tipos de mielitis

24
Acute Disseminated Encephalomyelitis, Transverse Myelitis, and Neuromyelitis Optica Dean M. Wingerchuk, MD, MSc, FRCP(C), FAAN; Brian G. Weinshenker, MD, FRCP(C), FAAN ABSTRACT Purpose of Review: This review defines current clinical criteria for diagnosis, differential diagnosis, and clinical evaluation of acute disseminated encephalomy- elitis, transverse myelitis, and neuromyelitis optica, and summarizes principles of treatment. Recent Findings: Consensus criteria for transverse myelitis and acute dissemi- nated encephalomyelitis have been proposed. A specific biomarker, aquaporin-4 autoantibody, has been discovered for neuromyelitis optica that allows for early and accurate diagnosis even in the absence of cardinal findings of optic neuritis and myelitis. The antibody is pathogenic and is facilitating an understanding of the pathophysiology of neuromyelitis optica and development of antigen-specific treatments. Summary: Clinical and radiologic findings combined with serologic findings may permit classification of syndromes of transverse myelitis and acute disseminated encephalomyelitis in ways that may predict risk of relapse, type of relapse, and prognosis. Treatment, especially to prevent relapse, is dependent on the specific disease context in which syndromes such as transverse myelitis occur. Continuum (Minneap Minn) 2013;19(4):944–967. INTRODUCTION: SYNDROME VERSUS DISEASE The nosology of demyelinating dis- eases of the CNS is complex. Multiple sclerosis (MS) has been an umbrella term for recurrent inflammatory dis- ease of the CNS after definable non- demyelinating mimics are excluded. Acute disseminated encephalomyelitis (ADEM), transverse myelitis, and neu- romyelitis optica (NMO) are inflamma- tory conditions that have not been well distinguished from MS or its presenting syndromes (termed ‘‘clini- cally isolated [demyelinating] syn- dromes’’) but are linked by their tendency to relapse and remit and by their inflammatory characteristics and overlapping pathology. Distinction between syndromes that reflect localization (eg, optic neu- ritis and transverse myelitis) and dis- ease entities (eg, ADEM, MS, and NMO) is now feasible. Distinction is important because of the prognostic and treat- ment implications of different disease entities. For example, transverse myeli- tis refers to a syndrome of acute or subacute myelopathy accompanied by indicators of inflammation, either ra- diologically or based on spinal fluid. It may occur as a stand-alone syndrome, a component of ADEM, a relapse of MS or NMO, or a nondemyelinating syndrome Address correspondence to Dr Brian G. Weinshenker, Department of Neurology, Mayo Clinic, 200 First St SW, Rochester, MN 55905, [email protected]. Relationship Disclosure: Dr Wingerchuk receives research support from Alexion, Genentech, Genzyme Corporation, Guthy-Jackson Charitable Foundation, and Terumo BCT, Inc. Dr Weinshenker serves on the data and safety monitoring board of Biogen Idec and Novartis and serves as a consultant regarding neuromyelitis optica therapeutics for Asahi Kasei Medical Co, Ltd, Elan Corporation, and Novartis. Dr Weinshenker receives royalties for licensed technology for the diagnosis of neuromyelitis optica from RSR Limited, and receives research funding from Guthy-Jackson Charitable Foundation. Unlabeled Use of Products/Investigational Use Disclosure: Drs Wingerchuk and Weinshenker discuss the unlabeled uses of corticosteroids and plasma exchange for the treatment of acute disseminated encephalomyelitis, transverse myelitis, and neuromyelitis optica; IV immunoglobulin for acute disseminated encephalomyelitis; and azathioprine, mycophenolate mofetil, rituximab, mitoxantrone, methotrexate, and eculizumab for neuromyelitis optica. * 2013, American Academy of Neurology. 944 www.ContinuumJournal.com August 2013 Review Article Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Upload: mario-alberto-ornelas-casillas

Post on 07-Nov-2015

19 views

Category:

Documents


7 download

DESCRIPTION

Artículo original

TRANSCRIPT

  • Acute DisseminatedEncephalomyelitis,Transverse Myelitis, andNeuromyelitis Optica

    Dean M. Wingerchuk, MD, MSc, FRCP(C), FAAN;Brian G. Weinshenker, MD, FRCP(C), FAAN

    ABSTRACTPurpose of Review: This review defines current clinical criteria for diagnosis,differential diagnosis, and clinical evaluation of acute disseminated encephalomy-elitis, transverse myelitis, and neuromyelitis optica, and summarizes principles oftreatment.Recent Findings: Consensus criteria for transverse myelitis and acute dissemi-nated encephalomyelitis have been proposed. A specific biomarker, aquaporin-4autoantibody, has been discovered for neuromyelitis optica that allows for early andaccurate diagnosis even in the absence of cardinal findings of optic neuritis andmyelitis.The antibody is pathogenic and is facilitating an understanding of the pathophysiologyof neuromyelitis optica and development of antigen-specific treatments.Summary: Clinical and radiologic findings combined with serologic findings maypermit classification of syndromes of transverse myelitis and acute disseminatedencephalomyelitis in ways that may predict risk of relapse, type of relapse, andprognosis. Treatment, especially to prevent relapse, is dependent on the specificdisease context in which syndromes such as transverse myelitis occur.

    Continuum (Minneap Minn) 2013;19(4):944967.

    INTRODUCTION: SYNDROMEVERSUS DISEASEThe nosology of demyelinating dis-eases of the CNS is complex. Multiplesclerosis (MS) has been an umbrellaterm for recurrent inflammatory dis-ease of the CNS after definable non-demyelinating mimics are excluded.Acute disseminated encephalomyelitis(ADEM), transverse myelitis, and neu-romyelitis optica (NMO) are inflamma-tory conditions that have not beenwell distinguished from MS or itspresenting syndromes (termed clini-cally isolated [demyelinating] syn-dromes) but are linked by theirtendency to relapse and remit and by

    their inflammatory characteristics andoverlapping pathology.

    Distinction between syndromesthat reflect localization (eg, optic neu-ritis and transverse myelitis) and dis-ease entities (eg, ADEM, MS, and NMO)is now feasible. Distinction is importantbecause of the prognostic and treat-ment implications of different diseaseentities. For example, transverse myeli-tis refers to a syndrome of acute orsubacute myelopathy accompanied byindicators of inflammation, either ra-diologically or based on spinal fluid. Itmay occur as a stand-alone syndrome, acomponent of ADEM, a relapse of MS orNMO, or a nondemyelinating syndrome

    Address correspondence toDr Brian G. Weinshenker,Department of Neurology,Mayo Clinic, 200 First St SW,Rochester, MN 55905,[email protected].

    Relationship Disclosure:Dr Wingerchuk receivesresearch support fromAlexion, Genentech,Genzyme Corporation,Guthy-Jackson CharitableFoundation, and Terumo BCT,Inc. Dr Weinshenker serveson the data and safetymonitoring board of BiogenIdec and Novartis and servesas a consultant regardingneuromyelitis opticatherapeutics for Asahi KaseiMedical Co, Ltd, ElanCorporation, and Novartis.Dr Weinshenker receivesroyalties for licensedtechnology for the diagnosisof neuromyelitis optica fromRSR Limited, and receivesresearch funding fromGuthy-Jackson CharitableFoundation.

    Unlabeled Use ofProducts/InvestigationalUse Disclosure: DrsWingerchuk and Weinshenkerdiscuss the unlabeled uses ofcorticosteroids and plasmaexchange for the treatmentof acute disseminatedencephalomyelitis, transversemyelitis, and neuromyelitisoptica; IV immunoglobulinfor acute disseminatedencephalomyelitis; andazathioprine, mycophenolatemofetil, rituximab,mitoxantrone, methotrexate,and eculizumab forneuromyelitis optica.

    * 2013, American Academyof Neurology.

    944 www.ContinuumJournal.com August 2013

    Review Article

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • such as an infectious myelitis or agranulomatous myelitis. If due to her-pes virus infection, it is best treated withantiviral treatment; if indicative of arelapse or harbinger of MS, an MSimmunomodulatory treatment may beappropriate; if a relapse of NMO, animmunosuppressive drug, such as aza-thioprine, would be appropriate,whereas interferon-" or natalizumabmay actually be deleterious. Noninfec-tious and noninflammatory disorders,such as arteriovenous fistula, mayproduce syndromes suggestive oftransverse myelitis but would requireentirely different treatment.

    ADEM, as currently defined, is char-acterized by acute encephalopathy butfrequently accompanied by optic neuri-tis or transverse myelitis. NMO is thefirst inflammatory demyelinating condi-tion to be defined, in part, by a bio-marker that is molding our evolvingconcept of this condition (an auto-immune aquaporinopathy) and ex-panding the spectrum of the diseaseto include nonYoptic nerve and spinalcord syndromes and MRI-detectedbrain lesions that, in the past, wouldhave excluded a diagnosis of NMO.

    Understanding of the distinction andinterrelationships of these syndromeshas been facilitated by insights into thepathophysiology, informed by the neu-ropathology, and illustrated by ad-vances in NMO over the past decadeas outlined below.

