Skip to main content
BMJ Case Reports logoLink to BMJ Case Reports
. 2020 Sep 18;13(9):e236545. doi: 10.1136/bcr-2020-236545

Bickerstaff’s brainstem encephalitis associated with anti-GM1 and anti-GD1a antibodies

Jonathan Cleaver 1,, Richard James 2, Gurjit Chohan 1, Paul Lyons 1
PMCID: PMC7511636  PMID: 32948528

Abstract

Bickerstaff’s brainstem encephalitis (BBE) is a Guillain-Barré syndrome (GBS) spectrum disorder associated with predominantly central nervous system predilection. Patients exhibit a variable constellation of depressed consciousness, bilateral external ophthalmoplegia, ataxia and long tract signs. Although the pathophysiology is not fully understood, it has been associated with anti-GQ1b antibodies in two-thirds of patients. We present a patient with clinical features consistent with BBE and positive anti-GM1 and anti-GD1a antibodies. A diagnostic approach to the acutely unwell patient with brainstem encephalitis is explored in this clinical context with a literature review of the aforementioned ganglioside antibody significance. Intravenous immunoglobulin therapy is highlighted in BBE using up-to-date evidence-based extrapolation from GBS.

Keywords: neurology, brain stem / cerebellum, cranial nerves, neuro ITU

Background

Bickerstaff’s brainstem encephalitis (BBE) is a Guillain-Barré syndrome (GBS) spectrum disorder. In 1951, BBE was described by Bickerstaff and Cloake in patients exhibiting ophthalmoplegia, ataxia and depressed consciousness.1 After 6 years, Bickerstaff expanded this condition in a case series aptly entitled ‘Brain-stem encephalitis: further observations on a grave syndrome with a benign prognosis’.2

BBE is typically associated with anti-GQ1b antibodies in two-thirds of patients. The largest retrospective clinical review of laboratory and clinical profiles of 62 patients with BBE revealed anti-GM1 and anti-GD1a antibodies present in 10% and 13% respectively.3

We highlight a case with a broad initial diagnostic differential of brainstem encephalitis, provide a practical approach to diagnosis and review the literature on BBE and intravenous immunoglobulin (IVIg) therapy. Ganglioside antibodies are important diagnostic corroborators as BBE may be a spectrum of parainfectious immunological diseases, which can include anti-GM1 and anti-GD1a antibodies.

Case presentation

A 39-year-old right-handed man was admitted, pre-COVID-19 pandemic, with a gradually progressive 3-day history of horizontal diplopia, unsteadiness on his feet and distal painful paraesthesia. He was previously fit and well with no medical history and took no regular medications. He denied alcohol excess, smoking and recreational drugs or supplements.

Closer questioning revealed a diarrhoeal illness 1 week prior to this presentation which resolved over several days. He denied unwell contacts, aberrant dietary intake and foreign travel. There were no features of meningism and he was keenly responsive to questions.

Initial examination revealed stable vital signs with no fever. Neurologically, his Glasgow Coma Score (GCS) was 15/15, he had a left internuclear ophthalmoplegia and impairment of right conjugate horizontal gaze palsy consistent with a one-and-a-half syndrome. His pupils were symmetrical and responsive to light. There was evidence of bilateral cerebellar dysfunction with intentions tremors, dysmetria, staccato speech and an ataxic gait with inability to tandem. Long tract signs were present manifesting with global hyperreflexia and positive Babinski’s bilaterally. Peripheral neurological, cardiorespiratory and abdominal examinations were unremarkable.

One day into his admission, he developed worsening diplopia and vertical nystagmus on upgaze progressing to complete ophthalmoparesis over the course of 48 hours.

His GCS dropped to 6 (M3 V2 E1) progressively over 72 hours with stertorous breathing requiring ventilatory support and admission to the intensive care unit (ITU). There was no evidence of preceding motor weakness and peripheral neurological examination continued to demonstrate the aforementioned long tract signs suggesting centrally mediated respiratory failure. He had evidence of autonomic dysregulation with tachycardia, hypertension (180mmHg systolic blood pressure) and urinary retention requiring catheterisation. He was given oral anti-hypertensives with adequate blood pressure control. He was afebrile throughout.

