Status Epilepticus
Status Epilepticus
What Do You Need to Know?
- Definition: continuous seizure ≥5 min (t1) or recurrent seizures without recovery; neuronal injury risk at 30 min (t2)
- First-line: BZDs — IM midazolam 10 mg (RAMPART: 73.4% vs. 63.4%, superior to IV lorazepam); underdosing in >75%
- Second-line: ESETT — fosphenytoin 20 mg PE/kg, VPA 40 mg/kg, LEV 60 mg/kg — ALL equivalent (~46–47%)
- VA Cooperative: stepwise efficacy decline 55.5% → 7.0% → 2.3%
- Refractory SE: failed 2 agents → anesthetics; target burst suppression on cEEG
- NCSE: Salzburg criteria — EDs >2.5 Hz = NCSE; ≤2.5 Hz needs secondary criteria; sensitivity 97.7%
- NORSE/FIRES: immunotherapy ≤72 h; ketogenic diet ≤7 days; anakinra: SE cessation 50–60%; mortality 16–22%
Definition & Stages
ILAE 2015 Operational Time Points
- t1 = 5 min: seizure regarded as continuous → initiate treatment
- t2 = 30 min: risk of long-term neuronal injury, hippocampal damage, network alteration
- Most self-terminating tonic-clonic seizures end within 2–3 min
Stages of Status Epilepticus
| Stage | Time Frame | Definition | Treatment Phase |
|---|---|---|---|
| Developing SE | 0–5 min | Not yet meeting SE; most seizures self-terminate | Stabilization (ABCs, glucose, IV access) |
| Established SE | 5–30 min | Ongoing ≥5 min or recurrent without recovery | 1st line: BZDs; 2nd line: IV ASMs |
| Refractory SE | >30–40 min | Failed 2 adequately dosed ASMs from different classes | Anesthetics or additional non-sedating ASM |
| Super-Refractory SE | ≥24 h on anesthetics | Persists/recurs despite ≥24 h continuous anesthetic | Multimodal ICU; immunotherapy if NORSE |
Epidemiology
- Incidence: 10–41 per 100,000/year; bimodal: highest in children (<1 yr) and elderly (>60 yr)
- ~20% of patients present with SE as their first-ever seizure
- 30-day mortality: ~10% adults, ~2% children
- RSE develops in 23–55% of SE patients; SRSE in ~10–15% of all SE
- Etiology is the strongest predictor of outcome (acute symptomatic = worst prognosis)
💎 Board Pearl
- >75% of SE patients receive subtherapeutic BZDs — apparent refractoriness may be iatrogenic underdosing; confirm doses before escalating
Convulsive SE Treatment Algorithm
0–5 min: Stabilization
- Recovery position; suction; O2 via non-rebreather; do NOT place objects in mouth
- 2 large-bore IVs; POC glucose, BMP, CBC, ASM levels, toxicology
- D50W 25–50 mL if hypoglycemia; thiamine 100 mg IV first if malnourished/alcohol
- Continuous cardiac telemetry + pulse oximetry; note exact seizure onset time
- Prepare for cEEG as soon as available
5–20 min: First-Line — Benzodiazepines
| Agent | Route | Adult Dose (>40 kg) | Onset | Repeat |
|---|---|---|---|---|
| Midazolam | IM | 10 mg | 3–5 min | Single dose |
| Lorazepam | IV | 4 mg (0.1 mg/kg) | 2–3 min | Yes, ×1 |
| Diazepam | IV / PR | 10 mg IV; 0.2 mg/kg PR | 1–2 min | Yes, ×1 |
RAMPART Trial (2012)
- 893 prehospital SE patients: IM midazolam 10 mg vs. IV lorazepam 4 mg
