Vous êtes sur la page 1sur 14

The new england journal of medicine

review article

medical progress

Chronic Inflammatory Demyelinating

Hubertus Kller, M.D., Bernd C. Kieseier, M.D., Sebastian Jander, M.D.,
and Hans-Peter Hartung, M.D.

hronic inflammatory demyelinating polyneuropathy is a com-

c mon, albeit underdiagnosed, and potentially treatable disease with an estimat-

ed prevalence of about 0.5 per 100,000 children1 and 1 to 2 per 100,000 adults.2,3
Clinical similarities to the acute variant of inflammatory demyelinating polyneurop-
athy (the GuillainBarr syndrome) and the beneficial effects of immunosuppressive
From the Department of Neurology, Hein-
rich-Heine University, Dsseldorf, Germany.
Address reprint requests to Dr. Hartung at
the Department of Neurology, Heinrich-
Heine University, Moorenstr. 5, D-40225
Dsseldorf, Germany, or at hans-peter.
therapies suggest an immune-mediated pathogenesis. Since the first descriptions of
patients with corticosteroid-responsive chronic polyneuropathies by Austin,4 Thomas N Engl J Med 2005;352:1343-56.
et al.,5 and Dyck et al.,6 the spectrum of clinical presentation and the diagnostic arma- Copyright 2005 Massachusetts Medical Society.

mentarium have enlarged, and further therapeutic options have evolved. The recog-
nition of this disorder as distinct from other common chronic sensorimotor polyneu-
ropathies that accompany diabetes, alcoholism, or malnutrition is important. This
review summarizes present knowledge about the clinical features of this condition, di-
agnostic criteria and diagnostic procedures involved in assessment, and current man-
agement strategies based on the results of randomized, controlled trials. Current con-
cepts of immunopathogenesis are also considered.

clinical presentation
classic chronic inflammatory demyelinating polyneuropathy
Classic chronic inflammatory demyelinating polyneuropathy is characterized by the
occurrence of symmetrical weakness in both proximal and distal muscles that pro-
gressively increases for more than two months (setting this condition apart from the
GuillainBarr syndrome, which is self-limited). The condition is associated with im-
paired sensation, absent or diminished tendon reflexes, an elevated cerebrospinal fluid
protein level, demyelinating nerve-conduction studies, and signs of demyelination in
nerve-biopsy specimens.7-9 The course can be relapsing or chronic and progressive, the
former being much more common in young adults.
As the disease has become better recognized and clinical trials have been consid-
ered, several groups have proposed clinical definitions of this neuropathy (Table 1).9-15
In all these definitions, the diagnosis is based primarily on clinical features and elec-
trophysiological studies, whereas the requirement for cerebrospinal fluid examination
and nerve biopsy varies, depending on the level of clinical diagnostic certainty, which
can range from possible to probable to definite. Obtaining both cerebrospinal fluid
and a nerve-biopsy specimen is mandatory to make a definitive diagnosis of the dis-
ease, according to criteria of the American Academy of Neurology,9 but not according
to the widely used criteria proposed by Saperstein et al.10 and by the Inflammatory Neu-
ropathy Cause and Treatment (INCAT) group.11 Classic chronic inflammatory demye-
linating polyneuropathy typically responds well to corticosteroid treatment an ob-

n engl j med 352;13 www.nejm.org march 31, 2005 1343

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

Table 1. Diagnostic Criteria.*

Feature AAN Criteria Saperstein Criteria INCAT Criteria

Clinical Motor dysfunction, sensory dysfunction Major: symmetric proximal Progressive or relapsing motor and sensory
involvement of >1 limb, or both and distal weakness; minor: dysfunction of more than 1 limb
exclusively distal weakness
or sensory loss
Time course (mo) 2 2 >2
Reflexes Reduced or absent Reduced or absent Reduced or absent
Electrodiagnos- Any 3 of the following 4 criteria: partial 2 of the 4 AAN electrodiagnos- Partial conduction block of 2 motor nerves
tic test results conduction block of 1 motor nerve, tic criteria and abnormal conduction velocity or dis-
reduced conduction velocity of 2 mo- tal latency or F-wave latency in 1 other
tor nerves, prolonged distal latency of nerve; or, in the absence of partial conduc-
2 motor nerves, or prolonged F-wave tion block, abnormal conduction velocity,
latencies of 2 motor nerves or the ab- distal latency, or F-wave latency in 3 motor
sence of F waves nerves; or electrodiagnostic abnormali-
ties indicating demyelination in 2 nerves
and histologic evidence of demyelination
Cerebrospinal White-cell count <10/mm3, negative VDRL Protein >45 mg/dl; white-cell Cerebrospinal fluid analysis recommended
fluid test; elevated protein level (supportive) count <10/mm3 (supportive) but not mandatory
Biopsy findings Evidence of demyelination and Predominant features of demy- Not mandatory (except in cases with electro-
remyelination elination; inflammation diagnostic abnormalities in only 2 motor
(not required) nerves)

* The criteria are those proposed by the American Academy of Neurology (AAN),9 Saperstein et al., 10 and Hughes et al.,11 for the Inflammatory
Neuropathy Cause and Treatment (INCAT) group. VDRL denotes Venereal Disease Research Laboratory.
According to AAN criteria, a partial conduction block is a drop of 20 percent or more in negative peak area or peak-to-peak amplitude and a
change of less than 15 percent in duration between proximal and distal site stimulation. A possible conduction block or temporal dispersion
is a drop of 20 percent or more in negative peak area or peak-to-peak amplitude and a change of more than 15 percent in duration between
proximal and distal site stimulation. A reduced conduction velocity is a velocity of less than 80 percent of the lower limit of the normal range
if the amplitude of the compound muscle action potential (CMAP) is more than 80 percent of the lower limit of the normal range or less than
70 percent of the lower limit if the CMAP amplitude is less than 80 percent of the lower limit. Prolonged distal latency is more than 125 percent
of the upper limit of the normal range if the CMAP amplitude is more than 80 percent of the lower limit of the normal range or more than 150
percent of the upper limit if the CMAP amplitude is less than 80 percent of the lower limit. An absent F wave or F-wave latency is more than
125 percent of the upper limit (INCAT criteria, more than 120 percent) if the CMAP amplitude is more than 80 percent of the lower limit or la-
tency is more than 150 percent of the upper limit if the CMAP amplitude is less than 80 percent of the lower limit.

servation that may serve to distinguish it from quired demyelinating polyneuropathy.16 Features
other forms of acquired demyelinating polyneu- of the disorder include an increased prevalence in
ropathies. men and in persons over the age of 50 years, a pre-
dominantly distal sensory loss, a mild distal weak-
demyelinating neuropathies distinct ness (as opposed to the more generalized motor
from classic chronic inflammatory deficits in classic chronic inflammatory demyelin-
demyelinating polyneuropathy ating polyneuropathy), and an unsteady gait. IgM
Refined clinical analysis has defined other forms paraproteinemia is present in nearly two thirds of
of acquired demyelinating polyneuropathies with patients with this condition.17 IgM-associated dis-
presumed autoimmune or dysimmune causes that tal demyelinating symmetric neuropathy seems to
differ from classic chronic inflammatory demye- respond poorly to immunosuppressive therapy.17
linating polyneuropathy, both with respect to clini-
cal presentation and to the response to treatment. Multifocal Motor Neuropathy
It is not clear whether these conditions are variants It is important to differentiate multifocal motor
of chronic inflammatory demyelinating polyneu- neuropathy from motor neuron disease. Multifo-
ropathy or distinct diseases. cal motor neuropathy is characterized by asymmet-
ric weakness without sensory loss, often starting
Distal Acquired Demyelinating Symmetric in distal arm muscles. A partial motor-conduction
Neuropathy block at multiple sites is a characteristic electro-
It has been suggested that distal acquired demye- physiologic feature, although not all patients have
linating symmetric neuropathy is a distinct ac- this finding. The same holds true for the detection

