the cortico-spinal tract was examined. Structural changes were found in the descending motor output pathway white matter in NMO along with abnormal TMS measures, suggesting that there is greater spinal cord involvement and more extensive axonal loss found in NMO compared to MS.  Next, OCT was used as a measure of the anterior visual pathway and myelin water imaging of the posterior visual pathway; the effects of damage to one part of the visual system on the other was studied. Retrograde degeneration to the retina and anterograde degeneration to the optic radiations from the optic nerve was observed in both MS and NMO subjects with optic neuritis history. A correlation between the measures indicating that damage to one part may cause damage to another part of the visual pathway. Finally, damage was observed in optic pathway in iii  MS patients without optic neuritis history suggesting that there is damage in the absence of lesions in the optic nerve.  Finally, myelin water imaging was used to investigate if the disease burden of lesions regulate the level of damage to the normal appearing white matter (NAWM) tracts. The lack of correlation between disease burden of lesions and NAWM myelin water imaging in MS suggested that damage to the NAWM was mediated by processes independent of lesions.  These techniques can be used to study and better understand demyelinating diseases such as MS and NMO.           iv   PREFACE  Chapter 2. A version of this material has been accepted for publication as Manogaran, P., Vavasour, I., Borich, M., Kolind, S., Lange, A., Rauscher, A., Boyd, L., Li, D., and Traboulsee , A. (2015). Cortico -spinal tract integrity measured using transcranial magnetic stimulation and magnetic resonance imaging in neuromyelitis optica and multiple sclerosis. Multiple Sclerosis. I. Vavasour and I share first authorship on this manuscript. I prepared the manuscript, tables and figures and did the original writing. The co-authors provided guidance and proof-reading for the manuscript. MRI acquisition was done by the UBC MRI technologists, and TMS acquisition was done by the UBC Brain Behaviour La b. I. Vavasour did the MRI analysis. I did part of the recruitment, and all of the TMS data analysis and statistical analysis. This project and associated methods was approved by the University of British Columbia ’s Research Ethics Board [certificate # H12 -00088  for “Study of Myelination Using MRI and MEP s” and #H09 -01938 for ‘Remyelination in Multiple Sclerosis with Glatiramer Acetate: Insights from Myelin Water Imaging and Evoked Potentials ’] .  Chapter 3. A version of this material is under preparation to be submitted with the title Quantifying visual pathway axonal and  myelin loss in neuromyelitis optica and multiple sclerosis. Co-authors include S. Kolind, A. Lange, I. Vavasour, Y. Zhao, R. Carruthers, K. McMullen, D. Li and A. Traboulsee. As first author, I prepared the manuscript, tables and figures and did the original writing. The co-authors provided guidance and proof-reading for the manuscript. I conducted the OCT examination on the NMO patients and healthy controls; the MS patients were examined by the UBC MS clinical trials under the OPERA study (A randomized, double -blind, double-dummy, parallel-group sub-study to evaluate the efficacy and v   safety of ocrelizumab in comparison to interferon beta -1a (Rebif®) in patients with relapsing multiple sclerosis). MRI acquisition was done by the UBC MRI technologists. I did part of the recruitment, all of the MRI analysis, OCT analysis and statistical analysis. This project and associated methods was approved by the University of British Columbia ’s Research Ethics Board [certificate # H12 -00088  for “Study of Myelination Using MRI and MEP s” and # H11 -02278 for ‘A Randomized, Double -Blind, Double -Dummy, Parallel -Group Study To Evaluate The Ef ficacy And Safety Of Ocrelizumab In Comparison To Inteferon-Beta-1a (Rebif ® ) In Patients With Relapsing Multiple Sclerosis ’] .  Chapter 4. Part of the MRI analysis was done by I. Vavasour  (creating the corpus callosum and cortico-spinal tract masks and initial myelin water fraction analysis of those regions). I created the optic radiation masks and the myelin water fraction analysis of the optic radiation. I also re-did all of the MWF analysis of the corpus callosum and cortico-spinal tract regions. Lesion identification was done by the UBC MS/MRI Research Group, specifically with the help of A. Riddehough and R. Tam. MRI acquisition was done by the UBC MRI technologists. Significant guidance and proof-reading for this chapter was provided by S. Kolind, I. Vavasour , and R. Carruthers. This project and associated methods was approved by the University of British Columbia’s Research Ethics Board [certificate # H12 -00088  for “Study of Myelination Using MRI and MEP s” and # H11 -02278 for ‘A Randomized, Double -Blind, Double -Dummy, Parallel -Group Study To Evaluate The Ef ficacy And Safety Of Ocrelizumab In Comparison To Inteferon-Beta-1a (Rebif ® ) In Patients With Relapsing Multiple Sclerosis ’ ] .  vi   TABLE OF CONTENTS   ABSTRACT....................................................................................................................................ii PREFACE .......................................................................................................................................iv TABLE OF CONTENTS ................................................................................................................vi LIST OF TABLES .......................................................................................................................viii LIST OF FIGURES ........................................................................................................................ix GLOSSARY ....................................................................................................................................x ACKNOWLEDGMENT S..............................................................................................................xi DEDICATION ..............................................................................................................................xiii 1  INTRODUCTION .......................................................................................................................1  1.1  CENTRAL NERVOUS SYSTEM ....................................................................................1  1.2  MULT IPLE SCLEROSIS ..................................................................................................3  1.2.1   MS Immunopathology.............................................................................................4   1.2.1.1  MS Immunopathology Model.....................................................................5  1.3   NEUROMYELITIS OPTICA ............................................................................................7  1.3.1  NMO Immunopathology.........................................................................................8   1.3.1.1  Humoral Immunity ......................................................................................8   1.3.1.2  Immunoglobulin G ......................................................................................9   1.3.1.3  Aquaporin-4 ..............................................................................................10   1.3.1.4  NMO Immunopathology Model...............................................................10  1.4   MAGNETIC RESONANCE IMAGING .........................................................................1 2  1.4.1  Myelin Water Fraction...........................................................................................14  1.5  CORTICO-SPINAL TRACT ..........................................................................................16  1.5.1  Transcranial Magnetic Stimulation........................................................................1 6    1.5.1.1  Motor Threshold.......................................................................................1 7    1.5.1.2  Recruitment Curve....................................................................................18  1.6  VISUAL PATHWAY ......................................................................................................1 9  1.6.1  Optic Neuritis.........................................................................................................20  1.6.2  Optical Coherence Tomography............................................................................22  1.7  OVERVIEW OF THESIS ...............................................................................................22  2  CORTICO-SPINAL TRACT INTEGRITY MEASURED USING TRANSCRANIAL MAGNETIC STIMULATION AND MAGNETIC RESONANCE IMAGING IN  NEUROMYELITIS OPTICA AND MULTIPLE SCLEROSIS ...................................................24  2.1  INTRODUCTION ...........................................................................................................24  2.2  METHODS ......................................................................................................................26   2.2.1  Parti cipants............................................................................................................26   2.2.2  TMS Protocol and Analysis ...................................................................................27    2.2.2.1  Motor Threshold.......................................................................................27    2.2.2.2  Recruitment Curve....................................................................................27   2.2.3  MRI Protocol and Analysis ...................................................................................28   2.2.4  Statistical Analysis.................................................................................................29    2.2.4.1  Independent MRI and TMS Analysis.......................................................29    2.2.4.2  Correlation Analysis.................................................................................30  2.3  RESULTS ........................................................................................................................30  2.3.1  Motor Threshold....................................................................................................3 1  vii   2.3.2  Recruitment Curves...............................................................................................31  2.3.3  MRI Assessment....................................................................................................3 2  2.3.4  Correlation Analysis..............................................................................................33  2.4  DISCUSSION ..................................................................................................................35  3   QUANTIFYING VISUAL PATHWAY AXONAL AND MYELIN LOSS IN  NEUROMYELITI S OPTICA AND MULTIPLE SCLEROSIS................................................... 40  3.1  INTRODUCTION ...........................................................................................................40  3.2  METHODS ......................................................................................................................43   3.2.1  Participants ............................................................................................................43   3 .2.2  OCT Protocol and Analysis ...................................................................................44    3.2.2.1   RNFL Thickness Protocol .........................................................................44    3.2.2.2   TMV Protocol ...........................................................................................45   3.2.3   MRI Protocol and Analysis ...................................................................................45   3.2.4   Statistical Analysis................................................................................................47    3.2.4.1   Independent MRI and OCT Analysis.......................................................47    3.2.4.2   Correlation Analysis.................................................................................48  3.3  RESULTS ........................................................................................................................48  3.3.1   RNFL Thickness ...................................................................................................48  3.3 .2   Total Macular Volume ..........................................................................................50  3.3.3   MRI Assessment...................................................................................................51  3.3.4   Correlation Analysis.............................................................................................52  3.4  DISCUSSION ..................................................................................................................55  4   CORRELATING LESION VOLUME AND MYELIN WATER FRACTION IN NORMAL APPEARING WHITE MATTER TRACTS OF MULTIPLE SCLEROSIS AND  NEUROMYELITIS OPTICA PATIENTS ....................................................................................60   4.1  INTRODUCTION ...........................................................................................................60  4.2  METHODS ......................................................................................................................62   4.2.1  Participants ............................................................................................................62   4 .2 .2  MRI Protocol and Analysis ...................................................................................62   4 .2.3  Statistical Analysis.................................................................................................65  4 .3  RESULTS ........................................................................................................................65  4 .3.1  MWF Assessment..................................................................................................65  4 .3.2  Correlation Analysis..............................................................................................66  4 .4  DISCUSSION ..................................................................................................................68  5  CONCLUSIONS .......................................................................................................................7 1  REFERENCES ..............................................................................................................................74  viii   LIST OF TABLES   TABLE 2.1: TMS/CST MWF Subject Character istics and Results..............................................26  TABLE 2.2: MRI, TMS and Clinical Measure Correlations for  the NMO and MS Cohort.........34  TABLE 3.1: OCT/OR MWF Subject Charact eristics and Results................................................44  TABLE 3.2: MRI, OCT and Clinical Measure Correlations for the NMO, MS and healthy control group separately and together............................................................................................54  ix   LIST OF FIGURES   FIGURE 1.1: The effects of demyelination on axonal conduction .................................................2  FIGURE 1.2: Current model of MS immunopathology ..................................................................6  FIGURE 1. 3 :  Humoral immunity ....................................................................................................9  FIGURE 1.4 : Current model of NMO immunopathology .............................................................12  FIGURE 1.5 : Myelin water fraction..............................................................................................15  FIGURE 1.6 : Diagra m demonstrating TMS stimulation...............................................................17  FIGURE 1. 7 : Recruitment curve of a healthy individual..............................................................18  FIGURE  1.8 : The visual pathway .................................................................................................20   FIGURE 2.1: The cortico-spinal tract region of interest...............................................................29  FIGURE 2.2:  Boxplot of average recruitment curve slope plotted for healthy controls, MS and NMO..............................................................................................................................................3 1  FIGURE 2. 3 : Average recruitment curve maximal amplitudes as a function of motor threshold  for healthy controls, MS and NMO...............................................................................................32  FIGURE 2. 