Refractory & Super-Refractory Status Epilepticus
Refractory status epilepticus (RSE) is defined as SE that persists after failure of two adequately dosed antiseizure medications (ASMs) from different drug classes, typically a benzodiazepine and one non-benzodiazepine agent. Super-refractory status epilepticus (SRSE) represents the most extreme end of the SE spectrum — seizures that persist or recur despite ≥24 hours of continuous anesthetic infusion, including cases that recur on weaning of anesthetics. RSE develops in 23–55% of all SE patients, with in-hospital mortality of 15–39% in adults. SRSE carries a mean duration of 36 days, in-hospital mortality of 24%, and only 27% of survivors achieve zero to slight disability at hospital discharge.
Bottom Line
- RSE definition: SE persisting after failure of two adequately dosed ASMs from different drug classes (e.g., a benzodiazepine + fosphenytoin/valproate/levetiracetam)
- SRSE definition: SE that persists or recurs ≥24 hours after initiation of anesthetic infusions, including cases that recur during anesthetic weaning
- Anesthetic agents: Midazolam, propofol, and pentobarbital/thiopental are the three principal agents; no Class I evidence favors one over another
- cEEG monitoring: Mandatory throughout anesthetic infusion; target burst suppression (5–15 second interburst intervals) for 24–48 hours before weaning
- NORSE/FIRES: Important causes of SRSE; require early immunotherapy within 72 hours and ketogenic diet consideration within 7 days
- Prognosis: Etiology is the strongest predictor of outcome; in nonanoxic SRSE, 24% in-hospital mortality and only 27% with favorable functional outcome at discharge
Definitions and Staging
The 2015 ILAE classification and subsequent AES publications established a clear staging framework for SE that distinguishes RSE and SRSE from earlier stages:
| Stage | Definition | Approximate Incidence (Among SE Patients) | Mortality |
|---|---|---|---|
| Established SE | Ongoing seizure activity ≥5 min (convulsive) or ≥10 min (nonconvulsive) that does not respond to initial stabilization | 100% (by definition) | 5–10% (lower in breakthrough epilepsy) |
| Refractory SE (RSE) | SE persisting after failure of two adequately dosed ASMs from different classes (benzodiazepine + second-line agent) | 23–55% | 15–39% (adults); lower in children |
| Super-refractory SE (SRSE) | SE that continues or recurs ≥24 hours after onset of anesthetic infusion, including seizures recurring during dose reduction or withdrawal of anesthetics | 10–15% of all SE; ~22% of RSE | 24% in-hospital mortality (nonanoxic) |
| New-onset RSE (NORSE) | RSE in a patient without active epilepsy or preexisting neurologic disorder, without a clear acute structural/toxic/metabolic cause | ~10–16% of RSE patients | 19–22%; full recovery in only 4% |
The Significance of "Adequately Dosed"
A critical nuance in the definition of RSE is the requirement that both ASMs must have been adequately dosed. As documented by the SENSE registry and multiple studies, benzodiazepine underdosing occurs in >75% of SE patients. When patients who appear to have "refractory" SE have actually received subtherapeutic first-line doses, the refractoriness may be partly iatrogenic. Before escalating to anesthetic infusions, clinicians should confirm that guideline-recommended doses of both first-line benzodiazepines and second-line ASMs were administered.