1344 n engl j med 352;13 www.nejm.org march 31 , 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

of circulating antiganglioside antibodies. Cerebro- such as infection with the human immunodefi-
spinal fluid protein levels and cell counts are usu- ciency virus or hepatitis C, Sjgrens syndrome, in-
ally normal. Although corticosteroids and plasma- flammatory bowel disease, melanoma, lymphoma,
pheresis are ineffective treatments, multifocal motor diabetes mellitus,26,27 and IgM, IgG, or IgA mono-
neuropathy improves with immune globulin18 or clonal gammopathy of unknown significance.13,28
cyclophosphamide19 therapy. The pathogenetic relevance of such concurrent dis-
eases is unclear. Furthermore, in contrast to distal
Multifocal Acquired Demyelinating Sensory acquired demyelinating symmetric neuropathy with
and Motor Neuropathy (the LewisSumner Syndrome) IgM paraproteinemia, the clinical presentation with
Multifocal acquired demyelinating sensory and mo- both proximal and distal muscle weakness is identi-
tor neuropathy (the LewisSumner syndrome) has cal to that of classic chronic inflammatory polyneu-
similarities to both chronic inflammatory demye- ropathy, and therapeutic guidelines are the same.
linating polyneuropathy (i.e., motor and sensory The association with diabetes mellitus is of spe-
deficits, an elevated protein content, and abnormal cial interest because, according to some estimates,
results on motor-nerve and sensory-nerve con- chronic inflammatory demyelinating polyneurop-
duction studies) and multifocal motor neuropathy athy occurs more commonly among patients with
(i.e., asymmetrical presentation of symptoms, often diabetes, generating diagnostic and management
starting from the arms and hands, and conduction challenges.27 Occasionally, chronic inflammatory
block).20-22 Some patients with the condition have demyelinating polyneuropathy may develop in a
antibodies to gangliosides,23 and these patients have setting of another polyneuropathy, even one with a
a reasonably good response to treatment with in- hereditary basis, such as CharcotMarieTooth
travenous immune globulin or cyclophosphamide. disease.29

other neuropathies similar to chronic central nervous system involvement

inflammatory demyelinating Magnetic resonance imaging (MRI) of the brain
polyneuropathy has revealed demyelinating lesions in the central
A number of other forms of acquired and chronic nervous system in some patients with chronic in-
polyneuropathy share features with chronic inflam- flammatory demyelinating polyneuropathy, despite
matory demyelinating polyneuropathy and have the rarity of cerebral or cerebellar symptoms.30,31
been classified as subgroups. These forms include Demyelination of visual pathways, however, as evi-
axonal chronic inflammatory demyelinating poly- denced by prolonged latencies of visual evoked po-
neuropathy, pure sensory chronic inflammatory de- tentials, were identified in nearly half of the patients
myelinating polyneuropathy,10 and pure motor and with chronic inflammatory demyelinating polyneu-
axonal chronic inflammatory demyelinating poly- ropathy in one study.30 Symptoms that are related
neuropathy (which is also termed multifocal ac- to cranial-nerve dysfunction are also seen in 5 to
quired motor axonopathy).24 Only a small number 30 percent of patients with the condition.30,31 Of
of patients within each subgroup have been report- interest, clinical symptoms that are based in the cen-
ed. Patients with peripheral-nerve demyelination tral nervous system as well as brain lesions that are
and a complete or partial response to immuno- visualized on MRI may resolve after treatment with
therapies are best regarded as having a disorder immune globulins.32
that is part of the larger family of chronic acquired
demyelinating polyneuropathies.10 Depending on diagnostic approach
the entire picture, some patients condition may also
fit the definition of possible, probable, or definiteThe diagnosis of distal acquired demyelinating
chronic inflammatory demyelinating polyneurop- symmetric neuropathy is based mainly on the clin-
athy. Chronic idiopathic axonal polyneuropathy is ical presentation and on nerve-conduction findings
a heterogeneous group of slowly progressing sen- that are consistent with demyelination (Table 1).
sorimotor neuropathies with or without pain, caus- Elevation of the protein content of the cerebrospi-
ing mild-to-moderate disability.25 nal fluid, without pleocytosis, and histologic proof
of demyelination and remyelination, often with in-
concurrent diseases flammation, in nerve-biopsy specimens provide ad-
Chronic inflammatory demyelinating polyneuropa- ditional supporting data. When the diagnosis is
thy may be also associated with concurrent diseases, not clear, we recommend nerve biopsy, given the

n engl j med 352;13 www.nejm.org march 31, 2005 1345

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

various therapeutic implications and the poten- fied.10,11 Thaisetthawatkul et al. emphasized the
tially serious adverse effects of long-term treatmentdispersion of the distal compound muscle action
with immunomodulatory or immunosuppressive potential as a very sensitive diagnostic criterion
drugs. A list of the most relevant elements of the for chronic inflammatory demyelinating polyneu-
differential diagnosis is provided in Table 2. ropathy.33 Although research criteria for enroll-
ment in clinical studies need to have a high speci-
electrophysiological diagnostic ficity, clinical criteria should be more sensitive to
procedures allow the identification of patients who may need
Nerve-conduction studies reveal the cardinal fea- treatment.14
tures of demyelination. An ad hoc committee of the
American Academy of Neurology included manda- laboratory examinations
tory physiological features as the presence of three Most experts recommend cerebrospinal fluid analy-
of the following four criteria for demyelination9: sis in order to demonstrate the typical findings in
partial motor-nerve conduction block (Fig. 1A), re- this condition: increased protein and a normal or
duced motor-nerve conduction velocity, prolonged only slightly elevated cell count. However, spinal
distal motor latencies, and prolonged F-wave la- taps are not mandatory, according to the criteria
tencies. To define inclusion criteria for clinical of the INCAT group (Table 1). More extended lab-
studies, the demyelination criteria have been modi- oratory testing may also be necessary in some pa-

Table 2. Differential Diagnosis.

Neuropathy Examples Remarks

GuillainBarr syndrome Muscular weakness progressing over
a period of 1 mo
Inherited neuropathy Hereditary motor and sensory neuropathy; hereditary neuropathy Family history and DNA analysis
with susceptibility to pressure palsies needed
Recessively inherited neuropathies Family history often negative
Metabolic neuropathy Diabetic neuropathy and neuropathy associated with impaired glu- Appropriate laboratory testing needed
cose tolerance; uremic, hepatic, and acromegalic neuropathy;
neuropathy associated with hypothyroidism
Paraneoplastic neuropathy Neuropathy associated with lymphoma or carcinoma Workup for underlying cancer needed
Neuropathy associated with Neuropathy associated with osteosclerotic myeloma, with mono- Workup for underlying cancer needed
monoclonal gammopathy clonal gammopathies of undetermined significance, and with
Waldenstrms macroglobulinemia
Neuropathy associated with Infection with the human immunodeficiency virus Appropriate laboratory testing needed
infectious diseases Leprosy Typically starts with sensory loss;
minor weakness in later stages
Borreliosis (including Lyme disease) Appropriate laboratory testing needed
Diphtheria Microbiologic culture of isolates
Neuropathy associated with Sarcoidosis; neuropathy associated with acquired amyloidosis; vas- Appropriate laboratory testing needed
systemic inflammatory or culitis, including polyarteritis nodosa, ChurgStrauss syn- and sural-nerve or muscle biopsy
immune-mediated diseases drome, rheumatoid arthritis, Sjgrens syndrome, Wegeners if condition is suspected
granulomatosis, systemic lupus erythematosus, systemic scle-
rosis, giant-cell arteritis, Behets syndrome, cryoglobulinemia,
Castlemans disease
Nonsystemic vasculitic neuropathy Sural-nerve or muscle biopsy needed
if condition is suspected
Toxic neuropathies Alcohol, industrial agents (e.g., acrylamide), metals (e.g., lead), Axonal more than demyelinating
drugs (e.g., platinum-based agents, amiodarone, perhexiline,
tacrolimus, chloroquine, and suramin)
Neuropathy due to nutritional Deficiency of vitamin B1, B6, B12, or E Appropriate laboratory testing needed
Porphyria-associated Appropriate laboratory testing needed
Polyneuropathy associated Polyneuropathy associated with sepsis, multiple-organ failure,
with critical illness or long-term ventilation

1346 n engl j med 352;13 www.nejm.org march 31 , 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

tients to search for other causes of a demyelinating myelinating polyneuropathy, for several reasons.
polyneuropathy, as well as concurrent diseases The most prominent abnormalities may lie in the
(Table 2). proximal segments of the nerves or roots or in mo-
tor nerves, which are areas not accessible to biopsy.
nerve biopsy Moreover, concomitant or secondary axonal chang-
The diagnostic value of nerve biopsy, usually of the es starting early in the disease processes may over-
sural nerve, has been extensively debated during
the past few years. Some experts believe that nerve
biopsy is of no diagnostic value,34 whereas others
view it as essential for diagnosis and management
in up to 60 percent of patients with chronic inflam- 5 mV
matory demyelinating polyneuropathy. Bosboom
5 msec
et al.36 compared signs of demyelination, axonal de-
generation, regeneration, and inflammation in bi-
opsy specimens from patients with chronic inflam-
matory demyelinating polyneuropathy with those
of patients with chronic idiopathic axonal polyneu-
ropathy. The biopsy specimens from the majority of
patients in both groups had similar or overlapping
abnormalities. In addition, nerve biopsies may have
a low diagnostic yield in chronic inflammatory de-

