diagnosis and management of acute myelopathies

ORIGINAL ARTICLE

Diagnosis and Management of Acute Myelopathies

Adam I. Kaplin, MD, PhD,* Chitra Krishnan, MHS,† Deepa M. Deshpande, MS,†Carlos A. Pardo, MD,† and Douglas A. Kerr, MD, PhD†

Background: Acute myelopathies represent a heterogeneous groupof disorders with distinct etiologies, clinical and radiologic features,and prognoses. Transverse myelitis (TM) is a prototype member ofthis group in which an immune-mediated process causes neuralinjury to the spinal cord, resulting in varying degrees of weakness,sensory alterations, and autonomic dysfunction. TM may exist aspart of a multifocal CNS disease (eg, MS), multisystemic disease(eg, systemic lupus erythematosus), or as an isolated, idiopathicentity.Review Summary: In this article, we summarize recent classifica-tion and diagnostic schemes, which provide a framework for thediagnosis and management of patients with acute myelopathy. Ad-ditionally, we review the state of current knowledge about theepidemiology, natural history, immunopathogenesis, and treatmentstrategies for patients with TM.Conclusions: Our understanding of the classification, diagnosis,pathogenesis, and treatment of TM has recently begun to expanddramatically. With more rigorous criteria applied to distinguishacute myelopathies and with an emerging understanding of immu-nopathogenic events that underlie TM, it may now be possible toeffectively initiate treatments in many of these disorders. Throughthe investigation of TM, we are also gaining a broader appreciationof the mechanisms that lead to autoimmune neurologic diseases ingeneral.

Key Words: transverse myelitis, myelopathy, neuroinflammatory

(The Neurologist 2005;11: 2–18)

Although several cases of “acute myelitis” were describedas early as 1882, it was not until 1948 that Dr Suchett-

Kaye,1 an English neurologist at St. Charles Hospital inLondon, first used the term acute transverse myelitis. Dr

Suchett-Kaye used this term to designate a case of rapidlyprogressive paraparesis with a thoracic sensory level, occur-ring as a postinfectious complication of pneumonia. Severalattempts at providing diagnostic criteria for acute transversemyelitis (TM) have been made over the past half-century,culminating in the nosology established by the InternationalTransverse Myelitis Consortium Working Group in 2002.2 Itis this set of criteria for TM that will be employed in thisreview article.

TM is a rare syndrome with an incidence of between 1and 8 new cases per million people per year.3 TM is charac-terized by focal inflammation within the spinal cord, andclinical manifestations are due to resultant neural dysfunctionof motor, sensory, and autonomic pathways within and pass-ing through the inflamed area. There is often a clearly definedrostral border of sensory dysfunction and evidence of acuteinflammation demonstrated by a spinal MRI and lumbarpuncture. When the maximal level of deficit is reached,approximately 50% of patients have lost all movements oftheir legs, virtually all patients have some degree of bladderdysfunction, and 80% to 94% of patients have numbness,paresthesias, or bandlike dysesthesias.3–8 Autonomic symp-toms consist variably of increased urinary urgency, bowel orbladder incontinence, difficulty or inability to void, incom-plete evacuation or bowel constipation, and sexual dysfunc-tion.9 Like MS,10 TM is the clinical manifestation of a varietyof disorders, with distinct presentations and pathologies.11

Recently, we proposed a diagnostic and classification schemewhich has defined TM as either idiopathic or associated witha known inflammatory disease (ie, multiple sclerosis, sys-temic lupus erythematosus, Sjogren syndrome, or neurosar-coidosis).2 Patients with TM should be offered immuno-modulatory treatment such as steroids and plasmapheresis,though there is yet no consensus as to the most appropriatestrategy. Most TM patients have monophasic disease, whileup to 20% will have recurrent inflammatory episodes withinthe spinal cord (JHTMC case series).12,13

TM exists on a spectrum of neuroinflammatory CNSconditions (Table 1), characterized by abrupt neurologicdeficits associated with inflammatory cell infiltrates and de-myelination. This can occur as a single episode (eg, TM,optic neuritis �ON�, or acute disseminated encephalomyelitis

From the *Department of Psychiatry and Behavioral Sciences, Johns Hop-kins University School of Medicine, Baltimore, Maryland; and the†Department of Neurology, Johns Hopkins Transverse Myelitis Center,Baltimore, Maryland.

Reprints: Adam I. Kaplin, MD, PhD, Department of Psychiatry and Behav-ioral Sciences, Johns Hopkins University School of Medicine, Osler 320,600 N. Wolfe Street, Baltimore, MD 21287-5371. E-mail: [email protected].

Copyright © 2005 by Lippincott Williams & WilkinsISSN: 1074-7931/05/1101-0002DOI: 10.1097/01.nrl.0000149975.39201.0b

The Neurologist • Volume 11, Number 1, January 20052

�ADEM�) or as a multiphasic condition (eg, recurrent TM,recurrent ON, neuromyelitis optica �NMO�, and MS). Thepathophysiological cause of recurrence is not currentlyknown but is of obvious clinical significance. This spectrumof neuroinflammatory CNS conditions also varies based onregional involvement of the CNS, ranging from monofocalinvolvement (eg, TM involving the spinal cord and isolatedON involving the optic nerve) to multifocal involvement (eg,ADEM involving the brain and spinal cord, NMO involvingthe optic nerve and spinal cord, and MS involving anywherein the central neuraxis). What accounts for this regionalspecificity is a subject of considerable research interest, forwhich there is no current consensus explanation. Presumably,this regional specification could result from differences in-herent in CNS tissue at different sites (such as varyingthreshold for injury or distinct localization of signal trans-duction machinery or antigens) or from differential access todistinct regions of the CNS by exogenous pathogenic mech-anisms.

IMMUNOPATHOGENESIS OF TMThe pathology of acute myelopathies reflects the het-

erogeneous nature of these disorders. Few studies to datehave described the pathology of acute myelitis, and themajority of these pathologic descriptions are clinicopatholog-ical case reports.14–16 Pathologic data from autopsies andbiopsies of suspicious spinal cord lesions from patients later

found to have TM have been studied at the JHTMC (unpub-lished data). Indeed, all patients who met criteria for thediagnosis of TM and had tissue sampling of the spinal cord(biopsy or autopsy) had inflammatory changes. These patho-logic abnormalities invariably included focal infiltration bymonocytes and lymphocytes into segments of the spinal cordand perivascular spaces and astroglial and microglial activa-tion (Fig. 1). The magnitude and extension of these inflam-matory features vary and are determined by the etiologicalfactors and the temporal profile of the myelopathic changes.The presence of white matter changes, demyelination andaxonal injury is prominent in postinfectious myelitis. How-ever, involvement of the central compartment of the cord,gray matter, or neurons is also prominent in some cases, afinding that supports the view that in TM both gray and whitematter compartments may be equally affected. In some biop-sies obtained during the acute phases of myelitis, infiltrationof CD4� and CD8� lymphocytes, along with an increasedpresence of monocytes, is quite prominent. In biopsies ob-tained during subacute phases of myelopathic lesions, prom-inent monocyte and phagocytic-macrophage infiltration isobserved. In some cases, autoimmune disorders such assystemic lupus erythematosus (SLE) lead to vasculitic lesionsthat produce focal areas of spinal cord ischemia withoutprominent inflammation.17 These immunopathological obser-

TABLE 1. Spectrum of Neuroinflammatory Conditions ofCNS

CNS Conditions RegionRecurrence

(%)

Transverse myelitis Spinal cord Yes (20)Optic neuritis Optic nerve Yes (24)Neuromyelitis optica Spinal cord and optic

nerveYes (100)

Acute disseminatedencephalomyelitis

Brain and spinal cord No

Multiple sclerosis Optic nerve, brain, andspinal cord

Yes (100)

We proposed a diagnostic and classification

scheme which has defined transverse myelitis

(TM) as either idiopathic or associated with a

known inflammatory disease.FIGURE 1. Histology of transverse myelitis (TM). A, Myelinstaining of cervical spinal cord section from a patient who diedduring a subacute stage of TM. There are a few myelinatedareas left (asterisk) and foci of cystic degeneration in theanterior horns (arrow). The remaining spinal cord showschronic inflammation and demyelination (LFB/HE stain). B,Perivascular infiltration by inflammatory cells in an area ofactive inflammation in a patient with TM. C, Infiltration bymicroglial cells in an area of inflammation (HLA-Dr immuno-stain). D, High-magnification view of few myelinated fibers leftin areas of active inflammation (arrows) (LFB/HE stain).

The Neurologist • Volume 11, Number 1, January 2005 Acute Myelopathies

© 2005 Lippincott Williams & Wilkins 3

vations further confirm that TM is an immune-mediateddisorder that involves cellular reactions and perhaps humoralfactors that injure compartments of the spinal cord.

Overview of Immunopathogenesis of TMThe immunopathogenesis of disease-associated TM is

varied. For example, pathologic data confirm that many cases oflupus-associated TM are associated with a CNS vasculitis,18–20

while others may be associated with thrombotic infarction of thespinal cord.21,22 Neurosarcoid is often associated with noncase-ating granulomas within the spinal cord,23 while TM associatedwith MS often has perivascular lymphocytic cuffing and mono-nuclear cell infiltration with variable complement and antibodydeposition.24 Since these diseases have such varied (albeitpoorly understood) immunopathogenic and effector mecha-nisms, instances of disease-associated TM will not be furtherdiscussed here. Rather, the subsequent discussion will focus onfindings potentially related to idiopathic TM.

Most patients have CSF pleocytosis and blood-brainbarrier breakdown within a focal area of the spinal cord, andconventional treatments are aimed at ameliorating immuneactivation. In 30% to 60% of idiopathic TM cases, there is anantecedent respiratory, gastrointestinal or systemic ill-ness.3–6,8,11,25 In TM patients, it is likely that there is abnor-mal activation of the immune system resulting in inflamma-tion and injury within the spinal cord. Thus, an understandingof the immunopathogenesis of TM must account for abnor-mal or excessive incitement of immune activation, and effec-tor mechanisms by which immune activation leads to CNSinjury.

