Activity Transcript

Atul Malhotra, MD: Hello. We're going to be talking today about addressing insomnia in patients with obstructive sleep apnea. My name's Atul Malhotra. I'm the research chief for pulmonary critical care, sleep medicine, and physiology at the University of California, San Diego, and I see many patients with sleep disorders.

Whether you realize or not, these are very common problems in the primary care clinic, so we're going to be talking today specifically to a primary care audience about these issues. By way of background, obstructive sleep apnea is characterized by repetitive collapse of the upper airway. The back of the throat will collapse during sleep, and that leads to 2 sets of issues. One is neurocognitive sequelae because sleep gets quite fragmented. If I were to come shake somebody and wake them up every minute or 2, their sleep would be nonrestorative. In addition, there are cardiovascular sequelae because when you stop breathing, your oxygen levels fall, you release catecholamines, and both those things can contribute to cardiovascular risk.

The severity of sleep apnea is often defined by the apnea-hypopnea index (AHI), the number of cessations plus reductions in breathing per hour of sleep. It turns out, both sleep apnea and insomnia can coexist and those can often present to the primary care physician. Insomnia is generally difficulty falling asleep or difficulty staying asleep and that can sort of intermingle with sleep apnea, as we're going to talk about.

The first-line treatment for obstructive sleep apnea is CPAP, which is continuous positive airway pressure. It's a mask over the nose. It has kind of a bad reputation in primary care circles, but I could tell you, according to recent data, the vast majority of patients will tolerate it if given modern technology and adequate education and support. There are alternative treatments for sleep apnea as well, including oral appliances and upper airway surgery, and other things we probably won't get into today. There are various treatments for insomnia as well, which we'll touch on.

In terms of the global prevalence, we did an estimate of this a couple of years ago in the Lancet Respiratory Medicine. We estimated, based on an apnea-hypopnea index of more than 5 events per hour, up to a billion people worldwide with obstructive sleep apnea. It's a really very common problem, up to a billion people by that definition. If you use a stricter definition of obstructive sleep apnea, you set the bar higher at an apnea-hypopnea index more than 15 events per hour, where there's really no debate that treatment is required, it's still almost about a half a billion people worldwide. You can see the worldwide map there where sleep apnea is really quite common.

You might ask the question, how do we identify patients with comorbid insomnia and sleep apnea? It turns out it is quite common. It could be a third of patients with obstructive sleep apnea that have insomnia as well. There are various different questionnaires, which we'll get into. The Functional Outcome of Sleep Questionnaire is one, the Pittsburgh Sleep Quality Index is another. The Epworth Sleepiness Score is one you'll hear about to look at the degree of sleepiness and also the SF-36, which is a measure of quality of life. Suffice it to say, unless you ask these sorts of questions, you don't find out. Patients don't often volunteer these sorts of issues. So, unless you ask them as part of your review of systems, how you're sleeping, they often don't volunteer these things. So, one take-home message here is to add a question to your review of systems about sleep. I'm often fond of saying the 3 pillars of health are diet, exercise, and sleep. If you ignore 1, the other 2 will suffer. So, I think it's incumbent on all of us, including primary care physicians, to at least enquire.

So think about how to identify patients with comorbid insomnia and obstructive sleep apnea. There's some red flags out there that are worth knowing about as a primary care physician. With run of the mill sleep apnea, you might look for snoring, daytime sleepiness, if there's a bed partner thinking about witnessed apnea, that's typical of sleep apnea. With comorbid insomnia, sometimes patients say, "Doc, I can't sleep at night," or "I have trouble falling asleep," or "I have trouble staying asleep." Those are characteristic of insomnia. And interestingly, the two are sometimes interwoven. So, some people say, "I can't fall asleep at night," because their breathing is not allowing them to fall asleep, so they're starting to drift off and then they wake up suddenly because of an inability to breathe and they present with insomnia. That's not always obvious to the patient and so it's worthwhile doing a high index of suspicion where if somebody says, "Doc, I can't sleep at night," if they have risk factors for sleep apnea like obesity or other features, hypertension, whatnot, think about sleep apnea with comorbid insomnia, because both conditions may be there.