Pathophysiology of Pharmacoresistance
The progression from established SE to RSE and SRSE reflects a time-dependent cascade of molecular and cellular changes that progressively reduce the effectiveness of standard ASMs:
| Mechanism | Timeline | Consequence | Therapeutic Implication |
|---|---|---|---|
| GABAA receptor internalization | Within minutes | Reduced synaptic GABAA receptors; loss of benzodiazepine binding sites; decreased chloride conductance | Explains declining benzodiazepine efficacy; rationale for early, full-dose benzodiazepine treatment |
| NMDA receptor externalization | Minutes to hours | Increased surface NMDA receptors; enhanced glutamatergic excitation; calcium influx; excitotoxicity | Rationale for ketamine (NMDA antagonist) use in RSE/SRSE |
| GABAA receptor subunit changes | Hours | Shift from benzodiazepine-sensitive α1/γ2 subunits to benzodiazepine-insensitive α4/δ subunits | Further benzodiazepine resistance; potential role for neurosteroids (allopregnanolone) that modulate δ-containing receptors |
| Neuropeptide depletion | Hours to days | Depletion of inhibitory neuropeptides (neuropeptide Y, galanin, dynorphin); accumulation of excitatory neuropeptides (substance P, neurokinin B) | Shifts the excitatory-inhibitory balance further toward excitation |
| Blood-brain barrier disruption | Hours to days | Increased BBB permeability; neuroinflammation; peripheral immune cell infiltration; albumin extravasation → astrocyte activation | Rationale for anti-inflammatory and immunomodulatory therapies in SRSE |
| Neuroinflammation | Days | Microglial activation; pro-inflammatory cytokine release (IL-1β, IL-6, TNF-α); astrogliosis; further excitotoxicity | Rationale for immunotherapy (steroids, IVIg, IL-1/IL-6 blockade) in cryptogenic SRSE/NORSE |
| Epigenetic changes | Days to weeks | Altered gene expression in ion channels and neurotransmitter receptors; epileptogenesis; development of chronic drug-resistant epilepsy | Prolonged SRSE creates a self-perpetuating epileptogenic state; early, aggressive treatment may prevent this |
Anesthetic Agents for RSE
When a decision is made to escalate to anesthetic infusions for refractory SE, intubation, mechanical ventilation, and transfer to an ICU with continuous EEG monitoring are mandatory. The three principal anesthetic agents used in RSE are midazolam, propofol, and pentobarbital (thiopental in Europe). No Class I evidence establishes the superiority of one agent over another.
Midazolam Continuous Infusion
| Parameter | Details |
|---|---|
| Loading dose | 0.2–0.5 mg/kg IV |
| Maintenance infusion | 0.1–2.0 mg/kg/h (or 2–40 μg/kg/min); titrate to EEG target |
| Half-life | 1–4.5 hours (child); 2–7 hours (adult); up to 24 hours with prolonged administration |
| Advantages | Least hemodynamic compromise of the three agents; relatively rapid onset; predictable pharmacokinetics in short-term use |
| Disadvantages | Significant tachyphylaxis with prolonged use (requires dose escalation); active metabolites accumulate in renal failure; prolonged recovery after extended infusions |
| Practical notes | Often used as the initial anesthetic agent in many centers due to its favorable hemodynamic profile; breakthrough seizures are common, requiring dose escalation or transition to pentobarbital |
Propofol Continuous Infusion
| Parameter | Details |
|---|---|
| Loading dose | 1–2 mg/kg IV bolus; may repeat every 3–5 minutes up to a total of 10 mg/kg |
| Maintenance infusion | 1–15 mg/kg/h initially; reduce to maximum 5 mg/kg/h for long-term use |
| Half-life | 0.67 hours initially; 4–7 hours with prolonged use; >24 hours after >10 days |
| Advantages | Rapid onset and offset (for short-term use); potent anticonvulsant; favorable for neurologic assessments due to rapid washout |
| Disadvantages | Propofol infusion syndrome (PRIS) is the major concern — see warning box below; hypotension; hypertriglyceridemia; high caloric load (propofol is in a lipid emulsion) |
| Practical notes | Avoid prolonged use in children (PRIS risk highest); monitor triglycerides, CK, lactate, and lipase daily; count propofol calories in nutritional plan |
Propofol Infusion Syndrome (PRIS)
- Definition: A rare but potentially fatal syndrome characterized by unexplained metabolic acidosis, rhabdomyolysis, hyperkalemia, hepatomegaly, renal failure, cardiac failure, and lipemia
- Mechanism: Impaired mitochondrial fatty acid oxidation and disruption of the electron transport chain