Putative Mechanisms of Immune ActivationPostvaccination TM

It is widely reported in neurology texts that TM is apostvaccination event, despite there being evidence of corre-lation but not causation at the present time. Several reports ofTM following vaccination have been recently published,including following an influenza26 and booster hepatitis Bvaccination.27 Autopsy evaluation of patients with postvac-cination TM revealed lymphocytic infiltration of the spinalcord with axonal loss and demyelination. The suggestion

from these studies is that a vaccination may induce anautoimmune process resulting in TM. However, it should benoted that extensive data continue to overwhelmingly showthat vaccinations are safe and are not associated with anincreased incidence of neurologic complications.26–32 There-fore, such case reports must be viewed with caution, as it isentirely possible that 2 events occurred in close proximity bychance alone or for reasons that are only incidentally relatedto the vaccination procedure.

Parainfectious TMIn 30% to 60% of the idiopathic TM cases, there is an

antecedent respiratory, GI, or systemic illness.4–6,8,11,25 Theterm parainfectious has been used to suggest that the neuro-logic injury may be associated with direct microbial infectionand injury as a result of the infection, direct microbialinfection with immune-mediated damage against the agent, orremote infection followed by a systemic response that in-duces neural injury. An expanding list of antecedent infec-tions is now recognized, including herpes viruses and Listeriamonocytogenes, though in most of these cases, causality hasnot been established.

Though in these cases the infectious agent is requiredwithin the CNS, other mechanisms of autoimmunity dis-cussed below, such as molecular mimicry and superantigen-mediated disease, require only peripheral immune activationand may account for other cases of TM.

Molecular MimicryMolecular mimicry as a mechanism to explain an in-

flammatory nervous system disorder has been best describedin Guillain-Barré syndrome (GBS). First referred to as an“acute postinfectious polyneuritis” by W. Osler in 1892, GBSis preceded in 75% of cases by an acute infection.33–36

Campylobacter jejuni infection has emerged as the mostimportant antecedent event in GBS, occurring in up to 41% ofcases.37–40 Human neural tissue contains several subtypes ofganglioside moieties within their cell walls.41,42 A character-istic component of human gangliosides, sialic acid,43 is alsofound as a surface antigen on C. jejuni within its lipopoly-saccharide (LPS) outer coat.44 Antibodies against C. jejunithat cross-react with gangliosides have been found in serumfrom patients with GBS45–47 and have been shown to bindperipheral nerves, fix complement and impair neural trans-mission in experimental conditions that mimic GBS.41,48–50

Susceptibility to the development of GBS is also dependenton host genetic factors, which are at least partly mediated byHLA alleles.43,60

Molecular mimicry in TM may be associated with thedevelopment of autoantibodies in response to an antecedentinfection. This etiology was postulated to be involved in thecase of a patient who contracted TM following infection with

In TM patients, it is likely that there is

abnormal activation of the immune system

resulting in inflammation and injury within the

spinal cord.

Kaplin et al The Neurologist • Volume 11, Number 1, January 2005

© 2005 Lippincott Williams & Wilkins4

Enterobium vermicularis (perianal pinworm) and was foundto have elevated titers of cross-reacting antibodies.61

Microbial Superantigen-Mediated InflammationAnother link between an antecedent infection and the

development of TM may be the fulminant activation oflymphocytes by microbial superantigens (SAGs). SAGs aremicrobial peptides that have a unique capacity to stimulatethe immune system and may contribute to a variety ofautoimmune diseases. The best-studied SAGs are staphylo-coccal enterotoxins A through I, toxic shock syndrome tox-in-1 and Streptococcus pyogenes exotoxin, though manyviruses encode SAGs as well.51–54 SAGs activate T-lympho-cytes in a unique manner compared with conventional anti-gens: instead of binding to the highly variable peptide grooveof the T cell receptor, SAGs interact with the more conservedV� region.55–58 Additionally, unlike conventional antigens,SAGs are capable of activating T lymphocytes in the absenceof costimulatory molecules. As a result of these differences,a single SAG may activate between 2% and 20% of circu-lating T-lymphocytes compared with 0.001% and 0.01% withconventional antigens.59–61 Stimulation of large numbers oflymphocytes may trigger autoimmune disease by activatingautoreactive T-cell clones.62,63

Humoral DerangementsEither of the above processes may result in abnormal

immune function with blurred distinction between self andnonself. The development of abnormal antibodies potentiallymay then activate other components of the immune systemand/or recruit additional cellular elements to the spinal cord.Recent studies have emphasized distinct autoantibodies inpatients with NMO64–68 and recurrent TM.12,13,69 The highprevalence of various autoantibodies seen in such patientssuggests polyclonal derangement of the immune system. Itmay also be that some autoantibodies initiate a direct andselective injury of neurons that contain antigens that cross-react with antibodies directed against infectious pathogens.

However, it may not just be autoantibodies, but highlevels of even normal circulating antibodies that have acausative role in TM. A case of TM was described in a patientwith extremely high serum and CSF antibody levels tohepatitis B surface antigen following booster immunization.70

Such circulating antibodies may form immune complexesthat deposit in focal areas of the spinal cord. Such a mecha-nism has been proposed to describe a patient with recurrentTM and high titers of hepatitis B surface antigen.71 Circulat-ing immune complexes containing HbsAg were detected inthe serum and CSF during the acute phase and the disappear-ance of these complexes following treatment correlated withfunctional recovery.

Several Japanese patients with TM were found to havemuch higher serum IgE levels than MS patients or controls (360

versus 52 versus 85 U/mL).72 Virtually all of the patients in thisstudy had specific serum IgE to household mites (Dermatopha-goides pteronyssinus or Dermatophagoides farinae), while lessthan one third of MS and control patients did. One potentialmechanism to explain the TM in such patients is the depositionof IgE with subsequent recruitment of cellular elements. Indeed,biopsy specimens of 2 TM patients with elevated total andspecific serum IgE revealed antibody deposition within thespinal cord, perivascular lymphocyte cuffing, and infiltration ofeosinophils.73 It was postulated that eosinophils, recruited to thespinal cord, degranulated and induced the neural injury in thesepatients.

Putative Mechanisms of Immune-MediatedCNS Injury

We recently have carried out a series of investigationsthat describe immune derangements in TM patients (Kaplin etal, unpublished data). We have found that interleukin 6 (IL-6)levels in the spinal fluid of TM patients were markedlyelevated compared with control patients and to MS patients.While relatively low levels of IL-6 in MS patients did notcorrelate with disability, IL-6 levels in TM patients stronglycorrelated with and were highly predictive of disability. IL-6levels in TM patients’ CSF correlated with nitric oxide (NO)metabolites, which also correlated with disability. We sug-gest, therefore, that marked up-regulation of IL-6 as a resultof immune system activation correlates with increased NOproduction and that this elevation is etiologically related totissue injury leading to clinical disability in TM.

DEFINING TM

Historical Classifications of TMAcute transverse myelopathy (which includes noninflam-

matory causes) and TM have often been used interchangeablythroughout the published literature. One report established thefollowing criteria for transverse myelopathy: bilateral spinalcord dysfunction developing over a period of � 4 weeks with awell-defined upper sensory level, no antecedent illness, andexclusion of compressive etiologies.11 Subsequently, these cri-teria were altered to include only those patients who developedmotor, sensory, and sphincter dysfunction acutely over � 14days, whereas patients with other neurologic disease or under-lying systemic diseases were excluded.5 Other authors thendefined TM as acutely developing paraparesis (no specificationof a time to maximum deficit) with bilateral sensory findings and

IL-6 levels in TM patients strongly correlated

with and were highly predictive of disability.



impaired sphincteric function, a spinal segmental level of sen-sory disturbance, a stable nonprogressive course (to distinguishfrom progressive spastic paraparesis), and no clinical or labora-tory evidence of spinal cord compression.3 Patients were ex-cluded if they had progressive spastic paraparesis, a patchysensory deficit or hemicord syndrome, syphilis, severe backtrauma, metastatic cancer, or encephalitis. To further separatediseases with distinct etiologies, suggested criteria for TM wererevised to include only those patients who progressed to maxi-mum deficit within 4 weeks and to exclude other known diseasesincluding arteriovenous malformations of the spinal cord, hu-man T-cell lymphotropic virus-1 infection, and sarcoidosis.4

With use of these criteria, cases of TM were classified asparainfectious, related to MS, spinal cord ischemia, or idio-pathic.

Most recently, acute noncompressive myelopathies wereclassified according to an etiologic scheme74: (1) those related toMS; (2) systemic disease (eg, SLE, antiphospholipid syndrome,Sjogren disease); (3) postinfectious; (4) delayed radiation my-elopathy; (5) spinal cord infarct; and (6) idiopathic myelopathy.The presence of MS or systemic disease was determined bystandard criteria,75–77 whereas parainfectious myelopathies werediagnosed on the basis of positive IgM serology or a 4-fold orgreater increase in IgG levels on 2 successive tests to a specificcandidate/infectious agent. Delayed radiation myelopathy wasdiagnosed according to clinical history, and spinal cord infarc-tion was diagnosed on the basis of appropriate clinical andimaging findings in the absence of other likely etiologies. Idio-pathic transverse myelopathy was defined in those individualsthat could not be otherwise categorized and constituted 16.5% ofthis series.

We recently proposed a set of diagnostic criteria thatserved to distinguish TM from noninflammatory myelopa-thies and to distinguish idiopathic TM from TM associatedwith multifocal CNS and multisystemic inflammatory disor-ders. These criteria are summarized in Table 2. A diagnosis ofTM requires evidence of inflammation within the spinal cord.Because spinal cord biopsy is not a practical option in theroutine evaluation of these patients, spinal MRI and CSFanalysis are the only tools currently available to determine thepresence of inflammation within the involved lesion. Gado-

linium-enhanced spinal MRI and a lumbar puncture aremandatory in the evaluation of suspected TM, and we pro-posed that abnormal gadolinium enhancement of the spinalcord or CSF pleocytosis or elevated CSF IgG index berequired for a diagnosis of TM.2 If none of the inflammatorycriteria are met at symptom onset, MRI and lumbar punctureevaluation should be repeated between 2 and 7 days follow-ing symptom onset to determine if these inflammatory criteriaare met. IgG synthesis rate is a less specific indicator of CNSinflammation than is CSF IgG index78,79 and should not beused in the diagnosis. Vascular myelopathies can be differ-entiated from TM by a progression of symptoms to maximalseverity in less than 4 hours and the lack of inflammation asdefined above. However, these criteria do not completelydistinguish vascular myelopathies from TM, since myelopa-thies associated with venous infarcts or with vascular mal-formations may be more slowly progressive and may meetthe other criteria for TM (though not an elevated IgG index).

