Abstract
While lithium is generally considered to be a first‐line treatment for mania, the position of the anticonvulsants and the antipsychotics in the treatment of this disorder is currently under debate. For that reason, this paper reviews the original literature, in particular addressing the randomized controlled trials (RCTS) on lithium, anticonvulsants and antipsychotics, and the methodological limitations therein. As the treatment of mania needs to anticipate the future course of the illness, the data on prophylaxis will also be reviewed, albeit briefly. It is concluded that antipsychotics are powerful antimanics, which are particularly beneficial for some clinical presentations of severe mania. However, in general their use should not be prolonged into the maintenance phase. Lithium is still to be considered the mood‐stabilizing drug par excellence, although it may be insufficient in mixed states and severe mania. The evidence for antimanic efficacy of valproate, in particular for mixed states, seems more convincing than that for carbamazepine, while the evidence for a prophylactic action of carbamazepine still exceeds that for valproate. Adjunctive treatment with benzodiazepines is often useful. Small sample sizes, highly selected study populations and high drop‐out rates seem to be the most important limitations of the RCTS on mania. Quasi‐experimental, naturalistic studies on unselected populations are needed to investigate how the various treatments work in clinical practice. Based on the available evidence, summary guidelines for treatment are proposed.