You are on page 1of 8

 

SPECIAL ARTICLE
Evi
Evide
dence
nce-ba
-based
sed gui
guide
delin
line:
e: Cl
Clini
inical
cal ev
evalu
aluati
ation
on and
trea
treatm
tmen
entt of tr
tran
ansv
sver
erse
se my
myelelit
itis
is
Report of the Therapeutics and Technology Assessment Subcommittee of the
 American Academy of Neurology 

T.F. Scott, MD ABSTRACT


E.M. Frohman,
MD, PhD Objective: To assess the evidence for diagnostic tests and therapies for transverse myelitis (TM)
and make evidence-based recommendations.
 J. De Seze, MD
Methods:  A review of the published literature from 1966 to March 2009 was performed, with
G.S. Gronseth, MD,
evidence-based classification of relevant articles.
FAAN
B.G. Weinshenker, MD Recommendations:  Level B recommendations: neuromyelitis optica (NMO)–immunoglobulin G (IgG)
antibodies should be considered useful to determine TM cause in patients presenting with clinical
acute
acut e com
complet
plete
e tran
transver
sverse
se mye
myelitis
litis (AC
(ACTM)TM) feat
features
ures.. The pres
presence
ence of NMONMO-IgG
-IgG anti
antibodi
bodies
es
 Address correspondence and (aquaporin-4–specific antibodies) should be considered useful in determining increased TM recur-
reprint requests to American rence risk. Level C recommendations: in suspected TM, distinction between ACTM or acute partial
 Academy of Neurology, 1080
Montreal Avenue, St. Paul, MN transverse myelitis may be considered useful to determine TM etiology and risk for relapse (more
55116 common with APTM). Age and gender may be considered useful to determine etiology in patients
guidelines@aan.com
presen
presentin
ting
g wi
with
th TM sy
synd
ndrom
rome,e, wi
with
th spi
spinal
nal inf
infarc
arcts
ts se
seen
en mo
more
re of
often
ten in old
older
er pat
patien
ients
ts and mo
more
re fe
femal
malee
thanmale
than malepatie
patients
nts havi
having
ng TM due to mul
multiple
tiplescle
scleros
rosis
is (MS
(MS).
). Brain
BrainMRI
MRIchar
characte
acterist
ristics
ics con
consist
sistent
ent with

those
gerofspin
Longer
Lon MSalmay
spinal be considered
lesions
lesions extendin
exte nding useful
g over to
3 ver
predict
verteb
tebral conversion
ral se
segme
gments
nts matoyMS
may be coafter
cons
nsidea red
firstus
idered partial
usef ul inTM
eful episode.
determ
det ermini
ining
ng
NMO
NM O vs MS
MS.. CS
CSF F exa
examin
minati
ation
on for cel
cells
ls an
andd oli
oligoc
goclon
lonal
al ban
bands
ds ma
may y be co
cons
nside
idered
red us
usef
eful
ul to det
determ
ermin
ine
e
the cause of the TM syndrome. Plasma exchange may be considered in patients with TM who fail to
improv
imp rovee aft
after
er cor
cortic
ticos
oster
teroid
oidtre
treatm
atment
ent.. Ri
Ritux
tuxim
imab
ab ma
mayy be con
consi
sider
dereded in pat
patien
ients
ts wit
with
h TM dueto NM NMO O
to dec
decrea
rease
se the nu
numbmberer of rel
relaps
apses.
es. Le
Level
vel U rec
recom
ommemendndati
ation
ons:
s: th
there
ere is ins
insufuffi
ficie
cient
nt evi
eviden
dence
ce to su
sup-
p-
portt or ref
por refute
ute the eff
effica
icacy
cy of oth
otherer TM the
therap
rapies
ies or the us
usefu
efulne
lness
ss of eth
ethni
nicit
cityy to det
determ
ermine
ine the cau
cause
se
of a suba
subacute
cute myel
myelopat
opathy.
hy. Neurology  2011;77:2128–2134
®

GLOSSARY 
ACTM      acute complete transverse myelitis;  APTM      acute partial transverse myelitis;  CI     confidence interval;  IgG   
immunoglobulin G; MS  multiple sclerosis; NMO  neuromyelitis optica; OCB  oligoclonal band; TM  transverse myelitis.

Transverse myelitis (TM), an inflammatory lesion of  sion typically spans multiple vertebral segments and is
the spinal cord, occurs in 1 (severe) to 8 (mild) cases/ not radiologically or pathologically transverse; the term
millio
mil lion
n per yea
year.
r.1–5 TM is usu
usuall
allyy acc
accomp
ompani
anied
ed by MRI transverse  has
  has been retained because of the importance
signal abnormality in the spinal cord, CSF pleocytosis, of a spinal sensory level in making the diagnosis. 7
or both
both.. Typi
Typical
cal mani
manifesta
festation
tionss and incl
inclusio
usion/ex
n/exclus
clusion
ion This guideline seeks to answer the following clin-
Supplemental
Supplemental data at criteria were outlined by the Transverse Myelitis Con- ical questions:
www.neurology.org
sortium Working Group in 20026 (table e-1 on the For patients with myelopathy, which demographic,
Neurology ®  Web site at www.
www.neur
neurolog
ology.or
y.org).
g). The le- clinic
clinical,
al, rad
radiogr
iograph
aphic,
ic, and labo
laborat
ratory
ory feat
feature
uress are use
useful:
ful:
SupplementalData

From Drexel University School of Medicine/Allegheny MS Treatment Center (T.F.S.), Pittsburgh, PA; Department of Neurology (E.M.F.),
University of Texas Southwestern, Dallas; Department of Neurology (J.D.S.), Strasbourg university hospital and centre d’investigation clinique,
Strasbourg, Cedex, France; Department of Neurology (G.S.G.), University of Kansas Medical Center, Kansas City; and Department of Neurology 
(B.G.W.), Mayo Clinic, Rochester, MN.
Podcast
 Appendices e-1-e-5, tables e-1-e-5, and references e1-e19 are available on the  Neurology ®  Web site at www.neurology.org.
 Approved by the Therapeutics and Technology Assessment Subcommittee on November 13, 2010; by the Practice Committee on May 11, 2011; and
by the AAN Board of Directors on October 14, 2011.
Study funding: This evidence-based guideline was funded by the American Academy of Neurology. No author received honoraria or financial support
to develop this document.
Disclosure: Author disclosures are provided at the end of the article.