Figure 1. Diagnostic Findings in Chronic Inflammatory

Demyelinating Polyneuropathy.
Panel A shows a partial motor-nerve conduction block
and abnormal temporal dispersion in a nerve-conduction
study, with a reduction of compound muscle action po-
tentials from the abductor digiti minimi muscle after
ulnar nerve stimulation at the elbow (bottom), as com-
pared with the amplitude after stimulation at the wrist
(top). Axial T1-weighted MRI scans of the lower thoracic
spine, shown before the administration of gadolinium in
Panel B and after the administration of gadolinium in
Panel C, reveal strong enhancement of ventral and dorsal
nerve roots (Panel C, arrows). Cross-sections of a sural D E
nerve in Panels D and E show typical features of chronic
inflammatory demyelinating polyneuropathy, with im-
munohistochemical staining mirroring the distribution
pattern of T lymphocytes and macrophages. Invading
CD3+ T cells can primarily be localized to perivascular
infiltrates (Panel D, arrows) in the epineurium and
perineurium, and CD68+ immunoreactive macrophages
(Panel E, arrows) can be seen within the endoneurium.
Panel F shows a semithin section in which the extent of
the inflammatory process is reflected by the loss of mye- F
lin (arrowheads indicate demyelinated axons and arrows
the remains of thinly myelinated fibers) and the invading
macrophages (open arrow). In Panel G, an electron micro-
graph shows the onion-bulb formation of Schwann cells
(arrow) around demyelinated axons. (MRI scans were pro-
vided by A. Saleh, Institute for Diagnostic Radiology,
Heinrich-Heine University, Dsseldorf; the semithin sec-
tion by E. Neuen-Jacob, Institute of Neuropathology,
University of Dsseldorf; and the electron micrograph by
J. Pollard, University of Sydney.)

n engl j med 352;13 www.nejm.org march 31, 2005 1347

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

shadow the initial signs of demyelination and in- to autoantigens is key for the maintenance of self-
flammation by the time biopsy is performed. tolerance. In chronic inflammatory demyelinating
Despite these limitations, nerve biopsy is still polyneuropathy, self-tolerance breaks down, and
considered useful by many specialists under certain autoreactive T cells and B cells, which are part of
conditions (Fig. 1D to 1G). Haq et al. observed that the normal immune repertoire, become activated,
examination of sural-nerve biopsy specimens had a causing the organ-specific damage characteristic of
higher sensitivity than electrophysiological stud- autoimmune disease.44 The concept of molecular
ies.37 Likewise, Vallat et al. reported that 8 patients mimicry may hold special relevance to the break-
in a series of 44 had pathological findings indicative down in tolerance associated with autoimmune
of chronic inflammatory demyelinating polyneu- neuropathies. Molecular mimicry refers to a process
ropathy on biopsy even though they did not have in which the host generates an immune response
electrophysiological evidence of demyelination.38 to an inciting factor, most frequently an infectious
It is important to note that five of these patients had organism that shares epitopes with the hosts affect-
a favorable response to therapy.38 ed tissue. However, in chronic inflammatory demy-
Biopsy is recommended especially for patients elinating polyneuropathy, specific targets for such
with clinically suspected chronic inflammatory de- a response have been convincingly identified only
myelinating polyneuropathy in whom electrophys- in rare instances.
iological proof of demyelination is absent or vascu- Although chronic inflammatory demyelinating
litis is suspected. In a series of 100 patients with polyneuropathy occurs rarely in the context of can-
chronic inflammatory demyelinating polyneurop- cer, an association with melanoma is of great inter-
athy, Bouchard et al.39 observed that axonal loss est, since both melanoma and Schwann cells derive
on nerve biopsy was the most sensitive prognos- from neural crest tissues and share antigens. Sev-
tic factor, predicting an unfavorable course of the eral cases of chronic inflammatory demyelinat-
disease. They found demyelinating changes in 71 ing polyneuropathy have been reported in associa-
percent of the patients, mixed axonal and demye- tion with melanoma; several carbohydrate epitopes
linating changes in 21 percent, and purely axonal shared by the myelin sheath and the tumor have
changes in only 5 percent. A diagnostic algorithm been implicated as target antigens.45,46 Neverthe-
is shown in Figure 2. less, the hypothesis of molecular mimicry cannot
explain the entire immunopathologic and labo-
mri ratory spectrum of this complex disorder. On the
MRI may be used to demonstrate gadolinium en- basis of current data, chronic inflammatory demye-
hancement (Fig. 1B and 1C) and enlargement of linating polyneuropathy appears to be an organ-spe-
proximal nerves or roots, reflecting active inflam- cific, immune-mediated disorder emerging from a
mation and demyelination in the cauda equina40 or synergistic interaction of cell-mediated and humor-
brachial plexus.41-43 Abnormalities of the brachial al immune responses directed against incomplete-
plexus with irregular swelling and increased signal ly characterized peripheral nerve antigens (Fig. 3).
intensity on T2-weighted images were detected in
about 50 percent of patients with chronic inflam- cellular immune response
matory demyelinating polyneuropathy.34 Of inter- Evidence of T-cell activation in the systemic im-
est, these changes have also been noted in patients mune compartment in patients with chronic in-
with distal demyelinating polyneuropathy associ- flammatory demyelinating polyneuropathy exists,
ated with IgM monoclonal gammopathy,42 point- although antigen specificity remains largely un-
ing to similarly widespread nerve disease in the lat- known.47-49 From studies of nerve-biopsy speci-
ter condition. mens and animal models, it is known that acti-
vated T lymphocytes can invade peripheral-nerve
pathogenesis tissue. The T-cell populations that have been iden-
tified are heterogeneous, belonging to both the CD4
A normal, well-balanced network of immunocom- and CD8 subgroups.50-54 In order to generate in-
petent cells and soluble factors meticulously reg- flammatory lesions in nerves, activated T cells must
ulates the immune system within the local tissue cross the bloodnerve barrier, a complex process
compartment of the peripheral nerves, sustaining that includes homing, adhesion, and transmigra-
its integrity. Protection against immune responses tion.55 Derangement of the bloodnerve barrier has

1348 n engl j med 352;13 www.nejm.org march 31 , 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

Progressive or relapsing proximal paresis, distal

paresis, or both, with or without hypoesthesia
and reduced or no tendon reflexes

Duration <2 mo Duration 2 mo

Electrophysiological Electrophysiological examination Laboratory examination Other family members with

examination may be for signs of demyelination (including CSF analysis) similar symptoms
negative in the first
12 wk

Electrodiagnostic and CSF criteria

Electrodiagnostic and supportive
fulfilled and no evidence of another
CSF criteria not fulfilled
cause of polyneuropathy

Signs of demyelination in sural-nerve

biopsy specimen and no evidence
of other cause of polyneuropathy

Consider GuillainBarr Start treatment for chronic inflammatory Consider hereditary motor
syndrome demyelinating polyneuropathy and sensory neuropathy

Figure 2. Algorithm of Diagnostic Procedures.

