A recent paper lists 5 therapies for treatment-resistant depression, but the research keeps rolling in, and we bring you 2 more to add to the list.
Published On: 04/18/2022
Duration: 20 minutes, 15 seconds
Related Article: “Thyroid Augmentation in Bipolar Disorder,” The Carlat Psychiatry Report, April 2022
Chris Aiken, MD, and Kellie Newsome, PMHNP, have disclosed no relevant financial or other interests in any commercial companies pertaining to this educational activity.
KELLIE NEWSOME: Not long ago – I’m talking turn of the century Y2K – we had only one medication with clear benefits in bipolar depression: Lithium. Now the list is much longer, and new research is bringing hope to those who haven’t responded to any of them. Today, we bring you 7 therapies for treatment resistant bipolar depression.
Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003.
CHRIS AIKEN: I’m Chris Aiken, the editor in chief of the Carlat Report.
KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue.
KELLIE NEWSOME: In 2020, a group of psychiatrists led by Michele Fornaro (Mikel For-naro) went searching for answers to a problem we face all too often in practice: Treatment resistant bipolar disorder. They pulled up all the studies on treatment-resistant bipolar, and then went one step further. They sifted through 800 trials to find any that included a lot of treatment-resistant cases, their cut off was at least 75% with treatment resistance. It was a heroic effort, but the findings were meager: Just 17 trials, 14 in treatment resistant bipolar depression, and 3 in treatment resistant mania.
Last week we walked you through the various therapies that work in bipolar depression, and we ended by highlighting the ones that made Fornaro’s list:
- The modafinils: Provigil and Nuvigil
- And the anti-diabetic pioglitazone
We’re guessing a lot of our listeners have not tried these options – maybe you’ve given modafinil, or referred someone to ECT, but pramipexole? Pioglitazone? These don’t even show up in Stahl’s encyclopedic Prescriber’s Guide.
CHRIS AIKEN: Today, we’re going to narrow that list down to the practical few, and introduce you to two more contenders. But first, a preview of the CME quiz for this episode.
KELLIE NEWSOME: What is the main limitation with ECT in bipolar depression?
A. The controlled trials were all conducted in unipolar depression
B. ECT may work in bipolar depression, but it less effective than it is in unipolar depression
C. It treats bipolar depression, but we don’t have a clear way to prevent episodes after ECT is completed
D. ECT is effective in bipolar depression but not in treatment resistant cases
CHRIS AIKEN: Back to the list. We’ll take ketamine off for everyday use because it’s really a special-use medication. It has a very large effect – patients were 10 times more likely to respond to IV ketamine than IV placebo after only 1 day. But the effects were short-lived, so it’s best reserved for times when you need rapid relief to keep someone out of the hospital.
Nor is ECT for everyday use, but it’s our go-to for severe treatment resistant cases. I think of one patient who suffered 10 years in a mixed-state depression until he finally recovered on ECT and is now able to work again. We should have gone to ECT a lot sooner, but he would seem like he was responding a little to each medication trial, only to fall back from that false hope to the same place.
Not long ago we didn’t even know if ECT treated bipolar depression, as the well-controlled studies were all done in unipolar depression. There was also a concern that ECT, or any treatment for depression, might trigger mania. That started to shift in 2010 when a large study compared outcomes within unipolar and bipolar depression – they found the treatment worked equally well in both groups. The more convincing data came in 2015, when a randomized controlled trial compared ECT with pharmacotherapy for bipolar depression. The pharmacotherapy was the best available, using an evidence-based algorithm, but ECT worked much better, with a 74% response rate vs. 35% with medication.
The ketamines and ECT are both very effective treatments – with extra-large effect sizes – and they work fairly quickly. Last week, a new analysis came out suggesting that ketamine works about as well as ECT in treatment resistant depression – now, most of these were not bipolar depression, so I’m extrapolating a bit here. But the analysis had two important takeaways
- Ketamine is about as effective as ECT
- Ketamine takes a little longer to work in the treatment resistant population, building gradually over 2 weeks, which is about the same rate as ECT. That’s with 3 ketamine treatments a week.
