Article Text

Download PDFPDF

Recommendations on diagnostic strategies for chronic inflammatory demyelinating polyradiculoneuropathy
  1. The French CIDP Study Group
  1. Dr J-M Vallat, Department of Neurology, University Hospital, 2 Av. Martin Luther King, 87042 Limoges Cedex, France; jean-michel.vallat{at}


Background: Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is an immune mediated treatable peripheral neuropathy, the diagnosis of which is straightforward in more than half of cases. Numerous sets of electrophysiological criteria have been published. However, in some cases, electrophysiological data are not sufficient and patients that may benefit from treatment escape accurate diagnosis.

Objective: To describe a step by step diagnostic procedure for neurologists facing a peripheral neuropathy of undetermined cause, to help make an accurate diagnosis of CIDP.

Methods: A group of French experts was established, neurologists and neurophysiologists being recruited on the basis of personal experience with patients suffering from CIDP and also on publications in the field. A full literature review was conducted on the topic of diagnostic criteria and procedures for the diagnosis of CIDP, and meetings were scheduled to reach a consensus on the best diagnostic workup in different clinical situations.

Results: Six meetings were conducted and a consensus was reached, based on the available literature and experience in the management of such patients. Discussions resulted in defining five clinical situations in which a diagnosis of CIDP may be considered, and procedures were detailed in each case, including the location of nerve biopsy and use of non-conventional electrophysiological testing and imaging procedures.

Conclusion: The guidelines in the diagnostic procedure reported here result from a consensus of French experts in the field of peripheral neuropathy and allow a diagnosis of CIDP to be made in the most frequently encountered situations. These recommendations may be of value for physicians as they rely on the rational use of available techniques in typical clinical situations.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Box 1 Clinical signs suggestive of chronic inflammatory demyelinating polyradiculoneuropathy

  • Sensorimotor deficit in all four limbs

  • Proximal motor deficit (strongly suggestive)

  • Generalised areflexia

  • Initial sensory symptoms affecting the upper limbs

  • Associated involvement of one or more cranial nerves

  • Course with relapses (strongly suggestive)

Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is an immune mediated disorder of peripheral nerves. Its diagnosis is based mainly on clinical and electrophysiological features.15 Accurate diagnosis is important as effective treatments are available.610 The spectrum of CIDP appears to be relatively heterogeneous, and many sets of diagnostic criteria have been proposed that are based largely on electrophysiological findings.1116 Since the first consensus criteria proposed by the American Academy of Neurology in 1991,11 numerous authors have proposed alternative ones.1720 These recent sets of criteria differ in some points, which may influence their specificity and sensitivity. However, as there is no specific “gold standard” test for diagnosing CIDP, the issue of sensitivity, specificity and predictive value of electrophysiological criteria has been addressed by comparing patients with a diagnosis of CIDP based on expert opinion (relapsing course, response to treatment) with patients suffering from other peripheral nerve disorders such as distal diabetic polyneuropathy, amyotrophic lateral sclerosis and other types of polyneuropathy. The last set of criteria to date was established by international experts on neuromuscular disorders and is largely based on literature review and compilation of previous criteria by experts on neuromuscular disorders.12

Therefore, it is not the aim of the present paper to address the issue of electrophysiological parameters that may help differentiate CIDP from other types of peripheral neuropathy, nor to discuss the respective sensitivity, specificity or predictive value of each individual test that may be used to confirm a diagnosis of CIDP. Rather, we sought to provide general guidelines to be used in the different situations where a diagnosis of CIDP may be considered. The following recommendations are the result of cooperation between French neurologists, particularly involved in the management of patients with peripheral neuropathy. Our emphasis is on ease of use in clinical practice and no data are presented here.

In the first part of this paper, the clinical, electrophysiological, biological and histopathological signs underlying the diagnosis of CIDP are presented. In the second part, we propose a strategy for the main diagnostic situations encountered in CIDP. We have chosen to address the problem of chronic progressive polyneuropathy, as the relapsing forms of the disease with occasional acute attacks resembling Guillain–Barré syndrome are usually easier to diagnose, although long term treatment remains problematic in these patients. Moreover, we decided not to consider demyelinating neuropathy associated with paraprotein, especially with anti-MAG antibody, as this syndrome may be considered a separate entity, as recently outlined.12


The French CIDP Study Group comprised French neurologists and neurophysiologists that were chosen based on experience in the diagnosis and management of patients with peripheral neuropathy and CIDP in particular, and also on publications on this topic. A full literature Medline search (1985–2005) was conducted on the topic of “CIDP”, “polyneuropathy”, “diagnosis”, “criteria” and “guidelines”. As the literature search missed a lot of relevant papers, authors were free to use their personal files as well. Subgroups of authors were established with the task of working on the clinical, biological, electrophysiological and histopathological features. An additional subgroup worked on the particular topic of diagnostic criteria.

After the subgroups had collected the available data, six meetings were scheduled to review data and propose a diagnostic strategy in typical clinical situations where a diagnosis of CIDP is highly suspected, as well as in atypical cases.


