Clear, engaging, and practical updates on clinical psychiatry.

Earn CME for listening to the podcast with a multimedia subscription. Listen for free here or using Apple Podcasts, Android, or Stitcher.

Previous Post
Next Post

What Works in Bipolar Depression

Podcast, Volume , Number ,

Print Friendly, PDF & Email

From psychotherapy to ECT, pramipexole to lamotrigine, a walk through 20 therapies for bipolar depression.

CME: Podcast CME Post-Tests are available using this subscription. If you have already enrolled in that program, please log in.

Published On: 04/11/2022

Duration: 25 minutes, 18 seconds

Related Article:A Practical Guide to Light Therapy,” The Carlat Psychiatry Report, November 2019

Chris Aiken, MD,  and Kellie Newsome, PMHNP, have disclosed no relevant financial or other interests in any commercial companies pertaining to this educational activity.


Today, we bring you 20 therapies that passed in at least one randomized-controlled trial of bipolar depression.

Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. I’m Chris Aiken, the editor in chief of the Carlat Report. And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.

KELLIE NEWSOME: Bipolar depression is not easy to treat. Antidepressants can trigger mania, mixed states, and rapid cycling, and most studies suggest they don’t even treat bipolar depression, including a new meta-analysis of 19 controlled trials released earlier this year. That leaves us with a hodge-podge mix of mood stabilizers and novel therapeutics to draw from, none of which have a balance of safety and efficacy that we are very comfortable with. Today, we’ll walk you through those options, and explore what to do in treatment resistant cases.

But first, a preview of the CME quiz for this podcast

Which antipsychotic is effective in bipolar depression?

A. Quetiapine

B. Olanzapine

C. Risperidone

D. Brexpiprazole

CHRIS AIKEN: Timing is everything when it comes to bipolar depression, and it’s the first thing you want to think about when assessing the patient. When did the depression begin? Did it come on after a manic episode? Is it part of a rapid cycling picture where the patient goes in and out of episodes every few months?

Each of those is going to shape the treatment plan. 

Post-manic depression

Depressions that start after a manic episode – I call these post-manic depressions – are difficult to treat. These patients usually have low energy and high inertia, anhedonia, and psychomotor slowing. They may have guilty ruminations of manic misdeeds. It’s as if the fire of mania has gone out and all that is left is the ash. 

In post-manic depression, the first goal is to prevent future manias – as these are what caused the problem to begin with. Lifestyle interventions are also critical – these patients are often in bed all day, and regulating their circadian rhythm with morning light, regular wake times, and evening darkness is a good first step, followed by mild exercise like brisk walking if you can get past that first step.

When it comes to medications you have to be careful, because the closer the patient is in time to a manic episode the more likely those meds are to trigger another episode. I’d avoid antidepressants for that reason, but even interventions that have a very low risk of mania, like pramipexole, the modafinils, and the antipsychotics, should be used carefully. For those 3 meds the risk of mania is so low that it doesn’t show up in controlled trials – only in case reports.

Rapid cycling

KELLIE NEWSOME: The next type of bipolar depression is rapid cycling. This is when the patient has at least 4 distinct episodes in a year, and they need not cycle between mania and depression; just having 4 episodes of depression will qualify. Maybe those with purely depressive rapid cycling really have some hypomania going on between the episodes that’s hard to detect during the interview, but nobody knows for sure.

Consider rapid cycling when your patient seems to vacillate between better and worse at each appointment, or when every medication you try seems to work at first – because they were going to cycle out of the episode anyway – only to leave them in another episode soon after.

If your patient has rapid cycling, you need to shift the focus of your treatment from the current episode to the rapid cycling pattern. I’ll tell the patient “Normally I try to help you with your current symptoms, but what’s going on right now is different – you’re having rapid cycling – which means your mood changes every few weeks or months. The depression you’re having is caused by this cycling, and we’re going to have to address that cause to get you better. We need a map to help us get out of this, and the best map is a mood chart, where you track your moods every day or week. If our treatment works, you’ll still have days of depression, but those days will become less intense and more spaced out over time. Right now they are happening every 2 weeks, so we want to get them down to every month, and then every 2-3 months until they fade away.”