So ketamine might work quickly in some, but take longer in more treatment resistant cases, but either way – getting better within 2 weeks is pretty impressive. The problem with both these treatments is that both have short-term benefits and we don’t have a sure way to keep people out of depression. Many meds have been tried without success to sustain ketamine’s benefits. With ECT, we know that monthly maintenance ECT and lithium plus an antidepressant works – as does psychotherapy actually – but that’s for unipolar. How do you keep patients with bipolar depression well after ECT? That is less clear. Lithium and psychotherapy are good ideas, but an antidepressant is probably not.
KELLIE NEWSOME: Then there’s the antidiabetic med pioglitazone. This glutamatergic medication with lamotrigine-like properties worked in a small study, but failed in another small study, so what do you make of that? Well, it did work in a few studies of unipolar depression, so maybe there’s something there, we could speculate that bipolar depression is just harder to treat than unipolar. But it has a black-box warning in the PDR that makes us a little nervous – pioglitazone may increase the risk of bladder cancer, particularly with chronic use over more than a year.
CHRIS AIKEN: Pioglitazone is a diabetes medication, although its antidepressant effects are unrelated to its metabolic effects, but I might consult with the endocrine team if my patient has bipolar depression and diabetes to see if it could cover both bases.
Either way, improving diabetes management is one way out of treatment resistant bipolar depression because insulin resistance and end-organ damage is not just something that happens to the eyes and the kidneys, it goes on in the brain as well. A new controlled trial showed that improving insulin sensitivity with metformin relieved treatment resistant bipolar depression in patients with diabetes.
KELLIE NEWSOME: You’ve narrowed the list down to 2 options: Pramipexole and the modafinils.
CHRIS AIKEN: The modafinils have a role, and I’ve used them in several thousand patients with mood disorders, but I don’t see them as lifting people out of treatment resistant bipolar depression. They failed to work in some of the studies, and in practice they seem to help isolated symptoms of energy and concentration rather than the whole episode. But if they can help your patient stay out of bed during the day that may make the difference between holding a job and losing one.
That leaves us with one option: pramipexole. Fornaro found two controlled trials supporting this dopaminergic medication in treatment resistant bipolar depression, with a large effect size. I agree. I’ve used pramipexole in over a thousand cases of bipolar depression, most of them treatment-resistant, and have seen full recoveries on it. Two of our expert interviewees have also recommended it – Andrew Miller and Joe Goldberg – and I wrote a practical How-To guide to using pramipexole in our January 2020 issue.
KELLIE NEWSOME: But those are just the treatments that Fornaro and his group tallied. We found two more candidates that we’ll discuss today: the NSAID Celecoxib and augmentation with high-dose thyroid. The Celecoxib paper came out after Fornaro’s review, and the thyroid papers seemed to have slipped through the meta-analytic cracks.
CHRIS AIKEN: First, celecoxib. This is a non-steroidal anti-inflammatory or NSAID that’s used for arthritis and various pain conditions. Within that category of NSAIDs, its subcategory is as a Cox-2 inhibitor, and it’s the only Cox-2 inhibitor that’s still on the market. Others, like Vioxx, were withdrawn in the early 2000’s because they increased the risk of heart attacks and strokes. Celecoxib also increases those risks, but no more than other NSAIDs like ibuprofen or naproxen – at least, that is what the FDA concluded when they analyzed all the data in 2018.
There’s a few other risks to know about before you even consider celecoxib. It can cause ulcers and GI bleeds, so don’t give it with aspirin and avoid it in patients who already have these problems. Some people give it with a proton-pump inhibitor like pantoprazole to reduce this risk.
Like other NSAIDs, it may increase the risk of renal failure, and the risk with celecoxib doesn’t seem to be any greater than that for other NSAIDs. Avoid it in patients taking ACE inhibitors, as those raise the renal risks even further. And use caution with lithium, as NSAIDs can raise lithium levels.
So why would you want to use this medication after all those warnings? Well, it is otherwise well-tolerated – better than just about any psych med we know of. And in a new trial celecoxib worked in treatment resistant bipolar depression with a big effect – 1 out of 2 patients had full remission of this difficult-to-treat depression on it.