4 : Boxplots of MWF  in the CST for healthy controls, MS and NMO........................33  FIGURE 2.5 : Correlations between recruitment curv e slope and EDSS .......................................34   FIGURE 3.1: The opt ic radiation region of interest......................................................................46  FIGURE 3.2: RNFL thickness in individuals with neuromyelitis optica, multi ple sclerosis and  healthy controls..............................................................................................................................49  FIGURE 3.3: Total macular volume in individuals with neuromyelitis optica, multiple sclerosis  and healthy controls......................................................................................................................50  FIGURE 3.4: MWF in the optic radiation of individuals with neuromyelitis optica, multiple   sclerosis and healthy controls.......................................................................................................51    FIGURE  3.5: A) A trend towards a positive correlation between MWF in the optic radiation and the RNFL thickness in the NMO popu lation. B) A positive correlation between MWF in the optic radiation and the TMV in healthy controls ...........................................................................52  FIGURE 3.6: A) Positive correlation between MWF in the optic radiation and RN FL thickness and B) a positive correlation between MWF in the optic radiation and the TMV for all groups  combined........................................................................................................................................53  FIGU RE 3.7: A negative correlation between EDSS score and TMV in the NMO population ....54   FIGURE 4.1: The corpus callosum region of interest...................................................................64  FIGURE 4.2: MWF in the A) corpus callosum, B) cor tico-spinal tract and C) optic radiation of individuals with NMO and MS......................................................................................................66  FIGURE 4.3: No correlation between myelin water fraction (MWF) in the corpus callosum (CC) and lesion volume was observed in grouped analysis (A) or in multiple sclerosis patients separately (B). Similarly no relationship was seen between MWF and lesion volume of the cortico-spinal tract (CST) in the grouped analysis (C) or MS patients (E) and NMO patients separately (G). Similarly, there was no relationship between MWF in the optic radiation (OR) and lesion volume was observed in the grouped analysis (D) o r MS patients (F) and NMO  subjects (H) separately ...................................................................................................................67  x   GLOSSARY   3D   three dimensional  ANOVA  univariate analyses of variance AMT   active motor threshold AQP4   aquaporin-4  CNS   central nervous system CST   cortico-spinal tract DAI   diffuse axonal injury DTI   diffusion tensor imaging ECR   extensor carpi radialis EDSS   expanded disability status scale EMG   electromyography GRASE  gradient echo and spin echo IgG   immunoglobulin G  LGN   lateral geniculate nucleus  M1   primary motor cortex MANOVA  multivariate analysis of variance MEP   motor evoked potential MRI   magnetic resonance imaging MS   multiple sclerosis MS-NON  multiple sclerosis patients without optic neuritis history MS-ON  multiple sclerosis patients with optic neuritis history MTI   magnetization transfer imaging  MVC   maximum voluntary contraction MWF   myelin water fraction NAWM  normal appearing white matter NMO  neuromyelitis optica NMO-IgG  neuromyelitis optica-immunoglobulin G  NMO-NON  neuromyelitis optica patients without optic neuritis history NMO-ON  neuromyelitis optica patients with optic neuritis history OCT   optical coherence tomography ON   optic neuritis OR   optic radiation PDw   proton density weighted RMT   resting motor threshold RNFL   retinal nerve fibre layer SD   standard deviation T1   spin-lattice relaxation time T2   spin-spin relaxation time T2w   T2 -weighted TE   echo time TMV   total macular volume TMS   transcranial magnetic stimulation TR   repetition time TSE   turbo spin echo factor xi   ACKNOWLEDG E MENTS  I want to thank my supervisor, Anthony Traboulsee for the wonderful opportunity you gave me working in this lab and also for your guidance throughout my degree. I would also like to thank the rest of my supervisory committee Roger Tam, Piotr Kozlowski and most especially Sh annon Kolind. You were the best  help anyone could have asked for and thank you for being there for PHHYHU\VWHSRIWKHZD\,KRQHVWO\FRXOGQ’WKDYHGRQHWKLVZLWKRXWDOOWKHVXSSRUWODXJKWHUand guidance you gave me. I would also like to thank Corree Laule my external examiner and t o my examining committee chair Dessa Sadovnick.  Special thanks to Michael Borich who took me under his wings when I first joined the lab. I learned so much from you and appreciated all the time you took out of your work to help me with patience and kindness. I had so much fun working with you and miss you since you left for Atlanta.  Irene Vavasour , thank you for putting up with all my silly questions, and last minute e-mails asking for favours that you always did. I enjoyed your humour very much and had a lot of fun working with you.  Thanks to Annie Kuan who was always there to give me a helping hand whenever I needed it. I loved our little hallway chats and I really felt like I could open up to you.  I would also like to thank Eric Zhao, Evan Chen, Julia  Schubert, Emily Campbell, Nolan Shelley, Chantal Roy-Hewitson , Henry Chen, Jen Moroz, Enedino Hernandez T orres, Vanessa Wiggermann, Sophie Huijskens, Elham Shahinfard, Erin MacMillan, Sandra Meyers, Bretta xii   Russell-Schulz, Alex ander Rauscher and Katrina Mc Mullen for all the fun times and also guidance through my degree.  I would also like to thank Alex Lange for his eminence help with the OCT studies , Yinshan Zhao for all the last minute and (probably tiresome) help you gave me with statistics. I have also had great advice on a variety of topics from Alex Mackay, David Li, Robert Carruthers, Andrew Riddehough and many others.  Thanks to all the amazing MR technologists, including, but not limited to Trudy Harris, Tracey Maile and Alex Mazur. Addition ally, I want to thank everyone at the Brain Behaviour Lab, including Tamara Koren and Lara Boyd. Many thanks to everyone from the MS Clinical Trials including Lara Harvey, Nancy Bogle, Carley Pope, Monica Gill and Heather Short. I would like to thank all the volunteers who participated in these studies. Finally, but not least, my wonderful friends and family, specifically Ramanan Manogaran, Indrani Manogaran and Ponniah Manogaran . Thank you for all the support, love and care you gave me during my studies. IILWZDVQ’WIRUDOORI\RX,ZRXOGQ’WEHZKHUH,DPQRZ7KDQN\RXto Egg Beaterz, you guys are the shining light at the end of the weave. Dreams  are free.        xiii   Dedication     To my family for all the love and support,  I wouldn’t be here without you.  1   Chapter 1  1  INTRODUCTION  A variety of non-invasive structural imaging and functional tools can be used to investigate mechanisms of white matter injury and secondary axonal injury in multiple sclerosis and neuromyelitis optica patients. Magnetic resonance imaging measures of myelin water fraction can be used to quantify myelin, in vivo, in white matter tracts such as the cortico-spinal tract, optic radiation and corpus callosum. Systems electrophysiology, specifically transcranial magnetic stimulation, provides a functional measure of general membrane excitability and measures the integrity of the cortico-spinal pathway. Optical coherence tomography can be used to obtain quantitative measures of retinal nerve fibre layer thickness and total macular volume in the retina. These techniques can be used to study and better understand demyelinating diseases such as multiple sclerosis and neuromyelitis optica.  The introduction reviews the central nervous system and diseases that affect it, particularly, multiple sclerosis (MS) and neuromyelitis optica (NMO). It also provides an overview on multimodal imaging approaches such as myelin water imaging, transcranial magnetic stimulation and optical coherence tomography. 1.1 CENTRAL NERVOUS SYSTEM  The central nervous system (CNS) includes the brain, spinal cord and optic nerve and is involved in integrating sensory information received from the environment and producing a response. The white matter consists primarily of myelinated axons while the grey matter contains neuronal cell bodies. A neuron is a nerve cell that processes and transmits information via electrical and chemical signals. The axons of these neurons are covered by a fatty layer called the myelin 2   sheath that speeds up conduction (approximately 15 times faster than unmyelinated axons ) of electrical information (Martin et al., 1992) . Demyelination results in damage to the myelin sheath which is caused by inflammation-inducing multifocal plaques in the brain, optic nerve, and spinal cord (Frohman et al., 2006 ; Figure 1.1 ). This demyelination leads to nerve signals being disrupted or slowed down. Both white and grey matter contain glial cells such as astrocytes or microglia which have a variety of supporting functions. In healthy individuals, astrocytes provide both pro- and anti-inflammatory molecules, and has been known to play a role in blood-brain barrier integrity (Sofroniew and Vinters, 2010).    Figure 1.1 : The effects of demyelination on axonal conduction. The left image depicts a healthy axon (orange) with normal signal conduction due to the presence of myelin (green). The right image depicts an axon damaged by demyelination resulting in a disruption or slowing down of the signal (Frohman et al., 2013).  3   1.2  MULTIPLE SCLEROS IS  CNS demyelination is the product of damage to the myelin sheath or oligodendroglia and can be produced by mechanisms including toxicities, infection, autoimmunity and metabolic disturbances (Martin, McFarland & McFarlin, 1992). The most prevalent demyelinating disease affecting the CNS is MS. MS is a serious demyelinating neurological disease that results in white matter lesions affecting mainly the myelin sheath and the myelinating cells such as the oligodendrocytes (Frohman et al., 2006; Fuhr and  Kappos, 2001) . Axonal damage may also be associated with these lesions and usually elicits more permanent disability (Fuhr and Kappos, 2001) . Fortunately, remyelination can occur and clinical recovery can be tremendous, especially in early stages of MS. Symptoms of MS include movement disability, pain, and disruption of bodily functions like bowel incontinence (Sergott et al., 2007) . The Kurtzke Expanded Disability Status Scale (EDSS) is a common method for quantifying disability in demyelinating diseases such as MS.  The score is based on neurological testing and integrity of the functional systems however the scale places a lot of emphasis on the ability to walk.  MS is diagnosed most often in young adults between the ages of 20 to 45 years and affects women more frequently than men (Goldenberg, 2012). It is a chronic debilitating disease that can reduce the lifespan of a patient by seven to eight years on average (Trapp and Nave, 2008). MS is a highly prominent disorder in Canada, and although several options exist for disease modifying treatments without a cure, many patients remain disabled for their entire life (Kallmann et al., 2006). Patients with MS follow their own disease course with great inter -individual variability; the rate of progression being different from patient to patient (Serbecic et al., 2011; Fuhr and Kappos, 2001). At present, disease course is used to define the clinical 4   subtypes of MS. While there is evidence implicating both environmental and genetic factors in the epidemiology of MS, the cause of the disorder is still unclear (Noseworthy, Lucchinetti & Rodriguez, 2000; Martin, McFarland & McFarlin, 1992).   1.2.1 MS Immunopathology Tissue injury and demyelination in MS is thought to be mediated by T-cell activity (Wingerchuk et al., 2001 ). T-cells are white blood cells, specifically lymphocytes, involved in identifying pathogens and cell abnormalities as a part of the adaptive immune system. The adaptive immune system provides antigen (substances that provokes the adaptive immune response) specific protection against foreign organisms. The role of cytotoxic (CD8+ ) and helper (CD4+) T-cells in MS is controversial. CD8+ T -cells are typically involved in the process of eliminating pathogens in healthy individuals. It is thought that CD8+ T -cells have a tissue-damaging role in MS, however they may also be involved in suppressing CD4+ T -cell activity (Siffrin et al., 2010). CD4+ T -cells assist other white blood cells in immunological processes. However, in MS, CD4+ T-cells are most likely involved in the maintenance of the autoimmune response against the myelin sheath. Additionally, microglia that usually have a protective role on neuronal compartments, are shifted towards a pro-inflammatory phenotype in MS resulting in the production of cytokines (such as tumor necrosis factor-∝ and interleukin 1 -ȕDQGQHXURWR[LQVsuch as nitrogen oxide, and proteolytic enzymes which have been shown to contribute to axonal injury (Costello, 2013; Siffrin et al., 2010).   The mechanisms of axonal injury are largely unknown; however, it is likely a secondary effect of demyelination and not entirely due to it (Wingerchuk et al., 2001). Axonal damage is strongly 5   related to the presence of inflammatory cells such as T-cells in MS lesions, but is also present in the absence of active demyelination, suggesting that axonal loss may occur independent of demyelination in MS (Saffrin et al., 2010) . Wallerian degeneration is also known as anterograde degeneration and results in degradation to the axon distal to the site of injury (Kanamori et al., 2012). Fibres passing through a lesion may result in Wallerian degeneration as a consequence of axonal transections (Trip et al., 2006). Therefore, axonal damage may occur locally within a lesion or distally as a result of Wallerian degeneration (Trip et al., 2006; Ciccarelli et al., 2005).    1.2 .1. 1  MS Immunopathology Model Peripherally activated T lymphocytes interact with endothelial cells  of blood vessels by tethering and rolling (Figure 1.2; Wingerchuk et al., 2007) . This initiates movement of T lymphocytes into the CNS parenchyma (Figure 1.2 ) and proteinases like MMP -9 increase  migration (Wingerchuk et al., 2007) . CD4+ T -cells are activated when their T-cell receptor interact with the class II major histocompatibility class (MHC  II) on antigen-presenting cells (Lassmann et al., 2007; Wingerchuk et al, 2007 ). The secretion of cytokines and chemokines (involved in cell signalling) leads to the movement of leucocytes like B lymphocytes, monocytes and CD8+ T -cells into the CNS (Lassmann et al., 2007) . CD8+ T -cells and macrophages (developed from monocytes) are involved in the proinflammatory cascade that results in demyelination and axonal injury (Saffrin et al., 2010; Wingerchuk et al., 2007) . Local antibody production through proliferation of B lymphocytes may increase complement activity and axonal injury (Wingerchuk et al., 2007) .   6    Figure 1.2 : Current model of MS immunopathology . T lymphocytes interact with endothelial cells resulting in CNS inflammation. Helper T -cells are reactivated by unknown antigens presented on antigen-presenting cells. Cytokine and chemokine secretion results in leucocytes migrating into the CNS. Cytotoxic T-cells and macrophages are involved in proinflammatory cascade resulting in demyelination and axonal injury. B lymphocyte proliferation increases complement activity and axonal injury. LFA -1/V LA -4: ĮLQWHJULQV, VCAM -1/ ICAM-1 : endothelial receptors, CNS: central nervous system, MMP -9: proteinases, CD4: T helper cells, TCR: T -cell receptor, MHC: class II major histocompatibility class, APC: antigen -presenting cells, B: B lymphocytes, CD8: cytotoxic T cells, MØ : macrophage , CSF: cerebrospinal fluid. (Wingerchuk et al., 2007).  7    1.3 NEUROMYELITIS OPTICA  NMO is an inflammatory, demyelinating disease of the CNS (Wingerchuk et al., 2007). The median age at NMO onset is around 37 years, although disease can occur as early as infancy or in elderly people (Weinshenker, 2014). There is a marked female preponderance among patients with NMO (4:1  ratio), and it also disproportionately affects African-Americans and Asians. NMO is a rare relapsing disease with predilection for damage to the optic nerve ± optic neuritis (ON) and spinal cord ± transverse myelitis (Jarius et al., 2008) . Symptoms range from mild sensory disturbances to complete transverse myelitis with tetraplegia or paraplegia, sensory impairments and bladder±bowel dysfunction.   For many years, NMO had been considered under the umbrella term of MS; it was previously termed optico-spinal multiple sclerosis and also Asian multiple sclerosis. Both MS and NMO show similar clinical and radiological features, but treatment and prognosis differ significantly (Love, 2006; Wingerchuk et al., 2007). NMO patients are usually treated with immunosuppressants while MS is treated with immunomodulatory treatments (Wingerchuk et al., 2007). Unlike MS, spontaneous remission of neurological disability is rare in NMO and usually presents with a build-up of irreversible deficits and a more rapid disease progression (Jarius et al., 2008). Additionally, MS has relative, but not absolute, preservation of axons with more direct damage to myelin compared to NMO (Naismith et al., 2009). Misdiagnosis of NMO patients is common and being untreated or treated inappropriately can result in worsening of symptoms and progression.   