Guillain–Barré syndrome mimics

The Brighton Collaboration criteria have standardized the clinical and laboratory‐supported diagnosis of Guillain–Barré syndrome (GBS) and Miller Fisher syndrome (MFS) in a way that is applicable in many parts of the world with variable resources. The caveat within the criteria, “absence of an identified alternative diagnosis for weakness” makes GBS a diagnosis of exclusion. Accurate diagnosis of GBS requires a good understanding of an updated, locally contextualised list of mimics, and features that distinguish them from GBS.


| INTRODUC TI ON
The diagnosis of Guillain-Barré syndrome (GBS), the leading global cause of acute flaccid paralysis (Yuki & Hartung, 2012), is largely based on clinical features, and supported by serological, electrodiagnostic, and immunological investigations. However, the manifestations of GBS are protean and include limited forms such as acute ophthalmoparesis (Wakerley, Uncini, & Yuki, 2014) and pharyngealcervical-brachial variant (Wakerley & Yuki, 2013). Central nervous system signs such as drowsiness, hyper-reflexia, and Babinski reflex can rarely be present, indicating the related Bickerstaff encephalitis (Odaka et al., 2003). Hence, a good appreciation of common GBS mimics is necessary for the accurate diagnosis of GBS; as articulated by the Brighton Collaboration criteria, "absence of an identified alternative diagnosis for weakness" (Sejvar et al., 2011). The list of disorders that is commonly mistaken for GBS varies with the setting.
The aim of this study, performed at National Neuroscience Institute, Singapore, was to describe the clinical features and investigations of patients who were initially diagnosed as GBS but subsequent evaluation led to the diagnosis of an unrelated disorder that mimicked GBS.
In our institution, a tertiary neurological institute, GBS patients have been prospectively and continuously data-based since early 2010. Of a total hundred and forty-one patients, twelve turned out to have a different diagnosis during follow-up (

| Patient 1
A 27-year-old woman presented with rapidly progressive tetraparesis over 3 days. She had ophthalmoplegia, upbeating nystagmus, bifacial weakness, bulbar palsy, flaccid tetraparesis, areflexia, and distal sensory loss. She was diagnosed as GBS overlapping with MFS. Two days later, she was encephalopathic. Nerve conduction study (NCS) showed severe, diffuse, acute-on-chronic, axonal sensorimotor polyneuropathy. Magnetic resonance imaging (MRI) brain showed restricted diffusion in both caudate heads, and T2 hyperintensity in the periaqueductal area and bilateral dorsomedial thalamus. Cerebrospinal fluid (CSF) analysis was normal. A history of chronic alcoholism was elicited. She was diagnosed as Wernicke's encephalopathy with beriberi neuropathy. She recovered fully with intravenous thiamine treatment.

| Patient 6
A 54-year-old man had pain, numbness, and weakness in limbs over 2 days. He had asymmetrical flaccid weakness in limbs with reduced

| Patient 9
A 57-year-old man with stage four diffuse large B-cell lym-

| Patient 11
A 51-year-old man presented with difficulty walking for 2 weeks that was preceded by upper respiratory tract infection. He had symmetrical tetraparesis and reduction in reflexes. Cranial nerves were normal. CSF showed raised WBC count (9 cells/μl), raised protein (>3.00 g/L) and normal glucose. NCS showed demyelinating polyneuropathy. He was diagnosed with GBS, and IVIg was administered. A month later, he was readmitted with progressive tetraparesis, areflexia, and distal sensory loss. His diagnosis was revised to acute-onset chronic inflammatory demyelinating polyneuropathy (A-CIDP). His condition eventually improved with corticosteroid and mycophenolate mofetil.

| Patient 12
A 53-year-old woman presented with vertiginous giddiness and unsteady gait for 2 weeks. She was treated with steroid prior to coming to our center. Clinical examination showed complex ophthalmoparesis, and bilateral lower motor neuron facial weakness. There was mild left upper limb dysmetria and she could not tandem walk. She had normal power, reflexes, and sensation. She was diagnosed with GBS overlapping with MFS. NCS showed only absent bilateral blink reflex.
CSF showed elevated protein (0.60 g/L), no pleocytosis, and normal glucose. However, MRI brain showed nonenhancing signal abnormalities in the facial colliculi region of pons, periventricular white matter, and subcortical white matter. Final diagnosis was clinically isolated syndrome of brain stem demyelination. Three months later, she only had mild residual left facial weakness.

| Patient 13
A 37-year-old man with systemic lupus erythematosus (SLE), presented with acute-onset of severe loss of sensation associated with areflexia in all limbs and inability to walk. He was diagnosed with GBS and treated with IVIg. CSF showed normal WBC count (3 cells/ μl) and raised protein (1.01 g/L). NCS within 1-week from the onset was normal but subsequent study in the second week detected nonlength-dependent reduction in sensory potentials. His mobility improved over few weeks but continued to have sensory loss and persistent postural hypotension. Anti-Ro (SSA) was weakly positive.
He was finally diagnosed as subacute sensory neuronopathy from underlying Sjogrens-SLE. With escalation of immunosuppression, he gradually improved but he remains disabled by severe sensory ataxia and orthostatic hypotension.

| Patient 14
A

| Patient 15
A Cytoalbuminologic dissociation in spinal fluid examined about 1 week after symptoms is useful in the diagnosis of GBS (Wong et al., 2015). However, the Brighton criteria use a CSF total white cell count of ˂50 cells/μl, considerably above the cutoffs used in many centers and originally proposed by Asbury and Cornblath (1990). Thirteen of our fifteen GBS mimics patients had CSF examination (performed at a median of 8 days). Five (Patients 4, 10, 11, 12, and 13) fulfilled criteria for cytoalbuminologic dissociation in concordance with the Brighton criteria. Using our laboratory normal value of less than 6 cells/μl, only three patients had cytoalbuminologic dissociation (Patients 4, 12, and 13). Spinal fluid examination performed at an appropriate time, and using stricter criteria than that proposed in Brighton criteria can help the accuracy of GBS diagnosis.
The spectrum of disorders that resembles GBS evolves with changes in the practice of medicine, patient demographics, and disease epidemiology. Local setting should also be considered. For instance, we expect acute beriberi neuropathy to be less frequent in countries where alcohol consumption is low for cultural-religious reasons. In our center, improvements in access to laboratory investigations and neuroimaging have removed previously common differential diagnoses of GBS such as acute spinal shock and hypokalemic periodic paralysis.
In conclusion, accurate diagnosis of GBS requires a good understanding of an updated, locally contextualised list of mimics, and features that distinguish them from GBS.