Guillain–Barré syndrome
Guillain, Barré syndrome is rare, striking only 1 or 2 people in every 100,000 each year, yet it can turn tingling in the hands and feet into paralysis. The weakness can build over a matter of hours, or it can take a few weeks. At the center of it is a strange act of confusion. The immune system, built to protect the body, begins damaging the peripheral nerves that carry commands out to the muscles. Why would the body turn on its own wiring? What sets that mistake in motion? How do doctors recognize a disease that has no single decisive test? And what happens to the people who live through it?
The first symptoms are numbness, tingling and pain, alone or in combination. Weakness follows in the legs and arms, affecting both sides equally and worsening over time. It can take half a day to more than two weeks to reach its worst. In one person in five, it keeps advancing for as long as four weeks.
About half of patients have trouble with the cranial nerves, which serve the head and face. That can mean a weak face, difficulty swallowing and sometimes weak eye muscles. In 8% of people, only the legs are affected. About a third of all patients can still walk.
After the progression stops comes the plateau phase, when weakness holds steady before recovery starts. It can last from two days to six months, though a week is the most common length. More than half of patients also have pain, including back pain, muscle pain, odd skin sensations and head and neck pain from irritation of the lining of the brain.
On examination, doctors find weak muscles and tendon reflexes that are reduced or absent. A small proportion have normal reflexes at first, and some even have exaggerated ones. The level of consciousness is normally unaffected.
For a child younger than six, the picture is harder to read. Pain and trouble walking can be mistaken for a viral infection or a bone and joint problem, sometimes for up to two weeks.
Then there is the danger to breathing. A quarter of all patients develop weakness in the breathing muscles, which can lead to respiratory failure. Among those who need artificial ventilation, 60% face further problems such as pneumonia, severe infections, blood clots in the lungs and bleeding in the digestive tract.
Doctors try to anticipate that crisis with two breathing tests. A forced vital capacity below 15 mL per kilogram of body weight marks severe failure. So does a negative inspiratory force weaker than 60 cmH2O.
The autonomic nervous system, which runs involuntary functions like heart rate and blood pressure, is affected in two-thirds of patients. In 20%, blood pressure swings sharply and the heartbeat becomes irregular. Sometimes the heart stops, and a pacemaker is needed. Sweating and the reactions of the pupils can also change, and this can happen even in people whose muscles are not severely weak.
Two-thirds of people with the syndrome had an infection before it began. Most often it was gastroenteritis or a respiratory infection, with a sore throat, cough, diarrhea or bloating appearing 3-6 weeks before the neurological symptoms.
In many cases the culprit can be identified. Approximately 30% of cases follow Campylobacter jejuni, a bacterium that causes diarrhea. Another 10% follow cytomegalovirus. Yet very few infected people develop the syndrome, at 0.25-0.65 per 1000 Campylobacter episodes and 0.6-2.2 per 1000 cytomegalovirus episodes.
The particular strain of Campylobacter may matter. Different forms carry different lipopolysaccharides on their surface, and some induce the illness while others do not.
Other links are less certain. Epstein, Barr virus, varicella zoster virus and the bacterium Mycoplasma pneumoniae have all been associated with it. It is known to occur after influenza. The tropical infections dengue fever and Zika virus have also been associated with episodes, and previous hepatitis E infection is more common in patients.
The 1976 swine flu outbreak produced the most famous vaccine episode. Among 45 million people vaccinated, 362 developed the syndrome in the following 6 weeks, an 8.8-fold increase over normal rates. That works out to 8.8 cases per million recipients. It was an outlier, and later campaigns showed only small increases.
The 2009 pandemic vaccine against H1N1 did not cause a significant rise. Studies found about 1 extra case per million vaccinated, similar to seasonal influenza vaccines. Natural influenza infection is a stronger risk factor than the vaccine, so vaccination lowers the overall risk by lowering the chance of catching influenza.
In the United States, the syndrome after seasonal influenza vaccination appears on the federal government's vaccine injury table. On the 24th of March 2021, the FDA required safety label changes for the zoster vaccine Shingrix, made by GlaxoSmithKline. Observational studies had shown an increased risk of the syndrome following vaccination.