- Risk factors: High doses (>5 mg/kg/h), prolonged infusions (>48 hours), concurrent catecholamine infusions, critical illness, young age (particularly children)
- Monitoring: Daily CK, lactate, triglycerides, and arterial blood gas; monitor for unexplained tachycardia, new Brugada-like ECG pattern, or worsening acidosis
- Management: Immediately discontinue propofol; aggressive supportive care; hemodialysis or hemofiltration for severe metabolic derangements; PRIS mortality approaches 30–80% once established
- Prevention: Limit infusion rate to ≤5 mg/kg/h whenever possible; limit duration; avoid in children when alternatives exist; monitor biomarkers daily
Pentobarbital (Thiopental) Continuous Infusion
| Parameter | Details |
|---|---|
| Loading dose | 5–15 mg/kg IV at a maximum rate of 50 mg/min |
| Maintenance infusion | 0.5–5 mg/kg/h; titrate to burst suppression on cEEG |
| Half-life | 15–22 hours (can be much longer with prolonged infusion due to tissue redistribution) |
| Advantages | Most potent seizure suppressor; reliable induction of burst suppression; long experience in SE management |
| Disadvantages | Profound hemodynamic instability (hypotension requiring vasopressors in most patients); immunosuppression; paralytic ileus; very prolonged recovery period (days to weeks); complicates neurologic assessment and prognostication |
| Practical notes | Generally reserved for cases refractory to midazolam and/or propofol; requires arterial line, central venous access, and often vasopressor support; invasive hemodynamic monitoring recommended |
Choosing an Anesthetic Agent
Practical Considerations for Agent Selection
- Midazolam is often the initial choice due to its relatively favorable hemodynamic profile and familiarity; however, tachyphylaxis limits its effectiveness in prolonged RSE
- Propofol is favored when rapid neurologic reassessment is desirable and for shorter anticipated treatment courses (<48 hours); avoid in children and monitor for PRIS
- Pentobarbital is reserved for the most refractory cases when midazolam and/or propofol have failed; its hemodynamic consequences require full ICU support with vasopressors
- A 2024 study comparing midazolam and propofol infusions in adults with RSE across high-income and low–middle-income countries found equally good outcomes with both agents
- In a 2024 Swiss study, anesthetic dose escalation was required in 57% of 111 adults with RSE; despite higher morbidity, survivors who required dose escalation had decreased odds of in-hospital death
- Combination strategies: Some centers use midazolam + propofol concurrently at lower doses to achieve burst suppression while minimizing the toxicity of either agent alone (limited evidence)
Continuous EEG Monitoring Requirements
Continuous EEG monitoring is the indispensable companion to anesthetic therapy in RSE and SRSE. Without cEEG, it is impossible to confirm seizure suppression, detect breakthrough seizures, guide anesthetic titration, or safely wean anesthetic infusions.
EEG Targets During Anesthetic Infusion
- Burst suppression: The standard target; interburst intervals of 5–15 seconds are typical. Achieved by titrating the anesthetic infusion upward until the EEG shows alternating bursts of activity and periods of electrocortical silence.
- Seizure suppression: An alternative, less aggressive target; the goal is complete cessation of electrographic seizure activity without necessarily achieving burst suppression. May be appropriate for NCSE with a less aggressive treatment paradigm.
- Complete electrocerebral inactivity: Rarely targeted; associated with higher complication rates without clear benefit over burst suppression.
Monitoring Protocol
- Initiate cEEG before starting the anesthetic infusion to document the baseline seizure pattern
- Maintain cEEG continuously throughout the entire anesthetic infusion period
- Continue cEEG for at least 24–48 hours after weaning of the anesthetic to detect seizure recurrence
- Document the time to burst suppression after each bolus and during infusion titration
- Monitor for breakthrough seizures, which may manifest as rhythmic activity during the burst phase or loss of suppression periods
- Use quantitative EEG (qEEG) trends to facilitate overnight monitoring and rapid detection of changes
Withdrawal Protocol from Anesthetic Infusions
Weaning of anesthetic infusions is one of the most challenging aspects of RSE management. Premature weaning leads to seizure recurrence, while prolonged infusions increase the risk of systemic complications.
