Guillain-Barré Syndrome

Meet the patient

A 55-year-old man presents with weakness that began in his feet four days ago and has crept up to his thighs; this morning he could not stand. Two weeks earlier he had a week of bloody diarrhoea after a barbecue. His reflexes are absent at the ankles and knees, his plantars are flexor, and sensation is intact. He can no longer count to ten in one breath.[2]

Two questions now own his next week, and they own every GBS case: how fast is he climbing, and how close is his breathing to failing? Everything below exists to answer those two questions — and to stop you reaching for the steroid that will not work, or missing the autonomic swing that kills him between observations.[1][2]

The commonest acute paralysis — and the danger is not the weakness

GBS is the commonest cause of acute generalised neuromuscular paralysis in the developed world, and it is a neurological emergency. It is monophasic and immune-mediated: provoked by a recent infection, the immune system attacks the peripheral nerves and spinal nerve roots, producing symmetric weakness that ascends from the legs, almost always with lost reflexes, reaching its nadir within four weeks before plateauing and recovering.[1][2]

The danger is not the weakness itself — it is the two complications. Respiratory failure weakens the diaphragm and intercostals in 20 to 30 percent, and autonomic instability drives the arrhythmias and blood-pressure swings that are the commonest cause of sudden death. The discipline of managing GBS is therefore to recognise the ascending pattern early, admit for serial respiratory monitoring, and treat with IVIg or plasma exchange while providing meticulous supportive care. Most patients recover well, but around one in five is left with significant residual disability.[2]

The umbrella — six subtypes, one antibody map

GBS is an umbrella term covering several distinct clinicopathological subtypes. They share the temporal profile (acute, monophasic, nadir within four weeks, post-infectious) but differ in the target of immune attack (myelin versus axon), the pattern of weakness, and the specific ganglioside antibody that drives the disease.[4]

[4]

How common, and the trigger to name first

GBS is rare but not negligible — about 100,000 people worldwide every year. Incidence rises steadily with age (doubling each decade after the second), with a slight male predominance, and it can spike sharply during outbreaks: the Zika epidemics in French Polynesia (2013) and the Americas (2015 to 2016) raised GBS incidence 5- to 20-fold in affected regions.[2][10]

[2]

The hallmark is the antecedent infection, one to six weeks (usually two to three) before the weakness begins. About 70 percent recall a respiratory or gastrointestinal infection in that window, and identifying the trigger predicts the subtype and the antibody.[4]

[4]

Molecular mimicry — why the antibody matters

The unifying mechanism is molecular mimicry: an antecedent pathogen expresses surface molecules that structurally resemble components of the peripheral nerve, so the antibodies raised to clear the infection cross-react with the nerve. The clearest example is Campylobacter jejuni, whose lipo-oligosaccharide bears epitopes that mimic GM1 and GD1a gangliosides concentrated at the motor node of Ranvier — producing the AMAN phenotype. In Miller Fisher syndrome the target is GQ1b (concentrated in oculomotor myelin), giving the ophthalmoplegia.[2][4]

The downstream cascade differs by subtype. In AIDP, antibodies and complement deposit membrane attack complex on the outermost myelin lamellae of the spinal roots and peripheral nerves; macrophages strip the myelin and produce segmental demyelination. Because the axon is largely preserved, conduction block predominates and is potentially reversible — which is why AIDP recovery tends to be faster and more complete. In AMAN, the antibodies bind the nodal and paranodal axolemma of the motor nerve, complement deposits, and macrophages invade the periaxonal space, producing Wallerian-like axonal degeneration; recovery requires axonal regrowth at about 1 millimetre per day, which is slow and often incomplete.[4]

Three consequences of the location of attack explain the clinical picture. Because the longest nerves are affected first, weakness ascends from feet to legs to arms — the cardinal feature. Because the inflammation is in the spinal nerve root rather than the meninges, the blood-nerve barrier leaks albumin into the CSF without a cellular response — the albuminocytological dissociation that is the CSF signature. And because autonomic fibres travel in the same roots, they are damaged alongside motor and sensory fibres — the autonomic instability that is the leading cause of sudden death.[1]

