Neuroimmunology in Clinical Practice - part 5 potx

Acute disseminated encephalomyelitis 103
presence of red blood cells (Leake and Billman,
2002). Imaging studies disclose fairly symmetrically
distributed hemorrhages in deep white matter, some
of which may coalesce into large somewhat asym-
metrical hemorrhagic areas. Thalami, hypothalamus,
brainstem, cerebellum, and spinal cord may also
be involved, while cerebral cortex and basal ganglia
tend to be spared. Involved tissues may be quite
edematous.
The disease is fulminant with death usually ensu-
ing within hours to days. A number of reports of
favorable response to high doses of intravenous
corticosteroids (Seales and Greer, 1991) and appro-
priate additional treatments of increased intracra-
nial pressure suggest that AHLE may be treatable if
recognized quickly. Plasmapheresis and cyclophos-
phamide have also been employed with apparent
effectiveness in a few cases. Rapid and effective treat-
ment requires recognition of the presence of AHLE, a
diagnosis that has been confirmed in some instances
by urgent brain biopsy, followed by successful treat-
ment (McLeod, 2001). Craniectomy appears to have
been beneficial in some cases with severe elevation of
intracranial pressure.
Severe cases have become rarer due to the avail-
ability and increased safety of vaccines, permitting
children to be effectively vaccinated against many
illnesses associated with the possibility of AHLE. How-
ever, small quantities of red blood cells (10–500 /mm
3

)
that are often found in lumbar CSF of children with
ADEM suggests the possibility that milder degrees
of AHLE, falling below the resolution of scan tech-
niques, continue to occur. Moreover, it has been
compellingly suggested that cerebral malaria is the
result of a hyperergic host response with the pro-
duction of AHLE (Toro and Roman, 1978).
Concentric sclerosis, Balò type (CSBT; encephalitis
periaxialis concentrica)
This illness, which may be a peculiar hyperacute
variant of MS, may in addition be influenced by
genetic factors that may admix features of both MS
and ADEM (Caracciolo et al., 2001; Itoyama et al.,
1985; Kira et al., 1996; Moore et al., 1985; Sotgiu
et al., 2001). The rarity of this illness appears to be
on the order of Schilder disease and its etiological
basis similarly uncertain. Peak incidence is in the
third decade of life, although cases may arise as early
as the second and as late as the sixth decade. The male
to female ratio is about 1:2. CSBT appears to have
higher incidence in Taiwan, Japan, or the Philippines
(Tabira and Nishizawa, 1990; Tabira, 1994).
Mild prodromal fever, malaise, and headache are
noted in approximately half of CSBT cases, followed
by behavioral withdrawal. The acute behavioral
change is associated with weakness and numbness,
initially on one side of the body (face, limb, or trunk)
that then worsens in extent and degree. Develop-
ment of pyramidal or cerebellar signs and deteriora-

tion of higher cortical and oropharyngeal functions
ensue. A Dévic syndrome phenotype is occasionally
found (Currie et al., 1970). Generalized convulsions
occur in approximately 25% of cases. The clinical
course may suggest ADEM, especially in younger
individuals with this condition. Blood and CSF tests
reveal little information of diagnostic importance;
mild CSF pleocytosis may be found (Tabira, 1994).
Historically, diagnosis was ascertained in severe
cases by postmortem or brain biopsy. Pathological
study of the concentric layers of greater or lesser
degrees of demyelinative and remyelinative inflamma-
tion suggest similarities with the appearance of the
• Para-infectious (e.g. Influenza A, RSV,
mycoplasma, varicella, measles and
other exanthematous illnesses)
• Postvaccination (especially Pasteur
rabies vaccine; hepatitis B)
• Encephalitis (HSV, HHV6, VZV)
• Fungal cerebritis (e.g. Scopulariopsis
phaeohyphomycosis)
• Drug or toxin exposure
(e.g. intrathecal methotrexate)
• Blood dyscrasias
(e.g. acute myeloid leukemia)
• Fat embolism
• Nutritional deficiencies
• Ulcerative colitis
• Acute rheumatic fever
• Membranoproliferative

glomerulonephritis, acute
tubular necrosis
• Asthma
Box 5.13 Etiological association of acute hemorrhagic leukoencephalitis.
NICP_C05 04/05/2007 12:26PM Page 103
104 ROBERT S. RUST
edge of MS plaques (Courville and Cooper, 1970;
Moore et al., 1985; Yao et al., 1994).
MRI scans have proven useful in identification of
milder cases of CSBT. Between 3–5 fairly symmet-
rical lesions 1.5–5 centimeters in diameter are typic-
ally found, more commonly in the deep cerebral
(frontoparietal > temporal) or cerebellar white matter
than in other rarer locations, such as spinal cord
(Currie et al., 1970). During the early acute stage of
CSBT gadolinium enhancement may clearly delin-
eate alternating concentric zones of greater or lesser
inflammation, quite distinct from what is seen in
typical ADEM or MS. In the late acute stage only a
single ring of enhancement at the outer margin of
lesions may be found more closely resembling MS
(which the age at presentation usually suggests) or
ADEM (which the febrile prodrome may suggest).
The concentric rings may reappear during the ensu-
ing subacute phase of illness (Caracciolo et al., 2001;
Chen et al., 1999; Chen, 2001; Ghatak et al., 1989).
MR spectroscopic abnormalities of CSBT resemble
those found in MS plaques, although similar changes
are found in other diseases, including mitochondrial
cytopathies (Chen, 2001). In addition to ADEM and

MS, tumor, vasculitis, or infection may resemble CSBT
(Caracciolo et al., 2001). Caution is especially import-
ant where the pattern of alternating enhancement is
less centrifugal and more irregular in arrangement.
Biopsy may sometimes prove misleading.
Reported CSBT cases, chiefly from the pre-MRI
era, have generally proved fatal within 2–60 weeks.
Early deaths are due to herniation, late deaths to
inanition and secondary infections (Courville and
Cooper, 1970; Tabira, 1994). Milder cases identified
by MRI and increased availability of supportive ther-
apies have modestly improved the overall outlook for
this condition. Early provision of immunosuppressive
treatments may ameliorate clinical and imaging
abnormalities. Some cases arising in fourth–sixth
decades have longer survival and more prominent
gliosis of demyelinated layers.
Optic neuritis (ON)
Optic neuritis is considered in detail elsewhere in this
volume. In this section childhood ON and its rela-
tionship to ADEM and MS will briefly be reviewed.
ON may also occur in association with various
inflammatory illnesses other than ADEM or MS
(Kazarian and Gager, 1978; O’Halloran et al., 1998;
Riedel et al., 1998). ON is rare prior to six years of age
and more common from six years of age to puberty.
Prior to puberty it usually occurs in association
with ADEM or NMO. In approximately 70% of cases,
acute visual loss occurs days to weeks after a non-
specific viral illness (especially measles, mumps, and

varicella) or immunization (Kazarian and Gager, 1978;
Kline et al., 1982; Purvin et al., 1988; Riikonen,
1989). After puberty, it may occur in isolation or in
association with NMO or MS and the association
with a prodromal illness is less common. In isolated
postpubertal cases the risk for subsequent diagnosis
of MS is approximately 50%.
Visual loss of childhood ON may be preceded
by headache (frontal or ocular), scintillating scot-
omata, or painful eye movements. Visual loss may be
unilateral or bilateral. Three-quarters of prepubertal
cases develop bilateral changes either simultaneously
or sequentially, the changes in the second eye lagging
behind the first by weeks to months. Degree of visual
loss is usually fairly symmetrical in bilateral cases,
however, in a significant minority of cases it is asym-
metrical (Riikonen et al., 1988). Initially the visual
disturbance may be limited to visual blurring with
progression over several days to partial to complete
visual loss. In cases with partial visual field loss
there may be a particularly dense central scotoma.
Swelling of the optic nerve head (papillitis) is more
common in children than it is in adults with ON,
occurring in at least two-thirds of cases (Kriss et al.,
1988; Parkin et al., 1984). Quite striking abnormal-
ities, including fiber layer hemorrhages at the optic
nerve margin, vascular tortuosity, or sheathing of
veins, are readily observable on funduscopy in many
cases (Riikonen et al., 1988).
These changes may suggest papilledema, how-

ever, the visual loss of ON can, in most cases, readily
be distinguished from that due to malignantly
increased intracranial pressure (ICP). Increased ICP
is usually associated with additional neurological signs
(e.g. sixth nerve palsy, meningismus). It manifests
slower onset and as it usually provokes less profound
degree of visual loss is less frequently associated with
an afferent pupillary defect. As in many cases of
adult ON, childhood ON may occur without observ-
able funduscopic changes, in which case the term
retrobulbar ON is applied.
The diagnosis of ON is made on the basis of a com-
bination of clinical and laboratory findings. In subtle
cases, diagnosis can be clinically supported by loss of
red vision (“red desaturation”), or by loss of duration
or variety of the “flight of colors” that are appre-
hended in a dark room immediately after 60 seconds
of stimulation of a retina with bright light. Greater
NICP_C05 04/05/2007 12:26PM Page 104
Acute disseminated encephalomyelitis 105
degrees of visual loss are signified by the presence
of an afferent papillary defect (APD – loss of the
reflexive constriction of the contralateral pupil when
the retina of the affected eye is illuminated). Visual
evoked responses (VER) are particularly useful
where visual loss is mild enough to be uncertain.
ON results in increased latency of the positive com-
ponent of this cortical response (Feinsod et al.,
1975). Delayed VER may persist for several years in
patients who have shown excellent clinical recovery

(Aicardi, 1992).
Abnormalities of other portions of the nervous
system should be sought and if found a more general
diagnosis (e.g. ADEM, Dévic syndrome, Guillain–
Barré syndrome (GBS), MS) should be applied on the
basis of clinical features and history. Positivity of the
CSF immune profile studies noted above (excepting
myelin basic protein) or of other studies such as
CSF free kappa chains (Rudick et al., 1986; Riikonen
et al., 1988a) favors the diagnosis of MS but does not
exclude the diagnosis of ADEM. MRI scanning of
brain and brainstem with appropriate weighting
(T1, T2, balanced, and with gadolinium admin-
istrations) and special orbital views is important.
MRI demonstrates swelling of the optic nerve in
most cases; the extent of optic nerve enlargement
may be alarming in some children who nevertheless
experience good recovery.
MRI is the most important tool in excluding altern-
ative diagnoses such as lesions compressing the optic
nerve. Disseminated T2 bright lesions may be found
by MRI elsewhere in the brain in as many as 70% of
patients (Riikonen et al., 1988), interpretation of
such changes in children is difficult and may not
indicate MS. Where such abnormalities are at the
gray–white junction and the patients are younger,
ADEM is suggested. Periventricular plaques (espe-
cially if perpendicular to the ventricular surface) are
more suggestive of MS (Ormerod et al., 1986). SSPE,
intoxications (e.g. methanol), leukodystrophies, and

stroke must occasionally be considered. Malingering
may be excluded on the basis of inconsistencies on
examination or with VER testing.
Generally, recovery from idiopathic childhood ON
is excellent, although the rate of recovery may be
slow (Good et al., 1992). The most common residua
include optic nerve atrophy and impairments of color
and stereoscopic vision (Parkin et al., 1984; Purvin
et al., 1988). Permanent severe visual loss is quite
exceptional. Bilateral presentation after an ante-
cedent illness or immunization usually (although not
always) implies good prognosis for visual recovery
(Parkin et al., 1984; Riikonen et al., 1988b; Good
et al., 1992). Postpubertal ON is more likely to have
residual deficits.
In adults, treatment with intravenous high-dose
methylprednisolone of first bout of ON hastens
recovery and may prolong the time to diagnostic
recurrence of MS. Oral prednisone provides no benefit
and may heighten odds of early recurrence of a bout
diagnostic of MS (Beck et al., 1992; Beck et al., 1993;
Silberberg, 1993). These cases were chiefly first
bouts of MS and the relevance of this data to
children, many of whom do not go on to develop
MS, is unclear. Recovery occurs with or without
anti-inflammatory therapy in most children and
there is little evidence that final recovery is favorably
influenced by treatment. High-dose intravenous
treatment for 3–5 days in cases of quite profound
optic nerve swelling may prevent ischemic injury

and other childhood ON may manifest more rapid
recovery with such treatment. Limited data sug-
gests treatment may reduce chances for ultimate
diagnosis of MS ( Jacobs et al., 1994).
ON may recur. Various studies, following up for
2–18 years, have estimated a 0–60% risk for MS if a
bout of ON occurs before 18 years of age (Kriss et al.,
1988; Parkin et al., 1984; Riikonen et al., 1988). A
more refined estimate suggests 15–30% overall risk
(ON Study Group, 1997), chiefly sustained by those
over 12 years of age, in whom there is about a 50%
risk. Poor or incomplete visual recovery, itself chiefly
a postpubertal phenomenon, also implies a 50%
risk for ultimate diagnosis of MS (Good et al., 1992).
The presence of lesions consistent with MS plaques
in typical locations (periventricular, forceps major
and minor) on MRI increases the risk for subsequent
diagnosis of CD-MS to at least 75–80%. The presence
of oligoclonal bands in the CSF also increases the
risk for diagnosis of MS within five years, although
not so decisively as the MRI features just noted.
Oligoclonality increases risk even where the MRI
is normal (Cole et al., 1998). Risk of MS is high in
unilateral ON (which is mostly postpubertal) and
trivial in bilateral prepubertal cases (Parkin et al.,
1984). In cases where ON is associated with ADEM,
GBS, or Dévic syndrome, the prognosis should be
determined on the basis of the more disseminated
illness, but visual recovery is usually good.
Acute transverse myelitis (ATM)

