Pharmacological interventions for acute bipolar mania: a systematic review of randomized placebo-controlled trials


  • Funding was provided to Medical Research Matters by Sanofi-Aventis in the form of an unrestricted grant. The protocol was designed and agreed by all authors (including an author from the sponsoring company) and the analysis was performed independently by LAS and VC. Data were then considered by all authors and a report was drawn up by LAS and DT. The final content of the manuscript was contributed to and agreed by all authors. Sanofi-Aventis did not have (and did not attempt to exert) editorial control over this manuscript.

David Taylor, MSc, PhD, MRPharmS, Chief Pharmacist, South London and Maudsley NHS Trust, Maudsley Hospital, London SE5 8AZ, UK. Fax: +44 20 7919 3448; e-mail:


Objectives:  We conducted a systematic review and meta-analysis of randomized, placebo-controlled trials in acute bipolar mania to summarize available data on drug treatment of mania.

Methods:  We included trials of medications licensed in the USA or UK for the treatment of any phase of bipolar disorder. Outcomes investigated were changes in mania scores, attrition, extrapyramidal effects and weight change. Data were combined through meta-analyses.

Results:  We included 13 studies (involving 3,089 subjects) and identified 2 studies for each of the following medications: carbamazepine, haloperidol, lithium, olanzapine, quetiapine, risperidone, valproate semisodium and aripiprazole. All drugs showed significant benefit compared with placebo for reduction in mania scores. Compared with placebo, for all antipsychotics pooled, response to treatment (≥50% reduction in Young Mania Rating Scale scores) was increased more than 1.7 times [relative risk (RR) = 1.74, 95% confidence interval (CI) = 1.54, 1.96]; for all mood stabilizers pooled, response to treatment was doubled (RR 2.01, 95% CI = 1.66, 2.43). Overall withdrawals were 34% fewer (24–43%) with antipsychotics, and 26% fewer (10–39%) with mood stabilizers. However, for carbamazepine, aripiprazole and lithium an increase in risk of withdrawal could not be excluded. Small but significant increases in extrapyramidal side effects occurred with risperidone and aripiprazole.

Conclusions:  Antipsychotics and mood stabilizers are significantly more effective than placebo for the treatment of acute mania. Their effect sizes are similar. Small differences between effect sizes may be due to differences in the patients included in the studies or to chance. Carbamazepine and lithium may be more poorly tolerated, and antipsychotics cause more extrapyramidal side effects.