Good afternoon. My name is Corinne Berger. I'm a neuro ICU pharmacist at New York Presbyterian Hospital. And I'll be your moderator today, along with Dr. Margaret Pisani. Before we begin, just a few housekeeping remarks. If you want, you can move to the center. The exit is behind us. Please silence your cell phones if you can. If you need to leave, please just do so quietly. I think the rest of this is pretty intuitive. The last part I'll just say is we are going to tweet this session. So if you're tweeting, please use the hashtag CCC49. We would definitely encourage that. And we will do Q&A after each presenter. And so unfortunately, Brian was unable to be here for the talk. And I'll let Dr. Pisani introduce the next speaker. Welcome, everyone. Thanks for coming. I know there's a bunch of competing sessions at the same time, so we're happy you're here. As Corinne said, Dr. Gelbach is sick. So Dr. Knauert has kindly agreed to step in and not only give her own talk, but she will be giving his talk as well. Dr. Knauert is an assistant professor at Yale University School of Medicine. Welcome. Thank you. All right, good afternoon, everyone. Forgive any awkward transitions. This certainly is a topic near and dear to my heart. And it's a shame that Dr. Gelbach's not here to give it, but I'll try to do justice to his wonderful slides and then go into my talk. These are Dr. Gelbach's disclosures. My disclosure is that I'm not Dr. Gelbach. He had funding through NHLBI, the Brain Research Foundation, and current funding in unrelated research in NINDS. Thank you. The objectives for this first talk, most specifically, are to review the two major systems responsible for regulating sleep and wake. We're going to then characterize the effects of circadian disruption on sleep and characterize the effects of circadian disruption on sleep and characterize the effects of abnormal sleep timing on circadian rhythms, understanding the very reflexive relationship between the two systems. And then briefly at the end, we'll discuss implications for clinical trial design. Certainly this is a field that it's challenging to do clinical investigations in. And I think this is a good place to start. It is not hard to imagine in these highly artificial, loud, disruptive, complex, anxiety-provoking, pain-provoking environments that people don't sleep well. But understanding the complex biology under it has taken quite a bit of work and several decades of investigation. And I think we're really at the cusp of understanding what's going on in the ICU from a sleep and circadian perspective as technology and knowledge advances. And so the first thing to understand is that really two processes in our body contribute to sleep. And the first is called process S. It is also known as process H. And that's what you see here on the slide. It is called the homeostatic drive. That's where you get the H from. And this really is the very simple, linear accumulation of sleepiness the more that you're awake. This is largely driven by adenosine accumulation in the brain. And the more you are awake, the more sleepy you are. In contrast, there's something called process C. This is a little more simply named for circadian. And this is an oscillating system that promotes arousal to combat the sleep process during the day and then cycles off to promote sleep or allow sleep during the night. Here on the right-hand panels, Dr. Gelbach has illustrated really that the homeostatic drive to sleep is very closely tied to slow wave activity. This is a metabolic and growth and recovery sleep stage. And many of those processes happen during that time. And here he's also drawn for us the melatonin curve. And he's done that to indicate that really the circadian system is driven by melatonin. And also a note here we'll get into a bit later, but this is really something that drives REM sleep or dream sleep, thought to be very important for our cognitive processes. Driven a little more, drawn a little more simply, I've added in here a two-process model, just drawn a different way. And I wanted to highlight this so that you understand this interconnection. Again, we see our homeostatic sleep drive going up and then down in the purple line, process S in this slide. And then we see this green circadian oscillation. And what I wanted to highlight for you is this idea of a sleep gate. And here when our homeostatic drive is peaking, we are the most tired. You'll also notice that just before we go to sleep, before habitual sleep time, there's this real peak of circadian activity. And what I like to help people remember this with is any of you have been around toddlers about bedtime? This is the toddler running around with their pajama pants on their head saying, I'm not sleepy. And actually, it's biologic. And what is underpinning that is this last surge of arousal from the circadian system that then cycles off. And what it creates is this thing called sleep pressure or sleep gate. And what's important and what applies to our ICU patients is that when we are sleep deprived or have poor sleep, this process S that I told you about just accumulates up and up and up and up. And it does not, as we see in the normal, it does not decrease with