[Systematic review to evaluate the efficacy, acceptability and safety of second-generation antipsychotics for the treatment of unipolar and bipolar depression.]

BACKGROUND: The World Health Organization (WHO) estimates that depression affects about 121 million people in the world and in terms of years of illness, by the year 2020 could become the second most prevalent disease in the world population after cardiovascular diseases. Second-generation antipsychotics (SGAs) seems to induce remission in depression when added to an antidepressant. AIM: To evaluate the efficacy and safety of SGAs for the treatment of depression.
METHODS: We searched five bibliographic databases. We assessed the quality of evidence using Cochrane and GRADE criteria.
RESULTS: We included 42 RCTs. Where possible, we made a statistical synthesis of results. For efficacy outcomes, in direct comparisons in unipolar and bipolar patients with acute depressive episode, and in unipolar patients who did not respond to previous treatments with antidepressants (non-responders), SGAs gave better results than placebo, moderate to low certainty of evidence. In the comparison with antidepressants, in unipolar and bipolar patients with acute depressive episode the evidence was in favour of SGAs high certainty of evidence; while in the non-responder unipolar patients (the included studies considered only this typology of patients) the evidence was in favour of the antidepressants, low certainty of evidence. For safety outcomes, the results were in favour of placebo for patients with at least one adverse event, and in favour of SGAs for the number of patients with serious adverse events, for both comparisons the certainty of evidence was moderate. Comparing the SGAs with antidepressants, no differences were observed for patients with at least one adverse event, high certainty of evidence; while serious adverse events were less acute in patients treated with SGAs, moderate certainty evidence. The results of direct and indirect comparisons made with the network meta-analysis showed no differences for most of the outcomes considered, not showing a clear superiority of a drug compared to the others.
CONCLUSIONS: These results showed a moderate effect in favour of SGAs compared to antidepressants in patients, unipolar and bipolar, with a new acute depressive episode and confirm that in patients non-responders, antidepressants may remain more effective.