Investigations

Serum chemistries and haematology blood work showed no evidence of systemic inflammation, infection or metabolic disturbance. Admission chest X-ray was normal. Intracranial imaging including CT and MRI brain (including angiography and venography) was normal (figure 1). Cerebrospinal fluid (CSF) interrogation yielded a white cell count 11/mm3 (100% lymphocytes), red cells 6/mm3, protein 0.2 g/L, normal CSF glucose and paired oligoclonal bands with a single unpaired IgG oligoclonal band. CSF viral PCR was negative for herpes simplex virus-1 and virus-2, enterovirus and varicella zoster virus. Cytology revealed normal cells. An electroencephalogram was recorded pre-ITU when his conscious level was impaired to a GCS of 8 (M5 V2 E1). This revealed predominantly slow waves suggesting an underlying organic disturbance of cerebral activity. There was nothing to suggest non-convulsive status epilepticus.

Figure 1.

Figure 1

Bickerstaff’s encephalitis. (A) Axial fluid attenuated inversion recovery (FLAIR) MRI of the brain. (B) Axial T2 MRI of the brain. (C, D) Diffusion-weighted imaging (DWI) B1000 and apparent diffusion coefficient (ADC) map, respectively, obtained at the level of the mid-pons. No abnormal brainstem or cerebellar signal and no abnormal restricted diffusion. Normal MRI of the brain appearances.

Brainstem encephalitis of unclear aetiology was suspected and a broad panel of investigations were sent (table 1).

Table 1.

Inflammatory, autoimmune and atypical infection investigation summary

Blood CSF Imaging
ANA, ANCA, complements, anti-dsDNA, serum ACE, rheumatoid factor, lupus anticoagulant and anticardiolipin Viral PCR—HSV-1, HSV-2, enterovirus, varicella zoster virus CT of the chest, abdomen and pelvis
HIV, EBV, CMV, Hep B/C/E lyme and syphilis serology; meningococcal/pneumococcal PCR; cryptococcal antigen Whipple’s PCR MR of the brain and angiogram
Viral swab PCR for influenza A and B Bacterial 16S rDNA real-time PCR identification
CA19-9, beta-hcg, AFP NMDA receptor, LGI1, Caspr2, AMPA receptor, glycine receptor, IgLon5, GABAA receptor and GABAB receptor
Antineuronal antibodies including Hu, Ri, Yo, Ma2, CRMP5, amphiphysin Cytology
Antibodies against NMDA receptor, LGI1, Caspr2, AMPA receptor, glycine receptor, GABAA receptor, GABAB receptor and MOG
Anti-GM1 and anti-GD1a positive (the remaining ganglioside antibody panel including anti-GQ1b were negative)

All negative or normal, other than one result in bold.

AFP, alpha-fetoprotein; AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid; ANA, antinuclear antibody; ANCA, Antineutrophil cytoplasmic antibodies; Hep B/C/E, hepatitis B/C/E; beta-hcg, beta-human chorionic gonadotropin; Caspr2, contactin-associated protein-like 2; CMV, cytomegalovirus; CRMP5, collapsin response mediator protein 5; CSF, cerebrospinal fluid; dsDNA, double stranded DNA; EBV, Epstein-Barr virus; GABAA, gamma-aminobutyric acid A; GABAB, gamma-aminobutyric acid B; HSV, herpes simplex virus; LGI1, leucine-rich glioma-inactivated 1; MOG, myelin oligodendrocyte glycoprotein; NMDA, N-methyl-D-aspartate; PCR, polymerase chain reaction.

Differential diagnosis

In view of the clinical syndrome of bilateral long tract signs, disturbance of consciousness, eye movement disorder and global cerebellar dysfunction, an overarching syndrome consistent with a brainstem or rhombencephalitis was made.

