Activity Transcript

Segment 1: Normal Sleep Architecture and the Impact of Sleep Disorders

Geert Mayer, MD: Hello, I'm Geert Mayer, associate professor in the Department of Neurology at the Phillips University, in Marburg, Germany. Welcome to this activity, "How Important Is Sleep Architecture for Sleep Quality? We'll go straight to the first talk, "Normal Sleep Architecture and the Impact of Sleep Disorders."

What is important to know about sleep structure? Sleep consists of non-REM and REM sleep. Non-REM sleep 2, as you can see here, covers 50% of the night. Heart rate and breathing are slow. There are sleep spindles and it serves memory consolidation.

Non-REM 3 is prevalent in the first half of the night, and it consists of slow wave sleep, it has delta waves on the EEG, and it serves a restorative function.

REM sleep finally is prevalent in the second part of the night and it's increasing until the morning, so we wake up dreaming. It occurs about every 90 minutes and it is associated with dreaming and activated EEG. Breathing and heart rate are similar to the wake state. There is low muscle tone, sometimes missing muscle tone, and it serves learning and memory.

Throughout the lifetime, sleep duration changes and also the composition of sleep changes. As you can see from 5 to 85 years, we have an increase of wake after sleep onset. We have less REM sleep. We have less slow wave sleep, whereas light sleep in 2 and stage 1 remain the same.

What is important to record about sleep? We need to know about sleep onset; we need to know about REM sleep onset; wake after sleep onset in minutes; total sleep time; the amount of non-REM and REM within total sleep time; the arousal index, meaning the changes of sleep stages and wake stages throughout the night; periodic limb movement, which is important to record for restless legs or periodic leg movements; and sleep apnea and hypopnea index, which disturbs night sleep by hypopneas or apneas.

Sleep and memory are very important parts, as you have heard before, for learning and memory and cognition. On the graph you have on the left side, slow wave sleep. And slow wave sleep consists of spindles, ripples, and these transform memory and consolidate memory throughout the night. And from the neurotransmitter side, you see a decrease of acetylcholine and cortisol. Sleep-dependent memory formation in the immune and central nervous system share common mechanisms that in both domains appear to be linked to slow wave sleep.

On the right side of the graph, you see REM sleep. And REM sleep also has spindles and it has PGO waves, and cortisol and acetylcholine are increased. Subsequent REM sleep behind non-REM sleep may then be involved in strengthening the reactivated and reorganized representation on a molecular and synaptic level, so much for cognition.

How fast does memory consolidation occur? It occurs after 3 hours of sleep, but after a complete night of sleep, consolidation is much better. And sleep stages are thought to be responsible: deep sleep for declarative memory, REM sleep for procedural and emotional memory. But you can also consolidate your memory by daytime naps. You need to have at least 60 to 90 minutes of daytime sleep. If people have sleep deprivation, it causes worsening of encoding. And also to note, age means less deep sleep, less slow wave sleep, and slower learning.

What do we know about the key transmitter systems in the arousal system? Here, you see a graph of the brain and the arousal system comes from the periphery, goes through the medulla, the pons, and in the pons, we have the raphe nuclei and the locus coeruleus, which are very important for REM on and REM off. And all these show projections to many brain structures, to the frontal basal of the brain and to the cortex, and they're mediated also by orexins.

The sleep-promoting systems on the other hand, start in the ventro preoptic nucleus (VPO), as you can see on this graph. It's GABAergic and it dampens all the activity of the noradrenergic, histaminergic, and the other alerting systems so we can fall asleep, and it also has an impact on muscle tone.

To show which impact sleep disorders have on sleep architecture, you can see here as examples: sleep apnea, insomnia, restless leg syndrome, Parkinson disease, and Alzheimer dementia. And as you can see, except for sleep apnea, all these disorders show a reduction in sleep time, an increase in sleep fragmentation, and the sleep stages show increased transitions, and a decrease in light sleep, in REM sleep, and also in non-REM sleep 3. That means cognition is impaired.

