UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

MRI lesion activity in relapsing-remitting patients with multiple sclerosis Zhao, Guo Jun 1996

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

Item Metadata

Download

Media
831-ubc_1996-0342.pdf [ 8.71MB ]
Metadata
JSON: 831-1.0087162.json
JSON-LD: 831-1.0087162-ld.json
RDF/XML (Pretty): 831-1.0087162-rdf.xml
RDF/JSON: 831-1.0087162-rdf.json
Turtle: 831-1.0087162-turtle.txt
N-Triples: 831-1.0087162-rdf-ntriples.txt
Original Record: 831-1.0087162-source.json
Full Text
831-1.0087162-fulltext.txt
Citation
831-1.0087162.ris

Full Text

MRI LESION ACTIVITY IN RELAPSING-REMITTING PATIENTS WITH MULTIPLE SCLEROSIS by GUO JUN ZHAO M.D. Henan Medical University, China 1982 M.Sc. Henan Medical University, China 1988 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Experimental Medicine) We accept this thesis as conforming to the relfiaired standard THE UNIVERSITY OF BRITISH COLUMBIA May 1996 ®Guo JunZhao, 1996 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Experimental Medicine The University of British Columbia Vancouver, Canada Abstract Multiple sclerosis (MS) is a disease of the central nervous system seen mainly in young adults in which demyelination is the most specific pathological change. Magnetic resonance imaging (MRI) is the most sensitive method for imaging MS lesions. Serial MRI can detect brain lesion activity in MS and provide an important tool for monitoring therapeutic trials. Fifty relapsing-remitting MS patients, as part of a beta-interferon trial, were examined every 6 weeks with MRI for a 2 year period in this study. The patients were randomized into three treatment arms: placebo (N=17), 1.6 mlU (N=17), and 8 mlU (N=16) beta-interferon self-administered subcutaneously every other day. Activity of MS lesions on MRI was defined as new (first appearance), enlarging (increase in size of preexisting lesion), recurrent (reappearance of lesions that had disappeared), and enhancing (lesion enhancement on Tl weighted MRI scan after a contrast agent, gadolinium, injection). Lesion activity as detected in serial MRI examination is much more common than the clinical relapse rate. Study of lesion activity is helpful in understanding MS. This thesis includes serial studies of lesion activity on MRI. The studies are (1) to detect the activity of the individual lesion in order to see if there was any characteristics that distinguished active lesions from each other and from stable ones; (2) to identify the features of enhancing and non-enhancing lesions in shape, size, and location to learn the difference between morphologically active and gadolinium enhancing lesions due to their pathological difference; (3) to describe the growth pattern of enlarging MS lesions as we believe that enlarging lesions differ pathologically from new ones; (4) to compare the sensitivity of gadolinium MRI with unenhanced serial MRI ii in detecting activity of lesions to see which one is the more sensitive technique for detecting active lesions. The findings from the studies disclose features of MS active lesions that provide more information in understanding the natural history of MS. To determine how much MS lesion activity information, as detected with frequent serial MRI, is lost if a less frequent scan interval is used, a study of effects of scanning frequency was carried out. Lesion activity was assessed in a blinded fashion at scan intervals of 6 weeks, 3 months, 6 months, and 1 year. We found that scanning in less than 6 week intervals detects fewer active MS lesions. Active lesions located in the corticospinal tract were studied to see any relationship between MRI activity and clinical course. The study found that there was a positive relationship between active lesions in the corticospinal tract and clinical relapses. To monitor lesion activity as an outcome to measure therapeutic trials, a study of 50 Vancouver patients was conducted. The result provides further evidence that lesion activity measurement is a useful and sensitive tool in a therapeutic trial even with small group of patients. iii TABLE OF CONTENTS Abstract ii Table of Contents iv List of Tables viii List of Figures ix List of Abbreviations x Acknowledgment xii Chapter 1: Multiple Sclerosis 1 1.1. General introduction 1 1.2. Pathology of multiple sclerosis 4 1.3. Blood-brain barrier in multiple sclerosis 10 1.3.1. The structure of the normal B B B 11 1.3.2. The B B B breakdown in M S 13 1.3.3. The possible mechanism of the B B B breakdown 14 Chapter 2: MRI of Multiple Sclerosis 19 2.1. Introduction of M R I 19 2.1.1. Basic physical and biological principles 20 2.1.2. Pulse sequences 21 2.1.3. Signal intensity 22 2.1.4. Gadolinium enhancement in M R I 23 2.2. M R I of multiple sclerosis 24 iv 2.3. The BBB and MRI 29 2.4. Serial MRI as a tool in monitoring therapeutic trials in multiple sclerosis 33 Chapter 3: MRI Active Lesions 38 3.1. Definition of active lesions 38 3.2. Materials and methods in this study 49 3.2.1. Patient population 49 3.2.2. MR scan technique 49 3.2.3. Quantitative analysis of MRI 50 3.2.4. Statistics 50 3.3. Assessment of the individual lesions in RR MS on MRI 51 3.3.1. Introduction 51 3.3.2. Materials and methods 51 3.3.3. Results 52 3.3.4. Discussion 55 3.4. Features of enhancing and non-enhancing lesions of MS on MRI 58 3.4.1. Introduction 58 3.4.2. Materials and methods 58 3.4.3. Results 59 3.4.4. Discussion 63 3.5. The growth pattern of enlarging lesions in MS: observation in serial MRI 67 3.5.1. Introduction 67 3.5.2. Materials and methods 67 3.5.3. Results 68 3.5.4. Discussion 68 3.6. Sensitivity of gadolinium enhanced MRI and serial MRI in detecting activity of lesions in RR MS • 70 3.6.1. Introduction 70 3.6.2. Methods 70 3.6.3. Results 71 3.6.4. Discussion 73 3.7. MR detection of MS lesion activity: effects of scanning frequency 75 3.7.1. Introduction 7 5 3.7.2. Materials and methods 75 3.7.3. Results 76 3.7.4. Discussion 79 3.8. Corticospinal tract lesions in MS: relationship between MRI activity and clinical course 81 3.8.1. Introduction 81 3.8.2. Materials and methods 82 3.8.3. Results 83 3.8.4. Discussion 87 Chapter 4: Monitoring Lesion Activity as An Outcome to Measure Therapeutic Trial 89 4.1. Introduction 89 4.2. Materials and methods 90 4.3. Results 92 vi 4.4. Discussion 100 Chapter 5: Future Application of MRI for the Understanding of MS Pathology 102 5.1. M R spectroscopy in M S 102 5.2. Summary and future direction 105 Reference 108 Appendix A. Expanded Disability Status Scale (EDSS) 125 Appendix B. Functional System 128 vii LIST OF TABLES Table 1: Sequence of events in the evolution of MS lesions 10 Table 2: Margin and shape of lesions 53 Table 3: Location of lesions 54 Table 4: Mean area of lesions 54 Table 5: Location, margin, and shape of lesions 61 Table 6: Signal intensity of enhancing lesions 62 Table 7: Lesion size in 3 groups of lesions 62 Table 8: Location of lesions 71 Table 9: Signal intensity of enhancing lesions 72 Table 10: Mean area of lesions 72 Table 11: Active lesions involving CST above the pons (group 1) 84 Table 12: Active lesions involving the CST above the pons with preexisting lesions in the brainstem (group 2) 85 Table 13: Active lesions not in the CST (group 3) 86 Table 14: Active lesion rate 94 Table 15: New lesion rate 95 Table 16: Enlarging lesion rate 96 Table 17: Recurrent lesion rate 97 Table 18: Percent of scans with activity 100 viii LIST OF FIGURES Figure 1: Pathological appearance of MS lesion 7 Figure 2: The BBB disturbance 30 Figure 3: New lesion 39 Figure 4: Enlarging lesion 41 Figure 5: Recurrent lesion 43 Figure 6: Enhancing lesion 46 Figure 7: Percentage of active lesion versus scan interval 77 Figure 8: Mean number of active lesions in placebo versus treated patients 78 Figure 9: Mean percent of patients with active scans per scan week 98 Figure 10: Means of MRI cumulative MRI active area in the Vancouver cohort over 2 years 99 IX LIST OF ABBREVIATIONS AO albino oxford BBB blood-brain barrier CNS central nervous system CP chronic progressive CT computered tomography CSF cerebrospinal fluid CST corticospinal tract DA dark august EAE experimental allergic encephalomyelitis EC endothelial cell EDSS expanded disability status scale 5-HT serotonin Gd gadolinium Gd-DTPA gadopentetate dimeglumine GPC glycero-3-phosphorylcholine HRP horseradise peroxidase ICAM-1 intercellular adhesion molecule-1 IFNB interferon beta-1 b JL-2 interleukin-2 IR inversion recovery MBP myelin basic protein MIU million international unit MRI magnetic resonance imaging X MRS magnetic resonance spectroscopy MS multiple sclerosis MTI magnetic transfer imaging NAA ./V-acetylaspartate NMR nuclear magnetic resonance NRS neurological rating scale PNS peripheral nervous system PP primary progressive RF radiofrequency RR relapsing-remitting SD standard deviation SE spin echo SP secondary progressive TE echo time TR repetition time X I ACKNOWLEDGMENTS I would like to give special thanks to Dr. Donald W Paty for all his help and encouragement during the course of my graduate work. His patient and constructive criticisms were invaluable. I would like to thank Dr. David KB Li for his endless support and reassurance throughout the years of my study. I am grateful to Dr. Robert A Koopmans and Dr. Beth L Tanton for their contribution to this study. I would like to also thank Wendy Morrision and Jill Nelson for providing me with their clinical data, Brenda Rhodes and Andrea Cook for their excellence in outlining MS lesions, and all technologists for providing the MRI data. I thank Mrs. Katie Koopmans for editing and preparing this manuscript. Last but not least, I wish to thank all the MS patients who participated so willingly and with excellent compliance and made many personal sacrifices in order for this study to be completed. X l l Chapter 1 Multiple Sclerosis To understand the active lesions detected in magnetic resonance imaging (MRI) , one should be acquainted with the basic properties of myelin, demyelination, gadolinium (Gd) enhancement in M R I and multiple sclerosis (MS). Based on this knowledge it is possible to discuss features of active lesions and the relationship between lesion activity and clinical status. Verification of these relationships can be obtained by postmortem M R I of M S brains. Serial M R I also can provide much information, which wi l l be discussed in this thesis. 1.1. General introduction Definition and Classification Nerve conduction depends on adequate axonal sheathing in order to maintain nerve saltatory conduction. In the peripheral nervous system (PNS) sheathing is performed by the Schwann cells while in the central nervous system (CNS) sheathing is performed by oligodendrocytes. In both cases nerve sheaths consist of myelin bi-layers. Insufficient myelination is associated with incapacitated nerve conduction. A shortage or abnormal function of myelin can have a primary (congenital) cause (dysmyelination) or can have a secondary (acquired) cause (demyelination). Demyelination of the peripheral nervous system occurs in chronic inflammatory demyelinating polyneuropathy and Guillain-Barre syndrome. Several demyelinating diseases of the C N S are known, of which M S is the most common. B y definition M S is a demyelinating disease of the C N S characterized by various plaque formation. 1 Clinical Presentation M S is clinically characterized by episodes of focal disorder of the optic nerves, spinal cord, and brain, which remit to a varying extent and recur over a period of many years. The clinical manifestations are protean, being determined by the varied location and extent of the foci of demyelination; nevertheless, the lesions tend to have a predilection for certain portions of the C N S , resulting in characteristic complexes of symptoms and signs that can often be readily recognized. Approximately 95 percent of cases of M S begin between the ages of 10 to 50 years and 5 to 6 percent of cases start between ages 50 and 60 years [Broman et al., 1981]. In general, patients who develop M S while young tend to have more intracranial lesions and a higher frequency of exacerbations, whereas patients with late-onset M S are more likely to have a progressive spinal cord disease [Andersen et al., 1980]. The mean age of onset of M S is about 30 years in most studies. The reason for this age susceptibility is unknown. The female to male ratio of cases is 2:1. The highest ratios noted in prevalence studies may be due to a greater mortality rate in males with M S [Sibley and Paty,1981]. Genetic factors have a causal role in M S . Several twin studies have been conducted to estimate the M S concordance rates between mono- and dizygotic twins [MacKay and Myrianthopoulos, 1966; Cendrowski, 1968; Bobovick et al., 1978; Will iams et al., 1980; Currier and Eldridge 1982]. Their conflicting results may be explained, at least in some studies, by biases such as public appeals, which lead to an excess of monozygotic pairs. In a large Canadian population-based study [Ebers et al., 1986] (where biases were minimized), higher concordance rates in monozygotic twins (25.9%, 27 pairs) than in dizygotic twins (2.3%, 43 pairs) were observed. A 7.5 years follow-up study [Sadovnick et al., 1993] of the population-based series provided further evidence. The survey in the British Isles had similar results [Mumford et al., 1994]. Conversely, a French study of 54 twin pairs (17 monozygotic and 37 dizygotic pairs) revealed no significant difference, and 2 the observed clinical concordance rate was low in both groups [French research group on M S , 1992]. These discrepancies might be explained by differences in the population risk according to the geographical area on population susceptibility. Relatives of M S patients have an increased risk of M S . 10-20% of M S patients have an affected relative. In high-risk areas, siblings of M S patients have an estimated risk of 1.2% compared to 0.1% in the general population. Subclinical demyelinating lesions may occur in the brains of asymptomatic individuals, and the first-degree relatives of M S patients are at particular risk. In recent studies using M R I , there is an indication that subclinical M S may be present in apparently normal members of multiplex families [Lynch et al., 1990; Tienari et al., 1992]. M S is unequally distributed over the world. There are geographic differences in incidence and prevalence that cannot be genetically explained. Overall the disease is more common in northern America, northern Europe, and New Zealand. In British Columbia, Canada, M S is the most common neurological disorder in young adults; its prevalence is 117.2 per 100,000 population [Sweeney et al, 1986]. Prevalence studies for migrants from high to low risk areas indicate the age of adolescence to be critical for risk retention: those migrating under age 15 years acquire the lower risk of their new residence. The migrant data, defines M S as an acquired exogenous (environmental) disease whose acquisition in ordinary circumstance takes place years before clinical onset. The most common initial clinical manifestations are sensory (40%) and visual (35%), followed by motor (21%), brainstem (16%), cerebellar (15%) and bladder (4%) symptoms [de Graaf, 1988]. Signs and symptoms in M S patients include fatigue, optic neuritis, internuclear ophthalmoplegia, Lhermitte's sign, chronic myelopathies and urge-incontinence. The frequent occurrence of psychiatric and cognitive defects in M S has recently been recognized [Ron, 1992; Stenager et al., 1990; Minden and Schiffer, 1990]. Fixed disability usually occurs at a given moment in time and is frequently expressed as a disability score. The most widely used disability score is the Expanded Disability Status 3 Scale (EDSS) score [Kurtzke, 1983] [Appendix A ] . On this 10-point scale, 0 means no abnormalities, 6 or more the inability to walk without aid, and 10 deceased. The interrater variability in assessing the E D S S is 0.5 point and subsequently a change of 1 point or more is needed to demonstrate a significant change in disability; this variation is even more marked at the lower of the scale. Severity of exacerbations can be classified by the Neurological Rating Scale (NRS) [Sipe et al., 1984]. The N R S is based on assessment of each component of the neurologic examination and reflects overall neurologic function. The assignment of points in the N R S receives the full "normal" point value, with a progressive loss in points for mild (-1; 1+), moderate (-2; 2+), or severe (-3, -4; 3+, 4+) involvement. Severe (-3; 3+) and maximal (-4; 4+) deficits are scored as severe on the N R S . Several disease courses can be seen in M S . The initial course usually is a relapsing-remitting (RR) one, followed in 2 out of 3 patients by a secondary progressive (SP) course. The progression has been defined as a development of steadily increasing disability over a period of at least 12 months. One third of the patients with a R R disease course, still have an E D S S score of 3 or less after 10 or more years, and their condition is referred to as benign M S . A small percentage of the patients (5 - 10%) wi l l follow a progressive course from the onset without relapse or remissions, referred to as primary progressive (PP) M S . A n single episode of clinical disease activity is referred to as isolated syndromes. 1.2. Pathology of multiple sclerosis The presence of multiple gray sclerotic plaques throughout the brain and spinal cord of patients dying with longstanding multiple sclerosis is so striking that the pathological appearances have endowed the disease with its commonly used name. Demyelination 4 refers to the removal of apparently normal myelin from axons of the C N S , usually against a background of perivenular infiltration by small lymphocytes, plasma cells, and large mononuclear cells. Biochemically, myelin stands apart from other membranes in both possessing its unique proteins and its unusually high percentage (70%) of lipids. Thirty percent of the membrane is made up of protein, one component in particular, myelin basic protein ( M B P ) , has characteristic properties that render the sheath highly susceptible to immunologic attack. M B P , an 18,000 molecular weight, 170 amino acid molecule, is cross-reactive and highly encephalitogenic when inoculated into most species. Even short sequences of this molecule are encephalitogenic, and it has been demonstrated that different polypeptides along the molecule influence cell-mediated and antibody-mediated immunity to the membrane. In normal white matter, myelin is the major structural element, and related to its biochemistry, a wide variety of l ipid stains exist which clearly differentiate white from gray matter. Ultrastructurally, white matter is made up of densely packed, electron-dense, lipid-rich myelin sheaths surrounding pale-staining axons. Each sheath has the characteristic spiral lamellar pattern. The sheath of one fiber is often contiguous with and separated from its neighbor by a single myelin period. Groups of unmyelinated axons are sometimes seen. Nerve fibers often lie in groups separated by fibrous astrocytic septae and interfascicular oligodendrocytes. B lood vessels in normal white matter display little or no space between the closely apposed basal laminae of the endothelial cells and the glia limitans, except for that occupied by an occasional pericyte, macrophage, smooth muscle cell, fibroblast, or collagen pocket. In the majority of cases, the M S brain displays no outward abnormalities. The brain weight is usually within normal limits. The optic nerves are frequently atrophic and display gray zones from which myelin has been depleted. Coronal section of the brain reveals large lesions which vary in appearance, texture, size, shape, number, and topography (figure 1). Variations in appearance and texture reflect the age and activity of 5 the lesions. Pink, soft lesions indicate areas of recent activity, while older lesions are gray, sometimes rather translucent, gelatinous, and firm to the touch. Lesions vary in size from less than a millimeter to several centimeters across. Lesions may unite and arborize throughout the white matter, presumably along the vasculature. Some cases of M S wi l l display large confluent lesions. The location of M S lesions usually defies clinical and anatomical explanation. The entire central neuraxis is vulnerable, and in the vast majority of cases the cerebrum is involved. The heaviest concentration of periventricular lesions appears to occur in relationship to subependymal vasculature. One commonly involved area is the angle between the caudate nucleus and the corpus callosum. Smaller lesions commonly occur in the white matter at the tips of gyri where they may spill over into gray matter. Lesions may also be found in the striatum, pallidum, and the thalamus. Lesions also occur around the margins of the third ventricle and hypothalamic pathways. The lesions in disseminated sclerosis, wherever they are situated, are distributed evidently without any relation to nerve tracts. Within the cerebrum the veins pass towards the wall of the ventricles and the choroid plexus towards the veins of Galen, and have in this way a distribution altogether different from that of the arteries. It is known that the lesions are deposited in relation to the distribution of the veins and to the walls of the ventricles. The lesions of relapsing remitting M S fall into four categorizes: 1) chronic, inactive plaques; 2) chronic, active plaques; 3) shadow plaques, and 4) acute plaques. The most common location of inactive chronic plaques are periventricular, the optic nerves and chiasm, and the spinal cord. Mye l in stains wi l l reveal demyelinated areas sharply demarcated from the adjacent myelinated white matter, imparting upon them a punched-out appearance. A x o n stains wi l l disclose a moderate decrease in the density of axons toward the periphery of the plaque, but deeper areas wi l l be more depleted. The loss of myelin from a fiber is abrupt at the lesion edge and is visualized as a myelinated 6 segment becoming naked and continuing into the lesion as a demyelinated fiber. Frequently, even in silent lesions, foamy macrophages can be seen. A s a result of loss of parenchymal elements and intense gliosis, considerable atrophy of affected regions is not uncommon. Histologically, most inactive lesions have a rather acellular appearance, consisting of a fibrillary astroglial feltwork that is responsible for the sclerosis, color, and texture of the tissue. The stroma of most chronic lesions contains a few sudanophilic, lipid-laden macrophages, fat granule cells, compound granular corpuscles, foamy cells, and histiocytes, which are randomly scattered throughout the parenchyma or exist in small collections around blood vessels. Inflammatory cells are usually rare within silent lesions. Plasma cells are the most common non-gliotic cellular element, and small lymphocytes are rare. Oligodendroglial depletion is severe in chronic lesions of M S . A chronic, active M S lesion is regarded as an established lesion along the edges of which is a broad zone of perivascular and parenchymal inflammation together with diffuse ongoing demyelination. Macrophage activity, glial hypertrophy, hypercellularity, and some edema are in evidence. On gross examination, such a lesion would appear to have a pink or white tinge about its margins, corresponding to inflammation and macrophage activity, respectively. Chronic active lesions have a less well demarcated margin on myelin staining. Small lesions can often be found centered on small veins within the white matter proper. Presumably such small demyelinating foci wi l l eventually coalesce to form larger lesions. The adjacent myelinated white matter is often hypercellular and contains an increased number of proliferated oligodendrocytes. Foamy macrophages are invariably found throughout the normal white matter far away from lesions. Such a phenomenon might be representative of persistent low-level activity. Within the center of active lesions, only a few T cells and many la-positive macrophages and B cells are present. Throughout active lesions, astrocytic hypertrophy is common, particular in the white matter adjacent to the lesion. Hypertrophic astrocytes extend for some distance into the surrounding white matter. It was seen that despite the presence of 8 ongoing demyelination, inflammation, and macrophages containing recognizable myelin debris, surviving and proliferating oligodendroglia are common. Shadow plaques contain diffusely distributed, thinly myelinated fibers. The consensus is that they represent areas of C N S remyelination. This conclusion is based on a preponderance of large diameter axons with thinner than normal myelin sheaths, attenuated sheaths containing aberrant collections of oligodendroglial cytoplasm, and a modest increaseln the number of oligodendrocytes. Evidence of previous damage can be found in the form of a moderate background gliosis, an abundance of nonspecific corpora amylacea, a few fibrotic blood vessels, and the occasional foamy macrophage. Inflammation cells are absent, as are prominent fibrous astroglial and macrophage responses. The major pattern of myelin disruption in acute lesion is extensive extracellular vesicular dissolution, a change seen light microscopically in myelin-stained sections as thickening and pallor of the sheath leading to fragmentation. Oligodendrocytes may survive in large numbers in acute plaques. Oligodendrocytes loss in M S is probably secondary to myelin damage which is a proliferative response which follows some weeks after acute destruction of both myelin and oligodendrocytes [Prineas, 1985; 1994]. Inflammation is usually seen in acute lesions. Plasma cells are most numerous in fresh lesions. In acute lesions in the acute form of disease, extensive remyelination can begin within a few days or weeks. It possibly contributes to clinical remission, and may even result in disappearance of the plaque. A s the cause of M S remains unclear, the exact sequence of events in the evolution of M S lesions can only be hinted at (Table 1). Prineas [1985] summarizes the following concepts: 1) both myelin and oligodendrocytes are targets of 2) macrophages and their proteolytic enzymes, 3) after being recognized by oligoclonal IgG antigens or T-lymphocytes, and 4) remyelination occurs in acute and chronic plaques. 