Status Epilepticus, Guillain-Barré and Myasthenia Gravis

Overview & definition

This domain covers three conditions that bring the patient to the ICU for a neurological reason — the status epilepticus (the prolonged or the recurrent seizure), the Guillain-Barre syndrome (the acute, the ascending paralysis), and the myasthenia gravis (the fatigable weakness). Each can cause the respiratory failure requiring the ICU, and each has a specific, the evidence-based management.[1][1]

Status epilepticus: definition and the phases

The status epilepticus is the seizure lasting over 5 minutes (continuous), or the recurrent seizures without the recovery of the consciousness in between. The operational ILAE definition (Trinka 2015) introduces two time-points: t1 — the threshold beyond which the seizure should be treated (5 minutes for the generalised convulsive SE, 10 minutes for the focal SE with the impaired consciousness); and t2 — the time beyond which the long-term consequences (the neuronal injury, the pharmacoresistance, the network plasticity) become likely (30 minutes for the GCSE). The operational implication: treat at t1, do not wait for t2.[4][1]

The refractory SE (the failure of the first-line benzodiazepine and the second-line antiepileptic drug — typically at the 40-to-60-minute mark) and the super-refractory SE (the persistence or the recurrence beyond 24 hours despite the anaesthetic infusion) carry a rising mortality — the refractory SE mortality approaches 30 to 40 per cent, the super-refractory 35 to 60 per cent in the adult series. The super-refractory SE affects roughly 10 to 15 per cent of the refractory cases.[1][12]

The four phases of the SE management

The phases of the SE management drive the staged drug ladder. The receptor pharmacology underpins the urgency — as the seizure persists, the GABA-A receptors internalise (the benzodiazepines lose efficacy) and the NMDA receptors externalise (the glutamatergic drive strengthens). The early ladder is therefore GABAergic; the late ladder adds the NMDA antagonist (the ketamine).[1][5]

The causes — the aetiology drives the prognosis

The cause is the dominant prognostic factor in the SE — more so than the drug choice. The five cause-groups: [1]

1. The acute symptomatic — the stroke, the trauma, the tumour, the intracranial haemorrhage, the CNS infection, the anoxia (after the cardiac arrest), the alcohol withdrawal, the drug toxicity, the metabolic (the hyponatraemia, the hypoglycaemia, the hepatic and the renal encephalopathy).
2. The remote symptomatic — the prior stroke, the prior trauma, the malformation of the cortical development.
3. The progressive — the tumour, the degenerative, the autoimmune (the anti-NMDA receptor, the anti-LGI1, the anti-GABA-A receptor encephalitis), the paraneoplastic.
4. The cryptogenic / the NORSE — the new-onset refractory SE of unknown cause after the exhaustive work-up; the FIRES subtype follows a febrile prodrome.
5. The known epilepsy with the subtherapeutic level — the non-adherence or the interaction-driven drop in the antiepileptic level. [1]

The NORSE and the FIRES are the cryptogenic syndromes that account for a disproportionate share of the super-refractory SE — the Hirsch consensus definitions (2018) standardise the diagnostic criteria and mandate the autoimmune and the infectious work-up.[9][5]

[1]

The SE management: the evidence

The benzodiazepine first. The benzodiazepines are the most effective drugs in the early SE (the first 10 to 20 minutes), acting on the GABA-A receptor to potentiate the chloride influx and the neuronal hyperpolarisation. Three landmark trials anchor the evidence.[1][5]

• The VA Cooperative Study (Treiman, NEJM 1998) — the open-label randomised trial of the first-line therapy in the overt and the subtle GCSE (n = 518; 384 overt). The IV lorazepam was the most effective monotherapy in the overt SE (the seizure stopped in 65 per cent), superior to the phenytoin (58 per cent), the phenobarbital (56 per cent), and the diazepam (42 per cent). In the subtle SE all four were poor (10 to 25 per cent). The lorazepam became the first-line IV agent.[3]
• The RAMPART trial (Silbergleit, NEJM 2012) — the double-blind non-inferiority RCT of the intramuscular midazolam 10 mg versus the intravenous lorazepam 4 mg for the prehospital SE (n = 893). The IM midazolam was actually superior — the seizure terminated before the arrival in 63 per cent versus 53 per cent — because the IV access was difficult and delayed (no IV access in 23 per cent). The fewer recurrences and the fewer intubations with the IM midazolam. The practice change: the IM midazolam is the first-line prehospital drug when the IV access is not available.[1]
• The Misra 2016 trial — the four-drug Indian comparison (the lorazepam, the levetiracetam, the valproate, the lacosamide) in the established SE — found no significant difference between the drugs, supporting the ESETT-equivalence conclusion in a different population.[8]

