Bipolar Depression: Antidepressants

By | March 27, 2015

Tricyclic Antidepressants

Literature is sparse on controlled studies of tricyclic antidepressants for bipolar depression. Tricyclic antidepressants are more effective in the treatment of unipolar depression than in bipolar depression. The risk of switching patients from bipolar depression to mania and of shortening cycle lengths makes tricyclic antidepressants unattractive choices for the treatment of bipolar depression.

Monoamine Oxidase Inhibitors

Monoamine oxidase inhibitors (MAOIs) are more effective in the treatment of bipolar depression than in unipolar depression. This may be due in part to the superior efficacy of MAOIs in treating patients with anergy and reversed neurovegetative symptoms of hypersomnia and hyperphasia. Patients who respond often have marked improvements.

Newer Generation Antidepressants

There are surprisingly few studies of the use of selective serotonin reuptake inhibitors (SSRIs) and other newer generation antidepressants in bipolar depression.


Paroxetine is the most well studied of the SSRIs. Three double-blind, controlled studies of paroxetine were conducted versus placebo, imipramine, venlafaxine, and combined lithium and divalproex. In the one large multicenter study of paroxetine plus lithium versus imipramine plus lithium versus lithium monotherapy, there were no differences between the treatment groups. The other studies were not powered to establish significant drug-drug or drug-placebo differences.

A double-blind, randomized study by Young et al. compared the efficacy of adding paroxetine (36 mg/day) to divalproex sodium with adding a second mood stabilizer to divalproex in 27 patients with bipolar depression types I and II. Patients treated with the combination of paroxetine and divalproex did not differ in efficacy from those treated with the add-on lithium. However, paroxetine was significantly better tolerated than the combination of two mood stabilizers, suggesting that adding paroxetine to a mood stabilizer may have clinical utility than adding a second mood stabilizer.

Vieta et al. administered paroxetine or venlafaxine to 60 patients in a 6-week, single-blind, randomized trial. All patients experienced breakthrough depression while being treated with a mood stabilizer and were in a current depressive episode of bipolar disorder. Although significant improvements were observed in patients treated with both paroxetine and venlafaxine, there were no differences between the groups. One patient switched into mania while on paroxetine, and four patients switched while on venlafaxine.


Citalopram, added to lithium, divalproex, and/or carbamazepine, was studied in 45 patients on an open-label basis. Sixty-four percentage of the patients responded within 8 weeks. In general, the patients tolerated the citalopram well.

Despite the lack of evidence of efficacy from large, placebo-controlled, randomized, double-blind clinical trials, SSRIs are used widely in the treatment of bipolar depression. Many clinicians believe that they are effective and have a lower switch rate than tricyclic antidepressants and MAOIs.


Several open-label studies and case reports have suggested that bupropion is effective in the treatment of bipolar depression, with a significantly lower switch rate than other antidepressants. A study by Sachs et al. found a substantially lower switch rate in bipolar patients treated with bupropion (10%) vs. desipramine (50%), with equal response rates between the two groups. However, the results of other studies have suggested that bupropion-treated patients do not have lower switch rates than patients treated with other antidepressants.


Venlafaxine has been studied in the treatment of bipolar depression in several studies. In a previously mentioned study, patients were randomized to paroxetine (starting at 20mg/day) or venlafaxine (at 75mg/day) for 6 weeks. The mean daily dose of venlafaxine was 179mg/day and of paroxetine 32mg/day. There were no differences between the treatment groups on reduction of the HAM-D score (the mean HAM-D change for paroxetine was — 6.9 and for venlafaxine — 9.0). Response rates were 43% for paroxetine and 48% for venlafaxine. Patients on venlafaxine experienced increased rates of dry mouth and insomnia, whereas those on paroxetine had more nausea and headache. Thirteen percentage of the venlafaxine patients switched to hypomania or mania, whereas 3% of the paroxetine patients did so.

In another study, patients with breakthrough major depressive episodes who were on ongoing adequately dosed mood stabilizers were randomized to bupropion, sertraline, or venlafaxine. The primary question in the study was switch rate into mania or hypomania on these antidepressants as a group. Therefore, individual response rate for each drug was not analyzed. However, an acute clinical response rate to these antidepressants as a group was 50% on CGI-Depression for Bipolar Illness. In the 1-year continuation phase, the response rate of much or very much improved was 44%. In the acute 10-week phase, 9% of trials were associated with switched into mania or hypomania, and another 9% with a week of more of hypomania alone (with no to minimal dysfunction).

Another study compared venlafaxine XR with lamotrigine when added to a mood stabilizer in the treatment of bipolar I and II patients with major depression for 8 weeks. Mean HAM-D-17 and CGI-I rating scores improved significantly from baseline to endpoint in both treatment groups. However, there were no differences in reduction in scores between the two agents. Response and remission rates were also equivalent.