sleep because you don't have that restorative sleep. And instead, what's happened and then underneath it all is process C just oscillating away. And this is why we call this self-maintaining oscillation. And what's important and what's interesting to understand is that at different points in the sleep deprivation curve, this sleep pressure or the sleep gate is actually different sizes. And it is not linear. And you are not necessarily able to sleep biologically or feel like sleeping biologically at different times of day. And if you look at this, you'll actually see that you can have more sleep pressure and less sleep pressure even as sleep deprivation accumulates. And this is important for our ICU patients. So let's talk a little more in depth about that process S, that sleep process. Normal sleep does have a timing pattern. And that is to say that you go through stereotypical patterns or sleep cycles. And so you go stage 1, 2, and 3, then up into REM, and over and over again. This is a 90-minute cycle that changes slightly, but in general, repeats itself in the same way three to four times overnight. And we'll highlight the significance of this in fragmented sleep a little bit later. But our patients are getting very brief periods to sleep. And these 90-minute cycles just are not occurring. There is a little bit of degradation just between young and old and slightly different patterns. But overall, those 90-minute cycles maintain. When we have sleep deprivation, and I have here and circadian, because I'm an honest sort and I can't separate them. So I'll talk a little more about that later. But when we have sleep disruption, we also have increased inflammation, impaired immune memory. We have cardiovascular dysfunction. We have an increase in sympathetic tone and a higher arrhythmia risk. And we also have metabolic dysfunctions. We have increased insulin resistance and poor glucose tolerance. If we then flip over and try just to talk about the circadian piece, we see that this is an incredibly complex interdependent system in which we have clocks telling us what time of day it is. And so this is one great illustration of both the central clock and the suprachiasmatic nucleus and also the peripheral clocks. But put in a different way, I would like you to understand that light is a main zeitgeber or external cue that tells us what time it is in the circadian world. This through the eye and into the brain is one of the main orienting forces for our circadian system. And that, in turn, tells our peripheral system what time it is as well. And so you could imagine that you get up in the morning, you see light, and you say, great, it's time to be awake, I'm gonna eat some breakfast, I'm gonna go to work. And all of a sudden, you're gastrointestinal, you're immune, and your cardiovascular system with the program and doing the right things at the right time. The reason I redisplayed this picture versus the prior one is that I really want you to notice this bidirectional relationship. And so it turns out if you wake up in the middle of the night and have a donut, your gut's gonna say to your brain, I think you might be wrong, I don't think it's the middle of the night, maybe it's morning. And so we wanna start to think in this ICU about how all these signals are going and the direction they're going in. And so if you have mobility at the wrong time, feeding at the wrong time, social interactions at the wrong time, and certainly light at the wrong time, the circadian system can become very impaired. We also have ways of measuring the circadian system and primary among them is melatonin, which is generally low during the daytime hours, up overnight, peaking around your sleep midpoint, and then coming down and going through what we call offset around the time that you wake up. We have also core body temperature and cortisol that also let us cues to what circadian time it is in your body. We are able to leverage a lot of these measurements in doing our measurements and doing our investigations of ICU sleep and circadian disruption. But as you might imagine, we need to take frequent measurements over the 24 hour clock and that this can be quite resource intensive. And so when we live in harmony, this must be Brian, it's Iowa. So when we live in harmony and what we get, we have the sleep and circadian system that are working together. And so we are sleeping and waking at the right time, we're doing rest activity at the right time, fasting and feeding and light, dark exposure. And this is what I was alluding to with that bi-directional relationship is that if you want to be efficient as an organism, it makes a lot more sense to only quote unquote power up when you're gonna be using systems and then power down when you're not. And so the sleep and circadian system often work together in that way. We know that when you have acute circadian misalignment, or excuse me, chronic circadian misalignment, we have night shift workers who have higher risk of cardiovascular disease, metabolic disease in the form of type two diabetes, and higher risk of cancer, notably breast and colon. We also know that poorly timed sleep induces circadian rhythm abnormalities. And we have between 6.4 