9 Table 1: Sequence of events in the evolution of M S lesions T-cell activation with production of cytokines Blood-brain barrier breakdown Inflammation Edema Demyelination Gliosis astrocytosis, proliferation of microglia (foam cells) and oligodendroglia Remyelination Axonal loss 1.3. Blood-brain barrier in multiple sclerosis The brain requires a stable internal environment different from that of other organs of the body in order to function at its highest integrative level. Small fluctuations in extracellular fluid concentrations of hormones, amino acids, nutrients, vitamins, and ions are tolerated by the peripheral organs. However, the C N S has a unique microenvironment, a disturbance of which would interfere with the processes of the C N S integration. This unique microenvironment of the C N S demands a dynamic homeostasis in order to insure that individual neurons, synapses, and neuronal systems may receive, process, store, transfer, integrate, retrieve, and utilize the billions of bits of information essential to normal neurological function. Furthermore, the entire internal environment of 1 0 the C N S , including its supportive tissues, must be maintained in isolation from that of the blood. The evolution in parallel of the highest integrative level of the C N S function and the most efficient blood-brain barrier ( B B B ) is supportive of the hypothesis that the homeostasis and isolation of the extracellular fluid microenvironment of central neurons facilitates brain integrative functions [Cserr and Bundgaard, 1986]. Thus far, all species that have a B B B are able to achieve and maintain both isolation and homeostasis of the internal environment of the brain. The B B B integrity is essential to normal brain function. It has been noted that there is striking local damage to the B B B in M S . The damage to the B B B may allow the cellular and humoral components in blood to penetrate the endothelium and induce the demyelinating changes in the C N S . M R I , as one of the modern techniques for brain imaging can easily detect the presence or absence of a focal disturbance in the B B B . 1.3.1 The structure of the normal B B B Brain capillaries are relatively impermeable as compared with capillaries of most other regions of the body. Characteristically, typical cerebral capillaries are composed of a single layer of endothelial cells (ECs). A narrow, uninterrupted basal membrane surrounds the outer surface of ECs . The end-feet of astrocytes may lie directly upon the perivascular basal membrane. These processes are frequently separated from the lamina by a narrow intercellular space. The astrocytic ensheathment of the perivascular basal membrane is discontinuous because of the frequent presence of gaps or space between adjacent astrocytic end-feet [Pollay and Robert, 1980]. The crucial roles of the B B B are contributed by the E C s and their continuous tight intercellular junctions. The electron microscopic studies [Reese and Karnovsky, 1967; Brightman and Reese, 1969] clearly showed overlapping lateral edges of endothelial cells 11 sealed together by tight junctions that formed a continuous belt around the capillary at each interendothelial junction. On the luminal side of the tight junction complex molecules of an electron dense marker could be visualized after intravenous injection in the proximal portion of the cleft, however their passage is usually stopped abruptly at the first tight junction. Also, following administration, molecules of the marker could be visualized in the abluminal basal lamina and into the first abluminal portion of the interendothelial cleft; again, the electron dense marker molecules were abruptly blocked at the first tight junction. The width of the tight junction along the interendothelial cleft, as measured in electron micrographs, is less than twice the width of the endothelial cell luminal or abluminal membrane, which suggests that fusion of apposing membranes was complete and permanent. The capillary ECs do not show transendothelial channels and have a very low frequency of pinocytotic vesicles under normal conditions. Transcapillary exchange of vascular endothelium in cardiac and skeletal muscle had been shown to transport material by vesicular filling on the capillary luminal side and vesicular discharging on the abluminal side of the capillary. However, there is no evidence of discharge of Horseradish peroxidase (HRP) from the luminal to the abluminal side of brain endothelium [Reese and Karnovsky, 1967]. Bundgarrd [1986] has shown electron microscopically, by three-dimensional reconstructions based on consecutive ultrathin sections, that vesicular structures observed in brain endothelium are points of invagination on the cell membrane. The B B B prevents the free diffusion of circulating molecules and cells into brain interstitial space [Pardridge et al., 1986]. The rate that substances penetrate the B B B is found to be related to molecular size, l ipid solubility, and the presence of a specific carrier-mediated transport system. However, the barrier function of the B B B may be damaged under some diseases, such as, in M S . 12 1.3.2 The BBB breakdown in MS The early lesions of M S usually developed around small cerebral blood vessels. The observation has long been considered to be an important clue to the pathogenesis of M S disease [Adams, 1978]. Using trypan blue as a indicator, Broman [1964] demonstrated that the trypan blue staining of the sclerotic plaque of brain appeared uniform but actually included a staining of veins but not of arteries. Leakage of plasma proteins had been found [Broman, 1964; Gay and Esir i , 1991]. The leakage was most marked around capillaries and the small veins of the venocapillary junctions. The E C s of the vessels within acute M S plaques frequently contained many vesicles. Albumin is a non-CNS protein synthesized only in the liver. The concentration of albumin in the C S F , corrected for intravascular albumin, is used to help evaluate the B B B in M S . Tourtellotte and M a [1978] revealed that albumin could diffuse through the extracellular space and sink into the C S F once the albumin had penetrated the disrupted B B B . Hence, a significantly elevated C S F albumin concentration reflects generalized damage of the B B B . Some have observed, the frequency of the B B B alteration to be 23% for M S patients [Tourtellotte and M a , 1978; Eickhoff et al., 1977]. This frequency differs from the study of Annunziata and co-workers [1988] who show B B B impairment in 35% of patients with M S and blood-retina barrier impairment in 45% of M S patients. The function of the B B B to some degree inflences the IgG content of C S F [Pardridge et al., 1986]. The elevation of IgG in C S F may be influenced by the concentration of IgG in serum, by damage of the B B B , and by IgG synthesis within the C N S . Therefore, a simultaneous protein analysis of serum and C S F is necessary to differentiate among these three conditions. B B B permeability to radiotracers of different molecular sizes has been studied in experimental allergic encephalomyelitis (EAE) , an animal model of M S , sensitized using a tissue sampling technique [Juhler et al., 1984]. It has been found that 13 the B B B permeability to small molecules (Na + and Cl" ) increases significantly, preceding clinical symptoms by one day in the lumbar spinal cord and to coincide with the onset of clinical disease in other regions. In all regions, increased B B B permeability preceded the occurrence of histological lesions. The increase of the B B B permeability to sucrose was found to be only slight, but no permeability increase to insulin could be demonstrated [Juhler et al., 1984]. This study can not only confirm the B B B damage in E A E , but also demonstrate evidence of the relationship between the clinical picture and B B B permeability. 1.3.3. The possible mechanism of the B B B breakdown The mechanism of the B B B damage in M S and E A E is at present unclear. There are some suggestions on how to induce disruption of the B B B . These suggestions include manitol, vasoactive amines and/or other humoral or cellular components in the immune response which can affect the B B B . Serotonin (5-HT) and histamine have shown evidence of increasing the permeability of the B B B . After 5-HT is perfused through the cerebral ventricular system, the permeability across cerebral vessels is found to be increased [Westergard, 1978]. However, the ECs are intact. H R P which is employed as a marker of any increase of the B B B alteration does not form a continuous line between ECs , from the vessel lumen to the subendothelial B M . Nevertheless, several vesicles, filled with H R P have been observed in the cytoplasm of ECs . Freely situated HRP-containing vesicles are also found. Exposure of conscious young rats to 4 h heat stress at 38 ° C is associated with increased B B B permeability and with increased plasma and brain 5-HT level [Sharma and Dey, 1987]. Histamine in the regulation of the B B B was studied [Gross et al, 1981; Gulati et al, 1985]. Intra-arterial administration of histamine produced increases in permeability of the B B B in rat cerebral cortex. The response was present in many regions 14 of cortex and was mediated by H 2 receptors. Histamine may produce parenchyma edema which is concentration-dependent. However, these studies are not on E A E or M S . The histamine level in plasma or C S F in E A E animal model or M S patients is unknown. That is, vasoactive aminic mechanisms which are employed to explain the B B B damage on either E A E or M S are not completely understood. While immune reactions against specific endothelial autoantigens may in some instances contribute to vascular damage, cytokines and other inflammatory factors are the central mediators in the regulation of endothelial permeability for cells and macromolecules [Lassmann et al., 1991]. Interleukin-2 (IL-2) may be produced by activated T cells. It has been found that levels of IL-2 in the blood and brain of E A E are increased. Other studies [Damle et al., 1987; Alexander et al., 1989; Vukmanovic et al., 1989; Ell ison et al., 1987] showed that IL-2 may induce B B B dysfunction. When adoptive immunotherapeutic trials in patients with brain tumors with IL-2-activated killer cells were studied, Damle, et al. [1987] found that infusion of high doses of IL-2 caused systemic toxicity in patients with tumor and experimental animals resulting in the development of a vascular leakage syndrome. In a dose-dependent manner, IL-2 caused lymphocytes to strongly adhere to ECs and IL-2-activated killer cells may cause lysis of ECs . Ell ison and co-workers [1987] used cats who were infused with recombinant IL-2 (rIL-2) and showed the B B B alteration in multiple foci throughout the brain, being most prominent within white matter regions. In contrast, Alexander and associates [1989] demonstrated that parenteral injection of IL-2 increased B B B disruption in tumor-bearing rat brain but did not increase the vascular permeability of normal brain after comparing rats with intracerebral 9L gliosarcomas to control group who were administrated high-dose parenteral IL-2. Using two strains of rats, Albino Oxford (AO) and Dark August (DA) which have a different capacity to produce IL-2 ( A O is lower than D A ) , it was shown [Hickey and Kimura, 1988; Kinrichs et al., 1987] that A O rats exhibited lower susceptibility to the induction of E A E , and IL-2 production by their spleen and lymph 15 node cells was also significantly lower. Hence, we believe that IL-2 in the blood is one of the factors which can induce B B B disruption. The mechanism of IL-2 on E C s is still to be elucidated. An t i -EC antibodies may be present in patients with M S and rheumatoid arthritis [Tanaka et al., 1987; Tsukada et al., 1989; Heurkeas et al., 1989]. In the sera of patients with M S , the concentrations of IgG which bind to cerebral E C s and of immune complexes were significantly increased using cultured brain ECs . The levels of IgG binding to ECs were still increased in the sera of exacerbated M S patients after blocking Fc receptors compared to these of controls. The antibodies to each protein fraction extracted from the rat cerebral endothelial cell membrane were studied in patients with M S [Tsukada et al., 1989]. The patients with active relapsing M S displayed significantly higher levels of IgG binding to the E C membrane fraction than did the controls. IgG antibodies reactive with ECs were also found in patients with rheumatoid arthritis [Heurkeas et al., 1989]. The results showed that anti-EC antibodies and immune complexes were present in cases of M S and that they may play a pathogenetic role in the B B B damage. The extent to which antigen-specific recognition events may take place on the luminal surface of cerebral vascular endothelium in vivo remains controversial. A l l encephalitogenic T lymphocyte lines recognize their target autoantigens, M B P , in the molecular context of class II products of the M H C , Ia determinants. Thus, in order to immunogenically present protein (auto-)-antigens, E C s must be inducible to express Ia determinants. After stimulation with y-interferon, B B B endothelia and astrocytes express class I at least in some in vitro models class II M H C antigens [Traugott et al., 1988; Lassmann et al., 1991; Frohman et al., 1988]. Some [Lassmann et al., 1991; Wekerle, 1988] suggested that antigen presentation by ECs may be incomplete and may lead to T-cell anergy and/or to actual lysis of the endothelium. 1 6 It has been observed in E A E that T cells in the circulating blood can cross the B B B under electron microscopy [Dorovini-Zis et al., 1992], but, the mechanism by which T cells and macrophages in the peripheral blood initially enter the brain is still unclear. Several studies have hypothesized [Frohman et al., 1989; Fontana and Fierz, 1985] over the mechanisms in which T cells induced B B B damage, and then crossed the barrier. Some authors [Wong and Dorovini-Zis, 1992; Wi lcox et al., 1990; Frohman et al., 1989] suggested that because T cells had the cell surface receptor for intercellular adhesion molecu le - l ( ICAM-l ) , and ECs can be induced to express I C A M - 1 , the adhesion interaction between these two cell types may regulate the margination of T cells before they cross the B B B and entered the brain. Astrocytes can play a regulatory role connecting the brain with the immune response during the B B B disruption. Astrocytes may transfer information because the astrocytes can induce the B B B properties in ECs . Initially, astrocytes could be involved in bidirectional transport of antigens between ECs and brain parenchyma. This can lead to exposure of antigens on the surface of E C s which can be recognized by T cells. The interaction between T cells and ECs could lead to the disruption of the B B B . Lymphocyte transmigration through the B B B is not related to antigen presentation and specificity but critically depends on whether or not the B B B is damaged. After B B B disruption, using hyperosmotic mannitol, lymphocytes can be detected in the brain parenchyma of normal rats [Kajiwara et al., 1990]. MBP-specif ic T cells did not require MHC-compatible ECs in inducibility of E A E in the model of bone marrow chimeras [Hickey and Kimura, 1988; Kinrichs et al., 1987], but compatibility with the transferred bone marrow progeny was essential. On the other hand, the activated state of T cells is important for their transmigration through the B B B . Using antigen-specific activated T cell lines which are labelled in vitro with [ 1 4C]thymidine, encephalitogenic MBP-specific T cells can cross the B B B within 24 h after i.v. injection. Therefore, this finding is evidence that lymphocytes can open the B B B themselves when they adhere the E C s with 17 or without l a antigen. Lymphocytes, monocytes, and some cytokines can enter the brain parenchyma through the damaged B B B to produce local inflammation. Eventually demyelination occurs. The B B B damage in M S and E A E is due to a complicated process and is a result of interactions between multiple factors. M H C class II antigen on the ECs and astrocytes and humoral and cellular components in the peripheral blood can inevitably take part in B B B disruption, but, la antigens on the ECs and astrocytes, anti-EC antibodies, and T cells probably play a more important role. M S is probably a kind of B B B disease. The occurrence of M S may be a consequence of the B B B disruption caused by some factors located outside the brain. Burnham, et al. [1991] using high-dose steroids in M S patients, provided evidence that steroids may reduce the B B B disruption in a more specific manner. Enhancement on M R I can detect the B B B disruption and the improvement in the B B B disruption is correlated with clinical improvement. More studies on the relationship between the B B B breakdown and clinical aspects of M S should be done. It is important to identify drugs which can maintain B B B integrity therefore improving the symptoms of M S in the hope of eventually finding a complete cure. 18 Chapter 2 MRI of Multiple sclerosis 2.1. Introduction of MRI M R I is the latest in a long and imaginative series of technological advances which have been put to work in the service of medical diagnosis. Although the images produced by this technique are superficially similar to computered tomography (CT), the physical and biological principles involved are very different. M R I is based on the physical phenomenon called nuclear magnetic resonance ( N M R ) . N M R was described in 1946 by Bloch and Purcell working independently [Bloch and Hansen, 1946; Purcell et al. 1946]. They discovered that certain atomic nuclei, when placed in a uniform magnetic field and subjected to a brief pulse of radiofrequency, emit a pulse of radiofrequency (RF) in response. This resonance can be measured and contains information about the stimulated nuclei. In the early seventies, Damadian [1971], Lauterbur [1973] and other researchers [Mansfield and Grannell, 1973] described methods of using these physical concepts to produce images representing body tissues in a manner similar to that of C T . The patient is placed inside a powerful magnet and a series of radiofrequency pulses are initiated. The body tissues contain atomic nuclei which behave as 'small magnets' and respond at resonance to produce a R F signal which is detected and stored. B y varying the magnetic field and the R F pulse all the resultant data can be processed to produce images of the body tissues. Since the hydrogen nucleus (proton) produces the strongest signal, and is found in all body tissues, this is the nucleus currently used for mist imaging. It is 19 possible to detect other nuclei, such as phosphorus, in the body, but imaging from these nuclei is still only in the research phase. 