We keep several versions of mood charts on our practice website at One tip: Most patients find daily mood charting overly burdensome, and you don’t need to know their mood every day of the week. I prefer to use the weekly mood chart on that site. It gives you the bigger picture, and prevents another problem. Patients often rate their emotions rather than their mood, and here’s a surprise – several studies have found that the most chief complaint during mania is – depression. Yes, mania may start with a bit of euphoria but that quickly turns to anxiety, irritability, and a feeling that you’re losing your mind which is downright depressing. Anyway, this weekly mood chart gets around that by having patients rate symptoms of mania, depression, and mixed states with a 10-item scale so they can chart their ups and down more accurately.

CHRIS AIKEN: Rapid cycling tends to last 3-5 years on average, and often resolves on its own, which is a euphemistic way of saying “This is very hard to treat.” You’ll want to address some of the modifiable causes of rapid cycling, which are:

  1. Substance use
  2. Thyroid abnormalities
  3. Circadian rhythm abnormalities, particularly irregular wake time and nocturnal lights
  4. Antidepressants

Antidepressants may be more likely to cause rapid cycling than overt manic switches. In fact, they rarely trigger mania when given with a mood stabilizer, so rarely that it didn’t show up in most of the controlled trials that used that strategy. But even when those studies added them in with a mood stabilizer on board, they were able to detect a distinct increase in cycle frequency. 

In terms of treatment, anticonvulsants like valproate, carbamazepine, and lamotrigine may work better in rapid cycling than lithium, but lithium should not be avoided, and in some studies it worked well for these cases in combination with carbamazepine. The atypical antipsychotics can also reduce rapid cycling, particularly quetiapine (Seroquel) and aripiprazole (Abilify).


KELLIE NEWSOME: Stress plays a big role in bipolar depression. The illness itself is demoralizing, particularly the way it whittles away identity, causing people to lose jobs, educational opportunities, relationships, marriages, even custody of their children. Guilt, shame, self-stigma, and the rocky lives that these patients often live can get in the way of recovery from depression. Trauma is 3-4 times more common in bipolar disorder than it is in the general population, and there are at least 2 reasons why. The disorder is highly genetic, and if the parents had untreated bipolar disorder the patient may have suffered abuse or neglect. Then, as adults, the manic episodes may place them in risky situations where trauma is more likely. They drive fast, jump into unsafe relationships, and walk through the downtown streets in the middle of the night.

All that’s to say that psychotherapy is an important part of the treatment. And it’s also important for many of the comorbidities that go along with bipolar disorder. Depression may be driven by anxiety disorders, PTSD, OCD, personality disorders, substance use disorders, and eating disorders – all of which are more common in the bipolar population, and all of those comorbidities respond better to psychotherapy than medication. 

CHRIS AIKEN: Stress, rapid cycling, recent mania, psychiatric comorbidities – each of these is going to influence your treatment plan. We could add physical health into the mix, as cardiovascular and metabolic problems are more common in bipolar disorder, and these cause depression in their own way. But putting all these modifiers aside, what medication do we have that offer hope for these difficult-to-treat depressions? 

First is lithium. You’ll hear mixed things about whether lithium can treat bipolar depression, but here’s a tip: Lithium is very effective for classic bipolar disorder – the ones that cycle between mania or hypomania and depression, without a lot of mixed states or complicated comorbidities (with the exception of panic disorder – which is more common in lithium responders). Only about 1 in 3 people with bipolar disorder have such a classic, lithium responsive presentation, and they may be bipolar type I or II, but either way what distinguishes them is the purity of their episodes – with pure DSM symptoms that are cleanly separated by periods of normal mood. Lithium responders also tend to have post-manic depressions – their depressions usually come after their depressions.

Next is lamotrigine. This one is great for patients who need a tolerable option, but the downside is it can take 6-8 weeks to work. Although it’s very well tolerated, it does have a safety issue: 1 in 3,000 can develop a life-threatening rash during the titration. 

