http://msj.sagepub.com/Multiple Sclerosis

http://msj.sagepub.com/content/11/4/373The online version of this article can be found at:

 DOI: 10.1191/1352458505ms1194oa

2005 11: 373Mult SclerThomas F Scott, Salima L Kassab and Shalu Singhrate to clinically definite multiple sclerosis

Acute partial transverse myelitis with normal cerebral magnetic resonance imaging: transition  

Published by:

http://www.sagepublications.com

On behalf of: 

European Committee for Treatment and Research in Multiple Sclerosis

Americas Committee for Treatment and Research in Multiple Sclerosis

Pan-Asian Committee for Treatment and Research in Multiple Sclerosis

Latin American Committee on Treatment and Research of Multiple Sclerosis

can be found at:Multiple SclerosisAdditional services and information for    

  http://msj.sagepub.com/cgi/alertsEmail Alerts:

 

http://msj.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:   at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://msj.sagepub.com/content/11/4/373.refs.htmlCitations:  

What is This? 

- Aug 1, 2005Version of Record >>

at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

Acute partial transverse myelitis with normal cerebral magneticresonance imaging: transition rate to clinically definite multiplesclerosis

Thomas F Scott*,1, Salima L Kassab2 and Shalu Singh2

1Drexel University College of Medicine, Allegheny General Hospital, Pittsburgh, PA, USA; 2Department of Neurology,Allegheny General Hospital, Pittsburgh, PA, USA

Objective: To determine the long-term risk of developing clinically definite multiple sclerosis (CDMS) in patients with acute partialtransverse myelitis (APTM) and normal cerebral magnetic resonance imaging (MRI) scans. Methods: We retrospectively studied 30consecutive patients with clinical evidence of APTM. Patients with symmetric severe acute transverse myelitis were considered to havecomplete transverse myelitis and were excluded. All patients underwent spinal and cerebral MRIs, 13 underwent cerebrospinal fluidanalysis and 11 patients underwent evoked potential studies. Various other studies were performed to assess for connective tissuedisease and causes of APTM other than demyelinating disease. Results: After an average follow-up of 61 months, all laboratory andclinical evidence, including relapse history, indicated that three patients developed lesions on cerebral MRI and could be classified asCDMS by either Poser criteria (two patients) or MacDonald criteria (one patient). Relapses limited to the spinal cord seen clinicallywere seen in 14/30 (46.6%) patients. Oligoclonal bands were seen in 8/13 (62%) patients; one patient transitioned to CDMS. Unifocallesions of the cord were seen in 19/30 (63%) patients, multifocal lesions were seen in 8/30 (27%) and 3/30 (10%) had negative MRIs.The three patients who converted to CDMS did so within five years of the onset of myelitis. Conclusion: APTM with normal cerebralMRI had a low rate of conversion to CDMS in this long-term study. To date, there have been only a few follow-up studies that haveaddressed this issue.Multiple Sclerosis (2005) 11, 373�/377

Key words: MRI; multiple sclerosis; prognosis; transverse myelitis

Introduction

Only a few long-term studies have addressed the risk of

the development of clinically definite multiple sclerosis

(CDMS) in patients with normal cerebral MRI at the

onset of acute partial transverse myelitis (APTM).1,2

When patients present with APTM and cerebral MRI

showing lesions typical for MS, the transition rate to

CDMS is known to be quite high, in the range of 80�/

90% within a few years.3,4 By comparison, patients with

APTM and normal cerebral MRI may have a transition

rate in the range of 20�/33% for development of CDMS

over a 5 to 10-year period.1,2 Another group of myelitis

patients, those with acute severe idiopathic transverse

myelitis (‘complete’) with negative cerebral MRI appear

to be at very low risk for development of MS, with a

transition rate of 5/2%.5,6 Thirty patients with APTM

were followed longitudinally in an attempt to clarify this

important issue regarding transition to CDMS in this

population.

