Discussion

MFS represents ophthalmoplegia, ataxia and areflexia. MFS associated with central nervous signs including pyramidal features and/or reduced level of consciousness (drowsiness) represents BBE.2 Diagnosis of BBE is made in the context of these clinical features and antecedent infective illness.

Before the landmark discovery of anti-GQ1b antibody presence in patients with MFS in 1992, MFS and BBE had been considered separate clinical entities.3 The term ‘anti-GQ1b antibody syndrome’ adequately represents the common serological profile that is now known to underpin the clinical spectrum of both MFS and BBE.4

Our patient presented primarily with a progressive ataxia, upward gaze palsy and global areflexia. She displayed additional features of extensor plantar reflexes, confusion and hypersomnolence, indicative of central nervous system (CNS) involvement. In the context of positive anti-GQ1b antibodies and by ruling out more common pathologies, a clinical diagnosis of BBE was made.

The state and level of consciousness gave clues to the site of pathology. Level of consciousness is governed by the cortex, disorders of which give rise to an awake-encephalopathic picture, for example, delirium. State of consciousness is controlled in a more binary fashion by the reticular activating system in the brainstem, for example, awake or asleep. The clinical finding of when engaged in conversation, the patient was able to remain awake, but as soon as this cognitive stimulus was removed, the patient would enter sleep, suggested a brainstem pathology.5

Odaka et al reported 62 cases of BBE and quantified the prevalence of individual symptoms: 100% involved external ophthalmoplegia, 92% an antecedent illness, 60% limb weakness, 58% absent/reduced reflexes, 45% consciousness disturbance, 40% Babinski’s sign, 34% hyper-reflexia, 27% nystagmus and 8% normal reflexes, to isolate a few of the displayed symptoms.2 Our patient developed all of the aforementioned symptoms over a 2-week course.

Solid organ transplantation can result in multiple neurological complications including post-transplant autoimmune encephalitis. The causes of which are diverse and include medication toxicity, posterior reversible encephalopathy syndrome, viral infections and PTLD.6 Our patient’s PTLD episode was in remission but has been reported to cause CNS manifestations that commonly have radiological abnormalities, which were absent in our patient.7 8 A viral CNS infection was more plausible given our patient’s preceding episode of shingles.

HZ encephalomyelitis (HZE) is rare and represents CNS involvement after acute dissemination of reactivated HZ (shingles), following childhood varicella (chicken pox).9–11 Our patient had an unremarkable viral PCR and absence of lymphocytic pleocytosis, making HZE unlikely.12 Even though an aseptic HZE subtype exists, it was the combination of clinical features, antecedent HZ infection, positive anti-ganglioside antibodies and negative viral screen, which pushed the diagnosis towards BBE rather than HZE.13

A broader set of differentials included other immune-mediated neuropathies (acute inflammatory demyelinating polyneuropathies, AMAN, acute sensorimotor neuropathy) encephalitis, encephalomyelitis, ADEM, cerebrovascular infarct/ischaemia, myasthenia gravis and Wernicke’s encephalopathy.

A key investigation was CSF analysis, which was completely normal in our patient. In comparison with GBS, the frequency of protein-cytological dissociation has been shown to be lower than in MFS or BBE. Despite this, the extent of protein-cytological dissociation has been shown to develop over time, in isolated cases of BBE.2 Interestingly, not only are anti-ganglioside antibodies present in the acute phase sera of MFS and BBE, but many cases report the presence of anti-GQ1b antibodies in the CSF itself, confirming that the antibodies can access the CNS. It is postulated that this occurs in areas where the blood–brain barrier is most deficient, such as the brainstem.14

MRI is important when ruling out alternative aetiologies of brainstem encephalitis.15 16 Serial MRI studies have shown hyperintense signal changes on T2-weighted images in BBE, though MRI abnormality is still uncommon, reported in 11%–30% of BBE cases.2 4 Our case report showed a completely normal MRI with and without contrast of the brain and spinal cord.

The pathophysiology of the underlying neuropathy is subject to ongoing debate, and electrophysiology studies have been used to differentiate GBS, MFS and BBE. The nerve conduction study of our patient at 6 weeks post-discharge and 8 weeks post-symptom onset was unremarkable. Alberti et al found that among 12 patients with overlapping Fisher-Bickerstaff syndrome, the day 10 nerve conduction studies revealed reduced sensory nerve action potential (SNAP) amplitude but no signs of demyelination or other motor neuropathies. At days 15 and 18, all showed improvement or resolution in SNAP, in line with a reversible sensory conduction block.17 These findings have moved the anti-ganglioside-mediated neuropathies closer to the axonal forms of GBS, rather than the traditional demyelinating type. The node of Ranvier has also been found to be a site of dysfunction in BBE and has given rise to the term ‘nodo-paranodopathy’, which perhaps better describes the neuropathy in the anti-GQ1b syndrome.18

Ganglioside antibodies are abundant in neural tissue. When anti-ganglioside antibodies are produced via molecular mimicry due to antecedent pathogens, characteristic neurological phenotypes manifest.19 GQ1b forms 5%–6% of the total gangliosides in human peripheral nervous tissue, but a greater 11%–13% in the cranial nerves.20 The IgG anti-GQ1b has been shown to be present in 68% of BBE cases (n=53) and 83% of MFS cases (n=466), but typically the prevalence in patients with GBS is much lower.21 Antibody-positive BBE has been shown to display distinctive clinical features, with preceding infection and sensory disturbance being more frequent. However, CSF cell counts and abnormal MRI of the brain findings are more rare.22 The antibody titre has also been shown to closely follow the clinical course of GBS, MFS and BBE, rendering them pathogenic antibodies.20

Recurrence of disease is more common in the anti-GQ1b syndrome (MFS or BBE) than in GBS (14% and 4%, respectively).23 Ito et al reported a case in 2018 with recurrent MFS–GBS overlap and subsequently BBE–GBS overlap who was initially seronegative for anti-GQ1b. This patient later became positive for other anti-ganglioside antibodies with each successive recurrence, while remaining negative for anti-GQ1b.24 Conversely, Barbagallo et al reported a case where initially GQ1b-seronegative MFS–BBE overlap recurred 12 years later with positive anti-GQ1b antibodies.25 Moreover, recurrence has been associated with increased clinical severity and anti-ganglioside antibody pattern change. The latter term implies individual antibodies among the numerous anti-ganglioside antibodies, for example, anti-GQ1b, anti-GT1a and anti-GM1 becoming positive or negative on serial testing. It is possible that clinical severity and anti-ganglioside antibody patterning are linked.

Cases of BBE may be self-limiting, require intravenous immunoglobulin or plasma exchange. In refractory cases, anti-CD20 monoclonal antibodies in the form of rituximab have been used.14 A 2007 Cochrane review found no randomised data for the treatment of MFS and BBE but found it clinically logical to extrapolate the evidence in treatment of GBS to both MFS and BBE.26 Our patient completed her course of intravenous acyclovir for the initially presumed herpetic encephalomyelitis but did not receive any new immunotherapies during this episode and regardless, recovered fully.

Owing to its rarity, research into the mechanisms, efficacy of current treatments and role of antibody titres to prognosticate the course of illness is limited. This case was unique as BBE has behaved as a mimic of herpetic encephalomyelitis and has shone light on the future of predicting polyphasic disease and clinical severity.