Clinical Commander

All dossiers
ent.vestibular-neuritis.core.v1

Vestibular neuritis / acute labyrinthitis (acute vestibular syndrome, peripheral)

general_internal_medicineacutesubacuteadultgeriatricacuteoutpatientinpatient

NEUROTOLOGY/EMERGENCY-NEURO-OTOLOGY-framed engine — two inseparable jobs: positively diagnose PERIPHERAL acute vestibular syndrome (vestibular neuritis = no hearing loss; labyrinthitis = with SNHL) AND robustly exclude the can't-miss central mimic (~25% of AVS is posterior-circulation/AICA stroke). The peripheral-vs-central decision is framed explicitly via TiTrATE (timing/triggers, continuous-spontaneous AVS only) + HINTS. This engine is the RECIPROCAL routing target referenced by ent.bppv.core.v1 (continuous spontaneous AVS) and ent.sudden-sensorineural-hearing-loss.core.v1 (shared AVS+HINTS exam). Bidirectional: post-vestibular-neuritis is a recognised secondary-BPPV substrate (utricular otoconia shed) — new positional spells route to ent.bppv.core.v1; new acute SNHL routes to ent.sudden-sensorineural-hearing-loss.core.v1; central HINTS routes to neuro.posterior-circulation-stroke.core.v1. Definitive stroke/SSNHL/Ménière/migraine/falls management is recognised then routed OUT by engine_id — not re-authored here. RxCUI status: meclizine 6750, dimenhydrinate 3409, prochlorperazine 8704, ondansetron 26225, diazepam 3322 (strictly short-course ≤48-72 h symptomatic axis), prednisone 8640 (the DEBATED corticosteroid step) are well-established RxNorm CUIs. Antivirals are given an explicit NON-PHARM do-not-use guardrail entry (Strupp NEJM 2004 — valacyclovir ineffective; combination not superior) — rxcui intentionally OMITTED rather than fabricated (omit-rather-than-fabricate rule). Early vestibular rehabilitation + early mobilisation are non_pharm — the single evidence-based recovery lever is a therapy programme, not a drug. Bayesian linkage (pre-test peripheral-vs-central AVS prior by HINTS + vascular risk + hearing; per-component HINTS likelihood ratios — normal head-impulse + direction-changing nystagmus + skew = central; conditional dependence — HINTS valid ONLY in continuous AVS with nystagmus, NEVER in episodic BPPV, stated explicitly; imaging decision thresholds when HINTS central / cannot be applied / negative early DWI does not exclude; bidirectional cross-engine routing edges by engine_id with ent.bppv.core.v1, ent.sudden-sensorineural-hearing-loss.core.v1, neuro.posterior-circulation-stroke.core.v1, ent.meniere-disease.core.v1) is documented in the co-located _design-brief.md + _research-bundle.md; first-class TS LR fields remain schema-blocked (same constraint as ent.bppv.core.v1 / derm.cellulitis.core.v1). Effect sizes (≥5): AVS central proportion ~25% (~76/101 central in the index HINTS cohort; Kattah Stroke 2009, PMID 19762709); HINTS for stroke in AVS 100% sens / 96% spec, > early MRI-DWI (Kattah 2009); HINTS sens 96.5% / spec 84.4% / LR+ 6.19 / LR− 0.04 vs ABCD2 sens 61% / spec 62% / LR+ 1.62 / LR− 0.62 (Newman-Toker AEM 2013, PMID 24127701); early MRI-DWI false-negative ~12-15% (<48 h) and ~50-53% for small strokes (Kattah 2009; Saber Tehrani Neurology 2014, PMID 24920847); methylprednisolone caloric-recovery 62.4 vs placebo 39.6 percentage points, P<0.001; valacyclovir no effect P=0.43 (Strupp NEJM 2004, PMID 15269315); corticosteroid 1-mo complete caloric recovery RR 2.81 (95% CI 1.32-6.00) but 12-mo RR 1.58 (95% CI 0.45-5.62, NS) and no symptomatic benefit (Fishman Cochrane 2011, PMID 21563170); vestibular rehabilitation dizziness-frequency