The differential diagnosis for brainstem encephalitis is wide. Untreated infections can lead to devastating sequelae and the most common cause is listeria. This was less likely in view of the normal inflammatory markers, temperature and intracranial imaging with no clear consumption of unpasteurised products. Blood and CSF cultures, including listeria PCR were negative.

Rheumatological causes such as neuro-Behçet’s syndrome is the most common immune-related cause of brainstem encephalitis but there was no history of mouth or genital ulcers and no parenchymal hyperintensities or evidence of dural venous sinus thrombosis to fulfil the diagnostic criteria.

Botulism was also suspected with dysautonomia and descending paralysis, although mydriasis was absent and there was no history of nutritional risk factors, intravenous drug use or cuts. This was thought unlikely and he clinically improved despite the lack of anti-toxin. Other rarer infections, such as Whipple’s disease, are possible but there was no evidence of chronic diarrhoea, joint pains and CSF PCR was negative.

Thiamine deficiency can cause eye movement disorders and cerebellar features but there were no risk factors and the upper motor neuron features are not typical. Pabrinex was given nonetheless.

Finally, BBE was suspected as most likely in view of the brainstem syndrome with previous diarrhoeal illness, distal paraesthesia and long tract signs combined with depressed conscious level. His ganglioside antibodies revealed anti-GM1 and GD1a antibodies that have been found present in a cohort of these patients corroborating the diagnosis. Anti-GQ1b was negative. Neuroimaging is typically normal in patients with BBE. Importantly, there was no limb weakness and reflexes were brisk suggesting away from a progressive peripherally mediated GBS overlap.

Treatment

A diagnosis of BBE was made fulfilling diagnostic criteria3 (box 1) and he was treated with 2 g/kg IVIg over the course of 5 days.

Box 1. Bickerstaff’s brainstem encephalitis diagnostic criteria.

Core clinical features

Progressive external ophthalmoplegia and ataxia by 4 weeks and disturbance of consciousness or hyperreflexia

Exclusion of the following: vascular disease of the brainstem, Wernicke’s encephalopathy, botulism, myasthenia gravis, brainstem tumour, pituitary apoplexy and acute disseminated encephalomyelitis, multiple sclerosis, neuro-Behçet’s syndrome, vasculitis and lymphoma.

He was successfully extubated after 5 days and stepped down to the ward several days later. Autonomic features resolved expeditiously 1 week later. A fortnight after being stepped down from ITU, he could independently mobilise but with some unsteadiness and long tract signs disappeared. Complete resolution of cerebellar features (staccato speech, intention tremors, dysmetria and tandem unsteadiness) and diplopia was evident 12 weeks after admission.

Outcome and follow-up

He was discharged 1 month after his original admission with improving mobility and resolving diplopia. He was reviewed in clinic 2 months later with complete resolution in his symptoms and independence of activities of daily living, walking up to 4 miles per day.

Discussion

BBE is a rare diagnosis with an unclear incidence. A prospective population-based survey in Lombardy, Italy (population 8,891,652), found 138 patients diagnosed with a GBS spectrum disorder in 1996. Miller Fisher syndrome (MFS) was diagnosed in 1%–5% of these cases.4 BBE was not mentioned but anecdotally its incidence is less common.

The pathophysiology of BBE is discordant to other GBS spectrum disorders, exhibiting a central nervous system (CNS) predilection reflected in long tract signs, brainstem involvement and altered consciousness. CNS infiltration may occur through targeted permeability, generated by inflammation, of the blood–brain barrier selectively identified around the area postrema.5–7 This would allow larger molecules to pass, including anti-ganglioside antibodies, translating to pathology in the ascending reticular activating system and brainstem, generating this typical clinical constellation described.

