Aripiprazole alone or in combination with other drugs for treating the acute mania phase of bipolar disorder

Bipolar disorder is a mental disorder that is seen as periods of high mood called mania, or hypomania if less severe, and periods of low mood (depression).

Medication is the main treatment for mania, with the first aim to decrease agitation, aggression and dangerous behaviour.

Antipsychotics and other antimanic medicines are included in guidelines for treating mania. This review considers the antipsychotic, aripiprazole, and assesses how effective it is in the treatment of acute mania. It also examines the side effects of aripiprazole and discusses whether people find aripiprazole to be an acceptable treatment for themselves.

Ten studies are included (3340 participants). Most studies compared aripiprazole versus placebo, but some researchers compared aripiprazole versus haloperidol (two studies) and versus lithium (one study). Two studies examined the effect of adding aripiprazole to another treatment (valproate or lithium) and compared this combination versus placebo combined with these other treatments. We assessed the overall risk of bias in the ten studies as unclear.

The main measure of effect was the mean change on the Young Mania Rating Scale from the start to the end of the trial; this tool is used by clinicians to assess the severity of mania. After three weeks of treatment, aripiprazole was better than placebo at reducing the severity of mania when used on its own or when added to other mood stabilisers. The effect was modest. However, aripiprazole caused more inner restlessness (akathisia), nausea, and constipation than placebo. Aripiprazole was similarly effective in reducing the symptoms of mania when compared with other drug treatments (haloperidol and lithium). Aripiprazole caused fewer movement disorders and less raised prolactin (a hormone secreted by the pituitary gland) than haloperidol. People taking aripiprazole were more likely to remain on treatment than those taking haloperidol but were no more or less likely than those taking placebo or lithium. The main reason for the difference in dropouts between aripiprazole and haloperidol groups was the adverse effects associated with haloperidol.

In summary, aripiprazole is an effective treatment for mania when compared with placebo. This finding is based on studies that included mixed populations (i.e. children, adolescents and adults). For the adult population, studies have directly compared aripiprazole versus haloperidol, lithium and placebo, but evidence obtained for treatment of the child and adolescent population is available only from placebo-controlled studies. Given the lack of evidence obtained by comparing aripiprazole versus other drugs, its exact place in therapy is unclear. Further studies focused on particular populations are needed to determine whether this treatment is equally effective in different age groups.

Authors' conclusions: 

Aripiprazole is an effective treatment for mania in a population that includes adults, children and adolescents, although its use leads to gastrointestinal disturbances and movement disorders. Comparative trials with medicines other than haloperidol and lithium are few, so the precise place of aripiprazole in therapy remains unclear.

Read the full abstract...
Background: 

Bipolar disorder is a mental disorder characterised by episodes of elevated or irritable mood (manic or hypomanic episodes) and episodes of low mood and loss of energy (depressive episodes). Drug treatment is the first-line treatment for acute mania with the initial aim of rapid control of agitation, aggression and dangerous behaviour. Aripiprazole, an atypical antipsychotic, is used in the treatment of mania both as monotherapy and combined with other medicines. The British Association of Psychopharmacology guidelines report that, in monotherapy placebo-controlled trials, the atypical antipsychotics, including aripiprazole, have been shown to be effective for acute manic or mixed episodes.

Objectives: 

To assess the efficacy and tolerability of aripiprazole alone or in combination with other antimanic drug treatments, compared with placebo and other drug treatments, in alleviating acute symptoms of manic or mixed episodes. Other objectives include reviewing the acceptability of treatment with aripiprazole, investigating the adverse effects of aripiprazole treatment, and determining overall mortality rates among those receiving aripiprazole treatment.

Search strategy: 

The Cochrane Depression, Anxiety and Neurosis Group's Specialised Register (CCDANCTR-Studies and CCDANCTR-References) was searched, all years to 31st July 2013. This register contains relevant randomised controlled trials from: The Cochrane Library (all years), MEDLINE (1950 to date), EMBASE (1974 to date), and PsycINFO (1967 to date). We also searched Bristol-Myers Squibb clinical trials register, the World Health Organization (WHO) trials portal (ICTRP) and ClinicalTrials.gov (to August 2013).

Selection criteria: 

Randomised trials comparing aripiprazole versus placebo or other drugs in the treatment of acute manic or mixed episodes.

Data collection and analysis: 

Two review authors independently extracted data, including adverse effect data, from trial reports and assessed bias. The drug manufacturer or the trial authors were contacted for missing data.

Main results: 

Ten studies (3340 participants) were included in the review. Seven studies compared aripiprazole monotherapy versus placebo (2239 participants); two of these included a third comparison arm—one study used lithium (485 participants) and the other used haloperidol (480 participants). Two studies compared aripiprazole as an adjunctive treatment to valproate or lithium versus placebo as an adjunctive treatment (754 participants), and one study compared aripiprazole versus haloperidol (347 participants). The overall risk of bias was unclear. A high dropout rate from most trials (> 20% for each intervention in eight of the trials) may have affected the estimates of relative efficacy. Evidence shows that aripiprazole was more effective than placebo in reducing manic symptoms in adults and children/adolescents at three and four weeks but not at six weeks (Young Mania Rating Scale (YMRS); mean difference (MD) at three weeks (random effects) -3.66, 95% confidence interval (CI) -5.82 to -2.05; six studies; N = 1819, moderate quality evidence) - a modest difference. Aripiprazole was compared with other drug treatments in three studies in adults—lithium was used in one study and haloperidol in two studies. No statistically significant differences between aripiprazole and other drug treatments in reducing manic symptoms were noted at three weeks (YMRS MD at three weeks (random effects) 0.07, 95% CI -1.24 to 1.37; three studies; N = 972, moderate quality evidence) or at any other time point up to and including 12 weeks. Compared with placebo, aripiprazole caused more movement disorders, as measured on the Simpson Angus Scale (SAS), on the Barnes Akathisia Scale (BAS) and by participant-reported akathisia (high quality evidence), with more people requiring treatment with anticholinergic medication (risk ratios (random effects) 3.28, 95% CI 1.82 to 5.91; two studies; N = 730, high quality evidence). Aripiprazole also led to more gastrointestinal disturbances (nausea (high quality evidence), and constipation) and caused more children/adolescents to have a prolactin level that fell below the lower limit of normal. Significant heterogeneity was present in the meta-analysis of movement disorders associated with aripiprazole and other treatments and was most likely due to the different side effect profiles of lithium and haloperidol. At the three-week time point, meta-analysis was not possible because of lack of data; however, at 12 weeks, haloperidol resulted in significantly more movement disorders than aripiprazole, as measured on the SAS, the BAS and the Abnormal Involuntary Movement Scale (AIMS) and by participant-reported akathisia. By 12 weeks, investigators reported no difference between aripiprazole and lithium (SAS, BAS, AIMS), except in terms of participant-reported akathisia (RR 2.97, 95% CI 1.37 to 6.43; one study; N = 313).