Various causes for acute childhood/adolescent myel-
opathy are listed in Box 5.14. The most common
NICP_C05 04/05/2007 12:26PM Page 105
106 ROBERT S. RUST
causes are inflammatory, traumatic, or vascular.
In children, post-infectious/post-vaccination inflam-
matory myelitis is a particularly important category.
These cases are often a form of ADEM, including Lyme
myelitis (Kerr and Ayetey, 2002; Rousseau et al.,
1986; Tyler et al., 1986). Infections precede ATM by
days to several weeks in 60% of cases (Paine and
Byers, 1953). The additional history of blunt trauma
to the spine is not infrequently recalled. The cervical
location is a common one as is thoracic. Many more
levels of the spine may be involved than is typical of
MS. In some instances the entire spine is involved
as well as some of the brainstem. In some instances
the inflammatory sensitization involves both central
and peripheral (e.g. spinal root) myelin (Abramsky
and Teitelbaum, 1977). The irreversible injury with
myeloclasia that complicates severe ATM is likely
to be vascular: due to the ischemia induced by
cord swelling within the confined space of the spinal
canal. ATM that occurs in the first few years of life
may be particularly malignant, but most cases occur
in children more than five years of age (Aicardi,
1992; Berman et al., 1981).
Pain and dysaesthesiae in the region of the devel-
oping ATM are the most common early symptoms.
Fever and meningismus may then follow. Paraplegia,

sensory loss, and sphincter dysfunction may develop
slowly over days to many weeks or paroxysmally
within several hours. The rate of onset is often pro-
portional to the intensity of the initial discomfort.
Intense pain in neck presaging hyperacute cervical
ATM is a medical emergency that may have a fatal
outcome due to cardiorespiratory compromise. In
most instances bilateral flaccid areflexic paraparesis
with a sensory level and sphincter dysfunction develop,
followed in a few days by spastic weakness in the
same distribution. Superficial reflexes (abdominal,
cremasteric, bulbocavernosus) are usually absent.
Partial spinal cord syndromes (e.g. Brown–Sequard
syndrome) or Dévic syndrome may be found. Rarely
ATM presents with the isolated complaint of urinary
retention (Ropper and Poskanzer, 1978). ATM with
febrile infectious prodrome and associated constitu-
tional symptoms is more common in prepubertal
patients and suggests ADEM. ATM without these
associated features is more common in adolescents
and is suggestive of MS. MS tends to provoke a less
complete form of myelitis than ADEM. The MRI of
patients with MS-related myelitis typically demon-
strates T2 bright signal abnormality of some, but not
all, areas of the cord that are enriched with myelinated
fibers. The outlook for recovery may be poorer and
that for severity of MS may be greater in adolescents
that are found to have signal abnormality through
many rather than few levels of the spinal cord.
Adolescent patients with purely spinal manifesta-

tions suggesting MS should be screened for tropical
spastic paraparesis/HTLV-1 associated myelopathy
(TSP/HAM), while men should be evaluated for
adrenomyeloneuropathy (Walther and Cutler, 1997).
• Isolated ATM
– Abscess*
– Hemorrhage
– Stroke (vascular malformation,
compressive, embolic*)
– Radiation
– Tumor (spinal,* spinal root,
meningeal, vascular, bone)
– Trauma*
• More widespread neurological disease
– Dévic syndrome (see below)***
– Encephalomyeloradiculoneuropathy***
– Multiple sclerosis***
– ADEM***
– Guillain–Barré syndrome
– Neurosarcoidosis**
– Tropical spastic paraparesis**
(Link et al., 1989)
• In association with systemic disease
– AIDS vascular myelopathy*,**
(Rosenblum et al., 1989)
– Chronic progressive (third stage)
Lyme neuroborreliosis**,***
– Systemic lupus erythematosus**
– Syphilis
*Seldom if ever found in children; **Imaging changes

may resemble ADEM or MS; ***May be manifestation
of ADEM or MS.
Interferon beta-1b in the treatment of multiple
sclerosis: final outcome of the randomized controlled
trial. The IFNB Multiple Sclerosis Study Group and The
University of British Columbia MS/MRI Analysis Group,
Neurobrucellosis.
Box 5.14 Causes of acute transverse myelitis.
NICP_C05 04/05/2007 12:26PM Page 106
Acute disseminated encephalomyelitis 107
Some combination of history, examination, imag-
ing studies, and CSF and serum tests discloses an
etiological diagnosis (from among those listed in
Box 5.14) for ATM in approximately two-thirds of
the cases encountered in children and adolescents.
MS-associated ATM is almost entirely confined to
postpubertal individuals, ADEM-related ATM to pre-
pubertal individuals. Imaging studies of brain and
spinal cord are important in order to disclose dis-
tribution and in some instances cause of disease, as
well as lesions or edema that require urgent therapy
in order to prevent irreversible ischemic injury.
Unsuspected brain lesions may be found in MS, ADEM,
neurosarcoidosis, and other diagnoses (Miller et al.,
1987; Sanders et al., 1990). Myelography is some-
times helpful (Narciso et al., 2001). CSF pleocytosis
is present in 25%, increased CSF protein in 50% of
presumed ATM cases (Aicardi, 1992).
No therapy, including corticosteroids, has been
rigorously proven to be efficacious in the treatment

of ATM. Management is largely symptomatic, with
particular attention to careful management of such
associated problems as urinary retention and impaired
breathing. Pain and dysaesthesiae may be trouble-
some and vigorous attempts should be made to
treat these symptoms, particularly where they inter-
fere with sleep. Some degree of recovery occurs in
80–90% of children, requiring weeks to months.
Approximately half of children with ATM will show
excellent recovery; 10–20% develop cord necrosis
and do not recover. Most of the remainder have vari-
able residua (Ropper and Poskanzer, 1978; Berman
et al., 1981). The most important prognostic factor is
acuteness of onset; recovery is poor after hyperacute
onset (Dunne et al., 1986).
A very small number of children with cervical
ATM die from cardiorespiratory arrest or upwards
herniation. Despite the lack of established efficacy,
very high doses of anti-inflammatory agents may
be tried in cases of progressive cervical ATM, par-
ticularly those with hyperacute and potentially life-
threatening presentation. Ultimate diagnosis of MS
is made in only about 10% of adults who experience
ATM; the diagnosis of MS after isolated childhood
ATM is probably even more exceptional (Aicardi,
1992).
Neuromyelitis optica (NMO, Dévic syndrome)
In children with NMO the signs of ATM and ON
may develop simultaneously or in rapid succession.
Most prepubertal cases develop within days to weeks

after a viral illness or immunization. The transverse
myelitis is typically sudden and severe, producing
paraplegia. ON often develops in both eyes bilateral,
onset in the second eye occurring days to months
after the first. Funduscopic changes of papillitis are
usually, but not always, present. The prodromal
illness is less commonly discerned in postpubertal
cases, the myelitis less complete, and there is a greater
tendency for the ON to be unilateral. More detailed
description of this entity and recent advances in
understanding of pathogenesis, classification, dif-
ferential considerations and diagnostic testing are
considered elsewhere in this volume. There is no
generally accepted therapy for DS. However, very
high doses of intravenous corticosteroids may be
considered where optic nerve or spinal cord swelling
is particularly alarming, in order not only to attenuate
the inflammatory process, but also to close blood–
brain barrier and prevent swelling that may lead to
tissue ischemia.
Encephalomyeloradiculoneuropathy
(Miller–Fisher/Bickerstaff encephalitis)
Some cases of ADEM with prominent myelitis will
have peripheral nerve signs, representing clinical
overlap with GBS. This overlap is particularly pro-
minent in patients with AIDS. Tumors with involve-
ment of the cauda equina or nerve roots must also be
considered. The anti-Hu-associated paraneoplastic
syndromes that should also be considered in adults
have not yet been shown to occur in children

(Dalmau et al., 1992).
Acute necrotic encephalitis (ANE)
The pathogenesis of this early childhood illness that
has chiefly been reported in Japanese and Taiwanese
(Mizuguchi et al., 1995; Voudris et al., 2001) remains
obscure. Although some type of inflammatory en-
cephaloclasia is a likely explanation, it is unclear
whether this might be due to a primary infectious
process. As with ADEM, there is a male predom-
inance (boy to girl ratio = 2:1). Most cases arise
between 6–24 months of age, although cases have
occurred in children as old as five years. The onset is
typically marked by fever and rapid deterioration of
mental status in association with convulsive seizures
and brainstem signs. Abnormalities of liver enzyme
testing may be found.
MRI scans of severely affected children disclose
symmetrical bright lesions on T2 weighting that
NICP_C05 04/05/2007 12:26PM Page 107
108 ROBERT S. RUST
involve the thalami, hypothalamus, brainstem teg-
mentum, and cerebellum. Bright signal may also
be found in cerebral white matter. Some of these
abnormalities (particularly those representing edema)
and clinical status may improve with high doses of
intravenously administered corticosteroids. Some
particularly malignant cases appear to have favor-
ably responded to heroical treatment with surgical
decompression of intracranial pressure. The gray
matter lesions may become cavitary and death has

occurred in approximately half of recognized cases.
This severe disease must be differentiated from
those cases of presumed ADEM whose imaging
manifestations may be confined to or emphasized in
white matter and thalami. The patients with these
ADEM-related manifestations tend to be older (0.4–
6 years of age) and their disease evolution is much
less severe than is observed by the infants with ANE.
The outcome is favorable, with resolution of scan
changes once recovery occurs (Cusmai et al., 1994;
Marcu et al., 1979; Suwa et al., 1999; Tenembaum
et al., 2002). It is not entirely clear whether some
reports of milder cases of ANE from which infants
recover without subsequent relapse are cases of mild
ANE or of ADEM. ANE must not be confused with
some cases that fall within the ADEM/childhood
MS/“recurrent ADEM” spectrum that manifest large
pseudo-cavitary lesions of deep white matter sparing
basal ganglia or thalamus, or with AHLE or Balò
disease, all diseases for which high doses of intra-
venous corticosteroids may prove beneficial. The
differential diagnosis also includes tumor, infectious,
metabolic, or vascular (e.g. sinovenous thrombosis)
diseases of brain.
Subacute-onset disseminated CNS illnesses that
may be forms of ADEM
Considerable recent interest has focused on the pos-
sibility that certain complex subacute-onset illnesses
with extra pyramidal, psychiatric, and behavioral
manifestations may represent post-infectious diseases

that have mechanisms similar to ADEM or might
respond to therapies advocated for ADEM and related
illnesses. The basis for such speculation derives in
part from the strong evidence that one such illness,
Sydenham chorea, is a post-infectious condition that
is known to be provoked by certain strains of Group
A β-hemolytic streptococci. Recent studies have
attempted to characterize another entity, termed
pediatric autoimmune neuropsychiatric disorder
(PANDAS). Attention has also been directed to the
study of the neuropsychiatric disturbances that may
develop in patients with rheumatic fever without
associated chorea, observations that some believe
will have pertinence to the development of isolated
psychiatric disturbances (Mercandonte, 2000) or such
controversial entities as “chronic fatigue syndrome.”
During the past two decades, a number of cases of
indolent psychiatric disturbances associated with
“ADEM-like” MRI abnormalities have been reported,
as has the gradual improvement of images and affect
with corticosteroid treatment. Investigators have
suggested that these are examples of “subacute”
limbic or disseminated encephalomyelitis ( Johnson
et al., 1985). These cases have tended to have no
clear association of deterioration with a preceding
febrile illness or vaccination and CSF immune profile
studies if obtained have been normal.
Meningoencephalitic encephalitis and other
infections with possible ADEM-related
complications

The role that inflammatory dysregulation plays
in the pathogenesis of CNS infectious conditions
has received increasing scrutiny in the past decade
and falls outside of the scope of this review. The
extent to which mechanisms closely resemble those
of ADEM or MS, the role of genetics, and the per-
tinence of anti-inflammatory therapy for these con-
ditions are as yet incompletely understood. These
conditions constitute part of the differential dia-
gnosis of ADEM and MS, and may produce similar
imaging or CSF abnormalities. The presence of pial
or cortical gadolinium enhancement on T1-weighted
images is suggestive of encephalitis rather than
ADEM. The borderland between HSV2 encephalitis
and ADEM is especially unclear when there are
widespread lesions, some such cases responding
very well to acyclovir treatment (Chu et al., 2002),
some apparently benefiting from the addition of anti-
inflammatory therapy.
Acute or “relapsing” HSV2 encephalitis may be
especially difficult to distinguish from ADEM (Chu
et al., 2002). Some widespread HSV2 “recurrence” is
a form of ADEM, responsive to corticosteroid treat-
ment with a good outcome (Tulyapronchote and Rust,
1992). Chronic progressive Lyme encephalomyelitis
or third-stage neuroborreliosis may also be a form
of ADEM (Braune, 1991; Pavlovic et al., 1993;
Reik et al., 1985). Japanese B, measles, and mumps
encephalitides, cerebral malaria, and the CNS
manifestations of Dengue all likely entail immuno

NICP_C05 04/05/2007 12:26PM Page 108
Acute disseminated encephalomyelitis 109
dysregulation that may involve ADEM-related mech-
anisms and may prove responsive to therapies aimed
at this form of pathogenesis. The presence of pial
enhancement, which is not a finding in ADEM, sug-
gests active meningoencephalitis. Metazoal parasitic
diseases of brain (e.g. cysticercosis) may produce
imaging changes that closely resemble ADEM,
although the pathogenic mechanisms of ADEM are
probably not involved.
Summary
Box 5.15 includes hypothetical criteria for various
ADEM spectrum diagnoses in children.
I Tentative ADEM (T-ADEM)
1 Preceding exogenous provocation
(required)
a Febrile, likely infectious illness
or vaccination within 28 days
b At least 12 hours afebrile
improvement prior to
ADEM-related deterioration
2 Neurological deterioration: required
a At least two separate clinical
lesions, otherwise unexplained
b At least three of the following:
(1) Recurrence of fever,
irritability, or lethargy
at onset
(2) Bilateral optic neuritis

(3) MRI typical for ADEM
(see text)
(i) Cortical ribbon–
subcortical white
matter junction
(ii) Indistinct margins
(“smudge”
appearance)
(4) Focal or generalized EEG
slowing
(5) Elevated myelin basis
protein with normal
CSF immune profile
(6) Clear improvement ≤ 24
hours after high-dose
intravenous steroids
3 Other relevant diagnoses,
including MS, excluded by
appropriate testing
II Probable ADEM (P-ADEM): Meet
T-ADEM criteria without recurrence
in 2 years*
III Definite ADEM (D-ADEM): Meet
P-ADEM criteria, no recurrence
for additional 10 years
IV Tentative Recurrent ADEM
(TR-ADEM): Initial bout +≤four
total bouts* each meeting criteria
for diagnosis T-ADEM
V Probable recurrent ADEM (PR-ADEM):

Meet TR-ADEM criteria followed by
a hiatus of ≥ 2 years without further
recurrence*
Not treated with immunomodulatory
prophylaxis during those 2 years
without bouts
VI Definite recurrent ADEM (DR-ADEM):
Meet PR-ADEM criteria followed ≥ 10
additional years without recurrence
Not treated with immunomodulatory
prophylaxis during those 10 additional
years
(*Excepting taper-related recurrences)
VII Type 1 steroid-dependent idiopathic
demyelinating illness (SDIPI)
1 Recurrent cases not satisfying
CD-MS or ADEM diagnostic
categories
2 Unavoidable recurrences provoked
at same approximate threshold
of steroid taper unless replaced
with at least monthly high dose
intravenous corticosteroids
3 Excludes alternative diagnoses
VIII Type 2 steroid-dependent idiopathic
demyelinating illness (SDIPI)
1 Recurrent cases not satisfying CD-MS
or ADEM diagnostic categories
Box 5.15 Criteria for various suggested diagnostic
groupings of ADEM family.