Despite clinical features of a brainstem syndrome, MRI of the brain is only abnormal in ~30% of patients and is a useful diagnostic tool to exclude alternative causes of rhombencephalitis such as infections, inflammatory disorders or space occupying lesions that can masquerade in a similar fashion. However, autopsy findings have shown perivascular inflammatory changes within parts of the brainstem in patients with BBE.3

CSF studies have revealed fewer than half of patients with BBE exhibit albuminocytological dissociation at 2 weeks increasing to 57% at weeks 3 and 4.3 This is in contrast to GBS and MFS, with albuminocytological dissociation occurring in half of patients on initial CSF analysis increasing to 90% at disease nadir.3 Our patient demonstrated paired oligoclonal bands (suggesting systemic inflammation) with an isolated unpaired CSF IgG band. The latter was of doubtful alternative significance—although possibly reflecting a degree of intrathecal antibody synthesis—in view of a clear and expected improvement with IVIg in keeping with BBE. Furthermore, a normal whole-body CT, unremarkable CSF cytology and flow cytometry strengthened our decision not repeat CSF to look for its persistence.

Anti-ganglioside antibodies assist diagnosis and anti-GQ1b antibody has been associated in two-thirds of patients with BBE.3 Unlike our case, anti-GM1 and anti-GD1a antibodies are usually associated with significant limb weakness, acute motor axonal neuropathy (AMAN), chronic ataxic neuropathy or multifocal motor neuropathy. GM1 and GD1a gangliosides are typically found in the peripheral nerve roots, whereas GQ1b is concentrated in the paranodes and neuromuscular junction of the oculomotor, trochlear and abducens nerves.8 Adult mouse brain models demonstrate abundant GM1 gangliosides within most brainstem nuclei and white matter tracts, GD1a within the brainstem autonomic nuclei and both enriched in the thalamus.9 This could suggest a central pathophysiological role for anti-GM1 and anti-GD1a antibodies to induce ophthalmoparesis, disorders of consciousness, autonomic dysregulation and long tract signs demonstrated in our patient. Anti-GM1 and anti-GD1a antibodies were present in 11% and 13% of respective patients with BBE in a retrospective clinical review of 62 patients3 and other reports have demonstrated anti-GD1a and anti-GM1 antibody associated ophthalmoparesis.10 11 Furthermore, molecular mimicry with anti-GM1 antibodies and campylobacter jejuni lipopolysaccharides has been discovered in rodents raising the significance of the prodromal diarrhoeal trigger highlighted in our case.12

Practically, these antibodies take several weeks to come back and therefore the combination of clinical signs, normal MRI of the brain and a history of preceding infective illness heightens suspicion of BBE. We recommend concomitantly empirically covering for CNS infection, which can be devastating if left untreated in a time-critical brainstem syndrome. Additional antimicrobial listeria cover should be considered with brainstem encephalitis, which occurs in immunocompetent people more frequently than listeria meningitis. Important differential diagnoses should be excluded due to the possibility of more malignant disease patterns (figure 2—brainstem encephalitis approach).

Figure 2.

Figure 2

Approach to the differential diagnosis of brainstem encephalitis. *Cocaine can erode the surrounding skull base and sinuses creating a pathway for all infections from the nasal cavity to the brainstem. ADEM, acute disseminated encephalomyelitis; BBE, Bickerstaff’s brainstem encephalitis; CMV, cytomegalovirus; CNS, central nervous system; EBV, Epstein-Barr virus; EV-71, enterovirus-71; Gly, glycine; HSV, herpses simplex virus; IgLon5, immunoglobulin-like cell adhesion molecule 5; NBS, neurobehçet’s syndrome; NMDA, N-methyl-D-aspartate; NMOSD, neuromyelitis optica spectrum disorder; SLE, systemic lupus erythematosus; TB, tuberculosis.