Four phases, and the bedside warning signs

GBS unfolds in four phases: a prodromal infectious illness one to six weeks earlier, a progressive phase of weakness (days to four weeks), a plateau (days to weeks), and recovery (weeks to months, sometimes years).[1][2]

Weakness is the defining feature — symmetric, ascending (feet to legs to thighs to trunk to arms to face), flaccid, reaching nadir within four weeks (most within one to two). Areflexia is early and characteristic: the deep tendon reflexes are lost (absent ankle and knee jerks, then upper-limb reflexes), distinguishing GBS from myopathy (reflexes preserved early) and spinal cord disease (brisk after spinal shock). Sensory symptoms are mild by comparison — distal paraesthesia in the hands and feet, and back and thigh pain that is under-recognised and frequently the presenting complaint before weakness becomes obvious.[1][2]

Cranial nerve involvement is common — bilateral facial (VII) palsy in up to half, bulbar weakness (dysphagia, dysarthria, weak cough — an aspiration risk), and ophthalmoplegia in Miller Fisher. Sphincter function is usually spared early — a key discriminator from cauda equina and acute spinal cord compression, where bladder and bowel involvement are early. Autonomic dysfunction is dangerous and under-recognised: sinus tachycardia or bradycardia, blood-pressure swings, cardiac arrhythmia, paralytic ileus, urinary retention, sweating disturbance and SIADH. Check the serum sodium daily in the first week.[1][2]

Respiratory involvement develops in 20 to 30 percent and is the indication for ventilation. The bedside warning signs are tachypnoea, accessory-muscle use, a weak cough, inability to count to ten in one breath, and a falling serial vital capacity. Pain — deep aching back and thigh pain, painful paraesthesiae — is neuropathic (root and nerve inflammation) and musculoskeletal (immobility), and is often the dominant early complaint; treat it actively.[1]

Atypical presentations are deliberate examiner traps. Pure motor GBS (no sensory loss, often anti-GM1) is distinguished from a myopathy by the rapid tempo, areflexia and CSF. The pharyngeal-cervical-brachial variant mimics brainstem stroke or myasthenia. Miller Fisher may present with ataxia and diplopia alone. Bickerstaff adds drowsiness and hyperreflexia. In the elderly, GBS can present as a fall or functional decline; in children, as refusal to walk, pain or ataxia.[4]

The face-off — what else climbs and areflexic?

The differential of acute ascending flaccid paralysis is broad, and the discriminator is usually pattern, tempo and reflexes — plus a few targeted tests.[1][4]

[1]

The two cannot-miss mimics are acute spinal cord compression (a surgical emergency that GBS is not) and tick paralysis (cured by removing the tick). The bedside rule: in any patient with rapidly progressive limb weakness, examine for a sensory level and test the plantars and sphincter tone first — an upgoing plantar, a sharp sensory level, or a distended flaccid bladder points away from GBS and toward the cord, and demands immediate MRI.[1]

The CSF signature — and the first-week trap

GBS is a clinical diagnosis, supported by three pillars: the clinical pattern, the CSF, and nerve conduction studies. No single test is necessary, but together they confirm the diagnosis, define the subtype, and exclude mimics.[1]

Nerve conduction studies define the subtype and predict prognosis. In AIDP the hallmarks are prolonged distal motor latencies, conduction block, slowed motor conduction velocity, temporal dispersion, and prolonged or absent F-waves — all reflecting demyelination. In AMAN the CMAP amplitudes are reduced with normal sensory studies and normal velocities, reflecting axonal loss. F-wave abnormalities are among the earliest changes because they test the proximal root, which is inflamed first. Antibodies (anti-GQ1b for Miller Fisher, anti-GM1 and anti-GD1a for AMAN) confirm the subtype but are not essential for classic AIDP.[4]

The Brighton Collaboration criteria standardise GBS diagnosis across trials and vaccine-safety surveillance: Level 1 adds NCS plus CSF plus a detected antecedent infection to bilateral flaccid paralysis; Levels 2 and 3 step down as confirmatory tests drop away. They are not for everyday use, but examiners invoke them in vaccine-safety and epidemiology stems.[3]