- Noninferior AND superior: 73.4% vs. 63.4% seizure cessation (p < 0.001)
- Advantage = faster drug delivery — IM avoids IV access delays in seizing patients
- IM midazolam = preferred first-line when IV access not immediately available
BZD Underdosing — The #1 Treatment Error
- Subtherapeutic BZD dosing in >75% of SE patients (ESETT cohort, SENSE registry)
- Underdosing → higher rates of refractory SE, intubation, and death
- PHTSE trial: placebo group had >2× respiratory dysfunction vs. lorazepam/diazepam
- Uncontrolled SE is more dangerous than BZDs — use full guideline-recommended doses
20–40 min: Second-Line — Non-BZD ASMs
| Agent | Loading Dose | Max | Key Considerations |
|---|---|---|---|
| Fosphenytoin | 20 mg PE/kg | 1500 mg PE | Cardiac monitor; hypotension/arrhythmia; avoid in conduction disorders |
| Valproic acid | 40 mg/kg | 3000 mg | Avoid in pregnancy, mito disease, hepatic failure; best hemodynamics |
| Levetiracetam | 60 mg/kg | 4500 mg | Safest profile; no cardiac/hepatic toxicity; no drug interactions |
ESETT Trial (2019)
- 384 patients aged 2–95 with BZD-resistant CSE
- Seizure cessation + improved consciousness at 60 min: LEV 47%, fosphenytoin 45%, VPA 46%
- No significant difference — all three equivalent; choose based on patient factors
VA Cooperative Trial (1998)
- Stepwise efficacy decline: 1st ASM 55.5% → 2nd ASM 7.0% → 3rd ASM 2.3%
- Fundamental rationale for aggressive, protocol-driven, early treatment
>40 min: Third-Line Options
- Option A: additional non-sedating ASM (lacosamide, brivaracetam) — may terminate RSE in ~50%
- Option B: anesthetic infusion (midazolam, propofol, pentobarbital) — requires intubation + cEEG
- No Class I evidence for either strategy; individualize based on clinical severity
Pathophysiology of Pharmacoresistance
| Mechanism | Timeline | Consequence | Therapeutic Implication |
|---|---|---|---|
| GABAA receptor internalization | Minutes | Reduced synaptic GABAA; loss of BZD binding sites | Declining BZD efficacy — treat early, full dose |
| NMDA receptor externalization | Minutes–hours | Increased surface NMDA; Ca2+ influx → excitotoxicity | Rationale for ketamine in RSE/SRSE |
| GABAA subunit shift | Hours | α1/γ2 (BZD-sensitive) → α4/δ (BZD-insensitive) | Further BZD resistance; neurosteroid target |
| Neuropeptide depletion | Hours–days | Loss of NPY, galanin, dynorphin; accumulation of substance P | Excitatory-inhibitory balance shifts |
| BBB disruption | Days | Increased permeability; neuroinflammation; immune cell infiltration | Rationale for immunotherapy in SRSE |
Clinical Pearl
GABAA receptor internalization within minutes is the single most important mechanism explaining declining BZD efficacy. This creates a rapidly narrowing treatment window and is the fundamental rationale for immediate, full-dose BZD therapy. The VA Cooperative showed each successive treatment phase is dramatically less effective: 55.5% → 7.0% → 2.3%.
