Antidepressant monotherapy in pre-bipolar depression; predictive value and inherent risk

Objective: To identify specific treatment-emergent symptoms in response to antidepressant therapy in depression preceding bipolar disorder. Methods: Retrospective chart review of response to antidepressants in "pre-bipolar" depression, compared to a matched unipolar sample. Results: Family history of completed suicide (p = 0.0003) and bipolar disorder (p = 0.004) were more common in the pre-bipolar subgroup. Earlier age of onset of diagnosed depression (p = 0.005) as well as even earlier episodes of untreated retrospectively diagnosed major depression (p < 0.0001) were associated with a future bipolar course. The pre-bipolar group was less likely to respond to antidepressant treatment (p = 0.009). Treatment-emergent "mixed" symptoms (two or more symptoms of DSM IV mania, mood lability, irritability/rage with co-existing depression) and in particular, "serious symptoms" (treatment emergent or increased agitation, rage or suicidality) occurred more commonly in the bipolar group. The two variables that best accounted for the between-group differences in logistic regression, were early age at first symptoms of depression and treatment-emergent agitation. Conclusions: Family history of completed suicide and/or bipolar disorder, early onset of depressive symptoms as well as treatment-emergent "mixed" symptoms are common in depression preceding the diagnosis of bipolar disorder.