Can Antidepressants Cause Problems for Bipolar Patients?

In one of the first research papers on an antidepressant, “The Treatment of Depressive States with G22355, Imipramine Hydrochloride,” the author noted that “in individuals who are predisposed” the drug could “give rise to manic-like states or even a manic state.” This observation has been thoroughly confirmed since that paper was published in 1958, and the “predisposed individuals” are now recognized to be persons with bipolar disorder.

Source: Jéshoots/Pexels

Antidepressants have been implicated in a whole range of problems for bipolar patients who take them. Psychiatrists spend a lot of time trying to persuade bipolar patients already taking antidepressants to stop and patients who request one not to start. Why?

As that early paper reported, full-blown manic episodes can be triggered by antidepressants, especially by the older tricyclic drugs, but SSRIs, SNRIs, and other antidepressants have also been reported to trigger mania.

More controversial is the assertion that antidepressants, when taken by bipolar patients over the longer term, cause an increase of mood cycling, with more frequent episodes of mania, hypomania, and depression, too. Talk about counter-intuitive!

In a classic research article from the Brentwood VA hospital published in 1995, experienced clinicians reconstructed the course of illness of 51 “treatment-refractory” patients using all available sources: medical records (some over 30 years old), patient interviews, and interviews of relatives. Carefully assessing the timing of treatment with antidepressants and the onset of mood episodes, they determined how often an increase in cycling could be attributed to starting on an antidepressant.

They found that over a quarter of the patients had a worsening of their course of illness (more frequent mood episodes) that could be attributed to antidepressants. They also reported that over a third of these patients had experienced an antidepressant-induced manic episode.

A number of meta-analysis papers have appeared on the subject but have shed little light on this controversy. A 2016 meta-analysis published in Lancet Psychiatry included 1383 patients in six studies and concluded that longer-term antidepressant use (up to 52 weeks) increased the risk of mania. Another meta-analysis that appeared just several months later in the Journal of Affective Disorders included 692 patients in 11 studies and came to the opposite conclusion, stating that they found “No significantly increased risk of new manic/hypomanic episodes.”

A 2008 paper that came out of the STEP-BD study, a long-term observational study of patients with bipolar disorder identified a more subtle but nevertheless disabling syndrome that they attributed to antidepressants. They had observed that after some patients with bipolar disorder take antidepressants and appear to recover from depression, they start having more problems getting to sleep, notice more of the feelings of tense uneasiness that psychiatrists call “dysphoria,” and experience a smoldering irritability that is with them much of the time. The STEP-BD researchers coined a term for this combination of symptoms, calling it “ACID,” short for antidepressant-associated chronic irritable dysphoria. They found that these patients were often quite impaired by these symptoms.

The researchers had already suspected that antidepressants were the cause of the problem, so they looked through the records of all STEP-BD patients with this triad of chronic irritability, dysphoria, and sleep disturbances, noting who had or had not taken an antidepressant along with a mood stabilizer. The results were unequivocal. They found that the patients with these symptoms who had taken an antidepressant outnumbered patients taking only a mood stabilizer by 10 to 1.

One might ask whether the risk of taking an antidepressant might be outweighed by the benefit of treating severe depression symptoms. The preponderance of the evidence appears to show that this is not the case—at least in groups of patients. Another STEP-BD paper published in the New England Journal of Medicine found that after a year, antidepressants were no more effective than placebo in treating depression symptoms in bipolar patients. The Lancet meta-analysis discussed earlier came to this same conclusion but another meta-analysis from the American Journal of Psychiatry found that “Antidepressants are effective in the short-term treatment of bipolar depression.”

So what can practicing psychiatrists and their patients conclude from all this? Dr. Fred Goodwin, former director of NIMH summed it up in a paper that predates nearly all those referenced above, saying in 1987: “On balance, the available evidence suggests that some bipolar patients become manic, and a few experience rapid cycling, when they are treated with antidepressants.” Obviously then, the converse is also true: Some bipolar patients do not experience these problems from antidepressants. The meta-analysis studies that lump these two groups together will find for or against antidepressants depending on how they choose which studies to analyze.

I think a fair approach is to avoid antidepressants as much as possible in bipolar patients but be willing to try them as a last resort and then discontinue them as soon as possible. And throughout the process, keep a sharp eye out for trouble.