    ACUTE DISSEMINATEDENCEPHALOMYELITISAlthough a number of definitions havebeen proposed for ADEM, it remains apoorly defined syndrome of symptom-atic diffuse or multifocal CNS inflam-mation that is typically, if not always, amonophasic illness. ADEM has beenhistorically regarded as the clinicalcounterpart of the experimental dis-ease experimental autoimmune en-

    cephalomyelitis. ADEM is also knownas postvaccinal encephalomyelitiswhen it follows vaccination. Accord-ingly, it is believed to be induced byan immune reaction directed at across-reacting myelin antigen. Itssomewhat unique pathology ofperivenous sleeves of inflammationand demyelination has been recog-nized for decades but has been re-cently rediscovered and expanded.1

    Pathologically, ADEM can be distin-guished from fulminant acute MS,which is a major consideration in thedifferential diagnosis. Acute MS isassociated with confluent demyelin-ation and prominent macrophagesadmixed with reactive astrocytes.When cases are identified based onpathologic features, these featuresseem to be associated with some ofthe key clinical characteristics that havebeen identified in consensus clinicaldiagnostic criteria (Table 3-12) to dis-tinguish ADEM from MSVin particular,encephalopathy.1 Pathology has re-cently been proposed as the refer-ence standard to distinguish ADEMfrom fulminant MS, although this hasnot been widely debated and cannotbe considered as generally accepted atthe present time. Consensus criteria2

    remain imperfect in distinguishing pa-tients who, in the course of follow-up,will remain free of future relapses.Furthermore, consensus criteria allowfor recurrent and even multiphasicADEM episodes with criteria that mightdistinguish it from MS (ie, a newepisode must also meet the criteriafor ADEM and not be those of anattack of MSVsee Table 3-1). Contro-versy persists about whether relapsesmay occur in ADEM and remain con-sistent with the diagnosis of ADEM,especially in adult patients.

    The key clinical features requiredfor a diagnosis of ADEM includediffuse encephalopathy but may also

    KEY POINTS

    h Distinguishing betweensyndromes (eg, opticneuritis and transversemyelitis) and diseaseentities (multiplesclerosis, neuromyelitisoptica, and acutedisseminatedencephalomyelitis)is vital. Demyelinatingsyndromes may be acomponent of differentdiseases and mayhave vastly differentprognoses dependingon the disease context(and may thereforerequire differenttreatment).

    h Although the proposedconsensus criteria foracute disseminatedencephalomyelitisallows for recurrent ormultiphasic formsof the disease, theexistence of relapsingforms remainscontroversial. Acriterion standardfor distinguishingacute disseminatedencephalomyelitisfrom multiple sclerosishas not been widelyaccepted, althoughpathology has beenproposed.

    945Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • include other symptoms characteristicof an inflammatory demyelinating dis-ease, such as optic neuritis or trans-verse myelitis, including transversemyelitis associated with longitudinallyextensive lesions. Clinical features mayinclude seizures or coma; coma ishighly suggestive of ADEM in the

    appropriate setting of extensive whitematter disease (Case 3-1). MRI usuallyshows multiple lesions, although asingle large lesion is still felt to becompatible in the consensus criteria.Usually, cerebral lesions occur in bothcerebral hemispheres in ADEM.3 In-volvement of the deep gray matter is

    TABLE 3-1 International Pediatric Multiple Sclerosis Study Group Consensus Definitionsa

    b Monophasic Acute Disseminated Encephalomyelitis (ADEM)

    A first clinical event with a presumed inflammatory or demyelinating cause

    Acute or subacute onset that affects multifocal areas of the CNS

    Polysymptomatic and must include encephalopathy, including one or more of the following:

    Behavioral change (confusion, irritability)

    Alteration in consciousness (lethargy, coma)

    Postevent improvement clinically, on MRI, or both; residual deficits permitted

    No prior clinical episode consistent with demyelinating event

    No other etiologies apparent

    New or fluctuating symptoms, signs, or MRI findings occurring within 3 months of the inciting ADEMevent permissible

    Brain MRI (T2-weighted) with focal or multifocal supratentorial or infratentorial lesions, often large (1 to2 cm), predominantly involving white matter, but frequently present in gray matter (especially basalganglia or thalamus), without evidence of previous destructive white matter changes; brain MRI showinga single large lesion (1 to 2 cm), predominantly affecting white matter also possible, although rare

    Spinal cord MRI may have confluent intramedullary lesion(s) with variable enhancement, in addition toabnormal brain MRI findings specified previously

    b Recurrent ADEM

    Recurrence of the initial symptoms and signs of ADEM 3 or more months after the first event

    No new lesions based on history, examination, or neuroimaging

    Event does not occur while on corticosteroids and occurs at least 1 month after completing therapy

    MRI shows no new lesions; original lesions may have enlarged

    No better explanation exists

    b Multiphasic ADEM

    New clinical event also meeting criteria for ADEM but involving new anatomic areas of the CNS asconfirmed by history, neurologic examination, and neuroimaging

    Subsequent event must occur:

    At least 3 months after the onset of the initial ADEM event

    At least 1 month after completing steroid therapy

    Subsequent event must be a polysymptomatic presentation, including encephalopathy, with neurologicsymptoms or signs that differ from the initial event (mental status changes may not differ from theinitial event)

    Brain MRI shows new areas of involvement and complete or partial resolution of prior ADEM-associatedlesion

    a Modified with permission from Krupp LB, et al; International Pediatric MS Study Group, Neurology.2 B 2007, American Academyof Neurology. www.neurology.org/content/68/16_suppl_2/S7.short?sid=d000124e-c5a6-4744-a896-c2c5cfcc4545.

    946 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • not uncommon and is more com-monly observed in ADEM than in MS(Figure 3-2).4 Lesions may also bepresent in the optic nerves and spinalcord. A key exclusionary criterion islack of clinical or radiologic evidenceof prior CNS pathology that wouldindicate previous inflammatory de-myelination. Typically, CSF pleocytosisis present.

    ADEM is more common in childrenthan adults, and diagnosis can bemade with greater confidence whenit follows an acute infectious illness orvaccination in a child or an adult. Abroad differential diagnosis of acute

    leukoencephalopathies exists that mayproduce similar features (Table 3-2).5

    Clinical and radiologic clues are im-portant in distinguishing ADEM fromother mimics. Investigations should betargeted to specific suspected compet-ing diagnoses based on demographic,neurologic, and non-neurologic symp-tomatology and radiologic clues.

    The diagnosis of ADEM remainsone of exclusion of other competingdiagnoses and monitoring of treat-ment response and clinical courseVinparticular, whether remission occursspontaneously or after corticosteroidtreatment and whether relapse occurs.

    Case 3-1A 30-year-old woman with no antecedent illness or vaccination presented with headaches andmigratory numbness. Four days later she reported nausea and was confused, asking the samequestion repetitively. She developed gaitunsteadiness, followed within a week byparaplegia and later by paresis of herarms. Examination revealed that she wassomnolent but opened her eyes aftervigorous stimulation. She had bilateralpapilledema, moderate upper extremityparesis, and plegia of the lower extremities.Bilateral Babinski signs and spasticity werepresent. MRI scan of the brain showedextensive white matter lesions (Figure 3-1).CSF analysis revealed 30 white blood cells(WBC)/2L, primarily mononuclear; elevatedprotein level of 200 mg/dL, and negative/normal results for IgG index and oligoclonalbands. In spite of high-dose corticosteroids,she continued to deteriorate and developedmultiple new lesions of the cerebralhemispheres and brainstem, as wellas a herniation syndrome. Pathologicanalysis revealed evidence of perivenousdemyelination consistent with acutedisseminated encephalomyelitis (ADEM).

    Comment. This case illustrates howfulminant a course ADEM may take in certainindividuals and how it may cause cerebralherniation. FIGURE 3-1 T2-weighted axial brain image shows

    multiple lesions of varying size scatteredthroughout the subcortical and deep cerebralwhite matter.

    947Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • Relapse is a strong indicator that adiagnosis of ADEM was incorrect, al-though, as noted above, pediatric con-sensus criteria do not consider arelapse as a strict exclusionary charac-teristic. The radiologic characteristicsof ADEM are nonspecific. In the major-ity of patients, they are difficult todistinguish from those seen in MS,although extensive and symmetric ce-

    rebral, cerebellar, and basal gangliaabnormalities have been reported. Se-rial follow-up in ADEM does not revealnew MRI findings, but this observationcan only be made retrospectively. CSFfindings are similarly nonspecific. Ele-vations of IgG index or the presenceof oligoclonal bands are usually absent,or when present, are transient. WhenADEM is strongly suspected and no

    FIGURE 3-2 Acute disseminated encephalomyelitis in a child. Fluid-attenuated inversion recovery (FLAIR) axialimages at the level of basal ganglia and pons (A, C) and T1-gadolinium axial images at the samelevels (B, D). Note the mostly symmetrical, thalamic gray matter T2 hyperintensity with minimal

    gadolinium enhancement as well as the bilateral, though asymmetrical, middle cerebellar peduncle T2hyperintensity with prominent gadolinium enhancement.

    948 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • other competing diagnosis seemslikely or is detected after a comprehen-sive evaluation, a therapeutic trial ofcorticosteroids should be administered.Generally, the course of corticosteroidsshould be initiated within a short time,typically within a few days, becausepatients with ADEM are often very ill.In patients with clinical deterioration inspite of treatment with high-dose corti-costeroids, a brain biopsy should beconsidered, which can evaluate for bothADEM and other diagnoses.

