The Pathology of Multiple Sclerosis and Its Variants

iseases of myelin comprise a varietyof illnesses with varied etiology.They range from clearly infectiousdiseases such as progressive multi-

focal leucoencephalopathy to hereditary metabolic dis-orders such as adrenoleucodystrophy and metachro-matic leucodystrophy to a group of disorders ofunknown etiology (Table 14.1). Whether multiple scle-rosis (MS) and its variants comprise a single disease witha highly varied course or a group of syndromes of diverseetiology remains controversial, although recent evidencesuggests that it actually involves at least four differentpathologic variants (1). Numerous clinical demyelinat-ing disorders may be part of the MS spectrum or mayrepresent different diseases with similar clinical andpathologic features. For years, many clinicians have feltthat primary progressive MS and Devic syndrome (neu-romyelitis optica) were distinct from relapsing remittingand secondary progressive MS. Balo concentric sclero-sis was felt to be a different disease, but recent evidencehas suggested that it is a variant of relapsing remittingMS. Whether the Marburg variant and acute dissemi-nated encephalomyelitis (ADEM) are part of the MSspectrum or represent different disease entities remainsa matter of debate. Recent immunopathologic studiesare beginning to clarify these issues, but, ultimately, clar-ification of the relationship of the variants to each other

will have to await determination of the etiology of theseconditions.

Recent developments in immunopathology havesuggested that MS is several different diseases with a com-mon clinical presentation (1). These immunopathologi-cally different disorders are likely to have different eti-ologies. Guillian-Barre syndrome has been shown to haveseveral variants caused by different agents includingCampylobacter jejuni, cytomegalovirus, Epstein-Barrvirus, Mycoplasma pneumonia, swine flu vaccine, andprobably several others (2–4). Typical relapsing remittingMS now appears to have several variants, and these arelikely to have different etiologies. The inflammatorydemyelination in different cases has immunologic featuresthat are sufficiently different from each other to suggestthat this is really a group of diseases (1).

These demyelinating disorders appear to have cen-tral myelin and the cells that form central myelin, theoligodendrocytes, as the target of attack. MS usually doesnot affect peripheral myelin, although there are cases ofa disease clinically and pathologically indistinguishablefrom MS in which there is also peripheral demyelina-tion in a pattern indistinguishable from chronic inflam-matory demyelinating polyneuropathy (5–7). Becausethere is good evidence that molecular mimicry is involvedin Guillian-Barre syndrome (3), molecular mimicry of ashared epitope could cause this rare combination of cen-

185

The Pathology of MultipleSclerosis and Its Variants

Robert M. Herndon, M.D.

14

D

tral and peripheral demyelination and lends further cre-dence to the idea that it is an autoimmune disease. Theincreased risk of MS in first-order relatives of MSpatients, the higher concordance rate in identical versusfraternal twins, and the overrepresentation of certainHLA types (8) point to significant genetic risk factors,although it is clear that no single gene is responsible forgenetic susceptibility.

ETIOLOGY

The etiology of MS remains unknown. The most widelyaccepted hypothesis is that it is an autoimmune diseaseinduced by a virus or other infectious agent. The possi-bility of it being a primary infectious process with or with-out an associated autoimmune reaction has not beenentirely ruled out despite repeated failure to identify acausative agent. Over the past four decades, there havebeen a number of reports of viruslike particles in brain tis-sue from MS patients (9). At least 14 different viruses have

been isolated from the brains of MS patients, yet none hasbeen shown to be etiologically related (10). Cook listed22 agents suspected of being related to MS for which sub-stantial evidence of a causative role has not thus farappeared (11). The demonstration of viral nucleic acidby polymerase chain reaction in MS brain and cere-brospinal fluid (CSF) (12,13) has created some interest,but it has never been possible to clearly demonstrate a con-nection between any of these agents and the diseaseprocess. No agent has been found in 100 percent of activeplaques. The demonstration of herpes viruses, in particu-lar herpes I, II, and VI, in a significant proportion of MSplaques over the past decade (12,13), evidence that theanti-herpes drug acyclovir will reduce the number ofattacks of MS (14), and, more recently, evidence forchlamydial infection (15) have renewed interest in infec-tious hypotheses. Because herpes viruses are activated byother infections and viral infections are known to precip-itate some MS attacks, reactivation of herpes viruses maybe related to some attacks or may be significant aggra-vating factor even if not causative. Only time will tell ifany of these agents plays a role in disease causation. Thereport of a patient with MS developing acute optic neu-ritis two months after an allogenic bone marrow trans-plant for chronic myelogenous leukemia suggests that itis not purely an autoimmune disease (16). The possiblerole of infectious agents in MS is discussed in more detailin Chapters 7 and 8. Part of the problem in determiningthe role of infectious agents in MS may be multiple cau-sation, as occurs in Guillian-Barre syndrome, in which sev-eral different agents can induce acute autoimmune neu-ropathies that are difficult to differentiate clinically (2).

Extensive effort has gone into the attempt to under-stand the role of the immune system in MS. Much of thiseffort has been directed toward understanding theimmunology of experimental autoimmune encephalo-myelitis (EAE), which is the most extensively studied ani-mal model of the disease (see Chapter 6). This work hastaught us an enormous amount regarding immunologyin general and immunology as it applies to the humannervous system. Unfortunately, attempts to apply to MSwhat has been learned about EAE have rarely been suc-cessful. Numerous treatments that work well in EAE havefailed completely when tried in MS, making it clear thatMS is not simply EAE. However, there is good evidencethat autoimmunity is involved in the disease, and immuno-modulating agents such as the � interferons and glatirameracetate and immunosuppressant drugs such as cyclophos-phamide, methotrexate, and mitoxantrone can slow thedisease process. With the development of a more detailedunderstanding of immunology and the ability to directlyassess various immunologic components in situ, it is pos-sible to begin to define the immune reaction in the acuteplaque of MS and to begin to understand the nature of theimmune reaction in MS, as discussed below.

PATHOLOGY186

TABLE 14.1Demyelinating Diseases

Demyelination due to infectious agents (see Chapter 9)Progressive multifocal leucoencephalopathySubacute sclerosing panencephalitis

Metabolic disorders of myelinMetachromatic leucodystrophyGloboid cell leucodystrophyPelizaeus-Merzbacher disease

Peroxisomal disorders affecting myelinGeneralized peroxisomal disorders

Cerebrohepatorenal (Zellweger) syndromeInfantile Refsum disease (phytanic acid storage)Neonatal adrenoleucodystrophyHyperpipecolic acidemia

Single peroxisomal enzyme deficiencies with wide-spread pathology

Thiolase deficiencyAcyl-CoA oxidase deficiencyRhizomelic chondrodysplasia

Single peroxisomal enzyme deficiencies with more restricted pathology

Adrenoleukodystrophy complexDiseases of unknown etiologyMultiple sclerosis

Relapsing remitting/secondary progressive multiplesclerosisMarburg variantNeuromyelitis optica (Devic disease)Primary progressive multiple sclerosis

Acute disseminated encephalomyelitisFibrinoid leucodystrophy (Alexander disease)

CLINICAL SPECTRUM OF HUMANINFLAMMATORY DEMYELINATING DISEASES

The entities described here have at different times beenconsidered distinct diseases and, largely because of over-lap syndromes, been considered part of the MS spectrum.Recent advances in pathology are clarifying the relation-ships, but definitive answers to the relationship of thesediseases remain to be established.