OR 2.67 (95% CI 1.85-3.86) and DHI SMD −0.83 (95% CI −1.02 to −0.64) (McDonnell & Hillier Cochrane 2015, PMID 25581507). Clinical-uncertainty notes: (1) the corticosteroid step is genuinely DEBATED equipoise — Strupp NEJM 2004 shows a caloric (objective) benefit but Cochrane (Fishman 2011) and Leong OHNS 2021 / Kim Clin Otolaryngol 2021 find NO symptomatic or long-term benefit and "insufficient evidence"; encoded as comorbidity_specific/shared-decision, NOT first_line. (2) HINTS operating characteristics derive from high-risk AVS cohorts examined by neuro-otology specialists; real-world non-specialist ED performance is lower — encoded as the "cannot reliably apply HINTS → image/consult" guardrail rather than over-claimed. (3) HINTS is valid ONLY in continuous AVS with spontaneous nystagmus, NEVER in triggered-episodic BPPV (where Dix-Hallpike governs) — conditional dependence stated explicitly and gated by the TiTrATE timing pivot. (4) ICD-10 H81.2x (vestibular neuronitis), H83.0x (labyrinthitis), H83.01-03 (laterality), R42 dizziness, H81.4 central vertigo, G45.0 vertebrobasilar TIA, I63.9 cerebral infarction are standard ICD-10-CM; per memory the terminology pipeline is ⚠ partial — codes are canonical but flagged for scripts/terminology/ revalidation. (5) Antivirals are explicitly NOT recommended (Strupp NEJM 2004) — encoded as a do-not-use guardrail, not merely omitted.

Entry points (5)

  • symptom
    Acute or subacute SUSTAINED spinning vertigo lasting ≥24 h with nausea/vomiting, head-motion intolerance and gait unsteadiness — acute vestibular syndrome (AVS); apply HINTS first, NOT positional testing (Bárány AUVP criteria, Strupp J Vestib Res 2022; Newman-Toker Neurol Clin 2015 TiTrATE)
    acute_continuous_spontaneous_vertigo_avs
  • symptom
    Spontaneous unidirectional horizontal-torsional nystagmus, direction-fixed, suppressed by visual fixation and enhanced by Frenzel/cover removal — the leading peripheral-AVS sign (Bárány AUVP criteria, Strupp J Vestib Res 2022; Kattah Stroke 2009)
    spontaneous_unidirectional_nystagmus
  • symptom
    Acute prolonged vertigo PLUS new unilateral sensorineural hearing loss / tinnitus / aural fullness — labyrinthitis (cochleovestibular); HINTS-plus positive — raises AICA-stroke risk and routes to the SSNHL pathway (Newman-Toker AEM 2013 HINTS-plus; Saber Tehrani Neurology 2014)
    avs_with_acute_hearing_loss_labyrinthitis
  • symptom
    Acute prolonged vertigo WITH any central feature (the 4 D's — diplopia, dysarthria, dysphagia, dysmetria; severe truncal/gait ataxia; direction-changing/vertical nystagmus; skew; new severe headache/neck pain) — posterior-circulation-stroke entry (Kattah Stroke 2009; Saber Tehrani Neurology 2014)
    avs_with_neuro_red_flags
  • history
    Recent viral prodrome / antecedent URI, OR prior vestibular neuritis with new positional spells (post-neuritis secondary BPPV) — etiologic/recurrent-substrate entry (Strupp NEJM 2004; Bárány AUVP criteria 2022)
    post_viral_or_post_neuritis_recurrent

Required inputs (15)

  • symptom_timing_continuous_vs_episodicrequired
    symptom • used at ENTRY
    TiTrATE pivot 1 — continuous sustained vertigo ≥24 h (AVS) is this engine's channel where HINTS governs; brief recurrent position-triggered spells are BPPV (route to ent.bppv.core.v1, Dix-Hallpike not HINTS) (Newman-Toker Neurol Clin 2015 TiTrATE; Bárány AUVP 2022)
  • spontaneous_vs_triggered_onsetrequired
    symptom • used at ENTRY