Differentiating idiopathic TM from TM attributed to anunderlying disease is also important. Many systemic inflamma-tory disorders (eg, sarcoidosis, SLE, Behcet disease, Sjogrensyndrome) may involve the nervous system and TM may be oneof the possible presentations. Therefore, all patients presentingwith TM should be investigated for the presence of systemicinflammatory disease. Important historical information shouldbe obtained from the patient regarding the presence of rashes,night sweats, oral or genital ulcers, sicca symptoms, shortness ofbreath, pleuritic pain, or hematuria. Examination should attemptto detect the presence of uveitis or retinitis, decreased lacrima-tion or salivation, skin rash (malar, livedo reticularis, erythemanodosum), oral or genital ulcers, adenopathy, pleuritic or peri-cardial friction rub, or organomegaly. Laboratory studies shouldinclude the following: CBC with differential and smear, ANA,SS-A, SS-B, ESR, ACE, and complement. Additional laboratorytesting may be required if signs of a systemic vasculitis aredetected.

From this evaluation, it may be possible to distinguishidiopathic TM from disease-associated TM (ie, TM associ-ated with multifocal CNS disease or systemic inflammatorydisease). This distinction is important since patients at highrisk of developing MS may be evaluated more closely or maybe offered immunomodulatory treatment.80 Similarly, pa-tients with disease-associated TM may need to be closelyfollowed for recurrent systemic and neurologic complicationsand should be offered immunosuppressive treatment to de-crease the risk of recurrence.

NATURAL HISTORY OF TM

Epidemiology and Clinical Presentation of TMTM affects individuals of all ages, with bimodal peaks

between the ages of 10 and 19 years and 30 and 39 years.3–6

There are approximately 1400 new cases diagnosed in the

TABLE 2. Diagnostic Criteria for Transverse Myelitis

Diagnostic criteriaSensory, motor, or autonomic dysfunction attributable to the

spinal cordBilateral signs and/or symptomsClearly defined sensory levelInflammation defined by CSF pleocytosis or elevated IgG index

or gadolinium enhancementProgression to nadir between 4 hours and 21 days



United States per year, and approximately 34,000 peoplehave chronic morbidity from TM at any given time. Approx-imately 28% of reported TM cases are in children (JHTMCcase series). There is no sex or familial predisposition to TM.

A preceding illness including nonspecific symptomssuch as fever, nausea, and muscle pain has been reported inabout 40% of pediatric cases within 3 weeks of the onset ofthe disorder (JHTMC case series).25,81 Thirty percent of allcases of pediatric TM cases referred to an academic centerhad a history of an immunization within 1 month of the onsetof symptoms (JHTMC case series). Although a history of animmunization preceding the onset of TM is commonly re-ported, the relationship to this event is unclear because ofinsufficient data.

TM is characterized clinically by acutely or subacutelydeveloping symptoms and signs of neurologic dysfunction inmotor, sensory and autonomic nerves, and nerve tracts of thespinal cord. Weakness is described as a rapidly progressiveparaparesis starting with the legs that occasionally progressesto involve the arms as well. Flaccidity maybe noted initially,with gradually appearing pyramidal signs by the second weekof the illness. A sensory level can be documented in mostcases. The most common sensory level in adults is themidthoracic region, though children may have a higher fre-quency of cervical spinal cord involvement and a cervicalsensory level.82 Pain may occur in the back, extremities, orabdomen. Paresthesias are a common initial symptom inadults with TM but are unusual for children.83 Autonomicsymptoms consist variably of increased urinary urgency,bowel or bladder incontinence, difficulty or inability to void,incomplete evacuation, or bowel constipation.9 Also com-monly the result of sensory and autonomic nervous systeminvolvement in TM is sexual dysfunction.84,85 Genital anes-thesia from pudendal nerve involvement (S2-S4) results inimpaired sensation in men and women. Additional malesexual problems with parasympathetic (S2-S4) and sympa-thetic (T10-L2) dysfunction in TM patients include erectiledysfunction, ejaculatory disorders and difficulty reachingorgasm. Corresponding female sexual problems include re-duced lubrication and difficulty reaching orgasm.

In addition to the signs and symptoms of direct spinalcord involvement by the immune system in TM, there alsoappears to be indirect effects manifested as depression andselective cognitive impairment that are reminiscent of whathas been described in MS (unpublished observations).86 Thisdepression or cognitive impairment does not correlate signif-icantly with the patient’s degree of physical disability and canhave lethal consequences, resulting in suicide in severe casesif left untreated. In fact, in our case series depression resultingin suicide is the leading cause of mortality, accounting for60% of the deaths we have seen in our clinic (unpublishedobservations).

When the maximal level of deficit is reached, approx-imately 50% of patients have lost all movements of their legs,virtually all patients have some degree of bladder dysfunc-tion, and 80% to 94% of patients have numbness, paresthe-sias, or bandlike dysesthesias.3–8 In more than 80% of cases,patients reach their clinical nadir within 10 days of the onsetof symptoms.81 Although the temporal course may vary,neurologic function usually progressively worsens during theacute phase from between 4 and 21 days.2

A spinal MRI and lumbar puncture often show evi-dence of acute inflammation.3–5,8,11,87,88 In our case series of170 idiopathic TM cases, spinal MRI showed a cervical T2signal abnormality in 44% and a thoracic T2 signal abnor-mality in 37% of cases. Five percent of patients had multi-focal lesions and 6% showed a T1 hypointense lesion. Thiscorresponded to the following clinical sensory levels: 22%cervical, 63% thoracic, 9% lumbar, 6% sacral, and no sensorylevel in 7%. The rostral-caudal extent of the lesion rangedfrom 1 vertebral segment in many to spanning the entirespinal cord in 2 patients. In 74% of patients, the lesion alsoenhanced with gadolinium. Forty-two percent of patients hada CSF pleocytosis, with a mean WBC count of 38 � 13 cells(range 0–950 cells). Fifty percent of the patients revealed anelevated protein level (mean protein level 75 � 14 mg/dL).Table 3 lists some of the radiologic features that distinguishvarious acute myelopathies.

Monophasic Versus Recurrent TMSeventy-five percent to 90% of TM patients experience

monophasic disease and have no evidence of multisystemicor multiphasic disease. Most commonly, symptoms will stopprogressing after 2 to 3 weeks, and spinal fluid and MRIabnormalities will stabilize and then begin to resolve. Thereare several features, however, that predict recurrent disease(Table 4). Patients with multifocal lesions within the spinalcord, demyelinating lesions in the brain, oligoclonal bands inthe spinal fluid, mixed connective tissue disorder, or serumautoantibodies (most notably SS-A) are at a greater risk ofrecurrence.89 Preliminary studies suggest that patients whohave persistently abnormal CSF cytokine profiles (notablyIL-6) may also be at increased risk for recurrent TM, thoughthese findings must be validated before they are used clini-

Seventy-five percent to 90% of TM patients

experience monophasic disease and have no

evidence of multisystemic or multiphasic

disease.



cally (Kaplin et al, unpublished data). At the current time, wedo not understand the relative contribution of these factors togauge whether chronic immunomodulatory treatment is war-ranted in high-risk patients.

PrognosisSome patients with TM may experience recovery in

neurologic function, regardless of whether specific therapy isinstituted. Recovery, if it occurs, should begin within 6months, and most patients show some restoration of neuro-logic function within 8 weeks (JHTMC case series).83 Re-covery may be rapid during months 3 to 6 after symptomonset and may continue, albeit at a slower rate, for up to 2years.81,82 Longitudinal case series of TM reveal that approx-imately one third of patients recover with little to no sequelae,one third are left with moderate degree of permanent disabil-ity, and one third have severe disabilities.5,6,11,25,81 Knebuschet al81 estimated that a good outcome with normal gait, mildurinary symptoms, and minimal sensory and upper motor

neuron signs occurred in 44%. A fair outcome with mildspasticity but independent ambulation, urgency and/or con-stipation, and some sensory signs occurred in 33%, and apoor outcome with the inability to walk or severe gaitdisturbance, absence of sphincter control, and sensory deficitin 23%. The patient cohort we follow at Johns Hopkins ismore severe, with only 20% experiencing a good outcome bythose definitions, likely a reflection of referral bias to atertiary-care center. Symptoms associated with poor outcomeinclude back pain as an initial complaint, rapid progression tomaximal symptoms within hours of onset, spinal shock, andsensory disturbance up to the cervical level.83

The presence of 14-3-3 protein, a marker of neuronalinjury, in the CSF during the acute phase may also predict apoor outcome.90 Our recent studies suggest that CSF IL-6levels at acute presentation are proportional to, and highlypredictive of, long-term disability (Kaplin et al, unpublisheddata). If confirmed by future studies, the finding that IL-6

TABLE 3. Suggestive Imaging Features to Diagnose Acute Myelopathies

Imaging Features Potential Diagnosis

Blood within the spinal cord (bright and dark T1 andT2 signal)

Vascular malformation such as cavernous angioma or dural AVfistula

Flow voids within spinal cord Dural AV fistula or AVMCentral T2 signal abnormality Venous hypertensionRing-enhancing lesion Infection or tumor (but consider course of IV steroids to rule

out inflammatory process before progressing to biopsy)Acute loss of vertical intervertebral disc height and

corresponding T2 signal abnormalityConsider fibrocartilaginous embolism

Fusiform lesion extending over � 3 spinal cord segments Consider neuromyelitis optica or disease-associated TMT2 bright lesion in white matter occupying less than 2

spinal cord segments in rostral-caudal extent and lessthan 50% of the cord diameter

Consider MS

T2 spinal cord lesion adjacent to disk herniation orspondylitic ridge but lack of spinal cord compression

Consider dynamic spinal cord compression only during flexionor extension (flexion-extension x-ray to determine thepresence of abnormal spinal column mobility; MRI in flexionor extended position instead of in neutral position)

TABLE 4. Distinguishing Features Between Recurrent and Monophasic Transverse Myelitis

Characteristics Monophasic Recurrent

Spinal MRI Single T2 lesion Multiple distinct lesions or fusiform lesion extending over� 3 spinal cord segments

Brain MRI Normal T2/FLAIR abnormalitiesBlood serology Normal � 1 autoantibody (ANA, dsDNA, phospholipid, c-ANCA)SS-A Negative PositiveCSF oligoclonal bands Negative PositiveSystemic disease None Connective tissue disorderOptic nerve involvement No Likely



levels are proportional to disability in subjects prior to treat-ment could provide a much-needed biomarker to help guidethe aggressiveness of interventions employed in treating pa-tients presenting with acute TM.

