Decision support only — not a substitute for neurology, the EAN/PNS guidance, or eTG. GBS is a clinical diagnosis you should act on early: normal early nerve conduction and normal first-week CSF don't exclude it, and a deteriorating patient is treated on the clinical picture. Two things kill in GBS and both can be silent until late — respiratory muscle failure (trend the FVC) and autonomic instability (monitor the rhythm). Verify all doses.
1 Recognise it early
The clinical picture
- PatternProgressive, relatively symmetric limb weakness with areflexia/hyporeflexia, often ascending; ± distal paraesthesiae, ± facial/bulbar/respiratory involvement.
- TempoMonophasic, worsening over days; nadir by 4 weeks (usually within 2).
- Trigger~2/3 have an infection 1–4 weeks before — Campylobacter jejuni (diarrhoea; axonal, worse), CMV, EBV, Mycoplasma, hepatitis E, Zika, SARS-CoV-2.
Confirm — but treat on clinical grounds
- CSFAlbuminocytologic dissociation (high protein, normal cells) — but may be normal in the first week and is absent in ~10%. Pleocytosis → think HIV, malignancy, other.
- NCSDemyelinating (AIDP) or axonal (AMAN); can be normal in the first 1–2 weeks — a normal study does not exclude GBS.
- AntibodiesNot needed for typical GBS; anti-GQ1b for the Miller-Fisher spectrum, anti-GM1/GD1a for axonal/Campylobacter forms. Brighton criteria grade certainty.
- Doubt GBS ifA sensory level, marked persistent asymmetry, bladder/bowel dysfunction at onset, fever at onset, or CSF pleocytosis (>50 cells) — look for a cord lesion, infection, or vasculitis instead.
- Don't waitIn a deteriorating or non-ambulant patient, start treatment on the clinical diagnosis — don't hold off for confirmatory tests.
2 Variants & atypical presentations
"Ascending weakness" is the textbook picture, but a meaningful minority present differently — and missing the variant is how the diagnosis gets missed. The antibody often tracks the phenotype.
| Variant | Features | Antibody |
|---|
| AIDP | Commonest in the West; demyelinating, sensorimotor | — |
| AMAN | Pure motor, axonal; post-Campylobacter; common in Asia; reversible conduction failure | GM1, GD1a |
| AMSAN | Axonal sensorimotor; severe, slower recovery | GM1, GD1a |
| Miller-Fisher | Ophthalmoplegia + ataxia + areflexia; usually good prognosis (§3) | GQ1b |
| Pharyngeal-cervical-brachial | Bulbar, neck and arm weakness — mimics botulism/MG | GT1a, GQ1b |
| Bickerstaff encephalitis | Miller-Fisher features + drowsiness/encephalopathy + hyperreflexia (central) | GQ1b |
| Other | Paraparetic (leg-predominant); bifacial weakness with paraesthesiae; pure sensory/ataxic | — |
3 Miller-Fisher & the GQ1b spectrum
The syndrome
- TriadOphthalmoplegia, ataxia, areflexia, usually after an infection; limb power often preserved.
- AntibodyAnti-GQ1b positive in ~85–90% — the most useful serology in GBS.
- CourseUsually self-limiting with a good prognosis; treat if severe, progressing, or overlapping with limb/respiratory weakness.
Why the overlaps happen
- GQ1bEnriched in oculomotor nerves, muscle spindles and the brainstem reticular formation — which is why the same antibody produces ophthalmoplegia, ataxia and (centrally) drowsiness.
- SpectrumMiller-Fisher ↔ Bickerstaff (add encephalopathy/hyperreflexia) ↔ pharyngeal-cervical-brachial ↔ typical GBS form one continuum. Recognise the overlaps rather than forcing one label.
4 Treatment
What works
- Two optionsIVIG 2 g/kg over 5 days OR plasma exchange — equivalent efficacy.
- TimingStart early in non-ambulant or rapidly progressing patients (within ~2 weeks, sooner is better) — IVIG works predominantly early.
- ChoosingIVIG usually preferred (no central access, fewer haemodynamic issues); PLEX if IVIG is contraindicated (IgA deficiency) or unavailable. Don't give PLEX after IVIG — it removes it.