2.1.1. Basic physical and biological principles M R I is based on the phenomenon that nuclei of certain atoms respond to a magnetic field by aligning with or against it. This process is called magnetization. The hydrogen nuclei in water and short-chain fatty acids, which are abundant in l iving tissue, undergo magnetization. The number of nuclei aligned with the magnetic field is slightly greater than that aligned against i t , thus, a net magnetization is created that can be represented as a vector in the direction of the magnetic field. The magnitude of the vector is related to the proton density. The magnetization vector is oriented in the direction of the magnetic field, which is designated the longitudinal direction. The direction of the magnetization vector may be altered by the addition of energy in the form of an R F pulse of appropriate frequency. In the most common type of M R I , a R F pulse with enough energy to rotate the magnetization vector 90° from the static magnetic field is applied. This "90° pulse" changes the direction of the vector from the longitudinal plane into the transverse plane. When the R F pulse is terminated, the hydrogen nuclei wi l l begin to realign in the direction of the external magnetic field. The rate at which the realignment is reestablished depends on the rate that the added energy is dissipated to the surrounding environment. The dissipation mechanism is referred to as longitudinal relaxation, and the time required for 63% of the magnetization vector to return to the longitudinal direction from the transverse plane is designed T p the longitudinal relaxation time. While the magnetization vector is in the transverse plane, it rotates around the axis of the magnetic field and induces a current in a receiver coi l . The amount of current generated initially is directly related to the size of the magnetization vector in the 20 transverse plane. Since the magnetization vector is the sum of many nuclei rotating at slightly different frequencies, the magnitude of the vector and the signal strength diminishes as the nuclei lose coherence or fan out within the transverse plane. The rate at which coherence is lost is designated transverse relaxation. The time at which 63% of the transverse magnetization dissipates is defined as T2, the transverse relaxation time. The loss of coherence is accelerated by inhomogeneities in the external magnetic field. This accelerated loss can be compensated for by using a spin-echo (SE) pulse sequence. In the SE sequence coherence is reestablished by using a pulse of radio waves, with twice the energy of the 90° pulse, called a refocusing, or 180° pulse. This pulse flips the nuclei 180° in the transverse plane and subsequently refocuses the transverse magnetization vector [Osborn, 1988]. 2.1.2. Pulse sequences A pulse sequence consists of one or more pulses of radio waves. A pulse sequence is repeated typically 128 or 256 times or more depending on the desired pixel size and number of signal averages to produce enough signals for an image. The most commonly used pulse sequence is the SE sequence, which includes a 90° pulse that rotates the magnetization vector 90° from the magnetic field and a 180° pulse that restores coherence. The time between the 90° pulses is called the repetition time (TR); the time between the 90° pulse and the restoration of coherence is called the echo time (TE). Another pulse sequence used in MR imaging is inversion recovery (IR). In this sequence a 180° pulse followed by a 90° pulse is applied at regular intervals (TR). IR was not used in this study. 21 2.1.3. Signal intensity The signal intensity in MRI is dependent upon at least four parameters intrinsic to tissue: T, relaxation time, proton density, T2 relaxation time, and flow. To acquire Tj-weighted images a SE pulse sequence with a short TR (200-1000 msec) and a short TE (20-25 msec) is used. The longer the TR, the greater the percentage of the magnetization vector that returns to the longitudinal plane before the next pulse and therefore the more signal that is generated. Tissues with a short Tj have a greater signal intensity than tissues with a long Ti at a given TR value. As TR decreases from 1,000 to 200 msec, the signal from cerebral tissues diminishes and contrast due to differences in Tj decreases. To obtain an image in which the proton density is the major determinant of intensity, the TR in a pulse sequence must exceed four times the T, of the tissue being studied to allow nearly complete remagnetization and the TE must be as short as possible to minimize the loss of signal intensity due to T 2 relaxation. A proton density image is typically acquired in a SE pulse sequence with a long TR (2000-2500 msec) and a short TE (20 - 25 msec), allowing near-complete T, relaxation. In a proton density image the signal intensity of fat is greater than gray matter, which in turn is greater than white matter. The signal intensity of CSF is less than the other tissues because its T, relaxation time is longer, allowing for a smaller percentage of protons to become magnetized prior to execution of the next pulse sequence. T2-weighted images may be acquired at the same time as proton density image by using a SE pulse sequence with a long TR combined with a long TE. To understand the differences between proton density and T2-weighted images, it is necessary to know the relationship of signal intensity to TE. As the TE lengthens, the signal intensity decreases exponentially because of loss of coherence in the transverse plane. The long T 2 of CSF 22 causes minimal signal decay, such that with long T E values C S F has a relatively greater signal intensity than cerebral tissue. 2.1.4. Gadolinium (Gd) enhancement in MRI Paramagnetic compounds have at least one unpaired electron. That electron has a magnetic moment approximately 1000 times stronger than the magnetic moment of a proton. As a result of motion (diffusion, rotation, etc), the paramagnetic compounds act at the atomic level to produce a rapid fluctuation in the local magnetic field. This process facilitates energy transfer among the excited protons and also from the protons to their environment, and the magnetic moments of the hydrogen nuclei deflected by the R F pulse return more rapidly to their initial state. Paramagnetic compounds can only shorten relaxation times; they cannot prolong them. Moreover, they always affect both T, and T 2 relaxation times, usually to an equal degree. In some circumstances T 2 may be shortened to a much greater degree than T,; the reverse is never true. A s a paramagnetic agent G d has become available for clinical use. G d is a rare-earth element, the ion of which has 7 unpaired electrons. Because it is poorly tolerated in its pure form, G d has been detoxified by complexation into stable chelates. One of the G d chelates, D T P A (diethylenetriaminepentaaceticacid dimeglumine), is • widely used. Although G d - D T P A enhances both T, and T 2 relaxation in vivo, with the clinical dose used in vivo the T, effect is very marked and the T 2 effect virtually absent [Weinman et al., 1990]. G d - D T P A strongly enhances relaxation in vivo at a dose of 0.1 - 0.2 mmol/kg, while the lethal doses in 50% of rats (LD50) is 10 mmol/kg [Weinmann, et al., 1984]. There is no evidence of dissociation of the G d - D T P A complex in vivo, and more than 90% of the G d - D T P A is excreted through the kidneys within 24 hours [Weinmann, et al., 1984]. G d -D T P A dimeglumine is well tolerated in vivo, and adverse drug reactions occur in 1.46% 23 of the patients [Niendorf et al., 1991], and in the majority of cases consist of minor side effects, such as nausea, local warmth/pain, headache, paraesthesia, etc. Severe side effects are very rare. 2.2. MRI of multiple sclerosis Traditionally, the role of radiographic imaging studies in the diagnosis of M S has been predominantly indirect (to rule out a space-occupying lesion). Although computered tomography (CT) has demonstrated abnormal findings in patients with M S , the incidence of positive C T findings in patients with M S varied between 9% and 80%. The C T abnormalities were often nonspecific areas of low density, atrophy, and/or contrast enhancement. Thus C T was used primarily to eliminate the possibility of other incranial lesions, such as a mass or arteriovenous malformation. Similarly, myelography was used to evaluated patients with spinal cord symptoms to exclude an intrinsic spinal cord mass or spinal cord compression. Before M R I , patients with clinical symptoms of M S had to undergo C T and/or myelography, a battery of paraclinical studies (including visual and auditory evoked responses and urodynamic studies), and C S F studies (CSF electrophoresis for detection of oligoclonal bands) to establish a diagnosis. Now in many cases, M R I and an appropriate clinical history allow the neurologist to make the diagnosis of M S based on clear cut dissemination in space with a high degree of confidence, thereby obliterating the need for other ancillary studies. Since the initial report by Young et al [1981]. describing the increased sensitivity of M R I compared with C T in the diagnosis of M S , many reports, including some from our research group, have documented the usefulness of M R I [Li et al, 1984; Mi l l e r et al., 1988; Grossman et al., 1986; Barnes et al., 1988; Poser et al., 1987]. M R I is easier to perform and less invasive than previous techniques to diagnose M S . Furthermore, unlike 24 paraclinical tests or C S F tests, M R I is site-specific in defining demyelinating plaques. This information potentially enables correlation with clinical symptoms, has prognostic significance, and permits M R I to be used for follow-up to determine regression or progression of lesions. This last advantage has proven important in evaluating new experimental methods of therapy [Paty, 1987; Paty et al., 1993b]. Although the M R I appearance may vary slightly from patient to patient or even among lesions in the same patient, certain characteristic findings are common to most M S lesions. M S lesions are generally detected as areas of increased T, and T 2 relaxation times relative to white matter of the brain. Since M S lesions have a prolonged T, relaxation time, they are well defined on IR scans. Indeed, in many of the initial M R reports, M S lesions were demonstrated predominantly with this technique. Using this technique, M S lesions are visualized as areas of diminished signal intensity within the white matter of the brain. A major disadvantage of IR scanning is its relative insensitivity for detecting lesions that lie immediately adjacent to the ventricles or subarachnoid spaces, which are very common in the M S brain, because of partial volume averaging with the low intensity of the adjacent C S F . Some lesions may also be missed if they are contiguous with the gray matter of the brain, which also has a lower intensity than white matter on IR scans. The spin echo (SE) technique is currently the best method for screening patients with suspected M S . In contrast to IR scans, long T R S E sequences demonstrate M S lesions as areas of high signal intensity. S E scanning sequences with a long T R of 2500 to 3000 msec, and dual echos using short and long TEs of 30 and 80 msec, are preferred. A T R of at least 3000 msec, is preferred when scanning at high field strengths of 1.5 T. The pathologic lesions demonstrate moderately increased signal intensity compared with brain or C S F on spin-density or mildly T 2-weighted images (TE of 30 msec.) and increase further in signal intensity on heavily T 2-weighted images (TE of 80 msec). Periventricular lesions can also be obscured by partial volume effects on heavily T 2 -weighted spin-echo images, on which both C S F and M S lesions are displayed as very 25 high signal intensity. However, these periventricular M S lesions can easily be distinguished from adjacent C S F spaces by comparison with proton density images on which the lesions wi l l have a greater signal intensity then does the adjacent C S F . Generally, lesions in the periventricular regions are best seen on 30 msec. T E images, whereas posterior fossa and deep white matter lesions are often best displayed on 80 msec T E images. Dual-echo sequences are particularly useful whenever there are subtle lesions that may be equivocal on one echo, since they can be confirmed as a real finding when also seen on the other echo. Since the T 2 relaxation time of demyelinating plaques is less than that of C S F , the lesions may decline in relative signal intensity compared with the ventricles or subarachnoid space when T E times are extended beyond 80 msec. Thus SE sequences with T E times that are greater than 80 to 100 msec may obscure some lesions. Min imum possible slice thickness should be used to decrease partial volume averaging from the surrounding low-intensity white matter. A 5 mm slice with a 0.5 - 1 mm gap is typically used, although the gap is primarily determined by cross-slice contamination or "cross talk." When routine imaging fails to demonstrate disease, some investigators include T 2-weighted 2 mm slices with increased lesion detection [Schima, et al. 1993] Although gradient-echo sequences have many applications in cranial and spinal imaging, the relatively decreased contrast resolution in comparison to spin-echo sequences, as well as the increased susceptibility artifacts of gradient-echo sequences, are significant disadvantages in imaging M S . Rarely i f ever wi l l gradient-echo sequences provide additional information in cases of M S . Characteristic, findings of M S consist of multiple, usually small, lesions within the white matter of the brain that have long T, and long T 2 relaxation times, as well as increased proton density, in comparison with normal white matter. The cause for the increased T, and T 2 relaxation times seen with M S lesions relates in part to the gliosis that occurs in chronic plaque formation. In acute lesions, edema resulting from the B B B 26 breakdown results in prolonged relaxation times and may also produce an area of abnormality considerably larger than the actual demyelinated area. Demyelination itself, with the loss of fatty myelin around the axonal sheaths within the M S lesion, does not contribute significantly to prolonged T 2 changes, since the amount of l ipid lost is not great enough to cause the magnitude of change demonstrated on M R I and because the l ipid protons of myelin have an extremely short T 2 relaxation time and are effectively invisible on M R I . The loss of myelin l ipid does, however, provide a more hydrophilic environment. The subsequent increase in water content leads to the observed increases in proton density, T, , and T 2 times [Ormerod, 1989]. The individual lesion of M S are usually less than 10 mm in size, most often between 1 and 5 mm. Confluent plaque formation from merging of multiple small individual demyelinating plaques that are contiguous with each other or secondary to an acute, active demyelinating plaque can occur. In either case, these types of lesions can sometimes become quite large simulating tumors. The most common location of lesions seen by M R I is in the periventricular region adjacent to the superolateral angles of the lateral ventricles. This distribution corresponds with pathologic descriptions [Steward, et al.1986]. Lesions in the corona radiata often appear oval or elongated in configuration, with the long axis of the demyelinating lesion oriented along the subependymal veins, perpendicular to the walls of the ventricles. This orientation corresponds to the pathologically described periventricular, perivenular location of M S lesions. The ovoid shape corresponds to the perivenular inflammation seen pathologically. This appearance is highly characteristic appearance for M S lesions and helps to distinguish M S from other white matter diseases, most notably deep white matter gliosis and infarction in older patients. Lesions are also often seen within the body of the corpus callosum, a site rarely affected by microinfarcts, further allowing these two entities to be distinguished. In addition, M S lesions tend to be more focal than deep white matter ischemic changes. 27 Additional .common sites of involvement include the walls of ventricles adjacent to the atrial trigones and occipital horns, the white matter of the centrum semiovale, the forceps major and minor, and the temporal lobes. Demyelinating plaques are also often demonstrated in the brainstem and cerebellum on M R I . There, they often abut the C S F spaces. Brainin et al. [1987] reported that among clinically definite cases of M S , pontine lesions were identified in 71%, Medullary lesions were identified in 50%, and midbrain lesions were identified in 25%. Lesions are also frequently seen within the cerebellum, although less commonly than in the brainstem. The middle cerebellar peduncles and the white matter of the corpus medullaris are other preferred locations. The acute lesions of M S have a characteristic temporal course. Acute lesions reach a maximum size in approximately 4 weeks and subsequently regress and decrease in size, leaving a smaller residual gliotic lesion demonstrating a prolonged relaxation time [Koopmans et al, 1988, Willoughby et al., 1989]. Acute M S plaques can demonstrate enhancement on M R I after intravenous injection of the M R contrast agent, G d - D T P A . Since this pattern of enhancement is consistent and can be seen even with clinically silent acute demyelinating lesions, M R I has proven much more sensitive in detecting new disease activity than is the clinical examination. Because of this great sensitivity, M R I promises to be useful in monitoring the results of therapy, and potentially distinguishing between benign and chronic progressive forms of M S [Paty, 1987]. Although M S is generally considered a white matter disease, it must be realized that approximately 5% to 10 % of lesions occur in gray matter of the brain, such as the cerebral cortex or within basal ganglia nuclei. 28 2.3. The BBB and MRI Using C T scan after the intravenous injection of a contrast medium, regions of abnormal enhancement, reflecting extravasation of iodine through a damaged B B B , have frequently been reported in M S patients [Hreshey, 1979]. Compared to C T , M R I is more sensitive in detecting the M S plaques in the C N S . To provide contrast enhancement in M R I , G d - D T P A is used as a marker for the B B B breakdown [Runge, 1985]. Using mannitol to reversably open the B B B in dogs, Runge and colleagues [1985] showed that gadolinium may significantly detect an alteration of the B B B . Gadolinium enhanced M R I (G-MRI) appears more sensitive in the detection of active lesions than clinical examination alone. Some studies [Grossman, 1988; Mi l le r , 1988; Grossman, 1986] showed that more patients had enhanced lesions than were considered to have clinically active lesions. The B B B impairment is a consistent finding in new lesions detected with M R I , but, this impairment can also develop in older previously nonenhancing plaques without evidence of an increase in size. The enhancing regions were often less extensive than that on the corresponding high signal on T 2-weighted images. Maximum intensity of enhancement occurrs from 4 to 120 minutes after G d -D T P A injection. In the great majority of lesions peaking around 29 minutes [Kermode et al., 1990]. The enhancement can persist for 3 to 5 weeks. Our study [Zhao et al., 1993] found that G - M R I is sensitive for detecting lesion activity. Some lesions can be detected on G - M R I which may not be seen on standard M R I on the same scan. Those lesions can be either new or enlarging lesions that appear stable on follow up standard M R I . The B B B disturbance may precede other M R I signs of M S lesions (figure 2). 29 Figure 2a. Proton density imaging. TR 28447TE 20. Arrow shows one of MS lesions. 30 Figure 2b. Tl-weighted image pre-Gd injection. T R 720/TE 15. Arrow shows M S lesion with hypointensity. 31 Figure 2c. TI-weighted image, TR 720ATE 15. Arrow 2 shows an enhancing lesion which was not detected in proton density image. 32 2.4. Serial MRI as a tool in monitoring therapeutic trials in multiple sclerosis In early experience with M R I , intermittent scans showed that chronic lesions could be seen to increase in size and asymptomatic lesions could be seen to come and go [Li et al., 1984; Johnson et al., 1984]. Using serial M R I combined with frequent neurological examinations, it quickly became apparent that disease activity as measured by M R I could be quite dramatic and often subclinical. Several serial studies were carried out in our center. Seven relapsing-remitting patients [Isaac et al., 1988] entered into the first study. The patients were examined by monthly M R I scans, physical examination, and immunological testing over six months. Neurological histories were taken at each visit. Five clinical relapses occurred in three patients. There were 17 new or enlarging M R I lesions seen during the study. Seventeen out of the thirty-six follow-up M R I examinations (48%) showed evidence of new and/or increasing disease activity. The mean clinical relapse rate was 1.4 relapses per patient per year. The rate of the appearance of new M R I lesions was 8.0 activity events per patient per year. The second study included 9 patients with minimally disabling but active relapsing disease [Willoughby et al., 1989]. Each patient had a careful interim history along with neurological and M R I examinations done once every two weeks for an average of five months. Clinically detected activity was minimal with only three findings. One patient had two minor spinal cord sensory relapses. M R I examinations showed that there were ten instances in which new M R I lesions appeared, and two instances in which there was enlargement of pre-existing lesions. A l l of the M R I activity was asymptomatic. The clinical relapse rate was 0.4 relapses per patient per year. The frequency of M R I activity 33 was 2.6 positive M R I examinations per patient per year. There were eighty-three follow-up M R I examinations, and ten of those examinations (12%) showed evidence of increasing disease activity. This study, along with the first study, showed that the degree of disease activity in relapsing patients as detected by M R I could be as high as five times the clinical relapse rate. Eight severely disabled patients in the progressive phase of M S were included in the third study [Koopmans et al., 1989]. Seven of the eight patients had begun with relapsing disease and could be considered R P or secondarily progressive. The other patient had chronic progressive (CP) or primarily progressive disease from outset. A s in the second study, all patients had histories, physical examinations and M R I examinations, in addition to immunological tests, once every two weeks over a period of six months. In this study there were ninety-eight follow-up M R I examinations, during which time no clinical relapses were seen. However, twenty-five new M R I lesions were seen. There were also sixty-one instances in which previously seen stable M R I lesions increased in size. There were thus eighty-six M R I activity events during the study. Forty-seven of the ninety-eight follow-up scans (48%) showed evidence of increasing disease activity. These studies provide the evidence that serial M R I is more sensitive in detecting disease activity than clinical exacerbation. Using serial M R I to monitor disease activity, three therapeutic trials in the past 5 years, all in our center, have been done. In the first study, M R I was used to evaluate the efficacy of systemic lymphoblastoid interferon therapy in chronic progressive M S [Koopmans et al., 1993]. Thirty-six patients with chronic progressive M S were treated with lymphoblastoid interferon daily for 6 months and 27 received placebo. Patients had M R I at the outset of the study and after 6 and 24 months. Lesion activity was evaluated by visual analysis. A semiautomated computer-assisted quantification of overall lesion load was used to determine the change of overall lesion burden. We found that both