The atypical antipsychotics have the opposite profile. They work quickly, usually within 2 weeks, but have too many safety and tolerability issues to list in this podcast. 

KELLIE NEWSOME: The atypicals that work in bipolar depression are, in alphabetical order: 

  • Cariprazine (Vraylar)
  • Lumateperone (Caplyta)
  • Lurasidone (Latuda)
  • Olanzapine-Fluoxetine Combo (Symbyax)
  • Quetiapine (Seroquel)

Which to choose? Here’s a quick guide. In terms of effect size, quetiapine and olanzapine-fluoxetine combo are the strongest, lumateperone and lurasidone are in the middle, and cariprazine is near the bottom, but we’ve seen patients respond only to cariprazine so there’s no reason to overlook it. Quetiapine also has an added benefit of improving sleep and anxiety.

Unfortunately, the most effective are also the least tolerable, particularly when it comes to sedation and metabolic problems. Olanzapine-fluoxetine has the additional problem that it only works with the antidepressant on board, and even though this didn’t who up in the studies we’ve seen patients developed mania and mixed states on it that improved when the fluoxetine was taken away.

Novel Therapies

CHRIS AIKEN: Warning: At this point in the podcast we’ve exhausted all the FDA approved options for bipolar depression. But there are some novel therapies worth trying. We’ve tried them all, and here’s the ones we’ve had the most success with.

Pramipexole (Mirapex). This dopaminergic treated bipolar and unipolar depression in a few controlled trials, and in one study it worked in patients who did not respond to ECT. It doesn’t seem to cause mania, but it can cause a syndrome called hedonistic dysregulation, which is where pursue pleasure to such an unstoppable degree that it’s classified as compulsive: Gambling, shopping on Amazon, and compulsive masturbation are some of the ways it can show up. The same warning exists for aripiprazole, by the way, another dopamine D3 agonist.

Hedonistic dysregulation usually occurs without any other manic symptoms. It is rare on pramipexole and people with bipolar disorder do not seem to be more at risk for it than any other patients. The most common tolerability issue with pramipexole is nausea, which improves by titrating it very slowly. I usually start at 0.25mg at night and raise by 0.25mg every week toward 0.75mg. Then I’ll check on their response, and raise if needed, usually settling between 1-2 mg at night.

KELLIE NEWSOME: The modafinils – armodafinil (Nuvigil) and modafinil (Provigil) – are another off-label option. These wakeful-ness promoting agents have both positive and negative trials in bipolar depression, but when they were all meta-analyzed together it averaged out to a small benefit. Patients seem to appreciate that benefit more than the small effect on depression rating scale would suggest, however, because these meds are good at targeting some of the symptoms that bother them the most: Fatigue, cognition, and overall functioning. I usually start with armodafinil, because it has smoother blood levels that last a little longer throughout the day, and titrate the dose between 50-250mg depending on its benefits and side effects – which are usually anxiety, headache, and insomnia.

KELLIE NEWSOME: Another novel approach is Pioglitazone. This anti-diabetic medication has glutamatergic effects that resemble lamotrigine’s mechanism of action. On the one hand, we were encouraged by a positive randomized controlled trial in bipolar depression that included a lot of treatment-resistant cases. On the other hand, it failed in a follow up study a few years later, so the jury is still out on this one and we haven’t used it very much. We’ll also add the neuroprotective antibiotic minocycline to the wait-and-see box, as it has also a mix of positive and negative studies in bipolar depression. We’ve used this one a bit more in practice though because it can be used to treat acne on lithium. The dose is 100-200 mg/day, titrating over 2-3 weeks. It’s well tolerated and does help acne, but we haven’t seen much in the way of antidepressant effects with it.

CHRIS AIKEN: An often overlooked option for bipolar depression is one that is sort of on-label, at least for mania: Valproate (Depakote). This med is so well known for its antimanic effects that some people loosely think of it as a “downer,” but don’t let stereotypes like that get in the way of depression. Treating bipolar is not about uppers and downers. Sedation is a side effect of antimanic agents and not part of the therapeutic effects. Valproate treated acute bipolar depression with a small effect size in 4 small randomized controlled trials. It’s definitely not a first-line therapy, but it may be right for someone with rapid cycling, mixed states, migraines, an alcohol withdrawal, or anxiety – as valproate has evidence to help all those problems. Little known fact: Valproate has gabaergic effects and even treated anxiety in non-bipolar populations.

