Hello everyone. Thank you for joining me today. I'm going to discuss the relationship between disordered sleep and rhythms and psychiatric illnesses. My name is Ruth Benka and I am the Chair of Psychiatry and Behavioral Medicine at Wake Forest University School of Medicine. So, psychiatrists, of course, have been very interested in sleep for centuries, probably, but the modern field of sleep research really began in the early 1950s when Aserinsky and Kleitman first described the neurophysiologic properties of rapid eye movement sleep and sleep and associated with the phenomenon of dreaming. After that time, there was a renewed interest in looking at now the relationship of sleep electrophysiology and actual sleep EEG and psychiatric illnesses, since so many psychiatric patients have complaints of sleep problems. One of the first things that was discovered was that individuals who had major depression were more likely than non-depressed individuals to have abnormalities in their REM sleep parameters. In particular, they would enter REM sleep early in the sleep period, usually in 60 minutes or less versus the usual 90 minutes or so, and that REM sleep amounts were relatively greater during the night and REM intensity, so the frequency of the eye movements was greater as well. This led to many, many studies of various groups of psychiatric patients in sleep laboratories, people with acute psychiatric illnesses, and a few years ago now, I had the opportunity to do a meta-analysis of studies done during usually the 1970s and 80s of almost a little over 7,000 subjects who were studied in sleep laboratories and that met research diagnostic criteria for various disorders. What we found was that in general, groups of patients with psychiatric disorders showed objective abnormalities in their sleep architecture, and in particular, they had reduced amounts of total sleep, their sleep efficiency was reduced, that's the amount of time they slept divided by the time in bed, they took longer to fall asleep, so sleep latency was prolonged, and in the case of mood disorders, we did find that groups of individuals with mood disorders had reductions in their latency to REM sleep consistent with the Kupfer finding. Also, it seemed that they had these studies decrements in slow wave sleep. Now, the problem with these studies was while groups of people with psychiatric illnesses tended to have objective sleep abnormalities compared to normal control subjects, these sleep architectural changes weren't really very useful in making psychiatric diagnoses and were not very useful in distinguishing one group of psychiatric disorders from another. However, I think some of the next important findings in the field were that people who reported sleep disturbance were more likely to go on and develop psychiatric illness, so it's not just that during acute episodes that people had sleep complaints, but that the sleep complaints tended to precede many psychiatric disorders, and in this recent meta-analysis, they found that not only was insomnia highly predictive of depression, which is probably the disorder that's been most studied along these lines, but also that insomnia significantly predicted onset of anxiety disorders, of alcohol abuse, and of psychotic disorders, and we know in the substance use disorder literature that people who have insomnia are more likely to abuse substances or relapse with substance use disorders. So, most sleep reports are subjective, and certainly many of these epidemiologic studies were based on subjective reports of insomnia. We had the opportunity a number of years ago to look at objective sleep disturbance and how that related to future onset of psychiatric disorders, and this is data from the Wisconsin Sleep Cohort Study where people in the general population were brought in at four-year intervals and given a variety of questionnaires and also studied in the sleep laboratory, and so what we did was we took individuals who had sleep studies in the laboratory and were not depressed at that particular time according to their questionnaire data, and we divided them up into the best and the worst quartile of sleepers and found that no matter which sleep parameter we looked at that kind of mapped onto insomnia, so sleep latency, how much time they spent awake after they fell asleep, the sleep efficiency, or the total sleep time, that the individuals in the lowest 25 percent of the population were far more likely to have incident depression four years later. So, not only does subjective sleep disturbance predict psychiatric illness, but objective sleep disturbance is also predictive. Not only is this true in adults, but it is true in children as well, and a number of studies have looked at the relationship between sleep disturbances even very early in childhood and later onset of psychiatric disorders as well as comorbidity of psychiatric illnesses and sleep problems in kids. This is a study where we assessed sleep problems as reported by parents in young children at the ages of four and a half and nine and then looked at, did a psychiatric assessment of them at age 18, and what we found was not only were children whose parents reported sleep difficulties consistently in early childhood more likely developed psychiatric illness, but there was some relative specificity between the types of sleep problems parents reported and the types of psychiatric disorders that these children developed later. So, individuals who had more