COVID-19 added new questions. The syndrome has been reported with the infection itself and as a very rare side effect of the Janssen and Oxford, AstraZeneca vaccines. The European Medicines Agency issued a warning. Early reports put the vaccine incidence below that after infection. More recent studies found no measurable link with COVID-19 infection, while first doses of the AstraZeneca and Janssen vaccines still correlated positively.
One drug has also been blamed. Zimelidine, an antidepressant with a very favorable safety profile, was withdrawn from the market after rare case reports of the syndrome.
Each nerve cell in the peripheral nervous system has its body, called the soma, in the spinal cord. A long projection called the axon carries electrical impulses to the neuromuscular junction, where the signal passes to the muscle. Schwann cells wrap the axon in a sheath containing myelin, with small gaps between them called nodes of Ranvier.
The type of immune attack depends on the variant. In the demyelinating form, white blood cells called T lymphocytes and macrophages damage the myelin sheath. A group of blood proteins known as complement is activated first. In the axonal form, IgG antibodies and complement attack the membrane covering the axon itself, without direct involvement of lymphocytes.
Many antibodies aimed at nerve cells have been reported. In the axonal subtype they bind to gangliosides, a group of substances found in peripheral nerves. Four have been central: GM1, GD1a, GT1a and GQ1b. Different antibodies go with different features, and GQ1b antibodies are linked with the Miller Fisher variant and related forms.
The likely explanation is molecular mimicry. The immune system reacts to substances on a microbe, but the resulting antibodies also react with substances that occur naturally in the body. After a Campylobacter infection, the body makes antibodies of the IgA class. Only a small proportion of people also make IgG antibodies against bacterial cell wall substances, and these cross-react with human nerve gangliosides.
A puzzle remains. Central tolerance is meant to suppress antibodies against the body's own substances, and it is not currently known how this process is escaped here.
Not every antiganglioside antibody causes disease. Some may bind more than one type of epitope at once, called heterodimeric binding, and that may determine the response. One further suggestion is that poor injection technique could injure the axillary nerves near the deltoid muscle. Nerve cells taking up and translating the vaccine might then provoke an autoimmune response.
Even the bowel may play a part. The development of harmful antibodies may depend on the presence of other strains of bacteria there.
A diagnosis rests on rapid muscle paralysis, absent reflexes, no fever and no likely alternative cause. Blood tests help rule out other explanations, such as a low potassium level. Low sodium is often found in the syndrome, which has been attributed to inappropriate secretion of antidiuretic hormone leading to water retention.
The lumbar puncture, also called a spinal tap, removes a small amount of cerebrospinal fluid with a needle between the lumbar vertebrae. The classic result is a protein level above 0.55 g/L with fewer than 10 white blood cells per cubic millimeter. This pattern is called albuminocytological dissociation. It separates the syndrome from lymphoma and poliomyelitis, where both protein and cell count are raised.
Timing matters. Elevated protein appears in approximately 50% of patients in the first 3 days after weakness begins, rising to 80% after the first week. Repeating the puncture is not recommended, because protein levels may rise after treatment.
Magnetic resonance imaging of the spinal cord is often done to rule out conditions such as spinal cord compression. Enhancement of the nerve roots may point to the syndrome. In children it appears in 95% of scans, but it is not specific, so other confirmation is needed.
Nerve conduction studies and needle electromyography can exclude other causes of sudden weakness and distinguish the types. In the first two weeks they may show nothing abnormal, and they are not required for the diagnosis. Testing for antiganglioside antibodies is often done, though its contribution is usually limited.
Formal criteria exist for the main subtypes, but they can misclassify some cases, particularly where conduction failure is reversible. Changes have been proposed, and repeated testing is sometimes helpful.
In Europe and the United States, 60-80% of patients have the demyelinating subtype, known as AIDP. The axonal subtype AMAN affects only 6-7% there. In Asia and Central and South America, AMAN accounts for 30-65% and is sometimes called "Chinese paralytic syndrome." The gap may reflect exposure to different infections, as well as the genetic characteristics of those populations.
AMAN causes isolated muscle weakness without sensory loss. Its cousin AMSAN brings similar severe weakness along with sensory loss. A pharyngeal-cervical-brachial variant weakens the throat, face, neck and shoulder muscles.
The Miller Fisher variant causes weak eye muscles, poor coordination and absent reflexes, usually without limb weakness. It is more common in men than women at a ratio of 2 to 1, cases typically occur in spring, and the average age is 43. It is thought to be more common in Southeast Asia.
