Recovery and recurrence following a first episode of mania: a systematic review and meta-analysis of prospectively characterized cohorts

J Clin Psychiatry. 2015 Sep;76(9):1241-8. doi: 10.4088/JCP.14r09245.


Objective: Information about recurrence rates is useful in informing clinical practice, but most data with regard to recurrence rates in bipolar patients come from cohorts at different stages of illness. These data are of limited utility in estimating risk of relapse in first-episode bipolar disorder. Therefore, the objective of this investigation was to synthesize available recurrence data after a first episode of mania.

Data sources: We searched MEDLINE, EMBASE, PsycINFO, and Cochrane Central Register of Controlled Trials (CENTRAL) from 1980 to January 24th, 2014, for articles in English, French, or Spanish using (1) bipolar disorder (MeSH term) OR manic/mania, AND (2) first* (episode*, hospitalization* OR admission*) OR time factor (MeSH term), AND (3) recovery, remission, recurrence OR relapse.

Study selection: 712 articles were screened. Prospective cohorts of first-episode mania were included.

Data extraction: Syndromal recovery, symptomatic recovery, and recurrence rates were extracted by 2 independent raters at 6 months, 1 year, 2 years, and 4 years and analyzed using random effects models and meta-regression.

Results: We identified 8 studies representing a total of 734 first-episode patients. The syndromal recovery rates were 77.4% at 6 months and 84.2% at 1 year. Only 62.1% of patients had achieved a period of symptomatic recovery within 1 year. Recurrence rates were 25.7% within 6 months, 41.0% by 1 year, and 59.7% by 4 years. Younger age at first episode was associated with risk of recurrence after 1 year.

Conclusions: The majority of patients with first-episode mania exhibit syndromal recovery and, to a lesser extent, symptomatic recovery. The risk of recurrence is high, although the rates are slightly lower than those in mixed cohorts, with greater risk of recurrence associated with younger age at onset. Given lower recurrence than among mixed cohorts, there may be a window of opportunity to provide optimal treatment early and alter disease progression.

Publication types

  • Meta-Analysis
  • Review
  • Systematic Review

MeSH terms

  • Bipolar Disorder / epidemiology*
  • Humans
  • Prospective Studies*
  • Recurrence
  • Remission Induction
  • Risk Factors