CLINICAL EVALUATION AND TREATMENT OFPATIENTS WITH TM

Evaluation of Patients With AcuteMyelopathies

We recently proposed a systematic diagnostic approachfor evaluating patients with acute myelopathies.2 The algorithmis shown in Figure 2. The first priority is to rule out a compres-sive lesion. If a myelopathy is suspected based on history andphysical examination, a gadolinium-enhanced MRI of the spinal

cord should be obtained as soon as possible. If there is nostructural lesion such as epidural blood or a spinal mass, then thepresence or absence of spinal cord inflammation should bedocumented with a lumbar puncture. The absence of pleocytosiswould lead to consideration of noninflammatory causes of my-elopathy such as arteriovenous malformations, epidural lipoma-tosis, fibrocartilaginous embolism, or possibly early inflamma-tory myelopathy (ie, a false-negative CSF). In the presence of aninflammatory process (defined by gadolinium enhancement,CSF WBC pleocytosis, or elevated CSF immunoglobulin in-dex), one should determine whether there is an infectious cause.Viral polymerase chain reaction assays should be performed todetermine whether there is the presence of viral particles withinthe CNS (herpes simplex 1 and 2, varicella zoster, cytomegalo-

FIGURE 2. Acute myelopathies: a diagnostic approach.



virus, Epstein-Barr virus, and enterovirus). Detection of Lymedisease of the CNS typically is based on antibody detectionmethods (ELISA with confirmatory Western blot) and the CSF/serum index is often helpful in determining whether there is trueneuroborreliosis.91 Evidence of M. pneumoniae infection maybe determined by seroconversion, which is defined by a 4-foldincrease in titer or a single titer of � 1:128.

The next priority is to define the regional distribution ofdemyelination within the CNS, since several disorders (ie, mul-tiple sclerosis or ADEM) may present with TM as the initialmanifestation of disease or in the setting of multifocal disease. Agadolinium-enhanced brain MRI and visual evoked potentialshould be ordered to look for these entities. The absence ofmultifocal areas of demyelination would suggest the diagnosisof isolated TM and lead to appropriate treatment measures.2

TM is often misdiagnosed as acute inflammatory de-myelinating polyradiculoneuropathy (AIDP) or Guillain-Barre syndrome (GBS), because both conditions may presentwith rapidly progressive sensory and motor loss involvingprincipally the lower extremities. Table 5 illustrates keydifferential points between these 2 conditions. A pure para-plegia or paraparesis with a corresponding distribution ofsensory loss may favor TM, while GBS may present with a

gradient of motor and sensory loss involving the lowerextremities greater than the upper extremities. When weak-ness and sensory loss involves both the upper and lowerextremities equally with a distinct spinal cord level, then TMinvolving the cervical region is more likely. Pathologicallybrisk deep tendon reflexes are supportive of TM. However,patients with fulminant cases of TM that includes significantdestruction of spinal cord gray matter may present withhypotonia and have decreased or absent deep tendon reflexes.Urinary urgency or retention is a common early finding inTM and is less common in GBS. In GBS, dysesthetic pain,involvement of the upper extremity and cranial nerve 7, andabsent deep tendon reflexes involving the upper extremitiesare more common findings. An MRI of the spinal cord mayshow an area of inflammation in TM but not in GBS.Although cerebrospinal fluid findings in TM are not consis-tent and an elevated cell count may be absent, there is usuallya moderate lymphocytic pleocytosis and elevated proteinlevel. This is in contrast to the albuminocytologic dissocia-tion of the CSF seen in GBS.81

Differential Diagnoses/NoninflammatoryMyelopathies

As indicated above, the suggested diagnostic algorithmand criteria first distinguish inflammatory from noninflam-matory myelopathies. If the history and evaluation do notsuggest a systemic or a CNS inflammatory process, thenconsideration should be given to ischemic, metabolic, orstructural causes of myelopathy. Vascular myelopathy maybe fairly easy to recognize in the setting of an anterior spinalartery infarct (sudden onset of symptoms with relative pres-ervation of posterior column function). Or it may be moredifficult to recognize in the setting of a venous infarct or a

Recovery may be rapid during months 3 to 6

after symptom onset and may continue, albeit

at a slower rate, for up to 2 years.

TABLE 5. Distinguishing Features Between Guillain-Barré Syndrome and Transverse Myelitis

Characteristics Transverse Myelitis Guillain-Barré Syndrome

Motor findings Paraparesis or quadriparesis Ascending weakness LE � UE in the early stagesSensory findings Usually can diagnose a spinal cord level Ascending sensory loss LE � UE in the early

stagesAutonomic findings Early loss of bowel and bladder control Autonomic dysfunction of the cardiovascular

(CV) systemCranial nerve findings None EOM palsies or facial weaknessElectrophysiologic findings EMG/NCV findings may be normal or may

implicate the spinal cord: prolonged centralconduction on somatosensory evoked potential(SEP) latencies or missing SEP in conjunctionwith normal sensory nerve action potentials

EMG/NCV findings confined to the PNS: motorand/or sensory nerve conduction velocityreduced, distal latencies prolonged; conductionblock; reduced H reflex usually present

MRI findings Usually a focal area of increased T2 signal with orwithout gadolinium enhancement

Normal

CSF Usually, CSF pleocytosis and/or increased IgG index Usually, elevated protein in the absence of CSFpleocytosis



vascular malformation. Venous infarction may be suspectedwhen a clinical history and serologic studies are suggestive ofa prothrombotic state (deep venous thrombosis, pulmonaryembolus, livedo reticularis, antiphospholipid antibodies, fac-tor V Leiden mutation, APC resistance, or prothrombin genemutation). A vascular malformation (dural AV fistula, AVM,cavernous angioma) may be suspected if the imaging sug-gests the presence of flow voids or bleeding into the spinalcord. A dural AV fistula is most likely to occur in men olderthan 40 years old and may present with a “stuttering” orprogressive myelopathy. Patients with a dural AV fistula mayreport a postural dependence of symptoms and pain is usuallya prominent feature. Spinal angiography is the diagnosticstudy of choice to define the presence of a vascular malfor-mation. Surgical or endovascular treatment may result instabilization or clinical improvement in a substantial propor-tion of patients.92–94

Fibrocartilaginous embolism is a rare (though likelyunderreported) cause of acute myelopathy.95–98 In most re-ported cases, there has been a sudden increase in intrathoracicor intraabdominal pressure prior to the onset of symptoms,and in several autopsies, fibrocartilaginous material wasfound to have embolized to the spinal cord. The most likelyexplanation for these findings is that the nucleus pulposusherniated vertically into the vertebral body sinusoids in re-sponse to markedly elevated pressure, followed by furtherherniation through vascular channels into the spinal cordparenchyma. Fibrocartilaginous embolism should be sus-pected in a patient with a sudden onset of myelopathy thatreaches its maximal severity within hours in a patient with anantecedent elevation of intraabdominal or intrathoracic pres-sure. Imaging may show acute loss of intervertebral diskheight and vertebral body end-plate changes adjacent to anarea of T2 signal abnormality within the spinal cord.

Radiation myelopathy may develop at any time up to 15years following ionizing radiation. Pathologic studies showpreferential involvement of myelinated tissue and blood ves-sels and it is likely that cellular death of oligodendrocytes andendothelial cells contributes to the clinical disorder.99 Pa-tients may present with slowly progressive spasticity, weak-ness, hyperreflexia and urinary urgency. There is often acorresponding T2 signal abnormality that is nonenhancingand preferentially affects the more superficial spinal cordwhite matter. Though anticoagulation100,101 or hyperbaricoxygen102–104 has been proposed as a treatment option, nei-ther has been clearly shown to be effective in patients withradiation myelopathy.

Discrimination from Multiple SclerosisTM can be the presenting feature of MS. Patients who

are ultimately diagnosed with MS are more likely to haveasymmetric clinical findings, predominant sensory symptomswith relative sparing of motor systems, MR lesions extending

over fewer than 2 spinal segments, abnormal brain MRI, andoligoclonal bands in the CSF.74,87,105–108 A patient withmonofocal CNS demyelination (TM or ON) whose brainMRI shows lesions consistent with demyelination109 has an83% chance of meeting clinical criteria for MS over thesubsequent decade compared with 11% of such patients withnormal brain MRI.110

MANAGEMENT OF TM

Intravenous SteroidsIntravenous steroid treatment is often instituted for

patients with acute TM. Corticosteroids have multiple mech-anisms of action including antiinflammatory activity, immu-nosuppressive properties, and antiproliferative actions.111,112

Though there is no randomized double-blind placebo-con-trolled study that supports this approach, evidence fromrelated disorders and clinical experience support this treat-ment.113–117 Additionally, there are several small studieswhich support the administration of corticosteroids in patientswith TM.118–121 A study of 5 children with severe TM whoreceived methylprednisolone (1 g/1.73 m2/d) for 3 or 5consecutive days followed by oral prednisone for 14 daysreported beneficial effects compared with 10 historic con-trols.120 In the steroid-treated group, the median time towalking was 23 days versus 97 days, full recovery occurredin 80% versus 10%, and full motor recovery at 1 year waspresent in 100% versus 20%. No serious adverse effects fromthe steroid treatments occurred.

Other investigations have suggested that intravenoussteroid administration may not be effective in TM pa-tients.81,83,122 The most significant of these manuscripts122

compared 12 TM patients seen between 1992 and 1994 whodid not get steroids with 9 patients seen between 1995 and1997 who did. Although the authors claimed that there wasno statistically significant difference in the outcomes betweenthe groups, it is evident that the TM patients who receivedsteroids were more likely to recover, and fewer had a pooroutcome on the Barthel Index (33% versus 67%). Therefore,the available evidence suggests that intravenous steroids aresomewhat effective if given in the acute phase of TM.However, these studies did not rigorously define TM and

The available evidence suggests that

intravenous steroids are somewhat effective if

given in the acute phase of TM.



therefore likely included patients with noninflammatory my-elopathies.

At our center, we routinely offer intravenous methyl-prednisolone (1000 mg) or dexamethasone (200 mg) for 3 to5 days unless there are compelling reasons to avoid thistherapy. The decision to offer continued steroids or add a newtreatment is often based on the clinical course and MRIappearance at the end of 5 days of steroids.