Bickerstaff's brainstem encephalitis is now regarded as part of the same family. It brings rapid onset of eye muscle paralysis, poor coordination and disturbed consciousness. The course is usually a single episode, though recurrent episodes have been reported, and brainstem abnormalities on MRI appear in 11%. A related condition, acute ataxic hypersomnolence, involves coordination problems and drowsiness with no detectable muscle weakness.
Many patients have overlapping symptoms, so classification can be difficult. Whether isolated acute sensory loss counts as a form of the syndrome is disputed.
Age and sex also shape risk. The relative risk rises by 20% for every decade of life, so the elderly are more likely to be affected than children and young adults. The relative risk for men is 1.78 compared to women.
Two immunotherapies dominate care: plasmapheresis and intravenous immunoglobulins, known as IVIG. Plasmapheresis filters antibodies out of the bloodstream. IVIG neutralizes harmful antibodies and inflammation. They are equally effective, and combining them is not significantly better than either alone.
Timing matters. Plasmapheresis speeds recovery when used within four weeks of onset. IVIG works as well when started within two weeks, with fewer complications. It is usually tried first because it is easy to give and safe, though risks include occasional liver inflammation and, rarely, kidney failure. Glucocorticoids alone are not effective and could delay recovery.
For pain, studies comparing medications are too limited to recommend any one drug. When the breathing muscles fail, treatment may mean intubation and mechanical ventilation, generally in an intensive care unit.
Around 40% of people then need intensive rehabilitation. A multidisciplinary team led by a neurologist or rehabilitation physician may include physical therapists, occupational therapists, speech language pathologists, dietitians, psychologists and nurses. Physiotherapy builds strength, endurance and gait. Occupational therapy targets daily tasks, driving and work, and speech therapy helps with swallowing and communication, including for people with a tracheostomy. Psychological support may be needed for anxiety, fear and depression.
Support groups exist as well. Inflammatory Neuropathies UK serves the United Kingdom, GBS/CIDP Foundation International serves the USA, and EPODIN is the umbrella for European organizations.
With prompt treatment, most people with severe weakness recover well, though recovery may take weeks to years. About a third have some permanent weakness. Globally, death occurs in approximately 7.5% of patients, and even with optimal care about 5% die from infections, blood clots or cardiac arrest.
Several factors predict a poorer course. They include being over 40, severe symptoms after two weeks and diarrhea before onset. A conduction block on nerve studies predicts worse outcomes at six months. A small rise in blood IgG two weeks after IVIG is linked to poorer mobility at six months.
Researchers grade outcomes on a scale from 0 to 6, where 0 is completely healthy and 6 is death. After six months, about a fifth cannot walk unaided. Many live with chronic pain, fatigue and difficulty with work, education, hobbies and social life, though quality of life improves significantly in the first year.
A course that keeps progressing beyond four weeks, or fluctuates more than twice in eight weeks, may be chronic inflammatory demyelinating polyneuropathy, which is treated differently.
In 1859, Jean-Baptiste Octave Landry first described the disorder. In 1916, Georges Guillain, Jean Alexandre Barré and André Strohl diagnosed two soldiers with it. They identified the key abnormality: more protein in the spinal fluid but a normal cell count. Guillain had reported some of these features in 1938, before the full description.
Other names followed. British neurologist Edwin Bickerstaff described the encephalitis type in 1951 and added another paper in 1957. C. Miller Fisher described his variant in 1956. The axonal subtype was first described in 1986, and further subtypes have appeared since.
Treatment advanced in steps. Plasma exchange was first used in 1978, and larger studies confirmed its benefit in 1985. IVIG arrived in 1988, and studies in the early 1990s showed it to be no less effective. Diagnostic criteria were developed in the late 1970s after cases linked to swine flu vaccination, and refined in 1990. The Brighton Collaboration revised the case definition in 2009, mainly for research.
Since the immunotherapies of the 1980s and 1990s, new treatments have been few. Researchers are testing whether people whose antibody levels rose only slightly after IVIG might benefit from a second course. Studies of mycophenolate mofetil, brain-derived neurotrophic factor and interferon beta have not shown enough benefit for widespread use.