Methods

All patients were identified consecutively, examined, and

followed at the Allegheny Multiple Sclerosis TreatmentCenter. Over the past decade, our centre has monitored all

patients longitudinally with clinically isolated syn-dromes, such as transverse myelitis, in a standardized

fashion with attempts made to retain all patients. Chartreview for this study was retrospective. Inclusion required

sensory dysfunction with a demonstrable sensory level orhemi-level (unilateral truncal) or a clearly defined acutely

symptomatic spinal cord lesion by MRI. All MRIs werereviewed by the lead investigator, including cerebral MRIs

(obtained in all patients). Although upper motor neuron orbladder dysfunction was seen in most patients, signs and

symptoms referable to these problems were not requiredfor inclusion. Patients presenting with complete trans-

verse myelitis, defined as acute onset of moderate tosevere symmetrical motor and bladder dysfunction asso-

ciated with a sensory level, were excluded. Patients withone or more cerebral lesions typical of MS were excluded

(cerebral MRIs were considered to be negative when nowhite matter abnormality exceeded 3 mm). Patients with

relapsing symptoms were re-examined after each relapseby the lead investigator. Relapse was defined as a demon-

strable change on examination, correlating with com-plaints of new symptoms, occurring after a period of at

*Correspondence: Thomas F Scott, 420 East NorthAvenue, East Wing Office Building, Suite 206, Pittsburgh,PA 15212, USA.E-mail: tscott@wpahs.orgReceived 18 June 2004; accepted 24 February 2005

Multiple Sclerosis 2005; 11: 373�/377www.multiplesclerosisjournal.com

# 2005 Edward Arnold (Publishers) Ltd 10.1191/1352458505ms1194oa at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

least 30 days of stability following the initial attack, withonset over days to weeks. Follow-up exams were per-formed at least yearly on all but two patients.

Statistics

Fisher’s exact test was used for comparison of relapsingand nonrelapsing patients in assessing gender, the pre-sence of multifocal versus unifocal lesions on spinal MRI,and the predictive value of the presence of oligoclonalbands (obtained by agarose gel electrophoresis). Indepen-dent sample t -testing was used to assess differences in ageof onset in the same groups. The Mann�/Whitney U -testwas used to assess differences in disease severity accord-ing to the final Expanded Disability Status Scale (EDSS)score.7

Additional data collection

All patients underwent spinal MRI at appropriate levelsat initial presentation. A variety of MRI scanners wereused, with most patients studied using 1.5-Tesla units (nominimum requirement). Cerebral MRIs were performed onall patients and 26/30 (87%) patients had follow-upcerebral MRIs at 12�/24 month intervals (two patientshad follow-up MRIs at �/24 month intervals, and twopatients declined follow-up cerebral MRIs). Follow-upspinal MRIs were performed in select patients as indicatedclinically (in patients with relapses and/or suspicion ofneoplasm). Paraclinical evidence for diagnosis was soughtin many patients (11 underwent evoked potentials), andall patients were screened for evidence of connectivetissue disease.

Results

After an average follow-up of 61 months (range 24�/132months), 3/30 (10%) patients had developed cerebral MRIevidence of demyelinating disease. These three patients,and no others, could be classified as CDMS by either thePoser criteria (two patients) or the MacDonald criteria (onepatient).8,9 Relapses limited to the spinal cord clinicallywere seen in 14/30 (46.6%) patients. Oligoclonal bandswere seen in 8/13 (62%) patients; one patient transitionedto MS. Unifocal lesions of the cord were seen in 19/30(63%) patients, multifocal lesions in 8/30 (27%) and 3/30(10%) had negative scans. The three patients who con-verted to CDMS did so within five years of onset ofmyelitis (of the 11 patients with follow-up �/5 years, nonehas developed CDMS). One of these patients developedoptic neuritis at 12 months after onset of symptoms, onepatient had recurrent symptoms of myelitis and onepatient converted by MRI criteria only.

Additional studies

Repeat spinal MRIs were obtained in three patients felt topossibly have spinal cord tumours (two were relapsing),and in the 12 other relapsing patients. A new or expandingspinal lesion was found in only one patient, and thispatient had the worst outcome (cane dependence). Of the11 patients who underwent visual evoked potential

studies, only two had abnormalities (unilateral). Onewas symptomatic, studies having been performed onlyafter symptoms of optic neuritis became apparent. Thispatient developed cerebral MRI lesions and was classifiedas CDMS. The other patient was among the relapsingpatients, had positive studies for oligoclonal bands, andtherefore could be classified as laboratory-supporteddefinite MS (not included in the small group withCDMS).8 Increased cerebrospinal IgG synthesis rate wasseen in 7/13 (54%) patients and five of these sevenpatients were positive for oligoclonal bands.