Plasma Exchange (PLEX)PLEX is often initiated if a patient has moderate to

severe TM (ie, inability to walk, markedly impaired auto-nomic function, and sensory loss in the lower extremities)and exhibits little clinical improvement after instituting 5 to 7days of intravenous steroids. PLEX is believed to work inautoimmune CNS diseases through the removal of specific ornonspecific soluble factors likely to mediate, be responsiblefor, or contribute to inflammatory-mediated target organdamage. PLEX has been shown to be effective in adults withTM and other inflammatory disorders of the CNS.123–125

Predictors of good response to PLEX include early treatment(less than 20 days from symptom onset), male sex, and aclinically incomplete lesion (ie, some motor function in thelower extremities, intact or brisk reflexes).126 It is our expe-rience that PLEX may significantly improve outcomes ofpatients with severe (though incomplete) TM and who havenot significantly improved on intravenous steroids.

Other Immunomodulatory TreatmentNo controlled information currently exists regarding

the use of other treatment strategies in patients with acuteTM. Some clinicians consider pulse dose intravenous cyclo-phosphamide (500–1000 mg/m2) for patients with TM thatcontinues to progress despite intravenous steroid therapy.Cyclophosphamide, a bifunctional alkylating agent, formsreactive metabolites that cross-link DNA. This results inapoptosis of rapidly dividing immune cells and is believed tounderlie the immunosuppressive properties of this medica-tion. It is the experience at our center that some patients willrespond significantly to intravenous cyclophosphamide, andthis treatment is worthy of consideration while we await

double-blinded placebo trials. However, cyclophosphamideshould be administered under the auspices of an experiencedoncology team, and caregivers should monitor the patientcarefully for hemorrhagic cystitis and cytopenias.

CSF filtration is a new therapy, not yet available in theUnited States, in which spinal fluid is filtered for inflamma-tory factors (including cells, complement, cytokines, andantibodies) prior to being reinfused into the patient. In arandomized trial of CSF filtration versus PLEX for AIDP,CSF filtration was better tolerated and was at least as effec-tive.127 Clinical trials for CSF filtration are currently beinginitiated.

Chronic immunomodulatory therapy should be consid-ered for the small subgroup of patients with recurrent TM.Although the ideal treatment regimen is not known, weconsider a 2-year course of oral immunomodulatory treat-ment in patients with 2 or more distinct episodes of TM. Wemost commonly treat patients with azathioprine (150–200mg/d), methotrexate (15–20 mg/wk) or mycophenolate (2–3g/d), though oral cyclophosphamide (2 g/kg/d) may also beused in patients with systemic inflammatory disease. On anyof these medicines, patients must be followed for transami-nitis or leukopenias.

Long-Term ManagementMany patients with TM will require rehabilitative care

to prevent secondary complications of immobility and toimprove their functional skills. It is important to begin occu-pational and physical therapies early during the course ofrecovery to prevent the inactivity related problems of skinbreakdown and soft tissue contractures that lead to loss ofrange of motion. The principles of rehabilitation in the earlyand chronic phases after TM are summarized in Table 6.During the early recovery period, family education is essen-tial to develop a strategic plan for dealing with the challengesto independence following return to the community. Assess-ment and fitting for splints designed to passively maintain anoptimal position for limbs that cannot be actively moved is animportant part of the management at this stage.

The long-term management of TM requires attention toa number of issues. These are the residual effects of anyspinal cord injury including TM. In addition to chronicmedical problems, there are the ongoing issues of orderingthe appropriate equipment, reentry into the school for chil-dren and community, and coping with the psychologic effectsof this condition by the patients and their families.

Patients with TM should be educated about the effect ofTM on mood regulation and routinely screened for the devel-opment of symptoms consistent with clinical depression. Patientwarning signs that should prompt a complete evaluation fordepression include failure to progress with rehabilitation andself-care, worsening fixed low mood or pervasive decreasedinterest, and social and professional withdrawal. A preoccupa-

It is our experience that plasma exchange

(PLEX) may significantly improve outcomes of

patients with severe (though incomplete) TM

and who have not significantly improved on

intravenous steroids.



tion with death or suicidal thoughts constitutes a true psychiatricemergency and should lead to prompt evaluation and treatment.All patients should be educated about 3 main points related todepression. First, patients should be educated that depression inTM is similar to other neurologic symptoms patients endure,being mediated by the effects of the immune system on thebrain. Depression is remarkably prevalent in TM, occurring inup to 25% of patients at any given time, and is largely indepen-dent of the patient’s degree of physical disability. Depression isnot due to personal weakness or the inability to “cope.” Second,depression in TM can have devastating consequences; not onlycan depression worsen physical disability (such as fatigue, pain,and decreased concentration) but it can have lethal conse-quences. Suicide is the leading cause of death in TM, accountingfor 60% of the deaths in the JHTMC since its inception. Third,despite the severity of the clinical presentation of depression inmany patients with TM, these patients generally show a veryrobust response to combined aggressive psychopharmacologicand psychotherapeutic interventions. Complete symptom remis-sion is the rule rather than the exception with appropriaterecognition and treatment of TM depression.

Spasticity is often a very difficult problem to manage. Thegoal is to maintain flexibility with a stretching routine usingexercises for active stretching and a bracing program with splintsfor a prolonged stretch. These splints are commonly used at theankles, wrists, or elbows. An appropriate strengthening programfor the weaker of the spastic muscle acting on a joint and anaerobic conditioning regimen are also recommended. Theseinterventions are supported by adjunctive measures that includeantispasticity drugs (eg, diazepam, baclofen, dantrolene, tizani-dine, and tiagabine), therapeutic botulinum toxin injections, andserial casting. The therapeutic goal is to improve the function ofthe patient in performing specific activities of daily living (ie,feeding, dressing, bathing, hygiene, mobility) through improv-ing the available joint range of motion, teaching effective com-pensatory strategies, and relieving pain.

Another major area of concern is effective managementof bowel function. A high-fiber diet, adequate and timelyfluid intake, and medications to regulate bowel evacuationsare the basic components to success. Regular evaluations bymedical specialists for adjustment of the bowel program arerecommended to prevent potentially serious complications.

Bladder function is almost always at least transientlyimpaired in patients with TM. Immediately after the onset ofTM, as in the aftermath of traumatic spinal cord injury, thereis frequently a period of transient loss or depression of neuralactivity below the involved spinal cord lesion. This phenom-enon is often referred to as “spinal shock,” which lasts about3 weeks, during which there is an interruption of descendingexcitatory influence with resultant bladder flaccidity. Follow-ing this period, bladder dysfunction can be classified into 2syndromes involving either upper motor neurons (UMN) orlower motor neurons (LMN).

Sympathetic input to the bladder, which promotes urinestorage, originates at levels T10-L2 of the spinal cord andtravels via the hypogastric nerve. Afferent input to the urinarytract is provided by the sacral (S2-S4) spinal cord through thepelvic nerve. Efferent parasympathetic input to the bladder,which mediates detrusor contractions, is carried by the pelvicnerve (S2-S4). UMN bladder dysfunction results from lesionsabove S1-S2 and is characterized by reflexive emptying withbladder filling if the injury is complete and urge incontinenceif the neurologic involvement is incomplete. In addition,detrusor-sphincter dyssynergia results from impaired commu-nication between the sacral and brain stem micturition cen-ters. In the case of UMN dysfunction, anticholinergic medi-cations, �-blockers, or electric stimulation is used to restoreadequate bladder storage and drainage. LMN bladder dys-function with either direct involvement of S2-S4 or indirectinvolvement including the conus medullaris and cauda equinaresults in detrusor areflexia and requires clean intermittentself-catheterization.

TM-induced sexual dysfunction involves similar inner-vation and analogous syndromes as those found in bladderdysfunction. Spinal cord segments S2-S4 relay afferent sen-sory fibers from the genitalia via the pudendal nerves andsupply parasympathetic input via the pelvic nerves. Parasym-pathetic stimulation initiates and maintains penile erection inmen and clitoral and labial engorgement and vaginal lubri-cation in women. Sympathetic fibers from T10-L2 providethe major stimulus for ejaculation and orgasm but can alsomediate erections through mechanisms that are less wellunderstood. Reflex erections in response to tactile stimulationare parasympathetically mediated through a local sacral (S2-S4) reflex arc and tend to occur in patients with UMN lesions.Psychogenic erections are sympathetically mediated throughsympathetic pathways that exit the spinal cord at T10-T12and allow many patients with LMN lesions of the sacralreflex arc to achieve erections through psychogenic stimula-tion. Treatment of sexual dysfunction should take into ac-count baseline function before the onset of TM and begins,with adequate education and counseling about the knownphysical and neurologic changes that TM has on sexualfunctioning. Patients should be encouraged to discuss theirconcerns with their doctors, as well as their partners. Becauseof the similarities in innervation between sexual and bladderfunction, patients with UMN-mediated sexual dysfunctionshould be encouraged to empty their bladders before sexualstimulation to prevent untimely incontinence. The mainstaysof treatment of erectile dysfunction in men are inhibitors ofcGMP phosphodiesterase, type 5, which will allow most ofmen with TM to achieve adequate erections for success inintercourse through a combination of reflex and/or psycho-genic mechanisms. Although less effective in women, thesesame types of medications have been shown capable ofenhancing sexual functioning in women.



TABLE 6. Chronic Management of Patients with Transverse Myelitis

Early Rehabilitation Late Rehabilitation

GeneralConsider inpatient rehabilitation Examine for scoliosisDaily land-based and/or water-based therapy for 8–12 wk Serial flexion/extension x-ray of back to follow angleDaily weight bearing for 45–90 min Fatigue: Amantadine, Methylphenidate, Modafinil, CoQ10Bone densitometry: vitamin D, Ca Bone densitometry: vitamin D, Ca, Bisphosphonate therapyScreen for depression Screen for depression

DemoralizationIndividual and group support. Useful resource is TM

Association (TMA) (www.myelitis.org)Biannual national TM symposiums organized through the

JHTMC and TMADevelopment of problem-focused coping skills. All

members of medical care providers could contribute(neurologist, physiatrist, PT, OT, psychiatrist,psychologist, SW, etc)

Establish local TMA support group if none exists

DepressionEducation about depression as a manifestation of brain

involvement in conjunction with stressful circumstanceReferral to psychiatrist if diagnosis is in doubt, initial trials of

antidepressant treatment is unsuccessful, or if there isconcern about suicide potential

Warning about the lethality of depression (with suicidebeing one of the leading causes of death in TM)

Ensure caregiver is receiving sufficient support to preventburnout

Treatment of depression with antidepressants and talktherapy. SSRIs are common first-line agents, butconsider TCA if there is the possibility ofsimultaneously treating depression, incontinence, andneuropathic pain with a single agent.