the interferon- and placebo-treated groups developed more active lesions as the study 34 progressed. There was no difference in lesion activity between the two groups. Comparison of lesion load, however, showed a trend toward improvement after 6 months for the interferon-treated group. This difference between the two groups had disappeared by the end of the study. However, the average overall increase in lesion load (burden of disease) was 10% per year. The conclusion in this study was that lymphoblastoid interferon was not effective in decreasing active MRTdetected lesions or in decreasing M R I lesion load in patients with chronic progressive M S , but that a yearly predictable increase in burden of disease occurred overtime. One hundred fifty seven C P M S patients from 6 centers in U S A and Canada, randomized for a 2 year cyclosporine therapeutic trial, were included in our second study [Koopmans et al., 1992]. Seventy-seven patients received cyclosporine and 80 placebo. Entry and exit M R I scans were obtained for each patient. Pooled M R I data from all centers again showed an increase in lesion load overtime. However, there was no significant difference between placebo and cyclosporine groups. Our third study was an interferon beta-lb (IFNB) therapeutic trial in 372 ambulatory patients with R R M S , [Paty et al., 1993; The I F N B Multiple Sclerosis Study Group, 1993; Arnason, 1993]. In the multicenter, randomized, double-blind, placebo-controlled trial, one-third of the patients received placebo, one-third 1.6 mill ion international unit (MIU) of I F N B , and one-third 8 M I U of I F N B . Yearly M R I was used to monitor the disease activity. In the serial MRIs , M S activity was significantly less in the high-dose I F N B group. Clinical and M R I results support that I F N B has made a significant impact on the natural history of M S in these patients. Other M R I serial studies had been reported, some using gadolinium [Miller, 1989, Kermode A G 1990, Bastiianello S 1990, Wiebe 1992, Harris 1991, Thompson A J 1990, Smith 1993]. U p to 90 percent of new M R I lesions enhanced with gadolinium, and occasionally an enhancing area was seen before the standard M R I lesion was seen [Kermode 1990]. Spinal cord M R I added about 20 percent to the activity seen on the 35 standard head scans [Wiebe 1992]. In general, two-thirds of gadolinium enhancing lesions showed enhancement on only one scan [Miller , 1989]. Mi l l e r and associates found that twenty minutes after G d - D T P A injection was the optimum time to see enhancement. Repeat G d - D T P A injections were well tolerated by their patients. There are some profound differences among several clinical categories of patients [Thompson 1990]. Primary progressive patients had the lowest rate of development of new lesions at 3.3 new lesions per patient per year. The next highest rate was for benign patients who had 8.8 new lesions per patient per year. Typical R R and R P patients had 17.2 and 18.2 new lesions per patient per year respectively. The data from serial studies have shown that the rate of lesion activity varies widely among individual patients. Unfortunately, the rate of development of new and/or otherwise active lesions also varies considerably over time in the same patient. Some studies showed [Harris et al, 1991, Smith, 1993] that there were "burst" activity periods in some patients. Some patients can be active over a three month period of time and then be totally inactive over the next two to three months. However, "burst" periods of activity can be seen in our yearly M R I study [Zhao et al, 1995 (in press)]. The yearly M R I study showed that the more active lesions detected in the first year in an individual patient, the more likely one sees higher activity rates in follow up scans. Our other studies [Paty et al, 1993; Zhao et al, 1991] showed that the rate of lesion activity varied widely among patients. Some patients did not have any M R I lesion activity over 2 or more years, but some patients had more than dozens of new or enlarging lesions. This variability implies that a "run i n " period of scanning prior to a clinical therapeutic study may predict the subsequent M R I activity in groups of M S patients making it helpful for selecting patients in designing new therapeutic trials. Nevertheless, the "run in" period of time could be a key in determining M R I activity in individual patients which could let researchers know how long and how many patients required before a new study starts. 36 Mil le r and his colleagues [1988] have seen high rates of asymptomatic changes in their serial G d enhanced M R I studies in relapsing progressive patients. New lesions reached a maximum size in two to four weeks and then faded over the subsequent six to eight weeks. In contrast to the pattern seen in R P M S by our group and by Queen Square group, Thompson et al. [1992] found primary C P M S to have a different pattern. Their primary C P patients had a low relative burden of disease and had a low tendency to develop new M R I lesions. Additionally, only one of the twenty small new M R I lesions that they saw enhanced. The optimum frequency of scanning, considering the information available, has been once every two weeks. Assuming no lesion lasts less than two weeks, 100 percent of new and active lesions are detectable at two weeks. Based on our experience only 67% percent of that lesion activity is seen with scans performed every four weeks and 40 percent of the lesion activity is seen with scans every six weeks. Conversely, 36 percent of active lesions had a duration of activity of less than four weeks, another 28 percent between four and six weeks, and another 7 percent between six and eight weeks. However, scanning at yearly intervals allows most of the new lesions to be seen. This data supports serial M R I ' s usefulness as a tool in monitoring therapeutic trials in M S . On the use of M R I to monitor clinical trials, some suggestions had been made [Miller, 1991]. (1) M R I monitoring of treatment trials is appropriate. (2) Such monitoring studies can be done in periods as short as six months. (3) G d - D T P A enhancement adds 10 percent or more to the activity determined by enhanced scanning and should be done routinely. (4) Groups of patients should be stratified into a), early relapsing-remitting M S ; b). benign M S ; c). Secondarily progressive M S ; d). Primary chronic progressive M S . 37 Chapter 3 MRI Active Lesions 3.1. Definition of active lesions "Aactive" lesions were identified as follows: 1. New lesion: A lesion that had not previously been seen (figure 3). 2. Enlarging lesion: Enlargement of a previously seen stable lesion (figure 4). 3. Recurrent Lesion: A lesion that develops at the same site where a previous lesion had been seen (figure 5). 4. Enhancing lesion: A lesion that enhances in the Tj-weighted image post G d - D T P A injection (figure 6). A continuously enlarging lesion over several examinations was counted as active only once. Each significant change indicating increased activity that was not a continuation of a previous change was noted as an " M R I event". There could be one or several M R I events in each active scan. One lesion could also have several M R I events during a study. In patients with R R M S there are many more preexisting lesions seen and the pattern of lesion change is great. 38 Figure 3a. Proton density image, T R 2133/TE 60. 39 Figure 3 b . Proton density image, T R 2133/TE 60. Arrows show new lesions 40 Figure 4a. Proton density image, T R 2133/TE 60. Arrow shows one of many M S lesions 41 Figure 4b. Proton density image, TR 2133/TE 60, 6 weeks later. Arrow shows the lesion enlargement. The lesion also merges with a neighbor lesion. 42 Figure 5a. Proton density image, T R 2133/TE 60. Arrow shows one of several M S lesions 43 Figure 5b. Proton density image, T R 2133/TE 60, 6 weeks later. The lesion marked by the arrow in the previous scan has disappeared. 44 Figure 5c. Proton density image, T R 2133/TE 60. 12 weeks later. The lesion that disappeared in the previous scan hasreappeared (arrow). Careful examination of the rest of the image shows that the changes in this lesion were not artefactual. 45 46 Figure 6b. T,-weighted image, TR 683/TE 26, on the same day as fig. 6a Pre - Gadolinium injection. No lesion was detected. 47 Figure 6c. T rweighted image, TR 683/TE 26, post gadolinium injection on the same day. Two enhancing lesions were found (arrows). These lesions correspond to two of the stable looking lesions marked by arrows on figure 6a. 48 3.2. Materials and methods in this study 3.2.1. Patient population We studied 50 patients with clinically definite relapsing remitting M S who had been entered into a therapeutic trial using beta-interferon. There were 12 men and 38 women whose ages ranged between 22 and 53 years (mean, 38.6 years). Mean duration of disease was 8.2 years (range, 1.1 to 20.0 years; S D = 5.0 years). Disability of patients was between 0 and 3.5 (mean, 2.0, S D = 1.0) on the Kurtzke Expanded Disability Status Scale (EDSS). A l l patients had an abnormal cerebral M R I characteristic of M S , with scattered areas of increased intensity on the spin-echo (SE) sequence. Patients were randomized into three treatment arms: Placebo (N=17), 1.6 m l U (N=17), 8 m l U (N=16) I F N B self-administered subcutaneously every other day. Each patient had a cerebral M R scan every 6 weeks for 2 years. Clinical examination was also done on patients every 6 weeks. 3.2.2. MR scan technique M R scans were performed using a Picker International Cryogenic M R 2000 scanner with a superconducting magnet operating at a static magnetic field strength of 0.15 Tesla and a 30 cm diameter receiver coil in the Vancouver Hospital and Health Science Center, University of British Columbia ( U B C ) Site. Twelve contiguous 10 mm thick axial slices were obtained through the brain from the upper cerebral hemisphere to the medulla. The inplane resolution was 1 mm. A dual echo S E sequence was used with echo delay times (TE) of 60 and 120 msec and a repetition time (TR) of 2133 msec. M R I serial studies require very careful repositioning. Careful attention is the key to the repositioning process. The repositioning error was minimized through alignment of 49 external and internal landmarks. Initial repositioning was based on two angles from external landmarks (the canthomeatal line and the nasion-tragal line). A midline sagittal slice (pilot) was then obtained and the head position verified using an internal angle (angle between top of the cerebellum and the anterior sphenoid sinus). If this angle differed by more than 2 degrees from the baseline study, the patient was repositioned and the pilot scan repeated. Slices for the actual scans were then programmed so that the middle slice of a simultaneous 12-slice series was centered on the top of the cerebellum. Halfway through an individual scanning session, the pilot scan was repeated to ensure that positioning was still adequate. If movement occurred, the patient was repositioned and the appropriate sequence repeated. 3.2.3. Quantitative analysis of M R I A n interactive computer program was designed to display M R images and permit manual tracing of lesions [Koopmans, et al. 1993]. The lesion borders were outlined on a computer monitor. The area of the individual lesions could then be determined and analyzed. 3.2.4. Statistics The differences in M R I data were evaluated by means of the chi-square test for categorical variable and by Student's t test or the A N O V A for quantitative variables. 50 3.3. Assessment of the individual lesions in RR MS on MRI 3.3.1. Introduction Activity of M S lesions as detected on M R I can be determined by performing serial examinations over time. Active lesions include new, enlarging, and recurrent lesions (see sect. 3.1). Some new lesions had totally resolved in 4 - 8 weeks. On pathology, chronic active plaques have a different pattern from that of chronic inactive ones [Adams, 1983] . A chronic active M S plaque has a broad zone of perivascular and parenchymal inflammation along the edges together with diffuse demyelination. However, the pathogenesis of the developing new lesion is still not known. The rate of M R I active lesions using this method of analysis was 2.4 per patient per year in one study [Willoughby et al., 1989] while the clinical relapse frequency was 0.4 per patient per year. Many active lesions are asymptomatic causing a higher frequency of M R I activity. Understanding the dynamics of these active lesions may in part explain the pathogenesis of M S . However, the identification of lesion activity using both standard and enhanced M R I involves extra time, cost, and invasiveness. One objective of this study was to observe any characteristics that could allow active lesions to be distinguished from stable ones on systematic standard unenhanced scans, thus avoiding the extra time, cost and invasiveness of G d imaging. 3.3.2 Materials and methods Lesion Analysis of Magnetic Resonance Scans Researchers , experienced in M R I and masked to the clinical course of the patients, identified lesions for activity. Changes in lesion size and number between scans were determined by comparative examination of previous scans. Only changes in lesions that 51 were independently agreed to by all observers were recorded. A lesion enlarging continuously over several examinations was counted as being active only once. A lesion seen in adjacent slices was also counted only as one active lesion. Lesions that did not change over at least 9 months were recorded as stable. A total of 150. active lesions (50 new, 50 recurrent, and 50 enlarging) and 50 stable lesions were analyzed. The first 50 of each category of lesions were identified and analyzed as to their margin (well defined vs. i l l defined), shape (round/oval vs. irregular), and location (periventricular vs. nonperiventricular; cerebral, cerebellum, and brainstem; and white matter vs. gray-white matter junction). Periventricular lesion was defined as any part of lesion which contacted with ventricle and non-periventricular lesion was defined as the lesion was away from ventricle. Quantitative Analysis of Magnetic Resonance Images The lesion borders were outlined on a computer monitor by a technician. The area of the individual lesions could then be determined and analyzed. Lesions were also grouped as small when the area was less than 30 mm 2 , medium when the area was between 30 and 100 mm 2 , and large when the area was greater than 100 mm 2 . 3.3.3 Results M R I morphological findings in the margin and shape of the lesions are summarized in table 2. The margins of most lesions were well defined. When they were i l l defined, they were more commonly new or recurrent lesions rather than stable or enlarging ones. New lesions were also more likely to be round and oval in shape. The location of lesions is summarized in Table 3. Enlarging and stable lesions were slightly more common in a periventricular rather than a non periventricular location. On the other hand, new and recurrent lesions were more commonly located away from the 52 Table 2: Margin and shape of lesions (n = 50 for each group) New Enlarging Recurrent Stable No. % No. % No. % No. % Margin well-defined 42 84 49 98 42 84 50 100 Margin ill-defined 8 16 1 2 8 16 0 0 Round and Oval 36 72 22 44 30 60 17 34 Irregular 14 28 28 56 20 40 33 66 Enl = enlarging lesions, Rec = recurrent lesions, Sta = stable lesions. 53 Table 3: Location of lesions (n = 50 for each group) New Enlarging Recurrent Stable Location No. % No. % No. % No. % periventricular 9 18 29 58 12 24 27 54 non-peri ventricul ar 41 82 21 42 38 76 23 46 cerebrum 41 82 48 96 42 84 48 96 cerebellum 3 6 0 0 4 8 0 0 brainstem 6 12 2 4 4 8 2 4 white matter 25 50 41 82 36 72 42 84 GW junction* 25 50 9 18 14 28 8 16 GW junction = junction of gray and white matter Table 4: Mean area of lesions (mm2, n = 50 for each group) New Enlarging Recurrent Stable No. % No. % No. % No. % Small 20 40 2 4 16 36 20 40 Medium 19 38 16 32 30 60 15 30 Large 11 22 32 64 4 8 15 30 Mean Area (mm2) 79 370 48 107 54 ventricles (for periventricular and non-periventricular location, new lesions vs. enlarging lesions, p < 0.005). While more stable, enlarging, and recurrent lesions were found in the white matter, new lesions were just as likely to be at the gray white matter junction as in the white matter. Table 4 summarizes the size and the mean maximum area of each type of active lesions. As a group, enlarging and stable lesions are larger than new and recurrent ones, most of the enlarging lesions were of the large category. Recurrent lesions tended to be of the medium or small category (mean area, new lesions vs. enlarging lesions, p < 0.0003). 3.3.4 Discussion Lesion activity as detected by serial M R I examinations is much more common than is the clinical relapse rate. Our previous studies showed 2.3 new lesions/patient/year in a relapsing remitting group of M S patients and 6.0 new lesions/patient/year in a secondary progressive group. There were 2.9 total active events/patient/year in the R R - M S group and 21.8 total active events/patient/year in the R P - M S group [Koopmans et al., 1989; Willoughby et al., 1989; Paty et al., 1988]. On the other hand, the clinical relapsing rate was only 0.3 to 0.4 attacks/patient/year [Adams, 1989]. The very high rate of activity as detected on M R I indicates that most of these active lesions are asymtomatic. Active lesions can be detected in both serial unenhanced M R I and gadolinium enhanced M R I [Koopmans et al., 1989b; Bastianello et al., 1990; Mi l l e r et al., 1988]. Gadolinium enhancement indicates blood-brain barrier breakdown in the sites of inflammatory demyelination while findings in serial M R I showed morphologic change in plaques in which inflammation may be an important factor [Raine et al., 1991; Hawkins et al., 1990; Kermode et al., 1990]. In this study, new and recurrent lesions tended to have well-defined margins and were round and oval in shape while enlarging and stable lesions 5 5 tend to have ill-defined margins and were irregular. New lesions probably represent acute inflammation where demyelination may or may not be seen. Nevertheless, Raine suggested that these new lesions represent not acute lesions but merely growing fingers from active chronic plaques [Raine et al., 1991]. Our findings show that the majority of new lesions were non-periventricular (new vs. enlarging lesions, p < 0.005). These data suggest that new and recurrent lesions are rarely fingers of active chronic plaques. However, detecting a new lesion on a conventional M R I scan can be difficult. Chronic, active plaques constitute a less demarcated margin and are centered on small veins within the white matter proper on pathology [Raine et al., 1991]. Foamy macrophages can be invariably found throughout the normal white matter far away from lesions. Most of enlarging lesions we saw (82%) were originally from stable lesions. Compared to other kinds of lesions, enlarging lesions were larger in size and were irregular in shape, similar to their original stable lesions in size and shape. The location of active lesions showed new and recurrent lesions were non-periventricular (82% and 76%) contrasting to enlarging and stable lesions that were periventricular (42% + 46%) in location. The results were similar to findings by Koopmans and colleagues [1989a]. To date, it is not clear whether new lesions can be distinguished from stable ones in location only or also in their pathogenesis. In contrast to the other groups, new lesion also tended to be detected at the junction of gray and white matter (50%). Gray matter has a higher blood supply than white matter. The junction of gray and white matter which is more vascular may influence, the greater chance of new lesion formation. This study provides evidence that there are some differences that distinguish new and recurrent lesions from enlarging and stable ones. New and recurrent lesions were smaller, more likely to be round or oval, often distant from ventricles, with a more ill-defined margin when compared with stable ones. Enlarging lesions were much like stable lesions 56 in shape, size, location, and area. However, it may be difficult to separate active from stable lesions without serial M R I and gadolinium enhancement. 57 3.4. Features of enhancing and non-enhancing lesions of MS on MRI 3.4.1. Introduction Conventional standard serial T 2 and P D M R I and gadolinium-enhanced M R I are now established as the most useful markers of disease activity in M S . Their role in monitoring therapeutic trials is currently being clarified. One of the many factors requiring consideration in establishing a research protocol is the clinical pattern of disease activity. G d - D T P A enhanced M R I can detect B B B breakdown. What this breach in the normal B B B to G d - D T P A means in histological terms is of considerable interest because of its implications for understanding the development and evolution of the plaque as well as its potential use in monitoring treatment. A s expected, lesion enhancement is not always seen in morphologically active lesions as seen on M R I , conversly a stable lesion on serial standard M R I can be enhanced by G d - D T P A as its only indication of activity. In this study we examined the morphological features of enhancing and nonenhancing active lesions. 3.4.2. Materials and methods Enhanced scans were performed on entry into the trial and whenever morphologically active lesions were identified on serial standard M R I . Pre and post contrast scans were obtained using a T R of 683 msec and T E of 26 msec. Enhanced scans were performed 3 and 15 minutes after G d - D T P A administration (Magnevist, 0.1 mmol/kg). The repositioning error on follow-up scans and gadolinium enhancement scans was minimized by use of our previously described procedure (see section 3.2.2). 58 Analysis of Magnetic Resonance Scans Three of us (GJ Zhao., D K B L i . , B L Tanton.), experienced in M R I and masked to the clinical course of patients, identified lesions for activity and enhancement. Changes in lesion size and number between scans revealing morphological activity were determined by comparative examination of previous scans. We only recorded changes in lesions that were independently agreed on by all observers. A lesion enlarging continuously over several examinations was counted as active only once. A lesion seen in adjacent slices was also counted as one active lesion only. Active lesions were defined as (1) new: (2) enlarging; (3) recurrent; and (4) stable (see section 3.1). A n enhancing lesion seen on study entry was defined as active i f morphological change was seen in follow-up. The degree of enhancement after G d - D T P A was graded as follows: 0. no enhancement, 1. questionable enhancement, 2. faint but definite enhancement, 3. moderate enhancement, and 4. intense enhancement. A l l lesions were examined as to their margin (well defined vs. i l l defined), shape (round/oval vs. irregular), and location (periventricular vs. non-periventricular). 