The four pillars of management — and the numbers that own the airway

The resuscitation of GBS has one overriding principle: admit to a monitored bed (HDU or ICU) the moment there is any respiratory, bulbar or autonomic involvement, because the trajectory is unpredictable and 20 to 30 percent will need ventilation.[1]

[1]

The bedside respiratory measurements are the most important numbers you will take in GBS. Perform serial FVC, peak expiratory flow and negative inspiratory force every four to six hours in any patient with weakness progressing or affecting the arms, face or bulbar muscles.[1][8]

[1]

When you intubate, plan it as an elective procedure. Pre-oxygenate fully, choose propofol or thiopentone at reduced dose (autonomic instability makes these patients cardiovascularly sensitive), and a short-acting neuromuscular blocker such as rocuronium (with sugammadex available) — suxamethonium carries a hyperkalaemia risk in any denervating illness after the first week, so prefer rocuronium in late-presenting GBS. Have atropine, a vasopressor and a transcutaneous pacer ready, because laryngoscopy can precipitate profound bradycardia or asystole.[1]

Autonomic resuscitation is the second pillar. Treat symptomatic bradycardia with atropine or isoprenaline (transcutaneous pacer ready); treat severe hypertension with a short-acting beta-blocker or nitrate (avoid rapid swings); decompress paralytic ileus nasogastrically; catheterise for retention; and correct SIADH-related hyponatraemia with fluid restriction. Analgesia is needed for neuropathic pain and procedural discomfort — fentanyl is preferred over morphine for less haemodynamic effect. DVT prophylaxis (enoxaparin 40 mg subcutaneously daily, dose-adjusted for renal function, plus compression stockings) starts on admission, because immobility and the prothrombotic state of inflammation make PE a leading cause of death.[1]

IVIg or exchange — equally effective, never combined, never steroids

The two equally effective first-line disease-modifying treatments are intravenous immunoglobulin and plasma exchange. The Cochrane meta-analyses confirm that each hastens recovery to the same degree, that combining them adds no benefit, and that corticosteroids alone do not work.[5][6][7]

[1]

Indications for IVIg or plasma exchange: any patient unable to walk independently (MRC sum score under 40), or with rapidly progressive weakness, respiratory involvement, bulbar weakness, or autonomic instability — started within 2 to 4 weeks of onset. Mild cases who can still walk may be observed, but most clinicians treat any patient whose trajectory suggests they will not be walking within days.[1]

Pain management is often under-treated. First-line is gabapentin 300 to 600 mg three times daily titrated, or pregabalin 75 to 150 mg twice daily; carbamazepine 100 to 200 mg twice daily for paroxysmal neuropathic pain; opioids (tramadol 50 to 100 mg, morphine 2.5 to 5 mg) for severe pain. NSAIDs alone are inadequate for neuropathic pain.[1]

The subtypes that change the answer

Miller Fisher syndrome presents with the triad of ophthalmoplegia, ataxia and areflexia; the ataxia is sensory (proprioceptive), not cerebellar (no dysarthria or nystagmus), and anti-GQ1b is positive in over 90 percent. It may progress to overlap with GBS and is treated the same way — IVIg or plasma exchange. Bickerstaff brainstem encephalitis overlaps Miller Fisher but adds altered consciousness and hyperreflexia or a Babinski sign; the same anti-GQ1b is positive, MRI may show brainstem signal change, and treatment is identical.[2][4]

AMAN is pure motor (no sensory loss), shows reduced CMAP amplitudes with normal sensory studies and velocities, is strongly linked to Campylobacter and anti-GD1a or anti-GM1, and is commoner in Asia; recovery depends on axonal regrowth and is slower and sometimes less complete than AIDP. AMSAN applies the same axonal mechanism to motor and sensory fibres.[4]

The pharyngeal-cervical-brachial variant produces weakness of the pharynx, neck and arms with relative leg-sparing, mimicking myasthenia or brainstem stroke. Paediatric GBS (aged 1 to 10) presents with refusal to walk, leg pain, ataxia and weakness, and recovers faster and more completely; IVIg is weight-based at 2 g/kg total. GBS in pregnancy is managed unchanged — IVIg is safe — with added DVT prophylaxis and multidisciplinary planning. Treatment-related fluctuation — a relapse after initial improvement within 8 weeks, in up to 10 percent — is re-treated with IVIg; fluctuations beyond 8 weeks demand investigation for CIDP.[1]