The takeaway: a benzodiazepine at five minutes is the single most effective intervention in the SE; the route (the IV lorazepam if the access is present, the IM midazolam if not) is less important than the timing.[5]

[1]

Second-line therapy — the IV antiepileptics (ESETT)

The ESETT (Kapur, NEJM 2019) is the definitive second-line trial. It randomised the patients with the established SE (the benzodiazepine-refractory convulsive SE within 60 minutes) to levetiracetam 60 mg/kg, fosphenytoin 20 mg PE/kg, or valproate 40 mg/kg. The trial was stopped early at the interim analysis for the futility — there was no difference between the three drugs. Each stopped the seizure in about half of the patients within 10 minutes (levetiracetam 47 per cent, fosphenytoin 45 per cent, valproate 46 per cent). The takeaway: any of the three is acceptable, the choice is driven by the patient, and all three are inadequate alone in roughly half.[2]

[1] [1]

Refractory SE — the anaesthetic infusions and the continuous EEG

When the seizure continues despite the benzodiazepine and the second-line agent, the SE is refractory and the patient requires the anaesthetic intubation, the ventilation, and the continuous infusion of an anaesthetic agent titrated to the burst suppression on the continuous EEG. The goal is the burst suppression (the bursts of the activity alternating with the flat suppression, the inter-burst interval of 5 to 10 seconds) maintained for 24 to 48 hours before the weaning and the re-assessment.[1][5]

The Brophy (Neurocritical Care Society 2012) guideline mandates the anaesthetic infusion and the cEEG in the refractory SE, the burst-suppression target for 24 to 48 hours, and the systematic work-up of the underlying cause.[5]

[1]

The continuous EEG — the mandatory monitor in the refractory SE

In the refractory and the super-refractory SE, the EEG is the only way to know whether the seizure has stopped, because the patient is paralysed and sedated and the clinical signs are abolished. The continuous EEG (the cEEG) serves three purposes:[6]

1. To confirm the seizure termination — the ictal rhythm resolves and the background returns, or (in the anaesthetised patient) the burst suppression is achieved.
2. To titrate the anaesthetic — the infusion is adjusted to maintain the burst suppression (the inter-burst interval of 5 to 10 seconds), maintained for 24 to 48 hours before the weaning. The depth targets: the burst suppression, the suppression-burst (a deeper level with longer suppression), or the complete suppression (the isoelectric EEG — reserved for the most refractory).
3. To detect the non-convulsive SE — the patient who fails to wake up after the convulsive SE may be in the NCSE (the continuing epileptiform activity without the motor convulsions). The cEEG detects it; the bedside examination cannot. [1]

The Claassen (Neurology 2004) study of the continuous EEG in the critically ill patient detected the non-convulsive seizures in 8 per cent and the NCSE in 18 per cent of the comatose patients monitored — many with no clinical sign of the ongoing seizure. The Towne (Neurology 2000) study found the 8 per cent prevalence of the NCSE in the comatose ICU patients without the clinical seizure activity, and recommended the cEEG for any comatose patient with the unexplained depressed consciousness.[6][7]

The weaning of the anaesthetic

After 24 to 48 hours of the burst suppression, the anaesthetic is weaned (the infusion reduced by a quarter to a third every 6 to 12 hours) with the continuous EEG running. If the seizures or the ictal pattern recur, the infusion is restarted, the duration extended to 48 to 72 hours, and an additional agent considered (the ketamine, the immunotherapy, the ketogenic diet). The super-refractory SE (the recurrence or the persistence beyond 24 hours of the anaesthetic) may require weeks of the coma and is associated with the high morbidity and the mortality.[1][12]

Super-refractory SE — the desperate options

The super-refractory SE (SRSE) — the persistence or the recurrence of the seizure beyond 24 hours of the anaesthetic infusion — is the rarest and the most challenging form, with a mortality of 35 to 60 per cent in the adult series and a significant long-term morbidity in the survivors. The Ferlisi Global Audit (2015) reported the international case series and the audit framework; the Kantanen (2015) Finland population-based study estimated the SRSE incidence at 0.4 to 2.0 per 100 000 per year and the mortality at 35 per cent.[13][12]

