Authors

B.A. Jacobs, M.D., Ph.D.
P.A. van Doorn, M.D., Ph.D.

Clinical features

In 1956, Dr. Charles Miller Fisher described three patients with the clinical triad of ophthalmoplegia, ataxia and areflexia without prominent signs of peripheral neuropathy.¹ After this publication, patients presenting with this clinical triad were referred to as having the Miller Fisher syndrome (MFS).^2,3 The initial symptom is usually diplopia and/or limb and gait ataxia.² The full clinical picture of ataxia, areflexia and ophthalmoplegia usually develops within 5 to 10 days.² The diplopia usually advances over several days and may result in a complete ophthalmoplegia (including pupil areflexia) or a pure external ophthalmoplegia (with normal pupil reflexes). Most patients also have a unilateral or bilateral ptosis. Ataxia in MFS is usually characterized as cerebellar, although cerebellar dysarhria, vertigo and nystagmus are not part of the syndrome. The reflexes may be preserved in the initial phase of disease, but most patients develop thereafter develop symmetrical areflexia or hyporeflexia, which may persist for longer periods. Patients frequently show only part of the classic clinical triad, although they otherwise fulfill the diagnostic criteria and demonstrate the typical monophasic disease course. To diagnose MFS presumably not all three of the clinical triad have to be present.

In addition to the clinical triad, MFS patients frequently have involvement of cranial nerves other than the oculomotor nerves.^2-4 The facial nerve is most frequently affected in half of patients, followed by lower cranial nerves, but other cranial nerves may also be involved.^3,4 In addition, many patients have mild sensory complaints, usually paresthesias and dysethesia of all four extremities. Severe sensory deficits however are rare. Extended weakness of neck, shoulder and arm musculature may also occur (cervical brachial variant). Some patients further develop a mild tetraparesis and approximately one-third of patients will progress to the Guillain-Barré syndrome (GBS) with more profound weakness and/or respiratory failure.³ The clinical features are summarized in Table 1. Most patients with MFS report features of an initially uncomplicated respiratory or gastro-intestinal infection in the weeks preceding the onset of neurological deficits.^2,3

MFS is considered to be a variant of GBS because they share the presence of (1) symmetric weakness and areflexia essential for the diagnosis of GBS,⁵ (2) increased protein levels in absence of increased cells in cerebrospinal fluid, (3) usual monophasic course of disease, (5) antecedent infection, (6) anti-ganglioside antibodies in serum and (7) because of the presence of cross-over cases in which MFS patients progress to GBS.

Natural course and prognosis

Most patients with MFS show a monophasic and benign course of disease leading to complete remission without residual deficits.^1-3 Most patients spontaneously start to improve within 2 to 4 weeks after onset of neurological symptoms. The recovery usually is completed after weeks to months with a mean recovery time of 10 weeks.² However, part of the patients will develop bulbar weakness with swallowing disturbances needing intubation or progress to GBS with severe tetraparesis or respiratory insufficiency.^4,9 In these patients specific treatment may be considered. Relapses of MFS are rare, but may occur with disease-free intervals of several years.²   

Pathogenesis and etiology
The research on the pathogenesis of MFS is characterized by several scientific breakthroughs in the last decade.6 It had been demonstrated that serum and monoclonal antibodies to GQ1b bind to human peripheral nerves, predominantly oculomotor nerves.10 In the mouse phrenic nerve/diaphragm preparation these antibodies induce a neuromuscular transmission block and other electrophysiological effects similar to a-latrotoxin, the toxin used by the black widow spider to paralyse its prey.11,12 The a-latrotoxin-like effects of anti-GQ1b antibodies are mediated by complement activation which also lead to damaging of the nerve terminal architecture and of perisynaptic Schwann cells.11-14 All these ex vivo effects can be prevented by intravenous immunoglobulin preparations.15 Antibodies to GQ1b are most likely induced during the infection preceding the onset of neurological symptoms.6 Campylobacter jejuni isolates from MFS patients have been identified which contain lipo-oligosaccharides showing molecular mimicry with GQ1b and other disialosyl gangliosides. Antibodies raised to lipo-oligosaccharides during C. jejuni infection therefore can cross-react to GQ1b and other gangliosides in peripheral nerves. Eradication of the bacteria will lead to a decrease of antibody titres and to clinical improvement, explaining the monophasic course of MFS. Although MFS is a rare disease, these scientific developments made MFS a model disorder for investigating the pathogenesis of post-infectious, antibody-mediated neurological disorders. 