3.4.3. Results Of the 308 morphologically active lesions that were identified on serial M R I , 33 had a concurrent gadolinium enhanced scan of which 14 showed enhancement and 19 did not. Twenty nine (29) other enhancing lesions were identified at entry for a total of 43 enhancing lesions. Fifty (50) morphologically stable lesions identified consecutively from the first 7 patients were studied as a control group. The location, margin, and shape of lesions were summarized (table 5). In enhancing lesions, 29 of 43 were both enhancing and morphologically active in which 23 (79%) of 29 lesions had a clear margin. Seventeen [17/29 (59%)] lesions were round or oval in shape, and 6 of 29 (21%) were periventricular in location. In contrast, 14 of 43 (33%) 59 were enhancing but morphologically stable in which 11 of 14 (79%) were clear in margin, 6 of 14 (43%) were round or oval in shape, and 9 of 14 (64%) were periventricular in location. 54% of the stable lesion were periventricular in location. Lesion location was interesting in that morphologically active lesions, whether or not they were non-enhancing or enhancing, presented a similar proportion to be periventricular. Morphologically stable lesions, whether enhancing or non-enhancing, were also seen in almost the same periventricular proportion. There was no significant difference in periventricular versus non-periventricular location of enhancing lesions versus non-enhancing lesions (p > 0.2). Of 43 enhancing lesions, 29 were identified at entry scan. Twelve (12) of 29 (41%) were periventricular and 17 of 29 (59%) were non-periventricular. In 14 enhancing but morphologically stable lesions, 10 (72%) were identified on the entry scan. In other words, of 29 enhancing lesion detected on the entry scan, 14 were morphologically stable in follow up. Signal intensity of enhancing lesions was studied 3 minutes and 15 minutes post-injection (table 6). Twelve (12) of 43 (28%) lesions had not been detected after 3 minutes but were seen in 15 minutes. The total number of lesions that were in grade 3 and 4 was obviously higher in 15 minutes (20 lesions) than in 3 minutes (10 lesions). Fourteen (14) of the 43 enhancing lesions (33%) remained stable on follow up by serial M R I . Eight (8) of 19 (42%) non-enhancing lesions disappeared at the 3 month follow up compared to only 7 of 43 (16%) enhancing lesions. The area of lesions was summarized in Table 7. Most non-enhancing lesions were less than 30 m m 2 in size. Twenty three (23) of 29 (79%) enhancing and morphological active lesions and 13 of 14 (93%) of the enhancing but morphologically stable lesions were greater than 30 m m 2 in size. In contrast, only 9 of 19 (47%) non-enhancing morphologically active lesions and 30 of 50 (60%) of the non-enhancing morphologically stable lesions were greater than 30 60 o m 00 ON T3 <U 0 c ea s: e <D 1 C o cn o c ca c II _ > '•4—» 0 1 c o CN _ > 0 1 c o ON > o o o o z o 2 o ca ~3 o c 1) _> u OH O VD r- in vo cn ON *T) <-< m CN in Q 00 ON CO 00 CN <N CN cn CN r- m oo — CN m CN cn r~-CN •4—' -4-* C3 £ c (D O o c 3 o o o m ON r-cn CN ca € c 'bb 1— ca ~ ON r-- CN >n CN r- cn vo cn 00 T j -ON —' l> CN - H cn CN VO ca TS c 3 C 'Kb >-ca cn cn -vi-vo ON m r-ca > o T3 C ca c 3 o •06 VO VO o cn —i cn — ON r-~ CN m CN CN oo ON 61 Table 6: Signal intensity of enhancing lesions (n = 43) Time Post-injection Grade 3 min (%) 15 min (%) 0 12 27.9 0 0.0 1 8 18.6 11 25.6 2 13 30.2 12 27.9 3 7 16.3 13 30.2 4 3 7.0 7 16.3 Table 7: Lesion size in 3 groups of lesions <30(mm2) (%) 30-100(mm 2 ) (%) >100(mm2) (%) Non-enhanced (n=19) 10 52.6 4 21.1 5 26.3 Enhanced (n=43) 8 18.6 26 60.5 9 20.9 Stable (n=50) 20 40 15 30 15 30 62 m m 2 in size. Compared non-enhanced lesions with enhanced lesions, The difference was 2 2 stastically significant in size less than 30 mm versus greater than 30 mm (p < 0.01). 3.4.4. Discussion The findings of this study show that enhancing lesions can be larger although enhancing and nonenhancing lesions usually cannot be distinguished on morphological features alone. The majority of enhancing lesions, 34 of 43 (81%) were greater than 30 2 mm . Obviously, larger lesions can be easily detected especially i f the lesion was just in the enlargement phase when the M R was done. If the M R scan was performed in the phase before the lesion's peak, the lesion would then be enhanced. In other words, a morphologically active but nonenhancing lesion could have been scanned just in the shrinking phase. This point can be indirectly confirmed in this study in which the rate of lesion disappearance was higher (42%) in morphologically active but non-enhancing lesions than with morphologically active and enhancing lesions (16%) on scans followed up to three months. The frequency of clinically silent M S has been estimated at the 25% rate of those diagnosed clinically [Engell, 1989]. In the silent group, the M S plaques were located mainly in the periventricular areas. In a study by Brownell and Hughes [1962], 40% of M S plaques in the cerebrum were found in the periventricular location in pathology. In contrast, 80 percent of pathologically active lesions were in the deep cerebral white matter or at the gray-white matter junction. On M R I only 20% of the active lesions were periventricular [Koopmans et al., 1989a]. Koopmans and co-workers [1989b] found that 77% of enhancing C T lesions were in the deep white matter or at the gray-white matter junction and only 23% were periventricular. The findings were in accordance with those seen in this study in which only 21% of enhancing and morphological active lesions were periventricular compared with 64% of enhancing but morphological stable ones. On the 63 other hand, most of the stable and morphologically non-active but enhancing lesions tended to be located in the periventricular region. It was not surprising that 41% of 29 enhancing lesions that were identified on the entry scan were periventricular since the majority (72%) of enhancing but morphological stable lesions were seen on the entry scan. Therefore, enhanced lesions identified on the entry scan may be similar to stable lesions in location and in duration. It is suggested that there may be some differences in pathogenesis and a linking process between the two groups of morphological stable and morphologically active lesions. Gadolinium enhancement seen on M R I indicateded B B B breakdown at the site of inflammatory demyelination. Serial standard M R I showed morphological changes in plaques in which inflammation may play an important role [Hawkins et al., 1990; Kermode et al., 1990]. Some studies [Kato et al., 1989; Juhler et al., 1985; Lossinsky et al., 1989] showed the ultrastructure of the B B B in chronic relapsing E A E in the inactive stage with gliosis and perivascular fibrosis. The basement membrane of the perivascular processes of astrocytes in the demyelinating lesion sites was only partially formed, and neural parenchyma has not fully separated from the perivascular mesenchymal tissues by the E M of astrocytic processes. However, the most important finding was that vesicular transport is increased and mitochondrial content decreased in endothelial cells; tight junctions were opened between endothelial cells; and the interendothelial space was widened when the B B B was dysrupted [Claudio et al., 1990; Claudio et al., 1989]. The mechanism of B B B damage in E A E is similar to that in M S . Gay and Esir i [1991] showed a considerable leakage of large plasma proteins from the capillary beds and venocapillary junctions around all the acute plaques examined. Moreover, this process takes the form of an annulus around the plaque edge, a feature of enhancing lesions in which gadolinium leakage is maximal at the plaque periphery and reduced in the plaque center [Kermode et al., 1988]. The appearance of considerable protein leakage in normally myelinated areas around lesions led Kermode, et al to believe that the leakage 64 per se was not necessarily demyelinating. This finding would suggest that M R images may occur in the absence of any demyelinating lesion. In other words, some morphologically enhancing active lesions can be be purely inflamatory and disappear totally and not leaving any trace in M R imaging and be with not demyelination. The time course of intensity changes was noted in this study. More enhancing lesions were detected when imaged 15 minutes than 3 minutes post-gadolinium. Kermode, et al. [1990a] found that maximum intensity of enhancement was variable occurring from 4 to 120 minutes after contrast injection, the great majority of lesions reaching maximum around 29 minutes. Our finding provides further evidence that scan time post contrast injection not only affects the intensity of enhancement, but also affects number of enhancing lesions detected which is important in designing serial gadolinium studies. What lesion can be called an enhancing or morphologically active lesion? The enhancing lesion indicates the status of the B B B breakdown which allows cytokines and immune factors to enter and leave the brain freely. If a patient is currently suffering a dramatic change of pathology in a brain lesion, an enhancing lesion can be detected in T i -weighted imaging post G d - D T P A . On the other hand, morphologically active lesions, which may or may not be enhanced, suggests that breakdown and recovery of the B B B has occurred. The study of enhancing lesions is important for therapeutic trials because increasing the rate of B B B repair and therefore reducing the number of enhancing lesions may decrease the chance of further loss of neurological function in M S patients. In addition, morphologically stable unenhanced lesions can not be resolved as active. Morphologically active lesions seen on serial M R I can provide information on lesion activity even though the pathological changes are not known. This study provides evidence of the features of enhancing and non-enhancing lesions on M R I which include larger size lesions, such as more round and oval shape in non-enhancing than that of enhancing lesions. However, distinguishing enhancing from non-enhancing lesions using conventional serial M R I is not possible using only morphological 6 5 features. Only enhanced M R I is capable of making this distinction, but may not be necessary in some therapeutic trials (the I F N B Multiple Sclerosis Study Group, 1993). 66 3.5. The growth pattern of enlarging lesions in MS: observations in serial MRI 3.5.1. Introduction There are three kinds of active lesions detected in serial T 2-weighted M R I scans: new lesions that had not previously been seen; enlarging lesions that increase in size from a preexisting stable lesion or active lesion; and recurrent lesions that develop at the same site where a lesion had been previously seen. Stable lesions do not change morphologically at follow-up. We had shown in a previous study that enlarging lesions are more similar in size and shape to stable lesions than are new and recurrent ones. Therefore, this study of enlarging lesions can provide new information on both pathological and morphological aspects of M S . 3.5.2. Materials and methods Analysis of Magnetic Resonance Scans Researchers, experienced in M R I and masked to the clinical course, identified lesion enlargement. Enlarging lesions were grouped in terms of (1) growth direction of the enlarging lesions, (2) size of the lesion before and after enlargement, (3) the speed of change during the enlargement. In this study, lesion size before and after enlargement was defined as follows: (1) small lesion: less than 80 m m 2 in diameter; (2) moderate size lesion: 80 - 300 mm 2 ; (3) large lesion: greater than 300 mm 2 ; (4) confluent lesion: two or more lesions merged together. 67 3.5.3. Results Sixty-three enlarging lesions were identified. Before enlargement, 94% (59/63) were stable and 6% (4/63) were new, 63% (40/63) were small, 27% (17/63) medium, and 10% (6/63) were large lesions. After enlargement, 20% (13/63) were still small, 48% (30/63) medium, and 32% (20/63) large. The time from original size to maximum size was 6 weeks in 83%, (47 of the total 63 lesions), 12 weeks in 13% (8/63), 24 weeks in 1% (1/63), and 30 weeks in 3% (2/63). A l l but 4 were seen in the final scan, 75% (44/59) were smaller on the study subsequent (6 weeks) to the scan in which maximal size was seen. Fading time from maximum size to minimum was variable. 34 (20/59) reached smallest size in 6 weeks, 31% (18/59) in 12 weeks, 23% (14/59) in 18 weeks, 9% (5/59) in 24 weeks, and 3% (2/59) in 30 weeks. In 44% (26/59) the lesions returned to their original size; in 49% (29/59) the lesions were smaller but still larger than were originally seen. The direction of lesion growth was analyzed. Of the 24 periventricular lesions, 11 (46%) enlarged predominantly away from the ventricle, 3 anteriorly, 2 rostrally, 2 caudally, and 4 equally in all directions. O f the 39 non-periventricular lesions, 11 (28%) enlarged away from ventricles, 8 towards the ventricles, 2 posteriorly, 7 rostrally, 4 caudally, and 7 in all directions. The lesions located in the internal capsules enlarged only rostrally or caudally. 3.5.4. Discussion Using M R I , active M S lesions were often seen. In general, new lesions reached their peak in 2 - 4 weeks, either receded in 4 - 8 weeks or remained for relatively long periods. New lesions are probably due to B B B breakdown at the site of inflammation. The 68 inflammatory disturbance may or may not result in demyelination. We believe that the lesions which totally disappear may be only inflammatory in nature. Enlarging lesions differ from new lesions because they originate from previously stable ones and therefore represent re-activation. The majority of morphologically stable lesions are inactive, however they can enhance as their only sign of activity. It is unclear what signals a stable lesion to become active. New lesions were often seen in the non-periventricular region, round or oval in shape, and small or medium in size. In contrast, enlarging lesions are often seen in the periventricular region, irregular in shape, and larger in size. These characteristics of enlarging lesions are similar to those of stable ones. The factors that control the pattern of growth in M S lesions are unclear. The process may spread along the course of small veins and venules by direct expansion at the edge of active plaques. Our study provides evidence that M S lesions enlarge asymmetrically more often than concentrically. The enlargement may be along the course of venules or along the projections of white matter tracts as we saw in the internal capsules [Zhao, 1993]. 69 3.6. Sensitivity of gadolinium enhanced MRI and serial MRI in detecting activity of lesions in relapsing-remitting MS 3.6.1. Introduction M S lesions can be seen to be active in serial standard M R I (S-MRI) studies and with gadolinium enhanced M R I (G-MRI) . Active lesions are indistinguishable in appearance, distribution, shape, and size from stable ones i f imaged only once using a standard M R scan. G d enhanced M R I activity is due to blood-brain barrier ( B B B ) impairment. However, both S - M R I and G - M R I are time consuming and costly. Therefore, finding not only the most sensitive technique but adequate technique for detecting active lesions is important. 3.6.2. Methods MRI Protocol The enhanced scans were performed at 3 and 15 minutes after G d - D T P A administration (0.1 mmol/kg). Pre and post contrast scans were obtained using a T R of 683 msec and T E of 26 msec. Analysis of Magnetic Resonance Scans Two of us (G.J.Z. and B . L . T . ) , experienced in M R I and masked to the clinical course, identified lesions for activity. We identified active lesions as (1) new; (2) enlarging; and (3) recurrent lesions. The grade of enhancement after G d - D T P A was rated as follows: 0. no enhancement, 1. suspected vague enhancement, 2. faint but definite enhancement, 3. moderate enhancement, and 4. intense enhancement. 70 3.6.3. Results 308 active lesions were identified on S -MRI . 33 were studied with gadolinium. 14 od 33 (42%) were enhancing and 19 of 33 (58%) were non-enhancing. 29 other lesions were enhancing at entry for a total of 43 enhancing lesions. Table 8: Location of lesions Location Enhanced % Non-enhanced % (n = 43) (n = 19) Periventricular 15 34.9 4 21.1 White matter 21 48.8 6 31.6 G W junction 1 7 16.3 9 47.3 Enhanced vs. non-enhanced 2 in periventricular vs. non-periventricular location: p > 0.2 1. G W junction = junction of gray and white matter 2. Non-periventricular location = white matter and G W junction 71 Table 9: Signal intensity of enhancing lesions (n = 43) Time Post-injection Grade 3 min (%) 15 min ( % ) 0 12 27.9 0 0.0 1 8 18.6 11 25.6 2 13 30.2 12 27.9 3 7 16.3 13 30.2 4 3 7.0 7 16.3 Table 10: Mean area of lesions <30(mm2) (%) 30-100(mm2) (%) >100(mm2) (%) Non-enhanced (n= 19) 10 52.6 4 21.1 5 26.3 Enhanced (n=43) 8 18.6 26 60.5 9 20.9 Non-enhanced vs. Enhanced in 2 2 size of < 30 mm vs. > 30 mm : p < 0.01 72 Fourteen of the 43 enhancing lesions (33%) remained stable on follow up by S -MRI . 8 of 19 (42%) non-enhancing lesions disappeared at 3 months follow up compared to only 7 of 43 (16%) enhancing lesions. We found 3 lesions that enhanced when there was no morphological evidence of activity by S - M R I . Morphological enlargement was seen in all lesions at 6 weeks follow up by S - M R I . 3.6.4. Discussion Gadolinium enhanced M R I can be used as a marker of disease activity, notably breakdown in the B B B and inflammation. Several lines of evidence (pathological, animal experimental and immunocytochemical studies) [Kermode et al., 1991; Thompason et al., 1991] suggest that gadolinium enhancement ( B B B disruption) of M S lesions indicates the presence of inflammation. Indirect evidence that gadolinium enhancement indicates inflammation is provided by longitudinal M R I studies [Barkhof et al., 1992], showing that the vast majority of M S lesions display an initial phase of gadolinium enhancement, while tranditional pathology tells us that new, active lesions are characterized by inflammation. Gadolinium enhancement is a time limited feature of M S lesions [Grossman et al., 1988; Mi l l e r et al., 1988; Bastianello et al., 1990]. The evolution of gadolinium enhancement in M S lesions in several studies indicates a rather uniform time-frame of B B B disruption and reintegration, despite differences among the studies in patient selection and scanning protocol. Using monthly follow-up M R scans 22% [Miller et al., 1988] to 33% [Harris et al., 1991] of gadolinium enhancing lesions still enhance after 4 weeks. Some of the variance in persistence of gadolinium enhancement reported can be explained by variance in study design. This study showed that both G - M R I and S - M R I can detect lesion activity, but neither G - M R I nor S - M R I can show all active lesions. S - M R I is sensitive for seeing lesions 73 change in size, particularly smaller lesions. G - M R I is sensitive for monitoring pathological alteration in the lesion itself, even though there may be no change in morphological criteria. Therefore, S - M R I and G - M R I are complementary methods for detecting active lesions. 74 3.7. MR detection of multiple sclerosis lesion activity: effects of scanning frequency 3.7.1. Introduction Clinical evaluation of outcome in therapeutic trials in M S is difficult. The spontaneously variable natural history of M S requires clinical trials to be large and of long duration. Clinical scoring methods are relatively crude and do not accurately reflect the extent of underlying pathology. Serial M R I can detect brain lesion extent and activity in M S , including that of asymptomatic lesions. Unenhanced serial T 2 and P D M R I is now generally accepted as an important tool for monitoring therapeutic trials. However, the optimum scanning frequency interval is not yet universally agreed upon. The expense of frequent scanning, patient compliance and the natural history of the evolution of M S lesions are all factors that need to be considered when determining the optimum scanning frequency. In this study we attempted to determine (a) how much M S lesion activity information, as detected with frequent serial M R I (6 weekly) over a 2 year period, is lost if increased scan intervals are used and (b) at what point the treatment effect is lost i f scan intervals are increased. 3.7.2. Materials and methods Patient Population Forty five of the 50 patients had a cerebral M R scan every 6 weeks for a total of 18 examinations each. Two patients exited from the study after 14 examinations, two patients after 15 examinations, and one patient after 16 examinations. 75 Analysis of Magnetic Resonance Scans and Patients Lesion activity was assessed by observers, experienced in M R I and masked to the clinical course of patients, at scan intervals of 6 weeks. Changes in lesion size and number between scans were determined by comparative examination of previous scans. Only changes in lesions that were independently agreed on by all observers were recorded. A lesion enlarging continuously over several examinations was counted as active only once. A lesion seen in adjacent slices was also counted as one active lesion only. The evaluation was then repeated using only selected scans simulating scan intervals of 12 weeks, 24 weeks, and 48 weeks. We also determined the total number of patients with one or more active lesions for scan intervals of 6 weeks, 24 weeks, and 48 weeks. Statistical Methods For continuous variables, treatment-group differences were analyzed using an analysis of variance ( A N O V A ) based on ranked data. Percent differences were tested by Chi-square. 3.7.3 Results Thirty eight of 45 patients (84%) had a total 247 active M S lesions. There were 141 new, 77 enlarging and 43 recurrent lesions. Seven patients (16%) did not have any active lesions. Figure 7 shows the number of total active, new, enlarging and recurrent lesions that were seen with increased scan intervals. Overall a greater number of previously detected enlarging and recurrent lesions had disappeared as compared to the detection of new lesions (p< 0.001). 76 Fig. 7. Percentage of Active Lesions Versus Scan Interval 100% 100% 100% 100% 100% 80% 4-60% - -40% - -20% - -0% 55% 12 35% 24 21% 48 74% 12 53% 24 33% 48 44% 12 18% 12% 24 + New Enlarging Scan Interval (weeks) 9% 7% 2% Total Active Recurrent 48 77 Fig. 8. Mean Number of Active Lesions in Placebo Versus Treated Patients • Placebo B8 mlU I11.6 mlU 4.4 4.9 3.9 3.1 2.4 2.1 1.9 1.5 1.5 + 12 24 Scan Interval (weeks) 0.9 48 0.9 78 A s previously reported [Paty, et al. 1993], over the 2 years both high and low dose I F N B showed a steady reduction of lesion activity compared to placebo (p < 0.001). With longer scan intervals this treatment effect remains, however, it lost statistical significance with a scan interval longer than 48 weeks. (6 week interval, p = 0.0294; 12 week interval, p = 0.0324; 24 week interval, p = 0.0302; 48 week interval, p = 0.1103) (figure 8). 