Other red flags for insomnia. The bed partner might say, "My partner is tossing and turning all night." And asking the question, "What do you think is waking them up?" Sometimes they'll volunteer, the bed partner will volunteer, well, the patient's struggling to breathe, they'll wake up gasping and then they're unable to go back to sleep. And they'll present with tossing and turning all night, but when you dig into the history, you can actually learn what's waking them up and that can be quite helpful in some cases. So again, a high index of suspicion that these two conditions can coexist is important to know. We're talking about very common conditions and so they'll coexist by chance alone, but also they're mechanistically linked, as we've discussed.

There's a woman named Lichuan Ye who has published on this area. They've defined various different clusters in obstructive sleep apnea. There's a cluster that has disturbed sleep or insomnia that's about a third of patients. There's a minimally symptomatic or asymptomatic group, about a quarter of patients. Then there's a group with severe excessive daytime sleepiness, which is about 43% of patients.

The reason it's interesting is that third group with daytime sleepiness appears to be the group primarily at risk for cardiovascular disease. Meaning, the other 2 groups are not. The reason that's interesting is we often do randomized trials and just pick patients willy-nilly and put them into randomized trials with the one-size-fits-all approach and often those trials have failed. Perhaps the reason is that there is differential susceptibility. Not everybody with sleep apnea is at risk of cardiovascular disease from sleep apnea. Some are at risk of cancer, some at risk of cardiovascular disease for other reasons. So, a one-size-fits-all approach really hasn't done us much good.

The point to make in the context of an insomnia talk is that disturbed sleep group, that first group there, is about a third of sleep apnea patients and that combination of sleep apnea and insomnia often goes undiagnosed or unrecognized.

There are different endotypes or mechanisms underlying sleep apnea. This is the main focus of my laboratory so I could go on about this forever, but I won't. Suffice it to say there are various different reasons that people get sleep apnea and those matter for various different reasons. One is it may guide therapeutic interventions. It turns out that oxygen received all night may be very effective treatments for a subset of sleep apnea patients defined by their control of breathing. It turns out there's certain risk factors for sleep apnea that may be related to underlying mechanisms. For example, patients with posttraumatic stress disorder often have a lower arousal threshold, that is a propensity to wake up, and that's perhaps why they have sleep apnea. Some patients with neuromuscular disease have dysfunction in the upper airway muscles and perhaps that's why they're at risk of sleep apnea. It turns out the sleep apnea endotype of mechanism might predict the phenotype, that is the clinical expression of disease. We know that sleep apnea in the elderly is frequently complicated by insomnia and those patients don't seem to be at as much risk of their sleep apnea as younger patients matched on severity. The reason for that's unclear, but we have a theory at least, that the underlying mechanism of the sleep apnea is different in older individuals compared to younger ones. Perhaps the reason that they have fewer clinical manifestations of disease is that it's from a different mechanism. Then sleep apnea endotypes may predict treatment response. As an example, if you give people CPAP, continuous positive airway pressure, in patients who are prone to wake up with a low arousal threshold, they may not do well. Their adherence to CPAP may not be great, and that may be a problem for those individuals. So, it can be helpful to recognize that ahead of time and plan accordingly. The take-home message from this slide is that obstructive sleep apnea occurs for a variety of different reasons. Even if we don't know clinically which mechanism is underlying, it is helpful to think of this as a heterogeneous disease and using a one-size-fits-all approach is perhaps not the best idea.

People with insomnia seem to have a low arousal threshold, that is, they tend to wake up quite easily. It turns out we could predict that arousal threshold using clinically available data from sleep studies, and this clinical prediction of the arousal threshold may help guide us in terms of treatment. We're not there yet, but perhaps there's a subset of individuals with sleep apnea who are raising the arousal threshold using a hypnotic drug, for example, might be a viable approach.

Again, this slide is somewhat complex, but you can see towards the bottom of the slide, we're talking about treatment of sleep apnea and the CPAP, continuous positive airway pressure is the treatment. This is summarized in a review that we had in the Lancet a couple years ago.