Comparing relapsing to nonrelapsing patients, no differ-ences were found in age of onset, presence or absence ofmultifocal cord lesions, presence or absence of oligoclonalbands, severity of disease, or gender (see Table 1). Therewas a trend (P�/0.101) toward more relapsing patientsbeing found in the group of patients with multifocal cordlesions. The presence or absence of oligoclonal bands wasonly weakly predictive of relapse and did not predicttransition to CDMS. These analyses were repeated afterexcluding the three patients designated as CDMS from therelapsing group, and the results did not change.

Initial attacks were mild in most patients, as were mostrelapses. Complete recovery at final follow-up was seen ineight patients; EDSS was 1.0�/2.5 in 12 patients; EDSSwas 3�/4.5 in nine patients; and only one patient had anEDSS �/4.5. One patient experienced a slow decline to amild paraparesis over several years.

Discussion

Implications for prognosisThis report is the largest available long-term study ofAPTM in cerebral MRI-negative patients. With an averagefollow-up of five years, we found a 10% (3/33 patients)transition rate to CDMS by either the Poser or MacDonaldcriteria. Given the paucity of long-term data concerningoutcomes in cerebral MRI-negative patients with APTM,we feel our results will be of significant prognostic valueto physicians managing these patients.

The three described clinically isolated syndromes thatoften herald MS, namely APTM, optic neuritis and acutebrainstem syndromes, have all been studied in terms oftransition rates. Some of these studies have focused oncases presenting with negative cerebral MRI, but thesestudies are few and have low patient numbers.2,4,13�16 Ofthese syndromes, optic neuritis has been the most studied,and isolated brainstem syndromes are perhaps the leaststudied.13,16 The 10% (3/33 patients) transition rate toCDMS found in our study is within the range of transitionseen in other clinically isolated demyelination syndromespresenting with negative cerebral MRIs.13,16

Although it has been shown in many studies that thepresence of oligoclonal bands at the onset of clinicallyisolated syndromes is highly predictive of later develop-ment of MS, this has not been studied in large numbers ofpatients with a clinically isolated syndrome and a nega-tive cerebral MRI. Our results suggest that the presence ofoligoclonal bands may be of limited value in this patient

Acute partial transverse myelitis with MRITF Scott et al.

374

Multiple Sclerosis at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

group, however, due to the retrospective nature of this

study, only a subset of patients underwent cerebrospinal

fluid (CSF) analysis (raising the possibility of selectionbias). The study most comparable to ours in terms of CSF

analysis involved an Italian cohort presented by Martinelliet al .17 They studied 17 patients with APTM and negative

MRIs, a relatively large patient group, with only a shortfollow-up period of approximately two years. Only three

patients in this group were found to have oligoclonal

bands, and two of these developed CDMS. In a smallerstudy, the results of Dalecky et al .18 were more similar to

ours: of their seven patients with partial myelitis andnegative cerebral MRIs who were tested for oligoclonal

bands, five were positive, and two of these developed MRIevidence of cerebral disease within a two-year follow-up

period. With regard to spinal and optic-spinal forms ofdemyelinating disease in the Asian population, they

appear to differ from that seen in our group in two

respects: 1) this pattern of disease constitutes a muchlarger proportion of the patients seen in demyelinating

disease clinics in Asia, ranging from 8.5 to 40%,19,20

compared with 2�/4% of our clinic (data not shown); and

2) the incidence of oligoclonal bands appears to be verylow (0% in two Japanese studies and one Korean study,

and likely 5/33% in Chinese clinics).19 �22 None of our

patients was of Asian descent.

Implications for disease classificationVarious definitions for acute transverse myelitis have beenproposed.10 The term ‘acute partial transverse myelitis’

has received somewhat less attention, but may also be

defined.4 A working definition was proposed in 2002 forthe diagnosis of idiopathic acute transverse myelitis by the

Transverse Myelitis Consortium Working Group, withcriteria that do not distinguish between partial and

complete myelitis syndromes.10 The results of our studysuggest that it may be beneficial to develop criteria for

both APTM and acute complete transverse myelitis

because these syndromes, often idiopathic, present asomewhat different differential diagnosis and prognosis.

The definitions used in this study could be used as astarting point for such criteria (see Table 2).