insomnia reports were more likely to develop anxiety problems and externalizing symptoms, whereas children whose parents reported them as having restless sleep or increased movement during sleep developed higher rates of ADHD symptoms at age 18. So, of course, one of the most significant and concerning outcomes in people with psychiatric illness is suicide, and we know that suicide is highly linked to many psychiatric disorders, but epidemiologic studies have now suggested that insomnia is also an independent risk factor for suicidal ideations, suicide attempts, and completed suicides, and this is a meta-analysis that was recently published taking data from a number of studies and showing the consistency of these findings. So, insomnia is predictive not only of many major psychiatric illnesses but also of suicidal behaviors. All right, so the next thing I want to talk about is the relationship of sleep interventions on psychiatric outcomes, and so there are two interesting categories of interventions that have been used, and they're kind of opposite of each other. One is sleep deprivation as a treatment for depression and insomnia treatment as a treatment for depression, and it seems sort of contradictory because, on the one hand, I'm saying that disturbed sleep is associated with and predictive of psychiatric illness and particularly depression, yet sleep deprivation may also be a treatment, and what we now think may be going on is that the reason that sleep deprivation may be effective in mood disorders is that it tends to increase brain plasticity and increases pressure for slow-wave activity, and we believe that slow-wave activity in sleep is likely an indicator of changes in brain plasticity. So, there are different approaches that have been used to do this, either total sleep deprivation, keeping people up all night, partial sleep deprivation, particularly of the latter part of the night, so waking people up hours before they would normally wake up, or selective slow-wave deprivation, which I'm going to show you some data from in a moment, and that these interventions can improve mood, again, through increasing plasticity. Now, the other approach that's been used is REM sleep deprivation and increasing REM sleep pressure, and that also seems to be associated with an antidepressant effect, although, interestingly, when individuals undergo, or animals, depending on the experimental model, undergo chronic REM sleep deprivation, they also start to develop slow-wave deprivation, so it's hard to deprive individuals of REM sleep without also possibly depriving them of some slow-wave sleep, and the explanation for this was based on the homeostatic pressure hypothesis of the antidepressant response suggested by Alexander Borbet a number of years ago, and the idea here is that, and we sometimes see this, particularly in our bipolar patients, some strong diurnal variation in mood, so individuals wake up in the morning feeling not particularly in a good mood, they're more depressed. As the day goes on, they might start to feel a bit better. If we sleep deprive them, their mood increases even more as their homeostatic pressure to sleep, and particularly to have slow-wave sleep, increases, but then, unfortunately, one of the problems with sleep deprivation therapies, besides the fact is that the person is also sleep-deprived, is that once they have recovery sleep, they start to feel depressed again. Now, there are a number of other findings that have suggested that there are abnormalities in slow-wave sleep in people with depression, and one of the consistent findings is that the pattern of slow-wave sleep across the night may be altered. Now, normally, when we go to sleep at night, we enter sleep through non-REM sleep, of course, and our first non-REM period has a lot of intense slow-wave sleep. In fact, our slow-wave sleep, our deep sleep, is most intense during the first non-REM period, and then that starts to dissipate across the night as we quench our homeostatic drive for sleep, which, again, is expressed by our slow-wave activity. So, the delta sleep ratio basically is how much slow-wave activity do you have in your first non-REM period divided by the amount of slow-wave activity in the second non-REM period, and we know that people whose ratio is lower, so there's not as steep a drop from the first to second non-REM period, have a poorer outcome for their depression versus those who have a higher ratio are more likely to remain remitted. Also, some studies have shown that normalizing this delta sleep ratio, so increasing that homeostatic decline of sleep across the night, is associated with improvement in depressive symptomatology, and that a low ratio is actually a trademarker for depression, so it may not necessarily indicate the person is depressed, but it indicates an increased risk for developing depression. Finally, we know that antidepressant responses to a number of interventions are associated with increased slow-wave activity and or an increased delta sleep ratio, so an increased slope of the decline of slow-wave activity, and that is seen in ketamine treatment, so ketamine very much increases slow-wave activity. Sleep deprivation, as I mentioned earlier, by driving the homeostatic sleep drive is going to increase that delta sleep ratio, and many antidepressant medications that are effective also alter the pattern of slow-wave activity across the night. As I mentioned earlier, REM sleep deprivation also appears