When using valproate, I’ll often add l-methylfolate 15mg in. Why? This supplement is effective in regular depression, and it has a small randomized controlled trial in bipolar depression as well. Plus there’s evidence that valproate works better when it’s used with a folate vitamin – in one controlled trial it had greater antimanic effects when combined with folate.

KELLIE NEWSOME: If your patient prefers natural therapies, l-methylfolate is certainly an option, but you might try some others that have more evidence in bipolar depression: Omega 3 (Fish oil), Inositol, and N-acetylcysteine (NAC). This last one, NAC, got some bad press a few years ago when it failed to work in a few studies, but we’ll sometimes use it anyway because there is a pattern in those studies: NAC treated depression when the study lasted 6 months or more, but did not work in the short-term. That’s a long time to wait, but many patients with bipolar disorder will linger in a depressive state year after year.

CHRIS AIKEN: Let’s not forget ramelteon (Rozerem). This melatonergic sleep medication reduced the frequency of depressive days in a randomized controlled trial of bipolar disorder with insomnia. I would only use it if the patient had insomnia, which most people with bipolar disorder do, and I wouldn’t count on it to treat a depressive episode. Like NAC, it’s more of a long-term preventative approach.

KELLIE NEWSOME: Finally there’s ketamine or esketamine – the ketamine version has evidence in treatment resistant bipolar depression. Ketamine worked rapidly – within 1-3 days – but the effects fizzled out within 2 weeks so this is more of an acute rescue therapy to help keep people out of the hospital than a practical, long-term solution. 

And then there are the devices. Light therapy, transcranial magnetic stimulation, and ECT all have good evidence in bipolar depression, and ECT is one we’ll often turn to if the symptoms are severe – especially if suicidality is involved – or the depression is treatment resistant. But don’t think that light therapy is lightweight. It had a large effect size in randomized controlled trials of bipolar depression, and in the trials of unipolar depression as well. Light therapy works best in winter depression but it can be effective in the summertime too. For a how-to guide to light therapy, check out our Feb 2019 issue online or scroll back to our July 2021 podcast.

CHRIS AIKEN: Treatment resistance is defined in bipolar depression as failure to recover fully after two therapies – each lasting at least 8 weeks.  That’s pretty similar to the definition of treatment resistance in regular depression, but when we get to this stage in bipolar disorder the options are fewer. 

KELLIE NEWSOME: As far as we know, only a small handful of the therapies we’ve mentioned today have evidence in treatment-resistant bipolar depression. They are

  • Pramipexole
  • Ketamine
  • ECT
  • Possibly Pioglitazone and the modafinils – these included a lot of treatment-resistant cases in their trials

And two others that we’ll save for next week: Celecoxib and thyroid augmentation.

And now for the word of the day….Inuit

CHRIS AIKEN: The Inuit are a group of indigenous peoples who live in the Northern regions of Greenland, Canada, and Alaska. Most indigenous people arrived in America about 15,000 years ago, but the Inuit were part of a later migration, crossing the Bering sea between Russia and Alaska around 5000-10000 years ago. From there they migrated Eastward, eventually reaching Greenland in the 1200s, around the same time that the Vikings were attempting to colonize the icy continent.

In our March podcast we used the terms “Inuit” and “Eskimo” interchangeably when reporting on the medical origins of omega-3 fatty acids, and we would like to make a correction to that. “Eskimo” is a term imposed on the Inuits by the Europeans, while Inuit is the preferred term, used by the people to describe themselves. Words matter. And when they are misused it contributes to the health disparities experienced by marginalized communities.

Get your CME credits for this episode by following the link in the show notes. The Carlat Report is one of the few CME publications that depends entirely on subscribers. Thank you for helping us stay free of commercial support


Leave A Comment