Here we have an example of what happens if you suffer from insomnia. In the upper panel, you have a normal sleep architecture. In the lower panel, you have a patient with insomnia. And what you can immediately see is in the upper part, you see the number of arousals has increased. The wake is the second line and wake after sleep onset has increased. And most important is slow wave sleep in the very lower panel is fragmented. And then in the red, you see that REM sleep is fragmented as well.

Here, you see the changes that occur when people suffer from sleep apnea. These are different hypnograms. On the top left level, you see a hypnogram of apnea hypopnea index of 21. On the lower right panel, you see an apnea hypopnea index of 72. So what you can see from top down is that the number of awakenings is increased, sleep fragmentation is increased, and that there is a deficit of slow wave sleep, deep sleep.

To summarize mammalian behavior consists of wake, non-REM, and REM sleep. Sleep is recorded by polysomnography, which defines sleep stages according to EEG patterns, then neurophysiology, and psychology. Sleep structure changes across the life cycle. Healthy sleep should be sufficiently long, about 7 hours in Europe, and have a sufficient amount of non-REM and REM sleep.

Many sleep disorders may cause a change of the healthy sleep structure by difficulties of sleep initiation and maintenance, for example, in insomnia; sleep fragmentation, for example, in obstructive sleep apnea, restless leg syndrome, PLMS; and a reduction of sleep duration. Neurological sleep disorders, like, for example, Alzheimer and Parkinson disease, are associated with changes of sleep patterns that often precede their manifestation.

Thank you for your attention. Please continue on to watch the next section.

Segment 2: Impact of Disrupted Sleep Patterns

Geert Mayer, MD: Hello, I'm Geert Mayer, associate professor in the Department of Neurology at the Phillips-University in Marburg, Germany. Welcome to this section, "Impact of Disrupted Sleep Patterns."

What are the external factors that influence sleep quality? It's on one hand the societal level: that means where do I come from and which environment do I live in; the social level: which work do I have? How's my family functioning? How big is my apartment, for example; and the individual level: what are my habits?

So, if these are disturbed, it may result in insufficient sleep, which we all have in the industrial society sleeping less than 7 hours during the week time. And this is causing adverse health outcomes in endocrinological, metabolic, cardiovascular fields. And, finally, it results in an increased mortality.

Poor sleep quality is also interrelated with wider physiological functioning, as insomnia is a risk factor for many other conditions or events like major depression, anxiety disorder, suicidality, substance abuse, and metabolic diseases, cardiovascular diseases, falls. And interestingly, the incidence of depression in insomnia is much higher than in healthy controls at 13.1% vs 4% in the general population.

What are the consequences of disturbed night sleep? First, we have to understand the pathophysiological mechanism or the physiological mechanism behind it. We have a circadian rhythm through 24 hours. And during the day, as you see at about 16:00 in the afternoon, sleep is building up as measured in slow wave sleep, and then it's declining at a certain level and people fall asleep. So if we have sleep deprivation, this is on the right side, the sleep pressure increases. And finally, the result is that we have decreased levels of delta sleep, sleep pressure increases, and we have reduced REM sleep and an arousal in REM sleep resulting in chronic sleep deficit, tiredness, and cognitive problems.

What are the precipitating factors? As we have shown, of course, individual and societal factors are very important. But if this is ongoing sleep deprivation, we have this continuous hyperarousal with a cognitive cortical decrease. We have emotional changes. We have autonomic changes. We have changes and impairment in memory consolidation, which on the other hand shows changes in psychopathology and shows emotional dysregulation. And finally we have in the right upper hand, behavioral adaptation. It's perpetuating and it's getting worse and worse, and we have a vicious cycle.

So, how can we restore? What impact? How can I show that the impact is really great? Here we have a mathematical model. People were deprived 5 days to 4 hours of time in bed. And as you see on the left panel, performance is declining. And then 1 day, the people had the chance to stay 10 hours in bed, and immediately it improves -- sleep improves, memory improves, performance improves. And then it goes on with sleep deprivation and it gets worse and worse.

What is the interaction between the circadian system disruption, the misalignment, and the sleep disruption? We have changes in the endocrine system, especially important in women. In metabolism, for example, for adipose tissue function, insulin resistance in food intake, in gut health, flora, immune system, etc. And finally it results in diabetes mellitus and obesity.