8   1.3.1  NMO Immunopathology 1.3.1.1 Humoral Immunity  Humoral immunity  is the aspect of adaptive immunity that is mediated by macromolecules ± as opposed to cell-mediated immunity ± found in extracellular fluids such as secreted antibodies and complement proteins (Tan and Coussens, 200 7). It refers to antibody production and the accessory processes that accompany it, including T helper cell activation and cytokine production (Mosmann and Sad, 1996). B cells are triggered when enco untering its matching antigens resulting in proliferation and differentiation of activated B cells that increase antibody production (such as NMO-immunoglobulin G) and memory cell production. These antibodies  which are involved in identifying and neutralizing foreign objects  bind to antigens (such as aquaporin-4 ) resulting in functions like pathogen neutralization and classical co mplement activation (Figure 1 .3 ). The complement system helps the ability of antibodies and phagocytic cells to clear pathogens (Carroll, 2004). It i s a part of the innate immune system, however, it can be recruited and brought into action by the adaptive immune system. When triggered, proteases (enzymes for breaking down proteins) in the system cleave specific proteins to release cytokines and initiate an amplifying cascade of further cleavages. The end result of this activation cascade is a massive amplification of the response and activation of the cell-killing membrane attack complex. Humoral autoimmunity has a significant role in NMO pathogenesis with patterns of tissue inflammation including complement activation, leukocyte infiltrates and hyalinised vessels in active lesions which is distinct from MS pathological features (Lucchinetti et al., 2002; Wingerchuk et al., 2007).   9    Figure 1.3: Humoral immunity. Effector T -helper cells from cellular immunity activates B cells that bind to antigens (such as AQP4). These B cells mature and proliferate to activated cells such as the plasma cell. The plasma cell secretes antibodies specific to the antigen (such as NMO-IgG)  which then bind to the antibodies neutralizing them.  Figure by P. Manogaran.    1.3.1.2 Immunoglobulin G  Immunoglobulin G  (IgG) is an auto antibody which is a protein that is targeted by the immune system usually as a result of an autoimmune disease caused by genetics and environmental factors. IgG is the most abundant antibody isotype found in the serum of the blood and are synthesize d and secreted by plasma B cells (Jarius et al., 2008). It is involved in complement activation (drills holes in cells), agglutination (attaches to multiple cells making a clump ± blood typing), opsonisation (candy coating of bacteria for phagocytosis) and neutralization (coats a toxin or virus inactivating it). NMO specific serum autoantibodies are collectively known as NMO-IgG , which localizes to the blood -brain barrier and is not present in patients with MS (Lennon et al., 2004).  1 0    1.3.1.3 Aquaporin-4  In a breakthrough study, Lennon and colleagues (2005) showed that NMO -IgG binds selectively to aquaporin-4 (AQP4) water channels. AQP4 is the most abundant water channel in the CNS located on astrocytes (Jung et al., 1994). Each AQP4 monomer consists of 6 membrane -spanning ∝-helices with both termini located intracellularly; these monomers constitute AQP4, a hometetramer (Jarius et al., 2 008 ). AQP4 is predominantly expressed within the brainstem, optic nerves, and gray matter of the spinal cord (Jarius et al., 2008). The identification of AQP4 as the target antigen of NMO-IgG, in parallel with the demonstration of a striking loss of AQP4 a s a distinctive finding in acute spinal cord lesions, strengthened the case of the involvement of this antibody in the pathogenesis of NMO (Lennon et al., 2005). The discovery of NMO-IgG and the subsequent identification of AQP4 as its target antigen, make s NMO the first inflammatory demyelinating disorder of the CNS to have a defined autoantigen. The current serological test for NMO-IgG is 58 -76 % sensitive and 85 -99% specific to NMO (Lennon et al., 2005; Takahashi et al., 2007).  However, even with the most  sensitive assay, 10 -25% of patients clinically diagnosed with NMO are seronegative for NMO-IgG (Wingerchuk et al., 2007).   1.3.1.4 NMO Immunopathology Model Recently, in vitro and in vivo studies support a pathogenic role for NMO-IgG ; furthermore, are beginning to define the cascade of events that lead to inflammation, demyelination, and ultimate tissue destruction (Wingerchuk et al., 2006; Lennon et al., 2005 ). Currently, the immunising event in NMO is still not known; however, the peripheral immunoglobulin pool does contain NMO-IgG. Immunoglobulin has limited access to the CNS parenchyma , but it is believed that it 1 1   can enter through endothelial transcytosis (transported across the membrane via vesicles), or at damaged areas of the blood-brain barrier. The extracellular domain of the AQP4 is accessible to any NMO-IgG entering the region because it is presented on the astrocytic foot process (Jarius et al., 2008). In the CNS, AQP4 regulates water, glutamate and potassium transport er and down regulation of AQP4 happens through NMO-IgG antigenic modulation  (Jacob et al., 2013) . This occurs in parallel with down regulation of the glutamate transporter resulting in disruption of water and glutamate homeostasis (Jacob et al., 2013). Add itionally, complement activation increases blood-brain barrier permeability and causes a massive infiltration of leukocytes. Complements form membrane attack complexes (transmembrane channels formed on the surface), causing cytolytic injury due to osmotic imbalance resulting in hyalinisation and irregular thickening of blood vessels in NMO lesions (Lu cchinetti et al., 2002). The cellular influx combined with complement activation causes demyelination, severe neuronal injury and necrosis (Jarius et al., 2008 ; Naismith et al., 2009; Lucchinetti et al., 2002; Figure 1. 4 ). 1 2    Figure 1.4: Current model of NMO immunopathology. The peripheral IgG p ool (blue) contains NMO-IgG (red ) and has limited access to the CNS parenchyma. Astrocyte foot processes make the extracellular domain of the AQP4 channels accessible to NMO -IgG. Complement activation increases blood-brain barrier permeability result in an infiltration of immune cells (eosinophils [EOS], neutrophils [N], B lymphocytes [B], monocytes and macr ophages [M ∅]) . The calcium cell influx results in demyelination, severe neuronal injury and necrosis. Cytolytic injury by assembly of membrane attack complex (MAC) results in hyalinisation of vessels. CSF: cerebrospinal fluid. (Wingerchuk et al., 2007)    1.4 MAGNETIC RESONANCE IMAGING   Magnetic resonance imaging (MRI) provides a powerful objective approach to characterization and quantification of the evolving pathology in patients with MS and NMO. MRI is a valuable 1 3   tool in understanding the nature of the disease process, and as a tool in clinical trials to provide rapid and objective means of assessing the influence of treatment on disease progression (Barnett et al., 2013; Goldenberg, 2012).  It is well-recognized that MRI has great sensitivity and provides significant information in assessing disease progression (Kandler et al., 1991;  Arnold and Matthews, 2002; Fuhr et al., 2001; Beer et al., 199 5; Filippi et al., 199 9 ; Filippini et al., 1994). Conventional T1 -weighted MR imaging provides anatomical information about the CNS and is useful for identifying more chronic lesions (Fazekas et al., 1999).  While lesion detection with conventional MRI has been a tremendous aid in MS and NMO diagnosis, it lacks pathologic specificity, particularly the severity of axonal and myelin loss (Barnett et al., 2013 ; Law et al., 2003 ). Complex changes in the brains of patients with these demyelinating diseases have not been fully characterized because of the relative subjective evaluation that is possible with conventional MR imaging (Fox et al., 2011) . This suggests that there is subclinical abnormalities in these diseases that extend beyond what is visible on conventional MRI (Liang et al., 2012).   Normal appearing white matter (NAWM) UHIHUVWRZKLWHPDWWHUWLVVXHWKDWDSSHDUV‘QRUPDO’RQconventional MRI tools, despite the concealed damage still occurring in the CNS (Werring et al., 1999) . This, in part, explains the weak correlation found between conventional MR imaging measures and disease progression (Barkhof, 2002 ). Additionally, histological and advanced MRI studies have found significant abnormalities in the NAWM such as changes to the blood-brain barrier, axonal injury, and an increased presence of immune cells (Fu et al., 1998; Allen and McKeown, 1979).   1 4   The limitation of conventional MRI has led to years of research and the development of advanced MRI measures such as magnetization transfer, diffusion tensor imaging and myelin water imaging (Neema et al., 2007). Advanced MR imaging is more objective and potentially more sensitive to subtle brain changes than conventional MR imaging techniques (Fox et al., 2011) . Quantitative measures obtained from these techniques provide a useful tool to study diseases in research and clinical settings. Recently, advanced MR imaging techniques have led to remarkable success in identifying brain abnormalities associated with other disorders like 3DUNLQVRQ’VDQG$O]KHLPHU’V (Seppi et al., 2005; Bozzao et al., 2001 ). These techniques have evolved to examine injury with greater specificity and sensitivity in lesions and normal appearing brain tissue (Ceccarelli et al., 2012). These findings suggest that quantitative MR imaging may offer sufficient sensitivity and accuracy to characterize the disease progression in demyelinating disorders objectively (Fox et al., 2011; Steen et al., 1997).   1.4.1 Myelin Water Fraction Myelin damage has become easier to observe in patients with MS and NMO due to the novel development of the T2 relaxation method (Mackay et al., 2009).  Myelin helps facilitate rapid and coordinated action potential conduction in white matter tracts, and myelin water fraction (MWF) may be employed to quantitatively monitor demyelination in MS and NMO. Short T2 relaxation arises from water trapped between membranes of the myelin sheath (Mackay et al., 1994). Myelin water imaging using 3D multi -echo T2 relaxation measurements can distinguish between water compartments in a single voxel. Water more tightly confined within the myelin bilayer have shorter T2 times (below 30ms) and water less tightly confined in the intra/extracellular spaces have longer T2 times (approximate ly 40 to 500ms; Figure 1. 5 ). MWF is defined as a 1 5   signal with T2 below 30 milliseconds divided by the total signal and is a valid measure of the P\HOLQGHQVLW\LQDSDWLHQW’VEUDLQ (Moore et al., 2000 ; Laule et al., 2006; Laule et al., 2008; Figure 1. 5 ). With this measurement we can better understand the effects of demyelination and potential recovery.   Figure 1.5: Myelin water fraction. Water that is more tightly confined in the myelin bilayer have shorter T2 while water less confined in the intra/extracellular spaces have longer T2. MWF is defined as a signal with T2  below 30 milliseco nds divided by the total signal. Figure adapted from slides by C. Laule and S. Kolind.    1 6   1.5 CORTICO -SPINAL TRACT   The cortico-spinal tract (CST) is a part of the pyramidal tract involved in controlling motor functions of the body, more specifically voluntary movement. It consists of upper motor neurons that travel down from the motor cortex to the white matter of the spinal cord and innervate at the vertebral level of the muscle. These nerve fibres originate from pyramidal cells ± the primary neural cell type of the CST ± in layer V  of the cerebral cortex and the axons form the bulk of the CST. Studies have found lower nerve fibre density and total number of fibres in the CST in patients with MS and NMO compared to healthy controls (Deluca et al., 2004; Ganter et al., 1999 ; Lin et al., 2006 ).   1.5.1 Transcranial Magnetic Stimulation Over the last couple decades, transcranial magnetic stimulation (TMS) has become an established method for providing information about CNS functional abnormalities. TMS is a unique tool to study the CST because of its ability to stimulate the brain and measure the response (Curra et al., 2002). TMS  is a non-invasive, safe and painless method of activating the human motor cortex and accessing the integrity of the central motor pathways (Groppa et al., 2012 ). A figure-of-eight coil connected to two single pulse stimulators is used to deliver focal stimulation to the primary motor cortex area (Figure 1. 6 ). The electric pulse generates a magnetic field that can penetrate the skull to reach the brain and results in secondary ionic changes in the motor cortex (Kobayashi and Pascual -Leone, 2 00 3). These ionic changes cause  action potentials that travel down the CST to the contralateral muscles located in the forearm, and this response (or motor evoked potential, MEP) can be measured using surface electromyography ( EMG, Siebner and Rothwell, 2003; Kobayashi and Pascual -Leone,  2003 ).  1 7    Figure 1.6: Diagram demonstrating TMS stimulation of the motor cortex resulting in a motor evoked potential that can be read using EMG surface electrodes placed on the forearm muscle. Figure by P. Manogaran.   1.5.1.1 Motor Threshold Motor threshold is a TMS measure used to assess the general membrane excitability at rest (resting motor threshold, RMT) or while the muscle is active (active motor threshold, AMT). It is defined as the lowest intensity needed to elicit MEPs of more than 5 0  ȝ9SHDNWRSHDNDPSOLWXGH for at least 50% of the trials. (Curra et al., 2002 ). Motor threshold is usually lower when the muscle is active compared to at rest, and also lower in muscles that are more distal from the point of stimulation (Chen et al., 2008; Curra et al., 2002). Motor threshold provides information on the efficacy of a chain of synapses from the cortical neurons to the muscle and is often increased or decreased in diseases that affect the CST like in MS, or stroke (Kobayashi  and Pascual -Leone, 2003 ).    1 8   1.5.1.2 Recruitment Curve TMS can be used to obtain a recruitment curve which examines the MEP response as a function of stimulus intensity and is a reliable index of functional integrity in the CST (Siebner and Rothwell, 2003) . In healthy individuals, neurons that are intrinsically less excitable or further away from the point of stimulation are more likely to be activated at higher stimulation intensities whereas at low levels, only easily excitable neurons are activated resulting in a signature “6”VKDSHGUHFUXLWPHQWFXUYH+DOOHW; Figure 1. 7 ). These recruitment curves may look abnormal in neurological diseases such as stroke (Jouvin et al., 2006; Ward et a., 2006), hereditary spastic paraplegia (Jørgensen et al., 2 00 5) or dystonia (Hallet, 20 00). Recruitment curves are considered a more reliable measure of CST excitability compared to motor threshold because more values are obtained thus it is less susceptible to variability (Ward et al., 2006).    Figure 1.7: Recruitment curve of a healthy individual. The curve is fit a three parameter sigmoidal function that begins with a steeply rising slope and a final plateau. (Jouvin et al., 2006)   1 9   1.6 VISUAL PATHWAY  The visual system can be used as a model of potential differences between MS and NMO. The retina is the light sensitive layer of the eye that contains cones and rods important for high-definition color vision and dim light conditions respectively. The macula is the oval-shaped highly pigmented yellow spot near the center of the retina where the retinal ganglion cell bodies originate. The fovea, located in the center of the macula, is a small pit that contains the largest concentration of cone cells in the eye. Scanning of the macula to obtain the total macular volume (TMV) images the retinal ganglion cell bodies and represents a potential method for capturing primarily neuronal degeneration (Ratchford et al., 2009; Costello, 2013).   The retinal nerve fibre layer (RNFL) is a unique CNS structure because it lacks myelin, and thinning of the RNFL around the optic nerve head, where it is the thickest, is representative of axonal loss (Talman et al., 2011; Trip et al., 2005). The parts of the axons that are beyond the retina are myelinated (Frohman et al., 2008). The optic nerve, also known as cranial nerve II, is composed of glial cells and the axons of retinal ganglion cells and relays visual information from the retina to the brain. Retinal ganglion cells represent the output CNS neurons for the deeper retinal cells such as the photoreceptors, horizontal cells and bipolar cells . They perform the initial steps for visual processing (Costello, 2013). Approximately 90% of these retinal ganglion cell axons synapse to neurons in the lateral geniculate nucleus (LGN) locate d in the thalamus, and these projections form the optic radiation (OR) that relays information to the visual cortex (Reich et al., 2006; Costello, 2013). The anterior visual pathway include s pre-geniculate structures such as the retina, and the posterior visual pathway includes post-geniculate structures such as the optic radiations (Figure 1. 