3.7.4. Discussion Previous longitudinal M R I scanning studies of M S show that disease activity patterns vary according to the clinical course. Isaac et al [1988] and Willoughby and coworkers [1989] reported disease activity in relapsing remitting M S , defined as the presence of new and/or enlarging lesions on biweekly follow-up non-enhanced M R I scans in 17 R R patients. Most of the M R I activity in that study was due to new lesions (83%). Subsequently in a serial M R study of the chronic progressive phase of M S , Koopmans and coworkers [1989a] reported a much more complex pattern of disease activity in 8 patients. In these patients most of the lesion activity (71%) was due to enlarging and recurring lesions and only 29% was due to new lesions. New lesions typically enlarged over 2 to 4 weeks with subsequent decrease in size over 4 to 6 weeks. The lesions usually left a small residual abnormality, but some of the smaller lesions faded entirely, leaving no definite trace. This is probably because the residual lesions are too small to be identified (beyond the resolution of the scanner). There is also a suggestion that more lesions disappear when thick (1 mm) slices are used on low field strength machines. The pattern of change in enlarging lesions was similar with a rapid increase to maximum size in about 4 weeks followed by a more gradual decline over a further 4 to 8 weeks leaving a residual abnormality. Recurrent lesions developed from previously stable, new, or enlarging lesions. Their changing pattern was similar to enlarging lesions. 79 Serial unenhanced M R I can play an important role for monitoring therapeutic trials of M S . Enhanced M R I wi l l show additional activity events as well . When evaluating new therapies for M S it is important to state the M R I outcome measure to be used. The expected outcome of a beneficial drug therapy for M S may include a decrease in the overall number of active lesions and less active scans or fewer patients with active disease in the treatment group. The major findings of this study were as follows: 1). Scanning at greater than 6 week intervals reduced the detection of active M S lesions. This decrease in activity was disproportionately more for enlarging and recurrent lesions than it is for new lesions. 2). The reduction in number of active lesions caused by increased scanning intervals was seen in all three groups (placebo, 1.6 M I U and 8 M I U IFNB) and was not significantly different among the groups. The treatment effect significance was lost when the scan interval was longer than 48 weeks. It is remarkable that a significant treatment effect could be seen with such a small sample size, using unenhanced M R I , at a scanning interval of 24 weeks. Our results have important implications for designing future therapeutic trials in which serial M R I is used to measure treatment outcome. For instance, i f one is studying patients with early R R - M S (in whom new lesions is most important for assessing disease activity) one may obtain adequate activity levels at scanning interval of 24 weeks. However, in R P M S patients ( i n whom enlarging and recurring lesions are important for assessing disease activity) more frequent scanning is a prerequisite. These findings suggest that the optimal scanning interval must vary depending on how the optimal treatment outcome is defined (number of active lesions or number of patients with active disease) and whether new or chronic M S patients are being studied. 80 3.8. Corticospinal tract lesions in multiple sclerosis: relationship between MRI activity and clinical course 3.8.1. Introduction M S lesions can be seen in any part of the C N S , including the corticospinal tract (CST). Active lesions which differ from stable ones in pathology, distribution, and shape can be identified on serial M R I studies. The C S T is a major descending pathway for controlling voluntary movement. We have examined the correlation between active lesions in the C S T and clinical findings. E D S S is a clinical standard for determining neurologic impairment in M S . Poor but significant correlation between M R I lesions and the clinical status of the M S patients was found in previous studies [Issac et al., 1988; Willoughby et al., 1989; Koopmans et al., 1989; Harris et al., 1991; Thompson et al., 1992; Wiebe et al., 1992]. The possible reasons for this lack of correlation between clinical and conventional brain M R I findings may be that there were only small numbers of patients in the studies [Harris et al., 1991; Capra et al., 1992]; the follow up was too short [Issac et al., 1988; Willoughby et al., 1989; Harris et al., 1991; Capra et al., 1992]; on the scan timing was not consistent with the appearance of active lesions so as to miss the time of blood-brain barrier breakdown in the active M S lesions [Kermode et al, 1990; Y o u l , 1991]. Furthermore, M S lesion location and size are very important factors as well and should be considered in the study analysis [Zhao et al., 1993; Plant, 1992]. This study, as part of the I F N B (Betaseron®) therapeutic trial [The I F N B Multiple Sclerosis Study Group, 1993; Paty et al., 1993], was carried out with 50 M S patients at the University of British Columbia. Patients were scanned every 6 weeks for 2 years. The study was to reveal (1) the dynamics of mostly asymptomatic active lesions; (2) 8 1 effectiveness of IFNB on M R I active lesions, and (3) correlation of M R I quantitative measure and clinical measure by E D S S and the correlation between location of active M R I lesions and clinical signs of activity. 3.8.2. Materials and methods Analysis of Magnetic Resonance Scans Lesions were identified by experienced and masked observers. Active lesions were identified as (1) new lesions; (2) enlarging lesions; and (3) recurrent lesions. According to active lesion location and any preexisting, inactive lesions, patients were divided into 3 groups: (1) active lesions involving the C S T above the pons only; (2) active .lesions involving the C S T above the pons with preexisting lesions in the brainstem and C S T ; (3) active lesions at other sites not in the C S T . After all active lesions were identified, clinical data was reviewed. In this study, lesion size was defined as (1) small, less than 1 cm in diameter; (2) moderate size, 1 - 2 cm; (3) large, greater than 2 cm; (4) confluent (see section 3.5.2). The active lesions associated with relapse were those which were detected in the scans just before or after 2 weeks of exacerbation. Clinical Measurement A relapse was to have occurred with a worsening of an old symptom or an appearance of a new symptom, attributable to M S ; accompanied by an appropriate new neurologic abnormality; lasting at least 24 hours in the absence of fever; and preceded by stability or improvement for at least 30 days. E D S S score was used to measure the disability. The progression of disability was defined as the persistent increase of one or more E D S S points. 82 3.8.3. Results Eleven M R I active lesions were identified in group 1, associated with 5 clinical relapses, all with an appropriate motor deficit (table 11). Fourteen M R I active lesions were seen in group 2 with 5 clinical relapses (2 motor, 2 visual, and 1 sensory) (table 12). Thirty active lesions were in group 3 with 8 relapses, none except one of which were appropriately motor in nature (table 13). Both active lesions and relapses were followed up by appropriate serial examinations in this study. The difference was stastically significant when the motor deficit rate in group 1 was compared with the rate in group 3 (p < 0.05). Of 11 active lesions in group 1, 5 (45%) active lesions were small; 6 (55%) active lesions were medium sized, 4 (36%) of the medium lesions were active with relapses. Ten (10) of the total of 14 active lesions in group 2 were medium sized or large and 4/14 were small. 4/10 medium sized and large lesions were active with relapses. Only 1 of the 4 small lesions was correlated with clinical relapses. In group 3, 18/30 active lesions were small and 12/30 were medium sized or large. 4/12 medium or large lesions were active with relapses and 4/18 small lesions were active with relapses, but none of the clinical relapses were appropriate to the CST. 83 a fl o tw O C/2 s o a OJ * J o a H U 0X3 a '> "o >• G Vi C j U cu #i> *3 u < H U l <U .5 o, o u. O , | O , I E-U <U <U I o o "ca o > 00 lU Si P . 3 1 "5 PH I—I Oi c 'oo I 1) Pi c O cu E-1 oo U <u > •*-> o < u O0 oo c co + + + + + m co vo . J •a cu > o U l DH e X ) (U > O i— P H e (U > o <U c o 5 .3 .3 bO bO vrj VO C VO ca o oo CO „ M 1) S ca C ca c « o ca u , C3 ca OH t i OH h ca T3 ca o u .22 o .22 C T3 T3 C T3 J 3 C/3 CO 0) C ca CD ca ca 00 "TH 00 8 8 r> ca > <u bO bO t5J0 U CU cu 1 *- >-> *^ *^ +-* j -S* pi o< j <-> 4 H * J * J * H Pi _) h j - I Pi -< N m m s s cu cu N #e c o "K <u E * 84 o, s o u O •4-> c '3 00 C o OO 13X1 s v a oo e o a <u >• o w H U £ c oo c o "K CJ > r H H 03 O u 4-* 03 00 0 o, ' 3 1 & "o PH o H oo U P i | l c, 3 PH HH N C O °Lo <D 00 P i 03 c o 'oo J H oo U > 4—1 u < 1) oo 00 c + + w oi 2 N - a - a > > o o I H O 0) >H <D P - c c 9/ e g o o S ° , VO VO oo c ^ o3 o 0O T3 "O « 2 « " i 03 J3 S 03 ra *2 O U Bl « & D OJ O <U <U T3 ~& C ~0 -O O 3 3 - .22 '5 > D D . O 03 03 «J I H oo o C 0) 3 - a ^ oo .00 •is oo -r £ £ > 03 00 W) <D 03 <D > <D * — 1 ^ VH £ £ I S 00 J j J J-> -4-J -4-J J J J J - N tn 1 - m QJ SD e #© "S W) u 03 e i -3 E N (9 £ 85 a s o u O H u o c 1/1 c #© 3 H ul 1) •*H .2 Oi o I H o, o, H OO U U H OJ OJ (U 3.1 o o P H 13 o O, OJ OJ oi 3 I O PH «4-l OJ o oo <u o lap > OJ * c ."2 oo OO OJ - 1 H u 00 N U oo + T j - i n r - ~ v o m c N v o i n T 3 > O OH E T 3 OJ > O OH OJ C o bO. 1 3 (U > O OJ > o I H OH e OJ C o 00 . T 3 > O OJ C 0 001 vo VO VO ^ VO vo l^ <N "O T 3 T3 T 3 ~0 OJ i) D u o u oo oo oo oo oo C 03 o3 03 cd o3 03 « 2 a a a a •§ g O CJ CJ o o ; o o u u o u o OO T3 T3 T3 -O C c o too 3 Cj c3 oo oo « £ to 00 -O OJ > g P -a 5 03 00 OJ 03 £ I * oo oo U C -O 00 oo OJ C X) oo £ oo 5 OJ 2 C c oo 00 OJ C x> c 13 S3 c 3 T3 OJ co - « M R 03 OJ a) J _ ) J Oi -O OOSSI-JSOO.OOOO 4 - * * H 4 _ j 4 _ » 4 H 4 H 4 J 4 J 86 3.8.4. Discussion MS is a disorder characterized by recurrent multifocal inflammation and demyelination within the CNS. CNS dysfunction is usually attributed to conduction block secondary to changes in membrane electrical properties resulting from demyelination [Waxman, 1982]. However, alterations in axons also occur [Barnes, 1991; Matthews, 1991]. The number of axons that traverse a plaque can be reduced to only a small fraction of those originally present and these remaining axons may be thin and exhibit structural abnormalities [Raines et al., 1989]. MRI can detect the natural history of pathology. Our previous MRI studies had found that active lesions can come and go at any time. The new lesions reach their peak in 2 - 4 weeks, either recede in 4 - 8 weeks or remain for relatively long periods. The active lesions are often seen in the deep cerebral white matter and at the gray-white matter junction. Gadolinium enhancing lesions are frequently seen which are considered to represent blood-brain barrier breakdown at the site of inflammation. The inflammatory disturbance may result in demyelination whereas the lesions which totally disappear may be only inflammatory in nature. Our follow-up in active lesions has found a correlation between neurological deficit and location of individual lesions in the corticospinal tract. The results from this study provide evidence that the active lesions associated with clinical exacerbations were larger than those not associated with exacerbation. These lesions associated with exacerbations also left larger abnormal areas in follow-up scan after 6 months. They, to the most extent, became stable and rarely were re-active. Pathologically, most stable lesions have a rather acellular appearance, consisting of a fibrillary astroglial feltwork that is responsible for the sclerosis, color, and texture of the tissue [Matthews, 1991]. Lesion location and size are important factors in the resulting neurological disfunction. Large active lesions, if located in the course of corticospinal 87 tracts and postchiasmal visual pathways, were likely to cause corresponding neurological defects in M S [Zhao et al., 1993; Plant et al., 1992]. This study supports the conclusions and provides further information between lesion activity and clinical status. In the present study of 50 patients followed for 2 years, a relationship between lesion location and activity in serial M R I and clinical relapses has been seen. The evidence provided in this study is that the many of clinical relapses were associated with medium sized or large lesions. If a lesion was large and/or located in a functional area such as C S T , there was a high possibility of neurological symptoms. The same correlation was also observed in previous studies [Filippi et al., 1995] and a recent study [Khoury et al., 1994]. In the study of 281 patients with M S , we had observed two conventional T2-weighted brain M R I scans, separated by an interval of 24 to 36 months. A t the time of each scan, clinical disability was rated using the E D S S . There was a significant correlation between the number of new or enlarging lesions and increasing disability in the R R M S [Filippi et al., 1995]. This result supports the present study. The larger total active area in scans associated with relapses provides evidence that accumulation of destructive pathological changes over time can result in symptoms. Lesion location is very important as well. Symptoms may not occur in patients i f an active lesion occurs in a silent area even i f the active lesion area is large. However, i f a lesion is of considerable size and located in a functional area, symptoms wi l l occur with a higher probability. This study supports the hypothesis that active lesions located in the course of the C S T are more likely to be associated with a motor deficit than are other active lesions [Zhao et al., 1993]. In conclusion, the present study has revealed a positive relationship between active lesions seen in the C S T on M R I and clinical signs. Large lesions are also more likely than small lesions to be expressed clinically. This result shows that brain serial M R I is a useful supplementary marker of disease activity with clinical correlation in clinical treatment trials in M S . 88 Chapter 4 Using Lesion Activity as An Outcome Measurement to Monitor Therapeutic Trials in MS 4.1. Introduction M R I is the most sensitive scanning technique for the demonstration of M S lesions in the C N S and is an objective quantitative outcome measure for assessing the response of M S patients in a therapeutic trial. M S is a common neurologic disease, clinically characterized by the presence of symptomatic lesions in the C N S and by progressive neurologic impairment. Serial M R I has revealed a high rate of lesion activity in M S . Active lesions, which are often silent, occur at a significantly higher frequency than do clinical relapses. M R I provides the method for monitoring burden of disease as well as lesion activity as outcome measurements. Burden of disease accumulates with active lesion appearance. The hypothesis in a therapeutic trial is that the drug slows or stops the progression of M S if it can slow or stop lesion activity. We previously reported a 2 year interim M R I analysis, using burden of disease measurement, which showed I F N B was effective [The I F N B M S Study Group, 1993; Paty et al., 1993]. In a five year study, using yearly M R I , the positive effects of I F N B on clinical outcome and M R I burden of disease measurement were documented [The I F N B Multiple Sclerosis Study Group and the U B C M S / M R I Analysis Group, submitted, Koopmans, Zhao et al., submitted] 89 The present study is to examine the efficacy of I F N B treatment by an independent M R I analysis of M S lesion activity based on 6 week M R I scans of our 50 Vancouver patients for a total of 2 years. 4.2. Materials and methods Patient Population We studied 50 patients with clinically definite relapsing remitting M S who had been entered into a therapeutic trial using beta-interferon. There were 12 men and 38 women whose ages ranged between 22 and 53 years (mean, 38.6 years). Mean duration of disease was 8.2 years (range, 1.1 to 20.0 years; S D = 5.0 years). Disability of patients was between 0 and 3.5 (mean, 2.0, S D = 1.0) on the E D S S . A H patients had an abnormal cerebral M R I characteristic of M S , with scattered areas of increased intensity on the spin-echo (SE) sequence. Patients were randomized into three treatment arms: Placebo (N=17), 1.6 m l U (N=17), 8 m l U (N=16) I F N B self-administered subcutaneously every other day. Each patient had a cerebral M R scan every 6 weeks for 2 years. Clinical examination was also done on patients every 6 weeks (detailed description on clinical examination and method of patient distribution into three treatment arms seen in our reports [The I F N B Multiple Sclerosis Study Group, 1993; Paty, 1993]). MR Scan Technique M R scans were performed using a Picker International Cryogenic M R 2000 scanner with a superconducting magnet operating at a static magnetic field strength of 0.15 Tesla and a 30 cm diameter receiver coil in Vancouver Hospital and Health Science Center, University of British Columbia Site. Twelve contiguous 10 mm thick axial slices were obtained through the brain from the upper cerebral hemisphere to the medulla. The 90 inplane resolution was 1 mm. A dual echo S E sequence was used with echo delay times (TE) of 60 and 120 msec and a repetition time (TR) of 2133 msec. The same receiver coil (30-cm diameter) was used for each patient. The repositioning error was minimized through alignment of external and internal landmarks. Initial repositioning was based on two angles from external landmarks (the canthomeatal line and the nasion-tragal line). A midline sagittal slice (pilot) was then obtained and the head position verified using an internal angle (angle between top of the cerebellum and the anterior sphenoid sinus). If this angle differed by more than 2 degrees from the baseline study, the patient was repositioned and the pilot scan repeated. Slices for the actual scans were then programmed so that the middle slice of a simultaneous 12-slice series was centered on the top of the cerebellum. Halfway through an individual scanning session, the pilot scan was repeated to ensure that positioning was still adequate. If movement had occurred, the patient was repositioned and the appropriate sequence repeated. Quantitative Analysis of Magnetic Resonance Images A n interactive computer program was been designed to display M R images and permit manual tracing of lesions. The lesion borders were outlined on a computer monitor by one technician and checked by a single radiologist in order to reduce interobserver variation. The area of the individual lesions could then be determined and analyzed. The lesion areas were summed slice by slice for a total lesion area and recorded in mm 2 . Lesions were identified for activity. Changes in lesion size and number between scans were determined by 3 observers by comparative examination of previous scans. Only changes in lesions that were independently agreed on by all observers were recorded. A lesion enlarging continuously over several examinations was counted as active only once. A lesion seen in adjacent slices was also counted as one active lesion only. Active lesions were defined as (1) new when a new lesion seen was identified; (2) enlarging when a previously stable lesion demonstrated increases in size; (3) recurrent 91 when a lesion reappeared in the same location as one that had previously faded. A n active scan was a scan in which active lesions were detected. Active lesions were marked on the films by radiologists and outlined using our computer program on a monitor by a trained technician. Total active area was summed slice by slice. The total active area was defined as the summed area of all active lesions found in all previous scans. Statistics A s sample averages and medians were similar, descriptive statistics were reported as medians, means, and standard errors (SEs). A N O V A in excel software was reported for testing p values. 4.3. Results In 50 M S patients, 287 active lesions were detected, in which 63% (181/287) were new, 27% (77/287) enlarging, and 10% (29/287) recurrent. Significant treatment group differences were detected between the placebo and both treatment groups for the active lesion rate in M S patients. The 8 M I U group had a median reduction rate of 83% (Table 14). The annual rate of new lesions was also significantly lower for both the 8 M I U and 1.6 M I U treatment groups than for the placebo group. The 8 M I U group showed a median reduction of 75% in the rate of new lesion formation (Table 15). We examined the effect of treatment on enlarging and recurrent lesions. No significant differences in the rate of enlarging lesions were observed among the three treatment groups (Table 16). There was no overall significant treatment-group difference in the recurrent lesion rate (Table 17). No significant difference between the placebo and 8 M I U groups was detected. 92 There were 182 active scans. The percent of scans with activity showed a significant difference between the placebo and the treatment groups (Table 18). The 8 M I U group had a median of 80% fewer active scans than were present in the placebo arm. Mean cumulative percent of active scans per patient per scan week showed a significant difference as well (Figure 9). Some new lesions come and go totally, but many of them leave a small abnormal area. Of the 139 lesions followed for 36 weeks, 28 out of 139 (20%) were in the 8 M I U group, 82% (23/28) of them disappeared. 32 out of 139 active lesions (23%) were in the 1.6 M I U group and 59% (19/32) of them resolved totally. 79 of the total lesions (57%) were in the placebo group and 59% (47/79) of them disappeared. It is difficult to determine active lesions on a single conventional unenhanced M R I scan. The first scan in a serial study is used as a baseline to compare with follow-up scans. Therefore, the first active lesion in this study could be detected only as a change in the second scan. The total active area starting with the second scan was accumulated scan by scan for a total of 17 scans. The difference in rate between high dose arm and placebo arm was statistically significant ( 8 M I U vs. Placebo, p = 0.0408; 1.6 M I U vs. Placebo, p = 0.3702). The means of the M R I cumulative active area is seen in Figure 10. 