to have an antidepressant effect, and this was a seminal study done in the mid-1970s by Jerry Vogel and colleagues, where he took individuals with depression and put them in the sleep lab for three weeks, and each night either gave them REM awakenings or non-REM awakenings to basically interrupt their and decrease their REM sleep amount, and what he found was that sleep deprivation achieved through waking people up through REM sleep or reducing their REM sleep was more effective than awakenings during non-REM sleep in terms of reducing depression, and he proposed that one of the mechanisms for antidepressant treatment is that it's via REM sleep suppression, and we know that most of the antidepressants that we use, including the older drugs, the tricyclics, the monoamine oxidase inhibitors, and of course our newer agents like SSRIs and the dual serotonin and norepinephrine reuptake inhibitors all lead to REM sleep suppression. Now, not every drug that suppresses REM sleep has an antidepressant effect, and not every antidepressant that's effective suppresses REM sleep. I think notable exceptions are mirtazapine and bupropion. There also was a review of non-pharmacologic or interventions or behavioral interventions that reduce REM sleep, which is basically waking up individuals during REM sleep, that also suggests that even this type of suppression of REM sleep is effective in improving depression, and that short-term REM deprivation achieved by arousals didn't seem to be associated with negative effects, nor does chronic use of antidepressants seem to have obvious negative effects, but on the other hand, studies where we would deprive rodents of REM sleep chronically invariably led to their death, so this might not be a useful approach, but it's an interesting one from a mechanistic standpoint. Finally, this is another approach that we and others have used, and this was basically just a one-night approach to see what would happen if we reduce slow-wave activity, thinking that if a problem with homeostatic build of slow-wave activity is associated with depression, if we build up the slow-wave drive without necessarily depriving people of sleep overall, would that be helpful? In this study, we took individuals with major depressive disorder and had them in the sleep lab for multiple nights, and on one of the nights, and we didn't tell them which one it was, we played tones when they would enter slow-wave sleep to suppress the production of slow waves, and what we were able to do, as you see from the graph here compared to the baseline night, was significantly reduce their slow-wave activity, and on the recovery night, they had a rebound of slow-wave sleep, showing that we did build up homeostatic pressure. What we found was that whether we did ratings by blinded raters using Hamilton scales or we had the subjects do a Beck depression inventory, the day after the slow-wave deprivation, we noted a significant antidepressant response, and that this response was correlated with how much we had increased the slow-wave drive on the recovery night. We also found that the antidepressant response was associated with REM sleep suppression on the slow-wave deprivation night and long REM latency on the recovery night, and that the slow-wave effects and the REM effects were independent of each other and both associated with the antidepressant response. So clearly, manipulations of sleep do seem to be associated with the antidepressant effects of sleep deprivation. Another thing that we have started to look at, we and others, is the expression of sleep oscillations in the brain in people with a variety of neuropsychiatric disorders. And so in the case of major depression, we can basically take two of the classic in non-REM sleep oscillations, the sleep spindles here, which are 12 to 15 hertz bursts of activity during non-REM sleep and these big slow waves, and we can map them. And we know that spindles are most prominent on the vertex of the scalp because they're driven by the reticular thalamic nucleus. And slow waves are most prominent in frontal and prefrontal regions because they, as I mentioned, are related to brain plasticity. And of course, our frontal brain regions are the most plastic. We found evidence of impairments of slow wave function in depression. So one approach is to look at auditory evoked potential. So we can play a tone and look at the brain's response to that tone. And we looked both before and after sleep. And in normal individuals, the components of the evoked potential are scaled down by slow wave activity across the night. We see here pre-sleep, we have a more intense change in a larger change pre to post sleep. We see no change in the depressives. And this is the P2 component, again, which is decreased in individuals that are normal healthy controls and not decreased in depression. And this decline in the auditory evoked potential is correlated with slow wave activity in normals, but doesn't seem to have any relationship to slow wave activity in the depressives. Similarly, normally our theta activity during waking increases during the day as we're awake and it declines during sleep. And it declines in proportion to slow wave activity. And we saw that in people with depression that we didn't see this decline in theta activity. We also didn't see a strong decline in fast activity either. And whereas the decline in the low frequency theta activity here, four to seven Hertz was correlated with slow wave activity