After collection and review of the data on the diagnostic procedure of CIDP, authors were asked to make proposals on the best diagnostic procedure based on available data and clinical experience. Each proposal was iteratively reviewed by the other members to reach a consensus. Authors agreed on clinical, biological, electrophysiological and histopathological features of CIDP. However, as patients with this condition may have different presentations, five clinical situations were outlined that correspond to the most frequently encountered situations where a diagnosis of CIDP may be envisaged.

Clinical signs

In the common form of CIDP, the diagnosis must be suspected in patients with deterioration in symptoms over at least 2 months, involving the upper and lower limbs, including motor deficit with a proximal and/or distal distribution along with a predominantly proprioceptive and/or superficial sensory deficit. The cranial nerves may be involved. Tendon reflexes are abolished or diminished in all four limbs (box 1).3 11

The clinical course of CIDP may present in two main forms: relapsing and progressive. However, CIDP is heterogeneous and the classical relapsing form involves less than 50% of cases.2123

There are several clinical variants (box 2).

Box 2 Clinical variants of chronic inflammatory demyelinating polyradiculoneuropathy

  • Distal motor deficit in lower limbs

  • Asymmetric sensorimotor deficit, with predominant upper limb involvement (Lewis–Sumner syndrome)

  • Pure sensory deficit

  • Pure motor deficit

  • Topographic: the deficit may be mainly distal,24 or asymmetric with a multifocal distribution as in the Lewis–Sumner syndrome.2527

  • Sensory: the clinical presentation may be limited to distal paresthesia and proprioceptive ataxia for walking.2831

  • Purely motor: the sensory signs are slight or absent.32

  • Paucisymptomatic, with only fatigability and/or some distal paresthesia.33


An electroneuromyographic (ENMG) study is decisive for confirming the diagnosis, and an electrophysiological examination must follow rigorous procedures. The aim of an ENMG study is to identify nerve conduction abnormalities indicative of demyelination, mostly in the motor nerves. Prolonged motor distal latency, reduced nerve conduction velocity and prolonged F wave latency are suggestive of the diagnosis. We emphasise the particular value of conduction block and temporal dispersion, which are highly suggestive of segmental demyelination.12 15 Distinction between true conduction block and temporal dispersion is of little interest in CIDP as both are indicative of demyelination. Consensus criteria for conduction block are widely accepted and may be used as a reference.34 Although values of nerve conduction velocities that indicate demyelination may differ between studies, we recommend referring to recent consensus guidelines.12 In some instances, non-conventional techniques such as somatosensory evoked potentials or high voltage motor stimulation may be of particular value, especially in cases with proximal lesions. Standard techniques are described in appendix 1 (available online).

Other investigations

CSF examination

Lumbar puncture is helpful. A raised protein level without cells (<10 cells/mm3) is generally found, but the diagnosis is not excluded by normal CSF.

Laboratory investigations

There is no specific biological indicator for CIDP. A range of investigations is recommended to identify pathologies associated with CIDP (diabetes,35 36 monoclonal gammopathy,3739 HIV infection,40 cancer,41 hepatitis B and C,42 43 sarcoidosis,44 sicca syndrome,45 mixed connective tissue disease46 and systemic lupus47) or to eliminate other causes of demyelinating polyneuropathy, such as Charcot–Marie–Tooth type 1 or Refsum’s disease.48 There are currently no recognised standards for biological investigations.

When an IgM monoclonal gammopathy is found, anti-MAG antibodies should be assayed (this polyneuropathy presents with a predominantly distal sensory and motor demyelination).49 50 Disialylosyl antibodies should be sought if the polyneuropathy is associated with an oculomotor disorder, which is suggestive of a CANOMAD (chronic ataxic neuropathy, ophthalmoplegia, IgM paraprotein, cold agglutinins and disialosyl antibodies).51

Radiological investigations

They are not systematically performed but dictated by the clinical and biological features (see below).

Nerve biopsy (see appendix 2 online)

Indications for nerve biopsy are discussed in the next section.

Diagnostic strategy

The diagnostic strategy is largely based on clinical history, examination and ENMG. However, there are some situations where it is not possible to confirm the diagnosis of CIDP after this first step. These include: demyelinating abnormalities in a proximal location on the sensory or motor roots,52 more pronounced lesions in the sensory fibres, marked secondary or associated axonal loss masking the presence of demyelinating features and conduction block.

The diagnostic strategy thus depends on the particular situation, and other investigations such as nerve biopsy may be required.

Five different situations can be outlined.

Situation A

The clinical presentation fits the classical CIDP. The electrophysiological criteria are present. Raised protein concentration demonstrated by CSF examination may be supportive but is not mandatory. Nerve biopsy is of no use in clinical practice.

Situation B

The clinical presentation does not fit with classical CIDP but with a clinical variant. Electrophysiological examination can confirm the diagnosis of CIDP according to standard criteria. CSF examination is useful if it demonstrates an elevated protein level. Nerve biopsy is of no use.

Situation C

The clinical presentation is suggestive of or compatible with CIDP. The initial electrophysiological study is also suggestive of a demyelinating polyneuropathy, but the course or the context of the neuropathy is not typical and response to treatment is poor.

