02. Antidepressants in Bipolar Depression: Addressing The Controversy

Text version

Do you prescribe antidepressants for people living with bipolar? Whether you say yes or if you are one of the few who say no, we all are frequently faced with important antidepressant questions.

• How risky are antidepressants in people with bipolar?
• How well do they work?
• Could they only be used short term?

A new review by Lars Kessing, a highly respected Danish mood specialist, provides a sober, balanced comparison of research evidence and clinical practice regarding antidepressants for people with a bipolar diagnosis.

Antidepressants Efficacy in Bipolar Depression

First, do antidepressants even work in bipolar depression? Kessing confirms antidepressants are better than placebos. Interestingly, the effect size in bipolar depression ranging in studies from 0.2 to 0.4 in different meta-analyses is very close to the effect size in unipolar depression, about 0.3 (you’ll recall that an effect size of 1.0 is large; 0.5 is medium; 0.3 is a small effect and 0.2 perhaps not even likely to be clinically significant).

For comparison, Kessing cites data on lithium’s efficacy in bipolar depression: a few studies and low confidence but a mean effect size of 0.18, about 0.2.

And lamotrigine, across 11 randomized trials, 6 were negative, but in meta-analysis the effect size is not zero. It’s 0.16, lower than lithium.

Atypical Antipsychotics Show Similar Efficacy

To complete the picture, how about effect sizes for atypical antipsychotics? Basically, it’s about 0.3 for the ones that you’d expect: quetiapine, olanzapine and lurasidone. For cariprazine, 0.2.

And if anything, quetiapine is slightly better than the rest at 0.35 and not a single negative trial. You might remember aripiprazole did not prove better than placebo and so did not get an FDA indication for bipolar depression as the others did, likewise ziprasidone.

Similar Efficacy, Different Risks

So summarizing thus far, the effect size of all the usual suspects is not very large: quetiapine has a slight edge in efficacy; too bad about the very high incidence of weight gain. Lamotrigine has the poorest efficacy data but the lowest incidence of side effects and no significant long-term risks.

The main point: antidepressants do have short-term efficacy in bipolar depression about the same as lithium and a little shy of the main atypicals that you might consider. Overall, the risk-benefit balance is tipped more by risk than efficacy.

CANMAT Recommends Eight-Week Limit

Before turning to the risk side of Kessing’s essay, what about long-term use of antidepressants in bipolar depression? Interestingly, Dr. Kessing does not address that question.

So let’s turn to another source, the Canadian Mood and Anxiety Treatment (CANMAT) Guidelines, which recommend discontinuing antidepressants within eight weeks of starting them. Does that surprise you? It certainly surprised me because it seems like a lot of people living with bipolar are taking an antidepressant. So I asked OpenEvidence.com for a percentage. And the answer: it’s at least 50%. About one person in two living with bipolar is taking an antidepressant.

But clinicians will understand how this might happen. Bipolar depression can overlie other reasons for depression: trauma history, attachment problems, depressive temperament, or depressing social circumstances like unemployment, social isolation, and institutional barriers. So once the bipolar reason for depression is addressed, significant depression symptoms could persist. A pill called an antidepressant seems an obvious consideration especially if relevant psychotherapies and lifestyle changes are out of reach as they are for so many. And once underway with an antidepressant, if depression hasn’t fully remitted, it’s hard to justify taking it away.

Antidepressants Nearly Double Mania Risk

So backing up just a little bit, the reason for the CANMAT recommendation to taper by eight weeks: it’s not because they’ll later stop working. It’s based on risk, not benefit.

The authors cite among many others an analysis of 35 studies which found that the one-year incidence of a manic recurrence was almost doubled when comparing patients who remained on an antidepressant versus those who tapered off. To be more precise, the relative risk was 1.8.

Which brings us to Dr. Kessing’s assessment of antidepressant risks in his new essay: suffice to say that he too is concerned about antidepressants’ potential destabilizing influence. He cites a string of studies suggesting antidepressants increase the risk of having manic symptoms, though he does note one contrary meta-analysis by Oliva et al. (2025) which I reviewed in a recent Quick Take.

Bottom Line: Risk Outweighs Benefit Long Term

To summarize then, if you’ve already concluded that antidepressant risks outweigh their potential benefits especially for long-term use, you’ll find little to change your views in this Kessing essay. If on the other hand, you frequently prescribe antidepressants beyond eight weeks for patients with bipolar disorder, note that the CANMAT Guidelines cited by Dr. Kessing did not support that approach.

On the other hand, the best practices for managing depression from other causes like trauma or poverty in a patient with bipolar are not addressed here.

---

References

1. Kessing L. V. (2025). The use of antidepressants for bipolar disorder – a controversy between science and clinical practice. Danish medical journal, 72(12), A07250539. https://doi.org/10.61409/A07250539
2. Yatham, L. N., Arumugham, S. S., Kesavan, M., Ramachandran, K., Murthy, N. S., Saraf, G., Ouyang, Y., Bond, D. J., Schaffer, A., Ravindran, A., Ravindran, N., Frey, B. N., Daigneault, A., Beaulieu, S., Lam, R. W., Kondapuram, N., Reddy, M. S., Bhandary, R. P., Ashok, M. V., Ha, K., … BEAM-BD Trial Group (2023). Duration of Adjunctive Antidepressant Maintenance in Bipolar I Depression. The New England journal of medicine, 389(5), 430–440. https://doi.org/10.1056/NEJMoa2300184
3. Tondo, L., Vázquez, G., & Baldessarini, R. J. (2010). Mania associated with antidepressant treatment: comprehensive meta-analytic review. Acta psychiatrica Scandinavica, 121(6), 404–414. https://doi.org/10.1111/j.1600-0447.2009.01514.x
4. Oliva, V., De Prisco, M., La Spina, E., Paolucci, S., Fico, G., Anmella, G., Hidalgo-Mazzei, D., Murru, A., Pompili, M., Fornaro, M., Solmi, M., Yildiz, A., Leucht, S., Vieta, E., & Radua, J. (2025). Switch to mania after acute antidepressant treatment for bipolar depression: a systematic review and network meta-analysis of randomised controlled trials. EClinicalMedicine, 87, 103413. https://doi.org/10.1016/j.eclinm.2025.103413

Abstract

The use of antidepressants for bipolar disorder – a controversy between science and clinical practice

Lars Vedel Kessing

Depressive episodes are more challenging for patients and clinicians than other bipolar disorder episodes. The reasons for this include: 1. the prevalence of depressive episodes is higher than for other episodes; 2. functioning; 3. cognition is more impaired; 4. suicide is more prevalent; 5. and treatment is more complex. This paper aims to highlight the controversy between the poor evidence from science on the effects of antidepressants for bipolar depression versus the high use of antidepressants in clinical practice, and to specify the clinical role of antidepressants in bipolar disorder in relation to other drugs.

Reference

Vedel Kessing, L. (2025). The use of antidepressants for bipolar disorder – a controversy between science and clinical practice. Danish Medical Journal; 72(12):A07250539.