Relapsing Remitting Multiple Sclerosis

This is the classic MS clinical pattern marked by exac-erbations, with a variable amount of improvementbetween attacks

Secondary Progressive Multiple Sclerosis

About 80 percent of relapsing remitting MS cases go onto develop a secondary progressive disease pattern, witha slowly progressive ascending paralysis from a few to20 or so years after onset. This represents a later stage ofrelapsing remitting disease rather than a separate variant,although not all relapsing remitting MS goes on todevelop secondary progression.

Primary Progressive Multiple Sclerosis

This disease pattern has a later onset, usually after age40 years, and begins with an insidious progression of dis-ability affecting primarily the spinal cord without exac-erbations or remissions. Unlike exacerbating remittingdisease, in which two-thirds of the cases are female, pri-mary progressive MS is only slightly more common infemales with a ratio of about 1.3:1 (17). Magnetic reso-nance imaging (MRI) of the brain is sometimes normalin these cases, and MRI of the spinal cord may show onlycord atrophy.

Devic Syndrome (Neuromyelitis Optica)

Devic syndrome is an acute disorder in which optic neu-ritis and transverse myelitis occur within a short time ofeach other, with little or no involvement of other partsof the central nervous system (CNS) (18). It has gener-ally been considered a monophasic disease withoutrelapses after the initial episodes; however, numerous casesthat begin with optic neuritis and transverse myelitis pro-ceed to develop a relapsing remitting course similar to typ-ical relapsing remitting MS, but with more severe residuafrom attacks and a more necrotic pathology. MRI in thesecases shows no lesions within the brain itself but gener-ally shows evidence of cord inflammation usually extend-ing over three or more segments. Whether this is a variantof MS or a distinct disease remains controversial. It may

be a distinct disease entity, or the course and intensity dif-ferences may relate more to the genetic makeup of the hostthan to the etiology.

Marburg Disease (Acute Multiple Sclerosis)

This acute, fulminant demyelinating disorder was firstdescribed by Otto Marburg in 1906 (19). It is a severe,unrelentingly progressive demyelinating disease that typ-ically leads to death within a few months to a year.

Balo Concentric Sclerosis

This has been thought to be an aggressive variant usu-ally leading to death in weeks to months. It is marked bylarge plaques of demyelination with concentric bands ofpreserved or regenerated myelin. These bands can some-times be seen on MRI. There is now good evidence thatsome cases with Balo-like lesions on MRI improve andhave a course typical of relapsing remitting MS. It may bethat the Balo type lesion is a feature of some early MSwith very active remyelination, features that disappearas the disease progresses so that the concentric lesions areobserved only pathologically in individuals who die soonafter the onset of the disease.

Acute Disseminated Encephalomyelitis

This is usually a monophasic demyelinating disorder thattypically follows a viral infection. It appears to be closelyanalogous to EAE. The most severe cases are seen aftermeasles infection, smallpox vaccination, or rabies vac-cine. It is marked by the rapid, sometime sudden, onsetof confusion, fever, and depressed consciousness, oftenwith seizures and multiple focal neurologic signs such asataxia, paraplegia, or cranial nerve signs. It can be fatalwithin days to weeks, although survivors often recoverremarkably well over a period of many months.

PATHOLOGY OF RELAPSING REMITTINGMULTIPLE SCLEROSIS

The pathology of typical relapsing remitting MS consistsof lesions (plaques) disseminated in location and varyingin age, as would be expected from the clinical features.In addition, there is a second demyelinating process con-sisting of the demyelination of individual fibers or smallgroups of fibers that is best seen in the spinal cord. Plaquescan be found wherever there is central myelin. They arepresent in white and gray matter, but the gray matterlesions are much less obvious and generally do not appearon MRI, possibly because of the relatively small amountof myelin present and a less intense inflammatory reaction.The lesions occur in different parts of the nervous system

THE PATHOLOGY OF MULTIPLE SCLEROSIS AND ITS VARIANTS 187

and are in different stages of activity or maturity. Lesionsrange from acute plaques with active inflammatory infil-trates and macrophages loaded with lipid and myelindegeneration products, through various degrees of lesseractivity to plaques that are active only at their margin, tochronic, inactive demyelinated shrunken glial scars.Although plaques can appear anywhere there is centralmyelin, there is a predilection for the periventricular whitematter, optic nerves, spinal cord, and juxtacortical areas.

Areas of white matter outside the plaques are notnormal. They show biochemical abnormalities that somefeel can be explained on the basis of Wallerian degener-ation (18) and gliosis, whereas others believe that theyrepresent an important aspect of the pathology, particu-larly in secondary progressive MS. (20). This appears tooccur concurrently with plaque formation and consists ofpatchy demyelination in one or a few fibers with mini-mal inflammation. This is best seen in the spinal cord.

Recent reports by Lucchinetti et al. (1) and Lass-man et al. (21) have dramatically altered our view of thepathology of MS. They described four different patternsof demyelination in active MS lesions based on a vari-ety of modern staining and labeling techniques, sug-gesting that MS is a group of diseases with similar clin-ical and pathologic features rather than a single entity.Three of these patterns are seen in exacerbating remit-ting MS, and the fourth has been seen only in primaryprogressive MS. These recent observations already havehad a major impact on our thinking regarding etiologyand pathogenesis. It seems likely that the different pat-terns relate, at least in part, to different etiologies andmay provide the basis for associating particular agentswith particular disease patterns. They also explain someof the discrepancies found in the description of variousinvestigators. The first two patterns appear to beautoimmune reactions against myelin, whereas the thirdand fourth patterns more closely resemble a disease pri-marily of oligodendrocytes. Pattern four appears to beassociated exclusively with primary progressive MS,although it is likely that some cases that appear clini-cally to be primary progressive will have one of theother pathologies, because many cases of relapsingremitting MS are essentially silent clinically, and someapparently primary progressive cases are likely actu-ally secondary progressive (22). The pathology of thefourth group is discussed below. The patterns can besummarized as follows.

Pattern 1: Active Demyelination Associated withT-Lymphocyte and Macrophage-Dominated

Inflammation without Significant Amounts ofAntibody or Complement Deposition

Plaques were perivenular in location, and loss of all typesof myelin protein appeared to occur simultaneously. There

was some diffuse immunoglobulin G (IgG) stainingthroughout the lesion. Staining for complement compo-nents was negative. This pattern comprised about 12 per-cent of the reported cases.