continued
NICP_C05 04/05/2007 12:26PM Page 109
110 ROBERT S. RUST
References
Abramsky, O. and Teitelbaum, D. 1977. The auto-
immune features of acute transverse myelopathy.
Ann Neurol, 2, 36–40.
Aicardi, J. 1992. Diseases of the Nervous System in
Childhood. MacKeith Press, London.
Apak, R.A., Anlar, B. and Saatci, I. 1999. A case of
relapsing acute disseminated encephalomyelitis with
high dose corticosteroid treatment. Brain Dev, 21(4),
279–82.
Arya, S.C. 2001. Acute disseminated encephalo-
myelitis associated with poliomyelitis vaccine. Pediatr
Neurol, 24(4), 325.
Atlas, S.W., Grossman, R.I., Goldberg, H.I. et al. 1986.
MR diagnosis of acute disseminated encephalomye-
litis. J Comp Assist Tom, 10(5), 798–801.
Au, W.Y., Li, A.K., Cheung, R.T. et al. 2002. Acute
disseminated encephalomyelitis after para-influenza
infection post bone marrow transplantation. Leuk
Lymphoma, 43(2), 455– 7.
Baum, P.A., Barkovich, A.J., Koch, T.K. et al. 1994.
Deep gray matter involvement in children with
acute disseminated encephalomyelitis. AJNR Am J
Neuroradiol, 15(7), 1275–83.
Beck, R.W., Cleary, P.A., Anderson, M.M., Jr. et al.
1992. A randomized, controlled trial of corticoster-
oids in the treatment of acute optic neuritis. The

Optic Neuritis Study Group (see comments). N Engl J
Med, 326(9), 581–8.
Beck, R.W., Cleary, P.A., Trobe, J.D. et al. 1993. The
effect of corticosteroids for acute optic neuritis on
the subsequent development of multiple sclerosis.
The Optic Neuritis Study Group (see comments).
N Engl J Med, 329(24), 1764–9.
Berman, M., Feldman, S., Alter, M. et al. 1981. Acute
transverse myelitis: incidence and etiologic consid-
erations. Neurology, 31(8), 966–71.
Boiko, A., Vorobeychik, G., Paty, D. et al. 2002.
Early onset multiple sclerosis: A longitudinal study.
Neurology, 59(7), 1006–10.
Braune, H.J. 1991. Lyme borreliosis – Epidemiology,
etiology, diagnosis and therapy. Fortschr Neurol
Psychiatr, 59(11), 456–67.
Bye, A.M.E., Kendall, B. and Wilson, J. 1985. Multiple
sclerosis in childhood: A new look. Dev Med Child
Neurol, 27, 215–22.
Caldemeyer, K.S., Smith, R.R., Harris, T.M. et al. 1994.
MRI in acute disseminated encephalomyelitis.
Neuroradiology, 36(3), 216–20.
Campistol, P.J., Cambra, F.J. and Guitet Julia, M. 2001.
Acute disseminated encephalomyelitis in childhood.
Report of 10 cases. Rev Neurol, 32(5), 409–13.
Caracciolo, J.T., Murtagh, R.D., Rojiani, A.M. et al. 2001.
Pathognomonic MR imaging findings in Balo con-
centric sclerosis. AJNR Am J Neuroradiol, 22(2), 292–3.
Chen, C.J. 2001. Serial proton magnetic resonance
spectroscopy in lesions of Balo concentric sclerosis.

J Comput Assist Tomogr, 25
(5), 713–18.
Chen, C.J., Chu, N.S., Lu, C.S. et al. 1999. Serial magnetic
resonance imaging in patients with Balo’s concentric
sclerosis: Natural history of lesion development. Ann
Neurol, 46(4), 651–6.
Chu, K., Kang, D.W., Lee, J.J. et al. 2002. Atypical
brainstem encephalitis caused by herpes simplex
virus 2. Arch Neurol, 59(3), 460–3.
Cole, G.F., Auchterlonie, L.A. and Best, P.V. 1995. Very
early onset multiple sclerosis. Dev Med Child Neurol,
37(8), 667–72.
Cole, S.R., Beck, R.W., Moke, P.S. et al. 1998. The
predictive value of CSF oligoclonal banding for MS
5 years after optic neuritis. Optic Neuritis Study
Group. Neurology, 51(3), 885–7.
Courville, J. and Cooper, C.W. 1970. The cerebellar
nuclei of macaca mulatta: A morphological study.
J Comp Neurol, 140(2), 241–54.
2 Unavoidable recurrences
provoked at same approximate
threshold of steroid taper
unless replaced with
immunomodulatory
interferon therapy
3 Excludes alternative diagnoses
IX Other idiopathic recurrent
demyelinating illness (OIRDI)
1 Recurrent cases not satisfying
CD-MS, ADEM, R-ADEM,

T-ADEM, or SDIDI criteria
2 Alternative diagnoses excluded
by specific testing
X Idiopathic progressive disseminated
encephalomyelitis
1 Requires exogenous stimulus
2 Excludes cases with identifiable
toxic, heritable, infectious condition
(e.g. heritable leukodystrophy,
Aicardi-Goutieres, TORCH infection,
AIDS, etc.)
XI Special cases
1 Lyme neuroborreliosis
Box 5.15 (Cont’d)
NICP_C05 04/05/2007 12:26PM Page 110
Acute disseminated encephalomyelitis 111
Croft, P.B. 1969. Para-infectious and post-vaccinal
encephalomyelitis. Postgrad Med J, 45(524), 392– 400.
Currie, S., Roberts, A.H. and Urich, H. 1970. The
nosological position of concentric lacunar leucoen-
cephalopathy. J Neurol Neurosurg Psychiatry, 33(2),
131–7.
Cusmai, R., Bertini, E., Di Capua, M. et al. 1994. Bilateral,
reversible, selective thalamic involvement demon-
strated by brain MR and acute severe neurological
dysfunction with favorable outcome. Neuropediatrics,
25(1), 44–7.
Dale, J.B. and Beachey, E.H. 1984. Unique and com-
mon protective epitopes among different serotypes
of group A streptococcal M proteins defined with

hybridoma antibodies. Infect Immun, 46(1), 267–9.
Dale, R.C., de Sousa, C., Chong, W.K. et al. 2000. Acute
disseminated encephalomyelitis, multiphasic dissem-
inated encephalomyelitis and multiple sclerosis in
children. Brain, 123(12), 2407–22.
Dalmau, J., Graus, F., Rosenblum, M.K. et al. 1992.
Anti-Hu-associated paraneoplastic encephalomyelitis/
sensory neuronopathy. A clinical study of 71 patients.
Medicine (Baltimore), 71(2), 59–72.
de Vries, E. 1960. Postvaccinial Perivenous Encephalitis,
Elsevier, Amsterdam.
Dunne, K., Hopkins, I.J. and Shield, L.K. 1986. Acute
transverse myelopathy in childhood. Dev Med Child
Neurol, 28, 198–204.
Epperson, L.W., Whitaker, J.N. and Kapila, A. 1988.
Cranial MRI in acute disseminated encephalomye-
litis. Neurology, 38(2), 332–3.
Feinsod, M., Hoyt, W.F., Wilson, W.B. et al. 1975.
Electrophysiological parameters in the evaluation
of occipital apoplexy. Eur Neurol, 13(5), 451–60.
Ghatak, N.R., Leshner, R.T., Price, A.C. et al. 1989.
Remyelination in the human central nervous
system. J Neuropathol Exp Neurol, 48(5), 507–18.
Good, W.V., Muci-Mendoza, R., Berg, B.O. et al. 1992.
Optic neuritis in children with poor recovery of
vision. Aust NZ J Ophthalmol, 20(4), 319–23.
Hahn, J.S., Siegler, D.J. and Enzmann, D. 1996.
Intravenous gammaglobulin therapy in recurrent
acute disseminated encephalomyelitis. Neurology,
46(4), 1173–4.

Hanefeld, F., Bauer, H.J., Christen, H.J. et al. 1991.
Multiple sclerosis in childhood: Report of 15 cases
(see comments). Brain Dev, 13(6), 410–16.
Hartung, H.P. and Grossman, R.I. 2001. ADEM: Distinct
disease or part of the MS spectrum? Neurology, 56(10),
1257–60.
Hasegawa, Y. 2001. MRI findings of acute dissemin-
ated encephalomyelitis associated with mycoplasma
infection. No To Shinkei, 53(6), 580–1.
Hesselink, J.R., Dunn, W.M. and Healy, M.E. et al. 1986.
Paramagnetic enhancement of cerebral lesions with
gadolinium-DTPA. Acta Radiol Suppl,
369, 558–60.
Hohlfeld, R. and Wekerle, H. 2001. Immunological
update on multiple sclerosis. Curr Opin Neurol, 14(3),
299–304.
Hollinger, P., Sturzenegger, M., Mathis, J. et al. 2002.
Acute disseminated encephalomyelitis in adults:
A reappraisal of clinical, CSF, EEG, and MRI findings.
J Neurol, 249(3), 320–9.
Honkaniemi, J., Dastidar, P., Kahara, V. et al. 2001.
Delayed MR imaging changes in acute disseminated
encephalomyelitis. AJNR Am J Neuroradiol, 22(6),
1117–24.
Hung, K.L., Liao, H.T., Tsai, M.L. 2000. Postinfectious
encephalomyelitis: Etiologic and diagnostic trends.
J Child Neurol, 15(10), 666–70.
Hung, K.L., Liao, H.T. and Tsai, M.L. 2001. The spec-
trum of postinfectious encephalomyelitis. Brain Dev,
23(1), 42–5.

Hynson, J.L., Kornberg, A.J., Coleman, L.T. et al. 2001.
Clinical and neuroradiologic features of acute dis-
seminated encephalomyelitis in children. Neurology,
56(10), 1308–12.
Ichiyama, T., Shoji, H., Kato, M. et al. 2002. Cerebro-
spinal fluid levels of cytokines and soluble tumour
necrosis factor receptor in acute disseminated
encephalomyelitis. Eur J Pediatr, 161(3), 133–7.
Ishihara, O., Yamaguchi, Y., Matsuishi, T. et al. 1984.
Multiple ring enhancement in a case of acute
reversible demyelinating disease in childhood
suggestive of acute multiple sclerosis. Brain Dev,
6(4), 401–6.
Itoyama, Y., Tateishi, J. and Kuroiwa, Y. 1985.
Atypical multiple sclerosis with concentric or lamel-
lar demyelinated lesions: Two Japanese patients
studied post mortem. Ann Neurol, 17(5), 481–7.
Jacobs, L., Goodkin, D.E., Rudick, R.A. et al. 1994.
Advances in specific therapy for multiple sclerosis.
Curr Opin Neurol, 7(3), 250–4.
Jacobson, S., Gupta, A., Mattson, D. et al. 1990. Immuno-
logical studies in tropical spastic paraparesis. Ann
Neurol, 27(2), 149–56.
Jaing, T.H., Lin, K.L., Chiu, C.H. et al. 2001. Acute
disseminated encephalomyelitis in autoimmune
hemolytic anemia. Pediatr Neurol, 24(4), 303– 5.
Johnson, L.N. 1998. The 5-year risk of MS after optic
neuritis [letter; comment]. Neurology, 51(4), 1237–8.
Johnson, R.T., Griffin, D.E., Hirsch, R.L. et al. 1984.
Measles encephalomyelitis – clinical and immuno-

logic studies. N Engl J Med, 310(3), 137–41.
Johnson, Ye, Duray, P.H., Steere, A.C. et al. 1985. Lyme
arthritis: Spirochetes found in synovial micorangio-
pathic lesions. Am J Pathol, 118, 26–34.
Kanter, D.S., Horensky, D., Sperling, R.A. et al. 1995.
Plasmapheresis in fulminant acute disseminated
encephalomyelitis. Neurology, 45(4), 824–7.
Kazarian, E.L. and Gager, W.E. 1978. Optic neuritis
complicating measles, mumps, and rubella vaccina-
tion. Am J Opthalmol
, 86, 544–7.
Kepes, J.J. 1993. Large focal tumor-like demyelinating
lesions of the brain: intermediate entity between
multiple sclerosis and acute disseminated encephalo-
myelitis? A study of 31 patients (see comments). Ann
Neurol, 33(1), 18–27.
NICP_C05 04/05/2007 12:26PM Page 111
112 ROBERT S. RUST
Kerr, D.A. and Ayetey, H. 2002. Immunopathogenesis
of acute transverse myelitis. Curr Opin Neurol, 15(3),
339–47.
Kesselring, J., Miller, D.H., Robb, S.A. et al. 1990. Acute
disseminated encephalomyelitis. MRI findings and
the distinction from multiple sclerosis. Brain, 113(2),
291–302.
Kira, J., Kanai, T., Nishimura, Y. et al. 1996. Western
versus Asian types of multiple sclerosis: immuno-
genetically and clinically distinct disorders. Ann Neurol,
40(4), 569–74.
Kleiman, M. and Brunquell, P. 1995. Acute dissemin-

ated encephalomyelitis: response to intravenous
immunoglobulin. J Child Neurol, 10(6), 481–3.
Kline, L.B., Margulies, S.L. and Oh, S.J. 1982. Optic
neuritis and myelitis following rubella vaccination.
Arch Neurol, 39, 443–4.
Kriss, A., Francis, D.A., Cuendet, F. et al. 1988. Recovery
after optic neuritis in childhood. J Neurol Neurosurg
Psychiatry, 51(10), 1253–8.
Leake, J.A., Billman, G.F., Nespeca, M.P. et al. 2002.
Pediatric acute hemorrhagic leukoencephalitis: Report
of a surviving patient and review. Clin Infect Dis,
34(5), 699–703.
Levine, S. and Wenk, E. 1965. A hyperacute form of
allergic encephalomyelitis. Am J Pathol, 547, 61–88.
Link, H., Cruz, M., Gessain, A. et al. 1989. Chronic
progressive myelopathy associated with HTLV-I:
Oligoclonal IgG and anti-HTLV-I IgG antibodies in
cerebrospinal fluid and serum. Neurology, 39(12),
1566–72.
Lucchinetti, C., Bruck, W. and Noseworthy, J. 2001.
Multiple sclerosis: Recent developments in neuro-
pathology, pathogenesis, magnetic resonance imag-
ing studies and treatment. Curr Opin Neurol, 14(3),
259– 69.
Maeda, Y., Kitamoto, I., Kurokawa, T. et al. 1989.
Infantile multiple sclerosis with extensive white
matter lesions (see comments). Pediatr Neurol, 5(5),
317–19.
Marcu, H., Hacker, H. and Vonofakos, D. 1979. Bilateral
reversible thalamic lesions on computed tomography.