8 ). The optic radiation in the visual pathway is a frequent 2 0   site for inflammation and demyelination (Hornabrook et al., 1992; Reich et al., 2009) . Axonal degeneration within these structures likely leads to permanent visual loss; since the optic radiation carries visual information, injury to that tract can contribute to visual dysfunction.    Figure 1.8. The visual pathway. Retinal ganglion cell bodies originate in the macula within the fovea, and the retinal nerve fiber layer is composed of these unmyelinated axons that converge to form the optic nerve. The axons of the retinal ganglion cells synapse to neurons in the lateral geniculate nucleus (LGN) and these proje ctions form the optic radiation which relays information to the visual cortex. The anterior visual pathway includes pre-geniculate structures like the retina, and optic nerve, whereas the posterior visual pathway includes post-geniculate structures such as the optic radiations. Figure by P. Manogaran.   1.6.1 Optic Neuritis  Optic neuritis (ON) is an inflammation of the optic nerve that usually results in subacute monocular visual loss (Toosy et al., 2014). E arly signs include reduced vision and contrast 2 1   acuities and pain due to eye moment is common but usually mild (Shams and Plant, 2009 ). There are two forms of ON, retrobulbar optic neuritis, when the orbital portion of the nerve is involved, and papillitis (or intraocular optic neuritis), when the head of the optic nerve is involved (Toosy et al., 2014). ON patients can be categorized as either typical (usually presents with normal or swollen optic disc with mild to moderate vision loss and unilateral symptoms) or atypical (usually presents with severe optic disc swelling, periocular pain and vision loss), furthermore, they can be classified as those with systemic disease or those without (Toosy et al., 2014).   The optic nerve lesions are predominantly T-cell mediated with possible B-cell involvement, and are pathologically similar to MS brain lesions (Shams and Plant, 2009; Toosy et al., 2014). Good visual recovery occurs in about 90% of patients and usually occurs within the first few weeks ; however, persistent residual symptoms such as decreased visual acuity, color vision, contrast sensitivity, pupillary reaction and stereopsis may occur (Toosy et al., 2014). Vision  loss is often incomplete, likely due to persistent axonal loss and demyelination (Shams and Plant, 2009 ). Recovery usually involves some level of remyelination and redistribution of sodium channels that improve conduction, but results in more vulnerable axons (Toosy et al., 2014).   Acute ON usually results in an initial increase in RNFL thickness  with the optic nerve swelling, and the subsequent reduction of RNFL  thickness after significant axonal loss (Toosy et al., 2014).     2 2   1.6.2 Optical Coherence Tomography Optical coherence tomography (OCT) is a non-invasive ocular imaging tool used to generate high resolution cross-sectional images of the retina (Costello et al., 2006). The retina provides a good model to study virtually all manifestation of brain tissue injury in MS, except demyelination; therefore, OCT can be used to study the anterior visual pathway to better understand MS pathology. OCT imaging is analogous to ultrasound B-mode imaging but uses infrared-light instead of sound. From multiple axial scans (A scans) at different transverse locations, a two-dimensional, cross-sectional image can be obtained. The OCT measurements of the RNFL are very sensitive and even small changes can be quantified (Trip et al., 2006). OCT has been proven to be a reliable and sensitive tool in measuring the integrity of the damaged neurons, as well as being predictive of clinical outcomes (Petzold et al., 2010; Frohman et al., 2008; Noval et al., 2011). They also offer an accurate analysis of the relation between the structural damage and the disability progression, while maintaining high reproducibility (Petzold et al., 2010). OCT has been demonstrated to be a valuable technique for detection and monitoring of a variety of macular diseases (Hee et al., 1995, Ophthalmology; Hee et al., 1 995, Arch Ophthalmology; Hee et al., 1996 ) as well as glaucoma (Schuman et al., 1995, Arch Ophthalmol; Schuman et al., 1995, Curr Opin Ophthalmol; Schuman et al., 1996).  In MS and NMO, lesions resulting in demyelination of the optic nerve can cause serious visual problems which can be detected early using an OCT analysis of the RNFL and TMV (Petzold et al., 2010).   1.7 OVERVIEW OF THESIS  The first objective is to functionally measure CST integrity using TMS, and structurally measure myelin in vivo using myelin water imaging to investigate any possible relationships between 2 3   these measures in MS and NMO. We hypothesized that individuals with NMO would have abnormal recruitment curves and lower MWF in the CST because of greater lesion involvement expected in this region compared to individuals with MS and healthy controls.   The second aim looked at quantifying myelin in the optic radiation of MS and NMO patients and comparing it to measures of the retina using OCT. We hypothesized that OCT measures of  the retina (RNFL, TMV) will be severely affected in NMO patients compared to MS, and to a greater degree in patients with ON history. Additionally, we hypothesize that ON will result in retrograde degeneration towards the retina and Wallerian degeneration in the optic radiation. Finally, we expect an association between the OCT measures of the anterior visual pathway and myelin water imaging of the posterior pathway because damage to one part of the visual pathway may cause alternations in another part of the visual system.   The final aim is to look at the relationship between lesion load in white matter tracts and NAWM MWF in the CC, CST, and OR of patients with MS and NMO.  If there is a relationship between the measures then we would expect to see only focal damage with some downstream damage due to lesions, whereas a lack of relationship would suggest that there are processes involved in the damage caused to the NAWM, independent of the lesions.  2 4   Chapter 2  2  CORTICO-SPINAL TRACT INTEGRITY  MEASURED USING TRANSCRANIAL MAGNETIC STIMULATION AND MAGNETIC RESONANCE IMAGING IN NEUROMYELITIS OPTICA AND MULTIPLE SCLEROSIS   2.1 INTRODUCTION   Neuromyelitis optica (NMO) is a rare, autoimmune disorder of the CNS which has overlapping clinical features with MS including optic neuritis and transverse myelitis (Jacob et al., 2013). NMO is characterized by episodes of inflammation and damage to astrocytes with secondary injury to myelin and axons (Love, 2006) . NMO relapses are typically more severe than MS and may lead to more extensive axonal loss (Wingerchuk et al., 2007) .   Transcranial magnetic stimulation (TMS) is a non-invasive method to investigate cortical excitability and functional integrity of central motor pathways in the cortico-spinal tract (CST, Siebner and Rothwell, 2003 ). TMS has been studied in numerous disease models such as stroke, movement disorders, and spinal cord injury, demonstrating a high sensitivity in quantifying dysfunction in the cortico-spinal tract (Nascimbeni et al., 2005; Di Lazzaro et al., 1999) . In particular, TMS can be used to obtain a recruitment curve that evaluates the cortical physiology and detects changes in cortical output maps (Siebner and Rothwell, 2003; Jø rgensen et al., 2005) . Motor evoked potentials (MEPs) elicited across a range of stimulus intensities are used to produce input-output recruitment curves (Siebner and Rothwell, 2003) . The recruitment curve is used as a measure of the functional integrity of the CST and provides information on the physiological strength of the cortico-spinal projections (Devanne et al., 199 7) .  2 5    Conventional MRI scans are sensitive for detecting both NMO and MS lesions but lack pathologic specificity, particularly for the severity of axonal and myelin loss (Miller et al., 1998) . Myelin water imaging is an advanced MRI technique that can distinguish between water compartments within a single voxel based on differences in T2 relaxation (Mackay et al., 1994) . The signal fraction with short T2 times (10 -40 ms) arises from water trapped between myelin membranes and is known as the myelin water fraction, MWF (Mackay et al., 1994; Whittall et al., 1997; Moore et al., 200; Laule et al., 2006; Laule et al., 2008 ). MWF has been validated histologically as a measure of myelin (Moore et al., 2000; Laule et al., 200 6; Laule et al., 2008). As a consequence of axonal damage, changes in MWF are also expected in NMO white matter, and have not been previously reported.   While both MS and NMO can present with transverse myelitis, NMO symptoms are usually more severe and thus a better understanding of the structural and physiological effects of cortico-spinal involvement may help guide treatment or rehabilitation. The current preliminary investigation utilized TMS -based measures of cortical excitability and MWF to evaluate differences in structural and functional status of CST in individuals with NMO or MS, as compared to normal healthy individuals. We hypothesized that individuals with NMO would have abnormal recruitment curves and lower MWF in the CST because of greater lesion involvement expected in this region compared to individuals with MS and healthy controls (Jacob et al., 2013; Kim et al., 2012) .    2 6   2.2 METHODS  2.2.1 Participants  Ten participants with MS (median EDSS=  2,0 (range 0.0 ±6.0), mean age=42  years, SD= 9 y ears, mean disease duration=8.7  years, standard deviation (SD)= 5 years, 3 males and 7 female s), 10 individuals with NMO (median EDSS= 2.5  (range 2.0 ±6.0), mean age= 43  years, SD= 11 years, mean disease duration=7.4  years, SD= 4 years, 3 males and 7 females), and 1 0 healthy controls (mean age= 42  years, SD= 10 years, 2 males and 8 females) were age and gender matched for this study (Table 2. 1). All but one participant with MS was on glatiramer a cetate (Copaxone®) treatment at the time of testing. Four participants with NMO were on mycophenolic acid (CellCept®), three on azathioprine (Imuran®), and 1 participant each on either interferon beta -1a (Rebif 44®), mitoxantrone or rituximab. No participa nts were on steroid therapy at least one month prior to entering the study. The University of British Columbia Clinical Research Ethics Board approved all study procedures.   Technique  Measurement  Control  MS  NMO  TMS RMT (%)  4 2.00 (8.27)  47.80 (9.24)  49.75  (8.89)   AMT (%)  3 3.75 (5.08)  40.55 (8.27)  39.85 (9.08)   5HFUXLWPHQW&XUYH6ORSHȝ9‚ 2 1.93 (8.52)  23.61 (11.06)  13.60 (5.68)  MRI 0:)&67‚Á 0.199 (0.017)  0.193 (0.016)  0.165 (0.022)  Clinical EDSS  - 2.0 (0.0 -6.0 ) 2.5 (2.0 -6.0 )  Disease Duration (years) - 8.65 (4.9)  7.4 (4.1)   Age (years) 4 2.3 (10.3)  42.4 (8.9)  43.2 (11.2)   Sex (M:F)  2 :8  3:7  3:7   Table 2.1:  TMS/CST MWF  Subject Characteristics and Results. The mean of each measure is presented except for sex (ratio of male to female) and EDSS (median). Standard deviation is in parentheses except for EDSS (range) . MS: Multiple sclerosis; NMO: neuromyelitis optica; TMS: transcranial magnetic stimulation; RMT: resting motor threshold; AMT: active motor threshold; CST: cortico -spinal tract; MWF: my elin water fraction; EDSS: expanded disability status scale.  6LJQLILFDQWPHDVXUHPHQWVLQGLFDWHGDVVXFK‚S06FRPSDUHGWR102S102FRPSDUHGWRFRQWUROVÁS102FRPSDUHGWRFRQWUROV  2 7   2.2.2 TMS Protocol and Analysi s  A figure-of-eight shaped hand-held coil connected to two Magstim 200 2  single pulse stimulators via a Bistim 2  connecting module (Magstim Co., Wales, UK) was used to deliver focal stimulation over the primary motor cortex (M1). Surface electromyography (EMG) was collecte d over the extensor carpi radialis (ECR). All TMS assessments were performed bilaterally  (Figure 1.5) .   2.2.2.1 Motor Threshold  To assess general neuronal membrane excitability, resting motor threshold (RMT) were determined as the percentage of maximum stimulator output required to elicit a MEP in the contralateral ECR of   • ȝ9LQDWOHDVWRXWRIWULDOV (Edwards et al., 2013) . Additionally, the active motor threshold (AMT) was determined as the percentage of stimulator output required to elicit an MEP of  • ȝ9LQRXWRIWULDOVZKLOHSDUWLFLSDQWVPDLQWDLQHGDQisometric grip force contraction ± 20% max imum voluntary contraction, MVC ( Zabukovec et al., 2013) .  2.2.2.2 Recruitment Curve  Single TMS pulses were delivered over M1 at six intensitie s (105, 115, 125, 135, 145, and 155% AMT). Six trials were collected at each intensity level while the subject performed an active grip force contraction (20% MVC; Zabukovec et al., 2013) . Trial by trial MEP maximal amplitudes were identified and trials wiWK0(3DPSOLWXGHVRIȝ9ZHUHH[FOXGHGRIDOOWULDOVThe average MEP maximal amplitude for each intensity level was plotted as a function of %AMT. A three parameter sigmoidal function was fit to the data to visualize the group 2 8   differences (Carroll et al., 2001) . To obtain the slope value, a linear equation was fit to the data and the slope of this line was used to compare responses between groups (Zabukovec et al., 2013) .  2.2.3 MRI Protocol and Analysi s  MRI scans were performed on a Philips 3.0T Achieva scanner (Best, the Netherlands). An axial 3D T 1 -weighted turbo field echo with repetition time (TR) of 7.8ms, echo time (TE) of 3.6ms, and 1mm x 1mm x 1mm voxel size was acquired for segmentation of the tr acts of interest. An axial combined gradient echo and spin echo (GRASE) T 2  relaxation with 32 echoes sequence, TR = 1000ms, TE 1 -32  = 10 -320 ms, echo spacing = 10ms, 1  mm x 1  mm x 5  mm voxel size reconstructed to 1 mm x 1 mm x 2.5 mm, sensitivity encoding f actor = 2 and an echo planar imaging factor = 3 was acquired for MWF estimation  (Prasloski et al., 2012) . The scanned volume consisted of a 10cm slab measured from the top of the brain caudally.   The signal decay curve obtained by the T2  relaxation sequence was modelled by multiple exponential components and the T2  distribution was estimated using non-negative least squares with the extended phase graph algorithm (Whitall et al., 1989 ; Prasloski et al., 2012) . MWF in each image voxel was computed as the ratio of the area under the T2  distribution with times of 10 -40ms to the total area under the distribution  (Laule et al., 2006). This analysis was performed using in-house software on Matlab® 2012 (MathWorks, Natick, Massachusetts, U.S.A).   3D T 1  images of each subject were registered to T2  GRASE images using FMRIB's Linear Image Registration Tool (FLIRT ; Smith et al., 2004 ). Automated segmentation of the CST only 2 9   included white matter and was performed by registration of the corresponding JHU ICBM atlas region to the T2  GRASE images using a combination of FLIRT and FNIRT (Figure 2. 1 ; Andersson et al., 2007). T hese regions were then manually edited to ensure only white matter was included and overlaid on the MWF images to obtain the mean MWF for each region.    Figure 2.1 : The cortico-spinal tract region of interest (in red) on T1 -weighted image in an individual with neuromyelitis optica. A) coronal view B) axial view C) sagittal view.   2.2.4 Statistical Analysis  2.2.4.1 Independent MRI and TMS Analysis  Statistical analysis was performed using R: a language and environment for statistical computing (R Core Team, 2013; http://www.r -project.org). TMS measures and myelin water fraction were analyzed independently to identify differences between the three stu dy groups (MS, NMO and controls). A Shapiro-Wilk normality test determined that the data was normally distributed for all measures. A multivariate analysis of variance (MANOVA) demonstrated no hemispheric differences for any of the TMS measures (the independent measures for each MANOVA preformed being RMT, AMT or recruitment curve slope); therefore the data points were averaged over hemispheres for each participant. A MANOVA was also employed to assess for differences in the active and resting motor threshold TMS measures between MS, NMO and controls. Follow-up univariate analyses of variance (ANOVAs) were conducted if the overall 3 0   MANOVA result was significant. An ANOVA was also performed with a main effect of group and recruitment curve slope as the dependent variable. Post -hoc analysis employed independent t-tests to compare dependent variables between two of the three study groups. The same statistical analysis was performed to compare mean MWF between the three study groups in the CST regions. p-values equal to or below 0.05 were taken to be significant. For ANOVAs, F -values were reported with degrees of freedom for independent variables, degrees of freedom for number of subjects and the p-value. For t-tests, t-value was reported with degrees of freedom and p-value.  