Pattern 2: Active Perivenular DemyelinationAssociated with T-Lymphocyte and Macrophage-

Dominated Inflammation with Extensive Antibody Deposition in the Tissue and

in Astrocyte Cytoplasm

Complement C9 neoantigen was deposited at sites ofactive demyelination, indicating that the antibody playsa role in the demyelinating process. Plaque borders werewell defined, and lymphocytic perivascular cuffs werefrequent. Loss of all types of myelin protein appearedto occur simultaneously. There was heavy staining ofmyelin degradation products in the macrophages. Inolder, inactive plaques, there was variable loss of oligo-dendrocytes at the plaque borders, with reappearance ofoligodendrocytes in the plaque centers. A high incidenceof shadow plaques was seen in these cases, indicatingthat some remyelination is common. This was the mostcommon pattern, comprising 53 percent of the reportedcases. The formation of clathrin coated pits (caveolae)seen on electron microscopy reported by Prineas andConnell is consistent with a role for antibody in thedemyelination; thus, these two cases were most likelytype II (Figure 14.1) (23).

Pattern 3: Active Demyelination with anInfiltrate of T Lymphocytes and Activated

Macrophages and Microglia

No IgG deposition, possible preservation of a rim ofmyelin around venules, and the plaque borders were illdefined. With this lesion pattern, there was preferentialloss of myelin-associated glycoprotein (MAG) relativeto other myelin proteins such as MBP and proteolipidprotein (PLP) as originally described by Itoyama et al.(24). A concentric (Balo type) pattern was seen in someof these cases. This pattern comprised about 30 percentof the cases.

Forty-three of the lesions examined by Lucchinettiand others were biopsies of initial clinical lesions. Thirty-three of these patients developed clinically definite MSin the follow-up period, including individuals with eachof the first three disease patterns. Multiple lesions wereexamined from most of the cases, and the pattern was thesame in all of the lesions from a given patient (1). Con-version of one pattern to another would seem possible,but, except for conversion of pattern 1 to pattern 2 withthe development of an antibody reaction, conversionbetween types seemed a priori unlikely.

PATHOLOGY188

THE ACUTE PLAQUE

According to Prineas (25), most MS plaques appear tobegin with margination and diapedesis of lymphocytesand macrophages forming perivascular cuffs about cap-illaries and venules. This is followed by diffuse parenchy-mal infiltration by inflammatory cells, edema,macrophage activity with stripping of myelin fromaxons, astrocytic hyperplasia, and the appearance ofincreased numbers of lipid-laden macrophages anddemyelinated axons. Prineas also described a markedincrease in the number of plasma cells in some cases,which would be consistent with a type 2 case. In con-trast, Lumsden described plaques in which myelin frag-

mentation preceded the appearance of macrophages.This would seem to correspond to a type 4 case ofLucchinetti et al. In type 2 cases, as plaques enlarge andcoalesce, the initial perivenular distribution of the lesionsbecomes less apparent. The inflammatory reaction in theplaques usually is less pronounced in gray matter, prob-ably because of the smaller amount of myelin in theseareas and the correspondingly fewer macrophagesneeded to remove cellular debris.

Acute plaques begin in a perivenular location(25,26), and are marked by perivascular cuffs (Figure14.2 and 14.3) of inflammatory cells including lym-


FIGURE 14.1

Electron micrograph showing a clathrin-coated pit on the sur-face of a macrophage (lower right center) and a remyelinat-ing axon. Original magnification, 39000X. Reprinted from:Neurology 1978;28:S68–S75 (Figure 9), with permission.

FIGURE 14.2

Hematoxylin and eosin stain of subacute multiple sclerosisplaque in the subcortical white matter, with a few perivascu-lar inflammatory cells. At the top, the plaque extends intothe cortical gray matter. The plaque edge stains more heavilywith hematoxylin due to the concentration of inflammatorycells.

FIGURE 14.3

Perivascular cuff of lymphocytes and inflammatory infiltratein an acute plaque. Hematoxylin and eosin stain.

phocytes, activated macrophages, microglia, and occa-sionally plasma cells. Polymorphonuclear leukocytesand eosinophils are rare except in Devic syndrome, inwhich they are much more common. Myelin breakdownoccurs with stripping of myelin from the axons bymacrophages, and many of the macrophages are filledwith myelin debris and neutral lipid. Extracellularmyelin debris is rarely, if ever, seen, but some of themyelin breakdown is extracellular, because recogniza-ble myelin fragments are frequently seen on electronmicroscopy of CSF sediment taken at the time of anacute attack (27,28).

CHRONIC ACTIVE AND INACTIVE PLAQUES

As plaques mature, the lipid-laden macrophages andmicroglia exit the central plaque area, which becomeshypocellular (Figure 14.4). Gliosis gradually increases,followed by gradual shrinkage of the plaque. The plaquemargin may remain active with continuing or recurringactivity, or the plaque margin may gradually lose inflam-matory cells and become inactive. Plaque reactivation cancommonly be seen on MRI during acute attacks, in whicha known plaque shows ring enhancement thus indicatingactive inflammation with breakdown of the blood-brainbarrier. The pathologic correlate of this recurring activ-ity is a plaque with a gliotic, sparsely cellular center anda band of lipid-filled macrophages and lymphocytes atthe plaque margin.

Considerable axonal interruption even in new acuteplaques. Axonal interruption is marked by axonal ovoidsthat are easily demonstrated with special stains and con-

focal microscopy. Over time, these interrupted axonsundergo retrograde degeneration and the ovoids disap-pear. Axonal damage is demonstrated on MR spec-troscopy by loss of N-acetyl aspartate (NAA), an axonalmarker (29). Trapp and colleagues reported an averageof about 11,000 axonal ovoids/mm3 in acute plaques,indicating that significant axonal loss begins early in thedisease process (30,31). How much of the NAA loss isdue to axonal interruption and how much is due to lossin axonal diameter and impairment of axonal transportremains to be determined. Demyelinated axons initiallybecome irregular in diameter but in later stages appear tohave a uniformly reduced diameter (Figure 14.4). Theymay lose as much as 50 percent or more of their diame-ter, which would reduce the axoplasmic volume by 75percent, so that, even if the concentration in the axonsremained the same, the amount would be markedlydecreased. At the same time, axon transport is seriouslyimpaired, which also could reduce the amount of NAAout of proportion to the actual axonal loss and might welldecrease the concentration distal to the site of demyeli-nation (see Chapter 16).

Wallerian degeneration due to axonal destruction inacute plaques can be extensive enough to be detected byMRI (32) (see Chapter 15). Additional axonal destruc-tion occurs when older plaques are reactivated withdemyelination at the plaque margins. The fact that spar-ing of axons is relative and not absolute has been knownsince the time of Charcot, but the extent of the damageto the axons has been only recently quantified. Cerebralatrophy and atrophy of the corpus callosum, which iscommonly seen in advanced disease, is probably due toa combination of myelin and axonal loss and shrinkagedue to scarring.