2128   Copyright © 2011 by AAN Enterprises, Inc.


 

1. To di
dist
stin
ingu
guis
ish
h TM frfrom
om ot
othe
herr ca
caus
uses
es of ac
acut
utee an
and
d[95% confidence
confidence interval (CI) 0.51– 0.81] and 80%
subacute
subacu te noncom
noncompressi
pressive
ve myelopa
myelopathy?
thy?  women for TM due to any inflammatory cause [95%
2. To determine
determine the cause of the myelitis?
myelitis? CI 0.51– 0.77]
0.77]). ).
3. To ide
identi
ntify
fy pat
patient
ientss at inc
increa
reased
sed ris
riskk for Conclusions.  In patients presenting with acute my-
recurrence? elopathy,
elopat hy, age is pos
possib
sibly
ly use
useful
ful in ide
ident
ntify
ifying
ing pa
patie
tient
ntss at
For patients with TM, which therapies higher risk for spinal cord infarcts, and female gender is
4. Allevia
Alleviate
te acute attacks?
attacks? possibly useful in identifying patients at higher risk for
5. Preven
Preventt future attacks?
attacks? inflammatory myelopathies (2 Class III studies).
Clinical features.   We did not identify studies de-
DESCRIPTION OF THE ANALYTIC PROCESS  A  scribing an association between clinical features of 
literature search of Medline was performed for rele- myelopathy (such as the time of onset to maximal
vant articles published from 1966 to March 2009, neurologic deficit) and the etiology of the myelopa-
using the following key words: myelitis, transverse thy (myelitis vs other types).
myelitis, Devic disease, neuromyelitis optica, diagno- Conclusions. There is insuffi
insufficient
cient evidence
evidence to deter-
sis, prognosis, outcomes, MRI, and treatments. The mine whether clinical features of the myelopathy are
search was limited to reports in humans and abstracts associated with myelitis vs other myelopathies.
available in English. Subheadings were applied as ap- Laboratory features.  Two Class III studies (n  79
iate. The exact search strategy employed is de- and n      28), both retrospective cohort surveys,
propriate.
propr
scribed in appendix e-1. A secondary search of review  found CSF pleocytosis (10 cells/m cells/mmm3) to be pres-
articles was done to find any missed citations. ent more often in inflammatory processes (86%
 We revie
r eviewed
wed all abstr
abstracts; p oten-- [95% CI 60%–96%]) than in spinal cord infarct (0
acts; the full text of poten
tially relevant articles was subsequently reviewed [95% CI 0%–20%]).8,10
by at least 2 committee members. We excluded review  Conclusions.   For patients with subacute myelopa-
articles and case reports. At least 2 committee members thies, an elevated CSF leukocyte count (greater than
independently rated each article for its class of evidence 10 cells/mm3) is possibly useful in identifying pa-
using the American Academy of Neurology diagnostic tients with inflammatory myelopathies (including 
(questions 1 and 2), prognostic (question 3), or thera- TM) as opposed to those with spinal cord infarcts (2
peutic (questions 4 and 5) classification of evidence Class III studies).
schemes (appendix e-2). Differences between reviewers For pati
patients
ents with myelo
myelopath
pathy,
y, whic
whichh demo
demograph
graphic,
ic,
 were resolved
resolved throug
throughh discussion
discussion with
with a third reviewer.
reviewer. clinical, radiographic, and laboratory features are use-
Recommendations were formulated and linked to the ful to determine the cause of the myelitis?  When the
strength of the evidence using the scheme described in diagnosi
diagnosiss of TM is est
establ
ablish
ished,
ed, det
determ
ermini
ining
ng the cau
cause
se of 
appendix e-3. the myelitis is useful. The main etiologies of TM-like
syndromes are MS, parainfect
parainfectious
ious myelitis, NMO, and
ANALYSIS OF EVIDENCE  The search yielded 136 myelitis due to systemic disease (such as systemic lupus
articles. All articles were reviewed in their entirety. erythematosus). However, even after several years of 
Sixty-five articles met inclusion criteria. follow-up, 15% to 36% of patients cannot be given a 
11,12

For pati
patients
ents with myelo
myelopath
pathy,
y, whic
whichh demo
demograp
graphic,
hic, more specific diagnosis than “idiopath
“idiopathic”
ic” TM.
clinical, and laboratory features are useful to distin- Demographic features. Of 4 Class III retrospective co-
guish TM from other causes of acute and subacute hort surveys, the 2 largest (n  36,79)8,13 reported that
noncompressive myelopathy?  Demographic features.  We more women than men are diagnosed with inflamma-
identified 2 retrospective cohort studies (Class III) tory
tory mye
myelop
lopath
athies
ies due to MS
MS,, but no gengender
der ass
associ
ociat
ation
ion
(n     33 and n     79) that described demographic  was foun
foundd in thes
thesee 4 stud
studies
ies in idio
idiopath
pathic
ic TM (95% CI
features of patients with inflammatory (idiopathic 0.23–
0.23 – 0.61
0.61;; see table e-2)
e-2)..8,9,13,14 Only Class IV studies
TM, postinfectious, systemic collagen vascular dis- are avai
availabl
lablee rega
regardi
rding
ng the asso
associat
ciation
ion betw
between
een ethni
ethnicity 
city 
ease, and neuromyelitis optica [NMO]) or nonin- and the cause of myelitis.  When compar
15,16
comparing
ing various
various
flammatory (spinal infarct) acute myelopathies (table types of myelitis, we found 2 studies showing no signif-
e-2).8,9 Both studies demonstrated that patients with icant age differences and 2 studies with insufficient data 
spinal cord infarcts were older (mean age 52 years to assess age differences concerning idiopathic TM vs
[first study] and 67 years [second study]) than pa- MS presenting as myelitis (table e-2).
tients with TM (mean age 31 years [first study] and Conclusions.   For patients with myelopathy, demo-
50 years [second study]). When data from both stud- graphic
graphic fea
featu
tures
res ar
aree pos
possib
sibly
ly not use
useful
ful in dis
disti
tingu
nguish
ishing 
ing 
ies were combined, they showed patients with in- causes of myelitis (multiple Class III studies).
flammatory myelopathy were more often women Clinical features.  TM is commonly divided into 2
(68% women for TM due to multiple sclerosis [MS] subgroups on the basis of the extent of spinal cord