2.2.4.2 Correlation Analysis  The relationships between the motor thresholds, recruitment curve slope, MWF and disease duration were assessed using a Pearson product -moment correlation test for NMO and MS together. The correlations with EDSS score were assessed using a Spearman rank correlation test for NMO and MS together. Ten pairwise comparisons were conducted and the p-values were adjusted for multiple comparisons using a Bonferroni-Holm correction. Corrected p-values equal to or below 0.05 were taken to be significant.   2.3 RESULTS  One individual with NMO completed TMS measures for only one hemisphere and did not complete the recruitment curve for both hemispheres due to time constraint. The data was still included in the analysis. The clinical characteristics for all participants are shown in Table 2. 1.    3 1   2.3.1 Motor threshold  The RMT and AMT values were slightly higher for individuals with MS (RMT: mean=47 .80%, SD=9.24% and AMT: mean=40.55 %, SD=8.27 % ) and NMO (RMT: mean=49.75%, SD=8.89 %  and AMT: mean=39.85 %, SD=9.08 % ) compared to healthy controls (RMT: mean=42.00%, SD=8.27 %  and AMT: mean=33.75 %, SD=5.08 % ) but were not significantly different (RMT (F(2,27)=2.10, p=0.1); AMT (F(2,27)=2.38, p=0.1), Table 2. 1).   2.3.2 Recruitment Curve  The slope of the recruitment curve was different between the three groups (F(2,26)=3.46, p=0.05). Post -KRFWHVWLQJVKRZHGDORZHUDYHUDJHVORSHLQWKH102JURXSPHDQ ȝ96' FRPSDUHGWRWKHKHDOWK\FRQWUROVPHDQ ȝ96' W S and MS group (Figure 2 .2 PHDQ ȝV/%, SD=11:06; t(17)=2.52, p=0.02).  The recruitment curve for the NMO group was different from the MS group and healthy controls (Figure 2.3).   Figure 2.2 :  Boxplot of average recruitment curve slope plotted for healthy controls, multiple sclerosis (MS) and neuromyelitis optica (NMO). Significant differences between groups are shown. * p< 0.05  3 2    Figure 2. 3: Average recruitment curve maximal amplitudes as a function of motor threshold for healthy controls, multiple sclerosis (MS) and neuromyelitis optica (NMO). Plots display means and standard deviation for each motor threshold interval.  2.3.3 MRI Assessment  MWF measures of the CST was different between the three groups (F(2,27)=9.84, p=0.0006; Figure 2. 4 ). The NMO group (mean=0.17, SD=0.022) had lower CS T MWF when compared to MS (mean=0.19, SD=0.016; t(18)=2.78,  p=0.01) and when compared to healthy controls (mean=0.20, SD=0.017; t(18)=3.44, p=0.003). The CST MWF did not differ between the MS cohort and healthy controls (t(18)=0.93, p=0.4).  3 3    Figure 2. 4:  Boxplots of myelin water fraction (MWF) in the cortico-spinal tract (CST) for healthy controls, multiple sclerosis (MS) and neuromyelitis optica (NMO). Significant differences between groups are shown.  ** p  < 0.01, *** p < 0.00 1    2.3.4 Correlation Analy sis  A negative correlation was observed between EDSS scores and recruitment curve slope before correcting for multiple comparisons (r(19 )= -0.50 , p=0.0 3 ). The correlation was no longer statistically significant after correction (p=0.3 ; Figure 2. 5 ) across the MS and NMO cohorts. When separate correlation tests were performed for MS and NMO, no significance was found. No other correlations were found between EDSS, TMS and MRI measures (Table 2. 2).   3 4   E D S SRecruitment Curve Slope(V/%)0 2 4 601 02 03 04 05 0  r ( 1 9 )  =  -0 .5 0 , p  =  0 .0 3M SN M O Figure 2.5 : Correlations between recruitment curve slope and expanded disability status score (EDSS). Data were fitted by Spearman regression and 95% confidence intervals of the slope are drawn. 6SHDUPDQ’VUDQN correlation coefficient and uncorrected p-value are shown.  Correlations  r MWF vs. RMT -0.05    AMT  0.05    Recruitment Curve Slope  0.07  DD  vs. RMT  0.26    AMT  0.19    Recruitment Curve Slope  0.12    MWF -0.003  EDSS  vs. RMT  -0.02    AMT  0.11    Recruitment Curve Slope  -0.50*    MWF -0.004   Table 2.2:  MRI, TMS and clinical measure correlations for the NMO and MS cohort. Product -moment correlation coefficients are shown IRU0:)DQG''FRUUHODWLRQV6SHDUPDQ’VUank correlation coefficients are shown for EDSS correlations .  S”5 , uncorrected.  TMS: transcranial magnetic stimulation; MWF: myelin water fraction; RMT: resting motor threshold; AMT: active motor threshold; EDSS: expanded disability status scale; DD : disease duration .     3 5   2.4 DISCUSSION  In this study, group differences in TMS recruitment curves and CST MWF were found in the NMO cohort compared to MS and healthy controls. Additionally, a trend for a negative correlation between EDSS scores and recruitment curve slope was observed across the MS and NMO cohorts.   TMS-based recruitment curves are a measure of CST excitability and provide information regarding axonal integrity (Siebner and Rothwell, 2003 ; Jørgensen  et al., 2005; Devanne et al., 1997). The curve usually resembles a sigmoidal shape with a steeply rising slope and final plateau (Carroll et al., 2001). This sigmoidal shape was not obtained for healthy controls or individuals with MS, which might be because the stimulus intensity necessary for saturation was higher than the maximum intensity used. Consistent with our results, Jørgense n et al. (2005) also reported that recruitment curves from individuals with MS did not differ significantly from those of healthy controls.   Individuals with NMO obtained a saturated recruitment curve at a lower stimulus intensity as compared to the MS group and controls (Figure 2. 3 ).  A reduced and flattened recruitment curve was also observed in the affected hemispheres of individuals with stroke, a condition that primarily presents with axonal damage (Ward et al., 2006).  Furthermore, shallower recruitment curves have been observed in individuals with severe traumatic brain injury resulting in diffuse axonal injury (DAI) compared to healthy co ntrols; this difference was even more pronounced with increased injury severity and motor impairments (Bernabeu et al., 200 9).    3 6   Additionally, individuals with NMO had significantly lower recruitment curve slopes (Figure 2. 2) compared to controls and indi viduals with MS. These findings are opposite to those noted in individuals with dystonia, a neurological movement disorder, who demonstrate an increased recruitment curve slope compared to healthy controls possibly due to the sustained involuntary muscle contractions (Hallett, 2000) . On the other hand, individuals with NMO usually present with weakness or muscle impairment that may be represented by a lower recruitment curve slope. Therefore, the type of motor disability may be reflected by different recruitment curve slopes.   While individuals with MS were also disabled, significantly lower recruitment curve slopes were only observed in the NMO population (Figure 2.2) . Due to signal leakage from demyelination in the CST, reduced MEP ampl itudes have been observed in MS (Vucic et al., 2012).  However, there was a lack of structural evidence for MS pathophysiology in the CST for our cohort, supported by no observed differences in the CST MWF compared to controls. Therefore, changes in cortico-spinal excitability were also not expected. Additionally, MS has relative, but not absolute, preservation of axons with more direct damage to myelin compared to NMO (Wingerchuk et al., 2001). MS plaques form throughout the CNS but for permanent injury to occur, the accumulation of episodic inflammation needs to go beyond plasticity and repairing capacities (DeLuca et al., 2004). NMO lesions are generally more destructive than those of MS, characterized by extensive macro phage infiltration often results in damage to all cellular components including axons of both grey and white matter (Jacob et al., 2013).  It is possible that our MS cohort may have had relatively mild axonal loss reflected by near-normal cortico-spinal excitability because they are in the earlier stages of the disease course, whereas with NMO, the 3 7   axonal damage is prevalent regardless of disease duration. Furthermore, transverse myelitis presents more frequently in NMO, therefore, there is a greater probability of CST involvement compared to MS, where the CST may not always be affected due to the scattered distribution of MS plaques in the CNS (Love, 2006).     Reduced cortico-spinal excitability found in individuals with NMO may also be linked to the lower MWF in the CST (Figure 2.4) . Myelin or axonal damage (causing secondary damage to myelin) to the CST would cause a decrease in MWF. Damage may be a result of lesions within the spinal cord and subsequent Wallerian degeneration of proximal neurons in the CST where MWF was measured (DeLuca et al., 2004).  However , CST MWF and cortico-spinal excitability were not correlated, suggesting that damage to CST myelin may not directly affect excitability. In NMO, the CST has been previously identified as a common site for lesions, yet the underlying immune-pathological mechanisms are still unknown (Kim et al., 2012).   Contrary to our hypothesis, the CST MWF did not differ between individuals with MS and healthy controls. The CST in MS has been studied previously, although not over its entire intracranial course and not with myelin water imaging. Lin et al. (2007) used d iffusion tensor tractography to investigate the pyramidal tract in remitting-relapsing MS and found abnormal diffusion characteristics compared to healthy controls; however, they did not investigate the CST region alone. Recent advancements in myelin water imaging have allowed us to obtain good volumetric coverage and analyze the CST throughout the brain for the first time. Interestingly, we obtained similar results in another cohort where CST MWF had the smallest difference between 58 individuals with remi tting-relapsing MS and 34 healthy controls compared to other 3 8   regions such as the corpus callosum (Kolind et al., 2014; Vavasour et al., 2014). The magnetisation transfer ratio which is related to macromolecule content also had insignificant differences between the groups in the CST (Kolind et al., 2014; Vavasour et al., 2014) .  DeLuca and colleagues (2004) found a size -dependent loss of axons in MS CST where small diameter fibres were reduced with relative preservation of large fibres implying differential sensitivity to damage of axons within this tract. This preferential loss of smaller axons may result in an underestimation of the extent of damage to the CST in individuals with MS and might be another reason why a decrease in MWF was not evident in the MS group as compared to healthy controls.  There are some limitations to this study. One limitation was the small sample size within each participant group. However, even with only ten subjects per group, a significant difference was found in CST MWF and recruitment curves between NMO, MS and healthy control groups. The plateau of the sigmoidal shape was not observed in the recruitment curves of the MS or control group; it is possible that to achieve this plateau, higher stimulation intensities would have been required. Yet, there was evidence of a plateau in the NMO group, and thus it is possible that the group differences in excitability would have been enhanced had we employed higher stimulation intensities. Another disadvantage was that spinal imaging was not performed. Lesions within the spinal cord could influence the excitability of the cortico-spinal pathway (Love, 2006). Lesion identification was not completed on the MS group because it was not included in the protocol, however, we did find that seven of the NMO participants had lesions present in the CST (data not shown). Nine of the MS participants were on glatiramer acetate treatment while the NMO participants were on a variety of different treatments. The effect of treatment on cortical 3 9   excitability and MWF is currently unknown. Due to the small number of cases of NMO worldwide, there is a paucity of research in the field limiting the understanding of the pathogenesis of NMO. No significant clinical correlations were observed in this study which may partly be due to the very small sample size. However, previous research suggests correlations between MRI measures and clinical outcomes may underestimate the true extent of axonal loss because the degree of reduction in nerve fibre loss may be dependent on the region of the CST (DeLuca et al., 2004).  Future work should evaluate more specific relationships between brain structure-function changes, perhaps with motor function by including lower extremity assessments using TMS as well as additional functional tests. The current study, although exploratory, suggest differences between NMO and MS that should be comprehensively characterized in upcoming investigations.   The combined results from TMS and myelin water imaging suggest that there are both neurophysiological and neuroanatomical changes found in individuals with inflammatory CNS disorders and that joint utilization of these techniques may offer a better understanding of the underlying pathophysiology in these diseases. In particular, individuals with neuromyelitis optica have decreased cortico-spinal excitability and lower cortical spinal tract myelin water fractions compared to MS which could account for some of their disability.      4 0   Chapter 3  3  QUANTIFYING VISUAL PATHWAY AXONAL AND MYELIN LOSS IN NEUROMYELITIS OPTICA AND MULTIPLE SCLEROSIS   3 .1 INTRODUCTION   Multiple sclerosis (MS) is a demyelinating neurological disease that results in white matter lesions, affecting the myelin sheath and the myelinating cells such as the oligodendrocytes (Frohman et al., 2006; Lassmann et al., 2007). Neuromyelitis optica (NM O) is a rare, autoimmune disorder of the central nervous system that is characterized by episodes of inflammation and damage to astrocytes and has overlapping clinical features with MS including optic neuritis (ON) and transverse myelitis. NMO typically presents more severe relapses than MS with greater axonal loss, which usually elicits more permanent disability (Merle et al., 2008). Distinction from MS is crucial as prognosis is usually worse in NMO and treatment options for both diseases differ considerably (Jacob et al., 201 3).   Visual dysfunction is a common cause of disability and reduced quality of life in MS and NMO (Mowry et al., 2009; Pfueller and Paul, 2011). The visual system can be used as a model of potential differences between MS and NMO. ON is an inflammation of the optic nerve and lesions are known to occur in the optic nerve of both MS and NMO patients (Toosy et al., 2014). These lesions can result in either retrograde degeneration ± D“G\LQJEDFN”SKHQRPHQRQFDXVLQJchanges to the cell body proximal to the point of injury along the axon ± or anterograde (Wallerian) degeneration ± D“G\LQJIRUZDUG”SURFHVVWKDWUHVXOWVLQGHJHQHUDWLRQWRWKHSDUWRI4 1   the axon distal to the injury site (Kanamori et al., 201 2; Siffrin et al., 2010). MS and NMO  eyes with ON are often thought to have good clinical recovery after an attack; however, subclinical abnormalities indicate incomplete recovery (Balcer et al., 2015).   The main goal of this cross-sectional study is to use optical coherence tomography (OCT) to examine damage to the anterior visual pathway and myelin water imaging to examine damage to the posterior visual pathway in MS and NMO.  OCT is a non-invasive imaging tool primarily used to visualize the internal microstructures of the eye (Costello et al., 2006). OCT can be used to examine the anterior visual pathway damage through the retina by obtaining measures of retinal nerve fibre layer (RNFL) and macula depicting neuro-axonal damage in these diseases (Henderson et al., 2008; Petzold et al., 201 0; Zimmermann et al., 2013). The RNFL is composed of unmyelinated axons from the retinal ganglion cells (Figure 1. 7 ; Toosy et al., 2014). Thinning of the RNFL around the optic nerve head (where it is the thickest) may be representative of axonal damage (Barkhof et al., 2009; Lange et al., 2013). The macular located in the center of the retina is where the retinal ganglion cell bodies originate (Figure 1. 7 ; Henderson et al., 2008; Trip et al., 2005). Scanning of the macula to obtain the total macular volume (TMV) images the retinal ganglion cell layer and represents a potential method for capturing both axonal and neuronal degeneration (Barkhof et al., 2009). In MS and NMO, optic neuritis can result in retrograde degeneration to the RNFL and macula which can be detected using OCT analysis (Petzold et al., 2010).     4 2   The optic radiation (OR) is a collection of myelinated axons from relay neurons in the lateral geniculate nucleus (LGN) and is known to carry visual information originating from the retina to the visual cortex (Figure 1. 7 ). The OR is a frequent site for demyelination and injury in individuals with ON contributing to visual dysfunction and may be related to OCT measures of neuronal and axonal health (Hornabrook et al., 1992; Reich et al., 2009; Yu  et al., 2008). An advanced magnetic resonance imaging (MRI) technique called myelin water imaging can be used to quantify the amount of myelin in vivo in the posterior visual pathway, particularly, the OR (Mackay et al., 1994; Pfueller and Paul, 2011). An terograde degeneration due to lesions in the optic nerve may result in upstream damage to the posterior visual pathway affecting areas such as the OR (Petzold et al., 2010).   