Drug-induced switching in bipolar disorder: Epidemiology and therapeutic implications

The use of antidepressants in patients with unipolar depression is associated with a negligible rate of switching into mania. In contrast, in patients with bipolar disorder, even those treated concomitantly with a mood stabiliser, such switching appears to represent a greater problem. Preliminary evidence from some controlled studies suggests that antidepressants may double the incidence of a switch, from some 25% in patients receiving placebo to 50% in those receiving tricyclic antidepressants. Other series in patients with rapid cycling bipolar disorder suggest an even higher switch rate; however, only about one-third of the observed switches are likely to be attributable to the antidepressant (i.e. unrelated to the natural course of illness). In addition to these phenomena, cycle acceleration has been observed consistently in a subgroup of patients with bipolar disorder who are receiving maintenance treatment with tricyclic antidepressants, and this has been verified by periods of discontinuation and reintroduction of the drug. Cycle acceleration appears to occur in approximately one-fifth of patients with refractory bipolar illness. Given these potential liabilities, it is unclear as to what is the most judicious treatment algorithm to follow for the bipolar patient with depression that is breaking through treatment with a mood stabiliser. In the absence of adequate studies, our own algorithm is to use antidepressants judiciously in nonrapid cycling patients, but to relatively avoid them in rapid and ultra-rapid cyclers. Instead, we prefer to use adjunctively a second mood stabiliser prior to the introduction of a unimodal antidepressant. Systematic controlled clinical trials are eagerly awaited, comparing not only the acute antidepressant efficacy of the newer antidepressants, but their liability in causing a switch or cycle induction when introduced into long term prophylaxis. With the advent of a series of new potential mood stabilising agents, direct comparison of the addition of a second mood stabiliser with that of a unimodal antidepressant will help establish the comparative risk: benefit ratios of these differential approaches. It is also hoped that clinical and biological markers of differential antidepressant responsivity will ultimately be identified and help to refine the process of choosing the best antidepressant modality for depression breaking through prophylaxis with mood stabilisers in patients with bipolar disorder.