Neurology
Neurology 77 Dece
December
mber 13, 2011   2129
 

involveme
involvement:nt: acute complete transve
transverse
rse myelitis nese patients with optico-spinal MS were more likely 
(ACTM)
(AC TM) and acu acute
te par
partia
tiall tra
transv
nsvers
ersee myel
myeliti
itiss to have NMO define
definedd by NMO antibody positivity 
positivity 
(APTM).  ACTM is an acute or subacute inflam-
7,8
(vs oth
other
er typ
types
es of spi
spinal
nal demy
demyelin
elinatin
atingg dis
disease
ease,,
matory process of the spinal cord causing symmetric NMO antibody negative) if they had longitudinally 
moderate or severe loss of function distal to that extensive spinal lesions ( p  0.0036).21  Another ret-
level. APTM is incomplete or patchy involvement of  rospective cohort study (n      22) of patients with
at least one spinal segment with mild to moderate short spinal cord involvement radiographically re-
 weakness, asymmetric or dissociated sensory symp- vealed a 4% (1/22) rate of developing NMO,20 and
toms (i.e., spinothalamic function lost but dorsal col- another prospective cohort study of 29 patients with
umn functi
function
on spared
spared),
), and occasi
occasionally
onally bladder a long spinal cord segment of myelitis radiographi-
17

involvement.  We reviewed


the potential usefulness the evidence regarding 
of distinguishing ACTM cally revealed a high rate (38%) of NMO-IgG23 sero-
positivity and conversion to NMO or relapse.
from APTM in determining the cause of TM. Conclusions. The longitudinal extent of MRI lesions
 We found no studies directly comparing the risk  is possibly useful in determining the cause of TM
of MS development in patients who have APTM (multiple Class III studies), specifically in distin-
 with that in patients who have ACTM. However, guishing between NMO spectrum disorders and MS
Class III evidence from multiple natural history stud- in patients with idiopathic TM.
ies of well-characterized patients (cerebral MRI nega-  MRIs demonstrating lesions typical of MS.   One pro-
tive) with APTM and thos osee with ACACTTM spective Class II study of 26 patients with APTM
demonstrate an increased risk of MS development in provides evidence for the value of the presence of 
the former group. Two studies of APTM (n      30
cerebral MRI lesions for predicting the development
and n  9) demonstrated that transition to MS oc-
of MS
MS.. MS wa wass di
diag
agno
nose
sed
d du
duri
ring
ng 5 ye year
arss of 
curs at a rate of 10.3% (95% CI 4.1%–23.6%) 18,19
follow-up in 10/17 (59%) patients with any cerebral
 whereas 2 studies of ACTM suggest a significantly 
lower rate of transition to MS of 0% to 2% 5,13 (dur-  without MRI lesions
such as compared
lesions with 1/9
( p  0.018). 24 (11%) patients

ing approximately
approximately 5 years of follow-up of these con-
 A Class III retrospective cohort study of 15 pa-
ditions). One study characterized APTM as being 
tients with APTM also noted a high transition rate to
rarely associated with NMO–immunoglobulin G
MS in patien
patients ts with cereb
cerebral
ral MRIs typical for MS.17
(IgG) antibodies.20
In 2 Class III studies the transition rate to MS was
Conclusions. Patients with myelopathy who present
80% to 90% in patients with APTM followed over 3
as having APTM possibly have a higher risk of tran-
to 5 years if cerebral MRIs showed 2 or more lesions
sition to MS vs those presenting as having ACTM
typical
typical for MS at presepresentati
ntation,
on, vs 10%–11
10%–11%% trans
transi-
i-
(multiple Class III studies).
tion rate to MS among patients presenting with nor-
Radiographic
Radiog raphic featur
features. Length of spinal cord lesion.   We
es.   Length
found 4 studies that address the length of MRI- mal cerebral MRIs. This finding is further
18,24,25

detected spinal cord lesions in relation to the etiology  supported by 4 Class III retrospective cohort studies
of TM.20–23 Two of these studies, involving Japanese of patients with APTM.
8,14,26,27

Despite the evidence that MRI lesions are pre-


patients,
NMO vs directly compared
MS in patients withtheTMriskwith
of developing 
longitudi- dictive of MS, cerebral MRI lesions also occur
nally extensive lesions (defined as extending over at fairly frequently in NMO. However, Barkhof ce-
28

least 3 vertebral segments identified by standard rebral MRI criteria are usually not satisfied in
strength [1.5 T] MRI scanning) with that of pa- NMO, indicating that these lesions are not charac-
tients with TM with shorter lesions.21,22 In Japan, teristic of MS.
29

patients with optic neuritis or myelitis, regardless of  Conclusions. In patients with TM, especially APTM,
lesion length, are classified as having “optico-spinal” MS-like brain MRI abnormalities possibly indicate a 
MS.22  A large (n  200) retrospective cohort study  higher risk of “conversion” to clinically defined MS
(Class III) suggested that Japanese patients with TM (approximately 80% by 3–5 years after onset) (1
have a greater chance of manifesting the relapsing  Class II study and multiple Class III studies).
optico-spinal form (also fulfilling criteria for NMO) Laboratory features.  Autoantibodies.  We found 1 Class
if they present with longitudinally extensive lesions I prospective study (n  29) examining the predic-
rather than with short lesions (65% of patients who tive value of serum NMO-IgG positivity in identify-
met NMO criteria were noted to have presented with ing the etiology of TM. 23 The presence of these
longitudinally extensive lesions vs only 32% of pa- autoantibodies (also termed aquaporin-4–specific
tients with myelopathic MS,  p  0.001).22 Likewise, autoantibod
autoantibodies)
ies) in patie
patients
nts with TM was associated
in another Class III retrospective cohort study Japa-  with subsequent development of NMO or NMO

