Clinical Commander

All dossiers
neuro.status-epilepticus.core.v1

Convulsive Status Epilepticus

neurologyacuteadultpediatricacuteinpatient
Start encounter →Print handoutPRODUCTION

Phase B deepening (2026-05-14): 4-stage time-explicit ladder (stage_1_early 5–20 min → stage_2_established 20–40 min ESETT 3-drug → stage_3_refractory 40+ min cIV → stage_4_super_refractory >24 h) encoded as discrete severity_triggers so downstream reasoners can fire stage-specific action bundles. 5 setting playbooks span the full SE journey: home (RAMPART IM/IN/buccal midazolam field) → ed (door-to-benzo ≤10 min, door-to-AED ≤30 min, cEEG ≤60 min if AMS) → icu (cIV + cEEG burst-suppression + PRIS surveillance + NORSE/FIRES workup) → inpatient (anesthetic taper + maintenance AED transition + family education) → outpatient (AED maintenance + driving restriction + SUDEP + WCBA counselling). Special-population carve-outs encoded as discrete severity_triggers: eclampsia (MgSO4 supersedes ladder per ACOG 2020); INH-induced (pyridoxine 5 g IV or gram-per-gram); alcohol withdrawal (thiamine BEFORE glucose + benzo, no AED); PRES (BP control + remove offending agent); autoimmune NORSE/FIRES (methylprednisolone + IVIG + rituximab + IL-1/IL-6 blockade); PRIS suspicion (STOP propofol, rotate sedative). Schema-blocked calculators surfaced as tickets in docs/framework-audit/shard-3-neuro-sym-state.md: calc.stess (Status Epilepticus Severity Score; Rossetti 2006), calc.emse (Epidemiology-based Mortality in SE; Leitinger 2015), calc.mstess (Modified STESS), calc.sheiner_tozer_phenytoin (albumin-corrected free phenytoin). Currently encoded as plain-English required_assessments and severity_trigger descriptions until registry entries land. Evidence anchor refreshed twice. (1) Phase B 2026-05-14 removed wrong-engine PMIDs (DELIVER, IMPACT, ProMISe, POINT). (2) 2026-05-18 CL-3 PMID-correction pass: every dossier + ros-ddx-seed PMID PubMed-MCP-verified — 4 of the prior 6 dossier anchors were MIS-ATTRIBUTED to unrelated articles and are CORRECTED: RAMPART 22323081→22335736 (22323081 = Azoulay "ICU birthday" paper); pediatric ESETT/EcLIPSE 32178769→31005385 (32178769 = Remuzzi COVID-19 Italy; 31005385 = Lyttle EcLIPSE Lancet 2019); pediatric ESETT arm 32531194→32203691 (32531194 = VanderWall retinal-organoid; 32203691 = Chamberlain age-group ESETT Lancet 2020). Verified-correct: ESETT 31774955 (Kapur NEJM 2019), NCS Brophy 22528274, AES Glauser 26900382, ILAE Trinka 26336950. Added Vignatelli 38606469 (Epilepsia 2024 SE-CPG systematic review — guideline-delta, no paradigm change). Full PMID-correction table in src/lib/dossiers/neuro.status-epilepticus.core.v1._research-bundle.md. last_reconciled bumped to 2026-05-18. ESETT effect sizes wired into §5.5.1 rationale strings (LEV 47%/fos 45%/VPA 46% at 60 min; RAMPART IM-mdz 73.4% vs IV-lzp 63.4%; propofol PRIS dose/duration hazard). Cross-dossier routing added (all engine_ids verified to resolve to shipped dossiers): sibling_differentiation now routes to neuro.first-seizure-eval.v1 (post-SE first-seizure workup), neuro.encephalitis.hsv.v1 (febrile/immunocompromised SE → empiric aciclovir + LP), neuro.encephalitis-anti-nmdar.v1 (NORSE/super-refractory → autoimmune antibody panel + immunotherapy), neuro.ischaemic-stroke.v1 (persistent focal deficit not resolving as Todd paresis → STAT vascular imaging). Regimen axes encoded: 4-step se_4_stage_ladder (lorazepam/diazepam/IM midazolam per RAMPART → LEV/fosphenytoin/valproate per ESETT → midazolam/propofol/pentobarbital cIV per NCS 2012 → ketamine + immunotherapy super-refractory). All RxCUIs RxNav-validated. Cause-specific carve-outs (eclampsia MgSO4 / INH pyridoxine / alcohol thiamine-first) folded as axis-level drugs.

Entry points (4)

  • symptom
    Continuous seizure ≥5 min (ILAE 2015 t1 definition, Trinka et al)
    continuous_seizure_5min
  • symptom
    Recurrent seizures without recovery to baseline (ILAE 2015 operational definition)
    recurrent_seizures_no_recovery
  • symptom
    AMS with seizure history — suspect NCSE (NCS 2012 cEEG indication)
    altered_mental_status_with_seizure_history
  • imaging
    cEEG showing electrographic seizure activity (NCS 2012 monitoring criteria)
    continuous_eeg_pattern

Required inputs (16)