Standard Weaning Approach
- Duration of burst suppression: Maintain burst suppression for 24–48 hours after the last electrographic seizure (some experts advocate 48–72 hours for particularly severe RSE)
- Optimize background ASMs: Before attempting to wean, ensure that adequate non-sedating ASM levels are established (typically 2–3 agents at therapeutic doses) to provide a "safety net"
- Gradual taper: Reduce the anesthetic infusion by 10–25% every 6–12 hours, with continuous EEG monitoring during each step-down
- Monitor for recurrence: If seizures recur during weaning, return to the previous effective dose, maintain for another 24–48 hours, optimize concurrent ASMs, and attempt weaning again
- Repeated weaning failures: If seizures recur on multiple weaning attempts, this constitutes SRSE; consider alternative strategies (see below)
Practical Tips for Anesthetic Weaning
- Consider adding a non-sedating ASM before each weaning attempt (e.g., add lacosamide before weaning attempt 2 if only levetiracetam and valproate were on board)
- Review the total number of concurrent ASMs; studies report a median of 5 ASMs used in NORSE/SRSE patients
- Avoid weaning during physiologic vulnerability periods (e.g., active infection, metabolic instability)
- If switching from one anesthetic to another (e.g., midazolam to pentobarbital), cross-titrate rather than abruptly stopping the first agent
- Some centers use a "sentinel dose reduction" approach — reduce the infusion by 50% to see if seizures recur within 6 hours before proceeding with full weaning
- Prolonged use of pentobarbital may require days to weeks for complete clearance; neurologic assessment may be limited during this period
NORSE and FIRES
New-onset refractory status epilepticus (NORSE) and its subcategory, febrile infection-related epilepsy syndrome (FIRES), represent the most distinctive and challenging causes of SRSE. They are defined as clinical presentations, not specific diagnoses, and require an aggressive diagnostic and therapeutic approach.
Definitions
- NORSE: New onset of RSE in a patient without active epilepsy or other preexisting relevant neurologic disorders, without a clear acute or active structural, toxic, or metabolic cause. It is a clinical presentation, not a final diagnosis.
- FIRES: A subcategory of NORSE applicable to patients of all ages, requiring a prior febrile infection starting between 2 weeks and 24 hours before the onset of RSE, with or without fever at the onset of SE.
Epidemiology and Etiologies
| Feature | NORSE | FIRES |
|---|---|---|
| Age distribution | Children and adults; mean age ~40 years in adult series | Predominantly children (peak 3–10 years); can occur in adults |
| Proportion of RSE | ~10–16% of adult RSE; ~16% of pediatric RSE | Subset of NORSE |
| Most common etiology | Cryptogenic (49.9%); autoimmune encephalitis (36.2%); paraneoplastic; viral; other | Presumed post-infectious immune-mediated; elevated proinflammatory cytokines in serum |
| CSF findings | Nonspecific mild pleocytosis and mildly elevated protein; send autoimmune encephalopathy panel | Often mild pleocytosis; elevated cytokines (IL-6, CXCL13, CCL2) |
| MRI | May be normal initially; may develop mesial temporal or cortical signal changes; significantly higher BBB permeability compared with controls | Often normal initially; may develop bilateral mesial temporal T2/FLAIR hyperintensity |
| Mortality | 19–22% (2023–2024 systematic reviews) | 10–15% in children; higher in adults |
| Full recovery | Only 4% of patients | ~10–15% of children |
| Long-term outcome | High rates of drug-resistant epilepsy, cognitive disability, and psychiatric morbidity | Severe drug-resistant epilepsy; cognitive and behavioral impairment |
Diagnostic Workup for NORSE/FIRES
- CSF analysis: Cell count, protein, glucose, oligoclonal bands, cytology, cultures, HSV PCR, autoimmune encephalitis antibody panel (NMDA-R, LGI1, CASPR2, GABA-B, AMPA, DPPX)
- Serum: Comprehensive autoimmune encephalitis panel, paraneoplastic antibody panel (ANNA-1/Hu, ANNA-2/Ri, CV2/CRMP5, amphiphysin), complement, cytokines if available
- MRI brain with contrast: Repeat at 48–72 hours if initial is normal; look for mesial temporal, insular, or cortical signal changes
- CT chest/abdomen/pelvis: To evaluate for occult malignancy (particularly ovarian teratoma in anti-NMDA-R encephalitis)
- Body PET/CT: If paraneoplastic syndrome suspected and CT is unrevealing; repeat periodically (tumors may emerge months to years later)
- Genetic testing: Consider in children without identified autoimmune etiology (SCN1A, PCDH19, POLG, and other epileptic encephalopathy genes)
- Important: Do not delay empiric immunotherapy while awaiting antibody results — autoimmune panels may take 2–4 weeks to return
Immunotherapy for NORSE/FIRES and Cryptogenic SRSE
International consensus-based recommendations for NORSE and FIRES (published 2022) provide 85 treatment recommendations. A central principle is the early initiation of immunotherapy based on clinical suspicion, without waiting for confirmatory antibody results.