Earlier work has suggested differences in OCT measures between NMO and controls (Burkhold et al., 2009; Costello et al., 2006; Grazioli et al., 2008; Gordon -Lipkin et al., 2007; Lange et al., 2013; Ratchford et al., 2009), but there has yet to be a comparison of TMV and RNFL thickness with myelin water imaging of the posterior visual pathway, such as the OR. A correlation between RNFL thickness and TMV and OR MWF may provide an opportunity to study how the damage to one part of the visual pathway may cause, via antero- or retrograde degeneration, an alteration in the other part of the visual pathway (Petzold et al., 2010; Yu et al., 2008). Using these imaging techniques to focus on the visual pathway and its anatomical correlations can help identify significant differences between NMO and MS and give further insight into NMO-specific damage processes. Additionally, if there is a relationship between the measures, OCT may be a useful surrogate for brain atrophy in the OR.    4 3   3.2 METHODS  3.2.1 Participants 42 participants with MS (EDSS : median =2.0 , range= 0.0 ±4.0, age : mean =40  years, standard deviation (SD)= 10  years, 17 males and 25 females), 10 individuals with  NMO (EDSS : median= 2.5, range= 2.0 ±6.0, age : mean= 43  years, SD= 11  years, 3 males and 7 females), and 12 healthy controls (age: mean= 31  years, SD= 10  years, 3 males and 9 f emales) were recruited for this study (Table 3 .1 ). Any subject who met criteria for an NMO spectrum disorder was included, regardless of whether they had been affected by a clinical episode of ON. All but three subjects in the NMO group were NMO-IgG positive. Two subjects in the NMO group and 26 subjects in the MS group did not have a clinical episode of ON. Subjects with MS included those with clinically definite MS by the Poser criteria  (2001) or the modified McDonald criteria (Polman et al., 2005) . All participants with MS except one (on minocycline) were not on treatment at the time of testing. Four participants with NMO were on mycophenolic acid (CellCept®), three on azathioprine (Imuran®), and one participant each on either interferon beta-1a (Re bif 44®), mitoxantrone or rituximab. No participants were on steroid therapy at least one month prior to entering the study. ON history was characterized as having one or more episod es of ON in one or both eyes prior to entering the study. The University o f British Columbia Clinical Research Ethics Board approved all study procedures.             4 4     Technique Measurement Control MS MS-NON MS-ON NMO NMO-NON NMO-ON Clinical EDSS  - 2.0 (0 -4.0)  1.75 (0 -4)  2.0 (0 -4)  2.5 (2.0 -6.0)  2.5 (2.5 -2.5)  2.5 (2.0 -6.0)   Age (years) 3 1.2 (9.6)  39.5 (9.6)  40.3(9.6)  38.3(9.7)  43.2 (11.2)  39.0(3)  47.9(13)   Sex (M:F)  3 :9  17 :25  10 :16  7:9  3:7  1:1  2:6   # of ON history - 16  - - 7  - - OCT 51)/7KLFNQHVVȝP 1 02.9(15.0)  90.4(12.7)  93.5(11.1)  81.0(12.8)  76.0(19.1)  8 7.3(20.9 ) 70.8(16.4 )  PLQ51)/ȝP 9 9.6(10.5)  87.0(13.0)  91.7(11.3)  79.6(12.5)  68.1(20.1)  94.0(4.2)  61.6(16.6)   TMV (mm 3 ) 8.85(0.4)  8.49(0.5)  8.59(0.4)  8.18(0.4)  8.11(0.4)  8.3 7(0.5 ) 7 .99 (0.4)   minTMV (mm 3 ) 8.76(0.3)  8.40(0.5)  8.54(0.5)  8.17(0.4)  7.98(0.5)  8.54(0.5)  7.84(0.3)  MRI MWF OR 0.113(0.01)  0.0969(0.01)  0.0982(0.02)  0.0947(0.01)  0.0977(0.01)  0. 1007 (0.009)  0.09 69 (0.01)   Table 3.1: OCT/OR MWF Subject Characteristics and Results. The mean of each measure is presented except for sex (ratio of male to female) and EDSS (median). Standard deviation is in parentheses except for EDSS (range).   MS: multiple sclerosis; NMO: neuromyelitis optica; RNFL: retinal nerve fibre layer; TMV: total macular volume; minRNFL: min imum RNFL thickness between two eyes; minTMV: minimum TMV between two eyes; MWF: myelin water fraction; OR: optic radiation; EDSS: expanded disability status scale; ON: optic neuritis.   MS: All MS eyes (for OCT) or all individuals with MS (for MRI and Clinical);  MS-NON: MS eyes without ON history (for RNFL/TMV) or MS patients without ON history (for minRNFL/TMV, MRI &  Clinical);  MS-ON: MS eyes with ON history (for RNFL/TMV) or MS patients with ON history (for minRNFL/TMV, MRI &  Clinical);  NMO: All NMO eyes (for OCT) or all individuals with NMO (f or MRI and Clinical);  NMO-NON: NMO eyes without ON history (for RNFL/TMV) or NMO patients without ON history (for minRNFL/TMV, MRI &  Clinical);  NMO-ON: NMO eyes with ON history (for RNFL/TMV) or NMO patients with ON history (for minRNFL/TMV, MRI  &  Clinical).  3.2.2 OCT protocol and analysis The OCT assessment was conducted on the Heidelberg Spectralis SD -OCT device (Software version 5.6.4; Heidelberg Engineering, Heidelberg, Germany) for all participants. An online tracking system was used to compensate for eye movements.  3.2.2.1 RNFL Thickness P rotocol  The RNFL protocol was in high -resolution mode (axial resolution= ȝPVFDQVSHUsecond). 16 consecutive circular B -scans (each composed of 1,536 A -scans) with a diameter of 3.4 mm were automatica lly averaged to reduce speckle noise (Lange et al., 2013). Sever al scans 4 5   were taken by experienced operators and the best centered scans with a quality of at least 25 were chosen for analysis. The software algorithm calculated the objective refraction (spherical equivalent) and the overall mean RNFL thickness was obtained.   3.2.2.2 TMV P rotocol  The macular volume protocol involved 61 consecutive B -scans (ART 9; 76 8 A -scans each) horizontally crossing the macula for individuals with NMO and healthy control s. For individuals with MS, the TMV was obtained using a built -LQPDFXODUYROXPHSURWRFRO“33ROH1”FRQVLVWLQJof 61 consecutive B -scans (ART 15; 768 A -scans each) vertically crossing the macula. The software algorithm calculated the total macular volume automatically.   3.2.3 MRI Protocol and Analysis MRI scans were performed on a Philips 3.0T Achieva scanner (Best, the Netherlands). A 3D T1 -weighted spoiled gradient echo with repetition time (TR) of 6.5 ms, echo time (TE) of 3.6ms,  flip angle of 13 ° and 1 .7 mm x 1 .7 mm x 1 .7 mm voxel size was acquired for segmentation of the tract of interest. An axial combined gradient echo and spin echo (GRASE) T2 relaxation with 32 echoes sequence, TR = 1000ms, TE  1 -32 = 10 -32 0 ms, echo spacing = 10ms, 1mm x 1mm x 5mm voxel  size reconstructed to 1 mm x 1 mm x 2.5 mm, sensitivity encodi ng factor=2 and an echo planar imaging factor= 3 was acquired for MWF estimation (Prasloski et al., 2012; Neuroimage). The scanned volume consisted of a 10cm slab measured from the top of the br ain caudally.   4 6   The signal decay curve obtained by the T2 relaxation sequence was modelled by multiple exponential components and the T2 distribution was estimated using non -negative least squares with the extended phase graph algorithm (Whitall et al., 19 89 ; Prasloski et al., 2012; Magnetic Resonance in Medicine). MWF in each image voxel was computed as the ratio of the area under the T2 distribution with times of 10 -40ms to the total area under the distribution (Laule et al., 2006). This analysis was perf ormed using in-house software on Matlab® 2012 (MathWorks, Natick, Massachusetts, U.S.A).   3D T1 images of each subject were registered to T2 GRASE images using FMRIB's Linear Image Registration Tool (FLIRT; Smith et al., 2004). Automated segmentation of t he OR was performed by registration of the corresponding Juelich Histological  atlas region to the T2 GRASE images using a combination of FLIRT and FNIRT ( Figure 3. 1 ; Andersson et al., 2007). These regions were then manually edited to ensure only white matter was included and overlaid on the MWF images to obtain the mean MWF for each region.   Figure 3.1: The optic radiation region of interest (in red) on T 1 -weighted image in an individual with neuromyelitis optica. A) sagittal view B) axial view C) coronal view.    4 7   3.2.4 Statistical Analysis 3.2.4.1 Independent MRI and OCT A nalysis  Statistical analysis was performed using R: a language and environment for statistical computing (R Core Team, 2013 ; http://www.r -project.org). OCT measures and MWF were analyzed independently to identify differences between the three study groups (MS, NMO and controls). A Shapiro-Wilk normality test determined that the data was normally distributed for all measures. For the analyses of RNFL thickness and TMV, a  linear mixed-effects model was used to account for intra-patient inter-eye dependencies, with group, age, gender and ON history included as covariates. Based on this model, we assessed the differences between study groups as well as between eyes with and without ON history. For each measure, six pairwise comparisons were conducted and the p-values were adjusted for multiple comparisons using a Hommel correction. Corrected p-values ൑ 0.05 were taken to be significant.  For optic radiation MWF, initial analysis showed no significant ON history, age or gender effect; therefore, a linear regression model was not used to account for these factors. Instead, a one-way analysis of variance (ANOVAs) was performed to compare the four groups: healthy controls, MS with ON history (MS-ON) and those without it (MS-NON), and NMO subjects. The NMO group only had two subjects without ON history so they were not split up (preliminary analysis showed no clear difference when split up). Post -hoc analysis employed )LVKHU’V/6'test to compare all pairs of the four study groups. P -values ൑ 0.05 were taken to be significant. For ANOVAs, F -values were reported with degrees of freedom for independent variables, degrees of freedom for number of subjects and the p-value. For t-tests, t-value was reported with degrees of freedom and p-value.   4 8   3.2.4.2 Correlation A nalysis  The relationship between the RNFL thickness, TMV, and MWF were assessed using a Pearson product-moment correlation for NMO, MS and healthy controls separately and together. OCT DQG05,PHDVXUHVZHUHFRUUHODWHGZLWK('66VFRUHVXVLQJ6SHDUPDQ’VUDQNFRUUHODWLRQcoefficient for NMO, MS and healthy controls, both separately and together. Based on initial analysis there was no ON history, age or gender effects on the dependent variable (OR MWF), therefore a linear regression model was not used to account for these factors. For the analysis in this section, the minimum RNFL and TMV values between the two eyes of each individual was used. Because this is an exploratory study meant to be hypothesis generating, the correlations were not adjusted for multiple comparisons. P -values ൑ 0.05 were taken to be significant.   3.3 R ESULTS  Four individuals with MS did not have MWF measures of the OR due to factors such as scheduling issues, claustrophobia and other MRI contradictions. Their data was still included in the OCT analysis but not in the MWF or correlation analysis. One individual with NMO did not have the OCT exam completed for the left eye due to poor vision making it difficult to fixate. The data RIWKLVVXEMHFW’VULJKWH\Hwas still included in all of the analysis. The clinical characteristics for all participants are shown in Table 3 .1 .  3.3.1 RNFL Thickness In total, there were 24 healthy control eyes, 13 NMO eyes with ON (NMO -ON), 6 NMO eyes without ON (NMO-NON), 21 MS eyes with ON (MS -ON), and 63 MS eyes without ON (MS -NON). As seen in Figure 3. 2 , RNFL thickness was lower in NMO -21PHDQ ȝP4 9   6' FRPSDUHGWRKHDOWK\FRQWUROVH\HVPHDQ ȝP6' S 7KHUHZDValso a decrease in RNFL thickness in MS -21PHDQ ȝP6' FRPSDUHGWRKHDOWK\control eyes (p=0.005; Figure 3. 2 ). For both NMO and MS, eyes with ON history had lower RNFL thickness than those without (NMO -121PHDQ ȝP6' S 06-NON: PHDQ ȝP6' S UHVSHFWLYHO\)LJXUH3. 2 ).  R N F L  T h ic k n e s sThickness (m)C o n tr o l N M O -O N N M O -N O N M S -O N M S -N O N05 01 0 01 5 02 0 0**********n  =  2 4 n  =  1 3 n  =  6 n  =  2 1 n  =  6 3 Figure 3.2: RNFL thickness in  individuals with neuromyelitis optica, multiple sclerosis and healthy controls.  The Hommel -corrected p-values were based on a linear mixed-effects model accounting for intra-patient inter-H\HGHSHQGHQFLHVDJHJHQGHUDQG21KLVWRU\S”**** p ”QLVWKHQXPEHURIH\HVIRUHDFKJURXS  RNFL: retinal nerve fibre layer, ON: optic neuritis, NMO-ON: neuromyelitis optic with ON history, NMO-NON: neuromyelitis optic with out ON history, MS-ON: multiple sclerosis with ON history,  MS-NON:  multiple sclerosis without ON history    5 0   3.3.2 Total Macular Volume  Similar to RNFL results, TMV was lower in both NMO-ON (mean=7.9 9 mm 3 , SD=0.3 6 ) and MS-ON (mean=8.1 8 mm 3 , SD=0.3 6 ) compared to healthy controls eyes (mean=8.85mm 3 , SD=0.41; p =0.0007 and p=0.002 respectively; Figure 3. 3 ). For MS eyes with ON history, there was lower TMV compared to those without it (MS -NON: mean=8.59mm 3 , SD=0.44, p=0.003; Figure 3. 3 ). Unlike with RNFL thickness, NMO eyes with ON history did not differ significantly compared to eyes without it (NMO-NON: mean=8.38mm 3 , SD=0.46, p=0.2; Figure 3. 3 ).  T o ta l  M a c u la r  V o lu m eVolume (mm3)C o n tr o l N M O -O N N M O -N O N M S -O N M S -N O N67891 01 1******n  =  2 4 n  =  1 3 n  =  6 n  =  2 1 n  =  6 3 Figure 3.3: Total macular volume in individuals with neuromyelitis optica, multiple sclerosis and healthy controls. The Hommel -corrected p-values were based on a linear mixed-effects model accounting for intra-patient inter-eye dependencies, age, gender and ON history.  SS”S” n is the number of eyes for each group.  ON: optic neuritis, NMO-ON: neuromyelitis  optic with ON history, NMO-NON: neuromyelitis optic without ON history, MS-ON: multiple sclerosis with ON history, MS-NON:  multiple sclerosis without ON history  5 1   3.3.3 MRI Assessment MWF measures of the OR were different between the four groups (F3,56 = 4.15 , p=0.0 1 ; Figure 3. 4 ). The NMO group (mean=0.098, SD=0.011), the MS -ON (mean=0.096, SD=0. 0126) and the MS-NON group (mean=0.097, SD=0.016) had lower OR MWF when compared to healthy controls (mean=0.113, SD=0.014; t 5 6 =2. 54 , p=0.01, t 56 =3. 15 , p=0.003, an d t 56 = 3.07 , p=0.00 3 , respectively; Figure 3. 4 ). The OR MWF did not differ between the MS-ON and NMO cohort (t 5 6 =0.28 , p=0. 8), MS -NON and NMO cohort (t 56 =0.035 ,p=1.0) or betwee n MS-ON and MS-NON cohort (t 5 6 =0.31 , p=0.8; Figure 3. 4 ).  O p tic   R a d ia t io nMWFC o n tr o l N M O M S -N O N M S -O N0 .0 00 .0 50 .1 00 .1 5n =  1 2 n =  1 0*****n =  2 2 n =  1 6  Figure 3.4: MWF in the optic radiation of individuals with neuromyelitis optica, multiple sclerosis and healthy controls.   S”S”. n is the number of subjects for each group.  MWF: myelin water fraction, Control: healthy control subjects, NMO: neuromyeliti s optic, ON: optic neuritis, MS-NON: multiple sclerosis  subjects without ON history, MS-ON: multiple sclerosis subjects with ON history. 5 2   3.3.4 Correlation Analysis A trend towards a positive correlation was observed between the OR MWF and RNFL thickness in the NMO population (r(10)=0.56, p=0.09; Figure 3. 5 A). Similarly, a positive correlation was observed with the OR MWF and TMV in the healthy control population (r(12)=0.59, p=0.04; Figure 3. 5 B). When grouped correlations were performed for MS, NMO and healthy controls, a significant positive correlation was observed between OR MWF and RNFL thickness (r(60)=0.29, p=0.02; Figure 3. 6 A) and  a positive correlation between OR MWF and TMV (r60)=0.41, p=0.001; Figure 3. 6 B). No other correlations were found significant between OCT and MRI measures (Table 3. 2). A negative correlation was observed between EDSS scores and TMV in the NMO cohor t (r(10)= -0.58, p=0.04 ; Figure 3.7 ); however, correlation did not exist when grouped with the MS cohort (r(52)= -0.12, p=0.4). No other correlations were observed between EDSS and MRI or OCT measures for MS and NMO, both separately or grouped together (Table 3. 2).   N M O : M W F  v s  R N F LT h ic k n e s s  (m )4 0 6 0 8 0 1 0 0 1 2 00 .0 60 .0 90 .1 20 .1 5MWFr ( 1 0 )  =  0 .5 6 ,  p  =  0 .0 9A H e a lth y  C o n tr o l:M W F  v s  T M VV o lu m e  (m m3)8 .0 8 .5 9 .0 9 .50 .0 60 .0 90 .1 20 .1 5MWFr ( 1 2 )  =  0 .5 9 ,  p  =  0 .0 4B  Figure 3.5: A) A trend towards a positive correlation between MWF in the optic radiation and the RNFL thickness in the NMO population. B) A positive correlation between MWF in the optic radiation and the TMV in healthy controls. Minimum RNFL and TMV were used for comparisons with MWF in the optic radiation. 95% confidence interval is shown in dotted line.  NMO: neuromyelitis optica, MWF: mye lin water fraction, RNFL: retinal nerve fiber layer thickness, TMV: total macular volume  5 3   A O R  M W F  v s  R N F LT h ic k n e s s  (m )4 0 6 0 8 0 1 0 0 1 2 00 .0 60 .0 90 .1 20 .1 5r ( 6 0 )  =  0 .2 9 ,  p  =  0 .0 2MWFM S -O NM S -N O NC O NN M O -O NN M O -N O Nn  =  1 6n  =  2 2n  =  1 2n  =  8n  =  2 B O R  M W F  v s  T M VV o lu m e  (m m3)7 8 9 1 00 .0 60 .0 90 .1 20 .1 5r ( 6 0 )  =  0 .4 1 ,  p  =  0 .0 0 1MWFM S -O NM S -N O NC O NN M O -O NN M O -N O Nn  =  1 6n  =  2 2n  =  1 2n  =  8n  =  2 Figure 3.6: A) Positive correlation between MWF in the optic radiation and RNFL thickness and B) a positive correlation between MWF in the optic radiation and the TMV for all groups combined. Minimum RNFL and TMV between the two eyes were used for comparisons with MWF in the optic radiation. 