down to 1.1% of our genes oscillating with circadian clock. So keeping in mind in every tissue that we have day night patterns in the genes that we express and that this can be abruptly reduced in the case of circadian abnormalities. So let's think a little bit about and put ourselves in perspective of what our ICU patients are experiencing. And so these are five ICU patients. Each horizontal rectangle gives us a full day and the black is denoting sleep while the white is denoting wake. And so what you're seeing that this person is having from noon, evening, midnight, and 6 a.m. So they're having daytime sleep, they're having sort of evening sleep, a little bit of night sleep, certainly very short sleep bouts, and certainly highly disrupted sleep in the worst example here down in number five. So this is a great visual example of what our patients are going through. And paper after paper studies show that we really are seeing folks sleeping about half of the time during the day, being really only in N1, N2 very light stages of sleep, and having what you see here, these very high number of arousals where they're being woken up as far as their brain waves every two to three minutes. We also know that our patients have some circadian disruption. And so this is a very nice, on the left we have a panel that Dr. Gelbeck led. And what he's depicting here, again on the x-axis, is clock time. And so he starts here in the evening at 6 p.m. or 1800, and then on through the middle of the night, and then into the following day period. He's drawn here in this black rectangle what you might call a typical or a normal sleep schedule. And then he's used melatonin onset. So if you recall that curve, that uptick right before habitual bedtime. He's depicted, if you take the patient's uptick by measuring their urine metabolites, and you say when they would sleep, if you use that to predict it, he shows here that two of his patients really were very advanced, having a sleep onset between 6 and 9 p.m. A few folks that are pretty normal, maybe at 11 p.m. or midnight sleep time, and then a whole bunch of folks who are more and more delayed, all the way to this last person who has a biologic night, a sleep onset time that is predicted to be right around 7 a.m. And it's really this work and some other similar work that tells us that most, we think that most patients in the ICU are delayed. And I'll talk a little bit more about why that might be in my talk. Similarly, in some other work, we see that, not surprisingly, you can see lower levels of melatonin in patients. And so here he has folks who are all critically ill, but he has delirious folks, and then normal folks. So normal are with the, excuse me, with the circles. So a little higher. And you see a little bit here at this nighttime period. Excuse me. We have here the evening hours, the middle of the night hours, early morning hours, and then daytime. And so you see our normal folks with a peak and a profile, but you notice that they're peaking very late into the morning. So 5 a.m. to 8 a.m. Maybe showing a similar phenomenon to these folks. And then our ICU people who are delirious with the triangles along here are just suppressed with no rhythm at all. The jury is very much out on etiology, though I think we do have some ideas. All right, and one of these ideas is what's going on in the ICU. And I alluded to this with my picture of the ICU room and the telly and the renal replacement and the ventilator. But our patients are incredibly disturbed. And so this is a little bit of data from some work at Yale with us. And what I've done here is I have time blocks. We have 8 p.m. to midnight, midnight to 4 a.m., 4 a.m. to 8 a.m., and 8 a.m. to noon. So trying to break up the day into times that you think should be more and less busy. And here in this middle column, I've just said how many minutes per hour is their activity. It doesn't matter how it could be. It could be one block, it could be 20 blocks. But here we see that in the evening hours, we have 26 out of 30 minutes are occupied. We do dip down into that middle night period. And then about 20 minutes and then about 30 minutes in that busy sort of rounding part of the morning. And I think this is not unexpected, though this is probably higher than most of us would want for our patients. But I think what is even more interesting is if you take the data, and we did this by doing videography of the room door, and you look at how much time patients have between each period to rest, how long are they left alone. Here in the evening period, it's about 19 minutes at a go. In the middle of the night period, they're getting about 30 minutes at a go to rest before someone comes back into their room. Again, even shorter during that 4 a.m. to 8 a.m. and then very, very short during the daytime. So we see that these activities, these presence of folks in the room where they're bringing with them sound, they're bringing with them light, they're bringing with them disturbance, and maybe they're doing something painful. This could really give us clues about what's making our patients sleep so poorly. Dr. Gelbich made a beautiful slide here really summarizing what's going on with these patients. And