2130   Neurology
Neurology 77 Decem
December
ber 13, 2011
 

spectrum disorder on the basis of clinical criteria (see idiopathic APTM within 5 years is reported as ap-
table e-3 for criteria for NMO diagnosis).30 proximately 40% (Class III evidence).18
In sev
severa
erall Cla
Class
ss III
IIIstu
studie
dies,
s, aqu
aquapo
aporin
rin-4
-4 autoantibo
autoantibodies
dies Conclusions. Relapse rates possibly differ in patients
 were deemed
deemed a moderately
moderately sensitive
sensitive and highly specific  with ACTM and patients with APTM (Class III evi-
highly specific
test for discriminating NMO from MS (see table e-4) dence from multiple studies), with relapse possibly 
using clinical criteria and follow-up as the reference being more common in APTM.
standard.31–38 However, these retrospective studies do Radiographic features.  No information about recur-

not always specifically address which of these patients rence was given in 2 Class III studies suggesting that
 with NMO
NMO presente
presented d with TM.
TM. long spinal lesions may herald NMO.21,22  Another
Conclusions.   Aquaporin-4–specific autoantibodies study (n      29) prospectively found a high rate of 
relapse (and development of NMO) in patients with
(NMO-IgG)
cause of TM (NMO are probably
or NMO useful to establish
spectrum disorder)the
in longitudinally extensive lesions (more than 3 seg-
patients with suspected TM (1 Class I study and sev- ments) at presentation; however, the study did not
eral Class III studies). involve a group of patients with short lesions for
CSF.  One Class III retrospective cohort study re-
comparison.23 One Class III study (n      30) ad-
vealed a high likelihood of TM due to causes other dressed whether multiple short lesions (vs a single
than MS if CSF pleocytosis was greater than 30 cells/ short lesion) increase risk of relapse or transition to
mm3 (seen in 35% of patients with myelitis,   p    MS and found no predictive value.
18

Conclusions. Longer lesions on spina


spinall MRI possibly 
0.005 by Fisher test).8  A Class III case control study 
predict a higher risk of developing NMO; therefore,
of CSF of 71 patients with NMO vs patients with
some risk of recurrent TM is suspected, but the risk 
MS showed a white cell count higher than 50/dL in
relative to that from short lesions has not yet been
18 of 52 NMO cases, 28 of which had more than
directly studied (Class III evidence from multiple
10% polymorphonuclear cells.39
studies). There is insufficient evidence regarding the
 We found 8 Class III studie studiess (30 to 79
patients) 8,9,12,13,18,39,40,e1 usin
usingg olig
oligocl
oclonal
onal ban
bands
ds value of multiple
or transition to MSshort lesions
(1 Class III in predicting relapse
study).
(OCBs) to differentiate etiologies of TM (partial and Laboratory features.  One prospective Class I study 
comp
co mplet
lete)
e) anandd 1 Cla Classss II ststud
udyy (p
(pro
rosp
spec
ecti
tive
ve found that the presence of aquaporin-4–specific auto-
follow-up of 55 patients)26 assessing the usefulness of  antibodies predicts recurrence of TM or conversion to
OCBs to predict transition to MS after APTM. NMO.23 In this study, 44% of patients with TM who
These studies found OCBs in 85%–90% of patients  were NMO posi positive
tive had a relap
relapse
se (myeli
(myelitis
tis or opti
opticc
 with MS and in 20%–30% of patients with NMO or neuritis) within 1 year as compared with 0% of the pa-
vasculitis but none in patients with parainfectious tients who were NMO negative ( p  0.012). Antinu-
myelitis or spinal cord infarct.8,9,26,39 clear antibodies were more frequent in the group with
Conclusions. CSF analysis for OCBs is possi possibly
bly use- relapses (25%) as compared with the group without re-
ful in determining MS vs other causes of TM, specif- lapses (12%), but the difference was not significant.
ically for the diagnosis of MS vs NMO, spinal cord The presence of antibodies to SSA/Ro antigen (60 kD
infarct, vasculitis, and parainfectious and idiopathic and 52 kD polypeptides complexed with Ro RNAs)

TM
sis of(1CSF
Class
forIIpleocytosis
study and is
8 Class III useful
possibly studies).
in Analy-
distin-  was also pred
predicti
ictive
ve of relap
relapses
ses (myel
(myelitis
itis)) afte
afterr TM in
75% to 77% of patients in 1 Class III retrospective co-
guishing NMO from MS (1 Class III study) and MS hort study (n  25) ( p  0.047).e2
from all other causes of TM (1 Class III study). Conclusions. The presen
presence
ce of NMO autoantibodies
autoantibodies
For pati
patients
ents with myelo
myelopath
pathy,
y, whic
whichh demo
demograp
graphic,
hic,
probably predicts relapse in patients with TM (1
clinical, radiographic, and laboratory features are use- Class I study). There is insufficient evidence con-
ful to identify patients at increased risk for recurrence? cerning whether the presence of SSA antibodies pre-
Demographic features. No studies address the associ- dicts recurrence after a first episode of TM (1 Class
ation between demographic features of patients and III study).
risk of TM recurrence. For patients with TM, which therapies alleviate acute
Conclusions. There is insuffi
insufficient
cient evidence
evidence to deter
deter-- attacks?   Steroids.  Only Class IV evidence exists con-
mine whether demographic features are associated cerning the utility of steroids in treating TM.
 with relapsing TM. Conclusions.  In patients with TM, there is insuffi-
Clinical features.  We found no studies that directly  cient evidence
evidence to determine the utilit
utilityy of corticoste-
corticoste-
compared the rate of recurrence in ACTM with that roids in alleviating TM attacks (Class IV studies).
in APTM. However, the rate of recurrence of idio- Clinical context.   Despite the absence of evidence,
pathic ACTM in the 5 years after onset is approxi- admin administrat
istration
ion of high-d
high-dose
ose IV methylp
methylprednis
rednisolone
olone
mately
mate ly 10%
10%,,  where
13
 whereasas the recurre
recurrence
nce rate of  (1 g daily for 3 to 7 days) is typically the first treat-