Immunotherapy Protocol
| Line | Timing | Agents | Details |
|---|---|---|---|
| First-line | Within 72 hours of SE onset | IV methylprednisolone ± IVIg | Methylprednisolone 20–30 mg/kg/day (max 1 g) for 3–5 days; IVIg 0.4–2 g/kg/day for 3–5 days; may use PLEX as alternative to IVIg |
| Second-line | Within 7 days if first-line fails | Rituximab or cyclophosphamide | If a pathogenic antibody is identified or highly suspected, rituximab is recommended; for cryptogenic NORSE, consider IL-1 receptor antagonist (anakinra) or IL-6 antagonist (tocilizumab) |
| Maintenance | Postacute phase | Continue effective immunotherapy ≥3 months | No specific ASM is recommended for the postacute phase; if immunomodulation was effective acutely, it should continue |
Evidence Supporting Immunotherapy in NORSE/FIRES
- A multicenter study comparing NORSE/FIRES patients with other RSE etiologies and controls showed significantly elevated proinflammatory cytokines and chemokines in the serum of NORSE patients, supporting an inflammatory pathophysiology
- Brain MRI studies demonstrated significantly higher blood-brain barrier permeability in adults with NORSE compared with encephalitis patients without SE and healthy controls
- In a detailed case of ANNA-1 encephalitis presenting as NORSE, methylprednisolone improved cognition, aphasia, and seizures; subsequent cyclophosphamide, rituximab, and mycophenolate were required for long-term control
- No randomized controlled trials on NORSE treatment exist; all recommendations are based on expert consensus, case series, and pathophysiologic rationale
- Anakinra (IL-1 receptor antagonist) has shown promising results in case series for FIRES, particularly in the pediatric population
- Tocilizumab (IL-6 antagonist) has been used successfully in refractory cases
Adjunctive Therapies for SRSE
When standard anesthetic infusions and immunotherapy fail to control SRSE, several adjunctive therapies may be considered. Evidence for most of these is limited to case series, retrospective studies, and expert opinion.
Ketamine
- Mechanism: NMDA receptor antagonist; addresses the pathophysiologic shift toward NMDA receptor upregulation that occurs during prolonged SE
- Dose: Loading dose 1–2.5 mg/kg IV; maintenance infusion 3–10 mg/kg/h, titrated to burst suppression on cEEG
- Evidence: A 2024 systematic review and meta-analysis of ketamine in pediatric RSE reported efficacy in approximately 75% of cases in a university hospital series. A 2023 systematic review of nonanoxic SRSE found that ketamine use was not associated with improved seizure cessation or in-hospital survival, though the studies were highly heterogeneous.
- Advantages: Favorable hemodynamic profile (does not cause hypotension; may cause hypertension via sympathomimetic effects); acts on a different receptor target than GABA-ergic agents; can be combined with benzodiazepines
- Cautions: Must combine with a benzodiazepine (to avoid psychotomimetic effects and neurotoxicity); avoid in neonates and third trimester of pregnancy; monitor for hypertension, increased intracranial pressure, and emergence phenomena
- Half-life: 2.5 hours
Ketogenic Diet
- Mechanism: High-fat, very low-carbohydrate diet that induces ketosis; modulates GABA metabolism, reduces glutamatergic excitation, provides alternative energy substrates, and has anti-inflammatory effects
- Evidence: A 2023 review reported that the ketogenic diet was considered effective in 197/276 (71.4%) patients with SRSE after a median of 6.5 days from initiation. International consensus recommendations include ketogenic diet consideration within 7 days of NORSE/FIRES onset.