in normals, there was no association in depression. So it suggests that even if slow wave activity is present, it's not doing what it normally does in normal individuals. And finally, just to kind of make things even more complicated, we started looking at sex related differences in slow wave activity in depression and found very surprisingly that when we use this high density EEG sort of covering the whole scalp and looking at local expression of slow wave activity, that women here with major depression actually had increases in slow wave activity compared to non-depressed controls where as there's no difference in men. So there is a difference, a sex related difference in slow wave sleep in depression. And we were not exactly sure what the significance is, but we of course know that women are more likely to develop depression. So to summarize, we know that insomnia is associated with and predicts depression, suicide, and other disorders across the lifespan. Insomnia and major depression probably share some common mechanisms. There's evidence of similar genotype expressions related to both sleep problems and depression, as well as the fact that similar brain systems may mediate sleep and that are involved in mood regulation. We know that slow wave activity and the homeostasis of slow wave sleep is impaired in depression, that antidepressant effects are often associated with improvement in slow wave activity, and that there are sex related differences. So hopefully we'll continue to learn more about how sleep and sleep deprivation are associated with mood disorders. Well, before I leave the topic of altered sleep oscillations and psychiatric illnesses, I just wanted to mention the data in schizophrenia, because I think this is really quite striking and exciting. And that is that this is a work of Fabio Farrelli, who's now at the University of Pittsburgh, is that what he found for the first time and identified this is that individuals with schizophrenia appear to have a significant deficit in sleep spindle activity during non-REM sleep. And here are healthy controls, here are psychiatric controls, people with mood disorders who are on the same antipsychotic medications as schizophrenics, so it's not a medication effect, and that the spindles are pretty much reduced. Now he and his group went on to show later that this abnormality appears to occur very early in the course of illness, so these were young individuals who were in their first psychotic episode who either had no exposure to an antipsychotic medication or had been on one for less than two months, and suggesting that this, you know, may be one of our most sensitive and specific diagnostic markers for schizophrenia. Finally, even first-degree relatives of people with schizophrenia tend to have deficiencies in their spindle activity, so these are first-degree relatives who do not have schizophrenia compared to age and sex match controls, and we see that there is a significant decrement in spindle activity in these individuals. All right, so sleep oscillations may start to give us some clues as hopefully to some of the mechanisms of psychiatric disorders, but in any discussion of sleep, I think it's also important to talk about rhythms because one aspect of sleep regulation is that it is a rhythm. It is both a circadian or daily rhythm, but also there are seasonal rhythm changes in sleep, and individuals with psychiatric illnesses tend to have abnormalities in their sleep and other circadian rhythms, and in fact, there are altered clock genes that have been associated with most major psychiatric disorders, and people with particularly bipolar disorder, schizophrenia, and even PTSD and other mood disorders tend to be more likely to have sort of eveningness traits or they're night owls. They like to stay up late, have trouble going to sleep early, and then they can't get up in the morning. They also tend to be more seasonal, showing larger changes in sleep amounts and sleep timing in the winter versus the summer months, and in fact, many of them will have very disrupted circadian rhythms where they don't seem to follow any particular schedule of day or night. Of course, we're all familiar with seasonal affective disorder, which is a qualifier we apply to either major depressive disorder or bipolar disorder, and that consists of more atypical depressive symptoms, so increased sleep and appetite. It's a disorder which tends to respond to light therapy, and its prevalence is more at extreme latitudes. There was even a study done in the southern hemisphere in Australia and found that people who lived at more extreme latitudes there had higher rates of seasonal affective disorder. Now, people who are more seasonal, this is a study that was done in Scandinavia where they administered a global seasonality scale and found that people who were the most seasonal, the more seasonal people were, whether both males and females, the more anxiety symptoms they expressed and also the more depressive symptoms they just expressed, so seasonality goes with eveningness, it goes with more anxiety and depression. Mood disorders tend to be quite seasonal, so in comparison to even just control individuals, in this study they looked at people who were diagnosed with non-seasonal depression, who are still more seasonal than controls. Seasonal depression is more seasonal. Non-seasonal bipolar disorder looks pretty comparable to seasonal depression, and the most seasonal changing individuals