The management of the SRSE adds five classes beyond the standard anaesthetic ladder: [1]

1. The ketamine infusion — the rationale (the NMDA-receptor externalisation late in the SE) and the favourable haemodynamics drive its early use in the SRSE. The Rosati (Neurology 2012) series reported the seizure reduction in roughly two-thirds of the paediatric refractory SE.[11]
2. The ketogenic diet — the 4:1 fat-to-carbohydrate ratio, achieving the ketosis over 3 to 4 days; the Thakur (Neurology 2014) case series of the adults in the SRSE reported the seizure cessation in roughly two-thirds of the treated patients. The practical barriers: the enteral access, the slow onset, the family and the dietitian engagement.[10]
3. The empirical immunotherapy — for the suspected autoimmune cause (the anti-NMDA receptor, the anti-LGI1, the anti-GABA-A receptor encephalitis), the high-dose methylprednisolone (1 g IV daily for 3 to 5 days), the IVIG (0.4 g/kg/day for 5 days), the plasma exchange (5 sessions). The early empirical trial is justified by the high mortality of the SRSE and the treatability of the autoimmune cause.
4. The targeted temperature management — the hypothermia to 32 to 36 C for 24 to 48 hours has been used in the case series; the randomised evidence is absent and the shivering, the infection, and the bradycardia are the complications.
5. The surgery — the resection of the focal epileptic lesion (the tumour, the dysplasia, the cavity of the prior stroke) when the imaging and the electrocorticography localise the focus. The resective surgery or the multiple subpial transection is reserved for the surgically-tractable SRSE.

[1]

The NORSE / FIRES subtypes deserve a special mention — the cryptogenic, the often-young patient with the super-refractory SE of unknown cause after the exhaustive work-up. The Hirsch consensus definitions (2018) standardise the diagnostic criteria; the early empirical immunotherapy (within 7 days) is recommended by the international consensus, the ketogenic diet is offered early, and the surgical work-up is pursued if the focus is localised.[9][13]

Neuromuscular blocking agents in the severe SE — the train-of-four monitoring

The neuromuscular blocking agent (NMBA) is occasionally required in the severe, the convulsive SE — not for the seizure control (the NMBA abolishes only the motor activity, not the cerebral epileptiform activity), but for the patient-ventilator synchrony in the patient with the violent, the uncontrolled motor activity, the ventilator dyssynchrony, the rising intra-abdominal and the intrathoracic pressures, or the rhabdomyolysis from the ongoing muscle activity. The NMBA is NEVER a substitute for the anaesthetic and the antiepileptic therapy — the cEEG is mandatory whenever the NMBA is used, because the only sign of the ongoing seizure is the EEG.[5][1]

The indications for the NMBA in the SE: [1]

• The refractory motor activity despite the adequate anaesthetic load (the propofol, the midazolam, the ketamine) — for the patient-ventilator synchrony.
• The severe rhabdomyolysis (the CK over 5 000 U/L) or the rising compartment pressure from the ongoing convulsions.
• The severe respiratory compromise from the muscular activity (the high intra-thoracic pressure, the impaired venous return, the hypotension).
• The transport of the intubated SRSE patient (the CT or the MRI) — for the safety and the airway protection. [1]

The drug choice. The cisatracurium (the Hoffman elimination, the organ-independent clearance — preferred in the multi-organ failure), the vecuronium (the hepatic clearance), the rocuronium (the rapid onset — for the intubation), or the propofol-and-vecuronium combination. Avoid the depolarising agent (the succinylcholine) for the prolonged infusion — the potassium release, the phase-II block, the malignant hyperthermia risk. [1]

The train-of-four monitoring. The depth of the neuromuscular blockade is monitored by the train-of-four (TOF) — the supramaximal stimulus of four twitches at 0.5-second intervals over the ulnar or the facial nerve, and the count or the ratio of the fourth twitch to the first: [1]