Neurology
Neurology 77 Dece
December
mber 13, 2011   2131
 

ment offered to hasten recovery, reduce disease activ- fits of a variety of other agents to abort TM attacks,
ity, and restore neurologic function. promote functional recovery, or influence the future
Plasma exchange.  The American Academy of Neu- predilection of additional attacks.e11–e19
rology recently published an evidence-based guide- Conclusions. There is insuffi
insufficient
cient evidence
evidence to deter-
line
line on th thee ef
effi
fica
cacy
cy of plplas
asma
ma ex exch
chan
ange
ge fo
forr mine the efficacy of azathioprine, cyclophosphamide,
e3
neurologic disorders, including TM. The guideline and IVIg in alleviating TM attacks (Class IV studies).
concluded: “Based on a single Class II study [reference For patients with TM, which therapies
therapies prevent future
e4] plasmapheresis is possibly effective for acute fulmi- attacks?  Many of the same Class III and IV studies
nant CNS demyelinating diseases (including…TM) mentioned above have addressed prevention of re-
that fail to respond to high-dose corticosteroid treat- current TM attacks.
ment.
men
 with t.[dif
Becaus
Bec ause
[differe e the
ferent]
nt] study
study
demyelinat
demye includ
included
linating
ing ed sub
subgro
group
diseases
dise , itups
ases, s of poss
is not patien
patients
ts
possible
ible
Conclusion. There is insufficient evidence regarding 
the use of other immunosuppre
immunosuppressivessive strategies
strategies to re-
to determine if plasmapheresis is more or less effective duce the risk of future TM attacks.
in patients with [TM].”
 We found no additional studies to warrant chang- RECOMMENDATIONS
ing this conclusion.e5–e7 In patients with suspected TM, distinction be-
 Mitoxantrone.   An open-label study prospectively  tween ACTM or APTM may be considered useful to
evaluated the risk for attack recurrence in patients determine the etiology of TM and the risk for relapse
 with NMO using mitoxantrone (12 mg/m 2 monthly  (more common in APTM) (Level C).
for 6 months and then every 3 months for 3 addi-  Age and gender may be considered useful to de-
tional doses) (Class III, MRI and clinical outcomes termine etiology in a patient presenting with TM
 well defined and considere
considered d objective
objective).).e8 Despite syndrome, with spinal infarcts seen more often in
treatment, 2 subsequent attacks occurred during the older patients and more female than male patients
initial 5 months of therapy, 1 designated as severe having TM due to MS (Level C). Due to consider-
and the other as moderate. Systematic and longitudi- able overlap between groups, patient demographic
nal clinical (Expanded Disability Status Scale) and characteristics are not definitive in establishing the
radiographic (MRI) measures of disease activity dem- cause of myelopathy.
onstrated improvements in 4 of the 5 patients. There is insufficient evidence to support or refute
Conclusions. There is insuffi
insufficient
cient evidence
evidence to deter
deter-- the usefulness of ethnicity to determine the cause of a 
mine the efficac
efficacyy of mitoxantrone
mitoxantrone in allevia
alleviating
ting TM subacute myelopathy (Level U).
attacks (single Class III studies). Brain MRI characteristics consistent with those of 
MS may be considered useful to predict conversion
Rituximab.   Two uncontrolled, open-label Class
to MS after a first episode of partial TM (Level C).
III studies evaluated the use of rituximab in a com-
Longer spinal lesions extending over more than 3
bined total of 26 patients with NMO meeting es-
vertebral segments may be considered useful in deter-
tablished diagnostic criteria.e9,e10 The majority of 
mining NMO vs MS (Level C).
patients included in both studies had experienced
NMO-IgG
NMO-I gG antib
antibodies
odies should be consi
considered
dered use-
relapses despite treatment with one or more im-
ful to determine the cause of TM in patients present-
munotherap
munotherapies ies prior
In the first study e9 to 8treatment
6 of with attack-free
patients were rituximab. ing with clinical features of ACTM (Level B).
CSF examination for cells and OCBs may be con-
over the follow-up period (mean      12 months),
sidered useful to determine the cause of the TM syn-
and the reported median attack rate (attacks/pa-
drome (Level C).
tient/year) fell from 2.6 to 0 ( p  0.0078). Two of 
The presence of NMO-IgG antibodies (aqua-
the patients experienced a TM episode subsequent
porin-4–specific antibodies) should be considered
to rituximab treatment. In the second study, e10
useful in determining an increased risk of TM recur-
 which
 whic h inclu
included
ded 7 of the patie
patients
nts enro
enrolled
lled in the
rence (Level B).
first study, the median annualized relapse rate de-
Plasma
Plasma exc
exchan
hange
ge may be co
consi
nsider
dered
ed in pat
patien
ients
ts wit
with
h
creased from 1.7 to 0 after treatment with ritux-
TM who fail to improve after corticosteroid treatment
imab ( p      0.001) during the median follow-up
(Level C). Rituximab may be considered in patients
period of 19 months. Two patients died during 
 with TM due to NMO to decrease
decrease the numb
numberer of re-
the follow-up period, 1 during a brainstem relapse
lapses (Level C). There is insufficient evidence to sup-
and 1 from suspected septicemia.
Conclusions. Rituximab is possibly effective in reducing  port or refute the efficacy of other therapies (Level U).
TM atatta
tack
ckss in pa
pati
tien
ents
ts wi
with
th NM
NMOO (2 Cl
Clas
asss II
IIII st
stud
udies
ies).
). RECOMMEN
RECOMMENDATI
DATIONS
ONS FOR FUTU
FUTURE
RE RESE
RESEARCH
ARCH
agents..   Case reports, small case series, and
Other agents The efficacy of acute therapies, aimed at rapid inter-
retrospect
retrospective
ive reviews have sugges
suggested
ted poten
potential tial bene- vention in acutely declining patients, should be ex-