in terms of their symptomatology are seasonal bipolar individuals. Nora Volkow recently published a study that kind of summarized seasonal changes that are not only seen in healthy subjects but that are exaggerated in individuals with psychiatric illness, so we see that in terms of mood disorders, we see more depression in the winter, we see more we see more mania in the spring and summer, there's also probably a little manic peak in the autumn, we see more suicide attempts in the spring, and we see more exacerbation of schizophrenia as well in the spring and here again in the winter. So focusing again on suicide, this is another important area, so not only is insomnia related to suicide but abnormalities in rhythms are also associated independently with suicide. So we know that suicide attempts tend to have a diurnal pattern, so they're more likely to occur in the evening and they're more likely to occur in people who tend to be awake and evening types. They also have a seasonal rhythm, again as I mentioned, suicide attempts tend to peak in the spring and anybody who's worked in the emergency room knows the spring is the time where we not only see a lot of suicidal patients but also individuals who come in with mania and psychosis. So again evening prototypes and people with greater seasonality have increased risks for suicidal behavior and that in studies that have looked at it can be an independent risk factor, so it's not just that they're more depressed. So again suicides have a diurnal rhythm, again peaking in the evening, this is a study of trained suicides and here is some data looking at suicides in people with mood disorders in the pink which really peak in the spring but even individuals in the general population or those without a history of mood disorder again peak in the spring and then have another peak in a smaller peak in the fall. Now this seasonal behavior is probably evolved in an adaptation to seasons because certain changes in behavior are important. A number of years ago we studied patterns of sleep and behavior in migratory sparrows, white crown sparrows, which really had a lot of features in common with people with bipolar disorder which is there was a seasonal component of their migration. So here we see in the spring and the fall this dramatic increase in activity. They also showed increases in goal-directed behaviors and impulsivity and showed some similar gene expression patterns to genes that were identified in association with bipolar disorder. And then if we look at people who have bipolar disorder and look and assess their sort of sleep tendencies again during periods of mania and depression in their depressive episodes, they very rarely report reduced sleep need and in fact they mostly report that they are excessively sleepy and that's here this blue bar. So the reduced need for sleep is very low, the increased need for sleep or hypersomnia is high in contrast to when they're manic where they report a reduced sleep need and very little excessive sleepiness. So similarities across the evolutionary breadth of life here. So again psychiatric disorders and suicidality are associated with eveningness, seasonality, and disrupted rhythms. There is probably an adaptive feature of seasonality but obviously the extremes are not so healthy. Bipolar individuals are the most seasonal so you should always suspect seasonality and bipolar disorder people who report a lot of seasonality and we know that sleep and rhythm disruptions can trigger manic episodes. All right so we talked about all these relationships between sleep and rhythm problems and psychiatric illness. What happens to psychiatric disorders if we treat the sleep problems? And most of the studies that have been done have looked at the effects of insomnia treatment on depressive symptoms and major depression and on the on the prevention of depression and very few studies have looked at the effect of antidepressant medications. Most of them or sleep medications, most of them look at the effects of cognitive behavioral therapy for insomnia which is the recommended first-line treatment for insomnia in people with insomnia disorder including those with comorbid psychiatric illnesses. So only a couple of studies have looked at pharmacotherapy. One study that was a relatively large randomized trial looked at the impact of open label fluoxetine with either esopiclone or placebo and found that the group that was treated with the hypnotic had improved sleep and they also had a faster onset and a larger antidepressant response to the treatment overall. Another study which was also relatively large compared the effects of extended release zolpidem or placebo added on top of open label escitalopram and while the group that received the hypnotic had improved sleep there wasn't really much difference in the depressive symptoms between the two groups so not clear how impactful that was. Now look at the literature that has been done looking at cognitive behavioral therapy. So these are non-pharmacologic interventions almost 6,000 subjects in 49 studies and what was found here is that depressive symptoms were reduced in individuals who had behavioral sleep interventions to reduce their sleep problem and that in the individuals who actually had mental health diagnoses that the effects were greater. That's not surprising because you have a sort of a bigger range of symptoms and interestingly they found that it was the change in subjective sleep quality