    AVS is SPONTANEOUS and continuous (no obligate positional trigger); an obligate head-position trigger with <1-min fatiguing spells redirects to BPPV — this gates which exam (HINTS vs Dix-Hallpike) is even valid (Bárány AUVP criteria 2022; Newman-Toker Neurol Clin 2015)
  • nystagmus_character_unidirectional_vs_direction_changingrequired
    symptom • used at CONTEXT
    Unidirectional horizontal-torsional, fixation-suppressed, direction-FIXED nystagmus = peripheral; direction-CHANGING gaze-evoked, pure-vertical or pure-torsional = central (Kattah Stroke 2009; Bárány AUVP criteria 2022)
  • hints_head_impulserequired
    symptom • used at RED_FLAGS
    In AVS an ABNORMAL (positive, corrective catch-up saccade) horizontal head-impulse toward the affected ear SUPPORTS peripheral vestibular neuritis; a NORMAL head-impulse with spontaneous nystagmus is the single most ominous central sign (Kattah Stroke 2009 — HINTS 100% sens / 96% spec)
  • hints_nystagmus_directionrequired
    symptom • used at RED_FLAGS
    Direction-CHANGING gaze-evoked / pure-vertical / pure-torsional nystagmus = central; unidirectional fixation-suppressible horizontal-torsional = peripheral (Kattah Stroke 2009; Newman-Toker AEM 2013)
  • hints_test_of_skewrequired
    symptom • used at RED_FLAGS
    Vertical ocular misalignment on alternate cover (skew) = brainstem/central; present in ~17% and rescues a false-localising abnormal head-impulse (lateral pontine stroke can mimic peripheral HIT) (Kattah Stroke 2009 — skew predicts brainstem)
  • central_neuro_4Ds_gaitrequired
    symptom • used at RED_FLAGS
    The 4 D's (diplopia, dysarthria, dysphagia, dysmetria), severe truncal/gait ataxia (cannot stand/walk unaided — peripheral AVS patients can usually walk with lateropulsion), limb ataxia or new severe headache/neck pain → posterior-circulation stroke until excluded; route OUT (Kattah Stroke 2009; Saber Tehrani Neurology 2014)
  • acute_hearing_loss_tinnitusrequired
    symptom • used at CONTEXT
    New unilateral SNHL/tinnitus/aural fullness in AVS = labyrinthitis OR AICA-territory infarct OR SSNHL — HINTS-"plus" positive raises central/AICA risk and is itself a time-critical ENT emergency; route to ent.sudden-sensorineural-hearing-loss.core.v1 (Newman-Toker AEM 2013 HINTS-plus; Saber Tehrani 2014)
  • vascular_risk_factorsrequired
    history • used at CONTEXT
    Age, HTN, DM, AF, smoking, prior stroke/TIA, recent neck trauma/manipulation (vertebral dissection) raise the pre-test central prior in AVS and lower the imaging threshold even with a peripheral-appearing HINTS (Kattah Stroke 2009; Saber Tehrani Neurology 2014 — nonlacunar mechanisms in ~half of small strokes)
  • recent_viral_prodrome
    history • used at CONTEXT
    Antecedent URI/viral prodrome supports the post-viral vestibular-neuritis hypothesis (HSV-1 reactivation) and frames the (debated) corticosteroid discussion; absence does not exclude (Strupp NEJM 2004; Bárány AUVP criteria 2022 — etiology section)
  • agerequired
    demographic • used at CONTEXT
    Elderly AVS carries higher vascular-stroke prior (lower imaging threshold, HINTS caveats), greater fall-injury risk, and slower central compensation modifying rehab/disposition (Kattah Stroke 2009; McDonnell Cochrane 2015 — community-dwelling adults)
  • pregnancy
    history • used at TREATMENT
    Pregnancy: minimise/avoid prolonged vestibular suppressants and gate antiemetic safety; favour early vestibular rehabilitation and avoid routine corticosteroids absent clear indication (Strupp NEJM 2004; safety)
  • diabetes_or_immunocompromise
    history • used at TREATMENT