Bladder dysfunctionIf postvoid residual is � 80 mL, consider clean

intermittent catheterizationUrodynamics study for irritative or obstructive symptoms

Anticholinergic Rx if urgency Anticholinergic drug if detrusor hyperactive; adrenergic blockerif sphincter dysfunction

Cranberry juice for urine acidification Cranberry juice/vitamin C for urine acidificationConsider sacral nerve stimulation

Bowel dysfunctionHigh-fiber diet, increased fluid intake High-fiber diet, increased fluid intakeDigital disimpaction Digital disimpactionBowel program: colace, Senokot, Dulcolax, docusate PR,

bisacodyl in water base, MiraLax, enemas PRNBowel program: colace, Senokot, Dulcolax, docusate PR,

bisacodyl in water base, MiraLax, enemas PRNWeakness

Passive and active ROM Passive and active ROMSplinting or orthoses when necessary Splinting or orthoses if needed

Continued land and water therapyAmbulation devicesDaily weightbearing for 45–90 minutesOrthopedics evaluation if joint imbalance

Pain or dysesthesiasROM exercises ROM exercisesGabapentin, carbamazepine, nortriptyline, tramadol Gabapentin, carbamazepine, nortriptyline, tramadolAvoid narcotics if possible Topical lidocaine (patch or cream)

Intrathecal baclofen or opioids

(Continued)



SPECULATIONS ON FUTURE TREATMENTSOF TM

Work over the last few years has begun to revealfundamental immune abnormalities in patients with TM andrelated neuroimmunologic disorders. The generation of auto-antibodies and the presence of abnormally elevated cytokinelevels in the spinal fluid are likely to be important immuno-pathogenic events in many patients with TM. Though TM isa heterogeneous syndrome that is associated with distinctpathologies, recent classification strategies have attempted toidentify patients with likely similar immunopathogenicevents. While current therapies are largely nonspecific, futuretherapies will be more specifically targeted to those criticalimmunopathogenic events in TM. For example, evolvingstrategies will more effectively identify autoantibodies andthe antigen to which they respond,128,129 making it possible todevelop specific targets to block the effects of these autoan-tibodies. Additionally, several strategies exist and more arecurrently being developed that specifically alter cytokineprofiles or the effects of these cytokines within the nervoussystem. However, a cautionary note exists from recent studiesexamining TNF-� modulation in patients with multiple scle-rosis or systemic rheumatologic disease: paradoxical demy-elination may be triggered by TNF-� reduction in theblood.130 These findings may suggest that secondary alter-ations in immune system function may occur in response toblockade of any single pathway and that a “cocktail ap-proach” aimed at halting multiple proinflammatory pathwaysmay be ideal.

Future research will attempt to elucidate individualpredispositions and environmental triggers to immune over-activation. A more comprehensive understanding of themechanisms of tissue insult will lead to the development ofrational therapeutics geared to intervene at various steps ofthe signal transduction pathways leading to injury. For thosepatients who have already undergone extensive neurologicinjury as a result of TM, neurorestorative treatments (perhapsinvolving stem cells) offer the best hope for meaningfulfunctional recovery.

CONCLUSIONTM is a clinical syndrome caused by focal inflamma-

tion of the spinal cord. Many cases are postinfectious and arethought to be due to a transient abnormality in the immunesystem that results in injury to a focal area of the spinal cord.Recent studies have emphasized the need to classify TMaccording to whether there is evidence of systemic disease ormultifocal CNS disease. The importance of this may be thatdistinct treatment strategies are offered to patients with dis-tinct forms of TM. Though the causes of TM remain un-known, recent advances have suggested specific cytokinederangements that likely contribute to sustained disability dueto injury of motor, sensory, or autonomic neurons within thespinal cord. Future research will attempt to define triggers forthe immune system derangements, effector mechanisms thatpropagate the abnormal immune response, and cellular injurypathways initiated by the inflammatory response within thespinal cord. Ultimately, this may allow us to identify patientsat risk for developing TM, specifically treat the injuriousaspects of the immune response, and/or offer neuroprotectivetreatments which minimize the neural injury that occurs inresponse to the inflammation.

ACKNOWLEDGMENTWe acknowledge the support and efforts of the Trans-

verse Myelitis Association (TMA) and its president SanfordSiegel. The TMA serves a critical role to the TM communityand to researchers striving to understand and treat this disor-der. We also acknowledge financial support of the KatieSandler Fund for TM research and the Claddagh Foundationto the Johns Hopkins TM Center. We thank the Noel P. RahnFellowship for support (AIK).

REFERENCES1. Suchett-Kaye AI. Acute transverse myelitis complicating pneumonia.

Lancet. 1948;255:417.2. Transverse Myelitis Consortium Working Group. Proposed diagnostic

criteria and nosology of acute transverse myelitis. Neurology. 2002;59:499–505.

3. Berman M, Feldman S, Alter M, et al. Acute transverse myelitis:

TABLE 6. (Continued)

Early Rehabilitation Late Rehabilitation

SpasticityROM exercises ROM exercisesAquatherapy AquatherapyBaclofen, tizanidine, diazepam, botulinum toxin,

tiagabineBaclofen, tizanidine, diazepam, botulinum toxin, tiagabine

Intrathecal baclofen trialSexual dysfunction

Phosphodiesterase V inhibitors Phosphodiesterase V inhibitors



incidence and etiologic considerations. Neurology. 1981;31:966–971.4. Jeffery DR, Mandler RN, Davis LE. Transverse myelitis: retrospective

analysis of 33 cases, with differentiation of cases associated with multiplesclerosis and parainfectious events. Arch Neurol. 1993;50:532–535.

5. Christensen PB, Wermuth L, Hinge HH, et al. Clinical course andlong-term prognosis of acute transverse myelopathy. Acta NeurolScand. 1990;81:431–435.

6. Altrocchi PH. Acute transverse myelopathy. Arch Neurol. 1963;9:21–29.7. Misra UK, Kalita J, Kumar S. A clinical, MRI and neurophysiological

study of acute transverse myelitis. J Neurol Sci. 1996;138:150–156.8. Lipton HL, Teasdall RD. Acute transverse myelopathy in adults: a

follow-up study. Arch Neurol. 1973;28:252–257.9. Sakakibara R, Hattori T, Yasuda K, et al. Micturition disturbance in

acute transverse myelitis. Spinal Cord. 1996;34:481–485.10. Lucchinetti CF, Brueck W, Rodriguez M, et al. Multiple sclerosis:

lessons from neuropathology. Semin Neurol. 1998;18:337–349.11. Ropper AH, Poskanzer DC. The prognosis of acute and subacute

transverse myelopathy based on early signs and symptoms. Ann Neu-rol. 1978;4:51–59.

12. Tippett DS, Fishman PS, Panitch HS. Relapsing transverse myelitis.Neurology. 1991;41:703–706.

13. Pandit L, Rao S. Recurrent myelitis. J Neurol Neurosurg Psychiatry.1996;60:336–338.

14. Nagaswami S, Kepes J, Foster DB, et al. Necrotizing myelitis: aclinico-pathologic report of two cases associated with diplococcuspneumoniae and mycoplasma pneumoniae infections. Trans Am NeurolAssoc. 1973;98:290–292.

15. Mirich DR, Kucharczyk W, Keller MA, et al. Subacute necrotizingmyelopathy: MR imaging in four pathologically proved cases. AJNRAm J Neuroradiol. 1991;12:1077–1083.

16. Katz JD, Ropper AH. Progressive necrotic myelopathy: clinical coursein 9 patients. Arch Neurol. 2000;57:355–361.

17. de Macedo DD, de Mattos JP, Borges TM. �Transverse myelopathy andsystemic lupus erythematosus: report of a case and review of theliterature�. Ar Qneuropsiquiatr. 1979;37:76–84.

18. Piper PG. Disseminated lupus erythematosus with involvement of thespinal cord. JAMA. 1953;153:215–217.

19. Adrianakos AA, Duffy J, Suzuki M, et al. Transverse myelitis insystemic lupus erythematosus: report of three cases and review of theliterature. Ann Intern Med. 1975;83:616–624.

20. Nakano I, Mannen T, Mizutani T, et al. Peripheral white matter lesionsof the spinal cord with changes in small arachnoid arteries in systemiclupus erythematosus. Clin Neuropathol. 1989;8:102–108.

21. Sinkovics JG, Gyorkey F, Thoma GW. A rapidly fatal case of systemiclupus erythematosus: structure resembling viral nucleoprotein strandsin the kidney and activities of lymphocytes in culture. Texas ReportsBiol Med. 1969;27:887–908.

22. Weil MH. Disseminated lupus erythematosus with massive hemor-rhagic manifestations and paraplegia. Lancet. 1955;75:353–360.

23. Ayala L, Barber DB, Lomba MR, et al. Intramedullary sarcoidosispresenting as incomplete paraplegia: case report and literature review.J Spinal Cord Med. 2000;23:96–99.

24. Garcia-Zozaya IA. Acute transverse myelitis in a 7-month-old boy.J Spinal Cord Med. 2001;24:114–118.

25. Paine RS, Byers RK. Transverse myelopathy in childhood. Am J DisChild. 1968;85:151–163.

26. Patja A, Paunio M, Kinnunen E, et al. Risk of Guillain-Barre syndromeafter measles-mumps-rubella vaccination. J Pediatr. 2001;138:250–254.

27. Schonberger LB, Bregman DJ, Sullivan-Bolyai JZ, et al. Guillain-Barresyndrome following vaccination in the National Influenza ImmunizationProgram, United States, 1976–1977. Am J Epidemiol. 1979;110:105–123.

28. Langmuir AD, Bregman DJ, Kurland LT, et al. An epidemiologic andclinical evaluation of Guillain-Barre syndrome reported in associationwith the administration of swine influenza vaccines. Am J Epidemiol.1984;119:841–879.

29. Merelli E, Casoni F. Prognostic factors in multiple sclerosis: role ofintercurrent infections and vaccinations against influenza and hepatitisB. Neurol Sci. 2000;21(4 suppl 2):S853–S856.

30. Ascherio A, Zhang SM, Hernan MA, et al. Hepatitis B vaccination andthe risk of multiple sclerosis. N Engl J Med. 2001;344:327–332.