2132   Neurology
Neurology 77 Decem
December
ber 13, 2011
 

amined prospectively and should be distinguished exclude any reasonable alternative methodologies. The AAN recog-
nizes that specific patient care decisions are the prerogative of the
from efficacy of long-term therapies aimed at preven-
patient and the physician caring for the patient, based on all of the
tion of relapse. These cohort studies should prospec- circumstances involved. The clinical context section is made available
tively examine, over at least a 3-year period, the in order to place the evidence-based guideline(s) into perspective with
clinical features of partial and complete TM, the lon- current practice habits and challenges. No formal practice recommen-
dations should be inferred.
gitudinal extent of MRI lesions, the presence of 
NMO antibodies or other laboratory information, CONFLICT OF INTEREST
and the presence or absence of cerebral lesions typical The American Academy of Neurology is committed to producing inde-
of MS to predict prognosis for recovery and relapse pendent, critical, and truthful clinical practice guidelines (CPGs). Signifi-
risk. Discriminant function analysis should be used cant efforts are made to minimize the potential for conflicts of interest to
influence the recommendations of this CPG. To the extent possible, the
to determine which clinical features of idiopathic
 AAN keeps separate those who have a financial s take in the success or
TM clearly differentiate that condition from MS failure of the products appraised in the CPGs and the developers of the
 with myelopathy. guidelines. Conflict of interest forms were obtained from all authors and
Randomized trials of therapeutic interventions for reviewed by an oversight committee prior to project initiation. AAN lim-
its the participation of authors with substantial conflicts of interest. The
TM, such as plasma exchange or immunosuppres-  AAN forbids commercial participation in, or funding of, guideline proj-
sants, should be performed using corticosteroid ther- ects. Drafts of the guideline have been reviewed by at least three AAN
apy as the gold standard for comparison and both committees, a network of neurologists, Neurology  peer
 peer reviewers, and rep-
resentatives from related fields. The AAN Guideline Author Conflict of 
recovery and relapse as outcomes to be analyzed.
Interest Policy can be viewed at www.aan.com.

AUTHOR CONTRIBUTIONS
Received May 16, 2011. Accepted in final form August 26, 2011.
Dr. Scott: drafting/revising the manuscript, study concept or design, anal-
ysis or interpretation of data, acquisition of data, statistical analysis, study 
supervision. Dr. Frohman: drafting/revising the manuscript, study con-
REFERENCES
cept or design, analysis or interpretation of data, acquisition of data, sta- 1. Basti
Bastian
an HC. Thrombotic
Thrombotic softeni
softening
ng of the spinal
spinal cord:
cord: a 
tistical analysis, study supervision, obtaining funding. Dr. De Seze: case of so-called “acute myelitis.” Lancet 1910;ii:1531–
drafting/revising the manuscript, study concept or design, analysis or in- 1534.
terpretation of data, acquisition of data, study supervision. Dr. Gronseth: 2. Rivers TM. Viruses
Viruses.. JAMA
JAMA 1929;92:1147–1
1929;92:1147–1152. 152.
drafting/revising the manuscript, analysis or interpretation of data, statis- 3. For
Fordd FR. The nervous
nervous compli
complicat
cation
ionss of mea
measle
sles:
s: with a 
tical analysis. Dr. Weinshenker: drafting/revising the manuscript. summary of literature and Publications of 12 additional
case reports. Bull Johns Hopkins Hosp 1928;43:140–184.
DISCLOSURE 4. Paine RS, RS, Byers RK.
RK. Transverse
Transverse myelopa
myelopathythy in childhoo
childhood.d.
Dr. Scott has received funding for travel or speaker honoraria from, served  AMA Am J Dis Child 1953;85:151–163.
195 3;85:151–163.
on the speakers’ bureaus and scientific advisory boards of, and performed 5. Ber
Berman
man M, Feldma
Feldman n S, Alter M, Zilber
Zilber N, Kahana
Kahana E.
consultation work for Acorda Therapeutics Inc., Avanir Pharmaceutic
Pharmaceuticals,
als,
 Acute transverse myelitis: incidence and etiologic consider-
Biogen Idec, Novartis, and Teva Pharmaceutical Industries Ltd.; served as
ations. Neurology 1981;31:966–971.
an associate editor for  BMC Neurology ; and has received research support
from Biogen Idec, National Multiple Sclerosis Society, Novartis, Pitts-
6. Tran
Transver
sverse
se Myelitis
Myelitis Consortiu
Consortium m Wor
Working
king Group.
Group. Pro-
burgh
burg h Foun
Foundatidation,
on, and Teva Pharmaceuti
Pharmaceuticalcal Indu
Industri
stries
es Ltd. Dr. posed diagnostic criteria and nosology of acute transverse
Frohman has received funding for travel and/or speaker honoraria from myelitis. Neurology 2002;59:499–505.
Biogen Idec, Teva Pharmaceuti
Pharmaceutical cal Industries Ltd., Genzyme Corporation, 7. Scott TF. Nosolog
Nosologyy of idiopathic
idiopathic transverse
transverse myeliti
myelitiss syn-
 Abbott, Acorda Therapeutics
Therapeutics Inc., and Bayer Schering
Schering Pharma; has served dromes. Acta Neurol Scand 2007;115:371–376.
on speakers’ bureaus for Biogen Idec, Teva Pharmaceutical Industries 8. de Seze J, Stojko
Stojkovic
vic T, Breteau
Breteau G, et al.
al. Acute myelop
myelopa-a-
Ltd., Bayer Schering Pharma, and Athena Diagnostics; and has served as a  thies: clinical, laboratory and outcome profiles in 79 cases.
consultant for Biogen Idec, Teva Pharmaceutical Industries Ltd., Athena  Brain 2001;124:1509–1521.
Diagnostics, Inc., Acorda Therapeutics Inc., and Abbott. Dr. de Seze
9. Jeffer
Jefferyy DR, Mandler
Mandler RN, Davis LE. LE. Transverse
Transverse myelitis
myelitis
serves on scientific advisory boards for and has received honoraria from
(retrospective analysis of 33 cases, with differentiation of 
Biogen Idec, LFB, Merck Serono, sanofi-aventis, and Bayer Schering 
Pharma; and serves on the editorial board of  Revue   Revue Neurologique . Dr. cases associated with multiple sclerosis and parainfectious
Gronseth serves as an editorial advisory board member of  Neurology Now , events). Arch Neurol 1993;50:532–535.
serves on a speakers’ bureau for Boehringer Ingelheim, and receives hono- 10. Mas
Masson
son C, Pruvo
Pruvo JP, Meder
Meder JF, et al, Study
Study Group
Group on
raria from Boehringer Ingelheim and the American Academy of Neurol- Spinal Cord Infarction of the French Neurovascular Soci-
ogy. Dr. Wein
Weinshenk
shenker
er serve
servess on data safety monitoring
monitoring boards for ety. Spinal cord infarction: clinical and magnetic resonance
Novartis and Biogen Idec; serves on the editorial boards of the  Canadian imaging findings and short term outcome. J Neurol Neu-
 Journal of Neurological Sciences 
Sciences , the Turkish Journal of Neurology , and Mul- rosurg Psychiatry 2004;75:1431–1435.
tiple Sclerosis ; has received research support from Genzyme Corporation 11. de Seze J, Lancti
Lanctin n C, Lebrun C, et al. Idiopat
Idiopathic
hic acute
acute
and the Guthy-Jackson Charitable Foundation; and receives license royal-
transverse myelitis: application of the recent diagnostic cri-
ties from RSR Ltd. for a patent re: Aquaporin-4 associated antibodies for
teria. Neurology 2005;65:1950–1953.
diagnosis of neuromyelitis optica.
12.. Har
12 Harzhe
zheim
im M, Schlegel
Schlegel U, Urb
Urbach
ach H, Klockgeth
Klockgetherer T,
Schmidt S. Discriminatory features of acute transverse my-
DISCLAIMER
This statement is provided as an educational service of the American elitis: a retrospective analysis of 45 patients. J Neurol Sci
 Academy of Neurology. It is based on an assessment of current scientific 2004;217:217–223.
and clinical information. It is not intended to include all possible proper 13. Scott TF,
TF, Bhagavatula
Bhagavatula K, Snyder
Snyder PJ, et al.
al. Transverse
Transverse my-
my-
methods of care for a particular neurologic problem or all legitimate crite- elitis comparison with spinal cord presentations of multi-
ria for choosing to use a specific procedure. Neither is it intended to ple sclerosis.
sclerosis. Neurol
Neurology
ogy 1998;50:429– 433.