Most investigators now believe that axonal lossaccounts for most of the chronic, irreversible disabilityin MS. Cumulative axonal loss in long pathways such asthe pyramidal tracts and the dorsal columns of the spinalcord, where there may be several plaques along theircourse, is often very substantial. This in part accountsfor the pallor that is frequently seen in these tracts atautopsy. It is important to recognize that axonal loss is animportant factor in MS and that experimental efforts toimprove conduction in surviving fibers will not affectthose symptoms due to axonal loss, even though they mayproduce dramatic improvement in symptoms that are dueto conduction failure in surviving axons.

Immunocytochemical studies of inflammatorycytokines indicate that most of the lymphocytes at theplaque margins are T-cells. CD4� T-cells (helper/inducer)are seen in the plaque margin and extend beyond the mar-gin into periplaque white matter. CD8� T-cells (suppres-sor/cytotoxic) are fewer in number and more confinedto the plaque margin. Occasional B cells and a fewplasma cells may be seen. Macrophages may be seen

PATHOLOGY190

FIGURE 14.4

Inactive plaque center, with the axon stain showing numer-ous demyelinated axons and very little inflammation. Theaxons have a relatively uniform diameter, unlike newlydemyelinated axons, which are irregularly dilated.

throughout the plaque but are rarely seen beyond theplaque margin (33).

GRAY MATTER PLAQUES

Although juxtacortical plaques extending into the adja-cent gray matter have long been recognized, isolated cor-tical plaques are rarely mentioned in the literature. Nev-ertheless, they are an important part of the pathologyof MS. The failure to detect these plaques arises fromtwo factors: they are much less inflammatory than whitematter plaques and cannot be detected with myelin stain-ing techniques. The instructions after staining with luxolfast blue, the most commonly used current myelin stain,is to destain until the gray matter no longer shows anystaining. Thus gray matter plaques do not stain formyelin and are difficult to see with myelin stains. Graymatter involvement is frequently reported when plaquesat the gray–white junction extend into the cortex, butisolated gray matter plaques were rarely discussed orcommented on until the recent report of Peterson et al.(34). They reported a careful analysis of gray matterplaques and classified them into three types: type 1 werecontiguous with subcortical lesions and accounted for34 percent of the cortical lesions; type 2 were small, usu-ally perivascular lesions confined to the cortex, and con-stituting 16 percent of the lesions; and type 3 extendedfrom the cortical surface to layer 3 or 4 of the cortex,constituting 50 percent of the lesions (34). These plaquesare much less inflammatory than those in the white mat-ter, with only one-thirteenth as many CD3� lymphocytesand one-sixth as many microglia and macrophages (34).They found activated microglia apposed to andensheathing neuronal perikarya and apical dendrites.Apoptotic neurons were increased in the cortical plaques.Transected neurites were found with an average densityof 4119/mm3, which is about one-third the numberfound in white matter plaques. They were about one-fourth as frequent in chronic gray matter plaques andmuch less common in inactive lesions. These corticallesions may account for many of the cognitive changesseen in MS.

In chronic active lesions, astrocytes in and aboutthe plaque express class I and class II histocompatibil-ity antigens, although expression of class II antigens wasboth more common, intense, and consistent than expres-sion of class I antigens (33,35). Class I and class II pos-itive astrocytes extended into the periplaque area, butastrocytes outside the plaque and peri-plaque areas arenegative for both antigens. Astrocytes in the plaque mar-gin and peri-plaque area often stain for interferon-� anda bit less frequently for interferon-� (33). Astrocytes inthe plaque center and in normal-appearing tissue awayfrom the plaques were consistently negative except in

patients dying with intercurrent infections in whomstaining was more generalized.

Staining for the intercellular adhesion molecule 1(ICAM-1) revealed diffuse staining of the plaque periph-ery, most intense at the plaque margin but extending wellbeyond the zone of inflammation. Staining also was seenin perivascular cuffs, which included diffuse and cellularstaining. LFA-1 is a ligand for ICAM-1 and is present onessentially all of the hematogenous cells in the plaque andperi-plaque areas. LFA-3, with its ligand CD2, is moreselective for T-cells than ICAM-1, but it has a very simi-lar distribution to that of ICAM-1 in the plaque and peri-plaque areas (33).

Old inactive plaques generally have sharp marginswithout hypercellularity, a few widely scattered T-cells, anddo not stain for adhesion molecules or interferon. They aremarked by gliosis and stain heavily for glial fibrillary acidicprotein. Inflammatory cells are scarce, and most cells in theplaque are astrocytes. A few nonmyelinating oligoden-droglia may be seen near the plaque margins.

Immunocytochemical studies have shown the inflam-matory cells in the actively demyelinating central area ofacute plaques to be mainly Ia-positive cells. These are mostlyactivated microglia and macrophages with a few T-cells andantibody-producing plasma cells. The T cells in the acuteplaque are a mixture of T4� (helper-inducer) and T8� (sup-pressor/cytotoxic) lymphocytes and are more numerousnear the center of the plaque, diminishing in numbersperipherally. As the lesion enlarges, T cells become relativelymore numerous peripherally, whereas macrophages taketheir place centrally. T4� (helper) cells invade the normalappearing white matter about the lesion, whereas T8+ (sup-pressor) cells are confined largely to the plaque margins andperivascular cuffs. The plaque margins contain increasednumbers of oligodendrocytes, astrocytes, and inflammatorycells (33–35). As the lesions become more mature, myelinremnants and macrophages progressively disappear fromthe central part of the plaque (Figure 14.4), which eventu-ally becomes a gliotic scar. At the plaque margin, a hyper-cellular “glial wall” contains lymphocytes, oligodendro-cytes, and a few macrophages and astrocytes. In manyinstances, the disease process appears to continue, withlow-grade activity at the plaque margins, manifested bylipid-laden macrophages and lymphocytes, often accom-panied by a few thin perivascular cuffs. In chronic-activeMS, inflammatory cells are scattered in small numbersthrough much of the normal-appearing white matter. Theseinclude T4�, T8�, and Ia� cells (35,36).

Chronic inactive MS plaques usually have a sharplydemarcated border with little, if any, hypercellularity.Occasional T4� and Ia+ cells are scattered throughout thelesions. At the edges, a few T4�, T8�, and Ia� cells, includ-ing macrophages and B lymphocytes, may be seen, andthese also occur in small numbers throughout the other-wise normal-appearing white matter.


REMYELINATION

Plaquelike areas of very pale myelin staining are a fre-quent occurrence in many cases of MS. On examination,these areas have increased cellularity and abnormallythin myelin sheaths of relatively uniform thickness (Fig-ure 14.1 and 14.5). Numerous studies in experimentalanimals have demonstrated the characteristics of regen-erated myelin with thin sheaths (Figure 14.6) and shortinternodes (Figure 14.7). The thickness of the myelinabout individual axons in these shadow plaques bearsno relationship to fiber diameter as it does in normallymyelinated areas. Some have regarded these shadowplaques as evidence of partial demyelination; however,it is now clear that most, if not all, represent areas ofremyelination (37).