What does it do in obstructive sleep apnea? Obstructive sleep apnea has so many fields which are influenced by the apnea. It's intermittent hypoxia, sleep fragmentation, brain structural changes, and intrathoracic pressure changes. And sleep architecture shows less REM, shows less spindles, and less slow wave oscillations. And this results in difficulty or impairment of memory processing, finally resulting in sleepiness, car accidents, absenteeism, etc.

To sum up, short-term disruption of sleep is normally well tolerated. We all know this, and we have a good outcome, but consequences of long-term sleep disruptions of more than 3 months in sleep patterns may cause metabolic, hormonal, cardiovascular, cognitive dysfunction. Fragmentation of sleep, such as in obstructive sleep apnea or restless legs syndrome may cause a reduction of non-REM and REM sleep with consequences for daytime functioning, like falling asleep and impairment in academic and social behavior. Short sleep duration as in insomnia may be predisposing to psychiatric disorders and difficulties in functioning in academic and occupational settings. Reduced slow wave sleep may impair declarative memory, and reduced REM sleep may impair procedural and emotional memory.

Thank you for your attention. Please continue on to watch the next section with Göran Hajak.

Segment 3: Insomnia Medication and Sleep Architecture: GABA-Targeting Therapies

Göran Hajak, MD, PhD, MBA: Hello, I'm Göran Hajak, professor of psychiatry at the University of Regensburg, and director of the Department of Psychiatry at the Social Foundation in Bamberg, Germany. Welcome to the section, "Insomnia Medication and Sleep Architecture: GABA-Targeting Therapies."

In this segment, we'll focus on the inhibitory neurotransmitter gamma-aminobutyric acid, in short called "GABA," which is involved in the sleep-promoting effects of the worldwide, most used hypnotic drugs, the benzodiazepines.

GABA acts mainly on subtype GABA-A receptors in the human brain. Activation of GABA receptors is always inhibitory. It increases chloride iron influx into the neuron, thereby it hyperpolarizes the neuron and reduces the activity and excitability of neural networks.

What are the main sleep pathways of the GABA system in the brain? Well, sleep is driven by GABAergic neurons from the ventrolateral preoptic area in the brain, with ascending inhibition of cortical areas, and descending inhibition of brain stem arousal pathways, with their activities switching the brain from wake to sleep mode.

GABA-targeting sleep medications increase that activity. Those drugs are benzodiazepines: tranquilizers, such as diazepam, lorazepam, or alprazolam; also hypnotics, such as temazepam, triazolam, and lormetazepam. Besides them, here are modern Z-drugs, non-benzodiazepine receptor agonists. Those are all hypnotics and the most famous are zolpidem, zopiclone, and eszopiclone. And they are more selective for certain subtypes of the GABA-A receptor compared to benzodiazepines.

Z-drugs have shown to be clinically very effective, as the benzodiazepine have been shown. This is just an example to show a long-term, 6-month treatment effect of eszopiclone on latency to persistent sleep, which is significantly shortened.

The same significant effect has been shown for wake time after sleep onset, though there's no doubt that those drugs improve sleep: in particular, subjective quality of sleep, which is felt by the patient.

How do those medications now affect our sleep architecture? Well, that can be explained by sleep profiles. This is an example of a healthy sleeper, and it shows in a sleep profile that there is a short latency to sleep, if you sleep normally.

Patients have a REM sleep, dream sleep, more often occurring during the morning hours. There is a reasonable amount of sleep in total sleep time, and also there are a certain number of awakenings, which are very short during the night. There is a reasonable amount of deep sleep stages 3 and 4, with the deepest sleep in the first half of the night and the lighter in the second half.

In acute insomnia, the pattern changes. There is a prolonged latency to fall asleep. You will have an increased number of awakenings, some of them from REM sleep, and other, of course, prolonged awake periods, mainly in the morning hours.

On the other hand, if a patient becomes a chronic insomniac, there will be a destruction in most of the patients of the non-REM/REM sleep pattern. There are multiple awakenings, and a reduced amount of deep sleep, and reduced amount of rapid eye movement sleep.