How GBS patients come to harm — the preventable list

• The crash intubation in a dysautonomic, bulbar patient — the preventable death. Intubate early and electively.[1]
• Missing the normal first-week CSF as "not GBS" — repeat the lumbar puncture.[3]
• Reaching for corticosteroids and delaying IVIg or plasma exchange.[7]
• Treating a spinal cord compression as GBS because no one examined for a sensory level or tested the plantars.[1]
• Ignoring autonomic swings — a suction catheter precipitating asystole because atropine was not at the bedside.[1]
• Forgetting DVT prophylaxis — PE is a leading cause of death in the immobile patient.[1]
• Missing SIADH-related hyponatraemia because no one checked the sodium daily in week one.[1]
• Labelling treatment-related fluctuation as treatment failure rather than re-treating or thinking about CIDP.[1]

Prognosis, disposition, and what to tell the family

Most patients reach nadir within 2 to 4 weeks, plateau for days to weeks, then recover over weeks to months (sometimes years for full recovery). About 80 percent recover fully or with minor residual deficits; about 20 percent have significant residual disability (unable to walk independently at six months). Mortality is 3 to 7 percent even with modern care, and the leading causes are autonomic dysfunction, respiratory failure, sepsis and pulmonary embolism.[1][2]

The poor prognostic factors load the EGOS and mEGOS: older age, antecedent diarrhoea (Campylobacter, often AMAN), severe weakness at nadir (MRC sum under 40), need for mechanical ventilation, axonal physiology on NCS, delayed treatment, and the EGRIS-predicted need for ventilation.[8][9]

[1]

Disposition follows the trajectory. Any patient with respiratory involvement, rapid progression, severe weakness, bulbar weakness or autonomic instability goes to HDU or ICU. Stable, mild cases (still walking, slow progression) may be managed on a neurology ward with serial FVC. After the acute phase, a rehabilitation unit supports recovery; community rehabilitation continues for residual deficits. The safety-net includes serial FVC during recovery, monitoring for treatment-related fluctuation, evaluation for CIDP if symptoms continue beyond 8 weeks, early physiotherapy, and vaccination advice — avoid the specific trigger vaccine if it preceded onset within 6 weeks; otherwise vaccination, including influenza, remains safe and recommended.[1]

Special populations

[1]

The evidence, the regions, and the cost delta

The evidence base is dominated by three Cochrane meta-analyses and the Brighton criteria. The Cochrane IVIg review (Hughes 2012) showed IVIg hastens recovery as much as plasma exchange, with simpler administration and fewer lines. The Cochrane plasma-exchange review (Raphaël 2012) confirmed plasma exchange improves outcome versus supportive care alone, with benefit strongest within two to four weeks. The Cochrane corticosteroid review (Hughes 2016) is the definitive evidence that corticosteroids alone are not effective. The Brighton criteria (Fokke 2014) standardised diagnosis across trials and surveillance. The IGOS cohort (Doets 2022) internationally validated EGRIS and mEGOS.[3][5][6][7][9]

The weak evidence for complement-inhibitor therapies (eculizumab, ravulizumab) in severe GBS — early-phase studies only, not standard care — is mentioned for completeness; do not invoke these as first-line.[4]

The mantra, and the mnemonic

[1]

The mantra: ascending, areflexic, four weeks — IVIg or exchange, never steroids, and watch the FVC.[1][7]

Ward-round test — three stems, thirty seconds each

Stem 1 — the man from the top of the topic (answer)

The 55-year-old with ascending leg weakness over four days after bloody diarrhoea, areflexia, and an inability to count to ten in one breath. What is the diagnosis, the two key investigations, and the treatment? Model: This is AIDP following Campylobacter jejuni gastroenteritis. The two key investigations are lumbar puncture (looking for albuminocytological dissociation — high protein, normal cells — but remembering it may be normal in the first week) and nerve conduction studies (prolonged distal latencies, conduction block, slow velocity, prolonged F-waves). Admit to HDU or ICU, measure serial FVC every four to six hours, and treat with IVIg 0.4 g/kg per day for 5 days (total 2 g/kg) — plasma exchange is equally effective. Give DVT prophylaxis (enoxaparin 40 mg daily), start nasogastric feeding, place atropine and a pacer at the bedside for autonomic instability, and check the sodium daily. Do not give corticosteroids.[1][7]