Epidemiology
MFS is a rare disease. The incidence is not exactly known and may vary between  geographic areas. Based on a survey study performed in the South-West of the Netherlands the incidence of MFS is estimated to be about 1-2 per million per year. Approximately one-third of MFS patients will progress to GBS and about 5% of GBS patients started as a MFS. Males are two times more frequently affected than females.² MFS may affect people of all ages, including children.²

Differential diagnosis

The diagnosis is based on the clinical symptoms and the presence of serum antibodies to the ganglioside GQ1b (and GT1a). Other causes of these clinical features should be excluded. The differential diagnosis of MFS is given in Table 1.

Table 1. Differential diagnosis of Miller Fisher syndrome.

Other diseases related to serum antibodies to GQ1b and other gangliosides

• Subacute ataxia (without ophthalmoplegia)
• Subacute ophthalmoplegia (without ataxia)
• Guillain-Barré syndrome with ophthalmoplegia
• Lower bulbar variant of Guillain-Barré syndrome
• Chronic ataxic neuropathy with ophthalmoplegia, M-protein, (cold-) agglutinins and disialosyl ganglioside antibodies (CANOMAD)
• Bickerstaff encephalitis

Other diseases not related to serum antibodies to GQ1b

• Brainstem lesions (ischemic, hemorrhagic, multiple sclerosis, vasculitis, encephalitis)
• Neuromuscular transmission disorders (myasthenia gravis, antibiotic-induced)
• Meningitis lymphomatosa and carcinomatosa
• Metabolic disorders (hypermagnesiemia, hypophosphatemia) 
• Infections (neurolues, neuroborrelia, botulism, diphtheria, tetanus)
• Intoxications (organofosfates, lead, thallium, arsenicum, lithium, hexocarbon)
• Deficiencies (thiamine)

Ancillary investigations

Serum antibodies to the ganglioside GQ1b (see links)
Serum in the acute stage of disease (before treatment) in more than 90% of MFS patients contains antibodies to GQ1b (and GT1a) which can be determined by ELISA.⁶ Anti-GQ1b antibodies in serum from MFS patients are usually of the IgG isotype, but IgM and IgA to GQ1b may also be present. These antibodies usually disappear within weeks with clinical improvement. Antibodies to other disialosyl gangliosides, including to GT1a, GD3, GD1b and GT1ba may also be detected in lesser frequencies. Anti-GQ1b IgG antibodies may also be found in patients with acute ataxia (without ophthalmoplegia), acute ophthalmoplegia (without ataxia), GBS with ophthalmoplegia, lower bulbar variant of GBS and Bickerstaff encephalitis (Table 2). Anti-GQ1b IgM antibodies may be found in patients with chronic paraproteinanemic sensory neuropathy, including the chronic ataxic neuropathy with ophthalmoplegia, M-protein, (cold-) agglutinins and disialosyl ganglioside antibodies (CANOMAD). The presence of antibodies to GQ1b (and GT1a) has a high sensitivity and specificity for MFS and overlapping syndromes. Demonstrating these antibodies may therefore help in diagnosing problematic cases. These antibodies less frequently are also present in cerebrospinal fluid (CSF) of MFS patients, usually in much lower titres than in serum. Testing of CSF for the presence of these antibodies, therefore, has no additional diagnostic value. 

Magnetic resonance imaging (MRI)
MRI is mainly performed to exclude brain stem disorders and meningitis carcinomatosa or lymphomatosa. In most cases of MFS the MRI results are normal. Some case studies have been published in which brainstem lesions on MRI were reported (usually hypointense on T1 weighted and hyperintense on T2 weighted images and with (slight) enhancement on T1-weigted imaging after gadolinium administration).⁷  

Cerebrospinal fluid (CSF)
In MFS patients CSF may (but not necessarily) contain increased protein levels in absence of cells. Examination of CSF is mainly performed to exclude other diseases (including meningitis carcinomatosa and lymphomatosa). 