Treatment is with IVIg 2 g/kg in total over 2–5 days and based on results from GBS data. A recent large multicentre international prospective cohort study of 1165 patients with GBS followed-up over 6 months revealed a second course of IVIg within 2 weeks was associated with more severe disability.13 However, this was confounded by the influence of investigators to administer a second course in the more severely disabled patients with GBS. Despite using a multivariable ordinal model to correct for this there was no observed benefit for a second course of treatment in the lower GBS disability scoring patients at 26 weeks (OR 1.26, 95% CI 0.35 to 4.60). There is currently a large randomised controlled trial (Second IVIg dose-GBS trial) underway to review this further in the more severe patients with GBS.14

We avoided a second course of IVIg despite an initial delayed improvement on this basis and opted not for additional plasma exchange (PEX), which is known to remove IVIg from the body within a 2-week timeframe. Beyond 2 weeks, nerve damage is putatively complete, highlighting a narrow therapeutic time window to decide additional treatment. Our patient showed significant improvement within this interval from monotherapy and had a significantly raised serum immunoglobulin G, which is a positive prognostic indicator following IVIg treatment.15

Where available, PEX may be favoured in septic patients over IVIg or due to its therapeutic action at physically removing ganglioside antibodies which are thought of as central pathogenic mediators.16 Currently, no randomised controlled trials have been conducted to explore treatment in BBE and only level 4 quality of evidence exists. Due to the clinical severity and time-critical nature of the illness either IVIg or PEX may be reasonable.

There is evidence that IVIg can be useful with infectious encephalitides due to its immunomodulatory actions supporting early treatment with IVIg in suspected BBE cases without complete exclusion of subtle infective chameleons. This has been demonstrated in patients with flavivirus encephalitis including West Nile and Japanese encephalitis who have been treated with IVIg due to its specific antibody neutralising and anti-inflammatory properties.17 The immunoglobulin in the treatment of encephalitis (IgNITE) trial is currently underway assessing the outcome of encephalitis of unclear aetiology.18

Overall, BBE is likely a continuum from GBS, MFS, AMAN and other variants with a heterogenous B-cell-mediated pathophysiology. Our patient fit Bickerstaff’s original description of the ‘grave syndrome with a benign prognosis’ with associated anti-GM1 and anti-GD1a antibodies responding completely over months following IVIg.

Learning points.

  • Bickerstaff’s brainstem encephalitis (BBE) clinical features include a varied combination of ophthalmoparesis, reduced level of consciousness, ataxia and hyperreflexia, with or without limb weakness and usually following a prodromal diarrhoeal illness.

  • To send antiganglioside antibodies prior to treatment with IVIg which can support the diagnosis.

  • Recognise some overlap with other antiganglioside antibodies such as anti-GM1 and anti-GD1a that may be found in BBE.

  • Treatment with IVIg or plasma exchange can expedite recovery, reduce ventilation time and duration of intensive care unit stay.

  • Consider covering for central nervous system infection, including listeria, if unable to clinically rule out due to possible devastating sequelae.

Footnotes

Contributors: JC, GC and PL contributed to the manuscript rationale and patient management. JC created the first draft. GC, PL and RJ revised the manuscript. RJ contributed to manuscript images and descriptions.

Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests: None declared.

Patient consent for publication: Obtained.

Provenance and peer review: Not commissioned; externally peer reviewed.