In rats, an animal model called experimental autoimmune neuritis is often used. Glatiramer acetate, quinupramine, fasudil and the heart drug flecainide have shown promise there. An antibody aimed at the anti-GD3 antiganglioside antibody helped in laboratory research. Because complement plays a role, complement inhibitors such as eculizumab may also be effective.
The disease is not limited to people. Dogs get a version called acute polyradiculoneuritis, or "coonhound paralysis," which can appear in a coonhound 7 to 10 days after contact with raccoons.
Continue browsing
Updated
Common questions
What is Guillain, Barré syndrome?
Guillain, Barré syndrome is a rapid-onset muscle weakness caused by the immune system damaging the peripheral nervous system. It usually starts with numbness, tingling or pain, followed by weakness on both sides of the body.
What causes Guillain, Barré syndrome?
The exact cause is unknown, but two-thirds of patients had an infection beforehand. Approximately 30% of cases follow Campylobacter jejuni and 10% follow cytomegalovirus, and the immune response likely involves molecular mimicry.
How is Guillain, Barré syndrome diagnosed?
Diagnosis rests on rapid paralysis, absent reflexes, no fever and no likely alternative cause. Cerebrospinal fluid analysis and nerve conduction studies support it, with protein above 0.55 g/L and fewer than 10 white blood cells per cubic millimeter.
How is Guillain, Barré syndrome treated?
Treatment is intravenous immunoglobulins or plasmapheresis, together with supportive care. The two are equally effective, and IVIG is usually used first because it is easy to give and safe.
Can people recover from Guillain, Barré syndrome?
Most people with severe weakness recover well with prompt treatment, though recovery may take weeks to years. About a third have some permanent weakness, and globally death occurs in approximately 7.5% of patients.
Does the flu vaccine cause Guillain, Barré syndrome?
The 1976 swine flu vaccine raised cases 8.8-fold, with 362 patients among 45 million vaccinated, but this was an outlier. The 2009 pandemic vaccine added about 1 case per million, and natural influenza infection is a stronger risk factor than vaccination.
How common is Guillain, Barré syndrome?
Guillain, Barré syndrome is rare, at 1 or 2 cases per 100,000 people every year. The relative risk rises by 20% for every decade of life, and men are at 1.78 times the risk of women.
Who discovered Guillain, Barré syndrome?
Jean-Baptiste Octave Landry first described the disorder in 1859. The condition is named after French neurologists Georges Guillain and Jean Alexandre Barré, who described it in 1916 with physician André Strohl.
All sources
68 references cited across the entry
- 1Guillain–Barré Syndrome Fact SheetJune 1, 2016
- 2JournalGuillain-Barré syndrome: pathogenesis, diagnosis, treatment and prognosisvan den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA — August 2014
- 3BookFerri's Clinical Advisor 2017: 5 Books in 1Fred F. Ferri — Elsevier Health Sciences — 2016
- 4JournalPopulation incidence of Guillain-Barré syndrome: a systematic review and meta-analysisSejvar JJ, Baughman AL, Wise M, Morgan OW — 2011
- 5JournalClinical features, pathogenesis, and treatment of Guillain-Barré syndromevan Doorn PA, Ruts L, Jacobs BC — October 2008
- 6JournalGuillain Barré syndrome and other immune mediated neuropathies: diagnosis and classificationEldar AH, Chapman J — April 2014
- 7JournalGuillain-Barré syndromeWillison HJ, Jacobs BC, van Doorn PA — August 2016
- 8JournalGuillain-Barré syndromeYuki N, Hartung HP — June 2012
- 9JournalPediatric Guillain-Barré syndromeRyan MM — December 2013
- 10JournalBickerstaff brainstem encephalitis and Fisher syndrome: anti-GQ1b antibody syndromeShahrizaila N, Yuki N — May 2013
