Antidepressants in Bipolar Depression: To Use or Not to Use?

October 26, 2018

ORLANDO, Fla.—A friendly debate between 2 thought leaders followed by a vote at Psych Congress 2018 exhibited the divide that persists on a controversial topic in mental health: whether antidepressants should be used in the treatment of patients with bipolar depression.

S. Nassir Ghaemi, MD, MPH, Professor of Psychiatry, Tufts University, Boston, Massachusetts, went up against Joseph F. Goldberg, MD, Clinical Professor of Psychiatry, Icahn School of Medicine at Mount Sinai, New Rochelle, New York.

After they presented their cases for 15 minutes each and weighed in on 3 discussion questions, attendees used their mobile devices to vote on whose argument was more persuasive. Dr. Goldberg received 132 votes (50.6%), just 3 more than Dr. Ghaemi’s 129 votes (49.4%).


    S. Nassir Ghaemi, MD, MPH

Dr. Ghaemi, who has focused on the issue throughout his academic career, said the answer is so clear it shouldn’t even be a debate any longer.

Dr. Ghaemi has focused on the use of antidepressants in bipolar depression throughout his 25-year academic career, receiving 2 grants totaling more than $2 million from the National Institute of Mental Health to study the issue over 10 years. He said the answer is so clear it shouldn’t even be a debate any longer.

“I have proven that the drugs don’t work and I’ll show you the data,” said Dr. Ghaemi, who is also a lecturer on psychiatry at Harvard Medical School in Boston. “The real question is not what the right answer is. The real question is why do we still have to debate it when the evidence is so clear.”

The core of the data he presented was a meta-analysis of all randomized, placebo-controlled trials on antidepressants in bipolar depression, which he coauthored. It covered 6 studies (3 industry-sponsored), most using newer drugs, and was published in The Lancet Psychiatry in 2016.
“Generally they find very little benefit, if any, with antidepressants in bipolar depression,” he said.

He described the review’s overall Cohen’s d effect size of 0.16 as “very small,” and noted the effect size of antidepressants in major depressive disorder is 0.3. Generally speaking, .5 is considered the threshold for showing a clinically meaningful difference, he noted.

“This is really the end of the story,” Dr. Ghaemi said. “The only thing you can say about this is we have 6 studies instead of 300 like we do with antidepressants in MDD.”

He also argued there is some evidence that antidepressants could cause long-term manic episodes in people with bipolar depression, and could worsen the condition of people with rapid cycling.

In addition, he believes many studies of antidepressants in people with bipolar type 2 depression have flawed enriched designs. “Don’t believe these bipolar type 2 studies. You need to dig into the weeds a little bit and you’ll see that they don’t prove anything,” he told attendees.

He urged attendees not to ignore the research and rely on their clinical experience, as confounding factors may influence outcomes they see in practice. Randomized controlled trials equalize those confounding factors, he said.

“If the randomized trials show you the drugs mostly don’t work or are mostly harmful … that means you shouldn’t give it to most people,” he said, noting studies show 50% to 75% of people with bipolar disorder are prescribed antidepressants, even though none of them is approved by the US Food and Drug Administration for use in bipolar disorder.

Symptom improvements clinicians may see when a patient with bipolar disorder is on an antidepressant may be just due to the passage of time or the end of an episode, he added.

Dr. Ghaemi said he didn’t expect to change the minds of people who have already made up their minds on the topic, but hoped he could have an effect on younger people who have not.

“People won’t change their minds but maybe generations will,” he said.

A better debate, Dr. Ghaemi said, would be on using antidepressants in any type of depression.

“That’s the debate we should be having because this one’s obvious,” he said. “That one would take a little more discussion. Maybe it’s something we can do next year.”


    Joseph F. Goldberg, MD

Dr. Goldberg said he agreed with a number of Dr. Ghaemi’s points, but challenged other arguments, as well as the overall debate question.

“I actually would like to take this opportunity to suggest that we think about this whole issue from a different standpoint, which is not ‘are antidepressants good or bad for the diagnosis of bipolar disorder’ per se, but rather to reframe the question: ‘under what circumstances might antidepressants be more likely to be helpful or to be harmful?’ ” he said.

“The concerns really have to do not with ‘are antidepressants good or bad.’ I think such a black and white split is very disingenuous.”

Bipolar disorder is not homogenous, and calls for different approaches in different patients, as is the case in patients with conditions such as cancer, hypertension, or an infectious disease, Dr. Goldberg said. He agreed with Dr. Ghaemi that the signal of antidepressants in bipolar disorder is weak, but said he believes the literature shows they may have value in a subgroup of patients.

He pointed out that only a few of the 30 or so antidepressants have been studied in bipolar depression, and certain subtypes of them have not been studied at all.

“We can’t generalize about class based on just a few findings,” he argued. “Class generalizations, I think, are a very risky thing. And if we are inclined to stop studying a class based on negative findings—admittedly of the first handful of drugs that have been studied—the field’s not going to advance.”

He suggests considering a number of factors when considering whether to use antidepressants for a patient with bipolar depression. Good candidates may be those with bipolar type 2, no rapid cycling, no recent mania or hypomania, no comorbid alcohol or substance use disorders, prior favorable antidepressant response, no history of antidepressant-induced mania, and/or the SLC6A4 l/l genotype.

“We really have to think not just of the diagnosis, but of the patient that has it,” he said.

—Terri Airov


“The great debate in contemporary psychiatry: the use of antidepressants in bipolar depression.” Presented at Psych Congress 2018: Orlando, Florida; October 26, 2018.