31. Confavreux C, Suissa S, Saddier P, et al. Vaccinations and the risk ofrelapse in multiple sclerosis: Vaccines in Multiple Sclerosis StudyGroup. N Engl J Med. 2001;344:319–326.

32. Moriabadi NF, Niewiesk S, Kruse N, et al. Influenza vaccination inMS: absence of T-cell response against white matter proteins. Neurol-ogy. 2001;56:938–943.

33. Dowling PC, Cook SD. Role of infection in Guillain-Barre syndrome:laboratory confirmation of herpesviruses in 41 cases. Ann Neurol.1981;9(suppl):44–55.

34. Sanders EA, Peters AC, Gratana JW, et al. Guillain-Barre syndrome aftervaricella-zoster infection: report of two cases. J Neurol. 1987;234:437–439.

35. Tsukada N, Koh CS, Inoue A, et al. Demyelinating neuropathy associatedwith hepatitis B virus infection: detection of immune complexes composedof hepatitis B virus surface antigen. J Neurol Sci. 1987;77:203–216.

36. Thornton CA, Latif AS, Emmanuel JC. Guillain-Barre syndrome as-sociated with human immunodeficiency virus infection in Zimbabwe.Neurology. 1991;41:812–815.

37. Rees JH, Soudain SE, Gregson NA, et al. Campylobacter jejuni infectionand Guillain-Barre syndrome. N Engl J Med. 1995;333:1374–1379.

38. Mishu B, Ilyas AA, Koski CL, et al. Serologic evidence of previousCampylobacter jejuni infection in patients with the Guillain-Barresyndrome. Ann Intern Med. 1993;118:947–953.

39. Hariharan H, Naseema K, Kumaran C, et al. Detection of Campy-lobacter jejuni/E. coli infection in patients with Guillain-Barre syn-drome by serology and culture. New Microbiol. 1996;19:267–271.

40. Jacobs BC, Endtz H, Van der Meche FG, et al. Serum anti-GQ1b IgGantibodies recognize surface epitopes on Campylobacter jejuni frompatients with Miller Fisher syndrome. Ann Neurol. 1995;37:260–264.

41. Kusunoki S, Shiina M, Kanazawa I. Anti-Gal-C antibodies in GBSsubsequent to mycoplasma infection: evidence of molecular mimicry.Neurology. 2001;57:736–738.

42. Jacobs BC, Endtz HP, Van der Meche FG, et al. Humoral immuneresponse against Campylobacter jejuni lipopolysaccharides in Guillain-Barre and Miller Fisher syndrome. J Neuroimmunol. 1997;79:62–68.

43. Lee WM, Westrick MA, Macher BA. High-performance liquid chro-matography of long-chain neutral glycosphingolipids and gangliosides.Biochim Biophys Acta. 1982;712:498–504.

44. Moran AP, Rietschel ET, Kosunen TU, et al. Chemical characterizationof Campylobacter jejuni lipopolysaccharides containing N-acetylneura-minic acid and 2,3-diamino-2,3-dideoxy-D-glucose. J Bacteriol. 1991;173:618–626.

45. Gregson NA, Rees JH, Hughes RA. Reactivity of serum IgG anti-GM1ganglioside antibodies with the lipopolysaccharide fractions of Campy-lobacter jejuni isolates from patients with Guillain-Barre syndrome(GBS). J Neuroimmunol. 1997;73:28–36.

46. Jacobs BC, Hazenberg MP, Van Doorn PA, et al. Cross-reactiveantibodies against gangliosides and Campylobacter jejuni lipopolysac-charides in patients with Guillain-Barre or Miller Fisher syndrome.J Infect Dis. 1997;175:729–733.

47. Hao Q, Saida T, Kuroki S, et al. Antibodies to gangliosides andgalactocerebroside in patients with Guillain-Barre syndrome with pre-ceding Campylobacter jejuni and other identified infections. J Neuro-immunol. 1998;81:116–126.

48. Goodyear CS, O’Hanlon GM, Plomp JJ, et al. Monoclonal antibodiesraised against Guillain-Barre syndrome-associated Campylobacter je-juni lipopolysaccharides react with neuronal gangliosides and paralyzemuscle-nerve preparations. J Clin Invest. 1999;104:697–708.

49. Plomp JJ, Molenaar PC, O’Hanlon GM, et al. Miller Fisher anti-GQ1bantibodies: alpha-latrotoxin-like effects on motor end plates. Ann Neu-rol. 1999;45:189–199.

50. O’Hanlon GM, Paterson GJ, Veitch J, et al. Mapping immunoreactiveepitopes in the human peripheral nervous system using human mono-clonal anti-GM1 ganglioside antibodies. Acta Neuropathol (Berl).1998;95:605–616.

51. Bohach GA, Fast DJ, Nelson RD, et al. Staphylococcal and strepto-coccal pyrogenic toxins involved in toxic shock syndrome and relatedillnesses. Crit Rev Microbiol. 1990;17:251–272.

52. Bohach GA. Staphylococcal enterotoxins B and C: structural require-ments for superantigenic and entertoxigenic activities. Prep BiochemBiotechnol. 1997;27:79–110.



53. Betley MJ, Borst DW, Regassa LB. Staphylococcal enterotoxins, toxicshock syndrome toxin and streptococcal pyrogenic exotoxins: a compar-ative study of their molecular biology. Chem Immunol. 1992;55:1–35.

54. Zhang J, Vandevyver C, Stinissen P, et al. Activation and clonalexpansion of human myelin basic protein-reactive T cells by bacterialsuperantigens. J Autoimmun. 1995;8:615–632.

55. Kappler J, Kotzin B, Herron L, et al. V beta-specific stimulation ofhuman T cells by staphylococcal toxins. Science. 1989;244:811–813.

56. Hong SC, Waterbury G, Janeway CA Jr. Different superantigensinteract with distinct sites in the Vbeta domain of a single T cellreceptor. J Exp Med. 1996;183:1437–1446.

57. Webb SR, Gascoigne NR. T-cell activation by superantigens. CurrOpin Immunol. 1994;6:467–475.

58. Acha-Orbea H, MacDonald HR. Superantigens of mouse mammarytumor virus. Annu Rev Immunol. 1995;13:459–486.

59. Brocke S, Gaur A, Piercy C, et al. Induction of relapsing paralysis inexperimental autoimmune encephalomyelitis by bacterial superantigen.Nature. 1993;365:642–644.

60. Racke MK, Quigley L, Cannella B, et al. Superantigen modulation ofexperimental allergic encephalomyelitis: activation of anergy deter-mines outcome. J Immunol. 1994;152:2051–2059.

61. Brocke S, Hausmann S, Steinman L, et al. Microbial peptides andsuperantigens in the pathogenesis of autoimmune diseases of thecentral nervous system. Semin Immunol. 1998;10:57–67.

62. Kotzin BL, Leung DY, Kappler J, et al. Superantigens and theirpotential role in human disease. Adv Immunol. 1993;54:99–166.

63. Vanderlugt CL, Begolka WS, Neville KL, et al. The functional signif-icance of epitope spreading and its regulation by co-stimulatory mol-ecules. Immunol Rev. 1998;164:63–72.

64. Fukazawa T, Hamada T, Kikuchi S, et al. Antineutrophil cytoplasmicantibodies and the optic-spinal form of multiple sclerosis in Japan.J Neurol Neurosurg Psychiatry. 1996;61:203–204.

65. Leonardi A, Arata L, Farinelli M, et al. Cerebrospinal fluid andneuropathological study in Devic’s syndrome: evidence of intrathecalimmune activation. J Neurol Sci. 1987;82:281–290.

66. O’Riordan JI, Gallagher HL, Thompson AJ, et al. Clinical, CSF, andMRI findings in Devic’s neuromyelitis optica. J Neurol NeurosurgPsychiatry. 1996;60:382–387.

67. Reindl M, Linington C, Brehm U, et al. Antibodies against the myelinoligodendrocyte glycoprotein and the myelin basic protein in multiplesclerosis and other neurological diseases: a comparative study. Brain.1999;122(pt 11):2047–2056.

68. Haase CG, Schmidt S. Detection of brain-specific autoantibodies tomyelin oligodendrocyte glycoprotein, S100beta and myelin basic pro-tein in patients with Devic’s neuromyelitis optica. Neurosci Lett.2001;307:131–133.

69. Garcia-Merino A, Blasco MR. Recurrent transverse myelitis with unusuallong-standing Gd-DTPA enhancement. J Neurol. 2000;247:550–551.

70. Renard JL, Guillamo JS, Ramirez JM, et al. �Acute transverse cervicalmyelitis following hepatitis B vaccination: evolution of anti-HBs anti-bodies�. Presse Med. 1999;28:1290–1292.

71. Matsui M, Kakigi R, Watanabe S, et al. Recurrent demyelinatingtransverse myelitis in a high titer HBs-antigen carrier. J Neurol Sci.1996;139:235–237.

72. Kira J, Kawano Y, Yamasaki K, et al. Acute myelitis with hyperIgE-aemia and mite antigen specific IgE: atopic myelitis. J Neurol Neuro-surg Psychiatry. 1998;64:676–679.

73. Kikuchi H, Osoegawa M, Ochi H, et al. Spinal cord lesions of myelitiswith hyperIgEemia and mite antigen specific IgE (atopic myelitis)manifest eosinophilic inflammation. J Neurol Sci. 2001;183:73–78.

74. de Seze J, Stojkovic T, Breteau G, et al. Acute myelopathies: clinical,laboratory and outcome profiles in 79 cases. Brain. 2001;124(pt 8):1509–1521.

75. Vitali C, Bombardieri S, Moutsopoulos HM, et al. Assessment of theEuropean classification criteria for Sjogren’s syndrome in a series ofclinically defined cases: results of a prospective multicentre study: theEuropean Study Group on Diagnostic Criteria for Sjogren’s Syndrome.Ann Rheum Dis. 1996;55:116–121.

76. Lockshin MD, Sammaritano LR, Schwartzman S. Validation of theSapporo criteria for antiphospholipid syndrome. Arthritis Rheum. 2000;

43:440–443.77. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the

classification of systemic lupus erythematosus. Arthritis Rheum. 1982;25:1271–1277.

78. Rudick RA, French CA, Breton D, et al. Relative diagnostic value ofcerebrospinal fluid kappa chains in MS: comparison with other immu-noglobulin tests. Neurology. 1989;39:964–968.