Nonconvulsive Status Epilepticus (NCSE)
Salzburg EEG Criteria for NCSE
- EEG abnormalities must be continuously present for ≥10 seconds
| Step | Criterion | Outcome |
|---|---|---|
| 1 | Epileptiform discharges (EDs) >2.5 Hz | = NCSE |
| 2 | EDs ≤2.5 Hz OR rhythmic delta/theta >0.5 Hz | Proceed to Step 3 |
| 3 | ≥1 secondary criterion: electroclinical correlation; spatial/temporal evolution (≥1 Hz change, morphology, location); IV ASM trial with BOTH EEG + clinical improvement | If met: NCSE; if not: possible NCSE |
Diagnostic Performance
- Sensitivity 97.7%, specificity 95.9%
- IV ASM trial must show BOTH EEG AND clinical improvement — EEG alone insufficient (BZDs suppress non-ictal patterns)
Key NCSE Facts
- NCSE found in 28% of elderly with unexplained delirium
- NCSE outnumbers CSE in registry data (SENSE: 592 vs. 457 patients)
- Median time to first treatment: 150 min for NCSE vs. 30 min for CSE — profound diagnostic delay
- BZD underdosing in 78% of NCSE patients (SENSE registry)
- 14–48% have ongoing NCSE after clinical cessation of convulsive SE ("subtle SE") — always get cEEG after CSE
- In-hospital mortality: 18–19%
- NCSE can mimic stroke (acute aphasia, hemiparesis) — "epileptic pseudostroke" in 1–3% of stroke presentations
💎 Board Pearl
- "Postictal" state >30–60 min after GTC → suspect NCSE, not prolonged postictal — order urgent EEG
- Salzburg Step 3 IV ASM trial: must show BOTH EEG + clinical improvement — EEG improvement alone does not confirm NCSE
Ictal-Interictal Continuum (IIC)
ACNS 2021 Critical Care EEG Terminology
| Pattern | Full Name | Clinical Significance | Treatment |
|---|---|---|---|
| LPDs | Lateralized periodic discharges (formerly PLEDs) | Acute structural lesions; seizures 40–60% | Treat if clinical change |
| GPDs | Generalized periodic discharges (includes triphasic waves) | Metabolic (hepatic, uremia) or post-anoxic | Treat cause first; ASM trial if doubt |
| LRDA | Lateralized rhythmic delta activity | Highest seizure risk (>60%); "pre-ictal" | Strong indication for ASM treatment |
| GRDA | Generalized rhythmic delta (includes FIRDA) | Usually non-ictal; diffuse encephalopathy | Does NOT need ASMs; treat encephalopathy |
"Plus" Features (+F, +R, +S)
- +F = superimposed fast activity; +R = rhythmic; +S = sharp/spike
- Plus features move ANY pattern toward the ictal end → lower threshold for treatment
TELSTAR Trial (2023)
- Post-cardiac arrest periodic/rhythmic patterns: aggressive ASM treatment vs. no treatment
- No difference in neurologic outcome — supports conservative approach for post-anoxic patterns
💎 Board Pearl
- LPDs = lateralized → think structural lesion; GPDs (includes triphasic) = generalized → think metabolic
- LRDA = highest seizure risk (>60%) of all IIC patterns — treat aggressively
- TELSTAR: no benefit treating post-cardiac arrest periodic patterns — prognosis = anoxic injury severity
Refractory & Super-Refractory SE
Anesthetic Agents for RSE
| Agent | Loading Dose | Maintenance | Key Risks |
|---|---|---|---|
| Midazolam | 0.2 mg/kg bolus | 0.1–2 mg/kg/h | Tachyphylaxis; least hemodynamic compromise of the three |
| Propofol | 2 mg/kg bolus | 1–5 mg/kg/h | PRIS (>5 mg/kg/h >48 h): metabolic acidosis, rhabdo, cardiac failure; avoid in children |
| Pentobarbital | 5 mg/kg bolus | 1–5 mg/kg/h | Most potent; profound hypotension (vasopressors), immunosuppression, ileus |
EEG Target & Weaning Protocol
- EEG target: burst suppression (interburst intervals 5–15 sec) for 24–48 h
- Wean: reduce by 10–25% every 6–12 h with continuous EEG monitoring
- Optimize 2–3 non-sedating background ASMs at therapeutic levels before weaning
- Seizures recur on wean → return to previous dose, maintain 24–48 h, add ASM, re-attempt
💎 Board Pearl
- PRIS: metabolic acidosis + rhabdomyolysis + cardiac failure; risk >5 mg/kg/h >48 h; monitor CK, lactate, triglycerides daily; Brugada-like ECG is a red flag
- Pentobarbital = most potent but worst hemodynamics; reserved for failure of midazolam/propofol
NORSE & FIRES
Definitions
- NORSE: new-onset RSE without active epilepsy or preexisting neurologic disorder, without clear acute cause — a clinical presentation, not a diagnosis
- FIRES: NORSE subtype requiring preceding febrile illness 2 weeks to 24 hours before SE onset