• The TOF count of 4 — no significant blockade; the patient may cough and breathe — for the analgesic, the light-sedation, or the weaning target.
• The TOF count of 1 to 2 — the deep blockade; suitable for the severe SE with the patient-ventilator dyssynchrony; the target for the SRSE with the violent motor activity.
• The TOF count of 0 to 1 — the very deep blockade; reserved for the profound motor suppression to enable the ventilator synchrony and the ICP control; the higher risk of the critical-illness myopathy and the prolonged weakness.
• The TOF ratio (T4/T1) over 0.9 — the standard for the adequacy of the reversal (the recovery from the blockade); the sustained TOF ratio below 0.9 at the cessation of the NMBA predicts the residual curarisation and the post-extubation respiratory failure. [1]

The clinical implication: the NMBA in the SE is titrated to the TOF count of 1 to 2, the adequate analgesia and the sedation (the BIS or the processed-EEG target, the propofol or the midazolam) maintained alongside, and the cEEG running continuously — the seizure activity is invisible when the patient is paralysed. [1]

Non-convulsive status epilepticus (NCSE)

The non-convulsive status epilepticus is the continuous or the recurring epileptiform activity on the EEG without the motor convulsions, producing a persistent impaired consciousness. It is common — affecting 10 to 30 per cent of the comatose ICU patients, especially after a convulsive SE, a severe traumatic brain injury, a subarachnoid haemorrhage, a cardiac arrest, or in the hepatic and the renal encephalopathy. It is detected only by the continuous EEG — the clinical examination cannot distinguish it from the metabolic encephalopathy, the post-ictal state, or the sedation. The Towne (Neurology 2000) prevalence study found the 8 per cent of the comatose ICU patients had the NCSE without the clinical sign.[7][6]

The clinical clues. The fluctuating conscious level, the subtle twitching (the eye deviation, the facial twitching, the rhythmic nystagmoid movement, the distal limb twitching), the unexplained coma after a convulsive SE, the delayed awakening after the anaesthetic, the abnormal eye movements. The diagnosis is the cEEG — request it for any comatose ICU patient with the unexplained depressed consciousness.[6]

The management is the same as the convulsive SE — the benzodiazepine trial (the clinical and the EEG response), the second-line antiepileptic, and the anaesthetic for the refractory. The benzodiazepine trial (the IV midazolam 2 to 4 mg or the lorazepam 1 to 2 mg, the EEG before and after) is both the diagnostic test (the EEG response confirms the NCSE) and the treatment. The over-treatment of the purely-NCSE with the prolonged anaesthetic is debated — the NCSE carries a lower mortality than the convulsive SE, and the aggressive anaesthetic therapy may worsen the outcome in the elderly and the frail.[1][5]

Guillain-Barre syndrome

The GBS is the acute, the immune-mediated demyelinating polyradiculoneuropathy — the ascending, the symmetrical, the flaccid paralysis with the areflexia, progressing over days to weeks. The respiratory failure (the diaphragm and the intercostal involvement), the autonomic dysfunction (the arrhythmia, the blood-pressure swings, the ileus) and the bulbar weakness (the aspiration) are the ICU concerns.[1][1]

The monitoring. The forced vital capacity (the FVC below 20 mL/kg, or the falling trend), the negative inspiratory force (the NIF below -30 cmH2O) and the bulbar weakness are the predictors of the intubation. The 20/30/30 rule: intubate if the FVC below 20 mL/kg, the NIF below -30, or the rapid progression over 2 weeks. The autonomic monitoring (the cardiac, the blood pressure) is essential.[1]

The treatment. The IVIG (0.4 g/kg/day for 5 days) or the plasma exchange (the 5 sessions over 1 to 2 weeks) are the equally effective disease-modifying therapies. The corticosteroids are NOT effective and should not be used. The supportive care (the ventilation, the prophylaxis, the nutrition, the pain, the DVT) is the mainstay.[1][1]

Myasthenia gravis

The MG is the autoimmune (the acetylcholine-receptor antibody) neuromuscular junction disease — the fatigable, the fluctuating weakness (the ocular, the bulbar, the limb, the respiratory). The myasthenic crisis (the respiratory failure from the weakness) is the ICU emergency.[1]

The management of the crisis. The intubation for the respiratory failure (the same FVC and NIF criteria as the GBS), the IVIG or the plasma exchange (the equally effective disease-modifying therapies), the withdrawal of the precipitant (the infection, the non-depolarising muscle relaxant, the aminoglycoside, the beta-blocker, the magnesium), and the long-term therapy (the pyridostigmine, the immunosuppression — the prednisolone, the azathioprine, the mycophenolate, the rituximab).[1][1]