What doesn't — and what not to repeat
- SteroidsNot effective in GBS — do not use, alone or added.
- CombiningIVIG + PLEX together is no better than either alone.
- Second IVIGA second IVIG course in poor-prognosis patients (mEGOS ≥6) gives no benefit and more serious adverse events, including thromboembolism SID-GBS. There is no proven rescue for non-responders.
5 New therapies — because current treatment isn't enough
IVIG and plasma exchange have been the only proven treatments for three decades, yet around a fifth of patients can't walk unaided at 6 months and mortality runs ~3–7%. There's no rescue for non-responders, which is what the trial pipeline is trying to fix.
Complement inhibition
- EculizumabC5 inhibitor; early-phase signals, but the phase 3 trial did not meet its primary endpoint (2024) — complement monotherapy underwhelmed.
- ANX005Anti-C1q; phase 3 with promising preliminary results (2025) — a different point in the cascade.
Antibody clearance
- ImlifidaseIgG-cleaving enzyme for rapid removal of pathogenic antibody — phase 2.
- EfgartigimodFcRn inhibitor; case reports in IVIG-refractory disease, no RCT yet.
Where this leaves you
- StandardIVIG or PLEX remains the answer; the rest is investigational.
- Non-responderReconsider the diagnosis (acute-onset CIDP, other), and support — not a reflex second course.
6 Monitoring, complications & prognosis
Respiratory — the priority
- Scale~20–30% need mechanical ventilation.
- MonitorFVC (± NIF) every 6–8 h; predict early failure with EGRIS (rapid onset, facial/bulbar weakness, low MRC sum score).
- ActIntubate on the trend and on bulbar/aspiration risk — the 20/30/40 rule (FVC <20 mL/kg, NIF worse than −30, or >30% FVC fall). Elective beats emergency; SpO₂/ABG fall late.
Autonomic — the quiet one
- Common~2/3 have dysautonomia: labile BP, arrhythmias, ileus, urinary retention, sweating.
- DangerBradyarrhythmia/asystole — especially with tracheal suctioning (vagal). Cardiac monitoring; sudden death risk.
- CautionExaggerated responses to vasoactive drugs — titrate carefully.
The rest of supportive care
- BulbarAspiration risk → swallow assessment, airway protection.
- PreventVTE prophylaxis (immobile), pressure and eye care, bowel/bladder.
- Pain/NaNeuropathic pain is common (gabapentinoids/carbamazepine); watch for SIADH/hyponatraemia.
Prognosis: mEGOS predicts the chance of walking at 6 months; worse with older age, rapid onset, Campylobacter/axonal disease, and the need for ventilation. Most improve over weeks to months, but recovery is incomplete in a substantial minority — which is why supportive care and rehabilitation carry as much weight as the immunotherapy.
Guidelines & reviews. EAN/PNS guideline on diagnosis and treatment of Guillain-Barré syndrome (van Doorn et al, 2023); "Diagnosis and treatment of GBS in ten steps" (Leonhard et al, Nat Rev Neurol 2019); AAN practice parameter on immunotherapy; Brighton collaboration diagnostic criteria; Yuki & Hartung, "Guillain-Barré Syndrome," NEJM 2012; Lancet seminar (Shahrizaila et al 2021); eTG.
Trials. Cochrane reviews — IVIG and plasma exchange equally effective, combination not beneficial, corticosteroids ineffective; SID-GBS (Walgaard et al, Lancet Neurol 2021 — no benefit of a second IVIG course in poor-prognosis GBS, more thromboembolic events); JET-GBS/eculizumab (phase 2 signal, phase 3 negative 2024); ANX005 anti-C1q (phase 3, preliminary 2025); imlifidase (phase 2). Prognosis/monitoring: EGRIS (respiratory), mEGOS/EGOS (walking at 6 months).
Caveats. Diagnose and treat clinically — early NCS and first-week CSF can be normal. Steroids don't work; a second IVIG course in poor-prognosis patients doesn't either. Respiratory failure and autonomic instability are the killers and both hide until late — trend the FVC, monitor the rhythm, and beware suction-induced bradycardia. Miller-Fisher and its overlaps are anti-GQ1b-driven. Verify all doses. Companion to the neuromuscular set (myasthenia gravis, myopathy, MND).