Until the late 1950s, it was accepted dogma thatremyelination did not occur and that oligodendroglia, likeneurons, were endstage cells incapable of regeneration.Since Bunge et al. (38) first unequivocally demonstratedcentral remyelination in the cat in 1958, remyelinationhas been shown to occur in the CNS in essentially everyspecies tested, including tadpole, mouse, rat, guinea pig,rabbit, cat, and dog (39). The characteristics of theshadow plaque are essentially identical to those ofremyelinated areas in experimental animals: an increasednumber of oligodendrocytes that, contrary to traditionalviews, have been demonstrated to be capable of prolifer-ation (40–42), thin myelin sheaths of relatively uniformthickness, and short internodes. The alternative pos-sibility that shadow plaques arise from partial demyeli-nation seems unlikely because partial demyelinationhas been convincingly demonstrated only during theacute phase of the demyelinating process or where the

myelin is under mechanical pressure. In addition, thesethin myelin sheaths, particularly those at plaque mar-gins, have been shown to have short internodes (37).There is abundant evidence for remyelination in exper-imental animals and humans, and the appearance ofshadow plaques and short internodes indicates thatthese are indeed remylinated fibers.

Regeneration of central myelin in some cases isaccompanied by Schwann cell invasion of the CNS; con-sequently, the peripheral myelin can be seen within ademyelinated or remyelinated area (43). Schwann cellsoccur in small numbers in normal spinal cord, but it isuncertain whether they come from peripheral nervesinnervating CNS blood vessels or occur independent ofblood vessels. In pathologic conditions, most of theSchwann cell remyelination within CNS occurs near root

PATHOLOGY192

FIGURE 14.5

Luxol fast blue stain of a shadow plaque in optic nerve. Notethe myelin pallor at lower left compared with the deeper stain-ing of the undemyelinated area at upper right.

FIGURE 14.6

Electron micrograph of remyelination 28 days after mousehepatitis virus inoculation. The newly regenerated myelinsheaths are very thin, and the myelin thickness bears no rela-tionship to axon diameter. An oligodendrocyte can be seenat upper left. Original magnification, 5800X. Herndon RM andWeiner LP, unpublished observations.

entry or exit zones. Breaches in the glial limitans causedby the demyelinating process allow Schwann cells in thenerve roots or autonomic nerves accompanying cerebralblood vessels to migrate into the central white matter,where they retain their capability to form myelin.

With progression of the pathologic process, recur-rent demyelination at the plaque margins causes slowenlargement of the plaques. Although remyelinationappears to be a regular event in early MS, it clearlybecomes less common and less effective as the disease pro-gresses, so that, at autopsy in patients with long stand-ing disease, evidence of remyelination may be quite lim-ited. Over time, demyelination of newly remyelinatedareas may result in scarring, so that further remyelina-tion cannot occur and the plaque becomes a hypocellu-lar glial scar. Shadow plaques are thus more likely to beseen in autopsies of younger patients with continuedactive remitting disease (37), and much less common inolder patients with inactive disease or chronically active,indolent disease progression. In most cases, the end resultis a nervous system riddled with chronically scarred, inac-tive plaques and a considerable amount of Walleriandegeneration resulting in the pallor of long tracts inmyelin-stained preparations.

SECONDARY PROGRESSIVE MULTIPLE SCLEROSIS

Secondary progressive MS is usually marked by a slowlyascending paralysis or, much more rarely, progressiveataxia. It commonly occurs in patients in their 40s or early

50s who have had MS for a number of years. The patho-logic substrate for this appears to be a progressive lossof axons in the spinal cord with cord atrophy. Examina-tion of the cord, in addition to chronic inactive or occa-sionally active plaques, shows demyelination and/ordegeneration of multiple individual fibers. Careful exam-ination of the cord shows widely scattered individual fiberdemyelination with a few scattered macrophages alongthe sheaths (20). This could result from an inadequatesupply of transported materials with a resulting inabilityof the axon to adequately signal the myelin-forming oligo-dendrocytes. Certainly such scattered fiber demyelinationcould be a part of a dying axonopathy and thus the sub-strate for the slowly progressive ascending paralysis ofsecondary progressive MS (44) (Chapter 16).

What mechanisms could be invoked for such aprocess? It might be a problem in axonal transport, withretrograde degeneration, or axonal interruption second-ary to scarring with Wallerian degeneration. The formerseems more likely. Extensive studies in animals haveshown that fast and slow axonal transport are markedlyslowed by demyelination. Such axonal damage upstreamwould lead to an inadequate supply of the structuralmaterials needed to maintain energy metabolism andaxonal structure. These conditions, particularly if theaxons metabolic resources are stretched due to inducedsprouting of the distal axon to take over synapses vacatedby death of other axons, could lead to a dying backaxonopathy. Clinically this would result initially in deathof the longest axons with gradual progression to shorterand shorter axons. The second major possibility is thatthey are undergoing Wallerian degeneration secondary to


FIGURE 14.7

Remyelinated axon 28 days after mouse hepatitis virus inoculation. Two nodes are seen only 6 �m apart. Original magnifica-tion, 19000X. Herndon RM and Weiner LP, unpublished observations.

glial scar formation, with death of axons occurring inplaques as the extensive astrocytic scarring causes shrink-age of the plaques, thus compressing and choking off theaxons. Either scenario could lead to an ascending paral-ysis, although one would expect that the second wouldcause a more scattered, less clearly ascending picture.

PRIMARY PROGRESSIVE MULTIPLE SCLEROSIS

In the studies of Lucchinetti et al., all of the cases of pri-mary progressive MS had pattern 4 (1). Inflammation wasdominated by T lymphocytes, and macrophages, IgG, andcomplement deposition were absent. Death of oligoden-drocytes ocurred in peri-plaque white matter adjacent toareas of demyelination. The dying oligodendrocytesdemonstrated DNA fragmentation without features ofapoptosis. Staining for the various myelin proteins showedan essentially simultaneous loss in staining for each of theproteins. Thus, it appears to be an oligodendrogliopathy.

Primary progressive MS constitutes about 10 per-cent of MS cases and is equally common in men andwomen. This pattern was the least common in the seriesof Lucchinetti et al., comprising 4 percent of the casesstudied, and was the only one associated with a partic-ular clinical disease pattern. Its association with primaryprogressive MS strongly supports the frequentlyexpressed view that true primary progressive MS is adisease distinct from relapsing remitting and second-ary progressive MS. Some cases that clinically appear asprimary progressive disease undoubtedly will have thepathology of secondary progressive disease becauserelapsing remitting cases can be asymptomatic or nearlyso (22) and thus may present clinically only when theymove into the secondary progressive phase. Such casesshow the pathology of secondary progressive diseaseeven though clinical attacks were never observed. Thiswill tend to confuse the clinical distinction between thedisease types but is explained by the fact that a personcan have multiple demyelinating events without symp-toms in relapsing remitting MS, so that secondary pro-gressive disease can masquerade clinically as primaryprogressive disease.