There are various different mechanisms shown in green that correspond with various different interventions shown in pink. You can see if you have a low arousal threshold, perhaps a set of hypnotic drug is the way to go. If you have a problem with control of breathing, then perhaps oxygen or acetazolamide, as I mentioned, maybe a way of stabilizing control of breathing. Just another way of depicting the heterogeneity underlying sleep apnea and that subset of sleep apnea with insomnia maybe the group that's most amenable to using a set of hypnotic agents. Now I know what you're thinking. You're saying, "Well, giving a sedative to people with sleep apnea is a bad idea." In general, that's true, but there may be a subset where hand-picking those patients may be a good idea to intervene pharmacologically.

So how does sleep apnea and insomnia interact? Insomnia may fragment sleep and that can lead to fluctuations and carbon dioxide and that can destabilize breathing. Sleep apnea may lead to recurrent insomnia so if somebody stops breathing they wake up. In some cases, they stop breathing, they wake up and have a sustained awaking that can be a problem. Some people with CPAP, if they have a propensity to wake up, will sleep quite poorly and they can get CPAP-induced insomnia. That can be an issue for the subset of patients, as I mentioned, with a low arousal threshold. Then hypnotic therapy in general should be used cautiously in sleep apnea. Giving an agent such as a benzodiazepine may suppress the activity of upper airway muscles, perhaps even worse in sleep apnea. It's kind of a theoretical concern, but it's a real one that comes up. But in some cases, actually, if you're giving somebody a CPAP and they're developing insomnia or risk of insomnia, the hypnotics may actually be a good idea and I'll show you some data on that.

As one example, Chris Lettieri had this paper in the Annals of Internal Medicine using a drug called eszopiclone as compared to placebo in trying to improve CPAP adherence. What they did is randomized patients to get either eszopiclone or placebo, give that to them for a period of time, then followed them afterwards to see what their CPAP adherence looked like. As you can see, CPAP adherence is actually improved with the drug as compared to placebo, suggesting a good initial experience with CPAP may be facilitated by the agent that may lead to longer-lasting improvements with that agent. So, the take-home message here is there may be a subset of patients where improving the experience with CPAP using hypnotic maybe a good idea.

We did a follow-up study on that. Chris Lettieri was kind enough to give us some data that we analyzed from that prior study. We published this in the American Journal of Respiratory and Critical Care Medicine and showed that the arousal threshold may be a drug target and this propensity to wake up or arousal threshold may be a drug target. I won't get into the nuances of that today, but suffice it to say, there are subsets of sleep apnea patients who may be amenable to sedative hypnotic therapy, and that might help improve CPAP adherence in select patients.

It does raise the question, how do we treat these patients? Sleep apnea treatment has several components. There's primary therapy for the sleep apnea, nasal CPAP, continuous positive airway pressure. Oral appliances or mandibular advancement devices can be quite helpful. Or upper airway surgery is helpful in a subset of patients. Particularly hypoglossal nerve stimulation is now being looked at in randomized trials, although existing data have led to FDA approval based on observational studies. You can manage risk factors. Weight loss is a very important component of sleep apnea management. About 70% of sleep apnea patients are obese. They may well benefit from diet and exercise. That said, about 30% of sleep apnea patients are not obese and so, obviously, obesity is not the only risk factor. Addressing comorbidities can be quite helpful as well.