which is actually what we measure when we assess insomnia which moderated the antidepressant effect. This is another study that actually looked at, did a meta-analysis of the effect of cognitive behavioral therapy for insomnia on individuals who actually had a diagnosis of depression and they looked at the impact on depression at the end of the treatment trial and then farther out and what they determined was that the individuals who were treated with behavioral approaches for their insomnia had a better antidepressant response at the end of the treatment trial although there was no difference between the treatment group versus the control group farther out although insomnia continued to be improved. So again these studies are probably suggesting that treating insomnia kind of speeds up the antidepressant effect. Now what about the prevention of depression because that's really the key. So if we're saying that insomnia is highly predictive of depression and sleep problems are predictive, does intervening and improving sleep early on prevent the eventual occurrence of depression and of course those are very difficult studies to do because you need huge populations and very long-term follow-ups but there have been a few studies that have tried to look at this and the results were that four out of six studies that tried to assess the effect of CBT for insomnia on development of depression found that there were decreased rates of depression in the treated individuals but it was really hard to parse out whether these individuals were simply being sort of caught very early in the course of a depression so it's hard to say whether it actually prevented depression but certainly there's evidence that it either slowed down or stopped the progression of depression. So again we now know from these studies people who have major depression who were treated for insomnia get better faster, people who might be sort of earlier, they're already having insomnia and maybe some depressive symptoms, the treatment of the sleep seems to slow down the progression of the depression. Other approaches for treating sleep issues, of course, include, you know, what do we do with our bipolar depressives who tend to sleep too much? And there are a number of approaches that have been used. One I want to mention is interpersonal and social rhythms therapy, which is a combination of behavioral treatments for insomnia, but also behavioral treatments for rhythms. This was developed by Ellen Frank and colleagues at the University of Pittsburgh, and she has evidence that it has been helpful for stabilizing people with bipolar disorder by stabilizing both their rhythms and their sleep, but there's probably not enough data to definitively say how helpful it is overall. Other things you'll read about are blue light walking glasses, because, of course, blue light is what impacts on the circadian clock. Total sleep deprivation and cognitive behavioral therapy have not been probably applied a lot in bipolar depression, but certainly bright light therapy does seem to be helpful and leads to significant improvement in people with depression. I now want to talk about a study that I was involved with with my colleagues, Vaughn McCall from the University of Georgia and Andy Crystal, who was at Duke at the time, and we wanted to sort of take the bull by the horns in a way and assess what happens when you treat insomnia in depressed suicidal people, because, of course, we know that most studies that are done looking at the effects of hypnotics on insomnia in general and certainly the effects of hypnotics in people with depression will exclude people who have suicidal ideation, and we also know that most hypnotics have warnings about, you know, they might increase suicidal behaviors. So the question is, you know, which is it? Does treating insomnia decrease suicidality, or do treatments for insomnia potentially make suicidality worse? So what we wanted to do was to see what happens if we take people who are depressed, have insomnia, and have suicidal ideation and treat their insomnia with a hypnotic and for this study we used zolpidem versus a placebo, so it was a double-blind placebo-controlled study. We used open label SSRI, fluoxetine, and almost all the subjects, and we studied about 100, we asked over 100 subjects for eight weeks, and we didn't have any suicide attempts, fortunately, and what we found, first of all, is the group that received zolpidem, which is the orange line over the eight-week treatment course, had significant reduction in insomnia compared to the group that had the placebo, although you'll see that both of these groups really showed significant improvement in their insomnia ratings, even though they started pretty much at the same point. I will say this was a what we call a high-touch study. We saw these patients very frequently, obviously because of the, you know, concerns about trying to manage individuals with suicidal ideation. We also found that the zolpidem group showed decreased suicidal intensity overall compared to the control group. Certainly, treating their insomnia did not make them more suicidal, and what we found, interestingly, across both groups, whether it was the placebo group or the zolpidem group, was that the individuals who reported more insomnia were more suicidal even when we controlled for their depressive symptoms. So, self-reported insomnia was independently associated with suicidality, consistent with some of the other