When patients now receive benzodiazepines as a treatment, most of them have a profit. The sleep profile has improved. There's a short latency to sleep. There's a sufficient total sleep time. However, what we see quite often is a reduced deep sleep stage 3 and 4 in those patients. And what is important, some changes in REM sleep as well, having a very short episode as shown here. And this patient typically shows what we see quite often, a hangover sleep episode in the morning, the latter being a significant problem for treated insomniacs.

To address this issue, I'd like to show you the next slide, which gives information on how a Z-drug might impact on car driving. This is experimentally done by measuring the so-called standard deviation of the lateral position (SDLP), in either simulation techniques or with real driving. And what can be measured is the sway within the lane while driving. And a couple of studies have shown that with GABAergic substances, even if they are taken hours before the measurement of the driving ability, they will have a negative impact on that SDLP, so that hangover has a result in daytime performance.

On the other hand, the microstructure of sleep should be addressed. One is the sleep spindles. Sleep spindles are electroencephalogram oscillatory events that occur during non-rapid eye movement sleep mainly, and deficits in those spindles relate to diseases with memory impairment.

Now benzodiazepine and Z-drugs have shown to consistently enhance sleep spindle activity, unlike other psychoactive drug classes. And that has made up the theory that Z-drugs may enhance memory function during the night by boosting sleep spindle activity. And at least some of the studies, as shown here with power spectrum analysis and memory tests, have indicated this might be true. So there's some changes maybe of profit for the patients.

On the other hand, the microstructure of sleep with those medications might also have other changes. One is what happens with slow waves. Slow-wave sleep is a pattern that shows up in non-REM sleep stages 3 and 4 mainly, and is supposed to be the most restorative sleep. It has been shown repeatedly that benzodiazepines reduce the amplitude. This is an example of an animal experiment and the same takes place in humans.

So, as a result, we see quite often with long-term benzodiazepine use, a destruction of the non-REM/REM sleep pattern, even when the patient, and especially when the patient is treated with medication. The multiple awakenings reduce deep sleep, the delta sleep is missing, and a reduction in total sleep time. So bad and non-restorative sleep is the result of long-term benzodiazepine intake.

So, besides that, GABAergic drugs have a lot of other actions, which most clinicians know. This is due to the fact that the GABA-A receptors are located in the whole body and the brain. And a lot of, either positive, but also negative effects might show up, like muscle relaxation or paradoxical effects, but also adaption, when taking benzodiazepines.

Summarizing what this means clinically, it is very clear that GABAergic medications improve sleep. They enhance spindle activity, which might have some impact on memory function. They increase light sleep of stage 2 and reduce awakenings. Patients feel there is an improvement of subjective sleep quality. However, they suppress deep sleep stages and REM sleep, and thereby impair the restorative function of sleep, which may lead to deficits in cognition. And what is important, they impair daytime performance during the next day, which is also true, due to hangover effects of that medication. They exhibit plenty of unwanted side effects, so it's highly recommended that they should be prescribed with caution and professional judgment.

Thank you very much for your attention. I would highly recommend you to continue to watch the next section with our colleague Thomas Scammell.

Segment 4: Insomnia Medications and Sleep Architecture

Thomas Scammell, MD: Hello, I'm Thomas Scammell, professor of Neurology at Harvard Medical School in Boston, United States. Welcome to this section, "Insomnia Medications and Sleep Architecture," and we're going to focus on orexin-targeting therapies.

In this segment, we'll talk about the neurotransmitters that regulate normal sleep and likely contribute to insomnia. We'll mainly focus on the orexin neuropeptides, which are also known as hypocretins. The orexins promote arousal and regulate REM sleep. And we think they're relevant for insomnia because many people with insomnia have some degree of hyperarousal, which can be improved by both psychological and pharmacological therapies for insomnia.

I'm not going to go through this in detail, but suffice it to say that there's all different physiologic manifestations, both in the brain and systemically that are reflections of this elevated level of arousal that we think is a key part of the cause of insomnia.

So, let's start off by talking about the main arousal pathways in the brain that might contribute to this hyperarousal.