so the biologic night, he's showing us it's delayed. He's showing us that sleep is all over the place and very broken up. And then he's giving us some idea of what might be causing this. You'll see in my talk, we did not confab about this, but we should have. You'll see a very nice example of this later. You also see change of shift in pre-rounds, maybe some of that reason for all that business, 4 a.m. to 8 a.m., a trial, the consult team, visitors. Our patients really have very little time to rest and are getting very abnormal cues. Okay, and this is where we come into our objectives of understanding that sleep and circadian are feeding upon each other. And so if we have that increase in daytime sleep pressure, remember that red line going up and up and up, but no daytime light exposure, poor nighttime sleep pressure, and nighttime sleep disruption, we get into a cycle of missed time sleep and circadian disruption where one is making the other worse. And all of this leads us really to what we wanna do, which is to understand this phenomena in the ICU and how can we study it. And so this is Dr. Gelbeck's list and it really is a great list, understanding what's going on. And so we need to certainly pay attention to Zeitgebers. The light-dark cycle has received a lot of attention, but a lot of the attention has been on making the ICU darker at night rather than lighter during the day. We'll talk a little bit about physical therapy and how important mobility can be to orienting our folks in a circadian manner. And also how nutrition may be a cue to correct circadian signaling. So you remember those peripheral clocks that I was talking about. We know that our clinical trials need to be multi-component and interprofessional because we've already named multiple disciplines who are experts in these fields. And we also, of course, need to think about things that we haven't touched upon, such as pharmaceuticals. We also need to do continuous QI with feedback. I'll tell you in our own experience of creating a nighttime protocol at our institution, we did great and then we turned around and six months later to realize that they had hired a whole bunch of ultrasound techs to do Dopplers throughout the night to make sure that people were getting their studies done quickly. And so you do have to continuously be feeding back and seeing what the health system is doing, really maybe with a different intent, certainly an intent to help patients, but with maybe a different framework. This idea of customized versus routinized. And so we have our folks who are early, those few folks that had biologic night earlier in the evening. We had some folks that were normal and some folks that were delayed. And so we'll talk a little bit more about this with light, is if we don't know where someone is in their circadian day, it is very hard to support them with chronotherapy. As I alluded to with those continuous curves and taking multiple measures, it is very difficult to know this because we need to take measurements over 24 hours. And so we're looking to new technologies to understand people's orientation. And then of course, it is very difficult to separate sleep and circadian. We cannot have one without the other, but it's hard to know which has influenced us. So with that, I'll end. Dr. Gelbeck shares this with us, but I think it is a wonderful quote. If sleep does not serve an absolutely vital function, then it's the biggest mistake that evolutionary process has ever made. Well, that'll thank you for your attention. We'll take questions too. If anyone has questions, I ask that you please come to the microphone. Dr. Knauert, I have a question. Is there a easy way or is there a way for people to know if their patient's circadian rhythms are disrupted in the intensive care unit? There is not. So I think that customized versus routinized question is I think you can understand that folks can't move their circadian clock tremendously, so maybe a few hours per day, even under extreme disruption. And so you could argue, I think, that early in the hospitalization, if you could get some idea into someone's typical habits, you could have a guess within four hours or three hours, and maybe at least know that someone was a night owl or a morning lark, and try at least to take that first step. Other than that, I think you're stuck until we get some rapid diagnostics on board. Thank you. Okay. Yep, I am next to you. So Dr. Knauert will actually present her own talk now. I know. Uh-oh. Okay, good. All right, and so this is certainly a specific sub-example of what I talked about, and we'll dig in a little bit and hopefully clarify some things that seemingly were glossed over in that first talk to really to give you that 50,000-foot view and now dig in. And so light is a really important zeitgeber, and we'll talk a little bit about the biology behind that. I have nothing to disclose. I am supported by a career development grant and by an ASM Foundation Strategic Research Award, both of which I'm very grateful for. So I will be focusing on light, and I would like to identify and modify, identify and discuss elements of ICU light environment that disrupt circadian health, that modify snuck in