Research in context
Evidence before this study
The use of antidepressants in the treatment of bipolar depression is a controversial area in psychiatry. We searched MEDLINE from inception to Jan 31, 2016, for meta-analyses using the search term “(bipolar disorder OR bipolar depression) AND antidepressive agents AND meta-analysis”. We found four previous meta-analyses of antidepressants in bipolar depression that included agents and interventions that are discouraged—notably, antidepressant monotherapy, tricyclic antidepressants, and monoamine oxidase inhibitors. Therefore, the clinical relevance of existing evidence is unclear, and additional examination restricted to trials of second-generation antidepressants adjunctive to adequate mood stabilisation, a strategy now widely used by clinicians, is necessary.
Added value of this study
Our systematic review identified six randomised placebo-controlled trials representing a total of 1383 patients treated with a second-generation antidepressant in addition to a mood stabiliser or an antipsychotic. Random-effect analyses showed a small benefit of antidepressant treatment in reducing depressive symptoms, and no evidence of acute risk of treatment-emergent affective switch. However, treatment with second-generation antidepressants over 52 week extension periods was associated with an increased risk of affective switch despite adequate mood stabilisation.
Implications of all the available evidence
Concomitant to adequate mood stabilisation, time-limited use of second-generation antidepressants is associated with a small clinical benefit and no acute risk of treatment-emergent mania or hypomania in the short term. However, prolonged treatment should be avoided.