Neurology
Neurology 77 Dece
December
mber 13, 2011   2133
 

14. Bakshi
Bakshi R, Kinkel PR,PR, Mechtler
Mechtler LL, et al.
al. Magnetic
Magnetic reso-
reso- 28. Pittock
Pittock SJ, Lennon
Lennon VA, VA, Krecke K,K, Wingerchuk
Wingerchuk DM, DM, Luc-
nance imaging findings in 22 cases of myelitis: comparison chinetti CF, Weinshenker BG. Brain abnormalities in neu-
between patients with and without multiple sclerosis. Eur romyelitis optica. Arch Neurol 2006;63:390–396.
 J Neurol 1998;5:35– 48. 29. Barkho
Barkhoff F, Filippi
Filippi M, Miller DH, DH, et al. Comparison
Comparison of 
15. Cabre P, Signat
Signatee A, Olindo
Olindo S, et al.
al. Role of
of return
return migra-
migra- MRI criteria at first presentation to predict conversion to
tion in the emergence of multiple sclerosis in the French clinically
clinically defin
definite
ite multi
multiple
ple sclero
sclerosis.
sis. Brain 1997;120:
 West Indies. Brain 2005;128:2899 –2910. 2059–2069.
16. Fukazawa T. What are the unique characteristics of optico-
optico- 30. Winge
Wingerchuk
rchuk DM, Lennon
Lennon VA, Pittock
Pittock SJ,
SJ, et al. Revised
Revised
spinal MS in Japanese and neuromyelitis optica in western
diagnostic criteria for neuromyelitis optica. Neurology 
populations. Rinsho Shinkeigaku 2006;46:863–865.
2006;6:1485–1489.
17. Ford B, Tampieri
Tampieri D, Francis
Francis G. Long-ter
Long-term m follow-up
follow-up of 
31. Lenno
Lennon n VA, Wingerchuk
Wingerchuk DM, DM, Kryzer
Kryzer TJ, et al.
al. A serum
acute partial transverse myelopathy. Neurology 1992;42:
autoantibody marker of neuromyelitis optica: Distinction
250–252. from multiple sclerosis. Lancet 2004;364:2106–2112.
18. Scott TF,
TF, Kassab
Kassab SL, Singh S. S. Acute partial
partial transve
transverse
rse my-
32. Tan
Tanaka
aka K, Tani T, Tanaka
Tanaka M, et al. Anti-aq
Anti-aquap
uapori
orin
n4
elitis with normal cerebral magnetic resonance imaging:
antibody in selected Japanese multiple sclerosis patients
transition rate to clinically definite multiple sclerosis. Mult
Scler 2005;11:373–377.  with long spinal cord lesions. Mult Scler 2007;13:850 –
19. O’Rio
O’Riordan
rdan JI,
JI, Thompson
Thompson AJ, Kinglse
Kinglseyy DP, et al.
al. The prog- 855.
nostic value of brain MRI in clinically isolated syndromes of  33.. Tak
33 Takaha
ahashi
shi T, FujFujiha
ihara
ra K, Nak
Nakash
ashima
ima I, et al. Anti-
Anti-
the CNS: a 10-year follow-up. Brain 1998;121:495–503. aquaporin-4 antibody is involved in the pathogenesis of 
20. Scott TF,
TF, Kassab
Kassab S, Pittock SJ.SJ. Neuromyeliti
Neuromyelitiss optica anti-
anti- NMO: a study on antibody titre. Brain 2007;130:1235–
bodiess and acute parti
bodie partial
al trans
transverse
verse myelitis.
myelitis. Arch Neurol 1243.
2006;63:1398–1340. 34. Paul F, Jarius
Jarius S, Aktas
Aktas O, et al. Antibody
Antibody toto aquaporin
aquaporin 4 in
21. Nakas
Nakashima
hima I, Fujihara
Fujihara K,K, Miyazawa
Miyazawa I, et al.
al. Clinical
Clinical and the diagnosis of neuromyelitis optica. Plos Med 2007;4:
MRI features of Japanese patients with multiple sclerosis 669–674.
positive for NMO-IgG. J Neurol Neurosurg Psychiatry  35. Sai
Saizz A, Zul
Zulian
ianii L, Bla
Blanco
nco Y, Tavolato
Tavolato B, Giomett
Giomettoo B,
2006;77:1073–1075. Graus F, Spanish-Italian NMO Study Group. Revised di-
22. Mats
Matsuoka
uoka T, Matsush
Matsushita
ita T, Kawano
Kawano Y, et al. Heterogen
Heterogene-e- agnostic criteria for neuromyelitis optica (NMO). J Neurol
ity of aquaporin-4 autoimmunity and spinal cord lesions in 2007;254:1233–1237.
multiple sclerosis in Japanese [see comment]. Brain 2007; 36..
36 Tana
Ta naka
ka M, Ta Tana
naka
ka K, KoKomo
moriri M, Sa Said
idaa T. Anti-
Anti-
130:1206–1223. aquaporin 4 antibody in Japanese multiple sclerosis: the
23. Wein
Weinshenk
shenkerer B, Wingerchu
Wingerchukk D, Vukusic
Vukusic S, et al.
al. Neuromy-
Neuromy- presence of optic-spinal multiple sclerosis without long 
elitis Optica IgG predicts relapse following longitudinally ex-
spinal cord lesions and anti-aquaporin 4 antibody. J Neu-
tensive transverse myelitis. Ann Neurol 2006;59:566–569.