there, because that's what I want to do. Sorry about that. And we want to apply current evidence to making changes in ICU light delivery that will support circadian health. And so first of all, why does light matter? And this is the same graphic. I actually stuck it into Dr. Gelbach's talk because I like this bidirectional piece so much, and I think it's very helpful to understand that we have things going in and feeding up, things going in and feeding down. But we really be focusing on light, which is by far the strongest circadian zeitgeber. And so when you put light against any of the other things I mentioned, the feeding, the mobility, the social interaction, sleep itself, you really can win with light. But when it's so disrupted, those things also become important. And so certainly it's a team effort biologically. And to talk about light, I have to introduce something that is inherently complex. And this is the circadian phase response curve, which summarizes the circadian effect of a zeitgeber. And in this case, this is light. And so this is a cartoon, and I'll orient it to you. Many things in the circadian world, we start at noon, we go through the nighttime, 2400 or midnight, through the early morning hours to noon and afternoon again. I've drawn here a typical sleep period from about 10 or 11 p.m. to 5 or 6 a.m. This is just, again, a cartoon. We have a dividing line here. And any time that this line is above it, it indicates that the exposure I'm discussing is gonna cause an advance, which means earlier, which means if you normally were gonna go to bed at 10 p.m., you're now biologically gonna wanna go to bed at nine, maybe eight, maybe seven. So this is an advance. Anything below the line is a delay, okay? And so you'll notice right away in this cartoon that there is sort of this period before sleep, which typically causes delay, this very sharp relationship during sleep, a little bit more of an uptick, where you transition to advance. And we'll talk about how we get these because I think it's helpful. And so if you were to take five people or five mice or five porcupines, and you were to get them into a regular routine, sleeping, waking, sleeping, waking on a 24-hour clock, and then at noon, you were to flash 10,000 lux for 30 minutes. In this particular dose of light, you would not get much effect. And you'd be like, that's fine. And then you would do it again. And so you would do it at four o'clock, 10,000 lux, different group of five little mice or monkeys or porcupines or whatever. And you'll notice now when you flash light in these eyes, what you get is a phase delay. And you're like, oh, that's interesting. So if I do it at this time, I get a phase delay. Again, a different group of five animals, more delay, more delay. And so what happens is if you keep doing this experiment at different times relative to when people are habitually sleeping and waking, you get a clear pattern. And what you get is this thing called a phase response curve. As I said, this is a cartoon. And what you say when you describe a phase response curve is say, well, this is the phase response curve for this amount of light for this many hours. Or this is a phase response curve for melatonin. Or this is a phase response curve for mobility. Or this is a phase response curve for feeding. So all we're saying is that you take some Zeitgeber, some cue, and you give it at different circadian times and you see what it does. And so this is the one for light and it's very important. And what I'd like you to notice, as I pointed out before, so we're oriented to time and we're oriented to delay, and excuse me, advance and delay. But notice that this is bigger and deeper and centered around habitual bedtime. And that this is lighter, excuse me, less amplitude and smaller. And so it tells you that if you give a little extra light here or you give a little extra light here, our biology's gonna favor delay, which is interesting because in the ICU our patients are delayed. And if you say flashlight inadvertently throughout the night giving care, but then you leave the shades down because everyone wants ICU patients to rest, you're gonna favor delay further. And so I think this is a lot of what underpins the delay in the ICU among other things. And so what do we see in the ICU? And I've already alluded to this. So this is a very nice study of about 17 patients, about 35 to 40 patient days. And all they did was measure light. And so again, this time starting at midnight, 8 or 9 a.m., noon, and back to midnight. We have the median lux was 64, which is quite low. If you remember going outside on a sunny day is 20, 30, 40, 50,000 lux. And you know that circadian influencing light levels are on the two to five to 10 lux range. But what you see here is that we're actually pretty good at keeping light levels low during the nighttime. I think hospitals in general have gotten that memo and they understand that they should keep it dark. But what they're utterly failing to do is to get that light high enough. And so this looks like a nice peak in the morning and you're like, Melissa, no problem. We'll have a nice advance. But if you look at the intensity of it, it's very, very low. So even the highest