We can define the closely related syndromes of acute

transverse myelitis and APTM as two distinct patterns of

spinal cord dysfunction. As the terms imply, the two

syndromes may be distinguished by the extent to which

spinal cord function is disrupted: acute transverse myeli-

tis referring to complete or near complete disruption

of function at a spinal level, and APTM referring to

mild, patchy or grossly asymmetrical spinal cord dysfunc-

tion. This simple dichotomy serves well to divide the

vast majority of idiopathic myelitis cases, with very

few patients falling into ‘grey areas’. This division

becomes extremely important in terms of prognosis, as

does the status of cerebral MRI scanning at presentation

(discussed below). The two syndromes involve a widely

overlapping differential diagnosis; however, patients

presenting with a severe (complete) transverse myelitis

appear much less likely to develop MS as compared to

patients with APTM. The syndrome of severe acute

transverse myelitis has received attention in several

studies each involving dozens of patients.5,6 By definition,

cerebral MRI is negative in this population and many

cases are considered to be post-infectious. Compared to

APTM patients, these patients appear less likely to present

with oligoclonal bands, less likely to relapse with a

second bout of myelitis and have a very low transition

rate to CDMS (likely 5/2%).5 Supportive of this finding,

one series of MS patients presenting as myelitis was

reported to demonstrate a complete acute transverse

myelitis syndrome in only one of 318 patients.11 A similar

study reported 0 of 26 patients presenting with complete

acute transverse myelitis in a series of MS patients with

myelopathy.12

A question arises concerning how to classify relapsing

forms of idiopathic myelitis when cerebral MRI remains

negative over the course of many years. By our interpreta-

tion, current criteria for CDMS do not distinguish between

unifocal and multifocal spinal cord lesions in terms of

separation of lesions in space, i.e., multifocal lesions of

the spinal cord are considered to represent a lesion at a

single site.8,9 Similarly, a second attack of myelitis, even

when occurring clinically several spinal levels above that

of an initial bout, does not represent a recurrence of

demyelinating disease at a spatially separate site.We propose this interpretation of current CDMS criteria

for the following reasons: 1) the available descrip-

tions of criteria for CDMS do not address this issue

specifically; and 2) ours is the most conservative approach

in interpretation of spinal cord lesions. Purely spinal

forms of demyelinating disease have been identified (such

as Devic’s disease) which are accepted to be distinct from

MS. This separate classification assumes that some pa-

tients with relapsing multifocal spinal demyelinating

disease will remain free of cerebral lesions. Similarly

accepted is a monophasic form of complete acute trans-

verse myelitis, which has been proposed to relapse only

occasionally, and does not subsequently involve the

cerebrum. Finally, although multifocal lesions are often

identified more clearly by MRI than by clinical criteria

(signs and symptoms), the limitations of MRI for identi-

Table 1 Characteristics of patients with relapsing versus non-

relapsing course of myelitis

Relapsingpatients(n�/14)

Nonrelapsingpatients(n�/16)

P value

Age at onset(years, mean/median)

41.7/43.5 38.0/39.0 0.245

Gender (female/male) 10/4 13/3 0.675Spinal MRI abnormality

(multifocal/unifocal)*6/14 2/16 0.101

Oligoclonal bands 5/7 3/6 0.51EDSS (mean) 2.5 1.8 0.26

*Three patients with normal spinal MRIs were included in the

unifocal group.

Acute partial transverse myelitis with MRITF Scott et al.

375

Multiple Sclerosis at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

fication of such lesions are not known (possible false

negatives studies).Although previously developed criteria for acute trans-

verse myelitis (Consortium Working Group) included

demonstration of inflammation by either enhanced MRI

scanning or CSF analysis,10 we avoided including this

criterion for the following reasons: 1) evidence of inflam-

mation is not observed uniformally in either complete

acute transverse myelitis or APTM in the existing litera-

ture;5 2) although we cannot rule out ischaemia in most

cases, ischaemia seems unlikely in the relatively young

patients with these syndromes (ischaemia should cer-

tainly be considered in elderly patients); 3) the primary or

initial pathophysiological events in essentially all demye-

linating syndromes remain unknown; and 4) we do not

require the presence of inflammatory markers in CSF to

make the diagnosis of MS, and in fact, such analyses are

often unrevealing in otherwise proven cases; and 5)

MRI abnormalities in idiopathic myelitis syndromes are

extremely variable and are not uniformly present (dis-

cussed below).We considered developing MRI criteria for complete

acute transverse myelitis and APTM, but rejected this

consideration given the heterogeneous MRI findings.