ACUTE MULTIPLE SCLEROSIS (MARBURG DISEASE)

Acute MS is fortunately rather rare. It is marked by therapid onset and almost continual progression of demyeli-nation. It can begin in the hemispheres or the brainstem,and typical cases display on MRI multiple high signalareas that increase in size and become confluent. Manypatients die within a few weeks to several months. Many

of those who survive go on to have a typical relapsingremitting course.

Pathologically, it is marked by intense and wide-spread inflammation with T lymphocytes, large numbersof lipid-laden macrophages, and a scattering of B lym-phocytes (18,19). The lipid-laden macrophages stainintensely for neutral lipid with oil red-O but show littlestaining with most myelin stains. Essentially, total demyeli-nation occurs within the lesions, although beginningremyelination can be seen. Extensive IgG and complementproduct is deposited in the lesions, which pathologicallyresemble type 2 lesions of Lucchinetti et al. There is noevidence of death of oligodendroglia in the area sur-rounding the plaques. In semithin sections, partly digestedbits of myelin are seen in macrophages. The demyelinatedaxons are irregularly constricted and dilated, unlike thosein more chronic plaques, which have a more uniformdiameter. Perivascular lymphocytic cuffs may be presentbut often are sparse.

Oligodendrocytes are usually considerably reducedin the acute lesions, although larger numbers may appearin somewhat older lesions (19), indicating that oligoden-drocyte regeneration occurs in humans as it does in exper-imental animals.

ACUTE DISSEMINATEDENCEPHALOMYELITIS

ADEM has a host of names, including perivenousencephalomyelitis, encephalitis periaxalis diffusa, acutehemorrhagic encephalomyelitis, postinfectious encephalo-myelitis, and postvaccinal encephalomyelitis. Of thedemyelinating diseases, this appears to correspond mostclosely to EAE. It is an acute diffuse inflammatory demyeli-nating disorder that often occurs 1 to 3 weeks after a viralinfection or immunization. Its severity and intensity vary,but it usually runs a monophasic course, with a very smallpercentage later developing into a relapsing remitting dis-ease. In those who survive the acute episode, slow improve-ment usually occurs over many months, often with remark-ably good recovery and little residua. Most fatal casesfollow smallpox vaccination or measles infection. Thehemorrhagic form has been considered to represent a dif-ferent entity by some (18), although some have consid-ered it to be a more severe form of ADEM (45) because,in addition to the perivenous hemorrhages, perivenousinflammation and demyelination are regularly observed.

The gross pathology is marked by swelling and con-gestion, often with evidence of uncal and foraminal her-niation. The fresh brain may show numerous petechia orsimply appear swollen. Microscopically, it is marked byextensive perivenous inflammation and demyelination.The venules are surrounded by prominent cuffs of lym-phocytes and macrophages. Often, long strips of perive-

PATHOLOGY194

nous demyelination occur, giving the tissue a reticulateappearance. In some cases, there is spread of macrophagesand lymphocytes into the parenchyma. The macrophagesmay be laden with neutral lipid even rather early in thedisease process.

In the hemorrhagic cases, in addition to the perivas-cular cuffs, there is edema and necrosis of blood vesselswith fibrin deposition and numerous ring or ball hemor-rhages. In these cases, there are usually numerous neu-trophils in the infiltrate, and there may be areas of franknecrosis of tissue. Attempts to recover virus in these casesor to demonstrate by antibody staining or in situhybridization have usually failed. Overall, the pathologicpicture is essentially identical to EAE as seen in experi-mental animals (45).

NEUROMYELITIS OPTICA (DEVIC DISEASE)

The place of neuromyelitis optica in the MS spectrum haslong been in dispute. It has been said to be a monophasicdisease, with optic neuritis and transverse myelitis occur-ring within a few weeks to a month or two of each other,without recurrences. However, it is clear that over half thecases that fit these criteria initially go on to developrelapses and remissions. Wingerchuk and coworkersreviewed 71 patients with neuromyelitis optica (46). Ofthese, 23 had a monophasic course and the other 48 hada relapsing remitting course. In the cases of monophasicillness, they noted that the initial optic neuritis and trans-verse myelitis occur quite close together, with a medianof 5 days between the two events, whereas the median was166 days for the relapsing cases. The relapsing cases hada poor prognosis with poor recovery between attacks.They frequently developed respiratory failure secondaryto cervical cord disease.

Clinically Devic syndrome differs from typical MSin its clinical pattern and in the relatively poor recoveryusually seen after the attacks. It also differs in its distri-bution in the population, being more common in Asiansand Africans than in Caucasians. In Japan, the Devic pat-tern makes up about 7 percent of reported cases (47).

In the acute stages there is necrosis of the cord, usu-ally extending over three or more segments. The cord isusually swollen and soft in the affected area. There maybe additional demyelinating lesions in other cord segmentsbut the brain itself is not involved. The CSF has a vari-able pleocytosis, often with 50 to 100 or more cells. Theseare predominantly lymphocytes with a variable numberof monocytes and neutrophil leucocytes and occasionallya few eosinophil leucocytes. There may be extensiveperivascular cuffs containing macrophages and granulo-cytes, including eosinophils and CD3� lymphocytes (48).The inflammation and necrosis are marked by extensivemacrophage infiltration with lymphocytes and often poly-

morphonuclear leucocytes. Extensive deposition of IgGand C9 neoantigen is a marker for complement deposi-tion in areas of myelin necrosis. Extensive oligodendro-cyte destruction occurs, and all the myelin proteins dis-appear at about the same time. Thus the pathologyresembles a more severe, necrotic pattern similar to thetype 2 of Lucchinetti et al. It differs in that much of theIgG and complement deposition is on and around bloodvessels in the Devic’s cases (48).

In the late stages, cavitation and necrosis of gray andwhite matter occurs in the cord, and most of the lengthof the cord is affected. There is extensive scarring withmarked hyalinization and thickening of blood vessels andperivascular fibrosis (49). Numerous macrophages arepresent in necrotic areas. In the optic nerves, similar necro-sis and thickening and hyalinization of blood vessels is notseen, but demyelination typically includes the entire opticchiasm.

Axonal destruction in the lesions is much greaterthan that usually seen in MS. The destructiveness of thelesions is emphasized by the clinical findings, with rela-tively poor recovery from the lesions. For example, com-plete blindness is extremely rare in exacerbating remittingMS but common in neuromyelitis optica. Similarly, per-manent paraplegia or quadriplegia is rare after a singleattack of relapsing remitting MS but fairly common inDevic disease.