93 Table 14. Active lesion rate in Vancouver patients (N = 50) Measurement Statistic Placebo 1.6 MIU 8 MIU Active lesion per year Mean 4.9 1.8 2.0 Median 3.0 1.0 0.5 SE 1.3 0.4 0.7 Overall: Placebo vs. 8 M I U Placebo vs. 1.6 M I U 1.6 M I U vs. 8 M I U p = 0.0234 p = 0.0089 p = 0.0412 p = 0.5070 94 Table 15. New lesion rate in Vancouver patients (N = 50) IFNB Measurement Statistic Placebo 1.6 MIU 8 MIU New lesion per year Mean 3.2 1.1 1.2 Median 2.0 0.5 0.5 SE 0.9 0.2 0.9 by ANOVA Overall: Placebo vs. 8 MIU Placebo vs. 1.6 MIU 1.6 MIU vs. 8 MIU p = 0.0085 p = 0.0026 p = 0.0317 p = 0.3207 9 5 Table 16. Enlarging lesion rate in Vancouver patients (N = 50) IFNB Measurement Statistic Placebo 1.6 MIU 8 MIU Enlarging lesion per year Mean 1.2 0.7 0.6 Median 1.0 0.0 1.0 SE 0.7 0.5 0.4 by ANOVA Overall: P = 0.2484 Placebo vs. 8 MIU P = 0.1441 Placebo vs. 1.6 MIU P = 0.2501 1.6 MIU vs. 8 MIU P = 0.7177 96 Table 17. Recurrent lesion rate in Vancouver patients (N = 50) IFNB Measurement Statistic Placebo 1.6 MIU 8 MIU Recurrent lesion per year Mean 0.7 0.3 0.4 Median 1.0 0.0 1.0 SE 0.4 0.2 0.2 by ANOVA Overall: P = 0.1648 Placebo vs. 8 MIU P = 0.2970 Placebo vs. 1.6 MIU P = 0.0945 1.6 MIU vs. 8 MIU P = 0.3273 97 98 99 Table 18. Percent of scans with activity in Vancouver patients (N = 50) IFNB Measurement Statistic Placebo 1.6 MIU 8 MIU Percent of active scans Mean 34.6 17.0 15.4 Median 29.4 11.8 5.9 SE 6.0 3.8 4.5 by A N O V A Overall: P = 0.0170 Placebo vs. 8 M I U P = 0.0062 Placebo vs. 1.6 M I U P = 0.0349 1.6 M I U vs. 8 M I U P = 0.4692 4.4. Discussion Natural interferon beta (IFNB) is produced by fibroblasts, and probably by many other cells, i f stimulated with synthetic or viral oligonucleotides. Human I F N B consists of a single molecular species that is a glycoprotein. Its gene is encoded on chromosome 9, and it reacts with a cell surface receptor in common with I F N alpha [ De Maeyer & De maeyer-Gignard 1988; Panitch, 1992; Clanet M , 1989]. Interferon beta has been reported to be effective in the treatment of R R M S [The I F N B M S Study Group, 1993; Paty et al., 1993]. A s part of an interferon beta therapeutic trial, this study provided further evidence 100 of the drug's effectiveness. The cumulative active area on M R I in the placebo arm increased much faster than those in the treatment arms (fig. 10). There was also a higher proportion of active lesions associated with relapses in the placebo arm than those in the treatment arms. The cumulative active areas in M R I included both new appearing lesions and the lesions which were seen as active ones in all previous scans. Therefore, the cumulative pathology contained the edema, inflammation, demyelination, and/or axon loss. Measurement of new lesions is important for a therapeutic trial. Increase in lesion load relies mainly on the appearance of new lesions and enlargement of previously stable lesions. In another study [Zhao et al., in preparation] a higher percentage of new lesions was detected in patients with lower initial lesion load. More enlarging lesions were detected in patients with higher initial lesion load. The lower lesion load group had a faster increase in percentage of lesion load and the higher lesion load group had a greater absolute increase in actual lesion load. I F N B decreased both the lesion activity and slowed down the progress of M R I detected pathology. In addition, using yearly M R I over 5 years [Koopmans, Zhao et al., submitted], we examined the efficacy of I F N B by an independent M R I analysis of yearly M S lesion activity. The data supported the present results, that I F N B decreased M S lesion activity and total lesion load. The serial M R I data further confirmed the clinical study [The I F N B M S Study Group, 1993] showing a significant reduction in disease activity. The serial M R I data provide a sensitive and objective method to measure lesion activity. B y using the frequent M R I method and fewer patients, future therapeutic trials and dosage trials in M S can be shorter in duration and less expensive. In summary, lesion activity measurement is a sensitive and objective method by which we can monitor a therapeutic trial. 101 Chapter 5 Future Application of MRI for the Understanding of MS Pathology 5.1. MR Spectroscopy in Multiple Sclerosis Whereas M R I is an anatomic imaging modality in a long tradition of anatomic imaging techniques in radiology, M R Spectroscopy (MRS) is unique. It is one of few techniques in clinical medicine that provide noninvasive access to living chemistry in situ. In vivo M R S can be used to measure relative concentrations and mobilities of different low-molecular-weight chemicals. Proton M R S can be used to monitor metabolites such as choline, creatine, /V-acetyl aspartate ( N A A ) , mobile lipids, and lactic acid. In vivo M R S can be used to measure relative concentrations and mobilities of different low-molecular-weight chemicals [Gadian et al., 1982]. The lipid-methylene protons resonate at 1.2 ppm in the proton spectrum and contain mainly contributions from triglycerides and other mobile lipids. L i p i d metabolism is important in the study of demyelination because the myelin sheath is composed of a lipid-bilayer membrane. Histopathologic techniques have allowed researchers to study lipid changes during demyelination because the lipids stain differently depending on their structural and biochemical state. These studies have shown that abnormal l ipid droplets that form during demyelination are a sign of an irreversible lesion. The signal intensity of membrane lipids in M R S is influenced by the fluidity of the membrane structure, which depends on temperature, cholesterol content, l ipid composition, lipid-water interactions, and membrane proteins. The proton spectrum of normal brain has little (if any) l ipid 102 signal [Frahm J et al, 1989] because the phospholipids in cell membranes and in myelin have limited mobility. However, it is hypothesized that as myelin degrades and the membrane structure changes, the lipids become less ordered and thus become able to give a detectable M R spectroscopic signal. The N-methyl groups of choline resonate at 3.2 ppm in the proton spectrum in vivo and contain contributions mainly from phosphorylcholine and glycero-3-phosphorylcholine (GPC), which are involved in l ipid metabolism. Phosphorylcholine is a precursor molecule that is incorporated into the polar head group of phosphatidylcholine and sphingomyelin [Tunggal B , et al. 1990]. These phospholipids are then incorporated into cell and myelin membranes of the C N S . G P C is a metabolite of the breakdown of phosphatidylcholine. Unfortunately, all the N-methyl groups from the choline-containing compounds resonate at the same frequency. The contribution of phosphatidylcholine to this resonance is unknown. Phosphorylcholine and G P C are small molecules with long T2 relaxation times, whereas phosphatidylcholine is a large molecule with a relatively short T2 relaxation time that is made even shorter by being incorporated into the highly ordered myelin membrane. However, recent evidence suggests that phosphatidylcholine contributes to the proton M R choline signal [Miller D L , et al. 1991]. The methyl group of 7V-acetylaspartate ( N A A ) has a sharp resonance at 2.0 ppm in the proton spectrum of the brain. The exact role of N A A in brain tissue is still unclear. Most researchers agree that N A A is a marker for neurons and may have a role in neuronal metabolism [Birken D L , 1989, G i l l SS, 1989]. Although some researchers claim that N A A is involved in myelin production, this seems to be contradicted by the fact that N A A is at a much higher concentration in the neurons than in the glial cells [Birken D L , 1989, G i l l SS, 1989]. The N A A resonance has been of interest in the study of M S and other brain diseases because of the decrease in NAAxrea t ine and NAA:cho l ine ratios localized to brain lesions. This decrease in N A A has been attributed to axonal loss [Arnold D L , 1990], active degradation of N A A in injured neurons [Bruhn H , 1989], and 103 gliosis [Van Heche P, 1991]. Although M S is primarily a demyelinating disease, injury to neurons becomes more apparent with decade-long progression of the illness, possibly because of the metabolic relationship between axons and myelin-forming oligodendroglia. Along with the injured neurons, the normal-appearing "healthy neurons" may be influenced by the pathologic materials arising from inflammatory foci and the altered B B B . M S patients have been studied with in-vivo proton spectroscopic techniques in which the spectroscopic localized volume was placed over a region of brain that included a plaque defined by M R I . The main findings have been a decrease in NAA/creatine and an increase in choline/creatine. In some cases, the decrease in N A A did not depend on relative plaque volume inside the spectroscopic localized volume, suggesting that the disease extended into normal-appearing white matter surrounding the active chronic M S plaques. Acute (edema, inflammation) and chronic (demyelination, gliosis) M S plaques appear to have different proton spectra, potentially allowing the number of irreversible chronic plaques in a patient to be determined spectroscopically. The hypothesis is that metabolite ratios are unchanged in hyperacute (edema only) plaque, that demyelination in acute plaques is accompanied by an increased choline/creatine ratio, and that irreversible plaques in the subacute to chronic stage have a decreased NAA/creatine ratio. Some claims have been made of l ipid increases in M S plaques [Koopmans, et al., 1990; Wolinsky, et al., 1990]; however, one of these investigators admits the possibility of l ipid contamination from scalp and bone marrow [Koopmans, et al., 1990]. Other proton metabolites that have been reported to change in M S are lactate and inositol. It is now possible to measure the proton metabolites in patients with M S with spectroscopic imaging techniques. These studies have added insight into the distribution of metabolite concentration within a brain lesion, den Hollander et al. [1987] reported that N A A and choline are not uniform over the whole area of the M S plaques and that certain 104 plaques have foci of high concentrations of choline or lactate. They hypothesized that these foci may indicate a different stage of the pathology. The ability to differentiate between early reversible lesions and irreversible lesions would be useful in planning treatment of M S . Some M S lesions are purely edematous or inflammatory and should respond to corticosteroids and other anti-inflammatory drugs. These early lesion should be detectable by an increased intensity on T2-weighted M R I and no changes in proton metabolites on M R spectroscopy. However, as M S progresses, lipids in macrophages and mobile lipids increase months or years later, myelin lipids are lost and are replaced by astrocytic gliosis (decrease in N A A ) . These older lesions most likely wi l l not respond to anti-inflammatory treatment. Thus, proton M R S along with M R imaging could be helpful in distinguishing those early lesions that might respond to therapy. 5.2. Summary and future directions M R I has become the most prominent imaging modality for lesions of the C N S . The many tissue and machine parameters involved in the image formation, give M R I a unique capability to separate different healthy tissues and to identify pathological features. It is clear that M R I has made a major impact in the diagnosis of M S and in monitoring therapeutical trials. It is also possible to compare the site of lesions identified on M R I with the clinical condition of the patient, to establish the "lesion load", the number of events, and to follow the natural history of the different types of lesions as well as different clinical categories of M S . A s a consequence, M R I has brought many new insights into the understanding of M S . G d - D T P A is a paramagnetic contrast agent, with minimal side-effects, which adds considerable potential to M R I . Due to the variation in G d - D T P A uptake by various 105 tissues, it creates the possibility to manipulate tissue contrast on M R I . The increase in contrast is especially effective in the C N S , where the B B B selectively controls tissue perfusion, and normally prevents G d - D T P A entering the brain. Breakdown of this barrier, for example in inflammation, allows G d - D T P A to enter the brain, which locally increases M R I contrast considerably. Therefore, in M S , G d - D T P A is capable of differentiating new, actively inflamed lesions from non-inflammatory, chronic lesions. This thesis illustrates that active lesions, as a marker of disease activity, reinforces the innovative role of M R I in M S studies. Study in scanning frequency provides a powerful reference in designing new therapeutic trials. The results in monitoring lesion activity as an outcome measure therapeutic trials provide evidence that lesion activity measurement in small group of patients is still a useful and sensitive tool. Furthermore, other important issues in the evolution of M S lesions need to be addressed as well . M R I is capable of providing information on a number of critical issues, such as differentiation of lesions, seen on non-enhanced images, in relation to impairment: partly demyelinated lesions, insufficiently remyelinated lesions, gliotic lesions with preservation of axons, and gliotic lesions with loss of axons. Future M R research is aimed at identifying specific pathologies and monitoring their response to therapy. Newly developed M R methods might be quite helpful in attaining this goal. Magnetic transfer imaging (MTI) is a new M R technique that can be used to obtain information about large molecules such as phospholipids in myelin membranes. This method offers the possibility to quantitate myelin loss, and possibly also remyelination. Combination of gadolinium-enhancement and M T I may provide synergistic information. Diffusion M R I can identify changes in myelin structure at a very early stage. M R spectroscopy identifies specific biochemical metabolites involved in the formation and breakdown of myelin, neuronal function, and neurotransmission. Proton spectroscopy already has shown differences between active lesions and stable ones. Changes in choline, N-acetyl aspartate, myoinositol, lactate and free lipids have been 106 demonstrated, but await further confirmation. M R spectroscopy wi l l become especially valuable when combined with imaging techniques. Each of these new M R techniques could open new insights into the tissue characteristics of M S lesions and their evolution through specific stages, just as lesion activity studies have increased our knowledge about M S . Hopefully, new M R techniques wi l l allow more precise differentiation of tissue alterations and response to therapy. Their use offers a short cut towards detecting and implementing an effective therapy for M S . 107 References Abbott NJ. Glia and the blood-brain barrier. Nature 1987; 325:195. Adams CWM. The general pathology of multiple sclerosis: Morphological and clinical aspects of the lesions, in Multiple sclerosis: pathology, diagnosis, and management. Hallpike JF, Adams CWM, Tourtellotte WW. ed. London, Chapman and Hall. 1983 Adams CWM. Vascular aspects of multiple sclerosis. In: A colour Atlas of Multiple Sclerosis and Other Myelin Disorders. Adams CWM (Ed). London: Wolfe Medical 184-187. Adams RD, Victor M. Multiple sclerosis and allied demyelinative diseases, in Principle of neurology. McGraw-Hill Info Serv Comp 1989; 755-774 Alexander JT, et al. The effect of interlukine-2 on the blood-brain barrier in the 9L gliosarcoma rat model. J Neurosurg 1989; 70:92-96. Andersen O. Restricted dissemination of clinically defined attacks in an MS incidence material. Acta Neurol Scand 1980;62(suppl 77): 1 Annunziata P, et al. Frequency of blood-retina and blood-brain barrier changes in multiple sclerosis. Ital J Neurol Sci 1988; 9:345-349. Arnason BGW. Interferon beta in multiple sclerosis. Neurology 1993;43:641-643 Arnold DL, Matthews PM, Francis G, Antel J. Proton magnetic resonance spectroscopy of human brain in vivo in the evaluation of multiple sclerosis: assessment of the load of disease. Magn Reson Med 1990;14:154-159 Barkhof F. Gadolinium enhanced magnetic resonance imaging in multiple sclerosis. VU University Press, Amsterdam. 1992;29-61 Barnes D, et al. The characterization of experimental gliosis by quantitative nuclear magnetic resonance imaging. Brain 1988; 111:83-94. Barnes D, Munro PMG, Youl BD, McDonald WI. The longstanding MS lesion: a quantitative MRI and electron microscopic study. Brain 1991 ;114:1271-1280 Bastianello S, Pozzilli C, Bernardi S, Bozzao L, Fantozzi LM, Buttinelli C, Fieschi C. Serial study of gadolinium-DTPA MRI enhancement in multiple sclerosis. Neurology 1990;40:591-595 Birken DL, Oldendorf WH. N-acetyl-aspartic acid: a literature review of a compound prominent in 1H-NMR spectroscopic studies of brain. Neurosci Biobehav Rev 1989;13:23-31 Bloch F,'Hansen WW, Packard M. Nuclear induction. Phys Rev 1946;70:460-474 Bobowick AR, Kurtzke JF, Brody JA, et al. Twin study of multiple sclerosis: an epidemiologic inquiry. Neurology 1978;28:978-987 \ Brainin M, Reisner T, Beuhold A, et al. Topological characteristics of brainstem lesions in clinically definite and clinically probable cases of multiple sclerosis: an MRI-study. Neuroradiol 1987;29:530 Brightman MW, Reese TS. Junctions between intimately apposed cell membranes in the vertebrate brain. J Cell Biol 1969; 40:648-677. Broman T. Blood-brain barrier damage in multiple sclerosis, soprevital test-observations. ACTA Neurol Scand 1964; 40(Suppl 10):21-24. 109 Broman T, Andersen O, Bergmann L: Clinical studies on multiple sclerosis. Acta Neurol Scand 1981;63:6 Brown WJ. The capillaries in acute and subacute multiple sclerosis plaques: a morphometric analysis. Neurol 1978; 28:84-92. Brownell B and Hughes JT. The distribution of plaques in the cerebrum in multiple sclerosis. J Neurol Neurosurg Psychiat 1962;25:315-320 Bruhn H, Frahm J, Gyngell ML, et al. Cerebral metabolism in man after acute stroke: new observations using localized proton NMR spectroscopy. Magn Reson Med 1989;9:126-131 Bundgaard M. Pathways across the vertebrate blood-brain barrier: Morphological viewpoints. In HF Cserr (Ed) The Neuronal micronment. Annals of the New York Academy of Science 1986; 481:7-19. Burnham JA, Wright RR, Dreisbach J, Murray R. The effect of high-dose steroids on MRI gadolinium enhancement in acute demyelinating lesions. Neurology 1991; 41:1349-1354. Cancilla PA, Berliner JA, Bready JV. Astrocytes and the blood-brain barrier. Kinetics of astrocyte activation after injury and induction effects on endothelium. In Johansson BB, Ovman C, Widner H (Ed). Pathophysiology of the blood-brain barrier. Elsevier 1990; 31-39. Capra R, Marciano N, Vignolo LA et al. Gadolinium-pentetic acid magnetic resonance imaging in patients with relapsing remitting multiple sclerosis. Arch Neurol 1992;49:687-689 110 Cendrowski WS. Multiple sclerosis: Discordance in three pairs of dizygotic twins. J Med Genet 1968;5:266-268 Clanet M, Blancher A, Calvas P, Rascol O. Interferon and multiple sclerosis. Biomed & Pharmacother 1989;43:355-360 Claudio L, Kress Y, Norton WT, Brosman CF. Increased vesicular transport and decreased mitochondrial content in blood-brain barrier endothelial cells during experimental autoimmune encephalomyelitis. Am J Pathol 1989; 135:1157-1168. Claudio L, Kress Y, Factor J, Brosman CF. Mechanisms of edema formation in experimental autoimmune encephalomyelitis. The contribution of inflammatory cell. Am J Pathol 1990; 137:1033-1045. Cserr HF, Bundgaard M. The neuronal microenvironment: a comparative view. In HF Cserr (Ed) The neuronal micronment. Annals of the New York Academy of Sciences 1986; 481:1-7. Currier RD, Eldridge R. Possible risk factors in multiple sclerosis as found in a national twin study. Arch Neurol 1982;39:140-144 Damadian R. Tumor detection by nuclear magnetic resonance. Science 1971 ;171:1151-1153 Damle NK, et al. Interleukin 2-activated human lymphocytes exhibit enhanced adhesion to normal vascular endothelial cells and cause their lysis. J Immunol 1987; 138:1779-1785. de Graaf J, van der Hoeven JH, Hoogstraten MC, Minderhoud JM. Multiple sclerosis. Utrecht: Bohn, Scheltema & Holkema 1988 De Maeyer E, De Maeyer-Guignard J. Chapter 2: The interferon gene family. In De Maeyer & De maeyer-Guignard (Eds) Interferons and other regulatory cytokines. John Wiley & Sons, New York 1988:5-38 1 31 den Hollander JA, Luyten PR. Image-guided localized H and P NMR spectroscopy of humans. Ann NY Acad Sci 1987; 508:386-398 Dorovini-Zis K, Bowman PD, Prameya R, et al. Adhesion and migration of human polymorphonuclear leukocytes across cultured bovine microvessel endothelial cells. J Neuropathol Exp Neurol 1992;51:194-205 Ebers GC, Bulman DE, Sadovnick AD, et al. A population-based study of multiple sclerosis in twins. N Engl J Med 1986;315:1638-1642 Eickhoff K, et al. Protein profile of cerebrospinal fluid in multiple sclerosis with special reference to the function of the blood-brain barrier. J Neurol 1977; 214-207-215. Ellison, et al. Blood-brain barrier dusfunction in cats following recombinant interleukine-2 infusion. Cancer Res 1987; 47:5765-5770. Engell T. A clinical patho-anatomical study of clinically silent multiple sclerosis. Acta Neurol. Scand. 1989;79:428-430 Filippi M, Paty DW, Kappos L, Barkhof F, Compston DAS, Thompson AJ, Zhao GJ, Wiles CM, McDonald WI, Miller DH. Correlations between changes in disability and T2-weighted brain MRI activity in multiple sclerosis: a follow up study. Neurology 1995;45:255-260 112 Fontana A and Fierz W. The endothelium-astrocyte immune control system of the brain. Sprin Semin Immunopathol 1985;8:57-70. Frahm J, Bruhn H, Gyngell ML, et al. Localized high-resolution proton NMR spectroscopy using stimulated echoes: initial applications to human brain in vivo. Magn Reson Med 1989;9:79-93 French Research Group on Multiple Sclerosis. Concordance rate is independent of zygotic. Ann Neurol 1992;32:724-727 Frohman EM, Vayuvegula B, van den Noort S, Gupta S. Norepinephrine inhibits gamma-interferon-induced MHC class II (Ia) antigen expression on cultured brain astrocytes. J Neuroimmunol 1988; 17:89-101. Frohman EM, et al. Astrocytes and intracerebral immune response. J clin Immunol 1989; 9:1-9. Gadian DG. NMR and its application to living systems. New York, Oxford University Press 1982 Gay D and Esiri M. Blood-brain barrier damage in acute multiple sclerosis plaques: An immunocytological study. Brain 1991;114:557-572 Gill SS, Small RK, Thomas DGT, et al. Brain metabolites as 'H NMR markers of neuronal and glial disorders. NMR Biomed 1989;2:196-200 Gross PM, et al. Blood-brain barrier and histamine: quantitative evidence for increased permeability in rat cerebral cortex. J cereb Bio Flow Metab 1981; LS379-380. Grossman RI, et al. Multiple sclerosis: gadolinium enhancement in MR imaging. Radiology 1986; 161:721-725. 