Normal spinal MRIs were seen in 3/30 (10%) of our

patients. Normal spinal MRIs have also been reported in

patients with severe complete acute transverse myelitis,

and these patients may have an improved prognosis.5 The

optimal timing of MRI scanning during the disease course

in patients with myelitis is unknown, and it is likely that

clinicians will sometimes miss periods during which MRI

scanning is transiently abnormal. Most myelitis patients

have MRI abnormalities at the time of initial scanning and

have residual lesions at follow-up. Complete acute trans-

verse myelitis tends to be associated with large oedema-

tous-appearing lesions, which may expand the cord over

multiple levels and produce T2-weighted abnormalities

involving the central cord, and entire cord when viewed

axially. MRI lesions of APTM tend to be smaller, oftendorsal or lateral. Exceptions to these usual findings aresufficiently frequent to preclude the use of MRI abnorm-alities for the development of diagnostic criteria.

Conclusion

Our results and those of others suggest that the risk ofdeveloping MS in patients with APTM is likely in therange of 10�/20% over 5�/10 years. Although fairly low,this risk is significantly higher than that seen in patientswith complete transverse myelitis. We recommend long-term follow-up of APTM patients with both physicalexaminations and periodic cerebral MRIs. We advocatethe use of the terms ‘complete acute transverse myelitis’and ‘acute partial transverse myelitis’ to describe patientsat presentation (for reasons relating to diagnosis andprognosis as described above). We have also identifiedand followed many patients with relapsing APTM whohave yet to develop CDMS, and who present difficultieswhen applying current classification schemes. Taking ourdata together with the few existing reports of the long-termprognosis of patients with relapsing myelitis, it appearsthe syndrome will likely assume a place in the nosology ofdemyelinating diseases as a distinct syndrome separatefrom MS, as has Devic’s and monophasic completetransverse myelitis.

AcknowledgementsThe authors are grateful to Drs Kevin Kelly and John Ricefor their critical review, and to Lisa Pappert for herpreparation of this manuscript. Assistance with statisticalanalysis was provided by Drs Carol Schramke and GaryCutter.

Table 2 Criteria for transverse myelitis

Standing criteria for idiopathic acutetransverse myelitis

Proposed criteria for complete acutetransverse myelitis

Proposed criteria for APTM

Sensory, motor, or autonomic dysfunctionattributable to the spinal cord

Moderate or severe symmetrical weaknessand autonomic (bladder) dysfunctionattributable to the spinal cord

Mild sensory and or motor dysfunctionattributable to the spinal cord, bilateral orunilateral; if severe deficits present,marked asymmetry is observed

Bilateral signs and/or symptoms (notnecessarily symmetric), clearly definedsensory level

Symmetrical sensory level Sensory signs or symptoms referable to asensory level or hemi-level or MRI lesiontypical of myelitis

Progression to nadir between 4 h and 21 days Progression to nadir between 4 h and21 days

Progression to nadir between 4 h and21 days

Exclusion of other causes including otherforms of demyelinating disease (no historyof optic neuritis, no cerebral lesionssuggestive of MS, consider: autoimmunitylaboratory tests, evoked potentials, othertests)

Same Same

Inflammation within the spinal corddemonstrated by CSF pleocytosis orelevated IgG index or gadoliniumenhancement

CSF or MRI evidence of inflammationwithin the spinal cord may or may notbe present

CSF or MRI evidence of inflammationwithin the spinal cord may or may not bepresent

Acute partial transverse myelitis with MRITF Scott et al.

376

Multiple Sclerosis at PRINCETON UNIV LIBRARY on November 14, 2014msj.sagepub.comDownloaded from

References

1 Bashir K, Whitaker JN. Importance of paraclinical and CSF

studies in the diagnosis of MS in patients presenting with

partial cervical transverse myelopathy and negative cranial

MRI. Mult Scler 2000; 6: 312�/16.

2 O’Riordan JI, Thompson AJ, Kingsley DPE, MacManus DG,

Kendall BE, Rudge P et al . The prognostic value of brain MRI

in clinically isolated syndromes of the CNS. A 10-year follow-

up. Brain 1998; 121: 495�/503.