part of interquartile range are only getting up to 140. And so these folks are not being exposed to any daytime light at all. We see similar data. This is data by Danielson et al with Dr. Gelbeck as the senior author. And he did something very nice. He looked at the median light levels in blue and you see these very low light levels. These are each individual different rooms in his ICU. And then with these low medians, we see really maybe 100, a little higher than the other paper. And this room was certainly a little brighter and these rooms certainly tremendously dim. And then he looked at the maximum brightness and he's getting up to a little higher numbers than us. So maybe around 200, as high as 600. And what you find out is driving this is a few rooms who have nice big windows. And these are the folks that are getting those high light levels. And in these rooms with small windows or deep recesses, they're not getting much light at all. And so really architecture does matter in this case. And then finally, this is a study at Yale. I apologize for the busy table. Dr. Bassani might jump up and strangle me. But I do wanna show you some of this detail. So in the middle of the night, this rest period between midnight and 4 a.m., again, we have these very nice low mean lux levels, 20. And then our sleep protocol arm, 15. A lot of variability. And we don't know biologically what happens if you're flickering the lights constantly. But then when you look at the maximum light levels, they're only 30 or 40. Again, with a lot of variability. But if you looked during the daytime, we're only getting up to 120, 130. And the maximum levels, very reminiscent of that work by Fan et al, we're really getting to 170 to 190. So again, highlighting this pattern of maybe too much light, too much variability at night, and too low light during the day. This is interesting. This is a single patient example of ICU light. And so this is the videography data I touched on earlier. And what we did was we just put a motion detecting nanny cam at the door. And when someone walked in, we marked it an entrance. And when someone walked out, we marked it an exit. And we just wrote them all down. And what I'd like to draw your attention to is this one from 2.30, 2.34 to 3.36 in the morning. This was the patient's bath. And then of course, the 4.16 to 4.33, the emergency trash can emptying. And so when folks say, well, you know, these folks are, these people are critically ill. You need to be in that room taking care of them. I would argue that we really don't need to be doing this piece. And so this is the same patient's light profile. And what we see here again is these, you know, a little bit of light during the evening hours as they're getting their meds, getting settled, 10 p.m. Maybe a little something happened here, an IV change, a pump change, I don't know. Then we have this tremendous light dose right in the middle, 2, 3 a.m. Influencing our patient quite a bit in their circadian, in terms of sleep and circadian rhythm. So this is gonna wake you up certainly to have your bath at this time. And this is certainly a very high dose in terms of Zeitgeber. If you look at just practical things and you say, well, what's in the room that's making these patients get more or less light? We see on the first set of yellow, green, and blue, we see that the bright lights are on about the same across the day. So 30% in the morning, 40% at noon, and 30% at night. Similarly, the television is on across the day and we have patients throughout the night who have it on, some by request and some not. And then when you look and see if the window shade is open or closed, we see that the window shade is closed most frequently during the morning hours. And so what can we do about this? Can we influence this behavior and change what our patients are exposed to? And one thing that's interesting is that when we look at patients who are ventilated versus not ventilated, and we look at how our protocols work, one of the biggest changes in turning lights on or off, televisions on or off, or window shades up or down, is that it's our ventilated patients. And so this, I think, has prompted the hypothesis by myself and others, that it's the patients who cannot express their preferences who may be the most at risk for not getting what they want. So they may have light levels, televisions on and so on, and not be able to say, please turn that off. Okay. This is a surgical group that did a bright light. I love this picture so you can understand a little bit of what's happening. So this is a 10,000 lux machine. Light decays across space, so by the time it's reaching the patient, probably on the 1,000 lux range, but that certainly is sufficient. And this was a very small pilot study, but they did show a nice decrease in delirium. After surgery, they did the light for three mornings in a row and followed delirium. Similarly, if you look at just hospitalized patients, so not our ICU patients, and you look, if you gave them bright daytime light, the elderly patients with and without delirium had a little bit of improved sleep. Our cardiology patients had a little bit of improved sleeps. And interestingly, our hospitalized