The idea is backed up by 7 randomized controlled trials in unipolar depression which – when analyzed together – also had a large effect size – 0.8 – which is on the order of the benefits we see with stimulants in ADHD. And unlike a lot of antidepressant options celecoxib does not seem to cause mania and may even treat it. In two small trials of inpatient mania, adding celecoxib had a small but positive effect after 8 weeks.
Most of these studies used the same dose – 400mg a day – sometimes given all at once and sometimes divided in two. The effects came on in 6-8 weeks, so it’s a little slower to work than an antidepressant. Since most of the downsides are long-term medical risks, I’d recommend that you try to taper off celecoxib after 6 months if your patient responds to it.
KELLIE NEWSOME: So far things are looking pretty good for celecoxib. It even has a few pathophysiologic theories to back it up. Depression and mania are inflammatory states – not for all patients but for a large number – around 30-50% - we see high levels of inflammatory markers circulating in their bloodstream, as though their bodies were at war with some unknown infection. These inflammatory markers go by names like interleukins, tumor necrosis factors, prostaglandin, and C-reactive protein. Celecoxib lowers inflammation by blocking the production of prostaglandin – in fact that’s how it treats pain too.
All this raised expectations that celecoxib might work better in patients with high levels of inflammation, but so far that hope has not paid off. We’re aware of two studies that tested that out and neither found a relationship between inflammation and response. Another theory has to do with arachidonic acid, which is one of the main fatty acids in the brain – along with the omega-3 fatty acids. Basically, a deficiency of arachidonic acid could cause depression, and celecoxib raises arachidonic acid levels by blocking their conversion into those inflammatory prostaglandins.
But whatever the mechanism, we’re still on the edge of our seats about a well-tolerated medication that can bring 50% of patients with treatment-resistant bipolar depression into remission. At least, we were on the edge of our seats, until this follow up study came.
CHRIS AIKEN: The study was one of the biggest disappointments in psychiatry of 2020, generating several letters to the editor from bewildered investigators who had pinned their hopes on celecoxib. This study was conducted over 3 months, across 4 clinics in Pakistan. It was the largest to look at celecoxib in bipolar – enrolling 200 patients with bipolar I or II depression, 4 times as many as the earlier trial. Actually, the study enrolled even more than that – 266 – because it also tested another anti-inflammatory that has had promising results in depression – the antibiotic minocycline, but the tally of patients on celecoxib or placebo was only 200.
They randomized the patients to 4 groups: Minocycline with placebo, Minocycline with celecoxib, Celecoxib with placebo, or just placebo. Everything went as planned, it was a well designed study, except for the outcome – there was no difference between any of the groups. They even subdivided the patients by their levels of inflammation but still found nothing.
The study was picked apart by several authors, but the only viable explanations for those who still believed in celecoxib came down to this. Maybe there was a high placebo response rate, washing out any treatment effect for the whole population. Or maybe it really was positive – but the study was too small to detect it – Celecoxib did bring about slightly greater remission rates, but they were just shy of statistical significance.
Or, maybe celecoxib works in unipolar depression but not bipolar depression. We just don’t know. We haven’t used it enough in our own practice to draw any meaningful impressions, but there is one more therapy for treatment resistant bipolar disorder that we haven’t gone into detail on: Thyroid augmentation, and next week we’ll interview a psychiatrist who can speak from experience. Tam Kelly has tried high-dose thyroid in over 2,000 patients with bipolar depression, and he’ll tell us how it works.
KELLIE NEWSOME: If you can’t wait until next week, check out our April issue online at the Carlat Report .com where we feature a full interview with Tam Kelly.
And now for the word of the day….T3 and T4
We better get this lingo down before next week’s podcast. T3 and T4 are natural thyroid hormones that circulate in the body. T3 is more active in the CNS. T3 is available on the pharmacy shelves as the synthetic hormone triiodothyronine, Cytomel; while T4 goes by the prescription name levothyroxine (Synthroid).