If a patient presents with a history of symptoms suggestive of chronic inflammatory demyelinating polyneuropathy of two months duration
or more, we perform nerve-conduction studies for signs of demyelination including partial conduction block, reduced motor-nerve con-
duction velocity, prolonged distal latency of the motor nerve, and the absence of F waves or a prolonged F-wave latency to differentiate be-
tween predominantly demyelinating or axonal disease of peripheral nerves. We also use laboratory tests including cell-count and protein
studies of cerebrospinal fluid (CSF) to evaluate supportive criteria and to rule out other causes. If these causes have been ruled out and
electrodiagnostic and supportive CSF criteria are fulfilled, patients may begin long-term antiinflammatory and immunosuppressive therapy.
To confirm the diagnosis, we recommend sural-nerve biopsy.

been shown by demonstrating that the tight-junc- and secrete cytokines such as tumor necrosis fac-
tion proteins claudin-5 and ZO-1 are down-regu- tor a, interferon-g, and interleukin-2.55,63 T cells
lated in sural-nerve biopsy specimens.56 Elevated thereby activate resident endoneurial or passenger
levels of soluble adhesion molecules,57,58 chemo- macrophages, which then discharge an array of
kines,59,60 and matrix metalloproteinases61,62 can neurotoxic and immunopotentiating molecules
be detected in serum, cerebrospinal fluid, or both (i.e., oxygen radicals, nitric oxide metabolites, ar-
findings that are indicative of active T-cell mi- achidonic acid metabolites, proteases, and comple-
gration across the bloodnerve barrier. ment components)64,65 or engage in increased
Once within the peripheral nervous system, these phagocytic and cytotoxic activity against myelin or
T cells may undergo clonal expansion after encoun- Schwann cells. On the other hand, specialized sub-
tering an antigen presented in the context of appro- populations of T cells may terminate the acute im-
priate major-histocompatibility-complex molecules munoinflammatory process by secreting down-reg-
and costimulatory signals. Such T cells then express ulatory cytokines (e.g., transforming growth factor

n engl j med 352;13 www.nejm.org march 31, 2005 1349

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

Systemic Immune Peripheral Nervous System

Nerve cell

Macrophage Apoptosis

cell Schwann cell
Autoreactive Lytic
Autoreactive T cell membrane-attack
T cell complex
Tumor necrosis
Macrophage factor a C5b-9
Adhesion Chemokines Reactivation
and expansion Reactive oxygen and
molecules Interferon-g nitric oxide
Type 1 Tumor necrosis Complement
factor a Proteases
helper T cell
T cells

metalloproteinases Interleukin-10
growth factor b
Interleukin-4 Type 2 Plasma cell Autoantibodies
Interleukin-6 helper T cell
Compact Myelin Noncompact Myelin
Myelin protein zero Myelin basic Ganglioside
Plasma cell

Bloodnerve associated
Myelin protein 2 glycoprotein Connexin 32

Figure 3. Immunopathogenesis of Chronic Inflammatory Demyelinating Neuropathy.

A schematic illustration of the basic principles of the cellular and humoral immune responses shows that autoreactive T cells recognize a spe-
cific autoantigen in the context of major histocompatibility complex class II and costimulatory molecules on the surface of antigen-presenting
cells (macrophages) in the systemic immune compartment. An infection might trigger this event through molecular mimicry, a cross-reaction
toward epitopes shared between the microbial agent and nerve antigens. These activated T lymphocytes can cross the bloodnerve barrier in
a process involving cellular adhesion molecules, matrix metalloproteinases, and chemokines. Within the peripheral nervous system, T cells
activate macrophages that enhance phagocytic activity, the production of cytokines, and the release of toxic mediators, including nitric oxide,
reactive oxygen intermediates, matrix metalloproteinases, and proinflammatory cytokines, including tumor necrosis factor a and interferon-
g. Autoantibodies crossing the bloodnerve barrier or locally produced by plasma cells contribute to demyelination and axonal damage. Au-
toantibodies can mediate demyelination by antibody-dependent cellular cytotoxicity, potentially block epitopes that are functionally relevant
for nerve conduction, and activate the complement system by the classic pathway, yielding proinflammatory mediators and the lytic mem-
brane-attack complex C5b-9. Termination of the inflammatory response occurs through the induction of T-cell apoptosis and the release of
antiinflammatory cytokines, including interleukin-10 and transforming growth factor b. The myelin sheath (inset) is composed of various pro-
teins, such as myelin protein zero, which account for more than 50 percent of the total membrane protein in human peripheral nervous sys-
tem myelin; myelin protein 2; myelin basic protein; myelin-associated glycoprotein; connexin 32; and gangliosides and related glycolipids.
These molecules have been identified as target antigens for antibody responses with varying frequencies in patients with this disease.

b) or other molecules. It is important to note that neuropathy, a finding that is underscored by the
the local immune environment of the peripheral observed expression of major-histocompatibility-
nerves appears to facilitate the apoptosis of invad- complex class II molecules and the class Ilike mol-
ing autoaggressive T cells,66 a process augmented ecule CD1a in nerve-biopsy specimens.68 Costimu-
by therapeutically administered corticosteroids.67 latory molecules B7-1 and B7-2 are essential for
Macrophages also serve as antigen-presenting effective antigen presentation and may determine
cells in chronic inflammatory demyelinating poly- the differentiation of T lymphocytes into a pheno-

1350 n engl j med 352;13 www.nejm.org march 31 , 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

type of type 1 or type 2 helper cells, thus modulating various antibodies and separate mechanisms are
the local immune response and the clinical course involved in individual patients.
of the disease. A spontaneous immune neuropathy
with clinical, electrophysiological, and morpholog- axonal loss
ic similarities to chronic inflammatory demyelinat- Chronic inflammatory demyelinating polyneurop-
ing polyneuropathy in humans develops in auto- athy, though a demyelinating polyneuropathy, is
immune nonobese diabetic mice that are deficient associated with a concomitant axonal loss attri-
in B7-2 costimulation.69 buted to the primary demyelinating process.39,48
The cellular immune response within the pe- This finding appears to be important, since the long-
ripheral nervous system is tightly regulated at the term prognosis in chronic inflammatory demyelin-
transcriptional level. One of its key regulators, the ating polyneuropathy depends on the magnitude
transcription factor nuclear factor-kB, is up-regu- of axonal loss rather than on demyelination. There
lated predominantly in macrophages in chronic in- are questions as to whether the release of neuro-
flammatory demyelinating polyneuropathy.70 toxic cytokines (e.g., tumor necrosis factor a) and
noxious mediators (e.g., nitric oxide and metallo-
humoral immune response proteinases) enhances axonal destruction, but it
The contribution of autoantibodies to the patho- has become clear that early, effective therapy mini-
genesis of chronic inflammatory demyelinating mizes axonal loss.
polyneuropathy was suggested more than 20 years
ago on the basis of immunoglobulin and comple- current treatment
ment deposition on myelinated nerve fibers71 and
the presence of oligoclonal IgG bands in the cere- In general, therapies are directed at blocking im-
brospinal fluid.72 Passive transfer experiments have mune processes to arrest inflammation and demy-
demonstrated that serum or purified IgG from pa- elination and to prevent secondary axonal degen-
tients with chronic inflammatory demyelinating eration. In patients who have a response, treatment
polyneuropathy induces conduction block and de- must be continued until maximum improvement
myelination in rat nerves.73 In these experiments, or stabilization occurs; thereafter, maintenance
the 28-kD myelin protein zero was identified as one therapy is required and must be tailored to the indi-
of the putative target antigens.74 vidual patient, with the goal of preventing or dimin-
Gangliosides and related glycolipids may also ishing the frequency of relapses or disease progres-
be target antigens (Fig. 3, inset). In a few patients sion. A positive response to therapy is determined
with chronic inflammatory demyelinating polyneu- by a measurable improvement in strength and sen-
ropathy, there is serologic evidence of recent in- sation and the patients ability to perform activi-
fection with Campylobacter jejuni. Given the shared ties of daily living. It is important to be aware that
expression of carbohydrate epitopes in nerve gly- infections and febrile conditions may also affect de-
colipids and microbial lipopolysaccharides, this myelination and thereby worsen the clinical symp-
finding may hint at molecular mimicry as the un- toms of chronic inflammatory demyelinating poly-
derlying cause of chronic inflammatory demyelin- neuropathy. Concomitant use of neurotoxic drugs
ating polyneuropathy in rare instances.75 GM1 an- or the presence of systemic conditions known to
tiserum from a patient with chronic inflammatory cause neuropathies may also theoretically influence
demyelinating polyneuropathy substantially sup- the clinical symptoms of the condition.
pressed sodium currents in single myelinated nerve The most widely used treatments for chronic
fibers from rats.76 Serum reactivity against presum- inflammatory demyelinating polyneuropathy (Ta-
ably nonmyelin antigens on Schwann cells has re- ble 3) consist of intravenous immune globu-
cently been reported in 12 of 46 patients studied.77 lin,11,80,81,84-86 plasma exchange,78,79 and corti-
Demyelination and conduction block may also costeroids.11,87,88 Therapy should be initiated early
result from serum constituents other than mye- in the course of the disease to prevent continuing
lin-directed antibodies, such as cytokines, comple- demyelination and secondary axonal loss leading
ments, or other inflammatory mediators (e.g., ni- to permanent disability. According to published
tric oxide). The low frequency of specific antibodies data, there appears to be no difference in efficacy
that is observed in patients with chronic inflam- among these three main therapies.28,85,87 The de-
matory demyelinating polyneuropathy suggests that cision to choose one of them is usually made on

n engl j med 352;13 www.nejm.org march 31, 2005 1351

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

Table 3. Current Therapy Based on the Results of Randomized, Controlled Studies.*