    The hallmark pathologic character-istic for ADEM is perivenous demye-lination, typically distributed in sleevessurrounding areas of perivascular in-flammation (Figure 3-3). This patho-logic picture is distinct from MS,which features confluent demyelination,prominent and confluent macrophageinfiltration, and reactive astrocytes.1

    Recently, a distinctive microglial acti-vation and aggregation without corti-cal demyelination similar to that seenin MS has been observed in patients

    TABLE 3-2 Selected Acute Leukoencephalopathies That Mimic Acute DisseminatedEncephalomyelitis and Investigations

    Condition Index of Suspicion Initial TestsConfirmatoryTests

    Neurosarcoidosis Extra-CNS disease Serum and CSF angiotensin-converting enzyme level

    CNS or non-CNSbiopsyPersisting enhancement

    on brain MRI Brain and spinal cord MRIs withand without contrast

    Imaging for pulmonary disease

    Vasculitis Extra-CNS disease Sedimentation rate CNS or non-CNSbiopsyResidual infarcts Vasculitis panel (myeloperoxidase,

    proteinase 3)

    Angiography

    Progressive multifocalleukoencephalopathy

    Immunosuppressedindividual

    CSF PCR for JC virus CNS biopsy

    Natalizumab treatmentMRI sometimes sequential

    HIV infection

    Gliomatosis cerebri History of brain tumor Brain MRI with and withoutgadolinium

    CNS biopsy

    CNS lymphoma Prior immunosuppression CSF cytology CNS biopsy

    Homogeneous, nodularenhancement

    Multifocal tumor

    Posterior reversibleencephalopathysyndrome

    Risk factor (eg,hypertension, cyclosporine)

    Brain MRI with and withoutgadolinium

    Follow-up MRI

    No gadoliniumenhancement

    Imaging may change rapidly

    Indistinct boundaries

    Paraneoplastic disorder History of cancer Paraneoplastic antibody CNS biopsy

    Erdheim-Chester disease Bone pain X-ray of long bones CNS or non-CNSbiopsyExophthalmos

    Diabetes insipidus

    949Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • with ADEM, which was highly associ-ated with coma.1

    The prognosis of ADEM is highlyvariable and depends in large part onthe accuracy of the diagnosis, whichhas been poor in the existing literaturebecause of the lack of a specific di-agnostic test and the rarity with whichbiopsy is obtained to differentiate fromfulminant MS or other acute leuko-encephalopathies. Therefore, firmguidance for patients and their familiesregarding prognosis is difficult. Ingeneral, the outcome in definite ADEMis favorable if the course is not compli-cated by supervening complications(eg, sequelae of status epilepticus),because demyelination and acute axo-nal pathology is usually less severe thanother fulminant inflammatory disor-ders such as Marburg variant MS (anacute, monophasic form of MS thattypically leads to death from a hernia-tion syndrome or brainstem dysfunc-tion within days or weeks from onset).

    No randomized clinical trials ofADEM have been conducted. Thestandard of treatment based on em-pirical studies is high-dose IV cortico-steroids, typically administered at 1 g/d(in children 10Y30 mg/kg/d) for 5 daysfollowed by a taper of oral prednisoneover 3 to 6 weeks. Plasma exchange iseffective in acute, severe demyelinating

    diseases of a variety of types, althoughvery limited experience is available forADEM. IV immunoglobulin and otherimmunosuppressive agents similarlyhave been supported in case reportsand small series.6,7

    TRANSVERSE MYELITISThe term transverse myelitis de-scribes a heterogeneous collection ofacute and subacute infectious andnoninfectious inflammatory spinalcord syndromes. Cases of myelitiswere first described in the 19th cen-tury. Inflammatory demyelination wasrecognized as the underlying pathol-ogy in fatal postvaccinal encephalomy-elitis (smallpox and rabies vaccines) inthe 1920s. The annual incidence ofpostinfectious or idiopathic forms oftransverse myelitis is estimated to be1.3 to 8 cases per million population.8

    When MS myelitis is included, theannual incidence approaches 25 permillion. Modern diagnostic neuroim-aging, CSF analysis, and laboratorytechniques enable a specific diagnosisand prognosis in most cases of trans-verse myelitis.

    The consensus inclusion criteriafor diagnosis of idiopathic transversemyelitis is outlined in Table 3-3 andincludes the clinical features commonto all potential etiologies of myelitis.9

    KEY POINT

    h Acute disseminatedencephalomyelitis isassociated withperivenous sleeves ofdemyelination andperivascular inflammation,rather than confluentdemyelination andmacrophages andreactive astrocytes, as inacute multiple sclerosis.

    FIGURE 3-3 Representative brain biopsies illustrating the observed patterns of demyelination that may be found in patientswith clinically defined acute disseminated encephalomyelitis. A, Perivenous sleeve of inflammation anddemyelination (20); B, Three coalescing perivenous lesions (60); C, Extensive region of confluentdemyelination with areas of perivenous demyelination in the periplaque white matter (4). Luxol fast bluemyelin stain and periodic acidYSchiff counterstain was used for all three images.

    Reprinted with permission from Young NP, et al Brain.1 B 2010, Guarantors of Brain. brain.oxfordjournals.org/content/133/2/333.long.

    950 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • Most patients present with a combina-tion of sensory, motor, and bladder-or bowel-related symptoms suggestiveof myelopathy. Lhermitte sign (anelectrical or dysesthetic sensation inthe spine or limbs, elicited by neckflexion) and paroxysmal tonic spasms(repeated, brief [30- to 90-second],stereotypic attacks of painful limb ortruncal muscle spasms, with or withoutsensory symptoms, often triggered bylimb movement) are hallmarks of de-myelinating disease and suggest thatthe myelitis syndrome may be causedby MS or NMO.

    Transverse myelitis is classified clin-ically based on whether it is completeor incomplete, which may assist withdifferential diagnosis.10 A completecord lesion, which manifests as arelatively symmetric moderate or se-vere loss of motor and sensory mo-dalities caudal to the level of thelesion, suggests a monophasic disorder(infectious, parainfectious, or idiopath-ic transverse myelitis) or relapsingNMO. In contrast, a partial myelitissyndrome (ie, incomplete or patchyinvolvement of at least one spinalsegment with mild to moderate weak-

    ness and asymmetric or dissociatedsensory symptoms) is more likely toherald MS with high risk for futurerelapses.11 Nonetheless, the clinicalpresentation may not be concordantwith neuroimaging findings, which arelikely more reliable than clinical man-ifestations in determining the underly-ing etiology. MS-related myelitis isusually characterized by short-segment,peripheral cord lesions, whereas NMOis strongly associated with a longitudi-nally extensive transverse myelitis(LETM) lesion that extends contiguouslyover three or more vertebral segments(seen best on sagittal T2-weighted MRI)and usually affects the central cord(Figure 3-4).12 However, other diseases(including infarction, viral and otherinfections, syringomyelia, and sarcoid-osis) can cause a long cord lesion inthe central cord that mimics LETM.13

    The differential diagnosis of trans-verse myelitis is extensive and summa-rized in Table 3-4. Figure 3-5 outlines asystematic diagnostic approach, whichfocuses on establishing that the mye-lopathy is inflammatory and thenidentifying a causative agent or under-lying disease (Case 3-2).

    KEY POINTS

    h Transverse myelitis isclassified clinicallybased on whether it iscomplete or incomplete,which may assist withdifferential diagnosis.

    h Neuromyelitis opticais strongly associatedwith a longitudinallyextensive transversemyelitis lesion.

    TABLE 3-3 Diagnostic Criteria for Transverse Myelitisa

    b Development of sensory, motor, or autonomic dysfunction attributable to thespinal cord.

    b Bilateral signs and/or symptoms (although not necessarily symmetric).

    b Clearly defined sensory level.

    b Exclusion of extra-axial compressive etiology by MRI.

    b Inflammation within the spinal cord demonstrated by CSF pleocytosis orelevated IgG index or gadolinium enhancement. If none of the inflammatorycriteria is met at symptom onset, repeat MRI and lumbar puncture evaluationbetween 2 and 7 days later.

    b Progression to nadir between 4 hours and 21 days following the onset ofsymptoms.

    a Modified with permission from Transverse Myelitis Consortium Working Group, Neurology.9

    B 2002, American Academy of Neurology. www.neurology.org/content/59/4/499.short?sid=68d5cae0-275d-41e5-9a2b-9834280d8f5d.

    951Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • TABLE 3-4 Concise Differential Diagnosis and Evaluationfor Transverse Myelitisa

    Etiology Diagnostic Tests

    Infection Blood serology

    CSF culture, serologies, PCR

    Chest radiograph; other imaging as indicated

    Systemic autoimmune/inflammatory disease

    Clinical examination findings

    Serologic studies

    Chest and joint radiographs

    Other tests and/or imaging directed by historyand examination

    Paraneoplastic Chest radiography, CT, and/or body imaging

    Comprehensive serum and CSF paraneoplasticantibody panel

    Acquired CNS demyelinatingdisease (multiple sclerosis/neuromyelitis optica)

    Brain MRI with gadolinium

    CSF examination for cell count/differential,oligoclonal bands, and IgG index

    Visual evoked potentials

    Postinfectious orpostvaccinal

    Clear, recent history of infection or vaccination

    Serologic confirmation of recent infection

    Exclusion of other causesa Data from Frohman EM, Wingerchuk DM, N Engl J Med.10 www.nejm.org/doi/full/10.1056/NEJMcp1001112.