Neuroradiology, 18, 201–4.
Mehler, M.F. and Rabinowitz, L. 1988. Inflammatory
myelinoclastic diffuse sclerosis. Ann Neurol, 23,
413–15.
Mehler, M.F. and Rabinowich, L. 1989. Inflammatory
myelinoclastic diffuse sclerosis (Schilder’s disease):
Neuroradiologic findings. AJNR Am J Neuroradiol,
10(1), 176–80.
Miller, A., Hafler, D.A. and Weiner, H.L. 1991. Tolerance
and suppressor mechanisms in experimental auto-
immune encephalomyelitis: implications for immuno-
therapy of human autoimmune diseases. Faseb J,
5(11), 2560–6.
Miller, D.H., McDonald, W.I., Blumhardt, L.D. et al.
1987. Magnetic resonance imaging in isolated non-
compressive spinal cord syndromes. Ann Neurol,
22(6), 714–23.
Miller, H.G., Stanton, J.B. and Gibbons, J.L. 1956. Para-
infectious encephalomyelitis and related syndromes.
Quart J Med, 25, 427–505.
Miller, H.G., Stanton, J.B. and Gibbons, J.L. 1957.
Acute disseminated encephalomyelitis and related
syndromes. Br Med J, 1, 668–71.
Mizuguchi, M., Abe, J., Mikkaichi, K. et al. 1995. Acute
necrotizing encephalopathy of childhood: A new syn-
drome presenting with multifocal, symmetric brain
lesions. J Neurol Neurosurg Psychiatry, 58, 555–61.
Moore, G.R., Neumann, P.E., Suzuki, K. et al. 1985.
Balo’s concentric sclerosis: New observations on
lesion development. Ann Neurol, 17(6), 604–11.

Morimoto, T., Nagao, H., Sano, N. et al. 1985. A case of
multiple sclerosis with multi-ring-like and butterfly-
like enhancement on computerized tomography.
Brain Dev, 7(1), 43–5.
Narciso, P., Galgani, S., Del Grosso, B. et al. 2001. Acute
disseminated encephalomyelitis as manifestation of
primary HIV infection. Neurology, 57(8), 1493–6.
Nishikawa, M., Ichiyama, T., Hayashi, T. et al. 1999.
Intravenous immunoglobulin therapy in acute dis-
seminated encephalomyelitis [In Process Citation].
Pediatr Neurol, 21(2), 583–6.
O’Halloran, H.S., Draud, K., Minix, M. et al. 1998. Leber’s
neuroretinitis in a patient with serologic evidence of
Bartonella elizabethae. Retina, 18(3), 276–8.
Oleszak, E.L., Lin, W.L., Legido, A. et al. 2001. Presence
of oligoclonal T cells in cerebrospinal fluid of a child
with multiphasic disseminated encephalomyelitis
following hepatitis A virus infection. Clin Diagn Lab
Immunol, 8(5), 984–92.
Ormerod, I.E., McDonald, W.I., du Boulay, G.H. et al.
1986. Disseminated lesions at presentation in patients
with optic neuritis. J Neurol Neurosurg Psychiatry,
49(2), 124–7.
Ormerod, I.E., Roberts, R.C., du Boulay, E.P. et al. 1984.
NMR in multiple sclerosis and cerebral vascular
disease [letter]. Lancet, 2(8415), 1334–5.
Paine, R.S. and Byers, R.K. 1953. Transverse myelopathy
in childhood. Am J Dis Child, 85, 151–63.
Parkin, P.J., Hierons, R. and McDonald, W.I. 1984.
Bilateral optic neuritis. A long-term follow-up. Brain,

107(3), 951–64.
Pasternak, J.F., De Vivo, D.C. and Prensky, A.L. 1980.
Steroid-responsive encephalomyelitis in childhood.
Neurology, 30(5), 481–6.
Pavlovic, D., Levic, Z., Dmitrovic, R. et al. 1993.
Rhombencephalitis as a manifestation of neurobor-
reliosis. Glas Srp Akad Nauka (Med), 43, 233–6.
Plesner, A.M., Arlien-Soborg, P. and Herning, M. 1998.
Neurological complications to vaccination against
Japanese encephalitis. Eur J Neurol, 5(5), 479–85.
Poser, C.M. 1985. Myelinoclastic diffuse sclerosis.
In P. Vinken (ed.), Handbook of Clinical Neurology.
Vol. 47. Elsevier, Amsterdam, pp. 419–28.
Poser, C.M. 2000. The pathogenesis of multiple
sclerosis: A commentary. Clin Neurol Neurosurg
,
102(4), 191–4.
NICP_C05 04/05/2007 12:26PM Page 112
Acute disseminated encephalomyelitis 113
Poser, C.M., Goutieres, F., Carpentier, M.A. et al. 1986.
Schilder’s myelinoclastic diffuse sclerosis (published
erratum appears in Pediatrics, 1986, 78(1), 138).
Pediatrics, 77(1), 107–12.
Poser, C.M. and van Bogaert, L. 1956. Natural history
and evolution of the concept of Schilder’s diffuse
sclerosis. Acta Psychiatr Neurol Scand, 31, 285–331.
Pouly, S. and Antel, J.P. 1999. Multiple sclerosis and
central nervous system demyelination. J Autoimmun,
13(3), 297–306.
Pouly, S., Antel, J.P., Ladiwala, U. et al. 2000. Mechan-

isms of tissue injury in multiple sclerosis: oppor-
tunities for neuroprotective therapy. J Neural Transm
Suppl, 58, 193–203.
Purvin, V., Hrisomalos, N. and Dunn, D. 1988. Varicella
optic neuritis. Neurology, 38(3), 501–3.
Reik, L., Smith, L., Khan, A. et al. 1985. Demyelinat-
ing encephalopathy in Lyme disease. Neurology, 32,
1302–5.
Riedel, P., Wall, M., Grey, A. et al. 1998. Autoimmune
optic neuropathy (letter). Arch Ophthalmol, 116(8),
1121–4.
Riikonen, R. 1989. The role of infection and vaccination
in the genesis of optic neuritis and multiple sclerosis
in children. Acta Neurol Scand, 80(5), 425–31.
Riikonen, R., Donner, M. and Erkkila, H. 1988a. Optic
neuritis in children and its relationship to multiple
sclerosis: A clinical study of 21 children. Dev Med
Child Neurol, 30(3), 349–59.
Riikonen, R. and von Willebrandt, E. 1988b. Lympho-
cytes subclasses and function in patients with optic
neuritis in childhood with special reference to multiple
sclerosis. Acta Neurol Scand, 78(1), 58–64.
Ropper, A.H. and Poskanzer, D.C. 1978. The prognosis
of acute and subacute transverse myelopathy based
on early signs and symptoms. Ann Neurol, 4(1), 51–9.
Rosenblum, M., Scheck, A.C., Cronin, K. et al. 1989.
Dissociation of AIDS-related vacuolar myelopathy
and productive HIV-1 infection of the spinal cord.
Neurology, 39(7), 892–6.
Rousseau, J.J., Lust, C., Zangerle, P.F. et al. 1986. Acute

transverse myelitis as presenting neurological feature
of Lyme disease (letter). Lancet, 2(8517), 1222–3.
Rudick, R.A., Jacobs, L., Kinkel, P.R. et al. 1986.
Isolated idiopathic optic neuritis. Analysis of free
kappa-light chains in cerebrospinal fluid and correla-
tion with nuclear magnetic resonance findings. Arch
Neurol, 43(5), 456–8.
Russell, D.S. 1955. The nosological unity of acute
haemorrhagic leukoencephalitis and acute dissemin-
ated encephalomyelitis. Brain, 78, 369–76.
Rust, R.S. 2000. Multiple sclerosis, acute disseminated
encephalomyelitis, and related conditions.
Semin
Pediatr Neurol, 7(2), 66–90.
Rust, R.S., Jr. 2003. Infectious myositis. In H.R. Jones,
Jr, D.C. DeVivo, B.T. Darras (eds.), Neuromuscular
Disorders of Infancy, Childhood, and Adolescence
A Clinician’s Approach. Butterworth Heinemann,
Philadelphia, pp. 939–1005.
Rust, R.S. and Dodson, W.E. 1989a. Phenobarbital:
Absorption, distribution, and excretion. In R.H. Levy,
R.H. Mattson and B.S. Meldrum (eds.), Antiepileptic
Drugs. 3rd edn. Raven Press, New York.
Rust, R.S., Jr., Dodson, W.E. and Trotter J.L. 1988.
Cerebrospinal fluid IgG in childhood: The establish-
ment of reference values. Ann Neurol, 23(4), 406–10.
Rust, R.S., Mathisen, J., Prensky, A.L. et al. 1989b.
Acute disseminated encephalomyelitis (ADE) and
childhood multiple sclerosis (MS). Annal Neurol,
26(3), 467.

Rust, R.S. and Fleming, J.O. 1996. Immunologic dis-
eases of the nervous system. In E.R. Stiehm (ed.),
Immunologic Disorders in Infants and Children. WB
Saunders, Philadelphia, pp. 889–915.4.
Sacconi, S., Salviati, L. and Merelli, E. 2001. Acute
disseminated encephalomyelitis associated with hep-
atitis C virus infection. Arch Neurol, 58(10), 1679–81.
Sanders, K.A., Khandji, A.G. and Mohr, J.P. 1990.
Gadolinium-MRI in acute transverse myelopathy.
Neurology, 40, 1614–16.
Schilder, P. 1912. Zur Kenntniss der sogenannten
diffusen Sklerose. Z Ges Neurol Psychiat, 10, 1–60.
Schwarz, S., Knauth, M., Mohr, A. et al. 2001. Acute
disseminated encephalomyelitis (ADEM). Nervenarzt,
72(4), 241–54.
Scott, T.F. 1967. Postinfectious and vaccinal enceph-
alitis. Med Clin North Am, 51(3), 701–17.
Seales, D and Greer, M. 1991. Acute hemorrhagic
leukoencephalitis. A successful recovery. Arch Neurol,
48(10), 1086–8.
Shaw, C.M. and Alvord, E.C., Jr. 1987. Multiple sclerosis
beginning in infancy. J Child Neurol, 2(4), 252–6.
Shintani, S., Fumimura, Y. and Shiigai, T. 2001. Acute
disseminated encephalomyelitis following aseptic
meningoencephalitis. Clin Neurol Neurosurg, 103(3),
155–9.
Silberberg, D.H. 1993. Corticosteroids and optic neuritis
(editorial; comment). N Engl J Med, 329(24), 1808–10.
Sommer, J.B., Erbguth, F.J. and Neundorfer, B. 2000.
Acute disseminated encephalomyelitis following

Legionella pneumophila infection. Eur Neurol, 44(3),
182–4.
Sotgiu, S., Piana, A., Pugliatti, M. et al. 2001.
Chlamydia pneumoniae in the cerebrospinal fluid
of patients with multiple sclerosis and neurological
controls. Mult Scler, 7(6), 371–4.
Stricker, R.B., Miller, R.G. and Kiprov, D.D. 1992. Role
of plasmapheresis in acute disseminated (post-
infectious) encephalomyelitis. J Clin Apheresis, 7(4),
173 –9.
Sugita, K., Suzuki, N., Shimizu, N. et al. 1993.
Involvement of cytokines in N-methyl-N’-nitro-N-
nitrosoguanidine-induced plasminogen activator
activity in acute disseminated encephalomyelitis and
multiple sclerosis lymphocytes. Eur Neurol, 33(5),
358–62.
Suwa, K., Yamagata, T., Momoi, M. et al. 1999.
Acute relapsing encephalopathy mimicking acute
NICP_C05 04/05/2007 12:26PM Page 113
114 ROBERT S. RUST
necrotizing encephalopathy in a 4-year-old boy.
Brain Dev, 21, 554–8.
Tabira, T. 1994. Concentric sclerosis (Balo). Nippon
Rinsho, 52(11), 2971–5.
Tabira, T. and Nishizawa, M. 1990. Molecular patho-
mechanism of autoimmune encephalitis. Tanpa-
kushitsu Kakusan Koso, 35(7 Suppl), 1283–90.
Takahashi, S., Shimomura, T. and Tohgi, H. 1992.
Serial changes of magnetic resonance imagings
in acute disseminated encephalomyelitis. Rinsho

Shinkeigaku – Clinical Neurol, 32(2), 182–6.
Tellis, V.A. 1998. Bypass grafts for hemoaccess. J Am
Coll Surg, 187(5), 569.
Tenembaum, S., Chamoles, N. and Fejerman, N. 2002.
Acute disseminated encephalomyelitis: A long-term
follow-up study of 84 pediatric patients. Neurology,
59(8), 1224–31.
Toivanen, A.L., Valanne, L. and Tatlisumak, T. 2002.
Acute disseminated encephalomyelitis following
nephropathia epidemica. Acta Neurol Scand, 105(4),
333–6.
Toro, G. and Roman, G. 1978. Cerebral malaria. A
disseminated vasculomyelinopathy. Arch Neurol,
35(5), 271–5.
Tourtellotte, W. 1970. On cerebrospinal fluid
immunoglobulin-G (IgG) quotients in multiple
sclerosis and other diseases. A review and a new
formula to estimate the amount of IgG synthesized
per day by the central nervous system. J Neurol Sci,
10(3), 279–304.
Trotter, J.L and Rust, R.S. 1989. Human cerebrospinal
fluid immunology. In R.M. Herndon and R.A.
Brumback (eds.), The Cerebrospinal Fluid, Kluwer
Academic Publishers, Boston, pp. 179–226.
Tselis, A. 2001. Acute disseminated encephalomyelitis.
Curr Treat Options Neurol, 3(6), 537–42.
Tulyapronchote, R. and Rust, R.S. 1992. Recurrent
herpes simplex encephalitis: Responsive to high dose
methylprednisone. Ann Neurol, 32(3), 448.
Tyler, K.L., Gross, R.A. and Cascino, G.D. 1986. Unusual

viral causes of transverse myelitis: Hepatitis A virus
and cytomegalovirus. Neurology, 36(6), 855–8.
van der Meyden, C.H., de Villiers, J.F., Middlecote, B.D.
et al. 1994. Gadolinium ring enhancement and
mass effect in acute disseminated encephalomyelitis.
Neuroradiology, 36(3), 221–3.
Vliegenthart, W.E., Sanders, E.A., Bruyn, G.W. et al.
1985. An unusual CT-scan appearance in multiple
sclerosis. J Neurol Sci, 71(1), 129–34.
Voudris, K.A., Skaardoutsou, A., Haronitis, I. et al.
2001. Brain MRI findings in influenza A-associated
acute necrotizing encephalopathy of childhood. Eur
J Paediatr Neurol, 5(5), 199–202.
Walther, M.M. and Cutler, G.B. 1997. Urodynamic
abnormalities in two brothers with adrenomyelo-
neuropathy. World J Urol, 15
(4), 262–5.
Yao, D.L., Webster, H.D., Hudson, L.D. et al. 1994.
Concentric sclerosis (Balo): morphometric and in situ
hybridization study of lesions in six patients. Ann
Neurol, 35(1), 18–30.
NICP_C05 04/05/2007 12:26PM Page 114
Guillain–Barré syndrome (GBS) is the most common
cause of nontraumatic acute and subacute general-
ized paralysis in adults in the industrialized world. It
is an acute acquired inflammatory neuropathy that
usually affects the motor nerve roots or its axons.
GBS probably represents several distinct diseases
that are grouped within a single syndrome. Typically
GBS presents with acroparesthesias or numbness, and

evolves to ascending weakness within days to weeks.
Areflexia is a hallmark. As the weakness progresses,
the patient may become completely paralyzed, requir-
ing assisted ventilatory support. Typically there is
normal cerebrospinal fluid (CSF) cellularity with an
increased CSF protein level. Several variants have
been reported.
Epidemiology
Throughout the world, the annual incidence of GBS
is 1.3 cases for 100,000, affecting children and adults.
GBS is the most common acute acquired neuropathy
of childhood. In adults the prevalence is higher in
older patients (>75 years) as compared to younger
patients (<30 years). Men are slightly more often
affected than women (1.5:1.0) (Kuwabara, 2004).
Recent anteceding infections are frequently associated,
most commonly Campylobacter jejuni, identified in up
to 25% of the cases. Viral agents (HIV, EBV, CMV) and
other microbial infections (Mycoplasma pneumoniae)
have also been reported preceding GBS. A history of
recent immunization has also been implicated (e.g.
vaccination against swine influenza, 1977). Associ-
ations with recent surgery, trauma, and puerperium
have been suggested, though more epidemiological
data is needed to confirm these associations with GBS
(Cheng et al., 1998).
Immunopathogenesis
The immunopathogenesis of GBS is incompletely
understood. As an autoimmune disease, self-tolerance
is impaired.