- Protocol: Initiate enteral ketogenic formula (4:1 or 3:1 fat-to-carbohydrate+protein ratio); restrict IV dextrose-containing fluids; monitor serum and urine ketones (target beta-hydroxybutyrate 2–5 mmol/L); switch all medications to non-carbohydrate formulations
- Contraindications: Fatty acid oxidation disorders, carnitine deficiency, pyruvate carboxylase deficiency, porphyria
- Monitoring: Daily serum beta-hydroxybutyrate, glucose, lipids, hepatic function, carnitine; monitor for hypoglycemia, metabolic acidosis, hypertriglyceridemia, GI intolerance
- Practical challenges: Requires dietitian expertise; many IV medications and propofol (lipid emulsion) interfere with ketosis; coordination with pharmacy is essential to eliminate hidden carbohydrate sources
Therapeutic Hypothermia
- Mechanism: Reduces cerebral metabolic rate, decreases excitotoxic neurotransmitter release, attenuates neuroinflammation, stabilizes the blood-brain barrier
- Evidence: A 2021 AES review determined that therapeutic hypothermia was possibly effective in adults with SRSE, but with insufficient evidence in neonates with hypoxic-ischemic encephalopathy (for SE specifically)
- Protocol: Target temperature 32–35°C for 24–48 hours using surface or intravascular cooling devices; monitor for complications during rewarming (seizure recurrence, electrolyte shifts)
- Risks: Coagulopathy, arrhythmias, immunosuppression, infection, electrolyte derangements, skin complications
- Practical role: Generally considered a rescue therapy after failure of multiple anesthetic agents and immunotherapy; limited centers have expertise
Other Adjunctive Therapies
| Therapy | Mechanism/Rationale | Evidence Level | Key Notes |
|---|---|---|---|
| Magnesium sulfate | NMDA receptor antagonist; membrane stabilizer | Insufficient evidence for RSE; first-line for eclampsia | Loading dose 50 mg/kg (max 4 g); infusion 20–40 mg/kg/h; monitor magnesium levels |
| Lacosamide | Enhanced slow inactivation of sodium channels | Possibly effective (AES review 2020) | 5–10 mg/kg IV over 15–30 min; monitor for cardiac arrhythmias (atrial, ventricular); ECG monitoring recommended |
| Brivaracetam | SV2A binding (higher affinity than levetiracetam) | Unknown/insufficient evidence | 100 mg IV twice daily (adult); faster onset than LEV; limited RSE-specific data |
| Inhaled anesthetics (isoflurane, desflurane) | General anesthesia; potent GABAA agonism | Insufficient evidence | Requires anesthesia machine in ICU; seizure recurrence on discontinuation; practical and logistic challenges |
| Vagus nerve stimulation (VNS) | Neuromodulation; anti-seizure and anti-inflammatory effects | Insufficient evidence for SRSE | Case reports of benefit; requires surgical implantation (may not be feasible in acute setting) |
| Electroconvulsive therapy (ECT) | Induces a controlled seizure followed by postictal suppression; may "reset" network excitability | Case reports only | Recent case report (2025) of successful treatment of cryptogenic NORSE resistant to immunosuppression with IV ganaxolone and ECT |
| Perampanel | AMPA receptor antagonist | Insufficient evidence | Only available orally or via NGT; theoretical benefit from targeting glutamatergic excitation |
| Pyridoxine | Cofactor for GAD (GABA synthesis); diagnostic and therapeutic for pyridoxine-dependent epilepsy | First-line in neonatal SE/SRSE | 100 mg IV every 5 minutes up to 500 mg; maintenance 15–30 mg/kg/day |
Special Situation: Medications for Infancy, Pregnancy, and SRSE
Certain clinical scenarios in SRSE require specific pharmacologic considerations:
- ACTH (adrenocorticotropic hormone): Used in infantile-onset SRSE; 40–80 units IM or SC for children >2 years; 150 units/m2 for children <2 years; t1/2 0.25 hours (IV)
- IVIg: 0.4–2 g/kg/day for 3–5 days; used in immune-mediated SRSE (NORSE, FIRES, autoimmune encephalitis)
- IV methylprednisolone: 10–30 mg/kg/day for 3–5 days (max 1 g/day); sometimes followed by oral prednisone 1 mg/kg/day
- Magnesium sulfate: First-line treatment for eclamptic seizures; also used as adjunctive therapy in SRSE