95% confidence interval is shown in dotted line.   OR: optic  radiation, MWF: myelin water fraction, NMO: neuromyelitis optic, MS: multiple sclerosis, ON: optic neuritis, NMO -ON: NMO with ON history, NMO -NON: NMO without ON history, MS-ON: MS with ON history, MS -NON: MS without ON history, RNFL: retinal nerve fibre layer, TMV: total macular volume, CON: healthy control  5 4   Correlations Control MS NMO Grouped MWF vs RNFL  r(12)=0.23, p=0.5  r(3 8 )=0.0 6 , p=0.7  r(10)=0.56, p=0.09  r(60 )=0. 29 , p=0.02 *  MWF vs TMV  r(12)=0.59, p=0.04*  r(3 8 )=0. 27 , p=0. 1  r(10)=0.32, p=0.4  r(60 )=0.4 1 , p=0.001 **  MWF vs EDSS  NA r(38)=0.02, p=0.9  r(10)= -0. 26 , p=0.4  r(48)= -0.0 08 , p= 1.0  RNFL  vs EDSS  NA r(42)= -0.0 9 ,  p= 0.6  r(10)= -0. 40, p=0.2  r(52)= -0.21, p=0.1  TMV  vs EDSS  NA r(42)= 0.0 5 , p=0. 8  r(10)= -0. 58, p=0.04*  r(52)= -0.1 2, p=0.4   Table 3.2:  MRI, OCT and clinical measure correlations for the NMO, MS and healthy control group separately and together. Product -moment correlation coefficients are shown for OCT and 0:)FRUUHODWLRQVDQG6SHDUPDQ’VUDQNHGFRUUHODWLRQFRHIILFLHQWDUHVKRZQIRUWKHUHOationship between EDSS and OCT/MRI measures . Minimum RNFL and TMV were used for all correlations. S”S”, uncorrected.  MS: multiple sclerosis; NMO: neuromyelitis optica; RNFL: retinal nerve fibre layer; TMV: total macular volume; MWF: myelin water fraction in the optic radiation; EDSS: expanded disability status scale; Grouped: All subjects grouped together   N M O :T M V  v s   E D S SE D S S   S c o reVolume (mm3)2 4 667891 0N M O -O NN M O -N O Nr ( 1 0 )  =  -0 .5 8 , p  =  0 .0 4 Figure 3.7:  A negative correlation between EDSS score and TMV in the NMO population. Minimum TMV was  used for comparisons with EDSS score . 95%  confidence interval is shown by dotted line.  NMO: neuromyelitis optica, TMV: total macular volume , EDSS: expanded disability status scale, ON: optic neuritis, NMO -ON: NMO patients with ON history, NMO -NON: NMO patients without ON history 5 5   3.4 DISCUSSION  No studies to date have assessed the entire visual pathway through a combination of several imaging techniques in NMO (Pfueller and Paul, 2011).   RNFL thinning and decreased TMV was most pronounced in NMO eyes wi th ON history followed by MS eyes with ON history compared to unaffected eyes in MS and NMO. This finding is consistent with the concept of retrograde degeneration from the optic nerve to the retina. Retrograde loss of retinal ganglion cells have been observed in primate studies when the optic nerve was transected (Quigley et al., 1977 ). Similar to our results, previous research found that the RNFL and TMV of unaffected eyes in MS patients were less abnormal compared to eyes with a history of ON (Frohman et al., 2008; Burkholder et al., 2009; Ratchford et al., 2009; Siger et al., 2008). The lack of difference in the TMV measure between the NMO -ON and NMO-NON eyes is likely due to the small sample size of the NMO -NON group, as previous research has found a significant difference between these groups (Lange et al., 2013).   RNFL and macular volume loss in non -ON MS eyes may also be due to lesions within the posterior optic pathway that could result in retrograde degeneration in the retina (Petzold et al., 2010;  Young et al., 2013; Noval et al., 2011). Another possibility is clinically silent demyelinating lesions within the optic nerve that result in retrograde degeneration to the RNFL/macula in unaffected MS eyes (Henderson et al., 2008).   Similarly, the observed decrease in OR MWF in the MS-ON and NMO group (predominantly affected by ON) compared to healthy controls may be due to anterograde degeneration from 5 6   damage to the optic nerve (Green et al., 2010). It is possible that in addition to  anterograde degeneration, MS OR could be affected by lesions that typically accumulate in the periventricular and subcortical white matter that involves OR tracks (Reich et al., 2009).   The decrease in OR MWF in the MS-NON group compared to controls suggest that underlying disease in MS may cause damage to the visual pathway in the absence of ON (Reich et al., 2009). The general neurodegeneration and ongoing progressive neuro -axonal loss in MS brain and retina may be compounded by the additional axonal loss due to ON which are not directly linked (Zimmerma nn et al., 2013; Noval et al., 2011).  There may be a greater presence of subclinical optic nerve damage in MS with axonal attrition independent of ON attacks or due to an increased probability of MS lesions appearing in the optic chiasm and tract (Naismith et al., 2009; Noval et al., 2011; Pfueller and Paul, 2011).    On the other hand, NMO is predominantly a result of relapses and rarely due to secondary progression or diffuse low grade inflammation throughout normal appearing CNS tissue (Ratchford et al., 2009).  Yu and colleagues observed abnormal diffusion in the brain white matter of NMO patients which was restricted to regions with connections to the spinal cord and optic nerve suggesting that it is likely due to lesions in those regions (Yu et al., 2008). No significant differences between NMO-NON eyes and healthy controls using OCT have been observed in a previous study (Lange et al., 2013) ; however, this is difficult to confirm in our study since our sample size was too small to observe a difference in the NMO -NON group.    5 7   OR abnormalities have been observed in other neurological diseases. OR damage suggestive of Wallerian degeneration was identified in patients with LGN lesions (Reich et al., 2009). The loss of neurons in the LGN of individuals with MS may also be due to anterograde trans -synaptic degeneration from optic nerve damage (Green et al., 2010). Diffuse abnormalities along the entire tract were observed in MS patients with ON history from the LGN, to the OR and finally in the occipital cortex (Barkhof et al., 2009).   The association between RNFL thinning, macular volume and decreased myelin in the posterior visual pathway provides further support for how damage to one part of the visual pathway may cause, via antero- or retrograde degeneration, alterations in another part of the visual system. Previous studies have found similar correlations between MRI measures (such as T1/T2 lesion volume, brain parenchymal fraction, and magnetization transfer imaging) and OCT measures in MS and patients with ON (Frohman et al., 2009; Siger et al., 2008; Ciccarelli et al., 2005; Barkhof et al., 2009).  The correlation between RNFL thickness and OR MWF is li kely driven by the relationship seen in the NMO group, which is primarily due to lesions in the optic nerve resulting in anterograde degeneration to the posterior visual pathway. Interestingly, the correlation between TMV and OR MWF in healthy controls may  indicate that there is a wide range of macular volume that can be considered normal. In this study, a correlation between the OR MWF and OCT measures was observed in the grouped correlation but not separately for individuals with MS. The relationship between the two measures may be exemplified by the range of visual disability when combining the three groups together.   5 8   TMV  measures were negatively related to clinical disability in the NMO cohort; however, this relationship did not persist when grouped together with the MS cohort. Previous studies have found that TMV  did not correlate with MRI features (Noval et al., 2011).  It is difficult to say if this is a true relationship because the correlation observed in our study was only borderline significant, in addition to not being corrected for multiple comparisons. Furthermore, EDSS is  a measure of general disability and we did not expect a correlation with this measure in the NMO population as it is primarily driven by focal damage to the spinal and optic nerve. OR MWF and RNFL  was not related to EDSS scores, howe ver previous research suggests that correlations between MRI measures and clinical outcomes may underestimate the true extent of axonal loss because the degree of reduction in nerve fibre loss may be dependent on the particular region of the OR that is involved and varies considerably within the tract (DeLuca et al., 2004).   There are some limitations to this study. One limitation was the small sample size within the NMO and healthy control group. However, even with the current sample size, a significant difference was found in the OR MWF and OCT measures between NMO, MS and healthy control groups. OCT measures were collected under slightly different protocols for the TMV measure in the MS and NMO group. The differences between the two groups may be influenced by this; however, they were consistent with the RNFL measures and previous studies. The MS and NMO participants were on a variety of different treatments. The effect of treatment on the visual pathway and MWF is currently unknown. Due to the small number of cases of NMO worldwide, there is a paucity of research in the field limiting the understanding of the pathogenesis of NMO. The current study, although exploratory, suggest differences between NMO and MS that should be comprehensively characterized in upcoming investigations.  5 9    The correspondence between reductions in OCT measures of neuronal and axonal health in the anterior visual pathway and MRI-based measures of myelin health in the posterior visual pathway suggests that these measures may be used to evaluate disease progression and treatment approaches that promote repair. Imaging techniques like OCT and myelin water imaging have not yet been used to their full potential, especially studies focusing on the visual pathway and its anatomical correlates. Using these imaging techniques t o focus on the visual pathway and anatomical correlates can help identify significant differences between NMO and MS and give further insight into NMO-specific damage processes as well as into trans-synaptic damage processes independent of their underlying condition. The positive correlation observed between the OCT and myelin water imaging measures in NMO, MS and healthy controls suggest that the use of OCT as a surrogate for myelin loss in the OR should be further investigated.       6 0   Chapter 4  4  CORRELATING LESION VOLUME AND MYELIN WATER FRACTION IN NORMAL APPEARING WHITE MATTER TRACTS OF MULTIPLE SCLEROSIS AND NEUROMYELITIS OPTICA PATIENTS   4 .1 INTRODUCTION  Multiple sclerosis (MS) and neuromyelitis optica (NMO) are complex autoimmune inflammatory disorders that target both white and gray matter elements of the central nervous system (CNS); however, prognosis is typically worse in NMO and treatment options differ considerably (Jacob et al., 2013). The primary fea tures of MS include demyelinating white matter lesions and relapses (Siffrin et al., 2010). However , MS relapses are not directly related to long-term disability progression (Confavreux et al., 2000; Scalfari et al., 2010) , and as a result, there is a discrepancy between clinical and magnetic resonance imaging (MRI) measures that do not fully explain disability in MS.  This may be partly due to diffuse damage in the normal appearing brain tissue that looks normal on conventional MRI but appears abnormal with histological studies and advanced imaging (Trapp et al., 1999; Traboulsee et al., 2003; Tortorella et al., 2000 ). Normal appearing white matter (NAWM) degeneration occurs early in MS disease and has been shown to be only partly related to white matter lesions (Filippi and Rocca, 2005).    Previous studies using magnetization transfer imaging ( MTI) and diffusion tensor imaging (DTI ) have looked at correlations between lesions and NAWM in MS and NMO but have obtained mixed results about the relationship (Liu et al., 2012; Yu et al., 2006; Werring et al., 1999). Magnetization transfer and diffusion tensor imaging are the most commonly used advanced MRI 6 1   measurements related to myelin, both of which have revealed differences in MS and NMO compared to healthy controls (Yu et al., 2008; Filippi et al., 1998, Werring et al., 1999; Rocca et al., 2004). Magnetization transfer imaging provides estimates of macromolecular-bound water assumed to be associated with myelin, however macromolecules are not restricted to myelin and are also strongly influenced by inflammation (Vavasour et al., 1998; Vavasour et al., 2011; Gareau et al., 2000). On the o ther hand, diffusion tensor imaging is non-specific and reflects not only myelin, but also fibre coherence, axonal density and membrane permeability (Beaulieu et al., 2002; Harsan et al., 2006).   An advanced MRI technique called myelin water imaging can be used to quantify the amount of myelin in vivo in white matter tracts and may be the most sensitive and specific MRI technique for measuring changes in myelin. Pre -clinical (Kozlowski et al., 2008; Odrobina et al., 2005; Stanisz et al., 2004) and post -mortem human studies (Laule et al., 2006; Laule et al., 2008) have shown that MWF quantitatively correlates with histological staining of myelin.   Therefore, the goal of this study is to determine if disease burden of lesions regulate the level of damage in otherwise NAWM using myelin water imaging. If the reduction in MWF of the NAWM tracts correlated with lesion volume, it would suggest that the reduction in myelin of normal appearing tissue was driven by the lesions present in the tract; and therefore, the damage would be downstream of the acute lesion. A lack of relationship might indicate that the decrease in myelin in the normal appearing tissue is affected by a disease process independent of the lesions. Therefore, it is important to also look at normal appearing tissue and not just count lesions.  6 2   4.1  METHODS  4.2.1 Participants  53 participants with MS (EDSS: median=2.0, range=0.0 ±4.0, age: mean=39 years, standard deviation (SD )=9  years, 18 males and 35 females) and 10 individuals with NMO were recruited  for this study. However, only seven out of the ten NMO subjects had lesions in the brain (EDSS: median=2.5, range=2.0 -5.0, age: mean=45 years, SD=8  years, 3 males and 4 females). Any subject who met criteria for an NMO spectrum disorder was included and all but three subjects in the NMO group were NMO-IgG positive. Subjects with MS included those with clinically definite MS by the Poser criteria (2001) or the modified McDonald criteria (Polman et al., 2005). All participants with MS were not on treatment at the time of testing. Two participants with NMO were on mycophenolic acid (CellCept®), three on azathioprine (Imuran®), one participant on interferon beta-1a (Rebif 44®), and one on rituximab. No participants were on steroid therapy at least one month prior to entering the study. The University of British Columbia Clinical Research Ethics Board approved all study procedures.   4.2.2 MRI Protocol and Analysis  MRI scans were performed on a Philips 3.0T Achieva scanner (Best, the Netherlands). A 3D T1 -weighted spoiled gradient echo with repetition time (TR) of 6.5ms, echo time (TE) of 3.6ms, flip angle of 13° and 1.7 x 1.7 x 1.7mm voxel size was acquired for segmentation of the tract of interest. An axial combined gradient echo and spin echo (GRASE) for calcul ating MWF T2 relaxation with 32 echoes sequence, TR=1000ms, TE 1 -32=10 -320ms, echo spacing=10ms, 1mm x 1mm x 5mm voxel size reconstructed to 1 mm x 1 mm x 2.5 mm, sensitivity encoding factor=2 and an echo planar imaging factor=3 was acquired for MWF estima tion (Prasloski et 6 3   al., 2012; Neuroimage). The scanned volume consisted of a 10cm slab measured from the top of the brain caudally.  The signal decay curve obtained by the T2 relaxation sequence was modelled by multiple exponential components and the T2 distribution was estimated using non -negative least squares with the extended phase graph algorithm (Whitall et al., 1989; Prasloski et  al., 2012; Magnetic Resonance in Medicine). MWF in each image voxel was computed as the ratio of the area under the T2 distribution with times of 10 -40ms to the total area under the distribution (Laule et al., 2006). This analysis was performed using in -house software on Matlab® 2012 (MathWorks, Natick, Massachusetts, U.S.A).   For lesion identification the T2 -weighted (T2w) and proton density weighted (PDw) scans were obtained from the DOME and OPERA study. For the NMO patients (DOME study) the T2w and PD w scans were acquired with a dual-echo sequence using TE1=8.42ms, TE2=80.0ms and TR=2800.0ms. All of the images from the DOME study have dimensions of 256 x 256 x 60 and voxel size 0.937 x 0.937 x 3.0mm, with no interslice gap.  For the MS patients (OPERA s tudy) the protocol was slightly different where the PDw was acquired with a turbo -spin echo sequence using TE=10ms, TR=200 0.0ms, and turbo spin echo factor (TSE)=3. T2w was also acquired with a turbo-spin echo sequence using TE=80ms, TR=6100ms, and TSE=8. For both PDw and T2w in the OPERA study the dimension was 256 x 256 x 60 and voxel size was 0.98 x 0.98 x 3.0mm, with no interslice gap.    6 4   3D T1 images of each subject were registered to T2 GRASE images using FMRIB's Linear Image Registration Tool (FLIRT; Smith et al., 2004). Automated segmentation of the regions were performed by registration of the corresponding JHU ICBM atlas regions (for corpus callosum, CC and cortico-spinal tract, CST) and Juelich Histological atlas region (for optic radiation, OR) to the T2 GRASE images using a combination of FLIRT and FNIRT (Andersson et al., 2007). The CC mask only included the middle 18.7mm of the corpus callosum (Figure 4. 