Autoimmune encephalitis and organic psychosis

Psychiatrists fail this topic when they treat every new psychosis as primary FEP, or when they chase isolated serum antibodies without a clinical syndrome. Examiners reward red-flag recognition, correct work-up hierarchy, honest limits of antibody testing, and early multidisciplinary immunotherapy — not a full neuroimmunology textbook recitation.[6][5][17]

Overview and definition

Organic (secondary) psychosis is psychosis caused by a medical, neurologic, toxic, or metabolic process rather than a primary psychotic disorder. Autoimmune encephalitis (AE) is a group of inflammatory brain diseases driven by immune responses against neuronal antigens; many present with prominent psychiatric features before hard neurology appears.[4][5]

Antibody-mediated encephalitis (cell-surface or synaptic antibodies such as NMDAR, LGI1, CASPR2) is the psychiatry-relevant core: antibodies are often directly pathogenic, immunotherapy response is often good, and tumours (especially ovarian teratoma in anti-NMDAR disease) are disease-modifying targets.[1][4] Classic intracellular paraneoplastic antibodies (Hu, Ma2, CV2/CRMP5, amphiphysin) more often mark T-cell–mediated injury with poorer immunotherapy response and stronger cancer linkage — still organic psychosis differentials, different prognosis conversation.[4][3]

DSM-5-TR / ICD-11 frame secondary psychotic syndromes by medical cause; AE is not a freestanding DSM code you invent — you diagnose the encephalitis syndrome and treat the cause.[5]

Classification frameworks examiners expect

Graus 2016 — autoimmune encephalitis

Graus and colleagues provide a practical clinical approach to possible, probable, and definite AE using syndrome, MRI/EEG/CSF support, and antibody confirmation, designed so treatment is not indefinitely delayed by serology turnaround.[3] Key operational message for psychiatry: probable AE can be treated while awaiting definitive antibody results when the clinical–paraclinical package fits.[3][12]

Pollak 2020 — autoimmune psychosis consensus

An international consensus (Pollak et al., Lancet Psychiatry) proposes structured criteria for possible, probable, and definite autoimmune psychosis, translating AE logic into psychosis-first presentations and guiding when to escalate investigations and immunotherapy discussions.[5] Use it as a psychiatry-facing scaffold that still depends on neurologic partnership and rigorous antibody methodology — not as a licence to immunosuppress every atypical FEP.[5][17]

Antibody map (high-yield)

Epidemiology and risk

[10] [2] [11] [15]

Dubey and colleagues showed AE is not vanishingly rare relative to infectious encephalitis in modern diagnostics — the psychiatric trap is rarity framing that delays work-up.[10] Titulaer's large anti-NMDAR cohort established treatment-response and prognostic benchmarks still used in viva answers.[2]

Risk clues: female sex and reproductive age (teratoma), recent HSV encephalitis, subacute tempo, seizures, movement disorders, hyponatraemia (think LGI1), and first presentation outside the usual FEP demography without substance explanation.[2][8][11]

Pathophysiology (psychiatry depth, not fluff)

In anti-NMDAR encephalitis, IgG antibodies target the GluN1 (NR1) subunit of the NMDA receptor, causing receptor cross-linking and internalisation with reversible reduction of surface NMDAR — experimental and clinical observations that explain both psychiatric phenotype and treatment responsiveness when injury is not yet fixed.[1][4] Triggers include ovarian teratoma (neural tissue expressing NMDAR) and herpes simplex encephalitis as a post-infectious autoimmune sequel; many cases remain idiopathic.[1][11][4]

Cell-surface diseases differ mechanistically from many intracellular-antibody paraneoplastic syndromes, where cytotoxic T-cell injury dominates and antibody titres are biomarkers more than primary effectors — prognosis and immunotherapy expectations diverge accordingly.[4]

Clinical presentation

Anti-NMDAR staged course

Fellowship stems often walk through: viral-like prodrome → psychiatric psychosis, mania, anxiety, insomnia, aggression → speech reduction or mutism → orofacial and limb dyskinesias, dystonia, catatonia → seizures and autonomic instability → central hypoventilation requiring ICU.[1][2][4] Not every patient is textbook sequential; pure or prolonged psychiatric phases occur and are exactly why psychiatry must own recognition.[7][6]

