Can A Sleep Med Make Antidepressants Work Better?

When you sleep better you feel better, so it shouldn’t come as a surprise that a sleep medicine like eszopiclone (Lunesta) helps depression and anxiety. But can other sleep meds do the same thing, or is it just eszopiclone? In this podcast we explore whether eszopiclone (and its European cousin zopiclone) has direct effects on depression and anxiety. 

Some men, no doubt, will, before sleep, consider
One thought: I am alone. But some,
In the mercy of God, or booze, do not
Long stare at the dark ceiling.

KELLIE NEWSOME: That’s the last stanza of a poem by Robert Penn Warren, and it tells us something about what it’s like to be depressed and unable to sleep. When the lights are down, the ceiling is dark, there is nothing to distract from the ruminative thoughts that make depression so painful. Last month we covered the long and complicated relationship of benzodiazepines and depression that culminated in over a dozen controlled trials where the benzos – in particular alprazolam (Xanax) – did a pretty good job of relieving depression, even when compared head-to-head against a tricyclic antidepressant.

Today we’re going to talk about the successor to the benzos – the z-hypnotics – and whether they can augment antidepressants in depression and anxiety disorders. They are called the z-hypnotics because all of them start with the letter z:

  • Zopiclone (Imovane), which is not available in the US but has been used in Europe since 1986
  • Zolpidem (Ambien), released in 1992
  • Zaleplon (Sonata), released in 1999
  • And eszopiclone (Lunesta), released in 2004 in the US

It’s that last one we’ll be focusing on today, because it has the most studies in depression and anxiety.

Eszopiclone vs. Zopiclone

CHRIS AIKEN: Eszopiclone is not widely available outside the US, but for our international listeners it’s pretty similar to the original z-hypnotic zopiclone. Zopiclone is composed of two mirror image stereo-isomers – the L and S zopliclones – in a 50/50 mixture, so the typical dose of zopiclone – 7.5 mg – contains 3.25 mg of eszopiclone, which is just about the same as the max dose of 3mg. We know of only one clinical study that compared these two -clone hypnotics. It was a randomized trial that compared those two dosages of the drugs in 199 patients – and it found they were virtually identical in their clinical and adverse effects, which included polysomnography data. However, there are a few non-clinical studies in healthy volunteers, and they found that eszopiclone works faster and does not linger as long as zopiclone, which would suggest the s- isomer is a little more favorable.

KELLIE NEWSOME: You have to wonder – why would anyone fund a head-to-head study of two similar drugs that are both generic and not even available in the same country? This study was funded by Eurofarma, who recently launched eszopiclone in Brazil. Brazil is the only country we know of where both eszopiclone and zopiclone are available – and Eurofarma wanted to reassure Brazilian physicians that the new version was at least non-inferior.

Not All z-Hypnotics are the Same

CHRIS AIKEN: OK so now that we’ve clarified that these two drugs are similar, let’s look at how the zopiclones might differ from the other z-hypnotics. Although they all start with the letter z, or es-z, these medications have different chemical structures – enough that each is in its own class.

  • Zolpidem (Ambien) is an imidazopyridine
  • Zaleplon (Sonata) is a pyrazolopyrimidines
  • And the zopiclones are cyclopyrrolones

All these drugs act on GABA-A receptors, as do the benzodiazepiens, but the benzos are used for sleep and anxiety while the z-hypnotics are only used for sleep. You’ll often hear that the reason for this difference lies in a simple pharmacodynamic fact: The z-hypnotics only affect the part of the GABA-A receptor that’s involved in sleep – the alph-1 subunit – while the benzos affect alpha-2,3, or 5 – and these are the ones that reduce anxiety. But that’s not quite accurate. One z-hypnotic is not selective for alpha-1, and spreads its activity out among the alpha 2, 3, and 5 subunits. And that one – you guessed it – is zopiclone and its American cousin, eszopiclone.

KELLIE NEWSOME: Which means we might expect the zopiclones to have more anxiolytic effects than the other z hypnotics, at least in theory, but let’s at what happens when we test it out in people – and animals. When zopiclone was first discovered, they tested it in mice to see if it affected them like the benzodiazepines do, and indeed it did. Mice on zopiclone were less aggressive and anxious, much as they are on benzodiazepines. Most animal studies concluded that zopiclone has a very similar behavioral profile to the benzos.

Eszopiclone (Lunesta) in Depression & Anxiety

CHRIS AIKEN: The human studies came next. In 1989 and 1990 two small controlled trials came out comparing zopiclone to benzodiazepines that are often used for sleep: Triazolam and nitrazepam. In both studies, the z-hypnotic and benzos had similar effects on sleep, but the patients who took the z hypnotic consistently had less anxiety – a surprising finding considering they were compared to a benzodiazepine. The same thing happened when Japanese researchers compared zopiclone to the benzodiazepine midazolam for anxiety before surgery. Only zopiclone reduced the preoperative anxiety; the benzodiazepine did not.

KELLIE NEWSOME: So far we have a few small studies with some early indicators that zopiclone may reduce anxiety much like the benzodiazepines do. But the bigger studies came later with eszopiclone, and they were sponsored by its manufacturer, Sepracor.