- 11JournalIVIG treatment and prognosis in Guillain-Barré syndromevan Doorn PA, Kuitwaard K, Walgaard C, van Koningsveld R, Ruts L, Jacobs BC — May 2010
- 12JournalGuillain-Barre Syndrome, Influenza, and Influenza Vaccination: The Epidemiologic EvidenceC. Vellozzi et al. — 15 April 2014
- 13JournalNeurological complications of dengue virus infectionCarod-Artal FJ, Wichmann O, Farrar J, Gascón J — September 2013
- 14JournalZika virus: a previously slow pandemic spreads rapidly through the AmericasGatherer D, Kohl A — February 2016
- 15JournalInvited Commentary: Influenza Vaccine and Guillain-Barre Syndrome--Is There a Risk?K. E. Nelson — 2012
- 16JournalGuillain-Barré syndrome after exposure to influenza virusLehmann HC, Hartung HP, Kieseier BC, Hughes RA — September 2010
- 17JournalPrincipal controversies in vaccine safety in the United StatesDeStefano F, Bodenstab HM, Offit PA — August 2019
- 18Vaccine Injury Table - HRSAMarch 22, 2017
- 20JournalNeurological associations of COVID-19Ellul MA, Benjamin L, Singh B, Lant S, Michael BD, Easton A, Kneen R, Defres S, Sejvar J, Solomon T — September 2020
- 21JournalGuillain-Barré Syndrome Associated with SARS-CoV-2Toscano G, Palmerini F, Ravaglia S, Ruiz L, Invernizzi P, Cuzzoni MG, Franciotta D, Baldanti F, Daturi R, Postorino P, Cavallini A, Micieli G — June 2020
- 22JournalThe emerging spectrum of COVID-19 neurology: clinical, radiological and laboratory findingsPaterson RW, Brown RL, Benjamin L, Nortley R, Wiethoff S, Bharucha T, Jayaseelan DL, Kumar G, Raftopoulos RE, Zambreanu L, Vivekanandam V, Khoo A, Geraldes R, Chinthapalli K, Boyd E, Tuzlali H, Price G, Christofi G, Morrow J, McNamara P, McLoughlin B, Lim ST, Mehta PR, Levee V, Keddie S, Yong W, Trip SA, Foulkes AJ, Hotton G, Miller TD, Everitt AD, Carswell C, Davies NW, Yoong M, Attwell D, Sreedharan J, Silber E, Schott JM, Chandratheva A, Perry RJ, Simister R, Checkley A, Longley N, Farmer SF, Carletti F, Houlihan C, Thom M, Lunn MP, Spillane J, Howard R, Vincent A, Werring DJ, Hoskote C, Jäger HR, Manji H, Zandi MS — July 2020
- 23JournalGuillain-Barré syndrome spectrum associated with COVID-19: an up-to-date systematic review of 73 casesSamir Abu-Rumeileh et al. — 2021
- 26JournalNeurological complications after first dose of COVID-19 vaccines and SARS-CoV-2 infectionMartina Patone — 2021
- 27Rise in Guillain-Barré syndrome following AstraZeneca vaccineUCL — 2022-05-30
- 28JournalPeripheral neuropathy in COVID-19 is due to immune-mechanisms, pre-existing risk factors, anti-viral drugs, or bedding in the Intensive Care UnitJosef Finsterer et al. — 19 July 2021
- 29JournalGuillain-Barré syndrome following zimeldine treatmentFagius J, Osterman PO, Sidén A, Wiholm BE — January 1985
- 30JournalA paradigm shift in brain researchCarlsson A — November 2001
- 32JournalAxonal Guillain-Barré syndrome: concepts and controversiesKuwabara S, Yuki N — December 2013
- 33JournalGlycolipid antigens and autoantibodies in autoimmune neuropathiesWillison HJ, Goodyear CS — September 2013
- 35JournalClinical practice guideline on diagnosis and treatment of hyponatraemiaSpasovski G, Vanholder R, Allolio B, Annane D, Ball S, Bichet D, Decaux G, Fenske W, Hoorn EJ, Ichai C, Joannidis M, Soupart A, Zietse R, Haller M, van der Veer S, Van Biesen W, Nagler E — March 2014
- 36JournalThe Guillain-Barré syndromeRopper AH — April 1992
- 37JournalUpdate on Guillain-Barré syndromeRinaldi S — June 2013
- 38JournalElectrodiagnostic criteria for Guillain-Barrè syndrome: a critical revision and the need for an updateUncini A, Kuwabara S — August 2012
- 39JournalGuillain-Barré and Miller Fisher syndromes--new diagnostic classificationWakerley BR, Uncini A, Yuki N — September 2014