Stem 2 — the normal first-week CSF trap (answer)

A 30-year-old with three days of ascending weakness and areflexia has a lumbar puncture on day 4 that shows protein 0.4 g/L and 2 white cells — entirely normal. The registrar concludes this is not GBS. What is the error? Model: The error is treating a normal first-week CSF as excluding GBS. Albuminocytological dissociation is found in only 50 to 66 percent of patients in the first week and rises to over 85 percent after week two. The diagnosis is clinical — symmetric ascending flaccid paralysis with areflexia reaching nadir within four weeks — supported by nerve conduction studies. Repeat the lumbar puncture after one to two weeks if the diagnosis remains suspected. Conversely, a CSF white cell count over 50 should prompt an HIV test, not reassurance. Treating this patient as GBS with IVIg and respiratory monitoring is correct; sending him home because the CSF is normal is the classic, dangerous trap.[1][3]

Stem 3 — double vision and unsteadiness, no limb weakness (answer)

A 40-year-old woman presents with diplopia and unsteadiness for three days. Examination shows ophthalmoplegia, gait ataxia and absent reflexes, but normal limb power and sensation. What is this, what is the antibody, and what distinguishes the ataxia from cerebellar disease? Model: This is Miller Fisher syndrome — the triad of ophthalmoplegia, ataxia and areflexia, with anti-GQ1b positive in over 90 percent. The ataxia is sensory (proprioceptive), not cerebellar: there is no dysarthria and no nystagmus. Miller Fisher may overlap with GBS (limb weakness can follow) and is treated identically — IVIg 0.4 g/kg per day for 5 days or plasma exchange. Distinguish it from Bickerstaff brainstem encephalitis, which adds altered consciousness and hyperreflexia or a Babinski sign but shares the same anti-GQ1b antibody. Admit for monitoring even without limb weakness, because progression to respiratory involvement is possible.[2][4]

References

1. [1]Leonhard SE, Mandarakas MR, Gondim FAA, et al. Diagnosis and management of Guillain-Barré syndrome in ten steps Nat Rev Neurol, 2019.PMID 31541214
2. [2]Willison HJ, Jacobs BC, van Doorn PA. Guillain-Barré syndrome Lancet, 2016.PMID 26948435
3. [3]Fokke C, van den Berg B, Drenthen J, Walgaard C, van Doorn PA, Jacobs BC. Diagnosis of Guillain-Barré syndrome and validation of Brighton criteria Brain, 2014.PMID 24163275
4. [4]van den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA. Guillain-Barré syndrome: pathogenesis, diagnosis, treatment and prognosis Nat Rev Neurol, 2014.PMID 25023340
5. [5]Hughes RA, Swan AV, van Doorn PA. Intravenous immunoglobulin for Guillain-Barré syndrome Cochrane Database Syst Rev, 2012.PMID 22786476
6. [6]Raphaël JC, Chevret S, Hughes RA, et al. Plasma exchange for Guillain-Barré syndrome Cochrane Database Syst Rev, 2012.PMID 22786475
7. [7]Hughes RA, Brassington R, Gunn AA, van Doorn PA. Corticosteroids for Guillain-Barré syndrome Cochrane Database Syst Rev, 2016.PMID 27775812
8. [8]Walgaard C, Lingsma HF, Ruts L, et al. Prediction of respiratory insufficiency in Guillain-Barré syndrome Ann Neurol, 2010.PMID 20517939
9. [9]Doets AY, Walgaard C, Lingsma HF, et al. International Validation of the Erasmus Guillain-Barré Syndrome Respiratory Insufficiency Score Ann Neurol, 2022.PMID 35106830
10. [10]Walteros DM, Soares J, Styczynski AR, et al. Long-term outcomes of Guillain-Barré syndrome possibly associated with Zika virus infection PLoS One, 2019.PMID 31369576