So insomnia, there are behavioral approaches to insomnia, as I mentioned, and there are pharmacological approaches to insomnia as well, which can be quite helpful in select patients. What do we know about treating insomnia and obstructive sleep apnea? You can treat the underlying cause. If they have pain, you can treat their pain. If they have shortness of breath, you treat their dyspnea. If they have restless leg syndrome and aching pain in their legs with sensory and motor complaints in the legs, you can treat that with a dopaminergic agent or with gabapentin or other agents like that. Sleep hygiene can be quite helpful for a subset of patients. Avoiding alcohol, tobacco, caffeine can be quite helpful. Avoiding excessive napping. So ,if I take a nap on a Sunday afternoon, sometimes I can't sleep Sunday night. Not because I have insomnia, it's just because I'm not tired. So sometimes just cleaning up the schedule and having people just try and sleep at defined times and these erratic schedules or napping during the daytime can affect sleep at night. Those sorts of interventions can be quite helpful as far as sleep hygiene is concerned in a subset of patients where there are issues to address. There are behavioral approaches to treating insomnia as well. Cognitive behavioral therapy for insomnia is generally done by psychologists, it involves various different interventions. One is stimulus control where you prevent people from stimuli while in bed. We generally say the bed is for intimacy and sleep and nothing else. Many patients will play video games or check email, or do whatnot in bed, watch television. Trying to avoid those things, that's stimulus control. There's also a component called sleep restriction where we limit the time in bed and those are components of what we call cognitive behavioral therapy for insomnia.

In this particular study, a randomized trial of trying to intervene on patients getting CPAP therapy with insomnia and sleep apnea, in that context, you can see some improvement in CPAP adherence as published by Jennifer Martin and others in this particular study.

What do we know about hypnotic therapy and obstructive sleep apnea therapy with insomnia? As I mentioned, benzodiazepines may have deleterious effects, at least in theory. We see this when we do bronchoscopies. If you give a benzodiazepine with an opioid narcotic, often you'll see the airway close. When you look at actual clinical data, does a little bit of benzodiazepine worsen sleep apnea? Theoretically, it's not a big problem. These Z drugs may also have a role in improving CPAP adherence. By Z drugs, I'm talking about zolpidem or eszopiclone, as I showed you, that may help with CPAP adherence. There are some newer agents that work through orexin-type mechanisms that are being used for insomnia. Orexin is a chemical in the brain that is important in sleep regulation, the orexin agonist being used in narcolepsy, orexin antagonist being used in insomnia.

This particular agent is called daridorexant. There's a study in a randomized trial in sleep apnea. As you can see, they saw some minor increase in total sleep time and, as a result of that, there was slightly increased respiratory events. But in general, the number of events and the severity of events were basically unchanged. For me, this was fairly reassuring that this agent could be given in sleep apnea with respiratory safety.

There's a different agent called suvorexant, which was also studied in a randomized trial, and this is 26 patients in a placebo-controlled study. You can see a very tiny increase in the apnea-hypopnea index, nothing to worry about. There was no effect on saturation and, in general, no clinically important respiratory effects are seen with this agent in sleep apnea. So, for me, again, this is reassuring these agents could be given safely in people with sleep apnea and insomnia.

This is a different agent called lemborexant, which is also studied in a randomized trial. This is a short duration study in sleep apnea, placebo-controlled crossover study, and saw no difference in the apnea-hypopnea index or in saturations and no real treatment-emergent adverse events. So, for me, again, this is reassuring that this agent could be given safely in that context.

And then pharmacological treatment of sleep apnea is also being studied. That is, there are primary drugs being looked at to treat sleep apnea per se and that might help with the sleep fragmentation and daytime symptoms as well. We're not there yet, it's sort of a holy grail in sleep apnea to try and come up with a drug treatment, we're not there yet, but it's an active area of research.

By way of conclusion, sleep apnea is a common disorder with major neurocognitive and cardiovascular sequelae. It's both underdiagnosed and untreated. CPAP, continuous positive airway pressure, is a really good treatment for sleep apnea. It's efficacious. Its adherence can be quite variable, but I think with education and support and modern technology, things can go quite well. Addressing sleep apnea with comorbid insomnia is an important thing to do. It's important for primary care doctors to think about that overlap. Just giving them CPAP may not be the solution, as I said, that may be a predictive poor adherence. Addressing underlying causes, improving sleep hygiene, thinking about behavioral therapy or pharmacotherapy is also perhaps in the wheelhouse. I don't expect a primary care physician to be able to identify all the underlying causes and endotypes and whatnot but being aware of this can be quite helpful. Just recognizing that not all sleep apnea is the same, recognizing there's phenotypic variability can be quite helpful.

Thank you all for your attention. Please continue on to answer any questions that follow and complete the evaluation.

This transcript has been edited for style and clarity.