Early-onset psychosis and mania — branching viva

Opening

Examiner: A 15-year-old boy is brought in believing classmates implanted a chip in his head, has not slept for four nights, and smokes high-potency cannabis daily. Walk me through your approach from the moment you meet him. [2]

Candidate: He is a medical patient first and a psychiatric patient second. I run the gate in order: assess, distinguish, exclude danger, refer early and treat. I greet him first, set a joint agenda, and secure time alone with conditional confidentiality stated aloud. Because a psychotic young person is an unreliable sole historian, I gather collateral from family and school on onset, course, function, family psychiatric history, substance use and trauma. I perform a mental-state examination, run HEEADSSS with focused probes, and screen suicide risk with plan, intent and means — then exclude organic disease before any label. [2]

Branch A — Distinguishing the three faces

Examiner: How do you tell psychosis from mania and from a mimic? [2]

Candidate: I place him on three axes at once. The psychosis spectrum runs from clinical high risk through brief and schizophreniform disorders to schizophrenia, unified by loss of reality testing. The bipolar spectrum is defined by episodic, distinct-from-baseline mood elevation with increased energy and reduced sleep. The third axis is mimics — substance intoxication or withdrawal, organic brain disease, trauma and dissociation, and mood disorder with psychotic features — which need different treatment. The discriminator is reality testing and course: persistent loss of reality testing with functional decline points to a primary psychotic disorder, while episodic elevation points to mania. [2] [3]

Probe: How do you avoid over-diagnosing paediatric bipolar disorder? [7]

Candidate: I require the episodic, distinct-from-baseline quality — a clear change from the young person's usual self — rather than chronic irritability. Chronic, severe, non-episodic irritability points to disruptive mood dysregulation disorder or trauma-driven dysregulation, not mania, and mislabelling it exposes a child to unnecessary medication. [7]

Branch B — Organic exclusion

Examiner: What organic causes must you exclude, and how? [2]

Candidate: Encephalitis (infectious and autoimmune), thyroid disease, central nervous system infection, electrolyte disturbance, autoimmune disease, central nervous system tumour and substance intoxication can all masquerade as primary psychosis or mania. I examine for fever, fluctuating consciousness, headache, focal neurology and signs of intoxication or withdrawal, and order a targeted set: full blood count, electrolytes, liver and renal function, thyroid function, inflammatory markers and a urine drug screen, with an electroencephalogram when seizures are possible and neuroimaging when onset is acute, focal neurology is present, or consciousness fluctuates. Missing an organic cause is catastrophic, so the physical examination is part of the psychiatric assessment. [2]

Probe: His cannabis use — is it relevant or incidental? [5]

Candidate: Highly relevant. Cannabis is a real, modifiable, dose-related risk factor. Moore's systematic review found cannabis associated with a clear increase in psychotic outcomes, and Arseneault's longitudinal study showed adolescent use raised the risk of adult psychosis dose-dependently. I ask about potency, frequency and age of first use, counsel reduction, and arrange supported abstinence with reassessment — persistent symptoms after abstinence warrant full treatment of a primary disorder. [5] [6]

Branch C — Treatment and evidence

Examiner: First-line treatment for early-onset schizophrenia, and what evidence guides the choice? [1]

Candidate: A specialist-led antipsychotic is first-line after assessment. The TEOSS trial compared risperidone and olanzapine with molindone and found efficacy was similar across groups with different side-effect burdens — so there is no clear efficacy advantage of second- over first-generation drugs, and the side-effect profile decides. Risperidone and aripiprazole are common first choices. Treatment continues for one to two years after a first episode, and discontinuation happens only with specialist oversight because early relapse worsens outcome. [1] [2]

Probe: And if the picture were mania instead? [9]

Candidate: Second-generation antipsychotics have the strongest evidence: Tohen showed olanzapine beat placebo in adolescent mania, Haas showed risperidone beat placebo, and aripiprazole and quetiapine are further supported. Geller found risperidone, lithium and divalproex each beat placebo, with risperidone most effective but most fattening. Oxcarbazepine did not separate from placebo in Wagner's trial, so it is not first-line. [9] [8]

Branch D — Duration of untreated psychosis and monitoring

Examiner: Why does the timing of referral matter? [4]

Candidate: Longer duration of untreated psychosis predicts poorer long-term symptom, social and functional outcome — Penttilä's meta-analysis confirms this. Shortening the DUP through prompt specialist referral is the single most evidence-supported action a general paediatrician can take; early referral is treatment, not paperwork. The adolescent brain's reward and socio-emotional circuitry matures ahead of cognitive control, magnifying stress sensitivity, so early treatment is neurobiologically justified. [4] [3]

Probe: What monitoring does an antipsychotic demand? [1]

Candidate: Real metabolic and endocrine costs. I capture baseline weight, body-mass index, waist circumference, blood pressure, fasting glucose, lipids, prolactin and an electrocardiogram, then review at 4–6 weeks, 12 weeks and six-monthly, with extrapyramidal side effects, sedation and sexual function checked and suicide risk reassessed at every contact. Counselling, baseline tests and a monitoring plan are part of every prescription. [1]

Close

Examiner: One-line take-home. [2]

Candidate: Run the gate — assess, distinguish psychosis from mania and from mimics, exclude organic disease and suicide risk before any label, then refer early and treat; TEOSS shows no efficacy advantage between drug classes so side effects decide the choice, second-generation antipsychotics are first-line for paediatric mania, and shortening the duration of untreated psychosis is the most evidence-supported thing a general paediatrician can do. [1] [2] [4]