In a subset of patients with acute motor axonal
neuropathy (AMAN) identified in China, there is
evidence of molecular mimicry with cross-reactive
antigens between bacterial lipo-oligosaccharides
(LOS) from C. jejuni, which causes a diarrheal illness,
and gangliosides of the peripheral nerve (Hafer-
Macko et al., 1996b). The genetics of the micro-
bial agent associated with the preceding infection
may play an important role in the likelihood of its
immunogenicity and cross-reaction against com-
ponents of the peripheral nerve (Koga et al., 1998).
Several C. jejuni genes may participate in the patho-
genic role of C. jejuni infection. Mutation of the
C. jejuni waaC gene alters both the bacterial LOS and
the capsular polysaccharides (Kanipes et al., 2006).
Gene racR may encode factors necessary for patho-
genicity in humans (Kordinas et al., 2005). Human
host genetics also play an important role. Human
leukocyte antigens (HLA) B54 and Cw1 were signific-
antly more frequently present in patients with GBS
and Miller–Fisher (MF) syndrome than in normal
controls (Koga et al., 2005). In Japanese patients,
immunoglobulin KM allotypes are associated with
an increased prevalence of anti-GD1a antibodies,
but are not a risk factor for developing GBS (Pandey
et al., 2005).
The study of GBS variants like MF syndrome
and AMAN has increased the understanding of the
pathogenesis of these diseases and the autoimmune
nature of GBS. The pathogenesis of acute inflammat-

ory demyelinating polyradiculoneuropathy (AIDP),
which is the most common type of GBS in the USA
and Europe, still remains uncertain. There is suggest-
ive evidence that increased circulating T-cell react-
ivity against gangliosides, such as GM1, occurs in GBS.
This is probably not related to nonspecific peripheral
nerve damage, since hereditary, toxic, and diabetic
neuropathy patients did not have increased T-cell
reactivity (Csurhes et al., 2005). More recently,
characteristics common to both European and Asian
C. jejuni strains associated with GBS have been
described (Koga et al., 2006).
6
Guillain–Barré syndrome
Eduardo A. De Sousa and Thomas H. Brannagan III
NICP_C06 04/05/2007 12:26PM Page 117
118 EDUARDO A. DE SOUSA AND THOMAS H. BRANNAGAN III
In AMAN, there is no inflammation. Complement
activation product C3d has been identified bound
to the axolemma of motor fibers (Hafer-Macko
et al., 1996a). Experimental axonopathy has been
replicated in animals immunized with liposac-
charide extracted from C. jejuni isolated from a GBS
patient (Caporale et al., 2006). High throughput
amplified fragment length polymorphism (htAFLP)
analysis has been used to identify strains of C. jejuni
more frequently associated to GBS, corroborating
the hypothesis that molecular mimicry of LOS
may be a major etiological determinant of AMAN
(Godschalk et al., 2006). There is axonal degenera-

tion of the motor roots and nerves without evidence
for demyelination, and IgG GM1 antibodies are
often present. Animal models of AMAN have shown
an immunoglobulin class switch of GM1 antibodies
from IgM to IgG, with associated axonal and motor
nerve injury.
Experimental autoimmune neuritis (EAN) is an
animal model for GBS. It supports the prominent role
of the T cells in the pathogenesis of GBS. EAN demon-
strates cell-mediated as well as antibody-mediated
attack to the peripheral nerve, with perivascular
lymphocytic infiltration and segmental demyelina-
tion. Secondary axonal injury is noted with extens-
ive inflammation and demyelination. Complement
deposits are seen, and antibodies can be measured,
although the antigen has not been identified. Inter-
leukin 23 (IL-23) may play an important role in the
early effector phase of immune-mediated peripheral
nerve demyelination in EAN, as well as in human
GBS. IL-23 is highly expressed in the endoneurium,
with immunoreactivity patterns similar to the anti-
CD68 (macrophage) marker. Western blot CSF
analysis of GBS patients also shows an increased
IL-23 expression (Hu, 2006). Tumor necrosis factor-
alpha converting enzyme (TACE), a member of the A
Disintegrin and Metalloproteinase (ADAM) family,
may be upregulated in EAN, and is expressed in
nerves of GBS patients (Kurz et al., 2005).
Diagnosis
Diagnosis is based on clinical presentation, neuro-

logical examination, CSF analysis, and electrodia-
gnostic testing. Usually the diagnosis of GBS can
be made on clinical grounds. Magnetic resonance
imaging (MRI) may also be performed, mainly when
the diagnosis is in question, so as to exclude other
possibilities. MRI findings of nerve root enhancement
have not shown good correlation with prognosis.
Clinical presentation and GBS variants
In the USA and Europe the clinical presentation is
typically of a patient with progressive, ascending,
symmetric weakness affecting first the feet, then legs,
hands, arms, and bulbar muscles. These are charac-
teristics of AIDP, also called acute motor and sensory
demyelinating polyradiculoneuropathy (AMSDN).
GBS is frequently preceded by a paresthetic
prodrome. On examination the tone is reduced, and
tendon reflexes are absent throughout. Breathing
and swallowing may be affected, so early airway
protection may be important. Neck flexor weakness
and fatigability usually correlates well with dia-
phragmatic weakness.
Symptoms progress and patients reach a nadir by
four weeks. There may be a plateau phase followed by
recovery, though the recovery may be incomplete.
Several other GBS variants have been reported. In
Japan and China both the MF syndrome and AMAN
seem to be more frequent than they are in the USA
and Europe. The MF variant usually presents with
diplopia (from ophthalmoplegia) followed by marked
limb or gait ataxia, and areflexia. Acute motor and

sensory axonal neuropathy (AMSAN) has a similar
presentation to AIDP and AMAN affecting both the
motor and sensory nerves. The pharyngo-cervico-
brachial (PCB) variant may have weakness restricted
to the named distribution and is frequently associ-
ated with ptosis and ophthalmoplegia (Ropper, 1986).
Other subtypes include pure motor, pure sensory, pure
ophthalmoplegic, predominant or pure dysauto-
nomic, paretic, and ataxic variants.
A minority of cases evolves to chronic inflammat-
ory demyelinating polyneuropathy (CIDP).
Laboratory/CSF
Laboratory tests are useful to confirm the clinical
diagnosis of GBS, but should not delay onset of treat-
ment. CSF protein is usually elevated with a maximal
elevation from day 4 to day 20 (Marshall, 1963;
McLeod et al., 1976). The CSF protein may be normal,
particularly in the first few days. The protein level
remains elevated after clinical recovery (Arnason,
1982). CSF cell count is usually normal (albumino-
cytological dissociation), though HIV-associated GBS
may demonstrate increased cellularity. Not all patients,
however, with HIV-associated GBS have pleocytosis
(Brannagan and Zhou, 2003).
GBS patients may have antiganglioside anti-
bodies, more commonly anti-GM1 antibodies of the
NICP_C06 04/05/2007 12:26PM Page 118
Guillain–Barré syndrome 119
IgG class. Anti-GM1 antibodies of the IgM class are
usually associated with multifocal motor neuropathy,

and not with GBS. Carbohydrate molecular mimicry
of C. jejuni LOS may be critical for the induction of
anti-GM1 antibodies (Shu et al., 2006). Anti-GQ1B
antibodies are frequently present in MF syndrome,
but also in many of the ophthalmoplegic GBS vari-
ants, regardless of the ophthalmoplegia occurring
in isolation or in association with weakness or ataxia
(Chan, 2004; Lyu and Chen, 2004; Odaka et al.,
2003). Anti-GT1a antibodies may be present in GBS,
MF syndrome, PCB variant as well as in Bickerstaff
brainstem encephalitis (Nagashima et al., 2004).
Serum antibodies to GM1, GD1b, GD1a, GM3, and
sulfatides may be associated with increased intrathe-
cal IgG production indexes without CSF oligoclonal
bands or other evidence of intrathecal antigangli-
oside antibody synthesis (Mata et al., 2006).
Neurophysiology
Nerve conduction studies (NCS) and needle elec-
tromyography are useful to confirm the diagnosis
and to distinguish GBS from other diseases. These
tests may also help determine prognosis. Severely
reduced compound muscle action potential ampli-
tudes carry a poor prognosis for recovery (Cornblath
et al., 1988). A reduction of the compound muscle
action potential (CMAP) amplitude below 20% of
normal is the best predictor of a poor recovery. Early
studies suggested that nerve conduction studies could
be normal, however, more recent studies, which have
tested more nerves and F-wave responses, are only
rarely normal (Albers et al., 1985; Ropper et al.,

1990). They may not, however, initially be diagnostic
of demyelination.
Absent F-waves are often an early finding due to
proximal conduction block. Low amplitude or absent
CMAPs may also occur from distal conduction block,
though this can also be due to axonal loss (Triggs
et al., 1992). The H reflex is the most sensitive test
in early GBS (Gordon and Wilbourn, 2001). Sural
sparing may be present, in which unlike a typical
length-dependent neuropathy, normal sural sensory
nerve action potential (SNAP) amplitudes are seen
in the context of reduced upper extremity SNAP
amplitudes (Bromberg and Albers, 1993). Early in the
disease there may also be prolongation of the F-wave
responses and later mild conduction velocity slowing
may be present. Marked motor conduction velocity
slowing, another marker for demyelination, is not
frequently present at the onset of the disease. The
presence of conduction blocks suggests an acquired
demyelinating neuropathy consistent with GBS.
Phrenic NCS have been used to predict respiratory
failure, though they are not clearly useful. Earlier
some authors suggested use of phrenic NCS to detect
respiratory involvement in GBS patients and to iden-
tify those at risk of respiratory failure (Zifko et al.,
1996), but more recently others have reported only
weak correlation of phrenic action potential ampli-
tude and latency with vital capacity, claiming that
these studies are probably not useful for predicting
respiratory failure, though the methods used were

not identical (Durand et al., 2005).
Needle electromyography examination is useful.
The presence of abnormal spontaneous activity
(fibrillation potentials and positive sharp waves) is
a sign of muscle denervation. Reduced recruitment
may be seen early in the course, but does not dis-
tinguish demyelinating from axonal nerve injury.
Blink reflexes may show markedly prolonged R1
response latencies. Because it evaluates a short nerve
segment, it may show changes early in the course,
especially in patients with the MF variant.
Differential diagnosis
Porphyric neuropathy can mimic GBS but has a nor-
mal CSF protein, recurrent abdominal pain, mental
symptoms, onset after barbiturates or other drug
exposure, and high urinary δ-aminolevulinic acid
and porphobilinogen levels.
Several infections can cause an acute neuropathy
that may be confused with GBS. Diphtheria can cause
a demyelinating neuropathy, though it is now rare
in the USA. Diphtheritic polyneuropathy can usually
be distinguished by the long latency period between
the respiratory infection and onset of the neuropathy
(5 to 8 weeks), blurry vision due to paralysis of accom-
modation, and a slower evolution of symptoms. A thick
gray exudative pseudomembrane can be seen when
examining the soft palate and pharynx. A cytoal-
buminological disassociation is seen like in GBS.
Poliomyelitis, which is also now rare in the USA can
be distinguished by asymmetric weakness, signs of

meningeal irritation, fever, and CSF pleocytosis. Acute
encephalitis is the most common West Nile neurolo-
gical manifestation, but an acute paralytic syndrome
may occur. Asymmetric weakness is characteristic, but
some cases develop in a GBS-like manner. Some cases
have a flu-like prodrome without notable encephalitis.
In botulism, ocular muscles and the pupils are usu-
ally affected. Nerve conduction studies in botulism
NICP_C06 04/05/2007 12:26PM Page 119
120 EDUARDO A. DE SOUSA AND THOMAS H. BRANNAGAN III
reveal normal nerve conduction velocities and a
facilitating response to repetitive nerve stimulation.
Toxic neuropathies caused by arsenic or thal-
lium ingestions or n-hexane inhalation may present
acutely like GBS. Tick paralysis, which occurs pri-
marily in children, can be misdiagnosed as GBS. A
careful scalp examination should be performed. The
weakness improves rapidly after the tick is removed.
Treatment
Both plasmapheresis (PLEX) and IVIg are effect-
ive treatments for GBS. PLEX is known to be more
effective than supportive care alone. In mild GBS two
sessions of PLEX are superior to none. In moderate
GBS four sessions are superior to two. In severe GBS
six sessions are no better than four. Continuous-flow
PLEX machines may be superior to intermittent-flow
machines and albumin to fresh frozen plasma as
the exchange fluid. PLEX is more beneficial when
started within seven days after disease onset, but it
is still beneficial up to 30 days after disease onset.