CHRIS AIKEN: In 2006 Maurizio Fava and colleagues wanted to see if patients with depression got better when their sleep was treated. Intuitively it makes sense. Insomnia is a risk factor for depression, and depression is associated with all sorts of sleep complaints. But the idea had never been shown to work before, so this study was a first of its kind. They recruited over 500 patients with depression to take part, and everyone was started on the antidepressant fluoxetine. Half of them were also started on eszopiclone at the same time, and the other half were given a placebo. At the end of the 8 week study, those who got the zopiclone were less depressed than the patients on placebo, and the difference held up even when they removed the sleep items form the depression rating scale. The sleep medicine also brought about a faster response.

KELLIE NEWSOME: So was it something about zopiclone, or was it just that anything that improves sleep will improve depression. After all, we’ve since learned that a behavior therapy for insomnia – CBT-I – treats depression augments antidepressants. And keep in mind, the patients in that study had insomnia AND depression.

CHRIS AIKEN: The next study came out in 2010, and it was led by Vaughn McCall who also contributed to the Fava study. We interviewed Vaughn on the podcast last year, where he talked with us about the psychology of insomnia. These patients are keyed up, on edge, anxiously waiting for sleep to come, long staring at the dark ceiling as Robert Penn Warren put it. Lack of sleep causes anxiety, and anxiety causes lack of sleep. Vaughn understands this vicious cycle and wanted to see if eszopiclone would help not just depression but quality of life in depressed patients with insomnia. So he randomized 60 of them to either placebo or the hypnotic, again augmenting a new trial of fluoxetine. Quality of life no doubt improved, as did depression.

KELLIE NEWSOME: So far eszopiclone is 2 for 2 in depression, but there were two other industry-sponsored studies that did not get published. We were able to dig up the results of one of them, and it was negative, despite having a similar design to the others – eszopiclone helped sleep but not depression. The final study was a smaller one in elderly patients with depression and insomnia, and we reached out to the lead investigator but did not hear back.

CHRIS AIKEN: But most of the therapies we use have negative studies, and studies fail for all kinds of reasons. The only other z-hypnotic with similar studies in depression is zolpidem – Amien. It has about the same number of studies, including one by Dr. Fava, but zolpidem did not turn out as well – generally it helped sleep, but not depression. So is there something unique about zopiclone that gives it this antidepressant boast? That’s what Andy Krystal came to believe. Dr. Krystal was one of the investigators in many of these studies, and here’s what he wrote about them:

There is some evidence that the insomnia therapy eszopiclone may have direct antidepressant and anxiolytic effects. Studies were carried out evaluating the effects of eszopiclone therapy vs. placebo as co-therapy with open-label SSRI treatment in patients with insomnia co-morbid with major depression and generalized anxiety disorder. Eszopiclone not only significantly improved sleep compared with placebo but it also had a significant advantage over placebo on the associated depression and anxiety outcomes both with and without the sleep items included in the scales used. The initial interpretation of these findings was that eszopiclone improved sleep compared with placebo and that the sleep improvement mediated improvement in depression and anxiety. However, there is evidence that this is not the case. The identical studies were repeated with the insomnia agent zolpidem modified-release and comparable improvement in sleep vs. placebo was observed but without an accompanying effect on depression or generalized anxiety symptoms. The most parsimonious explanation for this set of findings is that eszopiclone has direct anxiolytic and antidepressant effects not mediated through effects on sleep and zolpidem modified-release does not.” – Andrew Krystal, MD, 2020

Where Zolpidem (Ambien) Failed

CHRIS AIKEN: As he describes, zopiclone was also tested in a large trial for generalized anxiety disorder, and it worked, while its neighbor zolpidem (Ambien) did not in a similar study. This was a very large study – 600 patients – and it compared zopiclone to placebo as augmentation for escitalopram in generalized anxiety disorder, and Dr. Krystal was one of the authors.  There are also small studies of eszopiclone in PTSD – one of which we reported on the Carlat Report last month – and there the results are not as strong.

KELLIE NEWSOME: And in 2014 Dr. Krystal published a controlled trial of eszopiclone for lower back pain, along with Harry Goforth and Xavier Preud’homme. And it worked – the patients on eszopiclone had less pain and less depression. The more their sleep improved, the more than pain went down, suggesting that it was the benefit was related to sleep rather than a direct effect of eszopiclone.

But Can It Cause Depression?

CHRIS AIKEN: Unfortunately, we don’t have a firm conclusion to this story. Eszopiclone is now generic and there doesn’t seem to be any more research funding to figure out if it has direct antidepressant or anxiolytic effects. But we have enough data here to suggest it might, and it’s worth keeping an eye out for this in your patients. But there’s also data pointing the other way. Daniel Kripke analyzed the major trials of the z-hypnotics in primary insomnia, and he found that the patients were more likely to develop depression on the hypnotic than the placebo. This was a secondary analysis, and the difference was small: 1 % vs 2 %, but it’s a warning worth considering. We saw the same paradox with the benzodiazepines. Those medications can cause depression, particularly with long-term use, but they also treated depression in the short term trials.