79. Hung KL, Chen WC, Huang CS. Diagnostic value of cerebrospinalfluid immunoglobulin G (IgG) in pediatric neurological diseases.J Formos Med Assoc. 1991;90:1055–1059.

80. Jacobs LD, Beck RW, Simon JH, et al. Intramuscular interferon beta-1atherapy initiated during a first demyelinating event in multiple sclero-sis: CHAMPS Study Group. N Engl J Med. 2000;343:898–904.

81. Knebusch M, Strassburg HM, Reiners K. Acute transverse myelitis inchildhood: nine cases and review of the literature. Dev Med ChildNeurol. 1998;40:631–639.

82. Pidcock F, Krishnan C, Kerr DA. Acute transverse myelitis in childhood:center based analysis of 40 children. 2003. Cur Ther Neurol Dis. In press.

83. Dunne K, Hopkins IJ, Shield LK. Acute transverse myelopathy inchildhood. Dev Med Child Neurol. 1986;28:198–204.

84. Burns AS, Rivas DA, Ditunno JF. The management of neurogenicbladder and sexual dysfunction after spinal cord injury. Spine. 2001;26(24 suppl):S129–S136.

85. DasGupta R, Fowler CJ. Sexual and urological dysfunction in multiplesclerosis: better understanding and improved therapies. Curr OpinNeurol. 2002;15:271–278.

86. Patten SB, Metz LM. Depression in multiple sclerosis. PsychotherPsychosom. 1997;66:286–292.

87. Scott TF, Bhagavatula K, Snyder PJ, et al. Transverse myelitis: com-parison with spinal cord presentations of multiple sclerosis. Neurology.1998;50:429–433.

88. al Deeb SM, Yaqub BA, Bruyn GW, et al. Acute transverse myelitis:a localized form of postinfectious encephalomyelitis. Brain. 1997;120(pt 7):1115–1122.

89. Hummers LK, Krishnan C, Casciola-Rosen L, et al. Recurrent trans-verse myelitis associates with anti-Ro (SSA) autoantibodies. Neurol-ogy. 2004;62:147–149.

90. Irani DN, Kerr DA. 14-3-3 Protein in the cerebrospinal fluid of patientswith acute transverse myelitis. Lancet. 2000;355:901.

91. Wilske B. Microbiological diagnosis in Lyme borreliosis. Int J MedMicrobiol. 2002;291(suppl 33):114–119.

92. Ferch RD, Morgan MK, Sears WR. Spinal arteriovenous malformations: areview with case illustrations. J Clin Neurosci. 2001;8:299–304.

93. Moriarity JL, Clatterbuck RE, Rigamonti D. The natural history of cav-ernous malformations. Neurosurg Clin North Am. 1999;10:411–417.

94. Wityk RJ. Dural arteriovenous fistula of the spinal cord: an uncommoncause of myelopathy. Semin Neurol. 1996;16:27–32.

95. Schreck RI, Manion WL, Kambin P, et al. Nucleus pulposus pulmonaryembolism: a case report. Spine. 1995;20:2463–2466.

96. Bots GT, Wattendorff AR, Buruma OJ, et al. Acute myelopathy causedby fibrocartilaginous emboli. Neurology. 1981;31:1250–1256.

97. Toro G, Roman GC, Navarro-Roman L, et al. Natural history of spinalcord infarction caused by nucleus pulposus embolism. Spine. 1994;19:360–366.

98. Case records of the Massachusetts General Hospital: weekly clinico-pathological exercises: case 5–1991: a 61-year-old woman with anabrupt onset of paralysis of the legs and impairment of the bladder andbowel function. N Engl J Med. 1991;324:322–332.

99. Okada S, Okeda R. Pathology of radiation myelopathy. Neuropathol-ogy. 2001;21:247–265.

100. Liu CY, Yim BT, Wozniak AJ. Anticoagulation therapy for radiation-induced myelopathy. Ann Pharmacother. 2001;35:188–191.

101. Glantz MJ, Burger PC, Friedman AH, et al. Treatment of radiation-induced nervous system injury with heparin and warfarin. Neurology.1994;44:2020–2027.

102. Asamoto S, Sugiyama H, Doi H, et al. Hyperbaric oxygen (HBO)therapy for acute traumatic cervical spinal cord injury. Spinal Cord.2000;38:538–540.

103. Calabro F, Jinkins JR. MRI of radiation myelitis: a report of a casetreated with hyperbaric oxygen. Eur Radiol. 2000;10:1079–1084.



104. Angibaud G, Ducasse JL, Baille G, et al. �Potential value of hyperbaricoxygenation in the treatment of post-radiation myelopathies�. RevNeurol (Paris). 1995;151:661–666.

105. Ford B, Tampieri D, Francis G. Long-term follow-up of acute partialtransverse myelopathy. Neurology. 1992;42:250–252.

106. Miller DH, Ormerod IE, Rudge P, et al. The early risk of multiplesclerosis following isolated acute syndromes of the brainstem andspinal cord. Ann Neurol. 1989;26:635–639.

107. Ungurean A, Palfi S, Dibo G, et al. Chronic recurrent transversemyelitis or multiple sclerosis. Funct Neurol. 1996;11:209–214.

108. Bakshi R, Kinkel PR, Mechtler LL, et al. Magnetic resonance imagingfindings in 22 cases of myelitis: comparison between patients with andwithout multiple sclerosis. Eur J Neurol. 1998;5:35–48.

109. McDonald WI, Compston A, Edan G, et al. Recommended diagnosticcriteria for multiple sclerosis: guidelines from the International Panelon the diagnosis of multiple sclerosis. Ann Neurol. 2001;50:121–127.

110. O’Riordan JI, Losseff NA, Phatouros C, et al. Asymptomatic spinalcord lesions in clinically isolated optic nerve, brain stem, and spinalcord syndromes suggestive of demyelination. J Neurol NeurosurgPsychiatry. 1998;64:353–357.

111. Miller JA, Munro DD. Topical corticosteroids: clinical pharmacologyand therapeutic use. Drugs. 1980;19:119–134.

112. Hallam NF. The use and abuse of topical corticosteroids in dermatol-ogy. Scott Med J. 1980;25:287–291.

113. Elovaara I, Lalla M, Spare E, et al. Methylprednisolone reducesadhesion molecules in blood and cerebrospinal fluid in patients withMS. Neurology. 1998;51:1703–1708.

114. Sellebjerg F, Christiansen M, Jensen J, et al. Immunological effects oforal high-dose methylprednisolone in acute optic neuritis and multiplesclerosis. Eur J Neurol. 2000;7:281–289.

115. Williams CS, Butler E, Roman GC. Treatment of myelopathy inSjogren syndrome with a combination of prednisone and cyclophosph-amide. Arch Neurol. 2001;58:815–819.

116. Dumas JL, Valeyre D, Chapelon-Abric C, et al. Central nervous systemsarcoidosis: follow-up at MR imaging during steroid therapy. Radiol-ogy. 2000;214:411–420.

117. Bracken MB, Shepard MJ, Collins WF, et al. A randomized, controlledtrial of methylprednisolone or naloxone in the treatment of acute

spinal-cord injury: results of the Second National Acute Spinal CordInjury Study. N Engl J Med. 1990;322:1405–1411.

118. Defresne P, Meyer L, Tardieu M, et al. Efficacy of high dose steroidtherapy in children with severe acute transverse myelitis. J NeurolNeurosurg Psychiatry. 2001;71:272–274.

119. Lahat E, Pillar G, Ravid S, et al. Rapid recovery from transversemyelopathy in children treated with methylprednisolone. Pediatr Neu-rol. 1998;19:279–282.

120. Sebire G, Hollenberg H, Meyer L, et al. High dose methylprednisolone insevere acute transverse myelopathy. Arch Dis Child. 1997;76:167–168.

121. Kennedy PG, Weir AI. Rapid recovery of acute transverse myelitistreated with steroids. Postgrad Med J. 1988;64:384–385.

122. Kalita J, Misra UK. Is methyl prednisolone useful in acute transversemyelitis? Spinal Cord. 2001;39:471–476.

123. Weinshenker BG. Plasma exchange for severe attacks of inflammatorydemyelinating diseases of the central nervous system. J Clin Apheresis.2001;16:39–42.

124. Weinshenker BG. Therapeutic plasma exchange for acute inflammatorydemyelinating syndromes of the central nervous system. J Clin Aphere-sis. 1999;14:144–148.

125. Weinshenker BG, O’Brien PC, Petterson TM, et al. A randomized trialof plasma exchange in acute central nervous system inflammatorydemyelinating disease. Ann Neurol. 1999;46:878–886.

126. Keegan M, Pineda AA, McClelland RL, et al. Plasma exchange forsevere attacks of CNS demyelination: predictors of response. Neurol-ogy. 2002;58:143–146.

127. Wollinsky KH, Hulser PJ, Brinkmeier H, et al. CSF filtration is aneffective treatment of Guillain-Barre syndrome: a randomized clinicaltrial. Neurology. 2001;57:774–780.

128. Robinson WH, Steinman L, Utz PJ. Protein arrays for autoantibodyprofiling and fine-specificity mapping. Proteomics. 2003;3:2077–2084.

129. Robinson WH, Fontoura P, Lee BJ, et al. Protein microarrays guidetolerizing DNA vaccine treatment of autoimmune encephalomyelitis.Nat Biotechnol. 2003;21:1033–1039.

130. Mohan N, Edwards ET, Cupps TR, et al. Demyelination occurringduring anti-tumor necrosis factor alpha therapy for inflammatory ar-thritides. Arthritis Rheum. 2001;44:2862–2869.

84 BILLION IN 150,000 �/- YEARS*

Reliable Ways to Get Down to a Healthy,Peaceful, Worldly One Billion:

Drive 50 on the 65 MPH beltway.Eat bacon, butter, and eggs each breakfast.“Entertain” your neighbor’s wife when he is out of town.Leave loaded rifles in the grandchildren’s closet.Repair the rain spouts on your three-story house.Volunteer to look for that elusive water on the moon.Negotiate the intra-uterine onset week of personhood & humanness.Agree that the persistent vegetative state arrives earlier in Republicans.Truly patriotic seniors will stop eating at age 75, or maybe 73.Bomb uncivilized nations with contraceptives and Krispy-Kremes.Sell your stairstepper.

Ed Spudis, MD

*NEWSWEEK, JAN. 2003



diagnosis and management of acute myelopathies

Documents