epidemiologic studies I mentioned earlier, and finally, improvement of insomnia was also independently associated with decreased suicidal ideation, and I think this is probably one of the first studies that has actually shown that relationship. Now, the other thing that we were interested in was, of course, rhythms, and we did find that across the treatments and across the groups throughout the study, individuals who reported more eveningness, that they were more of a night owl type, had more suicidal ideation, again, consistent with other epidemiologic studies, and that at the end of the treatment, at week eight, those who reported a complete absence of suicidal ideation had a better activity pattern. This is their activity from actigraphy recordings, and so the black line here are individuals who had any suicidal symptomatology at the end of the study, and you see that they had lower activity overall. They sort of got up and going earlier, and they tended to stay up later at night so that the improvement in activity rhythms was also independently associated with decreased suicidal ideation. So, to summarize, we know that treating insomnia may reduce depressive symptoms and suicidality, and most of the studies, again, have looked at the impact of cognitive behavioral therapy for insomnia, and although this is directed at insomnia, one of the important questions is, is the therapy directly affecting depression as well as affecting sleep? We know it's improving sleep, but could the fact that individuals are getting this therapeutic interaction also having…is that having an antidepressant effect? Could mood improvements be related to the fact that people are just relieved because they're sleeping better, or are we really removing an exacerbating factor by improving sleep? And finally, and other interesting questions are, what are the objective effects of cognitive behavioral therapy or even medications on sleep oscillations, which we know are impaired in psychiatric illnesses, and how do those changes of sleep oscillations impact outcomes? So, to conclude, again, what I want to leave you all with today is that sleep and rhythm disturbances are integral components of psychiatric disorders and really should be assessed in every patient, that the pathologic processes, both for psychiatric disorders and sleep and rhythm, abnormalities likely overlap, that these contribute to the exacerbation of some of the symptoms of psychiatric disorders and probably the cognitive symptoms as well. And so, really, to conclude, there's a bi-directional relationship here. Disordered sleep and rhythms are both consequences of psychiatric disorders as well as contributors to their onset and exacerbation, and that attempt and anything that we can do to help improve sleep and rhythms are likely to help us get our patients better faster. And I am going to stop there. Thank you all. And now we have some time for questions and discussions, if I can figure out how to get at those. Let's back up here. I'm having a little, let's see, here we go at the bottom. I'm going to get into our Q&A. No open questions yet. Well, while we're waiting to see if anybody has any specific questions, again, I just want to emphasize that there's a very strong correlation between sleep problems and psychiatric illnesses in general. Today, I've talked a lot about insomnia, but people with psychiatric disorders have high rates of other sleep disorders as well that can be contributing to their psychiatric symptomatology. And I think one of the biggest ones these days that we're seeing is obstructive sleep apnea. And so this is a situation where people will have collapse of the upper airway during sleep associated with loud snoring, with obesity, with diabetes. Many of the drugs we give our patients cause them to gain weight and may increase their arousal threshold, contributing to them having sleep disordered breathing. And people with sleep apnea have increased depressive symptomatology. So also it's important to screen for that. I see a question in the chat. This is great. You discussed how antidepressants reduce REM sleep. What are your thoughts on how antidepressants increase the vividness of dreams? That's a great question. And it's certainly something that we see not infrequently, certainly with some of our newer antidepressants. And for those of you who have been around for a while, I remember back when we used monoamine oxidase inhibitors more frequently. So MAOIs are probably the most potent REM sleep suppressing drugs we have. And in fact, if we record individuals on MAOIs in the sleep lab, we could see that they have little or no REM sleep. And yet sometimes they report horrific dreams. And that's because what we probably are seeing is when people are on REM suppressing antidepressants, there is a buildup of REM sleep pressure. And that's often evidenced by the fact that when people are on these drugs for a while and then skip a dose, they may have a lot of REM sleep rebound, very vivid and extensive dreaming and waking up a lot from dreaming. And so REM rebound is part of the withdrawal effect. But we can also see this, we see dissociation of REM sleep elements in people who are chronically REM sleep suppressed. And so the reason that you may get these vivid or horrific dreams is that this buildup of REM pressure is kind of expressing itself either during the short episodes of REM sleep, or I have actually recorded years ago, people on MAOIs in the lab, and