STRUCTURES RESEMBLING INFECTIOUSAGENTS IN MS TISSUES

Despite a number of reports of “viruslike particles” seenby electron microscopy of MS biopsy and postmortemmaterial (9,50,51), credible morphologic evidence for thepresence of true virions has not been reported. Dense bod-ies surrounded by a membrane have been seen intracel-lularly, mainly in macrophages, in proximity to activelydemyelinating areas. These are of variable size, do notclosely resemble any particular virus or class of viruses,and are generally regarded as myelin breakdown prod-ucts. Many other investigators, beginning with Prineasin 1972 (9), have reported “paramyxoviruslike” tubulesin the nuclei of inflammatory cells in MS plaques. Theseconsist of strands of 18- to 20-nm intranuclear filamentsor tubules. In size and appearance, they resemble thenucleocapsid of paramyxoviruses. They have been seenin a variety of conditions including normal tissues fixedunder acidic conditions (R.M. Herndon, unpublishedobservations) and appear to be chromatin strands, whichhave an altered appearance due to the metabolic state ofthe cell or conditions of tissue fixation (52).

With newer techniques of polymerase chain reac-tion, representational difference analysis, in situhybridization, and immunocytochemical techniques,


viruses are frequently found in MS plaques. The effortto tie viruses found in brain to the disease process hasfailed thus far. The viruses of greatest interest are the her-pes viruses. These viruses are found latent in nervous tis-sue, are activated by other viral infections, just as newattacks of MS frequently follow viral infection, and mightactivate an inflammatory process in the nervous systemresulting in a new attack of MS. Challoner et al. (12)reported the detection of herpes type 6 in MS brain usingpolymerase chain reaction and representational differ-ence analysis followed by localization in plaques usingimmunocytochemistry. Subsequently, Sanders et al. (13)detected several herpes viruses in brain tissue (Table14.2). Although several were more common in MS brainsand more common in MS plaques, the differences did notreach statistical significance. Many of these viruses arecarried in macrophages, so they would be more likely toappear in sites with large numbers of macrophages, suchas MS plaques. Nothing in this work indicated an etio-logic relationship to MS, but the presence of latentviruses, which are easily activated in plaques, suggeststhat they could contribute to the pathology.

CHLAMYDIA PNEUMONIAE

The presence of Chlamydia pneumonia in the CSF ofindividuals with MS has been demonstrated by poly-merase chain reaction and culture (15,53) and is dis-cussed in detail in Chapter 10. Chlamydia has notbeen reported pathologically in MS tissue thus far, butstructures resembling microorganisms compatiblewith C. pneumonia have been seen in the CSF sedi-ment in MS cases (R.M. Herndon, unpublished obser-vations). Whether this will prove to be related to thedisease or yet another incidental occurrence remainsto be determined.

CONCLUSION

MS is an inflammatory demyelinating disorder or, morelikely, a group of disorders of unknown but most likelyautoimmune or infectious origin. Current pathologic evi-dence suggests that MS may have more than one etiologyand that primary progressive MS is a separate disease.Much of the permanent disability in MS results fromaxonal destruction, which falls most heavily on very longpathways such as the pyramidal tract supplying the legsand the dorsal columns carrying sensory informationfrom the legs. These long pathways take multiple hitsover the years, with increasing axonal destruction lead-ing to the loss of lower extremity function so commonin advanced MS. Other aspects of the disease, such asincoordination and imbalance are caused by delayed anddegraded information resulting from slowed conductionof proprioceptive information and the inability to mon-itor, on line, motor processes due to conduction delaysand signal dispersion occurring as the signals passthrough demyelinated areas, as discussed further in thechapter on pathophysiology. Every attack, even subclin-ical ones, causes some permanent damage, and it is theaccumulation of damage from repeated demyelinatingepisodes that accounts for most of the long-term disabil-ity. It is this progressive accumulation of damage that pro-vides the best rationale for early use of disease-alteringtherapies.

References

1. Lucchinetti C, Bruck W, Parisi J, et al. Heterogeneity of multiplesclerosis lesions:implications for the pathogenesis of demyelina-tion. Ann Neurol 2000; 47:707–717.

2. Hughes, RA, Hadden RD, Gregson NA, Smith KJ. Pathogenesisof Guillain-Barre syndrome. J Neuroimmunol 1999; 100:74–97.

3. Ho TW, McKhann GM, Griffin JW. Human autoimmune neu-ropathies. Annu Rev Neurosci 1998; 21:187–226.

4. Schonberger LB, Bregman DJ, Sullivan-Bolyai JZ, et al. Guillian-Barre syndrome following vaccination in the National InfluenzaImmunization Program, United States 1976–1977. Am J Epi-demiol 1979; 110:105–123.

5. Mendell JR, Kolkin S, Kissel JT, et al. Evidence for central nerv-ous system demyelination in chronic inflammatory demyelinat-ing polyradiculoneuropathy. Neurology 1987; 37:1291–1294.

6. Ormerod IE, Waddy HM, Kermode AG, Murray NM, ThomasPK. Involvement of the central nervous system in chronic inflam-matory demyelinating polyneuropathy: a clinical, electrophysio-logical and magnetic resonance imaging study. J Neurol Neuro-surg Psychiatry. 199053:789–793.

7. Stojkovic T, de Seze J, Hurtevent JF, et al. Visual evoked poten-tials study in chronic idiopathic inflammatory demyelinatingpolyneuropathy. Clin Neurophysiol 2000111:2285–2291.

8. Compston A. The genetics of multiple sclerosis. J Neurovirol2000; (suppl 2):S5–S9.

9. Prineas J. Paramyxovirus-like particles associated with acutedemyelination in chronic relapsing multiple sclerosis. Science1972; 178:760–763.

10. Johnson RT. The virology of demyelinating diseases. Ann Neu-rol 1994; 36:S54–S60.

11. Cook SD. Evidence for a viral etiology of multiple sclerosis. In:Cook SD (ed.), Handbook of Multiple Sclerosis. New York: Mar-cel Dekker; 2001.

PATHOLOGY196

TABLE 14.2Herpes Viruses in Brain Tissue

VIRUS MS BRAIN ACTIVE

PLAQUE PLAQUE BRAIN INACTIVE CONTROL

Herpes 37% 41% 20% 28%simplexHerpes 6 57% 32% 17% 43%Varicella 43% 14% 10% 32%zoster virusEpstein-Barr 27% 5% 10% 38%virus

Cytomegalovirus 16% 9% 10% 22%

Adapted from: J Neurovirol 1996; 2:249–258.

12. Challoner PB, Smith KT, Parker JD, et al. Plaque-associatedexpression of human herpesvirus 6 in multiple sclerosis. ProcNat Acad Sci USA 1995; 92:7440–7444.

13. Sanders VJ, Felisan S, Waddell A, Tourtellotte WW. Detection ofherpesvirdae in postmortem multiple sclerosis brain tissue andcontrols by polymerase chain reaction. J Neurovirol 1996;2:249–258.

14. Lycke J, Svennerholm B, Hjelmquist E, et al. Acyclovir treatmentof relapsing-remitting multiple sclerosis. A randomized, placebo-controlled, double blind study. J Neurol 1996; 243:214–224.

15. Sriram S, Stratton CW, Yao S, et al. Chlamydia pneumoniaeinfection of the central nervous system in multiple sclerosis.Ann Neurol 1999; 46:6–14.