3 Jacobs LD, Beck RW, Simon JH, Kinkel RP, Brownscheidle CM,

Murray TJ et al . Intramuscular interferon beta-1a therapy

initiated during a first demyelinating event in multiple

sclerosis. N Engl J Med 2000; 343: 898�/904.

4 Ford B, Tampieri D, Francis G. Long-term follow-up of acute

partial transverse myelopathy. Neurology 1992; 42: 250�/52.

5 Scott TF, Bhagavatula K, Snyder PJ, Chieffe C. Transverse

myelitis comparison with spinal cord presentations of multi-

ple sclerosis. Neurology 1998; 50: 429�/33.

6 Berman M, Feldman S, Alter M, Zilber N, Kahana E. Acute

transverse myelitis: incidence and etiologic considerations.

Neurology 1981; 31: 966�/71.

7 Kurtzke JF. Rating neurologic impairment in multiple sclero-

sis: an expanded disability status scale (EDSS). Neurology

1983; 33: 1444�/52.

8 Poser CM, Paty DW, Scheinberg L, McDonald I, Davis FA,

Ebers GC et al . New diagnostic criteria for multiple sclerosis:

guidelines for research protocols. Ann Neurol 1983; 13: 227�/

31.

9 MacDonald WI, Compston A, Edan G, Goodkin D, Hartung HP,

Lublin FD et al . Recommended diagnostic criteria for multiple

sclerosis: guidelines from the International Panel on the

Diagnosis of Multiple Sclerosis. Ann Neurol 2001; 50: 121�/

27.

10 Tranverse Myelitis Consortium Working Group. Proposed

diagnostic criteria and nosology of acute transverse myelitis.

Neurology 2002; 59: 499�/505.

11 Runmarker B, Andersen O. Prognostic factors in a multiple

sclerosis incidence cohort with twenty-five years of follow-up.

Brain 1993; 116: 117�/34.

12 Simnad VI, Pisani DE, Rose JW. Multiple sclerosis presenting

as transverse myelopathy: clinical MRI features. Neurology1997; 48: 65�/73.

13 Optic Neuritis Study Group. The 5-year risk of MS after optic

neuritis. Neurology 1997; 49: 1404�/13.

14 Optic Neuritis Study Group. High- and low-risk profiles for

the development of multiple sclerosis with 10 years after optic

neuritis. Arch Ophthalmol 203; 121: 944�/49.

15 Morrissey SP, Miller DH, Kendall BE, Kingsley DPE, Kelly

MA, Francis DA et al . The significance of brain magnetic

resonance imaging abnormalities at presentation with clini-

cally isolated syndromes suggestive of multiple sclerosis.

Brain 1993; 116: 135�/46.

16 Sastra-Garriga J, Tintore M, Rovira A, Grive E, Pericot I,

Comabella M et al . Conversion to multiple sclerosis after a

clinically isolated syndrome of the brainstem: cranial mag-

netic resonance imaging, cerebrospinal fluid and neurophy-

siological findings. Mult Scler 2003; 9: 39�/43.

17 Martinelli V, Comi G, Rovaris M, Filippi M, Colombo B,

Locatelli T et al . Acute myelopathy of unknown aetiology: a

clinical, neurophysiological and MRI study of short- and long-

term prognostic factors. J Neurol 1995; 242: 497�/503.

18 Dalecky A, Pelletier J, Ali Cherif A, Levrier O, Khalil R.

Myelopathies aigu du sujet jeune et sclerose en plaques. RevNeurol (Paris) 1997; 153: 569�/78.

19 Misu T, Fujihara K, Nakashima I, Miyazawa I, Okita N, Takase

S et al . Pure optic-spinal form of multiple sclerosis in Japan.

Brain 2002; 125: 2460�/68.

20 Kira J. Multiple sclerosis in the Japanese population. Lancet

Neurol 2003; 2: 117�/27.

21 Kim K. Idiopathic recurrent transverse myelitis. Arch Neurol2003; 60: 1290�/94.

22 Yu YL, Woo E, Hawkins BR, Ho HC, Huang CY. Multiple

sclerosis amongst Chinese in Hong Kong. Brain 1989; 112:

1445�/67.

Acute partial transverse myelitis with MRITF Scott et al.

377

Multiple Sclerosis