No. of
Reference Year Therapy Patients Duration Design Result
Dyck et al.78 1994 Plasma exchange vs. intrave- 15 42 days Randomized, observer- No significant
nous immune globulin blinded, crossover difference
Hahn et al.79 1996 Plasma exchange 15 28 days Double-blind, sham- Improvement in
controlled, crossover 80% of patients
Hahn et al.80 1996 Intravenous immune 30 28 days Double-blind, placebo- Improvement in
globulin controlled, crossover 63% of patients
Mendell et al.81 2001 Intravenous immune 53 42 days Double-blind, randomized, Improvement in
globulin placebo-controlled 76% of patients
Hughes et al.11 2001 Intravenous immune globu- 32 14 days Double-blind, randomized, No significant
lin vs. oral prednisolone crossover difference
Dyck et al.82 1985 Azathioprine in combination 30 9 mo Open, parallel-group, No significant
with prednisone vs. pred- randomized difference
nisone alone
Hadden et al.83 1999 Interferon beta-1a in treat- 20 28 wk Double-blind, randomized, No significant
ment-resistant disease placebo-controlled, benefit of
crossover treatment

* Most of the clinical trials in chronic inflammatory demyelinating polyneuropathy have been limited to several weeks,
which is a rather short time period for a disease that typically spans months or years.

the basis of cost, availability (e.g., plasmapheresis a week for six months.93 Thirty-five percent of the
and venous access), and side effects (most impor- patients had an improvement, and the disease sta-
tant, the serious long-term side effects of cortico- bilized in 50 percent, prompting the authors to
steroids).87 All these factors should be considered recommend a larger, placebo-controlled trial. How-
when costutility analyses are performed.89 In some ever, another study, in which four patients with
60 to 80 percent of patients with classic chronic in- chronic inflammatory demyelinating polyneurop-
flammatory demyelinating polyneuropathy, the con- athy were treated, showed that treatment with in-
dition improves while they are receiving one of the terferon beta-1a was effective only in combination
three therapies, but the long-term prognosis ap- with intravenous immune globulin.94 Furthermore,
pears to vary according to the time at which therapy a small, randomized, double-blind, placebo-con-
is initiated and the degree of associated axonal trolled, crossover study involving 10 patients with
loss. Azathioprine,82 cyclophosphamide, and cyclo- treatment-resistant chronic inflammatory demy-
sporine have long been used mainly as secondary elinating polyneuropathy and evaluating interfer-
agents in the therapy of chronic inflammatory de- on beta-1a (3 million IU for 2 weeks and 6 million IU
myelinating polyneuropathy, but reliable data on for 10 weeks, administered subcutaneously three
their efficacy from randomized, controlled trials are times per week) failed to show a significant treat-
not available.90 For unknown reasons, the efficacy ment effect.83 The role of interferon alfa in the con-
of these latter treatments is clearly less favorable in dition is also uncertain. Some case reports95,96 and
patients who have a neuropathy accompanied by an open-label prospective pilot study97 suggested
antibodies to myelin-associated glycoprotein.91,92 that interferon alfa was effective.
Given the presumed autoimmune cause of this Of concern, chronic inflammatory demyelinat-
condition and its suggested pathogenetic similari- ing polyneuropathy has been reported to develop
ties to multiple sclerosis, immunomodulatory ther- during treatment with interferon alfa98-100 or inter-
apies that are considered effective in that disorder feron beta.101 Furthermore, interferon was ineffec-
have been investigated. Twenty patients with treat- tive in patients with IgM monoclonal gammop-
ment-resistant chronic inflammatory demyelinat- athy102 and the GuillainBarr syndrome.103 These
ing polyneuropathy were enrolled in a prospective, disturbing observations raise the provocative ques-
multicenter, open-label study that evaluated intra- tion of whether interferons are causally related to
muscular interferon beta-1a at a dose of 30 g once the onset of chronic inflammatory demyelinating

1352 n engl j med 352;13 www.nejm.org march 31, 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

polyneuropathy, as opposed to being capable of more, controlled trials providing long-term data are
suppressing the disease.104 Hughes et al. conclud- lacking. The evidence concerning the efficacy of
ed that there is currently no adequate evidence to plasma exchange, intravenous immune globulin,
decide whether interferons are beneficial in the and corticosteroids derives only from short-term
treatment of this condition.90 studies. Plasma exchange and intravenous immune
Other forms of treatment have been tested in globulin are expensive therapies and must be con-
open-label studies with a small number of patients tinued over the long term to maintain benefit. Anec-
or in individual patients. Beneficial effects in pa- dotal experience suggests that the use of immuno-
tients with previously treatment-resistant chronic suppressive agents may allow therapy with plasma
inflammatory demyelinating polyneuropathy were exchange or intravenous immune globulin to be ad-
reported for the combination of plasmapheresis ministered less frequently or even phased out, with
and intravenous immune globulin,105 mycophen- subsequent substantial financial savings. There is
olate mofetil,106-108 cyclosporine,109-111 etaner- clearly a need for controlled studies to assess this
cept,112 cyclophosphamide,113,114 and autologous long-term aspect of therapy for chronic inflamma-
hematopoietic stem-cell transplantation.115 In pa- tory demyelinating polyneuropathy.
tients with multifocal motor neuropathy or chronic
inflammatory demyelinating polyneuropathy, the conclusions
combination of intravenous immune globulin and
mycophenolate mofetil may permit a reduction in It is important to recognize chronic inflammato-
the dose of immune globulin or corticosteroids, a ry demyelinating polyneuropathy in a patient with
finding that was recently suggested by an open- a chronic progressive or chronic relapsing neu-
label study of 6 patients116 and a retrospective ropathy, since therapies that are at least partially
analysis of the efficacy of mycophenolate mofetil effective including corticosteroids, intravenous
in 21 patients with chronic inflammatory demye- immune globulin, plasma exchange, and immuno-
linating polyneuropathy.108 Two recent open-label suppressants are available for this crippling dis-
studies involving 30 patients found improvement ease. Sets of diagnostic criteria have been devel-
in those with IgM-associated demyelinating poly- oped. The disorder appears to be heterogeneous in
neuropathy who were receiving treatment with terms of clinical presentation and immunopatho-
rituximab, a chimeric humanized monoclonal an- genesis. Further research should provide further
tibody against CD20 antigen that reduces B-lym- insight into the underlying mechanisms of nerve
phocyte counts.117,118 However, no data that have damage and may facilitate the development of more
been collected on the basis of randomized, con- effective treatments.
trolled studies with a sufficient number of patients We are indebted to Marinos C. Dalakas, of Bethesda, Md., David
R. Cornblath, of Baltimore, and John Pollard, of Sydney, for their
are available to allow conclusive recommendations critical review of the manuscript and many helpful suggestions.
about treatment with any of these agents. Further-

1. Connolly AM. Chronic inflammatory Hewer RL. Recurrent and chronic relapsing committee of the American Academy of
demyelinating polyneuropathy in child- Guillain-Barre polyneuritis. Brain 1969;92: Neurology AIDS Task Force. Neurology 1991;
hood. Pediatr Neurol 2001;24:177-82. 589-606. 41:617-8.
2. McLeod JG, Pollard JD, Macaskill P, Mo- 6. Dyck PJ, Lais AC, Ohta M, Bastron JA, 10. Saperstein DS, Katz JS, Amato AA,
hamed A, Spring P, Khurana V. Prevalence of Okazaki H, Groover RV. Chronic inflam- Barohn RJ. Clinical spectrum of chronic ac-
chronic inflammatory demyelinating poly- matory polyradiculoneuropathy. Mayo Clin quired demyelinating polyneuropathies.
neuropathy in New South Wales, Australia. Proc 1975;50:621-37. Muscle Nerve 2001;24:311-24.
Ann Neurol 1999;46:910-3. 7. Dalakas MC, Engel WK. Chronic relaps- 11. Hughes R, Bensa S, Willison H, et al.
3. Lunn MP, Manji H, Choudhary PP, ing (dysimmune) polyneuropathy: patho- Randomized controlled trial of intravenous
Hughes RA, Thomas PK. Chronic inflamma- genesis and treatment. Ann Neurol 1981;9: immunoglobulin versus oral prednisolone
tory demyelinating polyradiculoneuropa- Suppl:134-45. in chronic inflammatory demyelinating poly-
thy: a prevalence study in south east England. 8. Barohn RJ, Kissel JT, Warmolts JR, Men- radiculoneuropathy. Ann Neurol 2001;50:
J Neurol Neurosurg Psychiatry 1999;66:677- dell JR. Chronic inflammatory demyelinat- 195-201.
80. ing polyradiculoneuropathy: clinical char- 12. Latov N. Diagnosis of CIDP. Neurology
4. Austin JH. Recurrent polyneuropathies acteristics, course, and recommendations 2002;59:Suppl 6:S2-S6.
and their corticosteroid treatment; with five- for diagnostic criteria. Arch Neurol 1989;46: 13. Sander HW, Latov N. Research criteria
year observations of a placebo-controlled 878-84. for defining patients with CIDP. Neurology
case treated with corticotrophin, cortisone, 9. Research criteria for diagnosis of chron- 2003;60:Suppl 3:S8-S15.
and prednisone. Brain 1958;81:157-92. ic inflammatory demyelinating polyneuro- 14. Magda P, Latov N, Brannagan TH III,
5. Thomas PK, Lascelles RG, Hallpike JF, pathy (CIDP): report from an ad hoc sub- Weimer LH, Chin RL, Sander HW. Compar-