We'll start off by talking about what we refer to as the ascending arousal systems, and these can be broken down into 2 broad categories. Some of these neurotransmitter systems make monoamine neurotransmitters. These include things like norepinephrine, serotonin, histamine, and dopamine, and these cell groups are found in the brain stem. And you can see the names for these various nuclei as they're scattered from the brain stem up into the back part of the hypothalamus. All of these systems are active during wakefulness. They all project to the cortex and deeper structures. And you can think of them all as having generally excitatory effects to help activate the brain and putting it into a waking state.

There are also cell groups that make acetylcholine and other neurotransmitters. And perhaps the one that's familiar to many people is the basal forebrain, which projects directly to the cortex and helps activate it. There are also acetylcholine-producing neurons in the pons, in the upper part of the brain stem. These don't go to the cortex, but they do project to the thalamus, and they help activate the thalamus so that during wakefulness and in REM sleep you can have active signaling between the thalamus and the cortex to help produce the rich mentations of those behavioral states.

Now in non-REM sleep, these arousal systems are all turned off. We think some of the most important cell groups for doing this are in the preoptic area; that's the front part of the hypothalamus. And the brain region that's been studied the most is called the ventrolateral preoptic area. This uses GABA and some other neurotransmitters to turn off all of those wake-promoting systems. And so you can really think of this as orchestrating the inhibition or inactivation of these wake-promoting systems to help get you to sleep. There are also some other sleep-promoting cell groups in the basal forebrain and other brain regions, but certainly these preoptic cells are felt to be the most important.

Now we want to spend a little time on REM sleep here, too. And remember, REM sleep is characterized by dreaming and muscle paralysis. We're not sure where the dreaming comes from, but it probably is related to some activation of these cholinergic cell groups in the upper part of the pons that activate the thalamus and, again, allow active thalamocortical signaling during REM sleep, just like it is in wakefulness. There's also descending projections that go down through the sublateral dorsal nucleus, and some other brain regions ultimately produce the paralysis of REM sleep. So you get these 2 phenomena -- dreaming and muscle paralysis -- as part of REM sleep.

Now I'd like to introduce one more neurotransmitter in this whole scheme, which is the orexins. We refer to these in the plural because there are 2 peptides, orexin A and B, and these have 2 major effects. The first is that they activate arousal regions. And you can see that through these descending projections to the same brain regions that we just talked about earlier, that help promote wakefulness. And in addition, the orexins also have ascending projections up to the cortex. In addition, the orexins suppress REM sleep by turning on brain regions that inhibit REM sleep. Some of these are the familiar characters, for instance, the cells that make norepinephrine and serotonin, but there's also other REM-inhibiting cell groups that get activated by the orexins.

So, the net effect is that the orexins do 2 things. They activate arousal regions, and help produce long periods of wakefulness, and they suppress REM sleep.

So, we can look at this in a schematic fashion here where we can really just boil it down to very simple elements where on one hand you have the wake-promoting systems that use monoamine neurotransmitters and acetylcholine to promote wake. On the other side, you have the ventrolateral preoptic area and other sleep-promoting cell groups. And these 2 have a mutually inhibitory relationship. During wakefulness, the monoamines such as norepinephrine, serotonin, and histamine all turn off the sleep-promoting cells in the preoptic area. And then during sleep, just the opposite happens. The preoptic neurons are turning off those wake-promoting systems. And so this allows for what we call sort of a bistable mechanism, where when you're awake, you're fully awake; when you're asleep, you're fully asleep because each side is inhibiting the other.

What this little circuit doesn't fully explain is how is it that you can stay awake throughout the whole day and how is it that most people can sleep reasonably well at night? And we think that the orexins play an important role in this because the orexins have a strong excitatory influence on those wake-promoting cell groups. And we think that they produce sustained firing in some of those regions. And then during sleep, the orexin neurons themselves are inhibited by the preoptic neurons. So the important message here is that the orexins stabilize wake by activating these other wake-promoting cell groups.

