Lower rate of depressive switch following antimanic treatment with second-generation antipsychotics versus haloperidol

BACKGROUND: Treatment of acute mania with second-generation antipsychotics has been claimed to involve a lower risk of switch to depression than haloperidol. However, clinical guidelines clearly state that this is not a proven fact. METHODS: Meta-analysis of double-blind randomized controlled trials in acute mania, comparing rates of switch to depression with atypical antipsychotics and with haloperidol. Search was conducted in MEDLINE and CENTRAL databases (last search: September 2011). RESULTS: 8 randomized clinical trials fulfilled inclusion criteria. 2 of them were excluded because of low methodological quality or lack of data. 5 second-generation antipsychotics (aripiprazole, olanzapine, quetiapine, risperidone, and ziprasidone) were compared to haloperidol. In the mixed effects model the Risk Ratio for depressive switch was 0.71 (0.52, 0.96) favouring atypical antipsychotics. In the random effects model the difference did not reach statistical significance. In the heterogeneity analysis, exclusion of an outlying aripiprazole trial yielded a Risk Ratio of 0.58 (0.42, 0.82) with a non-significant heterogeneity test. Although no atypical antipsychotic was individually significantly superior to haloperidol, a trend could be seen favouring olanzapine (RR=0.56 [0.29, 1.08]), quetiapine (RR=0.36 [0.10, 1.33]), and ziprasidone (RR=0.51 [0.22, 1.18]). LIMITATIONS: All trials were industry supported, with some variability in dosage of haloperidol. Switch to depression was not the primary outcome of the trials. Heterogeneity could be explained as a lack of class-effect for atypicals. CONCLUSIONS: Treating acute mania with atypicals is associated to 42% less risk of switch to depression than with haloperidol. Nevertheless, caution should be taken when considering this a class effect, as only olanzapine, quetiapine, and ziprasidone may show a better profile. Copyright (c) 2012 Elsevier B.V. All rights reserved.