- Etiologies: cryptogenic ~50%, autoimmune ~36%, infectious ~5–10%
Pathophysiology & Key Features
- Cytokine storm: IL-1β, IL-6, TNF-α elevated in serum and CSF
- BBB permeability significantly higher than encephalitis without SE or controls
- Characteristic "wean-to-seize" pattern — seizures recur every time anesthetics are weaned
- Multifocal or migrating seizure onsets on EEG
- Median 5 ASMs used during acute phase
Treatment (International Consensus 2022)
| Timing | Intervention | Details |
|---|---|---|
| ≤72 h | 1st-line immunotherapy | IV methylprednisolone (max 1 g) ×3–5 d ± IVIg 0.4 g/kg/d ×5 d; PLEX as alt |
| ≤7 d | 2nd-line + ketogenic diet | Rituximab if antibody identified; ketogenic diet 4:1 ratio enterally |
| 7–14 d | Anti-cytokine therapy | Anakinra (IL-1RA) 100 mg SC bid; Tocilizumab (anti-IL-6) 8–12 mg/kg IV |
Targeted Therapies
- Anakinra (IL-1RA): SE cessation in 50–60% of FIRES patients; most promising targeted therapy
- Tocilizumab (anti-IL-6): SE termination after 1–2 doses; effective in anakinra-refractory cases
- Do NOT delay immunotherapy while awaiting antibody results (panels take 2–4 weeks)
Outcomes
- Mortality: 16–22%; full neurologic recovery: only 4%
- Chronic drug-resistant epilepsy in >90% of survivors
- Cognitive impairment >80%; psychiatric comorbidities 50–70%
- Mean SRSE duration in NORSE: 36 days
- Better outcomes: younger age, shorter SE duration, identified autoimmune etiology (treatable), early immunotherapy
- Worse outcomes: cryptogenic etiology, prolonged SRSE, delayed treatment, multiple anesthetic failures
STESS Score
Status Epilepticus Severity Score
- Pretreatment prognostic tool; total range 0–6
- Score ≤2 = favorable (high probability of survival); ≥3 = progressively worse
| Predictor | Features | Score |
|---|---|---|
| Consciousness | Alert / somnolent / confused | 0 |
| Stupor or coma | 1 | |
| Worst seizure type | Focal aware/impaired, absence, myoclonic | 0 |
| Generalized tonic-clonic | 1 | |
| Nonconvulsive SE in coma | 2 | |
| Age | <65 years | 0 |
| ≥65 years | 2 | |
| Prior seizure history | Yes (known epilepsy) | 0 |
| No or unknown | 1 |
💎 Board Pearl
- Highest single-item scores: age ≥65 (2 pts) and NCSE in coma (2 pts) — elderly comatose NCSE can score 6/6
Reversible Causes of SE
| Cause | Treatment | Key Detail |
|---|---|---|
| Hypoglycemia | D50W 25–50 mL IV | Check POC glucose immediately; always first |
| Hyponatremia | 3% NaCl 100 mL over 10 min (×2) | Severe (<120 mEq/L) can cause SE |
| ASM non-adherence | Stat drug levels; reload | Most common precipitant in known epilepsy |
| Isoniazid toxicity | Pyridoxine 5 g IV | Depletes pyridoxal phosphate → ↓GABA synthesis |
| Eclampsia | MgSO4 4–6 g IV | First-line, NOT standard ASMs |
| CNS infection | Empiric acyclovir + antibiotics | Start immediately; do not wait for LP results |
💎 Board Pearl
- Isoniazid-induced SE + pyridoxine 5 g IV is a classic board question — standard ASMs are ineffective without pyridoxine
- Eclampsia = MgSO4, NOT phenytoin or BZDs as first-line — another high-yield board favorite
- ASM non-adherence is the most common precipitant in known epilepsy and carries the best prognosis of all SE etiologies
Board Pearls & Clinical Pearls
💎 Board Pearl
- RAMPART: IM midazolam 10 mg SUPERIOR to IV lorazepam 4 mg (73.4% vs. 63.4%) — faster delivery, no IV needed
- ESETT: fosphenytoin, VPA, LEV all equivalent (~46–47%) — choose by patient factors, not drug superiority
- VA Cooperative: 55.5% → 7.0% → 2.3% stepwise efficacy decline = time is brain in SE
- BZD underdosing = #1 treatment error (>75%); uncontrolled SE more dangerous than BZD side effects
- Salzburg criteria: EDs >2.5 Hz = NCSE; ≤2.5 Hz needs secondary criteria; IV trial requires BOTH EEG + clinical improvement
- TELSTAR: aggressive treatment of post-cardiac arrest periodic patterns does NOT improve outcomes
- Isoniazid SE: pyridoxine 5 g IV is the specific antidote; Eclampsia: MgSO4, NOT standard ASMs
Clinical Pearl
"Wean-to-seize" pattern in NORSE/FIRES: seizures recur every time anesthetic infusions are weaned, often for weeks. This characteristic pattern should trigger aggressive immunotherapy workup and initiation of anakinra/tocilizumab plus ketogenic diet — continued anesthetic escalation alone will not resolve the underlying neuroinflammatory process.