    Diabetes/immunocompromise raise infective-labyrinthitis and vascular concerns, alter corticosteroid risk-benefit (glycaemic destabilisation), and lower the threshold to investigate a non-idiopathic cause (Bárány AUVP criteria 2022; clinical)
  • intractable_vomiting_dehydration
    symptom • used at DISPOSITION
    Severe persistent vomiting / inability to tolerate oral intake → short-term parenteral antiemetic + IV rehydration and an admission threshold; disproportionate vomiting should also re-prompt a central re-screen (Bárány AUVP criteria 2022; clinical)
  • mri_dwi_posterior_fossa
    imaging • used at BRANCHING_WORKUP
    MRI-DWI is indicated when HINTS is central / cannot be applied / red flags present — but a single early (<48 h) DWI is falsely negative in ~12-15% (~50% for small strokes), so a negative early scan does NOT exclude stroke when HINTS is central (Kattah Stroke 2009; Saber Tehrani Neurology 2014)

12-phase flow (12)

  1. 1FRAME
    Frame the encounter as a TWO-CHANNEL acute-vestibular-syndrome problem: positively diagnose PERIPHERAL AVS (vestibular neuritis / labyrinthitis) AND robustly exclude the can't-miss central mimic (~25% of AVS is posterior-circulation/AICA stroke). The peripheral-vs-central HINTS decision is the spine of this engine. Definitive stroke/SSNHL/Ménière/migraine/falls management is routed OUT by engine_id, not re-authored.
    advance: two-channel AVS scope confirmed; out-of-scope definitive management flagged for engine_id routing
  2. 2ENTRY
    Apply the TiTrATE timing pivot at the door: CONTINUOUS sustained spontaneous vertigo ≥24 h with nystagmus = AVS → this engine's HINTS channel; brief recurrent position-triggered (<1 min, fatiguing) spells = triggered-episodic → route to ent.bppv.core.v1 (Dix-Hallpike, NOT HINTS). Timing/trigger is captured first because it determines which bedside exam is even valid (Newman-Toker Neurol Clin 2015; Bárány AUVP 2022).
    inputs: symptom_timing_continuous_vs_episodic, spontaneous_vs_triggered_onset
    actions: workup.vertigo
    advance: TiTrATE category = AVS (continuous spontaneous) confirmed; triggered-episodic redirected to ent.bppv.core.v1
  3. 3CONTEXT
    Build the pre-test peripheral-vs-central prior: spontaneous unidirectional fixation-suppressed nystagmus (peripheral) vs direction-changing/vertical (central); new auditory symptoms (labyrinthitis / AICA / SSNHL — route by engine_id); vascular risk factors (raise central prior); viral prodrome (post-viral hypothesis, frames the corticosteroid debate); age (vascular + compensation + fall-risk modifier). This phase sets the Bayesian prior before HINTS.
    inputs: nystagmus_character_unidirectional_vs_direction_changing, acute_hearing_loss_tinnitus, vascular_risk_factors, recent_viral_prodrome, age
    actions: workup.vertigo
    advance: peripheral-vs-central pre-test prior assigned; auditory/vascular/viral modifiers captured
  4. 4RED_FLAGS
    Perform HINTS (Head-Impulse–Nystagmus–Test-of-Skew) + screen the 4 D's / gait / focal signs in EVERY continuous AVS. Dangerous (central, "INFARCT") pattern = NORMAL head-impulse, OR direction-CHANGING / vertical nystagmus, OR skew (ANY one) → posterior-circulation stroke until excluded → route OUT to neuro.posterior-circulation-stroke.core.v1. HINTS out-performs early MRI-DWI and ABCD2. Critical guardrail: HINTS is valid ONLY in continuous AVS with nystagmus — NEVER apply it to triggered-episodic vertigo (Kattah Stroke 2009; Newman-Toker AEM 2013; Saber Tehrani Neurology 2014).