The diagnostic tests. The anti-acetylcholine-receptor antibody, the anti-MUSK antibody (the MUSK MG — the prominent bulbar and respiratory involvement), the single-fibre EMG, the CT of the thymus (the thymoma).[1]

Management: the integrated approach

1. The SE — the IM midazolam (RAMPART) or the IV lorazepam first; the levetiracetam/fosphenytoin/valproate second (ESETT — no difference); the anaesthetic third. The continuous EEG for the refractory. The cause treatment.[1][2][3]
2. The GBS — the FVC/NIF monitoring for the intubation; the IVIG or the plasma exchange; the autonomic and the supportive care.[1]
3. The MG — the IVIG or the plasma exchange for the crisis; the precipitant withdrawal; the long-term immunosuppression.[1][1]

Monitoring

• The SE — the clinical cessation, the continuous EEG (the refractory), the cause.
• The GBS/MG — the FVC, the NIF, the bulbar weakness, the autonomic (the cardiac, the BP), the nutrition, the DVT. [1]

Prognosis

The SE mortality is 10 to 20 per cent (higher in the refractory, the older, the anoxic cause). The GBS mortality is 5 to 10 per cent, with 80 per cent recovering fully or with minor residual. The MG crisis mortality is under 5 per cent with the modern ICU management.[1][1]

[1] [1]

Clinical pearls — the high-yield exam points

Clinical pearl — the integrated weakness assessment

[1]

The SE-vs-NM mimics — the examiner's distinction

Red flags

References

1. [1]Silbergleit R, Durkalski V, Lowenstein D, et al.; NETT Investigators. Intramuscular versus intravenous therapy for prehospital status epilepticus N Engl J Med, 2012.PMID 22335736
2. [2]Kapur J, Elm J, Chamberlain JM, et al.; NETT and ESETT Investigators. Randomized Trial of Three Anticonvulsant Medications for Status Epilepticus N Engl J Med, 2019.PMID 31774955
3. [3]Treiman DM, Meyers PD, Walton NY, et al.; DVA Status Epilepticus Cooperative Study Group. A comparison of four treatments for generalized convulsive status epilepticus. Veterans Affairs Status Epilepticus Cooperative Study Group N Engl J Med, 1998.PMID 9738086
4. [4]Trinka E, Cock H, Hesdorffer D, et al. A definition and classification of status epilepticus--Report of the ILAE Task Force on Classification of Status Epilepticus Epilepsia, 2015.PMID 26336950
5. [5]Brophy GM, Bell R, Claassen J, et al.; Neurocritical Care Society Status Epilepticus Guideline Writing Committee. Guidelines for the evaluation and management of status epilepticus Neurocrit Care, 2012.PMID 22528274
6. [6]Claassen J, Mayer SA, Kowalski RG, Emerson RG, Hirsch LJ. Detection of electrographic seizures with continuous EEG monitoring in critically ill patients Neurology, 2004.PMID 15159471
7. [7]Towne AR, Waterhouse EJ, Boggs JG, et al. Prevalence of nonconvulsive status epilepticus in comatose patients Neurology, 2000.PMID 11087805
8. [8]Misra UK, Kalita J, Maurya PK. A comparison of four antiepileptic drugs in status epilepticus: experience from India Int J Neurosci, 2016.PMID 26456955
9. [9]Hirsch LJ, Gaspard N, van Baalen A, et al. Proposed consensus definitions for new-onset refractory status epilepticus (NORSE), febrile infection-related epilepsy syndrome (FIRES), and related conditions Epilepsia, 2018.PMID 29399791
10. [10]Thakur KT, Probasco JC, Hocker SE, et al. Ketogenic diet for adults in super-refractory status epilepticus Neurology, 2014.PMID 24453083
11. [11]Rosati A, L'Erario V, Ilvento L, et al. Efficacy and safety of ketamine in refractory status epilepticus in children Neurology, 2012.PMID 23197747
12. [12]Kantanen AM, Reinikainen M, Parviainen I, et al. Incidence and mortality of super-refractory status epilepticus in adults Epilepsy Behav, 2015.PMID 26141934
13. [13]Ferlisi M, Hocker S, Grade M, et al.; International Steering Committee of the Global Audit of Treatment of Refractory Status Epilepticus. Preliminary results of the global audit of treatment of refractory status epilepticus Epilepsy Behav, 2015.PMID 25952268