113 Grossman RI, et al. Multiple sclerosis: serial study of gadolinium-enhanced MR imaging. Radiology 1988; 169:117-122. Gulati A, et al. Evidence for the involvement of histamine in the regulation of blood-brain barrier permeability. Pharmaco Res Comm 1985; 17:395. Hallpike JF, Adams CWM, Tourtellote WW. Multiple sclerosis: Pathology, diagnosis and management. Baltimore: Williams & Wilkins, 1983. Harris JO, Frank JA, Patronas N, McFarlin DE, McFarland HF. Serial gadolinium-enhanced magnetic resonance imaging scans in patients with early, relapsing-remitting multiple sclerosis: implications for clinical trials and natural history. Ann Neurol 1991;29:548-555 Hawkins CP, Munro PMG, Mackenzie F, Kesselring J, Tofts PS, et al. Duration and selectivity of blood-brain barrier breakdown in chronic relapsing experimental allergic encephalomyelitis studied by gadolinium-DTPA and protein markers. Brain 1990;113:365-378 Heurkeas, AHM, et al. Anti-endothelial cell antibodies in patients with rheumatoid arthritis complicated by vasculitis. Clin Exp Immunol 1989; 78:7-12. Hickey WF, Kimura H. Perivascular microglial cells of the CNS are bone-marrow derived and present antigen in vivo. Science 1988; 239:290-193. Hreshey LA, et al. Computerized tomography in the diagnosis evaluation of multiple sclerosis. Ann Neurol 1979;5:32-39. The IFNB Multiple Sclerosis Study Group. Interferon beta-lb is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 1993;43:655-661 114 The IFNB Multiple Sclerosis Study Group and the UBC MS/MRI Analysis Group. Beta-interferon lb int he treatment of MS: final outcome of the randomised controlled trial. Neurology (in submitted) Isaac C, Li DKB, Genton M, et al. Multiple sclerosis: a serial study using MRI in relapsing patinets. Neurology 1988;38:1511-1515 Janzer RC, Raff MC. Astrosytes induces blood-brain barrier properties in endothelial cells. Nature 1987;325-253-257. Johnson MA, Li DKB, Bryant DJ, Payne JA. Magnetic resonance imaging: serial observations in multiple sclerosis. AJNR 1984;5:495-499 Juhler M , et al. Blood-brain and blood-spinal cord barrier permeability during course of experimental allergic encephalomyelitis in the rat. Brain Res 1984; 302:347-355. Juhler M, et al. A spatial analysis of the blood-brain barrier damage in experimental allergic encephalomyelitis. J cereb Bio Flow Metab 1985; 5:545- 553. Kajiwara K, Ito H, Fukumoto T. Lymphocyte infiltration into normal rat brain following hyperosmotic blood-brain narrier opening. J Neuroimmunol. 1990; 27:133-140. Kato S and Nakamura H. Ultrastructural and ultracytochemical studies on the blood-brain barrier in chronic relapsing experimental allergic encephalomyelitis. ACTA Neuropathol 1989;77:455-464. Kermode AG, Tofts PS, MacManus DG, Kendall BE, Kingsley DPE, Moseley IF et al. Early lesions of multiple sclerosis. Lancet 1988 ii; 1203-1204 Kermode AG, Tofts PS, Thompson AJ, et al. Heterogeneity of blood-brain barrier changes in multiple sclerosis: an MRI study with gadolinium-DTPA enhancement. Neurology 1990a;40:229-235 115 Kermode AG, Thompson AJ, Tofts PS, et al. Breakdown of the blood-brain barrier precedes symptoms and other MRI signs of new lesions in multiple sclerosis. Brain 1990b;l 13:1477-1489 Khoury SJ, Guttmann CRG, Orav EJ, et al. Longitudinal MRI in multiple sclerosis: correlationbetween disability and lesion burden. Neurology 1994;44:2120-2124 Kinrichs DJ, Wegmann KW, Dietsch GN. Transfer of experimental allergic encephalomyelitis to bone-chimeras: endothelial cells are not the restricting element. J Exp Med 1987; 166:1906-1911. Koopmans RA, Li DKB, Oger JJF, Kastrukoff LF, Jardine C, Costley L, Hall S, Grochowski EW, Paty DW. Chronic progressive multiple sclerosis: serial magnetic resonance brain imaging over six months. Ann Neurol 1989a;26:248-256 Koopmans RA, Li DKB, Oger JJF, Mayo J, Paty DW. The lesion of multiple sclerosis: imaging of acute and chronic stages. Neurology 1989b;39:959-963 Koopmans RA, Zhu G, Li DKB, Allen PA, Javidan M, Paty DW. In-vivo proton MRS of the acute and chronic lesion of multiple sclerosis. In: Book of abstracts. Society of Magnetic Resonance in Medicine 1990;2:1205 Koopmans, RA, Li DKB, Redekop WK, Zhao GJ, Palmer MR, Kastrukoff LF, Paty DW. The use of magnetic resonance imaging in monitoring interferon therapy of multiple sclerosis. JNeuroimag 1993;3:163-168 Koopmans RA, Li DKB, Zhao GJ, Redekop Wk, Paty DW. MRI assessment of cyclosporine therapy of MS in a multicenter trial. Neurology 1992;42(suppl 3):210 Koopmans RA, Zhao GJ, Li DKB, Paty DW. Interferon b-lb decreases multiple sclerosis lesion activity as detected by yearly MRI. Neurology (in submitted) 116 Kurtzke JF. Rating neurological impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444-1452 Lassmann H, Waksman B, Brosnan C. Mechanism of vascular and tissue damage in demyelinating diseases. J Neuroimmunol 1991; 32:83-85. Lauterbur PC. Image formation by induced local interaction: examples employing nuclear magnetic resonance. Nature 1973;242:190-191 Li D, Mayo J, Fache S, Robertson WD, Paty, D, Genton M. Early experience in nuclear resonance imaging of multiple sclerosis. Ann NY Acad Sci 1984;436:483-486 Lossinsky AS, et al. Sites of egress of inflammatory cells and horseradish peroxidase transport across the blood-brain barrier in a murine model of chronic relapsing allergic encephalomyelitis. Acta Neuropathol 1989; 78:359-371 Lynch SG, Rose JW, Smoker W, et al. MRI in familial multiple sclerosis. Neurology 1990;40:900-903 MacKay RP, Myrianthopoulos NC. Multiple sclerosis in twins and their relatives. Final report. Arch Neurol Psychi 1966;15:449-462 Mansfield P, Grannell PK. NMR "diffraction" in solids? J Phys 1973;[C]6:L422-L426 Matthews WB, Acheson ED, Batchelor JR, Weller RO. McAlpine's multiple sclerosis. New York: Churchill Livingstone, 1985. Matthews WB. Pathophysiology. In: Matthews WB, ed. McAlpine's multiple sclerosis. 2nd ed. New York: Churchill-Livingstone, 1991:245-246 Miller BL. A review of chemical issues in 'H NMR spectroscopy: N-acetyl-1-aspartate, creatine, choline. NMR Biomed 1991;4:47-52 Miller DH, Rudge P, Johnson G, Kendall BE, Macanus DG, Moseley IF, Barnes D, McDonald WI. Serial gadolinium enhanced magnetic resonance imaging in multiple sclerosis. Brain 1988; 111:927-939 Miller DH, Rudge P, Johnson G, et al. Serial gadolinium enhanced magnetic resonance imaging in multiple sclerosis. Brain 1989; 111:927-939 117 Miller DH, Barkhof F, Berry L, et al. Magnetic resonance imaging in monitoring the treatment of multiple sclerosis: concerted action guidelines. J Neurol Neurosurg Psych 1991;54:683-688 Miller DH, Barkhof F, Nauta JJP. Gadolinium enhancement increases the sensitivity of MRI in detecting disease activity in multiple sclerosis. Brain 1993; 116:1077-1094 Minden SL, Schiffer RB. Affective disorder in multiple sclerosis: review and recommendations for clinical research. Arch Neurol 1990;47:98-104 Mumford CJ, Wood NW, Kellar-Wood H, et al.British isles survey of multiple sclerosis in twins. Neurology 1994;44:11-15 Niendorf HP, Dinger JC, Haustein J, et al. Tilerance data of Gd-DTPA: a review. Eur J Radiol 1991;13:15-20 Ormerod IEC, Miller DH, du Boulay EPGH, McDonald WI. Magnetic resonance imaging in multiple sclerosis. Thieme Med Publisher, New York 1989 Osborn AG, Hendrick RE. Introduction to MR imaging, part I. basic physics and instrumentation, in Daignostic Radiology. MR Imaging. Ed. Stark DD and Bredley WG. 1988;7-25 Panitch HS. Interferon in multiple sclerosis: a review of the evidence. Drugs 1992;44:946-962 Pardridge WM, et al. Blood-brain barrier: interface between internal medicine and the brain. Ann Inter Med 1986; 105:82-95. Paty DW. Multiple sclerosis: assessment of disease progression and effects of treatment. Can J Neurol Sci 1987;14:518-520 Paty DW. Magnetic resonance imaging in the assessment of disease activity in multiple sclerosis. Can J Neurol Sci 1988; 15:266-172 Paty DW, Li DKB, The UBC MS/MRI Study Group, The IFNB Multiple Sclerosis Study Group. Interferon beta-lb is effective in relapsing-remitting multiple sclerosis. II. 118 M R I analysis results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 1993a;43:662-667 Paty D W , Koopmans R A , Zhao G J , L i D K B , Oger JJF, Petkau J. Magnetic resonance imaging (MRI) as an outcome measure in multiple sclerosis. J Neurol Rehab 1993b;7:117-129 Plant G T , Kermode A G , Turano G , et al Symptomatic retrochiasmal lesions in multiple sclerosis: clinical features, visual evoked potentials, and magnetic resonance imaging. Neurology 1992;42:68-76 Pollay M and Robert P A . Blood-brain barrier: a definition of normal and altered function. Neurosurg 1980; 6:675-685. Poser C M , et al. Neuroimaging and the lesion of multiple sclerosis. A J N R 1987; 8:549-552. Prineas JW. The neuropathology of multiple sclerosis. In Handbook of Clinical Neurology. Koetsier JC. Ed . Elsevier Science Pub. 1985;3(47):213257 Prineas JW, Kwon E E , Goldenberg P Z , et al. Multiple sclerosis: oligodendrocyte proliferation and differentiation in fresh lesions. Lab Invest 1989;61:489-503 Prineas JW, Barnard R O , Revesz T, et al. Multiple sclerosis: pathology of recurrent lesions. Brain 1993;116:681-693 Prineas JW. Multiple sclerosis: pathology of the early lesion. In Multiple Sclerosis: Current Status of Research and Treatment. Ed . Herndon R M and Seil FJ . Demos. New York 1994:113-130 Purcell E M , Torrey H C , Pound R V . Resonance absorption by nuclear magnetic moments in a solid. Phys Rev 1946;69:37-38 Raine CS , Cross A H . Axonal dystrophy as a consequence of long-term demyelination. Lab Invest 1989;60:714-725 Raine C S . Demyelinating disease, in Textbook of Neuropathology. Ed i by Davis R L , Robertson D M . Will iams & Wilkins 1991 119 Reese TS, Karnovsky MJ. Fine structural localization of a blood-brain narrier to exogenous peroxidase. J Cell Biol 1967; 34:207-217. Ron MA, Feinstein A. Multiple sclerosis and the mind. J Neurol Neurosurg Psychia 1992;55:1-3 Runge VM, et al. The use of Gd DTPA as a perfusion agent and marker of blood-brain barrier disruption. Mag Res Imaging 1985; 3:43-55. Sadovnick AD, Armstrong H, Rice GPA, et al. A population-based study of multiple sclerosis in twins: update. Ann Neurol 1993;33:281-185 Sage MR and Wilson AJ. The blood-brain barrier: an important concept in neuroimaging AJNR 1994;15:601-622 Sharma HS and Dey PK. Influence of long-term acute heat exposure on regional blood-brain barrier permeability, cerebral blood flow and 5-HT level in concious normotensive young rats. Brain Res 1987; 424:153-162. Schima W, Wimberger D, Stiglbauer R, et al. The importance of magnetic field strength in the MR diagnosis of multiple sclerosis: a comparison of 0.5 and 1.5T. Rofo. Fortschritte Auf Dem Gebiete Der Rontgenstrahlen Und Der Neuen Bildgebenden Verfahren 1993; 158(4):368-171 Sibley WA, Paty DW. A comparison of multiple sclerosis in Ontario, Canada and Arizona, USA. Acta Neurol Scand 1981;64(suppl 87):60 Sipe JC, Knobler RL, Braheny SI, et al. A neurologic rating scale (NRS) for use in multiple sclerosis. Neurology 1984;34:1368-1372 Smith ME, Stone LA, Albert PS, et al. Clinical worsening in multiple sclerosis is associated with increased frequency and area of gadopentetate dimeglumine-enhancing magnetic resonance imaging lesions. Ann Neurol 1993;33:480-489 Stenager E, Knudsen L, Jensen K. Psychiatric and cognitive aspects of multiple sclerosis. Semi in Neurol 1990;10:254-261 120 Steward PA, Wiley MJ. Developing nervous tissue induces formation of blood-brain barrier characteristics in invading endothelial cells: A study using quail-chick transplantation chimeras. Devi Biol 1981; 84:183-192. Stewart WA, Hall LD, Berry K, et al. Magnetic resonance imaging (MRI) in multiple sclerosis (MS): pathological correlation studies in eight cases. Neurology 1986;36(suppl 1):320 Sweeney VP, Sadovnick AD, Brandejs V. Prevaence of multiple sclerosis in British Columbia. Can J Neurol Sci 1986;13:47-51 Tanaka Y, et al. Anti-endothelial cell antibodies and circulating immune complexes in the sera of patients with multiple sclerosis. J Neuroimmunol 1987; 17:49-59. Thompson AJ, Kermode A G , MacManus DG, et al. Patterns of disease activity in multiple sclerosis: clinical and magnetic resonance imaging study. Bri Med J 1990;300:631-634 Thompson AJ, Miller D, Youl B, MacManus D, Moore S, et al. Serial gadolinium-enhanced MRI in relapsing/remitting multiple sclerosis of varying disease duration. Neurology 1992;42:60-63 Tienari PJ, Salonen O, Wikstrom J, et al. Familial multiple sclerosis: MRI findings in clinically affected and unaffected siblings. J Neurol Neurosurg Psychiatry 1992;55:883-886 Tourtellotte WW and Ma BI. Multiple sclerosis: the blood-brain barrier and the measurement of de novo central nervous system IgG synthesis. Neurology 1978; 28:76-83. Traugott U, and Lebon P. Mulitple sclerosis: involvement of interferons in lesion pathogenesis. Ann Neurol 1988; 24:243-251. 121 Tsukada N, et al. Autoantibodies to each protein fraction extracted from cerebral endothelial cell membrane in the sera of patients with multiple sclerosis. J Neuroimmunol 1989; 24:41-46. Tunggal B, Hofmann K, Stoffel W. In vivo 13C nuclear magnetic resonance investigations of choline metabolism in rabbit brain. Magn Reson Med 1990;13:90-102 van Hecke P, Marchal G, Johannik K, et al. Human brain proton localized NMR spectroscopy in multiple sclerosis. Magn Reson Med 1991; 18:199-206 Vukmanovic S, et al. Experimental autoimmune encephalomyelitis in "low" and "high" interleukin 2 producer rats. I cellular basis of induction. Cell Immunol 1989; 121:237-246. Waxman SG. Membranes, myelin, and the pathophysiology of multiple sclerosis. N Engl JMed 1982;306:1529-1533 Weinmann HJ, Brasch RC, Press WR, Wesbey GE. Characteristics of gadolinium-DTPA -complex: a potential NMR contrast agent. AJR 1984; 142:619-624 Weinmann HJ, Press WR, Raduchel B, et al. Characteristics of Gd-DTPA and new derivates. In: Bydder GM, Felix R, Bucheler E, et al. (Eds). Contrast media in MRI. Bussum: Medicom Europe 1990:9-14 Wekerle H. Intercellular interaction in myelin-specific autoimmunity. J Neuroimmunol 1988; 20:211-216. Westergard E. The effect of serotonin on the blood-brain barrier to protein. J Neurol Trans 1978; Suppll4:9-15. 122 Wiebe S, Lee DH, Karlik SJ, et al. Serial cranial ans spinal cord magnetic resonance imaging in multiple sclerosis. Ann Neurol 1992;32:643-650 Wilcox CE, Ward AMV, Evans A, et al. Endothelial cell expression of the intercellular adhesion molecule-1 (ICAM-1) in the central nervous system of guinea pigs during acute and chronic relapsing experimental allergic encephalomyelitis. J Neuroimmunol 1990; 30:43-51. Williams A, Eldridge R, MacFarland H, Houff S, et al. Multiple sclerosis in twins. Neurology 1980;30:1139-1147 Willoughby EW, Grochowski E, Li DKB, Oger J, Kastrucoff LF, Paty DW. Serial magnetic resonance scanning in multiple sclerosis: a second prospective study in relapsing patients. Ann Neurol 1989;25:43-49 Wolinsky JS, Narayana PA, Fenstermacher MJ. Proton magnetic resonance spectroscopy in multiple sclerosis. Neurology 1990;40:1764-1769 Wong D, Dorovini-Zis K. Upregulation of intercellular adhesion molecule-1 (ICAM-1) expression in primary cultures of human brain microvessel endothelial cells by cytokines and lipopolysaccharide. J Neuroimmunol 1992;39:11-21 Youl BD, Turano G, Miller DH, et al. The pathophysiology of acute optic neuritis. An association of gadolinium leakage with clinical and electrophysiological deficits. Brain 1991;114:2437-2450 Young IR, Hall AS, Pallis CA, Legg NJ, Bydder GM, Steiner RE. Nuclear magnetic resonance imaging of the brain in multiple sclerosis. Lancet 1981 ;ii: 1063-1066 123 Zhao GJ, Li DKB, Tanton BL, Paty DW. Assessment of activity of the individual lesions in relapsing-remitting multiple sclerosis on magnetic resonance imaging. Ann Neurol 1991;30:270 Zhao GJ, Li DKB, Tanton BL, Paty DW. Sensitivity of Gd-DTPA MRI and serial MRI in detecting activity of lesion in relapsing-remitting multiple sclerosis. Book of abstracts. Society of MR in Med. 1992; 1:1710 Zhao GJ, Li DKB, Tanton BL, Paty DW. Features of enhancing and non-enhancing lesions in multiple sclerosis on Gd-DTPA MRI. J Neuroimaging 1992;2:52 Zhao GJ, Li DKB, Koopmans RA, Tanton BL, Paty DW. The growth pattern of enlarging lesions in multiple sclerosis: Observation on serial MRI. Can J. Neuro Sci 1993;Suppl.4:S219 Zhao GJ, Koopmans RA, Li DKB, et al. Corticospinal tract lesions in multiple sclerosis: relationship between MRI activity and clinical course. Neurology 1993;43:A246 Zhao GJ, Li DKB, Koopmans RA, Paty DW, MS/MRI Research Group, MS Study Group. Changes in burden of disease and lesion activity in MS placebo patients: a 5-year study by yearly serial MRI. Ann Neurol 1995 (in press) 124 Appendix A. Expanded Disability Status Scale (EDSS) 0 = Normal neurologic exam (all grade 0 in Functional Systems (FS, appendix B); Cerebral grade 1 acceptable). 1.0 = No disability, minimal signs in more than one FS (more than one grade 1 excluding Cerebral grade 1). 2.0 = Minimal disability in one FS (one FS grade 2, others 0 or 1). 2.5 = Minimal disability in two FS (two FS grade 2, others 0 or 10. 3.0 = Moderate disability in one FS (one FS grade 3, others 0 or 1), or mild disability in three or four FS (three/four FS grade 2, others 0 or 1) though fully ambulatory. 3.5 = Fully ambulatory but with moderate disability in one FS (one grade 3) and one or two FS grade 2; or two FS grade 3; or five FS grade 2 (others 0 or 1). 4.0 = Fully ambulatory without aid, self-sufficient, up and about some 12 hours a day despite relatively severe disability consisting of one FS grade 4 (others 0 or 1), or combinations of lesser grades exceeding limits of previous steps. Able to walk without aid or rest some 500 meters. 4.5 = Fully ambulatory without aid, up and about much of the day, able to work a full day, may otherwise have some limitation of full activity or require minimal assistance; characterized by relatively severe disability, usually consisting of one FS 125 grade 4 (others 0 or 1) or combinations of lesser grades exceeding limits of previous steps. Able to walk without aid or rest for some 300 meters. 5.0 = Ambulatory without aid or rest for about 200 meters; disability severe enough to impair full daily activities (eg, to work full day without special provisions). (Usual FS equivalents are one grade 5 alone, others 0 or 1; or combinations of lesser grades usually exceeding specifications for step 4.0). 5.5 = Ambulatory without aid or rest for about 100 meters; disability severe enough to preclude full daily activities. (Usual FS equivalents are one grade 5 alone, other 0 or 1; or combinations of lesser grades usually exceeding those for step 4.0). 6.0 = Intermittent or unilateral constant assistance (cane, Crutch, or brace) required to walk about 100 meters with or without resting, (usual FS equivalents are combinations with more than two FS grade 3+). 6.5 = Constant bilateral assistance (canes, crutches, or braces) required to walk about 20 meters without resting. (Usual FS equivalents are combinations with more than two FS grade 3+). 7.0 = Unable to walk beyond about 5 meters even with aid, essentially restricted to wheelchair; wheels self in standard wheelchair and transfers alone; up and about in w/c some 12 hours a day. (Usual FS equivalents are combinations with more than one FS grade 4+; very rarely, pyramidal grade 5 alone). 7.5 = Unable to take more than a few steps; restricted to wheelchair; may need aid in transfer; wheels self but cannot carry on in standard wheelchair a full day; may 1 2 6 require motorized wheelchair. (Usual FS equivalents are combinations with more than one FS grade 4+). 8.0 = Essentially restricted to bed or chair or perambulated in wheelchair, but may be out of bed itself much of the day; retains many self-care functions; generally has effective use of arms. (Usual FS equivalents are combinations generally grade 4+ in several systems). 8.5 = Essentially restricted to bed much of the day; has some effective use of arm(s); retains some self-care functions. (Usual FS equivalents are combinations generally 4+ in several systems). 9.0 = Helpless bed patient; can communicate and eat. (Usual FS equivalents are combinations, mostly grade 4+). 9.5 = Totally helpless bed patient; unable to communicate effectively or eat/swallow. (Usual FS equivalents are combinations, almost all grade 4+). 10. = Death due to MS. 127 Appendix B. Functional Systems Pyramidal Functions 0. Normal. 1. Abnormal signs without disability. 2. Minimal disability. 3. Mild or moderate peraparesis or hemiparesis; severe monoparesis. 4. Marked paraparesis or hemiparesis; moderate quadriparesis; or monoplegia. 5. Paraplegia, hemiplegia, or marked quadriparesis. 6. Quadriplegia. V . Unknown. Cerebellar Functions 0. Normal. 1 . Abnormal signs without disability. 2. Mi ld ataxia. 3. Moderate truncal or limb ataxia. 4. Severe ataxia, all limbs. 5. Unable to perform coordinated movements due to ataxia. V. Unknown. X . Is used throughout after each number when weakness (grade 3 or more on pyramidal) interferes with testing. Brain Stem Functions 0. Normal. 1. Signs only. 128 2. Moderate nystagmus or other mild disability. 3. Severe nystagmus, marked extraocular weakness, or moderate disability of other cranial nerves. 4. Marked dysarthria or marked disability. 5. Inability to swallow or speak. V. Unknown. Sensory Functions 0. Normal. 1. Vibration or figure-writing decrease only, in one or two limbs. 2. Mild decrease in touch or pain or position sense, and/or moderate decrease in vibration in one or two limbs; or vibratory (c/s figure writing) decrease alone in three or four limbs. 3. Moderate decrease in touch or pain or position sense, and/or essentially lost vibration in one or two limbs; or mild decrease in touch or pain and/or moderate decrease in all proprioceptive tests in three or four limbs. 4. Marked decrease in touch or pain or loss of proprioception, alone or combined, in one or two limbs; or moderate decrease in touch or pain and/or severe proprioceptive decrease in more than two imbs. 5. Loss (essentially) of sensation in one or two limbs; or moderate decrease in touch or pain and/or loss of proprioception for most of the body below the head. 6. Sensation essentially lost below the head. V. Unknown. Bowel and Bladder Function 0. Normal. 1. Mild urinary hesitancy, urgency, or retention. 129 2. Moderate hesitancy, urgency, retention of bowel or bladder, or rare urinary incontinence. 3. Frequent urinary incontinence. 4. In need of almost constant catheterization. 5. Loss of bladder function. 6. Loss of bowel and bladder function. V. Unknown. Visual (or Optic) Functions 0. Normal. 1. Scotoma with visual acuity (corrected) better than 20/30. 2. Worse eye with scotoma with maximal visual acuity (corrected) of 20/30 to 20/59. 3. Worse eye with large scotoma, or moderate decrease in fields, but with maximal visual acuity (corrected) of 20/60 to 20/99. 4. Worse eye with marked decrease of fields and maximal visual acuity (corrected) of 20/100 to 20/200; grade 3 plus maximal acuity of better eye of 20/60 or less. 5. Worse eye with maximal visual acuity (corrected) less than 20/200; grade 4 plus maximal acuity of better eye of 20/60 or less. 6. Grade 5 plus maximal visual acuity of better eye of 20/60 or less. V. Unknown. X. Is added to grades 0 to 6 for presence of temporal pallor. Cerebral (or Mental) Functions 0. Normal. 1. Mood alteration only (Does not affect DSS score). 2. Mild decrease in mentation. 3. Moderate decrease in mentation. 130 4. Marked decrease in mentation (chronic brain syndrome - moderate). 5. Dementia or chronic brain syndrome-severe or incompetent. V. Unknown. Other Functions 0. None. 1. Any other neurologic findings attributed to MS (specify). V. Unknown. 131 

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.831.1-0087162/manifest

Comment

Related Items