rol Neurosurg Psychiatry 2007;78:990–992.
24. Morri
Morrissey
ssey SP,
SP, Miller DH,DH, Kendall
Kendall BE, et al. The The signifi-
signifi-
37. Banwel
Banwelll B, Tenembaum
Tenembaum S, Lennon
Lennon VA,
VA, et al. Neuromye
Neuromyeli-
li-
cance of brain magnetic resonance imaging abnormalities
tis optica-IgG in childhood inflammatory demyelinating 
at presentation with clinically isolated syndromes sugges-
CNS disorders. Neurology 2008;70:344–352.
tive of multiple sclerosis. Brain 1993;116:135–146.
38. Marig
Marignier
nier R, de Seze
Seze J, Vukusic
Vukusic S, et al.
al. NMO-IgG
NMO-IgG andand
25. Jaco
Jacobs
bs LD, Beck
Beck RW, Simon
Simon JH, et al. Intram
Intramuscula
uscularr inter-
inter-
feron beta
beta-1a
-1a thera
therapy
py init
initiate
iated
d duri
during
ng a firs
firstt demy
demyelin
elinatin
atingg even
eventt Devic’s neuromyelitis optica: a French experience. Mult
in mul
multitiple
plescl
sclero
erosi
sis.
s. N Eng
Engll J Me
Medd 200
2000;3
0;343:
43:898–904
898–904.. Scler 2008;14:440–445.
26. Cor
Cordon
donnie
nierr C, de Seze J, Breteau
Breteau G, et al. Prospec
Prospectivtivee 39. Wi
Winge
ngerch
rchuk
uk DM
DM,, Ho
Hogan
gancam
campp WF
WF,, O’B
O’Brie
rien
n PC,
PC,Wei
Weinsh
nshenk
enker
er
study of patients presenting with acute partial transverse BG. The clinical course of neuromyelitis optica (Devic’s
myelopathy. J Neurol 2003;250:1447–1452. syndrome). Neurology 1999;53:1107–1114.
27. Campi A,
A, Filippi
Filippi M, Comi G,G, et al. Acute
Acute transverse
transverse my-
my- 40. Perumal
Perumal J, Zabad
Zabad R, Caon
Caon C, et al.
al. Acute transv
transverse
erse myeli-
myeli-
elopathy: spinal and cranial MR study with clinical follow- tis with normal brain MRI: long-term risk of MS. J Neurol
up. Am J Neuroradiol 1995;16:115–123. 2008;255:89–93.

Endorsed by the Consortium of MS Centers on October 19, 2011. The American Academy of Neurology is to be
congratulated for thoughtfully assessing the current status of evidence-based information and for revealing the
need for much work before credible algorithms can be applied to the treatment of this syndrome.

2134   Neurology
Neurology 77 Decem
December
ber 13, 2011
 

Evidence-based guideline: Clinical evaluation and treatment of transverse myelitis:


Report of the Therapeutics and Technology Assessment Subcommittee of the American
Academy of Neurology
T.F. Scott, E.M. Frohman, J. De Seze, et al.
 Neurology 2011;77;2128-2134 Published Online before print December 7, 2011
DOI 10.1212/WNL.0b013e31823dc535
10.1212/WNL.0b013e31823dc535

This information is current as of December 7, 2011

Updated Information & including high resolution figures, can be found at:
Services http://www.neurology.org/content/77/24/2128.full.html 

Supplementary
Supplementary Material Supplementary material can be found at:
http://www.neurology.org/content/suppl/2011/12/11/WNL.0b013e3182
3dc535.DC2 
http://www.neurology.org/content/suppl/2011/12/07/WNL.0b013e3182
3dc535.DC1 
http://www.neurology.org/content/suppl/2013/01/03/WNL.0b013e3182
3dc535.DC3 
References This article cites 40 articles, 11 of which you can access for free at:  
http://www.neurology.org/content/77/24/2128.full.html##ref-list-1 
Citations This article has been cited by 2 HighWire-hosted articles:
http://www.neurology.org/content/77/24/2128.full.html##otherarticles 
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Transverse myelitis
http://www.neurology.org//cgi/collection/transverse_myelitis 
Permissions & Licensing Information about reproducing this article in parts (figures,tables) or in
  its entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions 
Reprints Information about ordering reprints can be found online:
online:
  http://www.neurology.org/misc/addir.xhtml#reprintsus 

 Neurology ® is the official journal of the American Academy of Neurology. Published continuously since
1951, it is now a weekly with 48 issues per year. Copyright Copyright © 2011 by AAN Enterprises, Inc.. All
rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.

You might also like