    FIGURE 3-4 Spinal cord MRI in multiple sclerosis (MS) and neuromyelitis optica. A, Sagittal T2-weighted MRI of the cervicalspinal cord demonstrates typical dorsal, short-segment signal abnormalities (arrows) characteristicof MS. B, In contrast, patients with acute myelitis in the setting of neuromyelitis optica typically have

    longitudinally extensive, expansile, centrally located cord lesions that may extend into the brainstem (arrows) (sagittalT2-weighted cervical spinal cord MRI). C, On T1-weighted sagittal MRI sequences, such acute lesions may be hypointense(arrows), suggesting necrosis and cavitation, while exhibiting enhancement with IV gadolinium administration (arrowheads),indicative of active inflammation.

    Reprinted from Wingerchuk DM, et al, Lancet Neurol.12 B 2007, with permission from Elsevier. www.thelancet.com/journals/laneur/article/PIIS1474-4422(07)70216-8/fulltext.

    952 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • FIGURE 3-5 Diagnostic algorithm for acute myelopathies and myelitis. A systematic approach to the evaluation ofacute myelopathy syndromes allows for diagnosis of transverse myelitis and its etiology.

    MS = multiple sclerosis; NMO-IgG = neuromyelitis opticaYimmunoglobulin G;TM = transverse myelitis.

    Reprinted from Frohman EM, Wingerchuk DM, N Engl J Med.10 B 2010, with permission from Massachusetts MedicalSoceity. www.nejm.org/doi/full/10.1056/NEJMcp1001112.

    953Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • Neuroimaging characteristics arecritical for diagnosis. Identification ofan intramedullary cord lesion, espe-cially if it exhibits postgadoliniumenhancement, is very helpful in sup-

    porting a diagnosis of myelitis. Asdiscussed above, the lesion pattern(short peripheral lesions versus longi-tudinally extensive central lesions) of-fers substantial diagnostic guidance. In

    Case 3-2A 52-year-old woman with a history of vitiligo and hypothyroidism experiencedprogressive truncal and bilateral lower extremity numbness over 9 days. She lostthe ability to walk and developed urinary retention at day 14. Examinationrevealed moderately severe paraparesis, a T8 sensory level, and repetitive,involuntary, painful right lower extremity spasms each lasting 30 to 40 seconds.MRI (Figure 3-6) revealed an active longitudinally extensive transverse myelitis(LETM) lesion. Brain MRI was normal, CSF showed a lymphocytic pleocytosis(44 white blood cells/2L) but normal IgG index and no oligoclonal bands.Laboratory testing revealed vitamin B12 deficiency (level 169 ng/L) with positiveparietal cell antibodies, positive antinuclear antibody (1:160), and positiveneuromyelitis optica (NMO)YIgG (aquaporin-4 antibody). Her spasms resolvedwithin 6 hours of receiving 200 mg oral carbamazepine. She made a completeneurologic recovery after corticosteroid infusions and vitamin B12 replacement.She was also treated with mycophenolate mofetil with no evidence of recurrentCNS disease 3 years later.

    Comment. This case illustrates a first-ever attack of LETM; aquaporin-4seropositivity denotes an NMO spectrum disorder with high risk for relapse andneed for immunosuppression. The paroxysmal tonic spasms are common inLETM and usually respond well to carbamazepine therapy. The vitamin B12deficiency and positive antinuclear antibody reflect coexisting systemicautoimmunity in the context of NMO.

    FIGURE 3-6 Sagittal thoracic spine MRI shows a longitudinally extensive transverse myelitislesion (A, T2-weighted sequence) and gadolinium enhancement (B, T1-weightedsequence).

    954 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • the setting of partial myelitis, detectionof brain MRI lesions characteristic forMS (eg, periventricular, juxtacortical,or posterior fossa lesions) stronglypredicts future conversion to MS. Re-vised MS diagnostic criteria permit adiagnosis of MS in the presence of acombination of gadolinium-enhancingand nonenhancing white matter le-sions.14 Detection of CSF pleocytosissupports an inflammatory etiology,and other tests on CSF, includingserology and PCR, may establish aspecific inflammatory or infectiouscause. CSF oligoclonal bands are anindependent predictive factor for laterconversion to MS.15 No clinical, bio-logical, or MRI factor at onset ispredictive of long-term disability. Labo-ratory studies for viral, rickettsial, andother infectious diseases; serologic test-ing; and selected cultures obtained atpresentation may be informative underspecific circumstances. Autoimmuneserology should be obtained, especiallyto detect aquaporin-4 (AQP4) anti-bodies (also known as NMO-IgG)associated with NMO.11 Some non-infectious inflammatory disorders re-quire systemic evaluation, such as bodyimaging (eg, for sarcoidosis) or para-neoplastic antibody panel for myelitisrelated to occult malignancy. Spinalcord biopsy should be reserved forsituations in which there is concern fora neoplastic, vasculitic, or other disor-der that has evaded diagnosis and inwhich no other site presents a likelyinformative biopsy target. Attentionhas recently been drawn to the sub-stantial risk of further neurologic defi-cit complicating spinal cord biopsy thatcan be eliminated by a positive testfor AQP4 antibodies in many patientswith NMO.16

    Up to half of myelitis events arepreceded or accompanied by an identi-fiable viral illness, a clinical prodromesuggestive of infection (eg, upper re-

    spiratory tract syndrome or fever), ora vaccination. Because a systemic in-fection or vaccination may trigger in-flammatory neurologic events inpatients with underlying diseases suchas MS or NMO, comprehensive diag-nostic evaluation is still indicated be-fore concluding that the infection orvaccination is the primary cause of themyelitis. A presumed diagnosis ofmonophasic parainfectious, postinfec-tious, or postvaccinal transverse myeli-tis may be appropriate when theworkup is negative, recognizing itsinherent limitations. In one-third toone-half of transverse myelitis cases, anextensive neurologic and medical inves-tigation reveals no underlying cause ordisease, and the term idiopathic trans-verse myelitis is applied. Patients inthis category generally have a low riskof either recurrent myelitis or otherdisorders such as MS or NMO.

    Treatment of transverse myelitis de-pends on its etiology.10,11 Acute treat-ment of noninfectious myelitis includesa trial of IV methylprednisolone (ie,1 g/d for 3 to 5 days), with an optionaloral prednisone taper afterwards. Se-vere attacks that do not respond wellto corticosteroids may improve with acourse of plasma exchange (five toseven exchanges over 10 to 14 days).11

    The need for plasma exchange wouldbe more common with NMO, post-infectious, or idiopathic myelitis thanwith MS. Immunomodulatory or immu-nosuppressive therapy is only indicatedif the cause of myelitis poses significantrisk of relapse. MS disease-modifyingtherapy is indicated after a single epi-sode of myelitis in which MS is thelikely cause. AQP4 antibodyYpositiveLETM is highly likely to relapse early(approximate 60% risk at 1 year), andimmunosuppressive therapy is recom-mended (see section on NMO treat-ment). Diagnosis of specific infections,connective tissue diseases, vascular

    KEY POINTS

    h Spinal cord biopsyshould be reserved forsituations in which thereis concern for a disorderrequiring pathologicdiagnosis, especially ifprogressive neurologicdysfunction continuesdespite therapy and noother site presents alikely informative biopsytarget.

    h Because a systemicinfection orvaccination maytrigger inflammatoryneurologic events inpatients with underlyingdiseases such asmultiple sclerosis orneuromyelitis optica,comprehensivediagnostic evaluationis still indicated beforeconcluding that theinfection or vaccinationis the primary cause ofthe myelitis.

    h Severe myelitis attacksthat do not respondwell to corticosteroidsmay improve with acourse of plasmaexchange.

    h Aquaporin-antibodyYpositive longitudinallyextensive transversemyelitis is highlylikely to relapse early(approximately 60%risk at 1 year), andimmunosuppressivetherapy isrecommended.

    955Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • causes, or paraneoplastic disease mayrequire interventions such as antiviralor other antimicrobial therapies, immu-nosuppression, or detection and re-moval of the underlying malignancy,as appropriate.

    The prognosis for a patient toregain function after transverse myeli-tis is highly dependent on its etiology.Partial cord syndromes seen in MS aretypically mild to moderate in severityand plateau after days to 2 to 4 weeks,and patients typically recover well,although achievement of final recov-ery may take several months. Patientswith complete cord syndromes mayalso recover, but these syndromes aremore likely to result in substantialresidual neurologic deficits. Overall,50% to 70% of patients achieve at leastpartial recovery and ability to walk withor without assistance. Neurologicfollow-up is required, particularly forthose patients judged to be at high riskfor recurrent CNS disease.