Clinical Pearl
Treat the patient, not the EEG: in the ictal-interictal continuum, aggressiveness of treatment should match the clinical context. Suppressing an EEG pattern with sedating medications without clinical improvement is not a therapeutic success — particularly relevant in post-anoxic patients (TELSTAR).
Clinical Pearl
NCSE in the elderly: NCSE should be suspected in any elderly patient with unexplained acute confusion, especially if fluctuating. In the SENSE registry, median age of NCSE patients was 72 years, and increasing age independently predicted failure to receive BZDs. A missed NCSE diagnosis was associated with OR 3.83 for failure to return to functional baseline.
References
- Trinka E, Cock H, Hesdorffer D, et al. A definition and classification of status epilepticus — ILAE Task Force. Epilepsia. 2015;56(10):1515-1523.
- Silbergleit R, Durkalski V, Lowenstein D, et al. Intramuscular versus intravenous therapy for prehospital status epilepticus (RAMPART). N Engl J Med. 2012;366(7):591-600.
- Kapur J, Elm J, Chamberlain JM, et al. Randomized trial of three anticonvulsant medications for status epilepticus (ESETT). N Engl J Med. 2019;381(22):2103-2113.
- Treiman DM, Meyers PD, Walton NY, et al. A comparison of four treatments for generalized convulsive SE (VA Cooperative). N Engl J Med. 1998;339(12):792-798.
- Glauser T, Shinnar S, Gloss D, et al. Evidence-based guideline: treatment of convulsive SE. Epilepsy Curr. 2016;16(1):48-61.
- Leitinger M, Trinka E, Gardella E, et al. Diagnostic accuracy of the Salzburg EEG criteria for NCSE. Lancet Neurol. 2016;15(10):1054-1062.
- Hirsch LJ, Fong MWK, Leitinger M, et al. ACNS Standardized Critical Care EEG Terminology: 2021. J Clin Neurophysiol. 2021;38(1):1-29.
- Ruijter BJ, Keijzer HM, Tjepkema-Cloostermans MC, et al. TELSTAR trial. N Engl J Med. 2023;388(10):906-916.
- Vossler DG, Bainbridge JL, Boggs JG, et al. Treatment of refractory convulsive SE: AES comprehensive review. Epilepsy Curr. 2020;20(5):245-264.
- Hirsch LJ, Gaspard N, van Baalen A, et al. Consensus definitions for NORSE and FIRES. Epilepsia. 2018;59(4):739-744.
- Wickstrom R, Bhatt SM, Gaspard N, et al. International consensus for NORSE/FIRES management. Epilepsia. 2022;63(11):2827-2839.
- Kenney-Jung DL, Vezzani A, Bhatt D, et al. FIRES treated with anakinra. Ann Neurol. 2016;80(6):939-945.
- Rossetti AO, Logroscino G, Milligan TA, et al. Status Epilepticus Severity Score (STESS). J Neurol. 2008;255(10):1561-1566.
- Alldredge BK, Gelb AM, Isaacs SM, et al. Lorazepam, diazepam, and placebo for out-of-hospital SE (PHTSE). N Engl J Med. 2001;345(9):631-637.
- Vossler DG. First seizures, acute repetitive seizures, and status epilepticus. Continuum. 2025;31(1, Epilepsy):93-128.