Diagnostic Challenges in Chronic Inflammatory Demyelinating Polyradiculoneuropathy

Filip Eftimov; Ilse M. Lucke; Luis A. Querol; Yusuf A. Rajabally; Camiel Verhamme


Brain. 2020;143(11):3214-3224. 

In This Article


CIDP has a very heterogeneous presentation, and consists of a spectrum of autoimmune diseases of the peripheral nerves based on presumed breach of tolerance leading to autoimmunity against nerve antigens. Different pathophysiological mechanisms have been identified, often sharing clinical features, which makes diagnosis challenging.

The term CIDP was introduced in 1982 describing some of the most common features of the disease in four different domains, namely time of onset, pathophysiology, tissue component involved and anatomical distribution. However, the use of this term has evolved over time to try to incorporate other patients with primary chronic immune-mediated neuropathies susceptible to disease-modifying therapies that share but do not necessarily include all of the original features. More importantly, as our knowledge of the disease deepens, it has become increasingly difficult to unite all clinical presentations and pathophysiological mechanisms under this term, some even contradicting the original terminology. First, CIDP may have an acute onset, that initially resembles Guillain-Barré syndrome (Vural et al., 2018). Also, some neuropathies meet various supportive criteria, including treatment response, but not the electrophysiological criteria (Lucke et al., 2019a; Herraets et al., 2020). However, the best examples are the recently discovered nodal and paranodal neuropathies that are caused by autoantibodies. Some are autoantibodies to axonal nodal structures, such as CNTN1, so that it became clear that autoimmunity can be primarily directed to axonal antigens, and not only to myelin antigens. In case of anti-CNTN1, electron microscopy examination showed a selective loss of the septate-like junctions at the paranodes and a detachment of the paranodal myelin loops from the axon, but there was an absence of inflammation and classical macrophage-induced demyelination (Kouton et al., 2020). Importantly, this disorganization of the nodes of Ranvier disrupts the saltatory conduction, which leads to electrophysiological findings that are traditionally interpreted as demyelinating features. The electrophysiological criteria for CIDP were originally intended to be able to detect the physiological substrate of demyelination, as this was the only known underlying pathology. However, in essence they are markers for functional disruption or slowing of the saltatory conduction of myelinated axons. In the context of CIDP, conduction block can result from paranodal abnormalities of the myelin sheath, but also from primary dysfunction of the axon at the nodes of Ranvier. Considering that inflammation and demyelination as core features are not present in all CIDP patients, while an autoimmune aetiology is presumed in all patients, the umbrella term 'chronic autoimmune neuropathies' may fit better (Figure 1). Further discussion will be needed on whether to include other entities such as multifocal motor neuropathy and anti-MAG neuropathy within this term. Recent studies suggested that anti-MAG antibodies can be present in few patients with CIDP without the presence of IgM paraproteinaemia. However, in an overwhelming majority of patients there is also IgM paraproteinaemia, suggesting that the underlying pathomechanism is plasma cell dyscrasia rather than breach of tolerance. Moreover, further studies are needed to determine the diagnostic cut-off of anti-MAG antibodies in those patients without IgM paraproteins as these can also be found in low quantities in other conditions.

Figure 1.

A conceptual framework for a diagnostic work-up in chronic autoimmune neuropathies. A conceptual framework for a diagnostic work-up in chronic auto-immune neuropathies, assuming future emphasis on immunological tests with high specificity to show evidence for autoimmunity. Combinations and number of tests required for diagnosis depend on specificity of clinical phenotypes, of immunological tests and of supportive tests, such as nerve conduction studies, imaging, CSF examination, pathology and response to treatment. NCS currently have the best diagnostic accuracy.

Regardless of the nomenclature, if we consider the autoimmune aetiology as the hallmark of the disease, in the future we should also probably focus more on introducing specific immunological tests in our diagnostic work-up. This would be in line with the diagnostic approaches of other, similarly heterogeneous, autoimmune diseases, in which access to tissue is difficult, such as autoimmune encephalopathies (Graus et al., 2016). Unfortunately, despite recent progress, finding proof of autoimmunity is currently difficult as antibodies in blood/CSF are only found in a minority of patients, while unambiguous pathology results from nerve biopsy supporting an autoimmune origin are uncommon. However, other autoantibody reactivities or markers of autoimmunity, including immunity-related genes, may be identified in the future (Nevo et al., 2013; Staudt et al., 2017).

Despite desirable future advances to prove autoimmunity in more patients, the diagnosis of CIDP currently remains a clinical one, with greater complexity and concurrently increased uncertainty for atypical versus typical forms. The absence of a golden standard challenges proper evaluation of diagnostic accuracy of tests. Still, taking surrogate standards, such as fulfilment of consensus criteria and treatment response, partially overcomes this problem. Recognizing the technical difficulties and caveats as outlined in this review, will hopefully reduce the number of patients with misdiagnosis and underdiagnosis, particularly in non-expert environments. Currently, in the majority of patients, the diagnosis is supported by electrophysiological evidence of impaired saltatory conduction, so that NCS still are the most reliable and widely available diagnostic test to support the diagnosis of CIDP. A slightly elevated CSF protein level has a poor specificity, while normal protein levels do not exclude the diagnosis. If imaging is required, nerve ultrasound is probably preferred as it is quicker, easier and has a higher diagnostic accuracy compared to MRI, but further multicentre studies are needed. The role of evoked potentials and nerve biopsy is very limited in most patients and should be reserved for selected cases. Finally, progression should be made in standardized detection of currently available autoantibodies, of new autoantibodies and of other evidence for autoimmunity to improve diagnostic accuracy, so that these tests may become more useful in standard clinical practice. With the last update of the guidelines almost 10 years ago and the introduction of the nerve ultrasound and the discovery of autoantibodies in CIDP, the update of the EAN/PNS diagnostic guidelines is much awaited.