- Pyridoxine: Essential in neonatal SE to rule out pyridoxine-dependent epilepsy; also used for isoniazid-induced SE
Long-Term Outcomes and Prognostication
Outcomes Data
A 2023 systematic review identified 266 individual patients with nonanoxic SRSE available for meta-analysis. Key findings:
- Mean SRSE duration: 36.3 days
- Mean age: 40.8 years
- Etiologies: Acute cerebral events (41.6%), unknown (22.3%), other identified causes
- In-hospital mortality: 24.1%
- Zero to slight disability at discharge: Only 26.8%
- Mortality vs. duration: Mortality increased with SRSE duration up to 28 days
- Successful seizure termination: Continued to increase with longer treatment, but the proportion of patients with substantial disability also increased significantly
- Reported treatment with ketamine, phenobarbital, barbiturates, VNS, and ketogenic diet were not associated with improved seizure cessation or in-hospital survival (though this may reflect selection bias — these treatments were likely used in the most refractory cases)
NORSE-Specific Outcomes (2024 Data)
- Mortality: 19% (consistent with earlier systematic reviews)
- Full neurologic recovery: Only 4%
- High rates of subsequent active drug-resistant epilepsy
- Significant vocational, cognitive, and psychiatric disabilities among survivors
- New treatment-resistant epilepsy occurred in 0.8% of all SE patients in the Korean nationwide study
Prognostic Factors in RSE and SRSE
| Factor | Favorable Prognosis | Unfavorable Prognosis |
|---|---|---|
| Etiology | ASM non-adherence, drug-related, autoimmune (if treated early) | Anoxic brain injury, CNS infection, malignancy, cryptogenic NORSE without treatment response |
| Duration | Shorter RSE/SRSE duration | SRSE duration >28 days |
| STESS | Score 0–2 | Score ≥3 |
| Age | Younger (<65 years) | Older (≥65 years) |
| MRI | Normal or resolving peri-ictal changes | Diffuse cortical, thalamic, or hippocampal injury; progressive changes |
| Treatment response | Response to first or second anesthetic agent; response to immunotherapy | Multiple anesthetic failures; lack of response to immunotherapy |
| Complications | Few systemic complications | Infection, multiorgan failure, need for prolonged vasopressors |
Prognostication Caution in SRSE
- Avoid premature prognostication: SRSE with treatable etiologies (autoimmune, infectious) may have dramatically better outcomes than predicted by early clinical findings
- Anesthetic agents confound neurologic examination: Pentobarbital may take days to weeks to clear; neurologic assessment during or shortly after high-dose anesthetics is unreliable
- Delayed recovery is possible: Some NORSE/FIRES patients who appear to have devastating outcomes in the acute phase show gradual improvement over months to years
- Goals-of-care discussions: Should be ongoing, transparent, and incorporate the uncertainty inherent in SRSE prognostication; involve palliative care early for support
- Exception — post-anoxic SRSE: Carries a consistently dismal prognosis; the TELSTAR trial showed no benefit from aggressive EEG-guided ASM treatment in post-cardiac arrest patients
ICU Management Considerations
Patients with RSE and SRSE require comprehensive ICU care that extends well beyond seizure management. Common complications include:
- Infections: Hospital-acquired pneumonia (especially ventilator-associated), urinary tract infections, central line-associated bloodstream infections; immunosuppression from barbiturates and corticosteroids increases risk; obtain surveillance cultures regularly
- Hemodynamic instability: Hypotension from anesthetic infusions (especially pentobarbital) often requires vasopressors; invasive arterial and central venous monitoring
- Thromboembolism: Pharmacologic DVT prophylaxis and sequential compression devices are essential in all immobilized patients
- Nutrition: Early enteral nutrition when feasible; account for propofol lipid calories; ketogenic diet formulation if indicated; avoid dextrose-containing IV fluids if on ketogenic diet
- Skin integrity: Prolonged immobility and barbiturate-induced vasodilation increase pressure injury risk