References

  • 1.Bickerstaff ER, Cloake PCP. Mesencephalitis and rhombencephalitis. Br Med J 1951;2:77–81. 10.1136/bmj.2.4723.77 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Bickerstaff ER. Brain-stem encephalitis; further observations on a Grave syndrome with benign prognosis. Br Med J 1957;1:1384–7. 10.1136/bmj.1.5032.1384 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Odaka M, Yuki N, Yamada M, et al. Bickerstaff's brainstem encephalitis: clinical features of 62 cases and a subgroup associated with Guillain-Barré syndrome. Brain 2003;126:2279–90. 10.1093/brain/awg233 [DOI] [PubMed] [Google Scholar]
  • 4.Bogliun G, Beghi E, Italian GBS Registry Study Group . Incidence and clinical features of acute inflammatory polyradiculoneuropathy in Lombardy, Italy, 1996. Acta Neurol Scand 2004;110:100–6. 10.1111/j.1600-0404.2004.00272.x [DOI] [PubMed] [Google Scholar]
  • 5.Shahrizaila N, Yuki N. Bickerstaff brainstem encephalitis and Fisher syndrome: anti-GQ1b antibody syndrome. J Neurol Neurosurg Psychiatry 2013;84:576–83. 10.1136/jnnp-2012-302824 [DOI] [PubMed] [Google Scholar]
  • 6.Faraci FM, Choi J, Baumbach GL, et al. Microcirculation of the area postrema. Permeability and vascular responses. Circ Res 1989;65:417–25. 10.1161/01.RES.65.2.417 [DOI] [PubMed] [Google Scholar]
  • 7.van Breemen VL, Clemente CD. Silver deposition in the central nervous system and the hematoencephalic barrier studied with the electron microscope. J Biophys Biochem Cytol 1955;1:161–6. 10.1083/jcb.1.2.161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chiba A, Kusunoki S, Obata H, et al. Serum anti-GQ1b IgG antibody is associated with ophthalmoplegia in Miller Fisher syndrome and Guillain-Barré syndrome: clinical and immunohistochemical studies. Neurology 1993;43:1911–7. 10.1212/WNL.43.10.1911 [DOI] [PubMed] [Google Scholar]
  • 9.Vajn K, Viljetić B, Degmečić IV, et al. Differential distribution of major brain gangliosides in the adult mouse central nervous system. PLoS One 2013;8:e75720. 10.1371/journal.pone.0075720 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Go T. Partial oculomotor nerve palsy associated with elevated anti-galactocerebroside and anti-GM1 antibodies. J Pediatr 2000;137:425–6. 10.1067/mpd.2000.107797 [DOI] [PubMed] [Google Scholar]
  • 11.Fusco C, Bertani G, Scarano A, et al. Acute ophthalmoparesis associated with anti-GM1, anti-GD1a, and anti-GD1b antibodies after enterovirus infection in a 6-year-old girl. J Child Neurol 2007;22:432–4. 10.1177/0883073807301921 [DOI] [PubMed] [Google Scholar]
  • 12.Yuki N, Susuki K, Koga M, et al. Carbohydrate mimicry between human ganglioside GM1 and Campylobacter jejuni lipooligosaccharide causes Guillain-Barre syndrome. Proc Natl Acad Sci U S A 2004;101:11404–9. 10.1073/pnas.0402391101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Verboon C, van den Berg B, Cornblath DR, et al. Original research: second IVIg course in Guillain-Barré syndrome with poor prognosis: the non-randomised ISID study. J Neurol Neurosurg Psychiatry 2020;91:113–21. 10.1136/jnnp-2019-321496 [DOI] [PubMed] [Google Scholar]
  • 14.Walgaard C, Jacobs BC, Lingsma HF, et al. Second IVIg course in Guillain-Barré syndrome patients with poor prognosis (SID-GBS trial): protocol for a double-blind randomized, placebo-controlled clinical trial. J Peripher Nerv Syst 2018;23:210–5. 10.1111/jns.12286 [DOI] [PubMed] [Google Scholar]
  • 15.Kuitwaard K, de Gelder J, Tio-Gillen AP, et al. Pharmacokinetics of intravenous immunoglobulin and outcome in Guillain-Barré syndrome. Ann Neurol 2009;66:597–603. 10.1002/ana.21737 [DOI] [PubMed] [Google Scholar]
  • 16.Willison HJ. Gangliosides as targets for autoimmune injury to the nervous system. J Neurochem 2007;103(Suppl 1):143–9. 10.1111/j.1471-4159.2007.04718.x [DOI] [PubMed] [Google Scholar]
  • 17.Rayamajhi A, Nightingale S, Bhatta NK, et al. Correction: a preliminary randomized double blind placebo-controlled trial of intravenous immunoglobulin for Japanese encephalitis in Nepal. PLoS One 2015;10:e0136008. 10.1371/journal.pone.0136008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Iro MA, Sadarangani M, Absoud M, et al. Immunoglobulin in the treatment of encephalitis (IgNiTE): protocol for a multicentre randomised controlled trial. BMJ Open 2016;6:e012356. 10.1136/bmjopen-2016-012356 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from BMJ Case Reports are provided here courtesy of BMJ Publishing Group

RESOURCES