they would wake up with these horrific dream reports and they were not in obvious REM sleep when this was occurring. So again, you may have this dissociation of dreaming from REM sleep. The other thing to keep in mind is that you don't have to be in REM sleep to dream. So we can dream in any stage of sleep. And in fact, I think in your parasomnia lecture, hopefully, if you attended that one, you heard about how night terrors and other non-REM parasomnias are associated with vivid and sometimes very frightening dream mentation. So we can be having intense dreams, both in non-REM and REM sleep. And again, it can be dissociated. Hopefully that was helpful. Any other questions? Okay, we still have a couple more minutes. Maybe talk a little bit about, ah, here's another one. Great. A couple more questions. Thanks for the presentation. Thank you for the compliment. I recently inherited a patient who carries a seasonal affective disorder, bipolar type diagnosis with hypersomnolence. Previous treater put her on caffeine pills at 2000 milligrams a day. Patient continues to complain of daytime fatigue. Is there a place for this amount of caffeine for treating the hypersomnolence? Well, caffeine is not a great treatment for hypersomnolence. So caffeine, of course, blocks adenosine receptors and caffeine is most effective probably when people are over-dosing. So caffeine is not a great treatment for hypersomnolence. A bit sleep deprived. So it's not necessarily a potent stimulant in and of itself. So I wouldn't probably use caffeine in that situation. I would think about for a bipolar, and I assume with depression, to think about things like light therapy. Now, if it's chronic hypersomnolence all the time, not just during episodes of depression, then a sleep evaluation is warranted because the individual should be assessed for other sleep disorders. Again, if there's any other risk factors like obesity, snoring, diabetes, type two, other risk factors, it could be a case of untreated sleep apnea. There are other hypersomnia disorders as well. And so an overnight sleep study to rule out a primary sleep disorder like sleep apnea followed by a daytime nap test or a multiple sleep latency test will quantify the degree of hypersomnolence. Another thing in people who have mood disorders associated with self reports of hypersomnia, although some of them can actually have true hypersomnolence, a lot of them are not necessarily excessively sleeping, meaning that they don't sleep excessive amounts or have abnormal tendencies to fall asleep. So that's why a sleep evaluation would be ideal. Okay, next question. Do the Z drugs have antidepressant effects even without SSRI administration? Somebody had a session about this a couple of months ago. I don't know that there is really solid data to that effect. I think decades ago, I think there were some studies that suggested that maybe benzodiazepines initially might have had a bit of an antidepressant effect. And again, I think the issue is that via an impact on sleep. I will say that the Z drugs vary in terms of their, you know, anxiolytic effects. And again, we, you know, generally would not treat a true mood disorder with a Z drug alone. Okay, another question here. I don't know the schizoaffective disorder, bipolar type. I'm not sure what the question is. If that person wants to add a little bit more, I can try to get that. Another question is to comment on low dose antipsychotics like Seroquel for insomnia. I think it, you know, really depends on the patient that you're treating. One of the advantages we have in psychiatry is that many of our psychotropic agents, the drugs that we use for psychiatric illnesses, also have sleep effects as sort of side effects. And so we can sometimes use a drug that does not have an FDA indication for insomnia, but does have an FDA indication for a disorder that we're trying to treat on our patients. So, for example, in people who have, you know, bipolar disorder or psychotic disorders, where insomnia can be quite severe, we can use sedating antipsychotics in those patients because we may be trying to kill two birds with one stone. We're going to fix their psychiatric illness, but also help their sleep. By the same token in people with depression, you know, we can sometimes choose a more sedating antidepressant in someone who has a lot of severe insomnia in combination with their mood disorder. I'm not a big fan of using an antipsychotic for what we would call a primary insomnia. So someone who has insomnia and does not have a psychiatric indication for that medication because of the associated risks. And I think there are a number of other agents that we can try that are a bit safer. Okay, any other questions? Looking at the chat here. Okay, I think the APA has just let everybody know it is coffee break time. And probably at this hour in the afternoon, a little caffeine to block your adenosine receptors will be helpful as we get ready for the next lecture, which I'm looking forward to as well. And again, I think it's going to raise the issue of not only assessing sleep, but assessing rhythms because our teenagers tend to be one of our most phase delayed populations. So next we'll hear about teens who can't sleep. Is it insomnia or delayed sleep phase? Thank you all for your attention and have a good afternoon. Thank you.

sleep disorders

psychiatric illnesses

REM sleep

mood disorders

insomnia

depression

sleep interventions

brain plasticity

suicide risk