Psychopathology synthesis (Al-Diwani)

Individual-patient-data synthesis of adult NMDAR-antibody encephalitis shows a mixed mood–psychosis–catatonia–behavioural phenotype rather than a single first-rank schizophrenia clone. Expect mood lability, delusions, hallucinations, aggression, insomnia, and catatonic signs — often with rapid evolution that outpaces ordinary FEP tempo.[7]

LGI1 and CASPR2 (do not forget older adults)

Faciobrachial dystonic seizures (FBDS) — brief ipsilateral face and arm jerks — are a highly specific clinical clue to LGI1 antibody disease and often precede frank limbic encephalitis; hyponatraemia is common.[8][9] CASPR2 spectrum includes neuromyotonia, Morvan syndrome features, severe insomnia, and autonomic disturbance.[9] Mislabeling these as late-onset primary psychosis loses the immunotherapy window.

Red flags and differential

[5] [6] [12] [11]

Discriminators, not laundry lists: clonus/hyperreflexia with serotonergics → serotonin toxicity pathway; lead-pipe rigidity after dopamine antagonists → NMS; pure environmental heat → heat stroke; CSF lymphocytic pleocytosis + subacute encephalitis → infectious and autoimmune tracks in parallel until HSV PCR and AE work-up clarify.[12][3]

Assessment at the bedside

1. Tempo chart — hour/day evolution from premorbid baseline; viral prodrome; sleep collapse.
2. Collateral — speech change, facial movements, nocturnal seizures, personality rupture.
3. MSE with examples — fluctuating alertness; catatonic excitement or stupor; poorly systematised psychosis evolving over days.
4. Neurologic screen — FBDS, orofacial dyskinesia, myoclonus, tone, pupils, autonomic vitals.
5. Risk — suicide, violence, vulnerability, absconding; capacity for LP/MRI/immunotherapy; local Mental Health Act principles (jurisdiction-specific statutes — do not invent foreign section numbers).
6. Apply structured red-flag and Pollak-style thresholds to decide investigation intensity.[6][5][7]

Investigations

Antibody testing pitfalls: commercial serum-only tests can miss or overcall; CSF increases diagnostic confidence in anti-NMDAR disease; isolated low-level serum positivity without syndrome is not automatic "autoimmune psychosis."[4][17][13] Lennox found serum neuronal cell-surface antibodies enriched in FEP versus controls, but clinical relevance requires full phenotype and methodologic care — Guasp and Dalmau caution against over-searching unselected psychiatric populations.[15][17] Australian early-psychosis testing guidance (Scott) and later FEP service analyses (Cohn) support thoughtful targeted testing, not dogma that every mild FEP needs a full AE panel.[13][14]

Acute management

Medical priorities first: airway and ventilation risk (especially anti-NMDAR hypoventilation), seizure control, autonomic instability, DVT prophylaxis, aspiration precautions, and ICU liaison.[2][12]

Symptomatic psychotropics: use the lowest effective doses for dangerous agitation or psychosis as a bridge. Prefer cautious titration; benzodiazepines are first-line for catatonia and often for agitation. Avoid reflexive high-potency parenteral antipsychotic stacking that risks obscuring NMS/catatonia differentials. Antipsychotics do not replace immunotherapy.[5][12][7]

Definitive management — immunotherapy principles

First-line immunotherapy

Observational cohorts (Titulaer) and best-practice recommendations (Abboud) support early first-line immunotherapy once infectious mimics needing antivirals are addressed in parallel as clinically indicated: high-dose corticosteroids (commonly IV methylprednisolone about 1 g daily for 3–5 days in adult protocols — confirm local neurology protocol and product information), IV immunoglobulin (IVIG) (commonly total 2 g/kg divided over 2–5 days), and/or therapeutic plasma exchange (often about 5 exchanges on alternate days in many centres).[2][12][18] Exact combinations are specialist-led; psychiatry's job is to not block timely treatment while capacity and legal pathways are secured.[12][5]

