Professor of Clinical Neurology, Weill Cornell Medical College
SYNOPSIS: Chronic inflammatory demyelinating polyradiculoneuropathy is commonly misdiagnosed. It is important to adhere to established diagnostic criteria and to regularly re-evaluate all patients given this diagnosis.
SOURCE: Broers MC, et al. Misdiagnosis and diagnostic pitfalls of chronic inflammatory demyelinating polyradiculoneuropathy. Eur J Neurol 2021; Mar 3. doi: 10.1111/ene.14796. [Online ahead of print].
With a prevalence of five cases per 100,000, and an incidence that increases with advancing age, chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a male-predominant, motor-predominant, symmetric neuropathy with no clearly identified predisposing risk factors and no genetic predisposition. Characterized by a progressive or relapsing-remitting course and symmetric proximal and distal muscle weakness, asymmetric and sensory forms also are recognized. Although thought to be immune mediated, the cause remains unknown. More than 15 sets of diagnostic criteria have been outlined, underscoring the challenge of correct diagnosis. Misdiagnosis is not uncommon. What factors underlie an erroneous diagnosis of CIDP? How can this be avoided?
Clinical, electrodiagnostic, laboratory, and treatment data were retrospectively collected on all patients referred to Erasmus University Medical Center Rotterdam, with a referral or final diagnosis of CIDP, between April 2011 and March 2017. Recognized as a Center of Excellence for Guillain-Barré syndrome (GBS) and CIDP, this tertiary academic hospital keeps an internal CIDP database, which was reviewed to ensure capture of all eligible patients. Data included electrodiagnostic studies, serum paraprotein and immunoglobulin M (IgM) antibody to myelin-associated glycoprotein (anti-MAG Ab), cerebrospinal fluid (CSF) analyses, magnetic resonance imaging (MRI) of spinal roots and lumbosacral plexus, nerve ultrasound, and nerve biopsy.
European Federation of Neurological Societies/Peripheral Nerve Society (EFNS/PNS) 2010 diagnostic criteria for CIDP were used to define patients as “definite,” “probable,” or “possible” CIDP, and all patients were classified as CIDP confirmed, CIDP underdiagnosed, or CIDP overdiagnosed. Statistical analysis comprised the Chi-square or Fisher’s exact test and the Mann-Whitney U test, with P values < 0.05 considered significant.
Among 122 patients referred with either a diagnosis of CIDP or a different diagnosis revised to CIDP, 10 were excluded for various reasons, including lack of diagnostic consensus, leaving 112 for analysis. Among 96 patients with a referral diagnosis of CIDP, the diagnosis was confirmed in 65 (68%), but revised in 31 (32% overdiagnosis), whereas among 81 patients ultimately diagnosed with CIDP at Erasmus University, 16 were referred with an alternative diagnosis (20% underdiagnosed). Age and gender were similar in all groups, but misdiagnosis was significantly more common when referral came from a non-teaching hospital.
Among the 31 overdiagnosed patients, almost all (n = 30) were diagnosed with another form of neuropathy and, compared to the CIDP-confirmed group, they were more likely to have asymmetric muscle weakness (20% vs. 5%) or purely distal weakness (29% vs. 11%), and less likely to have proximal weakness (52% vs. 86%). Failure to fulfill electrodiagnostic criteria for CIDP was found in 65%, and, although all overdiagnosed patients had neuropathy other than CIDP, 74% showed elevated CSF protein.
Among the underdiagnosed, 50% had clinically atypical CIDP, including predominantly distal or asymmetric phenotypes, all had proximal muscle weakness when seen at Erasmus, all fulfilled CIDP electrodiagnostic criteria, and 25% had elevated CSF protein.
Errors in CIDP diagnosis are common in both directions and may be explained by a variety of factors, including unawareness that proximal muscle weakness is characteristic of CIDP, unawareness of atypical CIDP phenotypes, erroneous interpretation of electrodiagnostic studies, nonadherence to accepted CIDP electrodiagnostic criteria, overreliance on CSF protein levels, and failure to exclude other causes of polyneuropathy.
Even among patients satisfying EFNS/PNS 2010 diagnostic criteria for CIDP, misdiagnosis may occur. Among seven such patients reported in a case series of CIDP mimics, five ultimately were diagnosed with POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, skin changes), and one each with neurolymphomatosis or chronic ataxic neuropathy ophthalmoplegia IgM paraprotein cold agglutinins disialosyl antibodies (CANOMAD). Neuropathic pain and leg swelling (n = 6) or significant weight loss (n = 4) were common, all had a monoclonal protein, most had elevated levels of vascular endothelial growth factor (VEGF), and some responded to steroids or intravenous immunoglobulin.
Be not fooled. Systemic symptoms and pain should raise suspicion for CIDP mimics. Patients should be reassessed and re-evaluated regularly for alternative diagnoses.