    inputs: hints_head_impulse, hints_nystagmus_direction, hints_test_of_skew, central_neuro_4Ds_gait
    actions: calc.abcd2, calc.nihss, protocol.stroke, workup.acute_stroke, workup.acute_headache
    advance: HINTS interpreted (central pattern routed OUT by engine_id) AND 4 D's/gait/focal screen complete
  5. 5INITIAL_WORKUP
    For the peripheral channel (continuous AVS, ALL-peripheral HINTS = abnormal head-impulse + unidirectional fixation-suppressed nystagmus + no skew, no central signs, no acute hearing loss): vestibular neuritis is a CLINICAL diagnosis (Bárány AUVP criteria) — routine imaging and lab panels are NOT required when criteria are met. CBC/CMP/inflammatory markers only if systemic/infective concern (febrile, immunocompromised, suspected suppurative/infectious labyrinthitis) or to gate corticosteroid use (glucose). Audiometry distinguishes neuritis (normal hearing) from labyrinthitis (SNHL → SSNHL pathway).
    inputs: nystagmus_character_unidirectional_vs_direction_changing
    actions: workup.vertigo, panel.cbc, panel.cmp, panel.inflammation
    advance: all-peripheral HINTS + clinical AUVP criteria met (imaging deferred), OR systemic/atypical features trigger labs/imaging
  6. 6BRANCHING_WORKUP
    If HINTS is central / cannot be applied (no nystagmus, uncooperative) / red flags present / high vascular risk → MRI-DWI + posterior-circulation-stroke pathway, recognising a single early DWI is falsely negative in ~12-15% (~50% small strokes) so a negative early scan does NOT exclude stroke when HINTS is central — delayed MRI / specialist if needed (route OUT). If acute SNHL accompanies AVS (labyrinthitis) → audiometry + route to ent.sudden-sensorineural-hearing-loss.core.v1 (time-critical steroid window). If recurrent positional spells emerge after the acute phase → post-neuritis secondary BPPV → route to ent.bppv.core.v1.
    inputs: mri_dwi_posterior_fossa, acute_hearing_loss_tinnitus
    actions: workup.acute_stroke, workup.vertigo
    advance: central pattern imaged/routed OUT, OR labyrinthitis routed to SSNHL, OR peripheral-isolated AVS confirmed for treatment
  7. 7DIFFERENTIAL
    Terminal vestibular differential with named pivots: vestibular neuritis (continuous AVS, unidirectional fixation-suppressed nystagmus, ABNORMAL head-impulse, NO skew, NO central signs, NO hearing loss) vs labyrinthitis (same + acute SNHL → ent.sudden-sensorineural-hearing-loss.core.v1) vs posterior-circulation/AICA stroke (NORMAL head-impulse OR direction-changing/vertical nystagmus OR skew OR 4 D's — route to neuro.posterior-circulation-stroke.core.v1) vs BPPV (triggered-episodic <1 min, positional, fatiguing canal-specific nystagmus — route to ent.bppv.core.v1) vs Ménière (recurrent spontaneous spells minutes-hours + fluctuating low-frequency SNHL/fullness — route to ent.meniere-disease.core.v1) vs vestibular migraine (episodic spontaneous + migraine features — route to neuro.vestibular-migraine.core.v1). The decisive pivot = HINTS + continuous-vs-episodic + hearing.
    actions: workup.vertigo
    advance: single best vestibular diagnosis selected; out-of-scope diagnoses routed by engine_id; co-existence (post-neuritis BPPV) flagged
  8. 8RISK_STRATIFICATION
    For confirmed peripheral AVS: stratify fall/injury risk (age, gait/balance impairment, anticoagulation, osteoporosis, no home support), dehydration severity (intractable vomiting), and central re-screen if any equivocal feature. For an unwell/systemic picture (suspected infective labyrinthitis, meningitis-associated): NEWS2 / qSOFA to detect a non-vestibular systemic cause needing a different pathway. The ABCD2 caveat: ABCD2 UNDER-performs HINTS — do NOT use ABCD2 to "rule out" stroke in AVS.