1). These regions were then manually edited to ensure only white matter was included and overlaid on the MWF. To obtain the NAWM value the lesion masks were subtracted from the MWF CC, CST (Figure 2.1)  and OR (Figure 3.2)  masks and a mean MWF for each region was calculated.   Figure 4.1: The corpus callosum region of interest (in yellow) on T 1 -weighted image in an individual with neuromyelitis optica. A) sagittal view B) axial view C) coronal view.   For the lesion masks, manual identification of the T2w lesions was done with the placement of seed points and the rest of the processing is fully automatic (Tam et al., 201 1; McAusland et al., 2010). After preprocessing the scans (Jones and Wong, 2002; Smith 2002) , an experienced MRI 6 5   radiologist was asked to place one or more seed points to mark the location and approximate extent of each lesion visible on the T2w and PDw scans. The lesion masks were multiplied by the NAWM CC, CST or OR masks to acquire the lesion load. Lesion volume was calculated by multiplying the image resolution (1.7 x 1.7 x 1.7mm) with the lesion load obtained for each region.   4.2.3 Statistical Analysis  For the NAWM MWF measurement of each region, an independent t-test was employed to compare the dependent variable between the two study groups. The relationship between NAWM MWF in the CC, CST, or OR and lesion volume in the particular regions were assessed using a 6SHDUPDQ’VUDQNHGFRUUHODWLRQFRHIILFLHQW for NMO, and MS patients separately and together. Eight pairwise comparisons were conducted and the p -values were adjusted for multiple comparisons using a Hommel correction . P -values ൑ 0.05 were taken to  be significant.  4.3 RESULTS  4.3.1 MWF Assessment  The NAWM MWF in the CC was not different between NMO (mean=0.05 0, SD=0.016, range=0.068 -0.028) and MS patients (mean=0.05 9, SD=0.015, range=0.09 8 -0.024, p=0.1; Figure 4. 2a). Similarly, the CST NAWM MWF were not different between NMO (mean=0.050, SD=0.015, range=0.067 -0.028) and MS (mean=0.059, SD=0.015, range=0.098 -0.024, p=0.2; Figure 4. 2b). However, there was a trend towards a decrease in NAWM MWF of the OR in the NMO group (mean=0.04 8, SD=0.017, range=0.06 8 -0.028) compared to MS (mean=0.059, SD=0.015, range=0.098 -0.024, p = 0.07; Figure 4. 2c).  6 6          Figure 4.2 : Myelin water fraction (MWF) in the A) corpus callosum, B) cortico -spinal tract and C) optic radiation of individuals with neuromyelitis optica (NMO) and multiple sclerosis (MS).  n is the number of subjects for each group.  4.3.2 Correlation Analysis  All 7 NMO subjects had no lesions in the CC tract. There was no significant correlation between lesion volume and NAWM in the CC when looking at the grouped relationship (r(60)= -0.05 , p= 1.0 ; Figure 4. 3a) and MS separately (r(53)= -0.0 9 , p=1.0 ; Figure 4. 3b). No  significant correlation was observed in the MS (r(53)= -0.1 7 , p=0.9 ; Figure 4. 3e)  and NMO subjects (r(7)=0. 73 , p=0.5 ; Figure 4. 3g ) separately or grouped together in the CST (r(60)= -0.06 , p=1.0 ; Figure 4. 3c). Similarly, there was no significant correlation observed in the MS (r(52)= -0. 33 , p=0.2; Figure 4. 3f ) or NMO (r(7)=0.51 , p=0. 9 ; Figure 4. 3h) population alone or grouped in the OR (r(59)= -0.1 9 , p=0.9 ; Figure 4. 3d).   C o r p u s  C a l lo s u mMWFN M O M S0 . 0 00 . 0 50 . 1 00 . 1 5p = 0 .1n  =  7 n  =  5 3C o r t ic o - s p in a l  T r a c tMWFN M O M S0 . 0 00 . 0 50 . 1 00 . 1 5p = 0 .2n  =  7 n  =  5 3O p t ic   R a d i a t io nMWFN M O M S0 . 0 00 . 0 50 . 1 00 . 1 5p = 0 .0 7n  =  7 n  =  5 2A B C6 7   C C  M W F  v s  L e s io n   V o lu m eL e s io n  V o lu m e  (m m3)MWF0 5 0 1 0 0 1 5 00 . 0 00 . 0 50 . 1 00 . 1 5M SN M Or (6 0 )= -0 .0 5 , p = 1 .0M S :  C C  M W F  v s  L e s io n lu m eL s io n  V o lu m e  (m3)MWF0 5 0 1 0 0 1 5 0. 0 0. 0 50 . 1 00 . 1 5r (5 3 -0 .0 9 , p = 1 .0C S T  M W F  v s  L e s io n   V o lu m eL e s io n  V o lu m e  (m m3)MWF0 2 0 0 4 0 0 6 0 0 8 0 00 . 0 00 . 0 50 . 1 00 . 1 5M SN M Or (6 0 )= -0 .0 6 , p = 1 .0O R  M W F  v s  L e s io n   V o lu m eL e s io n  V o lu m e  ( m3)MWF0 5 0 0 1 0 0 0 1 5 0 00 . 0 00 . 0 50 . 1 00 . 1 5M SN M Or (5 9 )= -0 .1 9 , p = 0 .9M S :  C S T  M W F  v s  L e s io n  V o lu m eL e s io n  V o lu m e  (m m3)MWF0 2 0 0 4 0 0 6 0 0 8 0 00 . 0 00 . 0 50 . 1 00 . 1 5r (5 3 )= -0 .1 7 , p = 0 .9M S :  O R  M W F  v s  L e s io n   V o lu m eL e s io n  V o lu m e  (m m3)MWF0 5 0 0 1 0 1 5 0 0. 00 . 0 5. 1 00 . 1 5r (5 2 )= -0 .3 3 , p = 0 .2N M O :  C S T  M W F  v s  L e s io n  V o lu m eL e s io n  V o lu m e  (m m3)MWF0 1 0 2 0 3 0 4 0 5 00 . 0 00 . 0 20 . 0 40 . 0 60 . 0 80 . 1 0r (7 )= 0 .7 3 , p = 0 .5N M O :  O R  M W F  v s  L e s io n   V o lu m eL e s io n  V o lu m e  (m m3)MWF0 5 1 0 0 1 5 0. 00 . 0 50 . 1 0. 1 5r (7 )= 0 .5 1 , p = 0 .9A BC DE FG HFigure 4.3 : No correlation between myelin water fraction (MWF)  in the corpus callosum (CC) and lesion volume was observed in grouped analysis (A) or in multiple sclerosis patients separately (B). Similarly no relationship was seen between MWF and lesion volume of the cortico-spinal tract (CST) in the grouped analysis (C) or MS patients (E) and NMO patients separately (G). Similarly, there was no relationship between MWF in the optic radiation (OR) and lesion volume was observed in the grouped analysis (D) or MS patients (F) and NMO subjects (H) separately.  6 8   4.4 DISCUSS ION To our knowledge, this is the first study to look at correlations between lesion volume and NAWM tracts using myelin water imaging in MS and NMO. In NMO, it is believed that there is focal damage, with minor downstream damage due to lesion. In comparison, MS has focal damage in addition to global damage which is partly independent of lesions. If lesions are observed, damage to the NAWM tracts would be expected. However, in MS, regardless of the presence of lesions, there appears to be damage to the NAWM tracts. This study investigated the CST and OR white matter tracts, regions that are commonly affected in both MS and NMO, as well as the CC, an area commonly affected in MS but not as prominently in NMO. No correlations were observed between lesion volume and MWF in the OR, CST or CC NAWM tracts of MS and NMO subjects.   The presence of occult damage to the NAWM in MS patients have been confirmed in previous studies (Cercignani et al., 2001; Rashid et al., 2004).  MR diffusion imaging suggests that NAWM abnormalities may be a result of structural lesions causing Wallerian degeneration to connected areas of NAWM tracts (Werring et al., 1999 ). Post -mortem analysis found a strong negative correlation between regional lesion load and the axonal density in the CC for patients with MS. This suggests that substantial diffuse loss of axons in the NAWM of MS patients is at least partly due to Wallerian degeneration of axons transected in the demyelinating lesions (Evangelou et al., 2000). However, only modest correlations have been observed previously between lesion load and NAWM abnormalities (Filippi et al., 2000; Filippi et al., 2001; Traboulsee et al., 2002) , suggesting that abnormalities in the NAWM tract of MS may be partly 6 9   due to processes independent of lesions. In this study, we found no correlation between lesion volume and MWF in the NAWM tracts for the MS group further supporting this notion.   Previous studies have found no difference between NMO and controls when  looking at the presence of occult brain damage, confirming the notion that brain white matter is relatively spared in NMO (Filippi et al., 1999; Rocca et al., 2004). We expected to see some correlation in NMO, because previous work has suggested that any damage in the white matter (WM) tracts would be driven by lesions and not an independent process. Yu and colleagues (2006) observed lower average fractional anisotrophy (FA; which reflects fiber density, axonal diameter and myelination in white matter) histograms in the WM of relapsing NMO patients compared to healthy controls but only in WM tracts with direct connections to the spinal cord or optic nerve. This suggests that abnormal diffusion of NAWM tracts is likely a result of secondary degeneration caused by lesions in the spinal cord and optic nerve (Yu et al., 2006; Yu et al., 2008). The lack of correlation in this study is likely due to our small sample size , and future work should look at this relationship with a larger group.  Liu and colleagues (2012)  observed wide-spread, subtle cerebral WM changes in NMO, including regions not limited with connections to the spinal cord and optic nerve. However, FA which is most commonly associated with changes in myelin and axons, did not detect any changes in their study. This may be due to the low resolution of the diffusion tensor imaging sequence making their regions of interest less conservative and potentially picking up non-white PDWWHUUHJLRQVLQWKHLUDQDO\VLVȜ1  DQGȜ2 3  were suggested as markers for axon and myelin respectively, but those interpretations are based on animal work (Song et al.,2002; Sun et al., 7 0   2008; Kim et al.,  2006) in acute injury and do not  necessarily translate well into MS and NMO studies (Sun et al., 2006; Lodygensky et al., 2010; Kim et al., 2007 ; Cheng et al., 2011; Xie et al., 2010). Additionally, the NMO population was not tested for NMO -IgG, a highly specific serum biomarker for NMO.   Lesions do not correlate well with disease progression in MS (Barkhof, 2002), therefore a measure sensitive to disease progression but independent of lesions may be used as a marker for progression. Currently, it is unclear if MWF can be used as a marker for progression (Laule et al., 2010), however this study does provide support for th is, since MWF in MS did not correlate with lesion volume. Future work will look into the relationship between disease development and MWF in progressive MS patients.   There were a number of limitations in this study. Different acquisition protocols were used for lesion identification in NMO and MS subjects. However, lesion identification is not a quantitative measure so identification should not depend too heavily on slight differences in intensity, nonetheless, it may result in lesion detection being more sensitive in one group over the other. Another limitation was the small sample size of the NMO group making it difficult to confirm if there was a relationship between lesion volume and MWF in the NAWM. However, NMO was not the main focus of this study for this reason and needs to be addressed in future studies with a larger group. Additionally, NMO patients were on a variety of treatments while the MS group were not on any treatment. This may have explained the lack of difference in the MWF between the two groups in the various white matter tracts observed because the treated NMO patients may have less myelin loss compared to those not on treatment.  7 1   Chapter 5  5  CONCLUSIONS  These pilot studies were exploratory and the results obtained were meant to be hypothesis generating. In chapter 2, structural changes were found in the descending motor output pathway white matter of individuals with NMO along with abnormal recruitment curves, which is a good indicator of axonal connectivity (Siebner and Rothwell, 2003). This provides support for the notion that NMO usually presents with greater spinal cord involvement and has more extensive axonal loss compared to MS (Jacob et al., 2013; Kim et al. 2012).  NMO lesions are generally more destructive than those of MS and are characterized by extensive macrophage infiltration often resulting in damage to all cellular components including axons of both grey and white matter (Jacob et al., 2013). Whereas in MS , plaques form throughout the CNS, but for permanent injury to occur, the accumulation of episodic inflammation needs to go beyond plasticity and repairing capacities (DeLuca et al., 2004).  Limitations of this st udy included the small sample size within each participant group. However, even with only ten subjects per group, a significant difference was found in CST MWF and recruitment curves between NMO, MS and healthy control groups. Another disadvantage was that spinal imaging was not performed. Lesions within the spinal cord could influence the excitability of the cortico-spinal pathway (Love, 2006). Future work should look into the relationship between spinal cord imaging and TMS functional measures in a larger sample size. The combined results from TMS and myelin water imaging suggest that there are both neurophysiological and neuroanatomical changes found in individuals with inflammatory CNS disorders and that joint utilization of these techniques may offer a better understanding of 7 2   the underlying pathophysiology in these diseases  Further support for this hypothesis came from chapter 3 ; the decrease in RNFL thickness, TMV and OR myelin provided evidence that NMO (specifically those with ON history) has more extensive axonal loss and optic nerve involvement than MS. RNFL thin ning and decreased TMV was most pronounced in patients with ON history compared to unaffected eyes suggesting that there may be retrograde degeneration retina secondary to ON. The decrease in OR MWF of patients with ON history was likely due to Wallerian degeneration from damage to the optic nerve. Additionally, the relationship seen between the OCT and MWF measure suggests that damage to one part may cause damage to another part of the visual pathway due to Wallerian or retrograde degeneration. Interestingly, the RNFL thinning and decrease in OR myelin found in MS patients without ON history suggests that MS patients may have damage in the absence of ON, possibly due to underlying disease processes not directly related to the lesions (Zimmermann et al., 201 3; Noval et al., 2011).  Conversely, NMO is predominantly a result of relapses and rarely due to secondary progression or diffuse low grade inflammation throughout normal appearing CNS tissue (Ratchford et al., 2009).  One limitation to our study was that the OCT protocol was collected under slightly different protocols for the TMV  measure in the MS and NMO group. The differences observed between these two groups may be influenced the protocol, although the results obtained in this study were consistent with the RNFL measures and previous research (Frohman et al., 2008; Burkholder et al., 2009; Young et al., 2013; Noval et al., 2011 ) . Future studies should look at comparing clinical measures that are more related to vision such as visual evoked potentials or visual acuity, and also obtain a larger population of NMO subjects without ON history. To the best of our knowledge, no other studies to date have assessed the entire visual pathway through a combination of several imaging techniques in NMO 7 3   (Pfueller and Paul, 2011). Using these imaging techniques to visualize the optic pathway and the anatomical correlates, it can help identify significant differences between NMO and MS. Additionally, it may give further insight into NMO-specific damage processes as well as into trans-synaptic damage processes independent of their underlying condition. OCT and MRI measures may be useful to evaluate disease progression and treatment approaches that promote repair.  In chapter 4, t he lack of relationship between disease burden of lesions and MWF of the NAWM tracts (CC, CST and OR) in MS patients (similar to findings in chapter 3)  suggests that processes independent of lesions may be involved. However, due to the small number of NMO subjects it was difficult to see if damage to the NAWM tracts were a result of secondary degeneration from lesions in the spinal cord and optic nerve (Yu et al., 2006). Future studies should look at the relationship between lesion volume and NAWM tracts in a larger sample of NMO subjects. Interestingly, lesions do not correlate well with disease progression in MS (Barkhof, 2002) therefore, if we find a measure that is sensitive to disease progression independent of lesions it may be used as a marker for progression. Currently, it is unclear if MWF can be used as a marker for progression (Laule et al., 2010), however this study does provide support for this notion si nce MWF in MS did not correlate with lesion volume. Future work should look into the relationship between disease development and MWF in progressive MS patients. Apart from the sample size, another limitation to this study was the different acquisition protocols used for lesion identification in MS and NMO. This may have resulted in more sensitive lesion detection in one group over the other. To our knowledge, this is the first study to look at correlations between lesion volume and NAWM tracts using myelin water imaging in MS and NMO.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0167701/manifest

Comment

Related Items