n engl j med 352;13 www.nejm.org march 31, 2005 1353

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
The new england journal of medicine

ison of electrodiagnostic abnormalities and Coexistent hereditary and inflammatory neu- 44. Quattrini A, Previtali SC, Kieseier BC,
criteria in a cohort of patients with chronic ropathy. Brain 2004;127:193-202. Kiefer R, Comi G, Hartung HP. Autoimmu-
inflammatory demyelinating polyneuropa- 30. Stojkovic T, de Seze J, Hurtevent JF, et al. nity in the peripheral nervous system. Crit
thy. Arch Neurol 2003;60:1755-9. Visual evoked potentials study in chronic Rev Neurobiol 2003;15:1-39.
15. Berger AR, Bradley WG, Brannagan TH, idiopathic inflammatory demyelinating poly- 45. Weiss MD, Luciano CA, Semino-Mora C,
et al. Guidelines for the diagnosis and treat- neuropathy. Clin Neurophysiol 2000;111: Dalakas MC, Quarles RH. Molecular mim-
ment of chronic inflammatory demyelinat- 2285-91. icry in chronic inflammatory demyelinating
ing polyneuropathy. J Peripher Nerv Syst 31. Rotta FT, Sussman AT, Bradley WG, Ram polyneuropathy and melanoma. Neurology
2003;8:282-4. Ayyar D, Sharma KR, Shebert RT. The spec- 1998;51:1738-41.
16. Katz JS, Saperstein DS, Gronseth G, trum of chronic inflammatory demyelinat- 46. Tsuchida T, Saxton RE, Morton DL, Irie
Amato AA, Barohn RJ. Distal acquired de- ing polyneuropathy. J Neurol Sci 2000;173: RF. Gangliosides of human melanoma. Can-
myelinating symmetric neuropathy. Neurol- 129-39. cer 1989;63:1166-74.
ogy 2000;54:615-20. 32. Fee DB, Fleming JO. Resolution of chron- 47. Hartung HP, Reiners K, Schmidt B, Stoll
17. Mygland A, Monstad P. Chronic acquired ic inflammatory demyelinating polyneuropa- G, Toyka KV. Serum interleukin-2 concen-
demyelinating symmetric polyneuropathy thy-associated central nervous system lesions trations in Guillain-Barre syndrome and
classified by pattern of weakness. Arch Neu- after treatment with intravenous immuno- chronic idiopathic demyelinating polyra-
rol 2003;60:260-4. globulin. J Peripher Nerv Syst 2003;8:155-8. diculoneuropathy: comparison with other
18. Federico P, Zochodne DW, Hahn AF, 33. Thaisetthawatkul P, Logigian EL, Herr- neurological diseases of presumed immu-
Brown WF, Feasby TE. Multifocal motor neu- mann DN. Dispersion of the distal compound nopathogenesis. Ann Neurol 1991;30:48-53.
ropathy improved by IVIg: randomized, dou- muscle action potential as a diagnostic cri- 48. Dalakas MC. Advances in chronic in-
ble-blind, placebo-controlled study. Neurol- terion for chronic inflammatory demyelin- flammatory demyelinating polyneuropathy:
ogy 2000;55:1256-62. ating polyneuropathy. Neurology 2002;59: disease variants and inflammatory response
19. Pestronk A. Multifocal motor neuropa- 1526-32. mediators and modifiers. Curr Opin Neurol
thy: diagnosis and treatment. Neurology 34. Molenaar DS, Vermeulen M, de Haan R. 1999;12:403-9.
1998;51:Suppl 5:S22-S24. Diagnostic value of sural nerve biopsy in 49. Van den Berg LH, Mollee I, Wokke JH,
20. Oh SJ, Claussen GC, Kim DS. Motor chronic inflammatory demyelinating poly- Logtenberg T. Increased frequencies of HPRT
and sensory demyelinating mononeuropa- neuropathy. J Neurol Neurosurg Psychiatry mutant T lymphocytes in patients with Guil-
thy multiplex (multifocal motor and sensory 1998;64:84-9. lain-Barre syndrome and chronic inflamma-
demyelinating neuropathy): a separate enti- 35. Gabriel CM, Howard R, Kinsella N, et al. tory demyelinating polyneuropathy: further
ty or a variant of chronic inflammatory demy- Prospective study of the usefulness of sural evidence for a role of T cells in the etiopatho-
elinating polyneuropathy? J Peripher Nerv nerve biopsy. J Neurol Neurosurg Psychiatry genesis of peripheral demyelinating dis-
Syst 1997;2:362-9. 2000;69:442-6. eases. J Neuroimmunol 1995;58:37-42.
21. Lewis RA, Sumner AJ, Brown AJ, Asbury 36. Bosboom WM, van den Berg LH, Frans- 50. Schmidt B, Toyka KV, Kiefer R, Full J,
AK. Multifocal demyelinating neuropathy sen H, et al. Diagnostic value of sural nerve Hartung HP, Pollard J. Inflammatory infil-
with persistent conduction block. Neurolo- demyelination in chronic inflammatory de- trates in sural nerve biopsies in Guillain-
gy 1982;32:958-64. myelinating polyneuropathy. Brain 2001; Barre syndrome and chronic inflammato-
22. Viala K, Renie L, Maisonobe T, et al. Fol- 124:2427-38. ry demyelinating neuropathy. Muscle Nerve
low-up study and response to treatment in 37. Haq RU, Fries TJ, Pendlebury WW, Ken- 1996;19:474-87.
23 patients with Lewis-Sumner syndrome. ny MJ, Badger GJ, Tandan R. Chronic inflam- 51. Winer J, Hughes S, Cooper J, Ben-Smith
Brain 2004;127:2010-7. matory demyelinating polyradiculoneurop- A, Savage C. gd T cells infiltrating sensory
23. Alaedini A, Sander HW, Hays AP, Latov athy: a study of proposed electrodiagnostic nerve biopsies from patients with inflamma-
N. Antiganglioside antibodies in multifocal and histologic criteria. Arch Neurol 2000; tory neuropathy. J Neurol 2002;249:616-21.
acquired sensory and motor neuropathy. Arch 57:1745-50. 52. Illes Z, Kondo T, Newcombe J, Oka N,
Neurol 2003;60:42-6. 38. Vallat JM, Tabaraud F, Magy L, et al. Di- Tabira T, Yamamura T. Differential expres-
24. Katz JS, Barohn RJ, Kojan S, et al. Axonal agnostic value of nerve biopsy for atypical sion of NK T cell Va24JaQ invariant TCR
multifocal motor neuropathy without con- chronic inflammatory demyelinating poly- chain in the lesions of multiple sclerosis and
duction block or other features of demye- neuropathy: evaluation of eight cases. Mus- chronic inflammatory demyelinating poly-
lination. Neurology 2002;58:615-20. cle Nerve 2003;27:478-85. neuropathy. J Immunol 2000;164:4375-81.
25. Hughes RA, Umapathi T, Gray IA, et al. 39. Bouchard C, Lacroix C, Plante V, et al. 53. Illes Z, Shimamura M, Newcombe J, Oka
A controlled investigation of the cause of Clinicopathologic findings and prognosis of N, Yamamura T. Accumulation of Va7.2-Ja33
chronic idiopathic axonal polyneuropathy. chronic inflammatory demyelinating poly- invariant T cells in human autoimmune in-
Brain 2004;127:1723-30. neuropathy. Neurology 1999;52:498-503. flammatory lesions in the nervous system.
26. Gorson KC, Ropper AH, Adelman LS, 40. Midroni G, de Tilly LN, Gray B, Vajsar J. Int Immunol 2004;16:223-30.
Weinberg DH. Influence of diabetes melli- MRI of the cauda equina in CIDP: clinical 54. Bosboom WM, Van den Berg LH, Moll-
tus on chronic inflammatory demyelinating correlations. J Neurol Sci 1999;170:36-44. ee I, et al. Sural nerve T-cell receptor Vb gene
polyneuropathy. Muscle Nerve 2000;23:37- 41. Duggins AJ, McLeod JG, Pollard JD, et utilization in chronic inflammatory demye-
43. al. Spinal root and plexus hypertrophy in linating polyneuropathy and vasculitic neu-
27. Haq RU, Pendlebury WW, Fries TJ, Tan- chronic inflammatory demyelinating poly- ropathy. Neurology 2001;56:74-81.
dan R. Chronic inflammatory demyelinat- neuropathy. Brain 1999;122:1383-90. 55. Gold R, Archelos JJ, Hartung HP. Mech-
ing polyradiculoneuropathy in diabetic pa- 42. Eurelings M, Notermans NC, Franssen anisms of immune regulation in the periph-
tients. Muscle Nerve 2003;27:465-70. H, et al. MRI of the brachial plexus in poly- eral nervous system. Brain Pathol 1999;9:
28. Gorson KC, Allam G, Ropper AH. Chron- neuropathy associated with monoclonal 343-60.
ic inflammatory demyelinating polyneurop- gammopathy. Muscle Nerve 2001;24:1312-8. 56. Kanda T, Numata Y, Mizusawa H. Chron-
athy: clinical features and response to treat- 43. Van Es HW, Van den Berg LH, Franssen ic inflammatory demyelinating polyneu-
ment in 67 consecutive patients with and H, et al. Magnetic resonance imaging of the ropathy: decreased claudin-5 and relocated
without a monoclonal gammopathy. Neu- brachial plexus in patients with multifocal ZO-1. J Neurol Neurosurg Psychiatry 2004;
rology 1997;48:321-8. motor neuropathy. Neurology 1997;48:1218- 75:765-9.
29. Ginsberg L, Malik O, Kenton AR, et al. 24. 57. Previtali SC, Archelos JJ, Hartung HP.