For children less than 12 years old the value of PLEX
has not been established (Raphael et al., 2002).
IVIg started within two weeks from onset hastens
recovery at least as much as PLEX (randomized trials
in severe disease), though some authors report higher
rates of recurrence. IVIg is typically given at a dose
of 2 g/kg divided over five days. IVIg treatments are
more likely to be completed than PLEX. IVIg after
PLEX does not confer significant extra benefit (PSGBS
Group, 1997). In children, intravenous immuno-
globulin probably hastens recovery compared with
supportive care alone (Hughes et al., 2006a).
Oral corticosteroids are not effective for the treat-
ment of GBS according to controlled studies, and may
slow recovery. Intravenous methylprednisolone
(IVMP) alone does not produce significant benefit or
harm (van Koningsveld 2004, Hughes et al., 2006b).
Patients with MF syndrome usually recover
completely without requiring immunomodulatory
therapy. While some authors report the benefits of
immunotherapy in MF syndrome (Zifko et al., 1994;
Yeh et al., 1999), controlled studies have not demon-
strated benefit (Mori et al., 2001; Mori et al., 2002).
Supportive, preventive, and intensive care
Patients with GBS often need management in the
intensive care unit because of the risk of respiratory
failure and frequent dysautonomia
Severe fluctuations of heart rate and blood pres-
sure mandate cardiac telemetry and blood pressure
monitoring. Prominent hypertensive peaks may be

treated with short-acting antihypertensives, but
caution is advised since fluctuations towards severe
hypotension may follow antihypertensive treatment.
Autonomic dysfunction may also cause mental status
abnormalities that differ from ICU delirium, and are
more likely due to a dissociated sleep and dream
disorder (Cochen et al., 2005). Other aspects of diag-
nostic testing and management for dysautonomia in
GBS are discussed in Chapter 8.
Up to 25–30% of GBS patients may develop a
severe generalized flaccid paralysis with respiratory
failure, requiring assisted ventilation. The advent of
mechanical ventilation has overwhelmingly changed
the course of GBS. Mortality rate now is 5–10%,
markedly different from the natural history of this
disease before ventilators became available.
GBS patients may develop pharyngeal, laryngeal,
and tongue weakness, which increases the risk
of aspiration pneumonia due to decreased clear-
ance of secretions. Early airway protection prevents
aspiration, which is currently one of the most import-
ant causes of hospital complications and death in
GBS patients. It is not advised to wait for physical
examination findings of respiratory failure (such as
neck weakness and fatigability or blood gas abnorm-
alities), since these are late markers for intubation.
Preferably patients should have a baseline arterial
blood gas and frequent respiratory function mea-
surements, including forced vital capacity and nega-
tive inspiratory force, which may sooner predict time

for early intubation. Because GBS patients may have
a decreased cough reflex, there is greater risk of aspira-
tion and atelectasis, which can rapidly progress to
respiratory failure. Patients should be intubated if
the vital capacity is less than 12–15 ml/kg (Ropper
and Kehne, 1985).
Compression elastic stockings and intermittent
pneumatic compression devices (IPCD) are used for
deep venous thrombosis (DVT) prophylaxis, though
IPCD is preferable. Lower extremity venous Doppler
ultrasonograms can be performed when DVT is
suspected. Proton pump inhibitors or antihistamine
H2-blockers are used for gastric and duodenal ulcer
prophylaxis. Physical therapy may be done for decu-
bitus ulcer prophylaxis. Ankle splints may prevent
Achilles tendon shortening and contractures.
Prognosis
Most patients recover from GBS, but approximately
20% of patients are left with significant disability.
Respiratory function recovers in the vast majority,
NICP_C06 04/05/2007 12:26PM Page 120
Guillain–Barré syndrome 121
if not all. Residual motor and sensory abnormalities
may persist. One year after the onset of GBS, two-
thirds of patients still perceived a decrease of power
and sensation that were often disturbing (Bernsen
et al., 2005), with an evident impact on daily life
and social well-being.
The prognosis for the MF variant is generally good
when there is no concomitant GBS.

References
Albers, J.W., Donofrio, P.D. and McGonagle, T.K. 1985.
Sequential electrodiagnostic abnormalities in acute
inflammatory demyelinating polyradiculoneuro-
pathy. Muscle Nerve, 8, 528–39
Arnason, B.G.W. 1982. Acute inflammatory demyelin-
ating polyradiculopathy. In P.J. Dick, P.K. Thomas
and E.H. Lambert (eds.), Peripheral Neuropathy, WB
Saunders, Philadelphia, pp. 2050–100.
Bernsen, R.A., de Jager, A.E., van der Meche, F.G. and
Suurmeijer, T.P. 2005. How Guillain–Barré patients
experience their functioning after 1 year. Acta Neurol
Scand., 112(1), 51–6.
Brannagan, T.H. III and Zhou, Y. 2003. HIV-associated
Guillain–Barré syndrome. J Neurol Sci, 208(1–2),
39–42.
Bromberg, M.B. and Albers, J.W. 1993. Patterns of
sensory nerve conduction abnormalities in demyeli-
nating and axonal peripheral nerve disorders. Muscle
Nerve, 16(3), 262–6.
Caporale, C.M., Capasso, M., Luciani, M. et al. 2006.
Experimental axonopathy induced by immunization
with Campylobacter jejuni lipopolysaccharide from
a patient with Guillain–Barré syndrome. J Neuro-
immunol, 174(1–2), 12–20.
Chan, Y.C., Wilder-Smith, E. and Chee, M.W. 2004. Acute
ophthalmoplegia with pupillary areflexia associated
with anti-GQ1b antibody. J Clin Neurosci, 11(6),
658–60.
Cheng, Q., Jiang, G.X., Fredrikson, S., Link, H. and

de Pedro-Cuesta, J. 1998. Increased incidence of
Guillain–Barré syndrome postpartum. Epidemiology,
9(6), 601–4.
Cochen, V., Arnulf, I., Demeret, S. et al. 2005. Vivid
dreams, hallucinations, psychosis and REM sleep in
Guillain-Barre syndrome. Brain, 128(11), 2535–45.
Cornblath, D.R., Mellits, E.D., Griffin, J.W. et al. 1988.
Motor conduction studies in Guillain–Barré syndrome:
Description and prognostic value. Ann Neurol, 23(4),
354–9.
Csurhes, P.A., Sullivan, A.A., Green, K., Greer, J.M.,
Pender, M.P. and McCombe, P.A. 2005. Increased
circulating T cell reactivity to GM1 ganglioside
in patients with Guillain–Barré syndrome. J Clin
Neurosci, 12(4), 409–15.
Durand, M.C., Prigent, H., Sivadon-Tardy, V. et al.
2005. Significance of phrenic nerve electrophysio-
logical abnormalities in Guillain–Barré syndrome.
Neurology, 65(10), 1646–9.
Godschalk., P.C., Bergman, M.P., Gorkink, R.F. et al.
2006. Identification of DNA sequence variation in
Campylobacter jejuni strains associated with the
Guillain–Barré syndrome by high-throughput AFLP
analysis. BMC Microbiol, 6, 32.
Gordon, P.H. and Wilbourn, A.J. Early electrodiagnostic
findings in Guillain–Barré syndrome. Arch Neurol,
58(6), 913–7.
Hafer-Macko, C., Hsieh, S.T., Li, C.Y. et al. 1996a.
Acute motor axonal neuropathy: an antibody-
mediated attack on axolemma. Ann Neurol, 40(4),

635–44.
Hafer-Macko, C.E., Sheikh, K.A., Li, C.Y. et al. 1996b.
Immune attack on the Schwann cell surface in acute
inflammatory demyelinating polyneuropathy. Ann
Neurol, 39(5), 625–35.
Hu, W., Dehme, T., Pirhonen, J., Hartung, H.P.
and Kieseier, B.C. 2006. Interleukin 23 in acute
inflammatory demyelination of the peripheral nerve.
Arch Neurol, 63(6), 858–64.
Hughes, R.A., Raphael, J.C., Swan, A.V. and van
Doorn, P.A. 2006a. Intravenous immunoglobulin
for Guillain–Barré syndrome. Cochrane Database Syst
Rev, 25(1), CD002063.
Hughes, R., Swan, A., van Koningsveld, R. and
van Doorn, P. 2006b. Corticosteroids for Guillain–
Barré syndrome. Cochrane Database Syst Rev, 19(2),
CD001446.
Kanipes, M.I., Papp-Szabo, E., Guerry, P. and Monteiro,
M.A. 2006. Mutation of waaC, encoding heptosyl-
transferase I in Campylobacter jejuni 81–176, affects
the structure of both lipooligosaccharide and cap-
sular carbohydrate. J Bacteriol, 188(9), 3273–9.
Koga, M., Yuki, N., Kashiwase, K., Tadokoro, K., Juji, T.
and Hirata K. 1998. Guillain–Barré and Fisher’s
syndromes subsequent to Campylobacter jejuni enter-
itis are associated with HLA-B54 and Cw1 independ-
ent of anti-ganglioside antibodies. J Neuroimmunol,
88(1–2), 62–6.
Koga, M., Takahashi, M., Masuda, M., Hirata, K. and
Yuki, N. 2005. Campylobacter gene polymorphism as

a determinant of clinical features of Guillain–Barré
syndrome. Neurology, 65(9), 1376–81.
Koga, M., Gilbert, M., Takahashi, M., Li, J., Koike, S. and
Hirata, K. 2006. Comprehensive analysis of bacterial
risk factors for the development of Guillain–Barré
syndrome after Campylobacter jejuni enteritis. J Infect
Dis, 193(4), 547–55.
Kordinas, V., Nicolaou, C., Ioannidis, A., Papavasileiou,
E., John Legakis, N. and Chatzipanagiotou, S. 2005.
Prevalence of four virulence genes in Campylobacter
jejuni determined by PCR and sequence analysis. Mol
Diagn, 9(4), 211–15.
Kurz, M., Pischel, H., Hartung, H.P. and Kieseier, B.C.
2005. Tumor necrosis factor-alpha-converting enzyme
is expressed in the inflamed peripheral nervous
system. J Peripher Nerv Syst, 10(3), 311–18.
Kuwabara, S. 2004. Guillain–Barré syndrome: Epi-
demiology, pathophysiology and management. Drugs,
64(6), 597–610.
NICP_C06 04/05/2007 12:26PM Page 121
122 EDUARDO A. DE SOUSA AND THOMAS H. BRANNAGAN III
Lyu, R.K. and Chen, S.T. 2004. Acute multiple cranial
neuropathy: A variant of Guillain–Barré syndrome?
Muscle Nerve, 30(4), 433–6.
Marshall, J.T. 1963. The Landry–Guillain–Barré
syndrome. Brain, 86, 55–66.
Mata, S., Galli, E., Amantini, A., Pinto, F., Sorbi, S.
and Lolli, F. 2006. Anti-ganglioside antibodies and
elevated CSF IgG levels in Guillain–Barré syndrome.
Eur J Neurol, 13(2), 153–60.

McLeod, J.G., Walsh, J.C., Prineas, J.W. and Pollard, J.D.
1976. Acute idiopathic polyneuritis. A clinical
and electrophsiological follow-up study. J Neurol Sci,
27(2), 145–62.
Mori, M., Kuwabara, S., Fukutake, T., Yuki, N. and
Hattori, T. 2001. Clinical features and prognosis of
Miller Fisher syndrome. Neurology, 56(8), 1104–6.
Mori, M., Kuwabara, S., Fukutake, T. and Hattori T.
2002. Plasmapheresis and Miller Fisher syndrome,
analysis of 50 consecutive cases. J Neurol Neurosurg
Psychiatry, 72(5), 680.
Nagashima, T., Koga, M., Odaka, M., Hirata, K. and
Yuki, N. 2004. Clinical correlates of serum anti-GT1a
IgG antibodies. J Neurol Sci, 219(1–2), 139–45.
Odaka, M., Yuki, N., Yamada, M. et al. 2003.
Bickerstaff’s brainstem encephalitis: Clinical features
of 62 cases and a subgroup associated with Guillain–
Barré syndrome. Brain, 126(10), 2279–90.
Pandey, J.P., Koga, M. and Yuki, N. 2005.
Immunoglobulin KM allotypes are associated with
the prevalence of autoantibodies to GD1a ganglio-
side, but not with susceptibility to the disease, in
Japanese patients with Guillain–Barré syndrome.
Neurogenetics, 6(4), 225–8.
PSGBS (Plasma Exchange/Sandoglobulin Guillain-Barré
Syndrome) Trial Group. 1997. Randomised trial of
plasma exchange, intravenous immunoglobulin, and
combined treatments in Guillain-Barré syndrome.
Lancet, 349(9047), 225–30.
Raphael, J.C., Chevret, S., Hughes, R.A. and Annane, D.

2002. Plasma exchange for Guillain–Barré syndrome.
Cochrane Database Syst Rev, 2, CD001798.
Ropper, A.H. and Kehne, S.M. 1985. Guillain–Barré
syndrome: Management of respiratory failure.
Neurology, 35(11), 1662–5.
Ropper, A.H. 1986. Unusual clinical variants and signs
in Guillain–Barré syndrome. Arch Neurol, 43(11),
1150–2.
Ropper, A.H., Wijdicks, E.F. and Shahani, B.T. 1990.
Electrodiagnostic abnormalities in 113 consecutive
patients with Guillain–Barré syndrome. Arch Neurol,
47(8), 881–7.
Shu, X.M., Cai, F.C. and Zhang, X.P. 2006. Carbohyd-
rate mimicry of Campylobacter jejuni lipooligosaccha-
ride is critical for the induction of anti-GM1 antibody
and neuropathy. Muscle Nerve, 33(2), 225–31.
Triggs, W.J., Cros, D., Gominak, S.C. et al. 1992. Motor
nerve inexcitability in Guillain–Barré syndrome.
The spectrum of distal conduction block and axonal
degeneration. Brain, 115(5), 1291–302.
van Koningsveld, R., Schmitz, P.I.M., van der
Meché, F.G.A., et al. 2004. Effect of methylpred-
nisolone when added to standard treatment with
intravenous immunoglobulin for Guillain-Barré
syndrome. Lancet, 373, 192–6.
Yeh, J.H., Chen, W.H., Chen, J.R. and Chiu, H.C. 1999.
Miller Fisher syndrome with central involvement:
Successful treatment with plasmapheresis. Ther Apher,
3(1), 69–71.
Zifko, U., Drlicek, M., Senautka, G. and Grisold, W.

