After a first psychotic manic episode, mood stabiliser regimens were linked to fewer relapses

Design

Target trial emulation using linked observational cohort data

Population

371 people aged 16 or over with a first manic episode with psychotic features, median age 22

Primary outcome

Psychiatric hospitalisation or emergency visit over 2 years

Effect

42.3% mood stabiliser, 42.2% combination, 49.5% antipsychotic alone; RR 0.82 for each mood stabiliser strategy vs antipsychotic alone

Researchers emulated a pragmatic trial using linked observational data from first-episode psychosis clinics in North America, Chile and Spain. They included 371 people aged 16 or over with bipolar I disorder presenting with a first manic episode with psychotic features, who had not recently taken a mood stabiliser or second-generation antipsychotic. Three strategies were compared over two years: mood stabiliser alone, second-generation antipsychotic alone, or both.

The two-year risk of a psychiatric hospitalisation or emergency visit was 42.3% with a mood stabiliser alone, 42.2% with the combination and 49.5% with antipsychotic monotherapy. Against antipsychotic monotherapy, both mood stabiliser strategies carried a risk ratio of 0.82, an absolute difference of about 7 percentage points. Mood stabiliser alone and combination therapy did not differ.

The median age was 22, so these are the patients whose early course shapes decades of illness. Many young people leave hospital on an antipsychotic alone because the psychosis is what was treated. This analysis argues for deliberately adding or anchoring on a mood stabiliser for maintenance, with the usual care around lithium monitoring and the avoidance of valproate in women who could become pregnant. A randomised trial is still needed.

• Review the maintenance plan of anyone discharged on an antipsychotic alone after a first psychotic manic episode.
• Consider a mood stabiliser for maintenance, alone or alongside the antipsychotic.
• Avoid valproate in women and girls who could become pregnant, unless no alternative is suitable and pregnancy prevention is in place.
• Arrange baseline kidney and thyroid tests and level monitoring before starting lithium.
• Plan follow-up for the first two years, when about four in ten in this cohort were readmitted or attended emergency services.

Why it matters

It challenges the common habit of continuing only the antipsychotic that settled the psychosis.

Don't overread it

This emulates a trial with observational data; residual confounding is possible and no randomised trial has yet confirmed it.

The statistics, in plain English

A risk ratio of 0.82 means about 18% fewer hospitalisations or emergency visits over two years. The confidence intervals are wide and come close to no effect at their upper end (0.97 for mood stabiliser alone), so the true benefit could be small. Target trial emulation reduces some biases of ordinary observational studies but cannot remove unmeasured reasons why clinicians chose one regimen over another.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.