Sound familiar? Sounds a bit like eszopiclone. If it acted just like a benzo in mice, why wouldn’t it do the same in humans?

Desmethylzopiclone: A Secret Ingredient

KELLIE NEWSOME: So it may not be that anything that improves sleep will also improve depression. That could be part of the mechanism here, but there could be some anxiolysis going on that is helping eszopiclone cut through the depression. And it may not even be eszopiclone that’s doing it. Eszopiclone has an active metabolite – desmethylzopiclone – which has antianxiety effects. This metabolite seems to reduce anxiety without causing sedation, which is a good thing because it lingers around in the body with its 7-10 hour half-life. The company that made eszopiclone – Sepracor – put a patent on this metabolite 20 years ago, hoping to develop it for anxiety and possibly depression. Whatever research came of those efforts, however, seems to have vanished without a trace.

Barbiturates, Benzos, and Eszopiclone: The GABAergics in Depression

CHRIS AIKEN: What’s the take home here? We hope you’ve learned that the z-hypnotics are not interchangeable, and one of them – zopiclone – acts more like the benzos at both the cellular and the clinical level, though it’s not as likely to cause tolerance or addiction as the benzos are. But this is a complicated subject, and one that’s part of a larger history of attempts to treat depression with GABAergic medications, from barbiturates to benzos to zopiclone. Check out part one of this episode from April 5 to learn more on that.

KELLIE NEWSOME: And while we’re suggesting that eszopiclone may treat depression through anxiolytic, benzo-like effects, even that is not clear. Among the benzos, it’s arguably alprazolam that has the most specific antidepressant actions, and those effects may not be mediated by GABA. In part one of this podcast we mentioned that alprazolam has a unique structure which resembles a tricyclic antidepressant structure, and one of our listeners sent in a comment that adds to that story. Alprazolam also affects serotonin 5HT1A signaling, much as vortioxetine, buspirone, and some of the atypical antipsychotics do.

CHRIS AIKEN: We started with what looked like a coherent story – that the gaba-ergic medications have long been used with some success to treat depression – and that the major advance from barbiturates to benzos to the z-hypnotics was one of safety not efficacy. But we ran into a lot of unanswered questions along the way.

KELLIE NEWSOME: So what’s the take home point?

What To Do In Practice

CHRIS AIKEN: If your patient has depression and needs a benzo, consider alprazolam. And if they have depression or generalized anxiety – or chronic pain – and need a hypnotic, consider one of the zopiclones. And if you use them, try to keep it to the short term. All of the studies we’ve talked about are short term, and we’ve uncovered some evidence that long-term use of these gaba-ergics can cause depression.

KELLIE NEWSOME: Yes and the researchers who discovered eszopiclone’s augmentation effects also proved that you could successfully stop the hypnotic as recovery kicks in. In the depression study, they secretly replaced the hypnotic with a placebo after 8 weeks in a randomized withdrawal phase, and the patients did just as well on the placebo – in terms of sleep and depression – suggesting that short term use is all that’s needed. In the generalized anxiety study, however, they had a little more trouble coming off eszopiclone, which makes sense if you consider it has benzodiazepine effects.

CHRIS AIKEN: It’s a little harder to do this in practice where you can’t replace your patients sleep med with a placebo, and placebos do have a big effect on sleep. But you can set expectations at the start. Tell your patient that the sleep medicine is there to get them back on the road when their sleep is really off track, and that they’ll need to make some lifestyle changes to help stay on the road once their body is back in the habit of regular sleep. Then, taper off the hypnotic, lowering the dose and the quantity, over a month or two. See our companion podcast for patients – The Pocket Psychiatrist podcast – for a recent episode on CBT insomnia called “How to Sleep.

Present Moment Awareness

And now for the word of the day…. Present moment awareness

KELLIE NEWSOME: Present moment awareness is an element of mindfulness. It involves monitoring and attending to your current experience rather than anticipating or worrying about the future or dwelling on the past. The idea is captured in the title of Jon Kabat-Zinn’s 1994 bestseller, Wherever You Go, There You Are, or in a lot of cliché’s like “Be here now” and “Seize the day.” In next week’s podcast we’ll detail the treatments that work best after a patient fails two antidepressant trials. On that list you’ll see some unique medications like minocycline that few of us are using, but you’ll also see psychotherapy, particularly mindfulness based therapy and psychoanalysis. That’s right, psychoanalysis worked for treatment resistant depression – and we got the manual here for the particular form of psychoanalysis – the Tavistock method – and I’m going to read it to you right now. Here’s what it says the analyst provides to the patient:

The analyst is receptive and open, makes emotional contact, and attends to the here and now.

People with depression have a hard time being in the here and now. Their inner experience is often painful, and they are driven to avoid it. They are disconnected from the outer world as well. Psychologist James McCullough describes patients with chronic depression as though they are encased behind a wall of concrete. It’s not easy to sit with that kind of pain, or to reach behind those walls, but the next time you see someone with depression, try to bring a dose of Present moment awareness into the room, and join us next Monday for part 2 of our series on What Works Best in Treatment Resistant Depression.