- 40JournalFisher syndrome: clinical features, immunopathogenesis and managementMori M, Kuwabara S, Yuki N — January 2012
- 41JournalAtypical Bickerstaff brainstem encephalitis: ataxic hypersomnolence without ophthalmoplegiaBR Wakerley et al. — 2013
- 42JournalImmunotherapy of Guillain-Barré syndromeShuang Liu et al. — 2021-06-02
- 43JournalIntravenous immunoglobulin for Guillain-Barré syndromeHughes RA, Swan AV, van Doorn PA — September 2014
- 44JournalPractice parameter: immunotherapy for Guillain-Barré syndrome: report of the Quality Standards Subcommittee of the American Academy of NeurologyHughes RA, Wijdicks EF, Barohn R, Benson E, Cornblath DR, Hahn AF, Meythaler JM, Miller RG, Sladky JT, Stevens JC — September 2003
- 45JournalIntravenous immunoglobulins: in-depth review of excipients and acute kidney injury riskDantal J — October 2013
- 46JournalCorticosteroids for Guillain-Barré syndromeRichard Ac Hughes et al. — 2016-10-24
- 47JournalGuillain-Barré syndrome and variantsDimachkie MM, Barohn RJ — May 2013
- 48JournalPharmacological treatment for pain in Guillain-Barré syndromeJia Liu et al. — 2015-04-09
- 49JournalRehabilitation interventions in patients with acute demyelinating inflammatory polyneuropathy: a systematic reviewKhan F, Amatya B — September 2012
- 50JournalMultidisciplinary care for Guillain-Barré syndromeKhan F, Ng L, Amatya B, Brand C, Turner-Stokes L — October 2010
- 52Home
- 53Epodin
- 54JournalImmunotherapy for Guillain-Barré syndrome: a systematic reviewHughes RA, Swan AV, Raphaël JC, Annane D, van Koningsveld R, van Doorn PA — September 2007
- 55JournalHealth-related quality of life in Guillain-Barré syndrome patients: a systematic reviewDarweesh SK, Polinder S, Mulder MJ, Baena CP, van Leeuwen N, Franco OH, Jacobs BC, van Doorn PA — March 2014
- 56JournalGeorges Guillain's neurologic exploration of AmericaGoetz CG — May 1998
- 57JournalNote sur la paralysie ascendante aiguëJean-Baptiste Landry — 1859
- 59JournalAn unusual variant of acute idiopathic polyneuritis (syndrome of ophthalmoplegia, ataxia and areflexia)Fisher M — July 1956
- 60JournalMesencephalitis and rhombencephalitisBickerstaff ER, Cloake PC — July 1951
- 61JournalBrain-stem encephalitis; further observations on a grave syndrome with benign prognosisBickerstaff ER — June 1957
- 62JournalRecurrent and chronic relapsing Guillain-Barré polyneuritis by P.K. Thomas, R.G. Lascelles, J.F. Hallpike and R.L. Hewer. From the Institute of Neurology and National Hospital, Queen Square, the Royal National Orthopaedic Hospital, and Guy's Hospital, London. Brain 1969; 92: 589–606; with An acute axonal form of Guillain-Barré polyneuropathy by T.E. Feasby, J.J. Gilbert, W.F. Brown, C.F. Bolton, A.F. Hahn, W.F. Koopman and D.W. Zochodne. Brain 1986; 109: 1115–1126Alastair Compston — July 2016
- 63JournalAssessment of current diagnostic criteria for Guillain-Barré syndromeAsbury AK, Cornblath DR — 1990
- 64JournalGuillain-Barré syndrome and Fisher syndrome: case definitions and guidelines for collection, analysis, and presentation of immunization safety dataSejvar JJ, Kohl KS, Gidudu J, Amato A, Bakshi N, Baxter R, Burwen DR, Cornblath DR, Cleerbout J, Edwards KM, Heininger U, Hughes R, Khuri-Bulos N, Korinthenberg R, Law BJ, Munro U, Maltezou HC, Nell P, Oleske J, Sparks R, Velentgas P, Vermeer P, Wiznitzer M — January 2011
- 65JournalEmerging drugs for Guillain-Barré syndromeWalgaard C, Jacobs BC, van Doorn PA — March 2011
- 66JournalTwo cases of acute polyradiculoneuritis in dogs consuming a raw poultry dietSe-Hoon Kim et al. — 2021
- 68BookGuillain–Barré syndrome: from diagnosis to recoveryGareth J. Steinberg et al. — Demos; ANN Press (American Academy of Neurology) — 2006