Second-line and refractory

If response is inadequate, rituximab and/or cyclophosphamide are standard second-line escalations in anti-NMDAR and related AE pathways (dosing is neurology/oncology protocol–specific; example rituximab schedules include 375 mg/m² weekly × 4 or 1 g fortnightly × 2 — always under specialist governance).[2][12][18] Refractory disease may involve additional agents such as tocilizumab in selected cohort experience — know it exists; do not claim large RCT proof.[16]

Tumour treatment

In anti-NMDAR encephalitis with ovarian teratoma, prompt tumour resection plus immunotherapy associates with better outcomes — delaying imaging in young women with probable disease is an exam-fail.[1][2]

Do not wait forever for the lab

When probable AE criteria are met, start immunotherapy while antibody results pend, after appropriate infectious evaluation — delayed treatment predicts worse outcomes in large cohorts.[2][3][12]

Subtypes and scenarios

Complications and pitfalls

• Escalating antipsychotics only while encephalitis progresses.
• Reassured by normal MRI.
• Serum-only testing.
• Overcalling disease from incidental antibodies.
• Missing teratoma.
• Missing post-HSV AE.
• ICU complications: status epilepticus, dysautonomia, hypoventilation, VTE, infection on immunotherapy.
• Premature diagnostic fatalism ("schizophrenia forever") within weeks of onset.[6][2][17][11]

Prognosis and disposition

In Titulaer's multi-institutional anti-NMDAR cohort, most patients improved with immunotherapy and tumour treatment when needed, but recovery often unfolded over months, with substantial early ICU utilisation and residual cognitive/psychiatric morbidity in a minority — early treatment and tumour removal are key modifiable factors.[2] Disposition needs joint neuroimmunology + psychiatry follow-up, relapse education, rehabilitation, and careful psychotropic de-prescription as the encephalitis remits.[12][5]

Special populations

• Children/adolescents: movement disorders and speech regression may dominate over adult-like delusions; paediatric consensus (Nosadini) prioritises corticosteroids plus IVIG/PLEX pathways.[18]
• Older adults: weight LGI1/CASPR2 and limbic phenotypes heavily.[8][9]
• Pregnancy: multidisciplinary fetal–maternal–neurology planning; still pursue diagnosis and indicated immunotherapy.
• Intellectual disability/autism: new regression or catatonia deserves organic work-up, not diagnostic overshadowing.
• Cultural formulation: content of psychosis may be culturally shaped; red flags and tempo still drive investigation.[5][6]

Evidence, guidelines, and regional practice

Named landmarks to drop in viva: Dalmau 2008 series; Titulaer 2013 outcomes; Graus 2016 criteria; Dalmau/Graus NEJM 2018; Pollak 2020 consensus; Herken/Prüss red flags; Al-Diwani psychopathology; Irani FBDS; Armangue post-HSV; Dubey epidemiology; Abboud 2021 recommendations.[1][2][3][4][5][6][7][8][11][10][12]

Exam pearls

[3] [5] [2] [6]

• Normal MRI ≠ not NMDAR.[4][2]
• Extreme delta brush supports but does not prove anti-NMDAR encephalitis.[4]
• Post-HSV behavioural relapse ≈ think AE, not only viral reactivation.[11]
• FBDS ≈ LGI1 until proven otherwise.[8]
• First-line: steroids ± IVIG ± PLEX; second-line: rituximab/cyclophosphamide; parallel: tumour resection.[2][12]
• Do not immunosuppress the lab result without the syndrome.[17]
• Legal pathway: least restrictive care enabling life-saving investigation/treatment under local law — never invent interstate section numbers.[5]

Self-test: 30-second stem

A 23-year-old woman develops persecutory delusions and insomnia over 10 days after a viral-like prodrome. On day 12 she is mute with orofacial dyskinesias; temperature is normal; MRI is reported normal. Staff request IM antipsychotics only. What is your plan? Model: This is a classic anti-NMDAR red-flag package despite normal MRI. Escalate medical work-up: EEG, LP with CSF (including HSV PCR as indicated) and paired serum–CSF NMDAR cell-based assay, pelvic imaging for teratoma, neurology/ICU as needed. Start first-line immunotherapy when probable AE criteria are met; use benzodiazepines/cautious psychotropics for safety only — not as definitive care.[1][2][6][12]