    NEUROMYELITIS OPTICAHistorical PerspectiveNMO was originally recognized as aclinical syndrome characterized bybilateral optic neuritis and severemyelitis that occurred in quick succes-sion. Although recognized by severalauthors in the late 19th century,Devics description of a single casewith neuropathologic analysis and athesis by his student, Gault, capturedthe most attention and led Acchiotteto propose the eponym Devic dis-ease.17,18 The similarities betweenMS and NMO have been recognized,and the potential differences were amatter of debate until the 21st cen-tury. In general, most pre-1990 reportsemphasized the nonrelapsing natureof NMO and its tendency to sparethe brain as key features differentiat-ing it from MS. Around the same time,reports from Asia, especially Japan,described a condition called optico-spinal MS, which had a higher fre-quency compared to classic MS.Opticospinal MS, like its conven-tional MS counterpart, was describedas a relapsing disorder, but was differ-entiated from MS by virtue of itsfrequent and severe attacks specifical-ly targeting the optic nerve and spinalcord and by infrequent detection ofCSF oligoclonal bands. At that time,the key differentiating feature be-tween opticospinal MS and NMO wasthe temporal course: monophasic forNMO and relapsing for opticospinalMS. In the late 20th century, severalgroups recognized that the majority ofpatients with severe optic nerve andspinal cord selective demyelinatingdisease in western countries typicallyhad a relapsing course, blurring thedistinction between NMO and opti-cospinal MS. Diagnostic criteria forNMO were proposed, emphasizingspecific clinical, MRI, and CSF featuresthat appeared to distinguish NMO

    TABLE 3-5 Diagnostic Criteriafor NeuromyelitisOpticaa

    b Optic Neuritis

    b Acute Myelitis

    b At Least Two of the ThreeSupportive Criteria Below:

    Contiguous spinal cord MRIlesion extending over threeor more vertebral segments(reliably assessed in thecontext of an acute myelitis)

    Brain MRI not meeting diagnosticcriteria for multiple sclerosis

    Neuromyelitis opticaYIgGseropositive status

    a Adapted from Wingerchuk DM, et al,Neurology.22 B 2006, American Academyof Neurology. www.neurology.org/content/66/10/1485.short?sid= c5435b8d-ac14-439a-9754-e244754ec58f.

    956 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • from MS. A landmark observation in2004 reported a high frequency of aspecific autoantibody in both westernNMO and typical Japanese opti-cospinal MS. One year later, theNMO autoantibody was found to bespecific for AQP4.19 Later immuno-pathologic studies and passive transferexperiments provided strong evi-dence for the pathogenic nature ofthis autoantibody.20 At the presenttime, no clear distinguishing charac-teristics between Asian opticospinalMS and relapsing NMO have beendocumented.

    Definitions and ClassificationThe most widely accepted diagnosticcriteria were those proposed by in-vestigators at Mayo Clinic originallyin 199921 and then revised in 2006(Table 3-5).22 The 1999 criteriapredated the novel biomarker. Fur-thermore, the 1999 criteria provedinsufficiently rigorous to distinguishNMO from some cases of MS withnormal brain MRI scan at presentation.The revised criteria in 2006 weresimplified and modified to include anindependent criterion of AQP4 auto-antibody seropositivity, although itsdetection was not required becauseof its imperfect sensitivity. Respectinghistorical tradition and clinical obser-vations at that time, those criteriarequire both optic neuritis and trans-verse myelitis for diagnosis. At approx-imately the same time, a variety ofclinical observations emerged in pa-tients with otherwise definite NMOand in some patients who had limitedsymptoms of NMO. These observa-tions suggested that an even broaderspectrum of syndromes existed underthe NMO designation, such as intrac-table vomiting and lesions (bothsymptomatic and asymptomatic) ofthe circumventricular organs. These syn-dromes sometimes preceded other,

    more typical symptoms of NMO andwere associated with AQP4-specificautoantibodies. Patients with NMOtypically present with isolated trans-verse myelitis or unilateral optic neuri-tis. They are frequently, if not usually,seropositive at initial presentation. Toembrace these patients under a moregeneral rubric, the term neuromyelitisoptica spectrum disorders was pro-posed to comprise a group of patientswho probably had the same biologicaldisorder even though they did notsatisfy the strict 2006 criteria.12 Aninternational panel is currently deter-mining the syndromes that warrantinclusion under this rubric. Several ofthe brain syndromes seem to be unitedby their tendency to target AQP4, whichis identified either by MRI lesionsin AQP4-rich areas or selective loss ofAQP4 demonstrated on brain biopsytissue.

    NMO is currently classified as beingeither monophasic (nearly simulta-neous bilateral optic neuritis and my-elitis with no subsequent relapses) orrelapsing (usually, but not always,presenting with unilateral optic neuri-tis or myelitis, followed by relapsesthat can include either optic neuritisor myelitis, or less frequently cerebralinvolvement). These subtypes seem todiffer in important ways; the relapsingform is more commonly associatedwith AQP4 autoantibodies and alsoaffects women and older individualsmore commonly than the monophasicform.23 Some have suggested that themuch less common monophasic formmay be a limited form of ADEM.Familial cases have been reported butconstitute fewer than 5% of cases.24

    PathophysiologyNMO has been suspected to haveunique pathogenesis and to be avasculocentric disease based on im-munopathologic studies reported in

    KEY POINTS

    h The term neuromyelitisoptica spectrumdisorders includes aseries of well-describedclinical and radiologicsyndromes associatedwith aquaporin-4seropositivity.

    h After an initialclinical presentationof a neuromyelitisopticaYtypical clinicalsyndrome such asintractable vomitingand hiccup, isolatedlongitudinally extensivetransverse myelitis, oroptic neuritis, detectionof aquaporin-4autoantibodiespredicts a high risk ofsubsequent attacks.

    957Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • the early 2000s.25 The presence ofperivascular immunoglobulin and acti-vated complement suggested ahumoral-mediated injury to an antigenexpressed on or near microvessels.With the discovery of a specific autoan-tibody, advances rapidly led to theidentification of AQP4-specific anti-bodies that target extracellular domainsof this protein. This targeting is partic-ularly avid when the protein forms largeaggregates, as occurs when a certainlong AQP4 isoform called M23 isexpressed. Although no active immuni-zation model has yet been reported,numerous investigators have been ableto use different passive transfer strate-gies when the blood-brain barrier isdirectly bypassed (ie, by intracerebralinjection) or opened either by concom-itant induction of T-cellYmediated ex-perimental allergic encephalomyelitis26

    or by injection of complete Freund ad-juvant.27 Lesions produced by passivetransfer have many of the characteristicimmunopathologic findings of NMO.AQP4-specific autoantibodies are ofIgG1 isotype and are therefore believedto be dependent on T-cell help. Avariety of studies support the impor-tance of T-helper 17 (Th17) cells in thiscondition, including similar pathology(frequent neutrophilic infiltration) toTh17-polarized experimental allergicencephalomyelitis,28 elevated serumand CSF levels of interleukin 6 (IL-6)29

    (which is a potent inducer of Th17differentiation), elevated levels of IL-17in CSF of NMO patients, and thepredilection of an immunodominantresidue of AQP4 to induce a Th17-polarized response.30

    Clinical FeaturesThe dominant manifestations in mostpatients are optic neuritis and myelitis.In a recent series of seropositive pa-tients, the frequencies of specific pre-sentations were as follows: 41% optic

    neuritis, 43% myelitis, 5% brain orbrainstem presentations in isolation,4% optic neuritis and simultaneousmyelitis, and 7% mixed presentations(eg, optic neuritis and brain).31 Opticneuritis and myelitis may be difficult todistinguish from those that occur inMS, and are variable in severity. Typi-cally, the clinical manifestations aremore severe than those in MS. Trans-verse myelitis in NMO may be partial(ie, unilateral, motor, or sensory) orcomplete (ie, bilateral, motor, and sen-sory); by contrast, transverse myelitisdue to MS is rarely complete. Radio-logically, the manifestations overlap,although lesions tend to be consider-ably longer, especially in the spinalcord. A lesion length of three or morespinal segments is widely accepted toeffectively distinguish NMO from MS.Some NMO exacerbations may not beassociated with long spinal cord le-sions, particularly in patients receivingimmunosuppressive treatment.

    NonYoptic nerve and spinal cordsymptoms include the following, ingeneral order of frequency:

    1. Intractable vomiting and/or hiccuptypically associated with lesionsin the area postrema, either incontiguity with a myelitis lesion oras an isolated lesion (Case 3-3)Vthis occurs in approximately 20% to30% of cases and can be the firstmanifestation of the disease32

    2. Symptomatic forms of narcolepsy orstates of altered consciousnessassociated with hypothalamic lesionsand reduced CSF hypocretin levels33

    3. Encephalopathy associated withdiffuse white matter CNS lesionsthat may appear similar to ADEM

    Other uncommon presentations orcomplications include posterior re-versible encephalopathy syndrome(PRES) and hyponatremia during at-tacks of NMO.

    KEY POINT

    h Patients withneuromyelitis opticawith concomitantsystemic lupuserythematosus orhigh-titer antinuclearautoantibodies havebeen commonly labeledas having lupus myelitisin the past and are nowincreasingly acceptedas having concomitantneuromyelitis optica andsystemic autoimmunedisease.

    958 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

  • NMO is commonly associated withother autoimmune diseases, especiallysystemic lupus erythematosus andSjogren syndrome.34 Myastheniagravis occurs more commonly thanexpected in NMO.35 The associationwith systemic autoimmunity has led toconfusion in the past. Patients whohave NMO with concomitant systemiclupus erythematosus or high-titer anti-

    nuclear autoantibodies have been com-monly labeled as having lupus myelitisin the past and are now increasinglyaccepted as having concomitant NMOand systemic autoimmune disease.