1994. High dose immunoglobulin therapy is effect-
ive in the Miller Fisher syndrome. J Neurol, 241(3),
178–9.
Zifko, U., Chen, R., Remtulla, H., Hahn, A.F., Koopman,
W. and Bolton, C.F. 1996. Respiratory electrophysio-
logical studies in Guillain–Barré syndrome. J Neurol
Neurosurg Psychiatry, 60(2), 191–4.
NICP_C06 04/05/2007 12:26PM Page 122
Chronic inflammatory demyelinating polyradiculo-
neuropathy (CIDP) is one of the most commonly
encountered, treatable peripheral neuropathies.
Though cases were reported previously under a
variety of names, the literature was reviewed and a
single patient was described responding repeatedly
to corticosteroids and relapsing when given placebo by
Austin in 1958 (Austin, 1958). Dyck and colleagues
described 57 patients, labeled as chronic inflam-
matory polyradiculoneuropathy (Dyck et al., 1975).
Later “demyelinating” was added to the diagnosis.
Other inflammatory demyelinating neuropathies
include multifocal motor neuropathy (Parry and
Clarke, 1988), anti-MAG (myelin-associated glyco-
protein) neuropathies (Latov and Renaud, 2004),
and a demyelinating neuropathy associated with
osteosclerotic myeloma (Dispenzieri et al., 2003),
which have distinct clinical characteristics, patho-
genesis, and response to treatment.
Epidemiology
The prevalence of CIDP has been estimated from
1–7.7 per 100,000 and rises with age (Lunn et al.,

1999; Mcleod et al., 1999; Mygland and Monstad,
2001). The prevalence in 70–79-year-old males is
1 in 10,000. The incidence in one study was 0.15/
100,000 (McLeod et al., 1999). These are likely under-
estimates, since the criteria to select cases were strict
and would not capture all cases. The mean age of
onset was 47.6 years and the prevalence is higher
in males than females. The prevalence is higher in
women in the 20–29 age range (McLeod et al., 1999).
CIDP accounted for 21% of initially idiopathic neuro-
pathy (Dyck et al., 1981) and 13% of patients seen in
another neuromuscular center (Barohn, 1998).
Immunopathogenesis
CIDP is thought to be an autoimmune disease.
This concept is supported by inflammation seen in
pathological specimens, the similarities to another
autoimmune disease, Guillain–Barré syndrome
(GBS), and improvement after immunomodulatory
treatment. How nerve damage results is incom-
pletely understood. Studies in humans as well as
animal models, such as experimental allergic
neuritis (EAN), support abnormalities in both cell-
mediated and humoral immunity (Hughes et al.,
2006a; Koller et al., 2005; Maurer et al., 2002).
An initial step is the loss of self-tolerance. This may
be due to molecular mimicry. The rare association
of melanoma with CIDP may relate to molecular
mimicry. There are shared carbohydrate antigens
between Schwann cells and melanoma cells, and
both derive from a neuroectodermal origin. Serum

from patients with CIDP and melanoma, as well
as GM2 antibodies, are cross-reactive with antigens
present on both nerve and tumor tissue (Bird et al.,
1996; Rousseau et al., 2005; Weiss et al., 1998).
Activated T cells can be demonstrated in sural nerve
biopsies, having crossed the blood–nerve barrier.
Damage to the endothelial cells comprising the blood–
nerve barrier can be demonstrated by loss of the tight
junction proteins claudin-5 and Z0-1 (Kanda et al.,
2004).
T cells can activate macrophages, which then can
release proteases and proinflammatory cytokines.
T cells that are not directed to a specific antigen may
be important to impair the blood–nerve barrier and
allow the passage of antibodies and macrophages
(Harvey et al., 1995; Pollard et al., 1995). The
macrophage is the major endoneurial inflammatory
cell seen in CIDP and can attack intact myelin around
normal axons resulting in demyelination (Prineas
and McLeod, 1976). Both resident macrophages,
within the nerve, as well as hematogenous macro-
phages arriving from the circulation are involved
in the pathogenesis of CIDP. Macrophages are also
involved in stopping the immune attack by promot-
ing T-cell apoptosis (Kiefer et al., 2001).
Antibodies are thought to contribute to nerve dys-
function in CIDP, since antibodies and complement
deposits have been demonstrated in nerve in patients
7
Immune-mediated chronic demyelinating polyneuropathies

Thomas H. Brannagan III
NICP_C07 04/05/2007 02:44PM Page 123
124 THOMAS H. BRANNAGAN III
with CIDP (Dalakas and Engel, 1980; Hays et al.,
1988) and many patients improve with plasma-
pheresis. Antibodies may lead macrophages to myelin,
resulting in segmental demyelination (Kiefer et al.,
2001). Antibodies may also activate the comple-
ment system and lytic membrane attack complex
(MAC). There is evidence that sublytic levels of com-
plement may result in decreased production of the
major myelin protein P0 production by Schwann
cells (David et al., 2006).
The costimulatory molecule B7-1 involved with
presentation to T cells is upregulated in macro-
phages in patients with CIDP. These costimulatory
molecules, which are typically expressed on antigen-
presenting cells, are also upregulated on myelinating
Schwann cells in patients with CIDP. Neighboring
T cells of B7-1 expressing Schwann cells have upregu-
lated receptors to these molecules (Murata and
Dalakas, 2000).
There is not an identified antigenic target in CIDP.
A small number of patients (29% in one study) have
antibodies to the major myelin protein P0 (Yan et al.,
2001). In a study of infiltrating T cells in sural nerve
biopsies of patients with CIDP, the T-cell receptor
variable (V)β utilization was characterized. There
was no evidence of clonally expanded T cells, though
this does not exclude that antigen-driven selection

of T cells was present (Bosboom et al., 2001). There
may be epitope spreading in CIDP (Jung et al., 2004;
Singh, 2004).
Immunogenetics
Weak human leukocyte antigen (HLA) associations
in patients with CIDP have been described, but not
confirmed (Feeney et al., 1990; van Doorn et al.,
1991; Vaughan et al., 1990). Cd1 molecules are
involved in the antigen presentation and processing
of glycolipids. A recent study noted 64% of patients
with CIDP were homozygous for the CD1e allele 1,
compared to 27% of normal controls, which may
enhance the ability to present lipid antigens in an
immune response (Lee et al., 2006).
Clinical presentation
CIDP is a chronic neuropathy, characterized by
greater than two months of progressive weakness,
sensory loss, and decreased or absent reflexes (Briani
et al., 1996). The weakness classically involves both
proximal and distal muscles (Dyck et al., 1975).
The weakness is typically symmetric, but can begin
asymmetrically (Verma et al., 1990). Most large
series have noted variants of this presentation, includ-
ing multifocal forms, predominantly sensory neuro-
pathies, or prominent cranial nerve involvement.
Patients may have a relapsing course or a steady
or stepwise progressive course. The relapsing and
remitting course has a peak onset with an age in the
20s, whereas the progressive form is more typical
in older subjects (Austin, 1958; Hattori et al., 2001;

McCombe et al., 1987; Thomas et al., 1969). Ninety-
one percent of patients over the age of 65 have a
chronic insidious course (Hattori et al., 2001). A pre-
ceding viral infection is reported less commonly than
with GBS in 20–30% of patients (McCombe et al.,
1987; Ropper et al., 1991; Simmons et al., 1993).
Muscle atrophy and fasciculations may be pre-
sent. Cranial nerve involvement is rare but can be
a presenting symptom (Donaghy and Earl, 1985).
Respiratory weakness occurs less commonly than
with GBS, and is seen in 8–28% of patients (Dalakas
and Engel, 1981; Dyck et al., 1975; Gorson et al.,
1997).
Clinically manifested central nervous system
involvement is seen in 3% (Barohn et al., 1989; Dyck
et al., 1975; Rubin et al., 1987; Thomas et al., 1987).
If magnetic resonance imaging (MRI) of the brain
or evoked potentials are used, one-third of patients
have subclinical involvement (Barohn et al., 1987;
Feasby et al., 1990; Pakalnis et al., 1988).
Laboratory
There is no laboratory test available for diagnosing
CIDP. Routine blood tests including the erythrocyte
sedimentation rate (ESR) are normal (Dyck et al.,
1975). An elevated CSF protein is present in over
80% of patients (Barohn et al., 1989; Bouchard
et al., 1999; Dyck et al., 1975; Gorson et al., 1997;
Rotta et al., 2000; Small and Lovelace, 1993;
Thaisetthawatkul et al., 2002) and is thought to reflect
nerve root involvement. Antitubulin antibodies had

been reported, but not replicated by two separate
groups. Antibodies to P0 protein were identified
in 29% of patients in one series (Yan et al., 2001).
Others have not yet replicated this finding.
Electrophysiological
Nerve conduction studies are very helpful in
the diagnosis by demonstrating signs of multifocal
demyelination. In inherited neuropathies (Charcot–
Marie–Tooth), nerve conduction studies can clearly
NICP_C07 04/05/2007 02:44PM Page 124
Immune-mediated chronic demyelinating polyneuropathies 125
differentiate between axonal and demyelinating
forms of neuropathy (Buchthal and Behse, 1977).
Axonal neuropathies have either normal conduc-
tion velocity or slowing or mild slowing, but not
more than 40% of normal, resulting from loss of
fast-conducting nerve fibers. CIDP, which causes
multifocal involvement of nerve fibers, rather than
the uniform involvement typical of most inherited
demyelinating neuropathies, does not always show
conduction velocity slowing in all segments measured
(Lewis and Sumner, 1982). In addition to slow con-
duction velocity, demyelinating neuropathies are
characterized by prolonged distal motor latencies,
prolonged F-wave latencies, conduction block, and
abnormal temporal dispersion. The conduction
changes result from paranodal and internodal
demyelination, which increase the transverse capa-
citance and reduce the resistance at the internode.
This increases the outward leakage current; increas-

ing the time the internal longitudinal current must
flow in order to generate an impulse at the next
node of Ranvier. If the transverse leakage current is
excessive, insufficient current may be available to
depolarize the next node of Ranvier and the impulse
transmission ceases (Briani et al., 1996).
An American Academy of Neurology (AAN) task
force developed criteria for the identification of
patients for research studies, which included clinical,
electrophysiologic and pathological criteria for the
diagnosis of CIDP. Their electrophysiological criteria
require three demyelinating range abnormalities
(either slow conduction velocity, prolonged distal
motor latencies, or F-wave latencies or conduction
block) in two nerves (Cornblath, 1991).
The AAN criteria are frequently used to diagnose
demyelination, however, they are not sensitive
and may miss more than 50% of patients with CIDP
(Bromberg, 1991; Haq et al., 2000; Magda et al.,
2003; Rotta et al., 2000; Van den Bergh and Pieret,
2004). They are useful and have a specificity ap-
proaching 100% (Bromberg, 1991), though they may
misidentify a few patients with other disorders such
as mitochondrial neurogastrointestinal encephalo-
myopathy (MNGIE) (Bedlack et al., 2004).
In a series of patients with clinically defined CIDP,
characterized by a progressive sensorimotor neuro-
pathy, with weakness and a response to immunomo-
dulatory treatment, the electrophysiological findings
common to all patients were looked for. All patients

had abnormalities in three nerves, with findings
characteristic of demyelination in at least one nerve
(Magda et al., 2003). These electrophysiological
criteria are included in the Neuropathy Association
diagnostic criteria proposed for use in clinical prac-
tice (Berger et al., 2003).
Since CIDP is a multifocal disease, testing multiple
nerves, including cranial nerves if clinically indic-
ated, is more likely to reveal signs of demyelination
(Sumner, 1994). Testing up to eight motor nerves
has been recommended (Nicolas et al., 2002).
Improvement with treatment is usually associated
with an increase in compound motor action potential
(CMAP) amplitude (Dyck et al., 1984), and improve-
ment in conduction block (Chaudhry et al., 1994).
There is a poor correlation between motor nerve con-
duction velocity and weakness (Brown and Watson,
2002; Prineas and Mcleod, 1976), and nerve con-
duction velocity often improves only minimally with
clinical improvement (Dalakas and Engel, 1981).
Pathology
Nerve biopsy may be helpful to identify patients with
CIDP. Findings include naked or thinly myelinated
fibers or onion bulbs on semi-thin sections. Inflamma-
tion may be seen. Teased fiber analysis reveals an
increased number of demyelinated or remyelinated
fibers (type C, D, F, G) (Hahn et al., 2005). In sural nerve
biopsy, findings of demyelination or inflammation,
however, are frequently absent. A nerve biopsy is often
nondiagnostic, showing nonspecific loss of nerve

fibers, and may even be normal. The nerve biopsy is
often not necessary for the diagnosis, but may be
helpful in atypical cases and is best done at special-
ized centers (Boukhris et al., 2004).
Though CIDP is a demyelinating polyneuropathy,
there is axonal loss that is thought to be secondary, but
may result in functional loss and impaired recovery
(Bouchard et al., 1999; Nagamatsu et al., 1999).
Diagnosis
Because of the clinical heterogeneity and the lack of
a diagnostic test, various diagnostic criteria have been
proposed (Barohn et al., 1989; Berger et al., 2003;
Cornblath, 1991; Dyck et al., 1975; Hughes et al.,
2001; Hughes et al., 2006b; Latov, 2002; Saperstein
et al., 2001) (see Tables 7.1 and 7.2). The diagnosis
may be difficult. For instance in one series of patients,
all of whom had proximal and distal weakness and
in whom 95% of patients had improvement with
treatment, only 30% had the classic triad of slow nerve
conduction velocity, elevated CSF protein, and demy-
elination on nerve biopsy (Barohn et al., 1989).
NICP_C07 04/05/2007 02:44PM Page 125
Table 7.1 Proposed diagnostic criteria for CIDP.
Clinical
phenotype
Motor or sensory
involvement
Reflexes
Progression
Diagnostic studies

Diagnostic
certainty
Dyck, 1975
Tendency to
symmetric, proximal,
and distal weakness,
though some have
predominantly
motor, sensory,
or autonomic
involvement
Large-fiber sensory
involvement
Hyporeflexia or
areflexia
>6 months
CSF protein >45
g/dl, nerve biopsy
(demyelination/
remyelination), NCS:
motor conduction
velocity may be
normal but more
commonly slowed
AAN, 1991
Motor and/or
sensory
dysfunction in
more than one
limb

Hyporeflexia or
areflexia
>2 months
Mandatory: CSF
(<10 cells/mm
3
,
VDRL negative),
NCS (Table 7.2)
Supportive:
elevated CSF
protein
Definite: clinical
phenotype,
NCS, CSF,
and biopsy
Probable:
clinical
phenotype,
NCS, and CSF
Possible: clinical
phenotype and
NCS
Saperstein, 2001
Major criteria:
symmetric,
proximal, and distal
weakness
Minor criteria: only
distal weakness or

sensory loss
Hyporeflexia or
areflexia
>2 months
Mandatory: CSF
protein >45 g/dl,
nerve biopsy
(demyelination),
NCS (Table 7.2)
Supportive: CSF
<10 cells/mm
3
,
nerve biopsy
(inflammation)
Definite: clinical
major, NCS, and
CSF
Probable: clinical
major, NCS or CSF,
and biopsy
Possible: (1) clinical
major and one out
of three diagnostic
studies, or (2)
clinical minor and
two out of three
diagnostic studies
INCAT, 2001
Motor and sensory

dysfunction in more
than one limb
Hyporeflexia
or areflexia
>2 months
Inclusion criteria:
CSF <10 cells/mm
3
,
NCS (Table 7.2).
Other: elevated
CSF protein,
nerve biopsy
(demyelination/
remyelination)
supportive not
required
Neuropathy Association, 2003
Classic CIDP: symmetric,
progressive, proximal, and distal
weakness and predominantly
large-fiber sensory impairment
Variants: purely motor,
predominantly sensory,
asymmetric or multifocal
weakness or sensory loss, cranial
nerve involvement, only proximal
or distal weakness
Hyporeflexia or areflexia
>2 months