Presentations include:

  • Systemic symptoms (marked weight loss, superficial or deep lymph node enlargement, skin lesions).

  • Rapid and severe initial axonal loss: marked amyotrophy, marked motor or sensory amplitude reduction, numerous fibrillation potentials at rest.

We strongly advise investigating for an associated disorder such as lymphoma,53 sarcoidosis or neoplasm. A monoclonal IgG or light chain (especially lambda) may indicate the presence of a POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy and skin changes) syndrome.54 55 CSF examination is highly recommended to look for abnormal cells.

Examination of a nerve biopsy specimen using standard techniques is recommended if no diagnosis can be made. Amyloid deposits or malignant cell infiltrates may be found. Recognition of other specific lesions (ie, endoneurial deposits of immunoglobulin, uncompacted myelin lamellae) may require the use of other techniques such as electron microscopy or immunohistochemistry.

Situation D

The clinical presentation is suggestive of classical CIDP although the electrophysiological criteria are not present after studying eight motor nerves (insufficient slowing of nerve conduction velocities, unexcitable nerves…). It is suggested to look for additional elements that may help to confirm the diagnosis of CIDP:

  • High CSF protein concentration.

  • Suggestive alterations in sensory pathways: more pronounced alterations in the upper limbs, highly reduced sensory nerve conduction velocity, marked sensory loss contrasting with normal sensory potentials and/or proximal alterations on somatosensory evoked potentials.

  • MRI enhancement or hypertrophy of plexus or roots.56 57

If one or more of the above abnormalities is present, the diagnosis of CIDP is highly probable. However, a nerve biopsy may be discussed on a case by case basis (see appendix 2 online). 5860

Situation E

This is the most difficult situation as the clinical presentation is not of a classical CIDP. The electrophysiological findings are not indicative of a demyelinating disorder. Investigations designed to identify the cause of this type of chronic polyneuropathy are largely negative and the clinical and electrophysiological features are not typical of a chronic idiopathic axonal polyneuropathy.

It is therefore important to look for clinical clues:

  • Young age at onset

  • Relapses

  • Rapid involvement of upper limbs or of proximal segments

  • More prominent motor than sensory symptoms

  • Face and/or trunk involvement

  • Generalised areflexia

  • Marked ataxia

In the ENMG study, it is important to look for:

  • Reduction in motor nerve conduction in one or two nerves, over and above that indicated by the axonal loss.

  • Normal distal compound muscle action potential in a weak muscle.

  • Sensory nerve action potentials more altered in upper than in lower limbs.

  • Highly reduced sensory conduction velocity.

  • Marked sensory complaint and normal sensory nerve action potentials.

If one or more of these clinical or electrophysiological items are present, the diagnosis of CIDP may be suspected. The same schedule of investigations as proposed in situation D may be envisaged.


CIDP is an immune mediated polyneuropathy that may become extremely disabling in some instances. It is considered a rare61 but perhaps underestimated60 condition. As patients with CIDP may improve with immunosuppressant or immunomodulatory treatments, it is important to recognise the disease. Numerous sets of criteria have been proposed but the diagnostic strategy may differ from one patient to another, according to clinical presentation, associated symptoms or electrophysiological findings.

The present guidelines are the result of a consensus between French neurologists involved in the treatment and follow-up of patients with peripheral neuropathy. They are based mainly on experience and literature knowledge. Therefore, these guidelines are designed to help physicians in detecting patients with CIDP based on practical situations that are encountered in clinical practice. We have attempted to define the main situations where a diagnosis of CIDP can be suspected, and to provide colleagues with the appropriate strategy to confirm this diagnosis. These guidelines describe the first steps of the diagnostic strategy, but clinicians may be aware that in each individual situation, response to immunomodulatory treatment may be decisive in confirming the diagnosis of CIDP.

Physicians and patients must be aware that these guidelines represent the authors’ views on the adequate strategies to diagnose CIDP, and that they are not intended to be used by lawyers or insurance companies to obtain any benefit or compensation.


We wish to thank SP Jarman for translation and N Couade for typing the manuscript.



  • ▸ Appendices 1 and 2 are published online only at

  • Competing interests: None.

  • The French CIDP Study Group: JC Antoine (Neurology Department, Saint-Etienne), JP Azulay (Neurology Department, Marseille), P Bouche (Neurophysiology Department, Paris), A Créange (Neurology Department, Créteil), E Fournier (Neurophysiology Department, Paris), G Gallouedec (Neurology Department, Limoges), A Lagueny (Neurology Department, Bordeaux), JP Lefaucheur (Neurophysiology Department, Créteil), JM Léger (Neurology Department, Paris), L Magy (Neurology Department, Limoges), T Maisonobe (Neurophysiology Department, Paris), G Nicolas (Neurology Department, Angers), J Pouget (Neurology Department, Marseille), P Soichot (Neurology Department, Dijon), T Stojkovic (Neurology Department, Lille), JM Vallat (Neurology Department, Limoges), A Verschueren (Neurology Department, Marseille), C Vial (Neurophysiology Department, Lyon), K Viala (Neurophysiology Department, Paris).