References

1. [1]Dalmau J, Gleichman AJ, Hughes EG, et al. Anti-NMDA-receptor encephalitis: case series and analysis of the effects of antibodies Lancet Neurol, 2008.PMID 18851928
2. [2]Titulaer MJ, McCracken L, Gabilondo I, et al. Treatment and prognostic factors for long-term outcome in patients with anti-NMDA receptor encephalitis: an observational cohort study Lancet Neurol, 2013.PMID 23290630
3. [3]Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis of autoimmune encephalitis Lancet Neurol, 2016.PMID 26906964
4. [4]Dalmau J, Graus F Antibody-Mediated Encephalitis N Engl J Med, 2018.PMID 29490181
5. [5]Pollak TA, Lennox BR, Müller S, et al. Autoimmune psychosis: an international consensus on an approach to the diagnosis and management of psychosis of suspected autoimmune origin Lancet Psychiatry, 2020.PMID 31669058
6. [6]Herken J, Prüss H Red Flags: Clinical Signs for Identifying Autoimmune Encephalitis in Psychiatric Patients Front Psychiatry, 2017.PMID 28261116
7. [7]Al-Diwani A, Handel A, Townsend L, et al. The psychopathology of NMDAR-antibody encephalitis in adults: a systematic review and phenotypic analysis of individual patient data Lancet Psychiatry, 2019.PMID 30765329
8. [8]Irani SR, Michell AW, Lang B, et al. Faciobrachial dystonic seizures precede Lgi1 antibody limbic encephalitis Ann Neurol, 2011.PMID 21416487
9. [9]van Sonderen A, Petit-Pedrol M, Dalmau J, et al. The value of LGI1, Caspr2 and voltage-gated potassium channel antibodies in encephalitis Nat Rev Neurol, 2017.PMID 28418022
10. [10]Dubey D, Pittock SJ, Kelly CR, et al. Autoimmune encephalitis epidemiology and a comparison to infectious encephalitis Ann Neurol, 2018.PMID 29293273
11. [11]Armangue T, Spatola M, Vlagea A, et al. Frequency, symptoms, risk factors, and outcomes of autoimmune encephalitis after herpes simplex encephalitis: a prospective observational study and retrospective analysis Lancet Neurol, 2018.PMID 30049614
12. [12]Abboud H, Probasco JC, Irani S, et al. Autoimmune encephalitis: proposed best practice recommendations for diagnosis and acute management J Neurol Neurosurg Psychiatry, 2021.PMID 33649022
13. [13]Scott JG, Gillis D, Swayne A, et al. Testing for antibodies to N-methyl-d-aspartate receptor and other neuronal cell surface antigens in patients with early psychosis Aust N Z J Psychiatry, 2018.PMID 29923413
14. [14]Cohn SL, Mohan A, Lappin JM, et al. Anti-N-Methyl-d-Aspartate Receptor Antibody Testing in First-Episode Psychosis: Universal or Targeted Testing J Neuropsychiatry Clin Neurosci, 2023.PMID 36128677
15. [15]Lennox BR, Palmer-Cooper EC, Pollak T, et al. Prevalence and clinical characteristics of serum neuronal cell surface antibodies in first-episode psychosis: a case-control study Lancet Psychiatry, 2017.PMID 27965002
16. [16]Lee WJ, Lee ST, Moon J, et al. Tocilizumab in Autoimmune Encephalitis Refractory to Rituximab: An Institutional Cohort Study Neurotherapeutics, 2016.PMID 27215218
17. [17]Guasp M, Dalmau J Searching for Neuronal Antibodies in Psychiatric Diseases: Uncertain Findings and Implications Neurology, 2023.PMID 37353340
18. [18]Nosadini M, Thomas T, Eyre M, et al. International Consensus Recommendations for the Treatment of Pediatric NMDAR Antibody Encephalitis Neurol Neuroimmunol Neuroinflamm, 2021.PMID 34301820