Elevated CSF protein is
supportive, nerve biopsy
(demyelination) may be
supportive, NCS (Table 7.2)
seen in most patients with
weakness, though not in all
patients
ENFS/PNS, 2005
Typical: chronically progressive,
stepwise, or recurrent symmetric
proximal and distal weakness
and sensory dysfunction of four
extremities. Cranial nerves may
be affected
Atypical: predominantly distal
weakness; purely motor or
purely sensory; asymmetric;
focal; CNS involvement
Typical: hyporeflexia or areflexia
Atypical: normal in unaffected
limbs
>2 months
Supportive criteria: elevated CSF
protein with <10 cells/mm
3
,
MRI with cauda equine, root,
or plexus enhancement and/or
hypertrophy, nerve biopsy
(demyelination/remyelination),

favorable response to
dysimmune therapy
Definite: clinical criteria (typical
or atypical) and definite NCS
(Table 7.2); or probable + at
least one supportive criterion;
or possible + at least two
supportive criteria
Probable: clinical criteria and
probable NCS (Table 7.2); or
possible CIDP + at least one
supportive criterion
Possible: clinical criteria and
possible NCS (Table 7.2);
or CIDP (definite, probable,
possible) associated with
concomitant diseases
NICP_C07 04/05/2007 02:44PM Page 126
Table 7.2 Electrodiagnostic (motor nerve conduction) criteria for demyelination in CIDP patients.
Nerve
conduction
studies (NCS)
General
guidelines
Conduction
velocity slowing
(CVS)
Conduction
block (CB)
Temporal

dispersion (TD)
Distal CMAP
(DCMAP)
dispersion
Severely
prolonged
distal motor
latency (DML)
Severely
prolonged
F-wave minimal
latency (F-min)
Absent F-waves
¶: LLN: lower limit of normal; @ULN: upper limit of normal; §: plus another demyelinating parameter in one other nerve; ¥: CVS, prolonged DML, absent or prolonged F-min.
AAN, 1991
≥3 out of 4 abnormalities: CVS
(≥2 nerves), CB and TD with
possible CB (≥1 nerve), DML
(≥2 nerves), absent or
prolonged F-min (≥2 nerves)
(1) CV <70% LLN if CMAP
amplitude <80% LLN, or
(2) CV <80% LLN if CMAP
amplitude >80% LLN
≥20% (forearm segment for
median and ulnar, lower leg
segment for peroneal nerves)
with proximal–distal duration
increase ≤15%
>15%

(1) DML >150% @ULN if
CMAP amplitude <80% LLN;
(2) DML >125% ULN if CMAP
amplitude >80% LLN
(1) F-min >150% @ULN if
CMAP amplitude <80% LLN;
(2) F-min >120% ULN if CMAP
amplitude >80% LLN
Saperstein, 2001
≥2 out of 4
abnormalities: CVS,
CB, DML, F-min
Same as AAN
>40–60% amplitude
drop, or >40–50%
area drop
>30% (median, ulnar,
radial, tibial, and
peroneal nerves)
Same as AAN
Same as AAN
If CMAP amplitude
>80% LLN
INCAT, 2001
≥1 of the following (1–4):
(1) CB/TD in three nerves +
abnormal conduction values¥
in one nerve, including one
of the nerves with CB/TD; or
(2) CB/TD in two nerves +

abnormal conduction values
in one nerve; or (3) CB/TD
in one nerve + abnormal
conduction values in two
nerves, or (4) No CB/TD, but
abnormal conduction values
in three nerves
Same as AAN
≥30% for the median, ulnar,
and peroneal nerves, or ≥50%
with Erb’s point stimulation
>15% (median, ulnar, and
peroneal nerves)
Same as AAN
Same as AAN
Neuropathy Association,
2003
Nerve conduction
abnormalities ≥3 nerves with
≥1 nerve with demyelinating
abnormalities (CVS, CB, TD,
DCMAP duration, DML, or
F-waves)
Same as AAN
≥30% (forearm segment for
median and ulnar, lower leg
segment for peroneal nerves)
>30% (median, ulnar,
and peroneal nerves)
DCMAP duration ≥9.0 ms

Same as AAN
Same as AAN
If CMAP amplitude ≥75% LLN
ENFS/PNS, 2005
Definite: ≥1 of the following (a–g):
(a) prolonged DML ≥2 nerves; (b)
CVS ≥2 nerves; (c) prolonged F-min
≥2 nerves; (d) absent Fs ≥2 nerves
plus §; (e) definite CB ≥2 nerves, or
definite CB in 1 nerve plus §; (f) TD
≥2 nerves; (g) prolonged DCMAP
duration in one nerve plus §.
Probable: either probable CB ≥2
nerves, or probable CB in one nerve
plus §.
Possible: As in definite, but in only
one nerve.
CV ≤70% LLN¶
Definite CB: ≥50% if CMAP
amplitude ≥20% LLN¶
Probable CB: ≥30% if CMAP
amplitude ≥20% LLN
>30% (median, ulnar, and
peroneal nerves)
DCMAP duration ≥9.0 ms
DML ≥150% @ULN
Same as AAN
If CMAP amplitude ≥20% LLN
NICP_C07 04/05/2007 02:44PM Page 127
128 THOMAS H. BRANNAGAN III

Treatment
Prednisone, intravenous immune globulin (IVIg),
and plasmapheresis have all been demonstrated to
be effective in controlled clinical trials (Dyck et al.,
1982a; Dyck et al., 1984; Hahn et al., 1996b; Mendell
et al., 2001). Treatment is usually continued, while
patients continue to improve; and when patients
no longer are improving, an attempt at tapering or
stopping treatment is made (Barohn et al., 1989;
Brannagan, 2002; Kissel, 2003). Some patients, includ-
ing those with severe axonal loss, may require pro-
longed treatment before seeing a response (Gorson
and Ropper, 2003; Midroni and Dyck, 1996).
Intravenous immune globulin has been demon-
strated to be beneficial in placebo-controlled double-
blind randomized controlled studies (Hahn et al.,
1996b; Mendell et al., 2001). The typical dose is
2 g/kg divided over two to five days. Maintenance
doses are required in the majority of patients. The
half-life of IVIg is 18–32 days, so a maintenance
dose of 0.5 g/kg every two weeks to maintain a con-
stant level and adjust as per patient benefit can be
used (Brannagan, 2002).
Mild, flu-like infusion-related reactions are frequent,
but these can often be controlled by slowing the infu-
sion rate or symptomatic medications (Brannagan,
2002). Anaphylaxis can occur in patients with IgA
deficiency. Rash and aseptic meningitis may occur
(Sekul et al., 1994). Though compared to alternative
treatment such as corticosteroids or chemotherapy,

IVIg is considered safe, rare serious side effects include
thromboembolic events such as stroke and heart
attack and renal failure, in patients with mild pre-
existing renal failure (Brannagan et al., 1996).
Two randomized placebo-controlled studies have
demonstrated benefit of plasmapheresis in CIDP
(Dyck et al., 1984; Hahn et al., 1996a). Side effects
include hypotension, electrolyte imbalance, allergic
reactions to infused plasma or plasma substitutes,
and infection or thrombosis at the site of venous
access (Mokrzycki and Kaplan, 1994; Rodnitzky and
Goeken, 1982).
Despite the lack of double-blind randomized
placebo-controlled trials for the use of corticosteroids,
there has been one randomized placebo-controlled
trial that showed short-term benefit (Dyck et al.,
1982a) and corticosteroids are commonly used to
treat CIDP (Mehndiratta and Hughes, 2002). At a
daily dose of 100 mg a day, response was seen from
within days to up to five months, with a mean of
1.9 months (Barohn et al., 1989). Alternate day
regimens are used (Dyck et al., 1982a) which may
reduce side effects.
Side effects associated with corticosteroid use are
common and serious and include hyperglycemia,
hypertension, aseptic necrosis of the hip, osteoporosis,
psychiatric disturbances, cataracts, susceptibility to
infection, obesity, and gastric ulcers (Schimmer and
Parker, 2001; Zonana-Nacach et al., 2000). There
are also long-term costs related to treatment of side

effects of chronic steroid use, such as diabetes, avas-
cular necrosis of the hip, hypertension, fractures, and
cataracts (Veenstra et al., 1999). Because of these
side effects, the medication is typically tapered and
discontinued as the clinical condition improves.
Less than a third of patients with CIDP remain
in remission without therapy (Barohn et al., 1989;
Gorson et al., 1997). Treatment should be continued
until the benefit is maximized (Barohn et al., 1989;
Brannagan, 2002; Kissel, 2003; Koller et al., 2005).
Once patients plateau, the dose may be able to
be reduced, however, the majority of patients need
ongoing treatment (Barohn et al., 1989; Dalakas
and Engel, 1981; Gorson et al., 1997; VanDoorn,
1996).
Other treatment options
Other agents are used for refractory patients.
Azathioprine, oral and intravenous pulse cyclophos-
phamide, interferons, cyclosporine, and mycophe-
nolate are all used for refractory patients (Briani
et al., 1996; Chaudhry et al., 2001; Good et al., 1998;
Gorson et al., 2004; Vallat et al., 2003).
A single patient has been reported with a long-
term remission after an autologous stem cell trans-
plant (Vermeulen and Van Oers, 2002). High dose
cytoxan without stem cell rescue resulted in long-
term remission for a small group of patients that had
been refractory to standard and multiple second-
line treatment modalities (Brannagan et al., 2002;
Gladstone et al., 2005).

Regional variants
The classical phenotype that suggests CIDP is the
presence of proximal and distal weakness, with large-
fiber sensory loss and areflexia. Some authors have
suggested subclassifications based on the clinical
phenotype. These have included the Lewis–Sumner
syndrome, distal acquired demyelinating sensory
neuropathy, and sensory CIDP (Katz et al., 2000;
Lewis et al., 1982). These are unlikely to be distinct
NICP_C07 04/05/2007 02:44PM Page 128
Immune-mediated chronic demyelinating polyneuropathies 129
diseases (Dyck and Dyck, 2000). Considering these
regional variants may be an aid to the clinical
approach to patients with neuropathy and their
recognition may avoid misdiagnosis (Van den Berg-
Vos et al., 2000; Saperstein et al., 2001).
Lewis–Sumner syndrome
Lewis and colleagues first described multifocal demy-
elinating neuropathy with persistent conduction
block. They reported several patients with a chronic
demyelinating sensorimotor neuropathy, character-
ized clinically by mononeuritis multiplex and electro-
physiologically by persistent multifocal conduction
block (Lewis et al., 1982). The patients responded
to corticosteroids. Pathologically there were similar
findings as seen with CIDP (Lewis et al., 1982). This
syndrome has been described by many authors under
a variety of designations including multifocal pseudo-
hypertrophic neuropathy (Adams et al., 1965),
multifocal inflammatory demyelinating neuropathy

(Van den Berg-Vos et al., 2000), upper limb predom-
inant, multifocal chronic inflammatory demyelinat-
ing polyneuropathy (Gorson et al., 1999), multifocal
acquired demyelinating sensory and motor neuro-
pathy (MADSAM) (Saperstein et al., 1999), and focal
upper limb demyelinating neuropathy (Thomas
et al., 1996).
Other than the multifocal involvement, no other
clinical, laboratory, pathological, or response to treat-
ment has been shown that differs between CIDP and
Lewis–Sumner syndrome. While many descriptions
of classic CIDP note that there is symmetrical weak-
ness, some degree of asymmetry is often seen. Patients
may initially present with involvement of single
nerves that progresses into confluent or symmetrical
involvement, as was the situation in one of the
patients described in the initial description by Lewis
et al. (Lewis et al., 1982; Verma et al., 1990). Over
50% of patients with Lewis–Sumner syndrome evolve
into a generalized typical CIDP pattern of involve-
ment (Viala et al., 2004). Nerve conduction studies
in all patients with CIDP show multifocal and asym-
metric involvement (Briani et al., 1996; Lewis and
Sumner, 1982).
Patients have responded to treatment with
corticosteroids and IVIg (Viala et al., 2004). It is
unlikely that Lewis–Sumner syndrome is a distinct
disease, but rather a multifocal form of CIDP. The
designation is useful to illustrate that a demyelinat-
ing neuropathy is in the differential diagnosis of a

mononeuritis multiplex.
Distal CIDP
A neuropathy with distal weakness and sensory loss
may be a primarily axonal neuropathy or may be
secondary to IgM monoclonal anti-MAG antibodies.
Some patients with CIDP may also have a distal neuro-
pathy. The designation distal acquired demyelin-
ating symmetric-idiopathic (DADS-I) neuropathy
has been proposed (Katz et al., 2000). These patients
respond to treatment similarly to patients with
classic CIDP, in comparison to patients with distal
demyelinating neuropathy with IgM monoclonal
antibodies with anti-MAG activity, who respond less
well (Hattori et al., 2001; Saperstein et al., 2001).
In the elderly (>65), CIDP typically has a slowly
progressive distal pattern, unlike in younger patients,
where proximal weakness is more common and a
relapsing course is seen more frequently. Predomin-
antly sensory (31%) or equal sensory and motor
involvement (51%) is more common in the elderly,
and a pure motor syndrome is less common (17%).
In contrast, adults (20–64) or juveniles (<20) more
frequently have a motor predominant pattern (16%
and 60%) or a motor equal to sensory pattern (71%
and 40%). Elderly patients may be misdiagnosed
with other causes of neuropathy or with radiculopa-
thies, but once diagnosed with CIDP respond to
treatment (Hattori et al., 2001).
Sensory CIDP
Patients with predominantly sensory findings may

also have an inflammatory demyelinating neuro-
pathy (Berger et al., 1995; Chin et al., 2004; Dyck
et al., 1975; McCombe et al., 1987; Oh et al., 1992;
Ohkoshi et al., 2001; Rotta et al., 2000; Simmons
and Tivakaran, 1996). Some of these patients are
identified by demyelinating changes on motor nerve
conduction studies, despite the lack of weakness.
Other patients are identified by sural nerve biopsy.
Demyelination is not usually detected with sensory
nerve conduction studies (Chin et al., 2004). Using
surface electrodes the sensory potentials are often
absent. Using near nerve recording, conduction
velocity slowing or conduction block are not seen as
frequently as these findings are seen in motor nerves
(Krarup and Trojaborg, 1996).
Some patients have later progressed to develop-
ing weakness (Berger et al., 1995; van Dijk et al.,
1999). An elevated CSF protein and abnormalities or
somatosensory evoked potentials may help to iden-
tify patients (Sinnreich et al., 2004). The sensory
NICP_C07 04/05/2007 02:44PM Page 129