Clinical features of antidepressant associated manic and hypomanic switches in bipolar disorder.

The present study investigated possible clinical differences between bipolar patients with and without manic or hypomanic switch during antidepressant (AD) treatment. The authors undertook a retrospective assessment of 169 individuals affected by bipolar disorder type I (BP I: n=96) and II (BP II: n=73) who experienced at least one manic or hypomanic episode following depression without any interposed normothymic period ("manic switch") during AD therapy. They were compared with a sex, age (+/-5 years), and ethnicity-matched group of 247 subjects, randomly selected from our pool of bipolar subjects who have never had manic switches. Only 2 of the 169 patients had had spontaneous switches before the AD-related one. Switched subjects were marginally older (t=-2.65, df=414, P=.008) compared to not switched and less frequently delusional (chi2=13.86, P=.0002). Polarity of the onset episode was more frequently depressive in switched patients (chi2=21.93, P=.00002), which had also less previous manic episodes than not switched (t=3.44, df=332, P=.0006). Those differences were more pronounced in the BP I subsample. Switched patients were more frequently BP I (chi2=29.66; P<.00001). Maintenance with mood stabilizers appears to be a strong protective factor; in fact, of the 124 individuals undertaking a mood stabilizer therapy, 21 had a switch and 103 had no switches (chi2=41.10, P<.000001). In conclusion, some clinical variables, such as the number of manic episodes, the presence of delusions, the polarity of onset episode, and the mood-stabilizing treatment, may be involved in AD-related switches. Further studies are required to investigate the causal relationships between those factors.