1354 n engl j med 352;13 www.nejm.org march 31, 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

Expression of integrins in experimental au- 71. Dalakas MC, Engel WK. Immunoglobu- ing neuropathies. Neurology 2002;59:Suppl
toimmune neuritis and Guillain-Barre syn- lin and complement deposits in nerves of pa- 6:S13-S21.
drome. Ann Neurol 1998;44:611-21. tients with chronic relapsing polyneuropa- 85. Van Schaik IN, Winer JB, De Haan R,
58. Previtali SC, Feltri ML, Archelos JJ, thy. Arch Neurol 1980;37:637-40. Vermeulen M. Intravenous immunoglobu-
Quattrini A, Wrabetz L, Hartung H. Role of 72. Dalakas MC, Houff SA, Engel WK, Mad- lin for chronic inflammatory demyelinating
integrins in the peripheral nervous system. den DL, Sever JL. CSF monoclonal bands polyradiculoneuropathy. Cochrane Database
Prog Neurobiol 2001;64:35-49. in chronic relapsing polyneuropathy. Neu- Syst Rev 2002;2:CD001797.
59. Kastenbauer S, Koedel U, Wick M, rology 1980;30:864-7. 86. Vermeulen M, van Doorn PA, Brand A,
Kieseier BC, Hartung HP, Pfister HW. CSF 73. Yan WX, Taylor J, Andrias-Kauba S, Pol- Strengers PF, Jennekens FG, Busch HF. In-
and serum levels of soluble fractalkine lard JD. Passive transfer of demyelination by travenous immunoglobulin treatment in
(CX3CL1) in inflammatory diseases of the serum or IgG from chronic inflammatory patients with chronic inflammatory demy-
nervous system. J Neuroimmunol 2003;137: demyelinating polyneuropathy patients. Ann elinating polyneuropathy: a double blind,
210-7. Neurol 2000;47:765-75. placebo controlled study. J Neurol Neuro-
60. Kieseier BC, Tani M, Mahad D, et al. 74. Yan WX, Archelos JJ, Hartung HP, Pol- surg Psychiatry 1993;56:36-9.
Chemokines and chemokine receptors in in- lard JD. P0 protein is a target antigen in 87. Mehndiratta MM, Hughes RA. Corti-
flammatory demyelinating neuropathies: a chronic inflammatory demyelinating poly- costeroids for chronic inflammatory demye-
central role for IP-10. Brain 2002;125:823-34. radiculoneuropathy. Ann Neurol 2001;50: linating polyradiculoneuropathy. Cochrane
61. Leppert D, Hughes P, Huber S, et al. Ma- 286-92. Database Syst Rev 2002;1:CD002062.
trix metalloproteinase upregulation in chron- 75. Melendez-Vasquez C, Redford J, Choud- 88. Sghirlanzoni A, Solari A, Ciano C, Mari-
ic inflammatory demyelinating polyneurop- hary PP, et al. Immunological investigation otti C, Fallica E, Pareyson D. Chronic inflam-
athy and nonsystemic vasculitic neuropathy. of chronic inflammatory demyelinating poly- matory demyelinating polyradiculoneurop-
Neurology 1999;53:62-70. radiculoneuropathy. J Neuroimmunol 1997; athy: long-term course and treatment of 60
62. Kieseier BC, Clements JM, Pischel HB, 73:124-34. patients. Neurol Sci 2000;21:31-7.
et al. Matrix metalloproteinases MMP-9 and 76. Takigawa T, Yasuda H, Terada M, et al. 89. McCrone P, Chisholm D, Knapp M, et al.
MMP-7 are expressed in experimental auto- The sera from GM1 ganglioside antibody Cost-utility analysis of intravenous immu-
immune neuritis and the Guillain-Barre syn- positive patients with Guillain-Barre syn- noglobulin and prednisolone for chronic in-
drome. Ann Neurol 1998;43:427-34. drome or chronic inflammatory demyelinat- flammatory demyelinating polyradiculoneu-
63. Mathey EK, Pollard JD, Armati PJ. TNF ing polyneuropathy block Na+ currents in ropathy. Eur J Neurol 2003;10:687-94.
alpha, IFN gamma and IL-2 mRNA expres- rat single myelinated nerve fibers. Intern 90. Hughes RA, Swan AV, van Doorn PA.
sion in CIDP sural nerve biopsies. J Neurol Med 2000;39:123-7. Cytotoxic drugs and interferons for chronic
Sci 1999;163:47-52. 77. Kwa MS, van Schaik IN, De Jonge RR, et inflammatory demyelinating polyradiculo-
64. Kiefer R, Kieseier BC, Stoll G, Hartung al. Autoimmunoreactivity to Schwann cells neuropathy. Cochrane Database Syst Rev
HP. The role of macrophages in immune- in patients with inflammatory neuropathies. 2003;1:CD003280.
mediated damage to the peripheral nervous Brain 2003;126:361-75. 91. Nobile-Orazio E, Meucci N, Baldini L,
system. Prog Neurobiol 2001;64:109-27. 78. Dyck PJ, Litchy WJ, Kratz KM, et al. Di Troia A, Scarlato G. Long-term prognosis
65. Hu W, Mathey E, Hartung HP, Kieseier A plasma exchange versus immune globulin of neuropathy associated with anti-MAG IgM
BC. Cyclo-oxygenases and prostaglandins infusion trial in chronic inflammatory de- M-proteins and its relationship to immune
in acute inflammatory demyelination of the myelinating polyradiculoneuropathy. Ann therapies. Brain 2000;123:710-7.
peripheral nerve. Neurology 2003;61:1774- Neurol 1994;36:838-45. 92. Gorson KC, Ropper AH, Weinberg DH,
9. 79. Hahn AF, Bolton CF, Pillay N, et al. Plas- Weinstein R. Treatment experience in pa-
66. Gold R, Hartung HP, Lassmann H. T-cell ma-exchange therapy in chronic inflamma- tients with anti-myelin-associated glyco-
apoptosis in autoimmune diseases: termi- tory demyelinating polyneuropathy: a dou- protein neuropathy. Muscle Nerve 2001;24:
nation of inflammation in the nervous sys- ble-blind, sham-controlled, cross-over study. 778-86.
tem and other sites with specialized immune- Brain 1996;119:1055-66. 93. Vallat JM, Hahn AF, Leger JM, et al. In-
defense mechanisms. Trends Neurosci 1997; 80. Hahn AF, Bolton CF, Zochodne D, Feas- terferon beta-1a as an investigational treat-
20:399-404. by TE. Intravenous immunoglobulin treat- ment for CIDP. Neurology 2003;60:Suppl 3:
67. Zettl UK, Gold R, Toyka KV, Hartung HP. ment in chronic inflammatory demyelinating S23-S28.
Intravenous glucocorticosteroid treatment polyneuropathy: a double-blind, placebo- 94. Kuntzer T, Radziwill AJ, Lettry-Trouillat
augments apoptosis of inflammatory T cells controlled, cross-over study. Brain 1996;119: R, et al. Interferon-beta1a in chronic inflam-
in experimental autoimmune neuritis (EAN) 1067-77. matory demyelinating polyneuropathy. Neu-
of the Lewis rat. J Neuropathol Exp Neurol 81. Mendell JR, Barohn RJ, Freimer ML, et al. rology 1999;53:1364-5.
1995;54:540-7. Randomized controlled trial of IVIg in un- 95. Harada H, Ohkoshi N, Fujita Y, Tamao-
68. Van Rhijn I, Van den Berg LH, Bosboom treated chronic inflammatory demyelinating ka A, Shoji S. Clinical improvement follow-
WM, Otten HG, Logtenberg T. Expression polyradiculoneuropathy. Neurology 2001; ing interferon-alpha alone as an initial treat-
of accessory molecules for T-cell activation 56:445-9. ment in CIDP. Muscle Nerve 2000;23:295-6.
in peripheral nerve of patients with CIDP 82. Dyck PJ, OBrien P, Swanson C, Low P, 96. Sabatelli M, Mignogna T, Lippi G, et al.
and vasculitic neuropathy. Brain 2000;123: Daube J. Combined azathioprine and pred- Interferon-alpha may benefit steroid unre-
2020-9. nisone in chronic inflammatory-demyelin- sponsive chronic inflammatory demyelinat-
69. Salomon B, Rhee L, Bour-Jordan H, et al. ating polyneuropathy. Neurology 1985;35: ing polyneuropathy. J Neurol Neurosurg
Development of spontaneous autoimmune 1173-6. Psychiatry 1995;58:638-9.
peripheral polyneuropathy in B7-2-deficient 83. Hadden RD, Sharrack B, Bensa S, Sou- 97. Gorson KC, Ropper AH, Clark BD, Dew
NOD mice. J Exp Med 2001;194:677-84. dain SE, Hughes RA. Randomized trial of RB III, Simovic D, Allam G. Treatment of
[Erratum, J Exp Med 2001;194:1393.] interferon beta-1a in chronic inflammatory chronic inflammatory demyelinating poly-
70. Andorfer B, Kieseier BC, Mathey E, et al. demyelinating polyradiculoneuropathy. Neu- neuropathy with interferon-alpha 2a. Neu-
Expression and distribution of transcription rology 1999;53:57-61. rology 1998;50:84-7.
factor NF-kB and inhibitor IkB in the in- 84. Dalakas MC. Mechanisms of action of 98. Anthoney DA, Bone I, Evans TR. In-
flamed peripheral nervous system. J Neuro- IVIg and therapeutic considerations in the flammatory demyelinating polyneuropathy:
immunol 2001;116:226-32. treatment of acute and chronic demyelinat- a complication of immunotherapy in malig-