Nerve conduction studies
The predominant neurophysiological abnormality present in the limbs of most patients with MFS is amplitude reduction or absence of sensory nerve action potentials.⁸ In addition, some patients may have neurophysiological features of mild demyelination or axonal degeneration of motor nerves.^2,3,8 Some patients may have mild-to-moderate reduction of facial compound muscle action potential amplitude and the blink reflex may be abnormal.⁸

Further laboratory studies
Mainly performed to exclude infectious, metabolic and toxic causes of clinical symptoms.

Genetics
MFS does not occur in families. Polymorphisms in immune response genes may be involved in the susceptibility to develop a cross-reactive antibodies after infection leading to MFS. No genetic susceptibility genes have been identified yet, but are currently investigated.

Therapy
Supportive treatment usually is sufficient in patients with mild MFS, considering the monophasic and benign course of disease. Patients with more severe MFS, developing swallowing disturbances or progression to GBS, may benefit from specific treatment although no randomised controlled trials have been performed to demonstrate treatment efficacy. Based on the parallels with GBS and on the results of case studies in MFS patients, treatment with intravenous immunoglobulin or plasma exchange may be effective in these patients. The effect of the combination of intravenous immunoglobulin and methylprednisolone is unknown. Treatment with methylprednisolone alone has not been studied in MFS, but the results in GBS indicate that it is not effective.

References

1. Miller Fisher. An unusual variant of acute idiopathic polyneuritis (syndrome of ophthalmoplegia, ataxia and areflexia). N Engl J Med 1956;255:57-65.
2. Berlit P, Rakicky J. The Miller Fisher syndrome. J Clin Neuro-opthalmology 1992;12:57-63.
3. Ropper AH, Wijdicks EFM, Truax BT. Guillain-Barré syndrome. Conemporary Neurology Series. F.A. Davis Company, 1991.
4. ter Bruggen JP, van der Meché FGA, de Jager AE, Polman CH. Ophthalmoplegic and lower cranial nerve variants merge into each other and into classical Guillain-Barré syndrome. Muscle Nerve 1998;21:239-242.
5. Asbury AK, Cornblath DR. Assessment of diagnostic criteria for Guillain-Barré syndrome. Ann Neurol 1990(suppl);27:21-24.
6. Willison HJ, Yuki N. Peripheral neuropathies and anti-glycolipid antibodies. Brain 2002;125:2591-2625.
7. Petty RKH, Duncan R, Jamal GA, et al. Brainstem encephalitis and the Miller Fisher syndrome. J Neurol Neurosurcg Psychiatry 1993;56:201-203.
8. Fross RD, Daube JR. Neuropathy in the Miller Fisher syndrome: clinical and electrophysiological findings. Neurology 1987:37:1493-1498.
9. Blau I, Casson I, Lieberman A, Weiss E. The not-so-benign Miller Fisher syndrome: a variant of the Guillain-Barré syndrome. Arch Neurol 1980;37:384-385.
10. 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;43:1911-1917.
11. Plomp JJ, Molenaar PC, O'Hanlon GM, et al. Miller Fisher anti-GQ1b antibodies: alpha-latrotoxin-like effects on motor end plates. Ann Neurol 1999;45:189-199.
12. Jacobs BC, Bullens RW, O'Hanlon GM, et al. Detection and prevalence of alpha-latrotoxin-like effects of serum from patients with Guillain-Barré syndrome. Muscle Nerve 2002;25:549-558.
13. O'Hanlon GM, Plomp JJ, Chakrabarti M, et al. Anti-GQ1b ganglioside antibodies mediate complement-dependent destruction of the motor nerve terminal. Brain 2001;124:893-906.
14. Halstead SK, O'Hanlon GM, Humphreys PD, et al. Anti-disialoside antibodies kill perisynaptic Schwann cells and damage motor nerve terminals via membrane attack complex in a murine model of neuropathy. Brain 2004;127:2109-2123.
15. Jacobs BC, O'Hanlon GM, Bullens RW, et al. Immunoglobulins inhibit pathophysiological effects of anti-GQ1b-positive sera at motor nerve terminals through inhibition of antibody binding. Brain 2003;126:2220-2234.