  • agerequired
    demographic • used at CONTEXT
    Pediatric vs adult dosing of benzodiazepines + AEDs (per ILAE 2020)
  • weightrequired
    demographic • used at TREATMENT
    Weight-based benzo + AED loading per AES 2016 dosing tables
  • seizure_onset_timerequired
    symptom • used at TREATMENT
    t1 (5 min) starts abortive Rx; t2 (30 min) marks neuronal injury risk (ILAE 2015 Trinka et al)
  • sbprequired
    vital • used at RED_FLAGS
    Sedative-induced hypotension during cIV anesthetic (NCS 2012 hemodynamic monitoring)
  • spo2required
    vital • used at RED_FLAGS
    Hypoxia from prolonged seizure / aspiration / sedation (AES 2016 airway assessment)
  • temperaturerequired
    vital • used at CONTEXT
    Fever as provoking factor — CNS infection, sepsis (NCS 2012 workup recommendation)
  • glucoserequired
    lab • used at RED_FLAGS
    Hypoglycaemia is reversible cause — give thiamine BEFORE glucose in alcoholics (AES 2016)
  • sodiumrequired
    lab • used at INITIAL_WORKUP
    Hypo/hypernatraemia provokes seizures; correction rate matters (NCS 2012 metabolic workup)
  • calciumrequired
    lab • used at INITIAL_WORKUP
    Hypocalcaemia provokes seizures; ionised preferred (NCS 2012 metabolic panel)
  • magnesiumrequired
    lab • used at INITIAL_WORKUP
    Hypomagnesaemia + eclampsia (give MgSO4) provoking factors (NCS 2012)
  • aed_levels
    lab • used at BRANCHING_WORKUP
    Phenytoin / valproate / phenobarbital / levetiracetam levels — albumin-correct phenytoin (NCS 2012 AED monitoring)
  • albumin
    lab • used at BRANCHING_WORKUP
    Free-phenytoin correction (Sheiner-Tozer equation) when hypoalbuminemic (NCS 2012)
  • ct_head_noncontrastrequired
    imaging • used at INITIAL_WORKUP
    Structural cause (ICH, mass, abscess); precedes LP if focal/AMS (AES 2016 workup)
  • eegrequired
    imaging • used at BRANCHING_WORKUP
    cEEG within 60 min if persistent AMS post-control; rule out NCSE (NCS 2012 cEEG recommendation)
  • pregnancy_eclampsia_screen
    history • used at RED_FLAGS
    Eclampsia → MgSO4 first-line; alters AED choice (ACOG 2020 eclampsia management)
  • inh_or_toxic_ingestion
    history • used at RED_FLAGS
    INH overdose → pyridoxine; cocaine/sympathomimetic; alcohol withdrawal (AES 2016 toxin-provoked SE)

12-phase flow (12)

  1. 1FRAME
    Convulsive SE per ILAE 2015 classification — stage 1 (5–30 min) abortive, stage 2 (30 min+) refractory, stage 3 (24h+ on cIV) super-refractory (Rossetti Lancet Neurol 2011)
    advance: SE confirmed and stage assigned per ILAE 2015 criteria
  2. 2ENTRY
    Activate stop-the-seizure clock per AES 2016 time-anchored treatment protocol
    inputs: age, weight, seizure_onset_time
    advance: time-zero anchored per AES 2016 protocol
  3. 3CONTEXT
    Capture AED history, alcohol/drug withdrawal, pregnancy, INH/toxin exposure, recent illness (NCS 2012 etiology workup)
    inputs: temperature, pregnancy_eclampsia_screen, inh_or_toxic_ingestion
    advance: trigger context captured per NCS 2012 etiology checklist
  4. 4RED_FLAGS
    ABCDE + thiamine + finger-stick glucose per AES 2016; eclampsia → MgSO4; INH → pyridoxine; airway protection
    inputs: sbp, spo2, glucose
    actions: status_epilepticus
    advance: reversible cause addressed + airway/IV access secured per AES 2016
  5. 5INITIAL_WORKUP
    BMP + Ca/Mg/PO4, glucose, ABG, CBC, LFT, tox screen, AED levels, pregnancy test, head CT non-contrast per NCS 2012 workup recommendations
    inputs: glucose, sodium, calcium, magnesium, ct_head_noncontrast
    advance: reversible metabolic + structural causes screened per NCS 2012
  6. 6BRANCHING_WORKUP
    cEEG within 60 min if AMS persists (NCS 2012 cEEG guideline); AED levels with albumin correction; MRI when stable
    inputs: eeg, aed_levels, albumin
    advance: NCSE ruled in or out per NCS 2012 cEEG criteria
  7. 7DIFFERENTIAL
    Provoked vs unprovoked per ILAE 2015 classification; convulsive vs non-convulsive vs focal SE; PNES exclusion
    advance: ILAE classification assigned
  8. 8RISK_STRATIFICATION
    STESS (Rossetti 2006) / EMSE / mSTESS for outcome prediction; refractory / super-refractory staging
    advance: stage + risk score documented per STESS (Rossetti 2006)
  9. 9TREATMENT
    4-stage AES/NCS ladder: (1) IV lorazepam per AES 2016 or IM midazolam per RAMPART (Silbergleit NEJM 2012); (2) LEV/fosphenytoin/VPA per ESETT (Kapur NEJM 2019); (3) cIV anesthetic per NCS 2012; (4) super-refractory per Rossetti Lancet Neurol 2011
    inputs: weight, seizure_onset_time, sbp
    advance: seizure terminated clinically and electrographically per NCS 2012 cEEG endpoint
  10. 10DISPOSITION
    ICU with cEEG per NCS 2012 monitoring recommendation; transfer if no neuro-ICU capability
    advance: ICU bed + cEEG available per NCS 2012 disposition
  11. 11MONITORING
    Continuous EEG until 24–48h seizure-free off cIV per NCS 2012; serial AED levels; vasopressor support during sedative wean
    inputs: sbp, spo2
    advance: monitoring plan documented per NCS 2012 cEEG protocol
  12. 12FOLLOWUP
    Long-term AED plan, driving counsel, epilepsy clinic referral per AAN/AES post-SE recommendations
    advance: epilepsy clinic + driving + AED tapering plan set per AAN/AES guidelines