So, how does orexin signaling actually affect sleep architecture? This is a good graphical depiction of what happens in the absence of orexin signaling. On the top, you can see a sleep recording from a healthy wild-type mouse. And on the bottom, you can see a mouse that lacks orexins. And what you can see is how fragmented that pattern is in the bottom trace where every few minutes the mouse is alternating between sleep and wakefulness. It occurs during the active period and the rest period of this mouse and people have now studied in addition to mice, rats and dogs that lack of orexin signaling, and they all have difficulty maintaining long periods of wake. And they also tend to have shorter awakenings even during the sleep period.

And so inspired by this, several companies have now developed medicines that can block the orexin receptors. And these are known as dual orexin receptor antagonists or DORAs.

So, let's look at some clinical data of how these orexin antagonists actually affect sleep architecture. So when people have done a systematic review of clinical studies of orexin receptor antagonists, what they generally find is that these drugs increase total sleep time, both in healthy controls and in people within insomnia. And in addition, they tend to shorten the latency to enter REM sleep. So once you fall asleep, how long does it take to actually enter REM -- that tends to get shorter when people take a DORA. And some studies have also shown in increase in the amount of REM sleep, though that's a little bit less consistent.

And so here's an example with one of these DORAs, daridorexant. And what you can see here is that in 2 large studies of people with insomnia, daridorexant dose-dependently reduced the latency to REM sleep. Lemborexant, which is another dual orexin receptor antagonist, shows an increase in REM sleep, but it's mainly in the first part of the night, presumably when the drug levels are at their highest. And so you can see that with placebo, there's no change from baseline in the amount of REM sleep as a percentage of total sleep time.

Benzodiazepines and other Z-drugs, like zolpidem, similarly have very little effect on REM sleep, but on the bottom panels, you can see this increase in the amount of REM sleep. That's mainly in the first quarter of the night and maybe a little bit in the later quarters, but the effect is so small it's not of great consequence.

So, if there is an increase in the amount of REM sleep, it's probably mainly due to this shortening of REM latency. So you're going into REM faster and having more REM mainly in that first quarter.

So, we spoke earlier about how orexins are involved in sustaining long periods of wakefulness. And so, one of the other interesting aspects of these dual orexin receptor antagonists is that they seem to reduce the amount of time that people spend in long awakenings during the night. So in an analysis of 2 clinical trials, people treated with the DORA suvorexant had fewer long awakenings during the night compared to those who got placebo. Overall, these folks spent more time in brief awakenings that lasted less than 2 minutes rather than in the longer periods of wake. And the good thing is this seems to be associated with subjectively improved sleep quality, maybe because the brief awakenings are just less likely to be remembered.

Interestingly, with this DORA, the total number of awakenings was about the same. And so that suggests that the threshold to wake up from sleep is probably pretty normal, but your ability to return to sleep is better so that you see less of these long awakenings.

So, what does all this mean clinically? These dual orexin receptor antagonists provide clinicians with new tools for helping patients with insomnia. Their unique mechanism of action is appealing because it might enable improved sleep without some of the side effects that can occur with conventional sleep-promoting medications.

Earlier, we talked about how people with insomnia have hyperarousal and if enhanced orexin signaling contributes to this, then DORAs might help counter this hyperarousal and enable sleep.

In addition, we know that sedatives and Z-drugs increase the risk for falls and confusion in the elderly, but as DORAs have no direct effect on GABA signaling, then these problems might be less common with DORAs.

In addition, a clinician who's hesitant to give a benzodiazepine or a Z-drug to an insomnia patient with a history of substance abuse might feel a little different about a DORA because from what we know of orexin signaling, you would expect a lower risk of abuse with a DORA. Now, whether the DORAs are clearly safer remains to be established, but their novel mechanism of action does raise the possibility that these medications could be especially helpful in certain populations.

So, in conclusion, we've talked about how orexins normally promote arousal, help produce long periods of wakefulness, and they suppress REM sleep. We also talked about how dual orexin receptor antagonists or DORAs block orexin signaling. They can increase total sleep time, and they usually shorten the latency to enter REM sleep. Normal orexin signaling promotes long periods of wake and DORAs likely improve sleep by blocking this effect, enabling natural sleep drive to kick in so people can fall back to sleep more quickly.

So, thanks very much for your attention. Please continue on to answer the questions and complete the evaluation.

This is a verbatim transcript and has not been copyedited.