    Differential DiagnosisThe differential diagnosis is largelythat of optic neuritis and of myelitisas most patients present with one of

    Case 3-3A 34-year-old man presented with recurrent vomiting that remained unexplained after thoroughgastroenterologic evaluation. The vomiting subsided but was shortly followed by imbalance and legweakness. A lesion was detected in his cervical spinal cord extending from the medulla through C6 onT2-weighted sequences, and focal gadolinium enhancement was evident in the medulla, as well assome subpial/leptomeningeal enhancement (Figure 3-7). He improved after corticosteroid treatment.CSF revealed 12 white blood cells/2L, absent oligoclonal bands, and a total protein level of 80 mg/dL.He had multiple recurrent events of myelitis over the ensuing months, all associated withlongitudinally extensive spinal cord lesions. Episodes of myelitis were prevented by azathioprinetreatment, but stopping azathioprine was followed by recurrent episodes of myelitis. He respondedsatisfactorily to treatment with corticosteroids after each attack. Repeated tests for aquaporin-4autoantibodies in serum and testing of CSF on one occasion yielded negative results. Chest x-rayand serum angiotensin-converting enzyme were normal.

    Comment. This case illustrates that recurrent attacks of longitudinally extensive myelitis, especiallywhen preceded by intractable vomiting (a signature syndrome of neuromyelitis optica [NMO]),may lead to a working diagnosis of NMO spectrum disorder; this is true even when there has beenno history of optic neuritis. Similarly, recurrent optic neuritis alone may be an indicator of NMOsyndrome when NMO-IgG is detected; the proportion of cases with recurrent optic neuritis that areseropositive for NMO-IgG is lower than cases of recurrent myelitis. Often NMO-IgG may be negativein cases of clinically definite NMO; depending on the specific assay used, the seronegative rate inhighly suspect cases is between 30% and 50%.

    FIGURE 3-7 Cervical spinal cord MRI. Sagittal imaging reveals a longitudinally extensive transverse myelitis lesion(A, T2-weighted sequence) associated with cord swelling (B, T1-weighted sequence) and focal gadoliniumenhancement (C, T1-weighted sequence).

    959Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

    CasillasResaltado

  • these syndromes. Once other ophthal-mic disorders such as acute glaucomaor retinal venous occlusion have beenexcluded, the major entities to be con-sidered in the differential diagnosis ofoptic neuritis are ischemic optic neu-ropathy (sometimes associated withgiant cell arteritis), toxic and metabolicamblyopia, compressive lesions of theoptic nerve, sarcoidosis, and otherinfiltrative disorders of the optic nerve.The entities to be considered in thedifferential diagnosis of acute longitu-dinally extensive myelitis are spinalcord infarction, viral myelopathies,postinfectious inflammatory transversemyelitis (which is clinically and radio-logically indistinguishable from trans-

    verse myelitis in NMO), intrinsic tumorsof the spinal cord (which rarely presentacutely), and other inflammatory andparaneoplastic disorders.

    Occasional patients present withsimultaneous optic neuritis and myeli-tis, in which case the differentialdiagnosis is more limited. Rare pa-tients have independent causes foroptic neuropathy and myelopathy syn-dromes (eg, ischemic optic neuropa-thy and a viral myelitis). Conditionsother than NMO that result in bothinflammatory optic neuropathy andmyelopathy include other CNS demy-elinating syndromes (MS, ADEM), sar-coidosis, paraneoplastic disorders (ie,syndromes associated with collapsing

    KEY POINTS

    h A broad range of brainlesions have beendescribed in associationwith neuromyelitisoptica and aquaporin-4antibodies.

    h Aquaporin-4 antibodytesting is approximately70% sensitive and morethan 90% specific forneuromyelitis optica.

    TABLE 3-6 Diagnostic Utility of Characteristics of Neuromyelitis Optica

    Characteristic Findings in Neuromyelitis OpticaRelative DiagnosticUtility

    Race/ethnicity NMO affects any race/ethnicity but accountsfor a smaller proportion (approximately 1% to 2%)of cases of CNS demyelinating disease in white patientsas compared to other groups (eg, 30% or more ofAfrican, Asian, and American Indian patients)

    ++

    Gender Predilection for women (80% in NMO versus 65%in MS)

    +

    Attack severity Optic neuritis and transverse myelitis attacks are moresevere than in MS

    ++

    Attack characteristics Paroxysmal tonic spasms ++

    Greater risk for neurogenic respiratory failure fromacute, ascending cervical transverse myelitis in NMOcompared with MS

    ++

    Attack residual Greater residual impairment than in MS attacks ++

    Brain MRI Normal or nonspecific findings not meeting MS MRIcriteria

    +++

    Hypothalamic lesions ++++

    Spinal cord MRI T2-weighted lesion extending contiguously over atleast three vertebral segments

    ++++

    CSF cell count anddifferential

    950 white blood cells/2L; neutrophil predominance +++

    CSF immunoglobulinstudies

    Negative IgG index and absence of unique oligoclonalbands

    +

    NMO-IgG Seropositive ++++

    NMO = neuromyelitis optica; MS = multiple sclerosis.

    960 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • response mediator protein-5 [CRMP-5],which can include optic neuritis andmyelitis), and rarely CNS malignancy.

    The potential overlap with systemiclupus erythematosusYassociated opticneuritis andmyelitis was discussed above,and most such cases probably reflectcoexistent NMO rather than a causativerole of systemic lupus erythematosus.

    Clinical, radiologic, and other labo-ratory characteristics that should beused in differentiating MS from NMOare outlined in Table 3-6.

    Diagnostic MethodsBeyond recognition of clinical syn-dromes associated with NMO andexclusion of competing diagnoses,the main contributors to NMO diag-nosis are brain and spinal cord MRIand serologic assessment.

    Neuroimaging. At or near clinicalonset, brain MRI in roughly 90% ofpatients is either normal or revealsonly nonspecific white matter lesionsthat do not fulfill MS criteria.21 Serialimaging reveals new asymptomaticbrain lesions in more than 60% ofpatients, but the pattern usually re-mains subcortical and nonspecific. Infact, the MRI pattern is typically notdistinguishable from those caused bysmall vessel cerebrovascular disease.36

    In the context of acute optic neuritis,increased T2 signal and gadoliniumenhancement of the affected opticnerve or chiasm may be seen andsometimes is longitudinally extensive.The discovery of AQP4 antibodies hasallowed recognition of a broader spec-trum of brain MRI lesions in thedisease (Figure 3-8). Cerebral whitematter lesions include large, confluentsubcortical lesions, sometimes withcloudlike gadolinium enhancementor transient lesions reminiscent ofPRES.37 Corpus callosum lesions tendto be block-shaped rather thanthe perpendicular Dawson fingers

    typical of MS. Diencephalic andperiaqueductal lesions are also typicalin NMO. Lesions may occur in the areapostrema within the dorsal medulla,particularly in patients with bouts ofnausea, vomiting, or hiccups.

    Detection of a LETM lesion onspinal cord T2-weighted sagittal MRIis a common and very specific neuro-imaging finding for NMO. In the acutephase, the lesion usually exhibits gad-olinium enhancement, and on preYgadolinium T1-weighted imaging mayappear hypointense. Acute cervical cordlesions sometimes ascend into the brain-stem. After several weeks or months,LETM lesions may disappear or resolveinto several small, patchy lesions thatcan bemistaken for smaller MS plaques.

    Serology. The detection of serumAQP4 antibodies is approximately 70%sensitive and more than 90% specificfor NMO.19 In the setting of opticneuritis, LETM, or both, the presenceof AQP4 antibodies provides strongevidence for the diagnosis of NMO ora NMO spectrum disorder and in-dicates high risk for recurrent opticneuritis or myelitis. The likelihood of apositive result diminishes if the pa-tient is already receiving immunosup-pressive therapy or after plasmaexchange. Antibody levels may rise inassociation with disease activity. There-fore, it is reasonable to retest a previ-ously seronegative patient during a newrelapse. Rare patients who have beenrepeatedly seronegative have had AQP4antibody detected in CSF.38

    Several immunologic assays havebeen developed for AQP4-antibodydetection. The original test was anindirect immunofluorescence assay.19

    At present, the most widely availabletest is an enzyme-linked immunosor-bent assay (ELISA), and laboratoriesmay report either dichotomous results(ie, positive or negative) or a quanti-tative titer. Other assay methods

    KEY POINTS

    h Permanent neurologicdisability fromneuromyelitis optica isalmost all attack-related(secondary progressiveneuromyelitis optica israre); therefore,attack-preventionstrategies are thekey to preservationof function.

    h Standard multiplesclerosis therapiessuch as interferon-",natalizumab, orfingolimod may worsenneuromyelitis optica. If itis unclear whether apatient has neuromyelitisoptica or multiplesclerosis, animmunosuppressivetreatment strategytypically used forneuromyelitis opticashould be consideredto avoid inadvertentdisease aggravation.

    961Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

    CasillasResaltado

  • FIGURE 3-8 Brain lesions in patients with neuromyelitis optica (NMO) and NMO spectrumdisorders. Panels A, B, and C show cerebral hemispheric white matter lesions; panelsD, E, and F show diencephalic lesions; panels G, H, and I show brainstem lesions.