    inputs: age, vascular_risk_factors, intractable_vomiting_dehydration
    actions: calc.abcd2, calc.nihss, calc.qsofa, calc.news2
    advance: fall-risk / dehydration / systemic-screen stratified; ABCD2-as-rule-out guardrail enforced
  9. 9TREATMENT
    Three-lever peripheral-AVS management: (1) SHORT-COURSE vestibular suppressant + antiemetic for the acute symptomatic phase ONLY, strictly ≤48-72 h — prolonged use RETARDS central compensation (the core caveat); (2) EARLY vestibular rehabilitation begun once acute vomiting settles — the single evidence-based recovery lever (McDonnell Cochrane 2015 — dizziness OR 2.67, DHI SMD −0.83); (3) the DEBATED corticosteroid step — methylprednisolone improves CALORIC recovery (Strupp NEJM 2004, 62.4 vs 39.6 pp) but Cochrane (Fishman 2011) / Leong OHNS 2021 find NO symptomatic/long-term benefit and "insufficient evidence" — present as shared-decision equipoise, weigh diabetes/immunocompromise/pregnancy. ANTIVIRALS are NOT recommended (Strupp NEJM 2004 — valacyclovir ineffective). Acute SNHL (labyrinthitis) → route to ent.sudden-sensorineural-hearing-loss.core.v1 (its own time-critical steroid pathway).
    inputs: pregnancy, diabetes_or_immunocompromise, recent_viral_prodrome
    advance: short-course suppressant time-limited (≤48-72 h); early vestibular rehab initiated; corticosteroid shared-decision documented; antivirals withheld
  10. 10DISPOSITION
    All-peripheral HINTS + safe gait + tolerating oral intake → discharge with early vestibular-rehab plan, suppressant-stop instruction, and central return precautions. Central HINTS / 4 D's / cannot apply HINTS / acute hearing loss → admit / emergent neuro / route OUT (neuro.posterior-circulation-stroke.core.v1 or ent.sudden-sensorineural-hearing-loss.core.v1). Intractable vomiting/dehydration or unsafe gait/high fall risk (frail elderly, no support) → short-stay/admit for IV rehydration + supervised mobilisation + PT.
    inputs: central_neuro_4Ds_gait, intractable_vomiting_dehydration, age
    actions: protocol.stroke
    advance: disposition documented; central/SNHL cases routed OUT by engine_id; fall-risk/hydration disposition set
  11. 11MONITORING
    Re-examine within the first 24-72 h: peripheral AVS should show GRADUAL improvement (nystagmus decreasing, gait improving) as central compensation begins. NON-resolution, worsening, NEW neuro/auditory features, or evolving direction-changing nystagmus → re-triage as CENTRAL (small posterior-fossa strokes are missed on early MRI-DWI ~50% — Saber Tehrani Neurology 2014); image / route OUT. STOP the short-course suppressant by 48-72 h regardless; persisting symptomatic suppressant use is itself a compensation-retarding error.
    inputs: nystagmus_character_unidirectional_vs_direction_changing, central_neuro_4Ds_gait
    actions: workup.vertigo
    advance: gradual peripheral recovery confirmed OR re-triaged central; short-course suppressant stopped by 48-72 h
  12. 12FOLLOWUP
    Reassess at ~1-4 weeks: most recover via central compensation over weeks; persistent imbalance/oscillopsia → structured vestibular rehabilitation (clinician- or home-based — McDonnell Cochrane 2015). Pursue secondary substrate: post-neuritis BPPV (utricular otoconia shed → new positional spells → route to ent.bppv.core.v1), incomplete compensation in the elderly (fall-injury risk → route to geriatrics.falls.core.v1), and audiometry follow-up if labyrinthitis. Educate on recovery trajectory, that residual unsteadiness for weeks is expected and NOT failure, and central return precautions.
    inputs: age, post_viral_or_post_neuritis_recurrent
    actions: workup.falls
    advance: 1-4 week reassessment booked; vestibular-rehab + secondary-substrate plan documented; falls/BPPV/audiometry referral made if criteria met