16. Jeffrey DR, Alshami E. Allogenic bone marrow transplantationin multiple sclerosis. Neurology 1998; 50(suppl 4):A147.

17. Cottrell DA, Kremenchutzky M, Rice GP, et al. The natural his-tory of multiple sclerosis: a geographically based study. 5. Theclincal features and natural history of primary progressive mul-tiple sclerosis. Brain 1999; 122:625–639.

18. Prineas JW, McDonald WI. Demyelinating diseases. In: GrahamDI, Lantos PL (eds.), Greenfields Neuropathology. 6th ed.813–896.

19. Marburg O. Die sogenannte “akute multiple Sklerose” Jahrb.Psychiat Neurol (Leizig) 1906; 27:213–311.

20. Adams CWM. The general pathology of multiple sclerosisi: mor-phological and chemical aspect of the lesions. In: Hallpike JF,Adams CWM, Tourtellotte WW (eds.), Multiple Sclerosis. Bal-timore: Williams & Wilkins; 1983; 203–240.

21. Lassman H, Raine CS, Antel J, Prineas JW. Immunopathologyof multiple sclerosis: Report on an international meeting heldat the Institute of Neurology of the University of Vienna. J Neu-roimmunol 1998; 86:213–217.

22. Gilbert JJ, Sadler M. Unsuspected multiple sclerosis. Arch Neu-rol 1983; 40:533–536.

23. Prineas JW, Connell F. The fine structure of chronically activemultiple sclerosis plaques. Neurology 1978; 28:S68–S75.

24. Itoyama Y, Sternberger NH, Webster HD, et al. Immunocyto-chemical observations on the distribution of myelin-associatedglycoprotein and myelin basic protein in multiple sclerosislesions. Ann Neurol 1980; 7:167–177.

25. Prineas JW. Pathology of the early lesions of multiple sclerosis.Hum Pathol 1975; 6:531–535.

26. Dow RS, Berglund G. Vascular pattern of lesions of multiple scle-rosis. Arch Neurol Psychiat 1942; 47:1–18.

27. Herndon RM, Johnson MA. A method for the electron micro-scopic study of cerebrospinal fluid sediment. J Neuropathol ExpNeurol 1970; 29:320–330.

28. Herndon RM, Kasckow J. Electron microscopic studies of cere-brospinal fluid sediment in demyelinating disease. Ann Neurol1978; 4:515–523.

29. Stefano N de, Narayanan S, Francis GS, et al. Evidence of axonaldamage in the early stages of multiple sclerosis and its relevanceto disability. Arch Neurol 2001; 58:65–70.

30. Trapp BD, Peterson J, Ransohoff RM, et al. Axonal transection inthe lesions of multiple sclerosis. N Engl J Med 1998; 338:278–285.

31. Trapp BD, Bo L, Mork S, Chang A. Pathogenesis of tissue injuryin MS lesions. J Neuroimmunol 1999; 98:49–56.

32. Simon JH, Kinkel RP, Jacobs L, et al. A Wallerian degeneration pat-tern in patients at risk of MS. Neurology 2000; 54:1155–1160.

33. Traugott U. On the role of astrocytes for lesion pathogenesis. In:Herndon RM, Seil FJ, (eds.), Multiple Sclerosis: Current Statusof Research and Treatment. New York: Demos, 1994; 13–31.

34. Peterson JW, Bö L, Mörk S, Chang A, Trapp BD. Transectedneurites, apoptotic neurons, and reduced inflammation in cor-tical multiple sclerosis lesions. Ann Neurol 2001; 50:389–400.

35. Traugott U, Reinherz EL, Raine CS. Multiple sclerosis: Distri-bution of T-cells, T-cell subsets and Ia positive macrophages inlesions of different ages. J Neuroimmunol 1983; 4:201–221.

36. Traugott U, Raine CS. Further lymphocyte characterization inthe central nervous system in multiple sclerosis. Ann NY AcadSci 1984; 436:163.

37. Prineas JW, Connell F. Remyelination in multiple sclerosis. AnnNeurol 1979; 5:22–31.

38. Bunge MB, Bunge RP, Ris H. Ultrastructural study of remyeli-nation in an experimental lesion in adult cat spinal cord. J Bio-phys Biochem Cytol 1961; 10: 67–94.

39. Hommes OR. Remyelination in human CNS lesions. Prog BrainRes 1980; 53:39–63.

40. Herndon RM, Price DL, Weiner LP. Regeneration of oligoden-droglia during recovery from demyelinating disease. Science1977; 195:693–694.

41. Arenella L, Herndon RM. Mature oligodendrocytes: Divisionfollowing experimental demyelination in adult animals. ArchNeurol 1984; 41:1162–1165.

42. Ludwin SK. Proliferation of mature oligodendrocytes after traumato the central nervous system. Nature 1984; 308:274–276.

43. Itoyama Y, Webster H deF, Richardson EP, Trapp BD. Schwanncell remyelination of demyelinated axons in spinal cord multi-ple sclerosis lesions. Ann Neurol 1983; 14:339–346.

44. Herndon RM. Why secondary progressive multiple sclerosis is arelentlessly progressive illness? Arch Neurol 2002; 59:301–304.

45. Hart MN, Earle KM. Haemorrhagic and perivenous encephali-tis: A clinico-pathological review of 38 cases. J Neurol Neuro-surg Psychiatry 1975; 38:585–591.

46. Wingerchuk DM, Hogancamp WF, O’Brien PC, WeinshenkerBG. The clinical course of neuromyelitis optica (Devic syn-drome). Neurology 1999; 53:1107–1114.

47. Kuroiwa Y, Igata A, Itahara K, Koshijima S, Tusbaki T. Nation-wide survey of multiple sclerosis in Japan. Clinical analysis of1,084 cases. Neurology 1975; 25:845–851.

48. Lucchinetti CF, Mandler RN, McGovern D, et al. A role forhumeral mechanisms in the pathogenesis of Devic’s neu-romyelitis optica. Brain 2002; 125:1450–1461.

49. Mandler RN, Davis LE, Jeffery DR, Kornfeld M. Devic neu-romyelitis optica: a clinicopathological study of 8 patients. AnnNeurol 1993; 34:162–168.

50. Raine CS, Powers JM, Suzuki K. Acute multiple sclerosis—con-firmation of “paramyxovirus-like” tubules. Arch Neurol 1974;30:39–46.

51. Field EJ, Cowshall S, Narang HK, Bell TM. Viruses in multi-ple sclerosis? Lancet 1972; ii:280–281.

52. Raine CS, Schaumberg HH, Snyder DK, Suzuki K. Intranuclear“Paramyxovirus-like” material in multiple sclerosis, adreno-leukodystrophy and Kuf’s disease. J Neurol Sci 1975; 25:29–41.

53. Sriram S, Stratton CW, Yao S-Y, et al. Chlamydia pneumoniaeinfection of the central nervous system in multiple sclerosis. AnnNeurol 1999; 46:6–14.


Health & Medicine

The Pathology of Multiple Sclerosis and Its Variants