    Extensive bihemispheric subcortical nonenhancing white matter fluid-attenuated inversion recovery(FLAIR) signal abnormality (A). Large confluent FLAIR signal abnormality in the right parietal area (B)with diffuse gadolinium enhancement (C). Images from one patient show FLAIR abnormality in thehypothalamus (F, arrow) and right cerebral peduncle (H, arrow). FLAIR signal abnormality in thethalamus (E, arrow), hypothalamus, and optic chiasm extending into the superior cerebellarpeduncle and the floor of the fourth ventricle. Images from a patient with a confluent nonenhancingsignal abnormality from the anterosuperior thalamus-hypothalamus (D, arrow) to the optic tractsbehind the chiasm to the superior surface of the mesencephalon extending to the periaqueductalarea (right to left) to the superior cerebellar peduncles, and the pontine tegmentum (I, arrows).Extension of T2-weighted MRI signal abnormality into the medulla (G, arrow).

    Reprinted with permission from Pittock SJ, et al, Arch Neurol.36 B 2006, American Medical Association. archneur.jamanetwork.com/article.aspx?articleid=790890.

    962 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • include immunoprecipitation and cell-based assays, the latter of which in-volves cells that express human AQP4.In a blinded, direct comparison ofthese methods, all assays had strongspecificity, but the cell-based assaysdemonstrated the highest sensitivityand may eventually become the refer-ence standard.39

    Pathologic examination. The vastmajority of NMO cases can be identifiedusing clinical criteria, MRI, and serologictesting. Some perplexing cases haveundergone spinal cord or brain biopsyor come to autopsy, revealing immuno-pathologic differences between NMOand MS. In particular, NMO is asso-ciated with vasculocentric depositionof immunoglobulin and complement.

    Lesions may reveal eosinophils withina vigorous cellular infiltrate. More-over, in actively demyelinating NMOlesions, AQP4 is depleted, whereas inMS, similarly active plaques revealAQP4 upregulation (Figure 3-9).40,41

    The authors do not generally advocatebiopsy for NMO diagnosis, but in un-usual situations, such as seronegativepatients with progressive leukoen-cephalopathy or extensive, treatment-unresponsive cord lesions, biopsy maybe considered.

    PrognosisNMO is generally a more severedisease than MS. Individual attacksare more likely to result in a perma-nent neurologic deficit. In one study,

    FIGURE 3-9 Comparison of aquaporin-4 (AQP4) immunoreactivity (IR) in active neuromyelitisoptica (A, B) and multiple sclerosis (C, D) optic nerve lesions. A, Activedemyelination with macrophages containing myelin oligodendrocyte

    glycoprotein (MOG)Yimmunoreactive myelin debris (arrowheads), adjacent to periplaque whitematter (asterisk). B, AQP4 is lost in the active lesion but retained in the periplaque white matter(asterisk). C, Active demyelination with macrophages containing proteolipidprotein-immunoreactive myelin debris (arrowheads), adjacent to periplaque white matter(asterisk). D, AQP4 IR is increased in both the active lesion and periplaque white matter (asterisk).Immunohistochemistry: A, MOG; B, D, AQP4; C, proteolipid protein.

    Reprinted with permission from Roemer SF, et al, Brain.40 B 2007, Guarantors of Brain. brain.oxfordjournals.org/content/130/5/1194.long.

    963Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

    CasillasResaltado

    CasillasResaltado

  • 50% of NMO patients were blind in atleast one eye or required ambulatoryassistance within 5 years of diseaseonset.42 Unlike typical MS, in whichmost disability accrues during a sec-ondary progressive disease phase,NMO disability occurs as a result ofindividual attacks, and gradually pro-gressive disability like that seen insecondary progressive MS is rare.43

    The expansion of the NMO spectrumhas probably uncovered milder andmore heterogeneous cases. Recently,French investigators asked whetherbenign NMO exists by evaluatingoutcomes of 175 patients with NMOover 10 years.44 Although mild disabil-ity was noted in about 12% of thesepatients (compared with 22% of MSpatients over that time frame), severalof them experienced a disabling attackbefore year 15. Therefore, all NMOpatients should be considered at riskof disabling attacks. Together with therarity of a secondary progressive NMOcourse, this attack-related threat em-phasizes the importance of attack-prevention strategies to prevent disability.

    TreatmentTreatment of NMO is evolving becauseof collective longitudinal experience inlarge care centers and advances inunderstanding disease pathobiology.There are no randomized controlledtrials for NMO therapies.

    Acute attacks are typically treatedwith corticosteroids and, if necessary,rescue plasma exchange as outlined inthe discussion of transverse myelitis.Relapse prevention strategies aremeantto reduce or eliminate the effects ofpathogenic AQP4 antibodies, eitherdirectly or indirectly. The importanceof accurate diagnosis has been height-ened recently because of a number ofreports indicating that the standard MStherapy interferon-" aggravates NMO.45

    Furthermore, worsening of NMO has

    been reported with other MS therapiessuch as natalizumab and, less convinc-ingly, fingolimod.46,47 Therefore, insituations where the diagnosis remainsuncertain between NMO and MS, theauthors favor starting with an NMOimmunosuppression strategy becauseit is likely to provide benefit in treatingeither disease, and the subsequentclinical course will usually reveal thecorrect diagnosis.

    General immunosuppression strat-egies. Patients with established relaps-ing NMO and those deemed to be athigh risk for relapse (eg, patients withseropositive NMO spectrum disordersuch as those with first-ever LETM oroptic neuritis) require long-term im-munosuppressive therapy.12,48 Themost common approaches to immu-nosuppression include oral drugs (eg,azathioprine or mycophenolate mofetil)or parenteral drugs (eg, rituximab). Inretrospective series, these agents appearto reduce relapse rates by 30% to 70%,but no controlled or comparative stud-ies have confidently established themagnitude of treatment effect. Azathio-prine and mycophenolate have delayedonset of action and typically requirebridge therapy for 4 to 6months, usuallywith oral prednisone (40 to 60 mg/d).Rituximab (1000 mg/d twice, 2 weeksapart, with retreatment approximatelyevery 6 months) is fully active withinabout 2 weeks but is substantially moreexpensive. In some countries,mitoxantrone, methotrexate, or chronicoral prednisone represent the mainstayof preventive therapy; a consensus sum-mary review of the use of these therapieswas recently published.48

    Regardless of which drug is selected,the goal of therapy is elimination ofacute relapses by optimizing drug dos-age, retreatment frequencies, and com-pliance. Patient compliance can bemonitored unequivocally for rituximabbut only indirectly for azathioprine (ie,

    964 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

    CasillasResaltado

  • elevation in mean corpuscular volume).The key outcome evaluation is theoccurrence of breakthrough attacks.Unfortunately, no predictive or thera-peutic biomarkers in NMO have beenvalidated. Repeat neuroimaging willusually not yield treatment-changinginformation. AQP4 antibody titer tendsto decrease with immunosuppression,but this is not consistently associatedwith clinical course.

    Rapid advances in our understandingof immunopathogenic mechanisms inNMO are informing therapeutic strate-gies.49 A recent open-label study ofeculizumab, a monoclonal antibodythat affects cleavage of complement,showed a marked reduction in on-study attack rate and resumption ofattacks in some patients after the drugwas discontinued.50 Inference fromhuman immunopathology and animal-model data suggests potential roles fortherapies aimed at interrupting B-cell,T-cell, complement, or cytokine func-tion. Moreover, particularly interestingstrategies are now being developed tointerfere with AQP4 antibody structureor antigen binding. Aquaporumab is anonpathogenic recombinant antibodythat competes with anti-AQP4 for anti-gen binding.51 In cell cultures and ina passive-transfer animal model, aqua-porumab eliminated complement-mediated and cell-mediated cytotoxicityinduced by AQP4 antibody. Small-molecule screening has identified aseries of compounds that can alsointerfere with antibody binding. Someof these agents, including antivirals,flavonoids, and berbamine alkaloids,are available for testing.52 Selectivedeglycosylation of the AQP4 antibodyheavy chain mitigates its pathogeniceffects and may convert it to anaquaporumablike blocking antibody.53

    Antigen-specific treatment strategiesare some distance from clinical usebut illustrate rapid application of

    emerging pathophysiologic informationtoward therapy.

    REFERENCES1. Young NP, Weinshenker BG, Parisi JE, et al.

    Perivenous demyelination: associationwith clinically defined acute disseminatedencephalomyelitis and comparison withpathologically confirmed multiple sclerosis.Brain 2010;133(pt 2):333Y348.

    2. Krupp LB, Banwell B, Tenembaum S;International Pediatric MSSG. Consensusdefinitions proposed for pediatric multiplesclerosis and related disorders. Neurology2007;68(16 suppl 2):S7YS12.

    3. Kesselring J, Miller DH, Robb SA, et al.Acute disseminated encephalomyelitis.MRI findings and the distinction frommultiple sclerosis. Brain 1990;113(pt 2):291Y302.

    4. Rossi A. Imaging of acute disseminatedencephalomyelitis. Neuroimaging Clin N Am2008;18(1):149Y161.

    5. Weinshenker BG, Lucchinetti CF. Acuteleukoencephalopathies: differentialdiagnosis and investigation. Neurologist1998;4:148Y166.

    6. Menge T, Hemmer B, Nessler S, et al. Acutedisseminated encephalomyelitis: an update.Arch Neurol 2005;62(11):1673Y1680.

    7. Tenembaum S, Chitnis T, Ness J, et al. Acutedisseminated encephalomyelitis. Neurology2007;68(16 suppl 2):S23YS36.

    8. Bhat A, Naguwa S, Cheema G, Gershwin ME.The epidemiology of transverse myelitis.Autoimmun Rev 2010;9(5):A395YA399.

    9. Transverse Myelitis Consortium WorkingGroup. Proposed diagnostic criteria andnosology of acute transverse myelitis.Neurology 2002;59(4):499Y505.