n engl j med 352;13 www.nejm.org march 31, 2005 1355

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.
medical progress

nant melanoma. Ann Oncol 2000;11:1197- resolution of quadriplegic CIDP by com- py for chronic inflammatory demyelinating
200. bined plasmapheresis and IVIg. Neurology polyneuropathy. J Neurol Sci 2003;210:19-
99. Meriggioli MN, Rowin J. Chronic in- 2004;62:155-6. 21.
flammatory demyelinating polyneuropathy 106. Chaudhry V, Cornblath DR, Griffin JW, 113. Brannagan TH III, Pradhan A, Heiman-
after treatment with interferon-alpha. Mus- OBrien R, Drachman DB. Mycophenolate Patterson T, et al. High-dose cyclophospha-
cle Nerve 2000;23:433-5. mofetil: a safe and promising immunosup- mide without stem-cell rescue for refractory
100. Marzo ME, Tintore M, Fabregues O, pressant in neuromuscular diseases. Neurol- CIDP. Neurology 2002;58:1856-8.
Montalban X, Codina A. Chronic inflamma- ogy 2001;56:94-6. 114. Good JL, Chehrenama M, Mayer RF,
tory demyelinating polyneuropathy during 107. Umapathi T, Hughes R. Mycopheno- Koski CL. Pulse cyclophosphamide therapy
treatment with interferon-alpha. J Neurol late in treatment-resistant inflammatory neu- in chronic inflammatory demyelinating
Neurosurg Psychiatry 1998;65:604. ropathies. Eur J Neurol 2002;9:683-5. polyneuropathy. Neurology 1998;51:1735-
101. Pirko I, Kuntz NL, Patterson M, Kee- 108. Gorson KC, Amato AA, Ropper AH. 8.
gan BM, Weinshenker BG, Rodriguez M. Efficacy of mycophenolate mofetil in patients 115. Vermeulen M, Van Oers MH. Success-
Contrasting effects of IFNb and IVIG in chil- with chronic immune demyelinating poly- ful autologous stem cell transplantation in
dren with central and peripheral demyelin- neuropathy. Neurology 2004;63:715-7. a patient with chronic inflammatory demye-
ation. Neurology 2003;60:1697-9. 109. Matsuda M, Hoshi K, Gono T, Morita linating polyneuropathy. J Neurol Neurosurg
102. Mariette X, Brouet JC, Chevret S, et al. H, Ikeda S. Cyclosporin A in treatment of Psychiatry 2002;72:127-8.
A randomised double blind trial versus pla- refractory patients with chronic inflamma- 116. Benedetti L, Grandis M, Nobbio L, et al.
cebo does not confirm the benefit of alpha- tory demyelinating polyradiculoneuropa- Mycophenolate mofetil in dysimmune neu-
interferon in polyneuropathy associated with thy. J Neurol Sci 2004;224:29-35. ropathies: a preliminary study. Muscle Nerve
monoclonal IgM. J Neurol Neurosurg Psy- 110. Barnett MH, Pollard JD, Davies L, 2004;29:748-9.
chiatry 2000;69:279-80. McLeod JG. Cyclosporin A in resistant chron- 117. Pestronk A, Florence J, Miller T, Chok-
103. Pritchard J, Gray IA, Idrissova ZR, et al. ic inflammatory demyelinating polyradicu- si R, Al-Lozi MT, Levine TD. Treatment of
A randomized controlled trial of recombi- loneuropathy. Muscle Nerve 1998;21:454-60. IgM antibody associated polyneuropathies
nant interferon-beta 1a in Guillain-Barre 111. Mahattanakul W, Crawford TO, Griffin using rituximab. J Neurol Neurosurg Psychi-
syndrome. Neurology 2003;61:1282-4. JW, Goldstein JM, Cornblath DR. Treatment atry 2003;74:485-9.
104. Lisak RP. Type I interferons and chron- of chronic inflammatory demyelinating poly- 118. Renaud S, Gregor M, Fuhr P, et al. Rit-
ic inflammatory demyelinating polyneurop- neuropathy with cyclosporin-A. J Neurol Neu- uximab in the treatment of polyneuropathy
athy: treatment or cause? Muscle Nerve 2000; rosurg Psychiatry 1996;60:185-7. associated with anti-MAG antibodies. Mus-
23:307-9. 112. Chin RL, Sherman WH, Sander HW, cle Nerve 2003;27:611-5.
105. Walk D, Li LY, Parry GJ, Day JW. Rapid Hays AP, Latov N. Etanercept (Enbrel) thera- Copyright 2005 Massachusetts Medical Society.

powerpoint slides of journal figures and tables

At the Journals Web site, subscribers can automatically create PowerPoint slides
of Journal figures and tables. Click on a figure or table in the full-text version
of any article at www.nejm.org, and then click on PowerPoint Slide for Teaching.
A PowerPoint slide containing the image, with its title and reference citation,
can then be downloaded and saved.

1356 n engl j med 352;13 www.nejm.org march 31, 2005

Downloaded from www.nejm.org at UNIVERSITY OF CALGARY on February 5, 2009 .

Copyright 2005 Massachusetts Medical Society. All rights reserved.