- Critical illness myopathy/neuropathy: Prolonged ICU stay and neuromuscular blockade (if used) increase risk; minimize paralytic use; begin physical therapy when hemodynamically stable
Summary Algorithm: RSE and SRSE Management
| Phase | Action | Key Details |
|---|---|---|
| 1. Confirm RSE | Verify that 2 appropriate ASMs at adequate doses have been given and failed | Check actual doses administered against guideline recommendations; correct underdosing before escalating |
| 2. Consider non-anesthetic ASM | A second non-sedating IV ASM (lacosamide, brivaracetam, or unused second-line agent) | May terminate RSE in up to 50% of cases; avoids anesthetic complications |
| 3. Anesthetic infusion | Intubate; initiate midazolam, propofol, or pentobarbital; start cEEG | Target burst suppression; load concurrent non-sedating ASMs; avoid PRIS with propofol |
| 4. Identify and treat cause | Full etiologic workup: imaging, labs, CSF, autoimmune panels, toxicology | Do not delay empiric immunotherapy in suspected NORSE/autoimmune encephalitis |
| 5. Immunotherapy (if NORSE/FIRES or cryptogenic) | IV methylprednisolone ± IVIg within 72 hours; second-line within 7 days | Rituximab if antibody identified; anakinra/tocilizumab for cryptogenic NORSE |
| 6. SRSE (>24h on anesthetics) | Switch or add anesthetic agent; consider ketamine, ketogenic diet, hypothermia | Continue aggressive etiologic workup; repeat MRI; goals-of-care discussion |
| 7. Wean anesthetics | After 24–48h of burst suppression; gradual 10–25% reduction q6–12h | Ensure adequate non-sedating ASM coverage; resume if seizures recur; patience is key |
References
- Vossler DG, Bainbridge JL, Boggs JG, et al. Treatment of refractory convulsive status epilepticus: a comprehensive review by the American Epilepsy Society Treatments Committee. Epilepsy Curr. 2020;20(5):245-264.
- 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;56(10):1515-1523.
- Shorvon S, Ferlisi M. The treatment of super-refractory status epilepticus: a critical review of available therapies and a clinical treatment protocol. Brain. 2011;134(Pt 10):2802-2818.
- Gaspard N, Hirsch LJ, Bhatt SM, et al. New-onset refractory status epilepticus (NORSE) and febrile infection-related epilepsy syndrome (FIRES): state of the art and perspectives. Epilepsia. 2018;59(4):745-752.
- 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;59(4):739-744.
- Wickstrom R, Bhatt SM, Gaspard N, et al. International consensus recommendations for management of new onset refractory status epilepticus (NORSE) including febrile infection-related epilepsy syndrome (FIRES): summary and clinical tools. Epilepsia. 2022;63(11):2827-2839.
- Vossler DG. First seizures, acute repetitive seizures, and status epilepticus. Continuum (Minneap Minn). 2025;31(1, Epilepsy):93-128.
- Glauser T, Shinnar S, Gloss D, et al. Evidence-based guideline: treatment of convulsive status epilepticus in children and adults. Epilepsy Curr. 2016;16(1):48-61.
- Fung FW, Jacobwitz M, Vala L, et al. Electroencephalographic seizures in critically ill children: management and adverse events. Epilepsia. 2019;60(10):2095-2104.
- Rossetti AO, Logroscino G, Milligan TA, et al. Status Epilepticus Severity Score (STESS): a tool to orient early treatment strategy. J Neurol. 2008;255(10):1561-1566.
- Strzelczyk A, Ansorge S, Grollmuss O, et al. Costs, length of stay, and mortality of super-refractory status epilepticus: a population-based study from Germany. Epilepsia. 2017;58(9):1533-1541.
- Ruijter BJ, Keijzer HM, Tjepkema-Cloostermans MC, et al. Treating rhythmic and periodic EEG patterns in comatose survivors of cardiac arrest (TELSTAR). N Engl J Med. 2023;388(10):906-916.
- Samanta D, Garrity L, Bhatt S, et al. International consensus recommendations for management of NORSE including FIRES: methodology and definitions. Epilepsia. 2022;63(11):2812-2826.
- Sculier C, Gainez M, Bhatt S, et al. Treatment of NORSE/FIRES: a systematic review and meta-analysis. Epilepsia. 2023;64(9):2303-2318.
- Kang BS, Kim DW, Kim KK, et al. Nationwide study on new-onset status epilepticus in Korea. Neurology. 2025.