    10. Frohman EM, Wingerchuk DM.Transverse myelitis. N Engl J Med 2010;363(6):564Y572.

    11. Scott TF, Frohman EM, De Seze J, et al.Evidence-based guideline: clinical evaluationand treatment of transverse myelitis.Report of the Therapeutics and TechnologyAssessment Subcommittee of the AmericanAcademy of Neurology. Neurology2011;77(24):2128Y2134.

    12. Wingerchuk DM, Lennon VA, Lucchinetti CF,et al. The spectrum of neuromyelitis optica.Lancet Neurol 2007;6(9):805Y815.

    13. Kitley JL, Leite MI, George JS, Palace JA.The differential diagnosis of longitudinally

    965Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • extensive transverse myelitis. Mult Scler2012;18(3):271Y285.

    14. Polman CH, Reingold SC, Banwell B, et al.Diagnostic criteria for multiple sclerosis:2010 revisions to the McDonald criteria.Ann Neurol 2011;69(2):292Y302.

    15. Bourre B, Zephir H, Ongagna JC, et al.Long-term follow-up of acute partialtransverse myelitis. Arch Neurol 2012;69(3):357Y362.

    16. Ringelstein M, Metz I, Ruprecht K, et al.Diagnostic biopsy of spinal tumours: needfor prior aquaporin-4 antibody testing. MSJ2012;18(S4):69Y70.

    17. Acchiote P. Sur un cas de neuromyelitesubaigue ou maladie de Devic. Bulletinofficiels de la Societe de neurologie de Paris1907a;8Y9:273Y275.

    18. Miyazawa I, Fujihara K, Itoyama Y. EugeneDevic (1858Y1930). J Neurol 2002;249(3):351Y352.

    19. Lennon VA, Wingerchuk DM, Kryzer TJ,et al. A serum autoantibody marker ofneuromyelitis optica: distinction frommultiple sclerosis. Lancet 2004;364(9451):2106Y2112.

    20. Jarius S, Wildemann B. AQP4 antibodiesin neuromyelitis optica: diagnostic andpathogenetic relevance. Nat Rev Neurol2010;6(7):383Y392.

    21. Wingerchuk DM, Hogancamp WF, OBrienPC, Weinshenker BG. The clinical course ofneuromyelitis optica (Devics syndrome).Neurology 1999;53(5):1107Y1114.

    22. Wingerchuk DM, Lennon VA, Pittock SJ,et al. Revised diagnostic criteria forneuromyelitis optica. Neurology 2006;66(10):1485Y1489.

    23. Jarius S, Ruprecht K, Wildemann B, et al.Contrasting disease patterns in seropositiveand seronegative neuromyelitis optica: amulticentre study of 175 patients.J Neuroinflammation 2012;9:14.

    24. Matiello M, Kim HJ, Kim W, et al. Familialneuromyelitis optica. Neurology 2010;75(4):310Y315.

    25. Lucchinetti CFMR, McGavern D, Bruck W,et al. A role for humoral mechanisms inthe pathogenesis of Devics neuromyelitisoptica. Brain 2002;125(pt 7):1450Y1461.

    26. Bradl M, Misu T, Takahashi T, et al.Neuromyelitis optica: pathogenicity ofpatient immunoglobulin in vivo. Ann Neurol2009;66(5):630Y643.

    27. Kinoshita M, Nakatsuji Y, Kimura T, et al.Anti-aquaporin-4 antibody induces

    astrocytic cytotoxicity in the absence ofCNS antigen-specific T cells. BiochemBiophys Res Commun 2010;394(1):205Y210.

    28. Axtell RC, de Jong BA, Boniface K, et al.T helper type 1 and 17 cells determineefficacy of interferon-beta in multiplesclerosis and experimental encephalomyelitis.Nat Med 2010;16(4):406Y412.

    29. Chihara N, Aranami T, Sato W, et al.Interleukin 6 signaling promotesanti-aquaporin 4 autoantibody productionfrom plasmablasts in neuromyelitis optica.Proc Nat Acad Sci U S A 2011;108(9):3701Y3706.

    30. Varrin-Doyer M, Spencer CM,Schulze-Topphoff U, et al. Aquaporin4-specific T cells in neuromyelitis opticaexhibit a Th17 bias and recognizeClostridium ABC transporter. Ann Neurol2012;72(1):53Y64.

    31. Kitley J, Leite MI, Nakashima I, et al.Prognostic factors and disease course inaquaporin-4 antibody-positive patients withneuromyelitis optica spectrum disorder fromthe United Kingdom and Japan. Brain2012;135(pt 6):1834Y1849.

    32. Popescu BFG, Lennon VA, Parisi JE, et al.Neuromyelitis optica unique area postremalesions: nausea, vomiting, and pathogenicimplications. Neurology 2011;76(14):1229Y1237.

    33. Kanbayashi T, Shimohata T, Nakashima I,et al. Symptomatic narcolepsy in patientswith neuromyelitis optica and multiplesclerosis: new neurochemical andimmunological implications. Arch Neurol2009;66(12):1563Y1566.

    34. Pittock SJ, Lennon VA, de Seze J, et al.Neuromyelitis optica and non-organ-specificautoimmunity. Arch Neurol 2008;65(1):78Y83.

    35. Kister I, Gulati S, Boz C, et al. Neuromyelitisoptica in patients with myasthenia graviswho underwent thymectomy. Arch Neurol2006;63(6):851Y856.

    36. Pittock SJ, Lennon VA, Krecke K, et al. Brainabnormalities in neuromyelitis optica. ArchNeurol 2006;63(3):390Y396.

    37. Magana SM, Matiello M, Pittock SJ, et al.Posterior reversible encephalopathy syndromein neuromyelitis optica spectrum disorders.Neurology 2009;72(8):712Y717.

    38. Klawiter EC, Alvarez E 3rd, Xu J, et al.NMO-IgG detected in CSF in seronegativeneuromyelitis optica. Neurology 2009;72(12):1101Y1113.

    39. Waters PJ, McKeon A, Leite MI, et al.Serologic diagnosis of NMO: a multicenter

    966 www.ContinuumJournal.com August 2013

    ADEM, Transverse Myelitis, and NMO

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

  • comparison of aquaporin-4-IgG assays.Neurology 2012;78(9):665Y671.

    40. Roemer SF, Parisi JE, Lennon VA, et al.Pattern-specific loss of aquaporin-4immunoreactivity distinguishes neuromyelitisoptica from multiple sclerosis. Brain 2007;130(pt 2):1194Y1205.

    41. Misu T, Fujihara K, Kakita A, et al. Loss ofaquaporin 4 in lesions in neuromyelitisoptica: distinction from multiple sclerosis.Brain 2007;130(pt 5):1224Y1234.

    42. Wingerchuk DM, Weinshenker BG.Neuromyelitis optica: clinical predictors of arelapsing course and survival. Neurology2003;60(5):848Y853.

    43. Wingerchuk DM, Pittock SJ, Lucchinetti CF,et al. A secondary progressive clinical courseis uncommon in neuromyelitis optica.Neurology 2007;68(8):603Y605.

    44. Collongues N, Cabre P, Marignier R, et al.A benign form of neuromyelitis optica: doesit exist? Arch Neurol 2011;68(7):918Y994.

    45. Palace J, Leite MI, Nairne A, Vincent A.Interferon Beta treatment in neuromyelitisoptica: increase in relapses and aquaporin 4antibody titers. Arch Neurol 2010;67(8):1016Y1017.

    46. Kleiter I, Hellwig K, Berthele A, et al. Failureof natalizumab to prevent relapses inneuromyelitis optica. Arch Neurol 2012;69(2):239Y245.

    47. Min J-H, Kim BJ, Lee KH. Development ofextensive brain lesions following fingolimod(FTY720) treatment in a patient withneuromyelitis optica spectrum disorder.Mult Scler 2012;18(1):113Y115.

    48. Kimbough D, Fujihara K, Jacob A, et al.Treatment of neuromyelitis optica: reviewand recommendations. Mult Scler RelatDisord 2012;1(4):180Y187.

    49. Papadopoulos MC, Verkman AS. Aquaporin4 and neuromyelitis optica. Lancet Neurol2012;11(6):525Y544.

    50. Pittock SJ, McKeon A, Mandrekar JN,et al. Pilot trial of eculizumab inAQP4-IgG-positive NMO. Boston, MA:American Neurological Association,2012.

    51. Tradtrantip L, Zhang H, Saadoun S, et al.Anti-aquaporin-4 monoclonal antibodyblocker therapy for neuromyelitis optica.Ann Neurol 2012;71(3):314Y322.

    52. Tradtrantip L, Zhang H, Anderson MO,et al. Small-molecule inhibitors of NMO-IgGbinding to aquaporin-4 reduce astrocytecytotoxicity in neuromyelitis optica. FASEB J2012;26(5):2197Y2208.

    53. Tradtrantip L, Ratelade J, Zhang H,Verkman AS. Enzymatic deglycosylationconverts pathogenic neuromyelitis opticaanti-aquaporin-4 immunoglobulin G intotherapeutic antibody. Ann Neurol 2012.doi: 10.1002/ana.23741.

    967Continuum (Minneap Minn) 2013;19(4):944967 www.ContinuumJournal.com

    Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.