cirrhotic patients did not. And so a little bit of mixed data here. And finally, Dr. Begalbek et al. looked at a group of their ICU patients. And this is the same group that I showed you, those nice horizontal bars earlier where two folks were advanced, a few folks were normal, and most of them were delayed. He split that group into five and six patients each and exposed six of them to bright light. And what he saw is that at baseline, they were both delayed. And then when the usual care group was measured again on day three, they were two hours later. So from day one to day three, they're shifting really from wanting to fall asleep, say normal bedtime was 11 p.m., they had shifted to two or three a.m., and then they shifted further to three or four or five a.m. However, when he exposed folks to bright light between day one and day two, the patients corrected three or four hours towards their normal. So that's a very promising, though, small study. So I'll remind you of this important biologic truth that really informs a lot of what we think about when we think about sleep and circadian influence in the ICU. I certainly have focused more on the circadian piece. I think this big delay in the high variability, high light exposure here with a loss of exposure here is probably what's underpinning that and really informs for us our interventions. With that, I'll summarize that light is a key factor in sleep and circadian disruption. The phase response curve really helps us understand what's going on and helps us design interventions. I think, as I said, that the ICU promotes delay, that the interventions are possible, but our evidence is limited, and we certainly need to be able to measure folks in real time to get that bright light shining on our patients correctly. Thank you so much for your attention, and of course, I'll also thank my mentors and my advisors. Thank you. That was a wonderful talk. I just had two questions. Perhaps you touched on it. I apologize in advance. Oh, my name's Michelle McMoon. I'm from Duke University. I'm an advanced practice PA in the critical ICU. I was wondering how our computer screen lights, what their loose is, because we have all those computers in the room now, and people turn the computer screen to the patient in the night, and second, I was wondering about our sedation medications, where they fall in your scale, and if they're a zeitgeber. Thank you. No, that's wonderful. So as far as, we call them WOWs, because we got in trouble for calling them cows. As far as WOWs are concerned, the data is very similar to the television data, and that is, as I was talking in my other talk, I mentioned in my other talk, that QI piece, looking and looking and looking, and so when we went from one medical, electronic medical record to another, and the WOWs needed to be in the room for the barcoding, that was a piece we had to bring on board, and so we try to turn the screens off. They're very bright, and our cycle, so it's like light, dark, light, dark, light, dark, and so we've tried to address that. The pharmacologic piece, some of the anesthetics directly suppress the melatonin curve, and that can be very important, and then a lot of the opiates and benzos, as many of you here know, drastically alter sleep stages, and that sort of is the next sleep is its own zeitgeber, and so you mess up sleep, you mess up circadian, and that's part of that intertwined problem that I alluded to. Excellent talks. Thanks for filling in on the first one. Michael Connor from Emory. So I struggle, I think, with probably most everybody in the room with the culture of lights being off all the time, and patients being allowed to rest. Had there been any thoughts about engineering solutions where the lights are automatically on, and the television and lights automatically turn off at a certain hour every night in every single room in the ICU? Yeah, so there have been a couple groups who have engineered full rooms. There's groups that have engineered, there's a cocoon bed that a group in Australia has that literally just sort of controls everything. Taking it sort of piece by piece, I'm having a meeting with the Yale fire marshal about the light thing, so there's just logistics there. You have to convince folks to think about architecture, and they don't, and so there are certainly, you can have an app, and you can control the light, and you can do a really nice sunrise, daytime, sunset, and so on. It's just a lot of logistics, I think would be the easiest way, and you've gotta prove it to convince a hospital. The television piece is complicated. I talked about the vent patients. So they don't usually ask for the television to be on, but it turns out that the patients who are able to communicate, vented or not, often about a third to a half of them want the television on, and they sleep with the television at home, and so that's sort of one of those, I don't know what you do. And so it's hard to find something that's one size fits all. I think we can do a show of hands who, when you're on rounds in the ICU, go in and take the curtains up, and you wake the, right? So we've already, it's preaching to the choir, but I think folks are aware of the daytime piece, but we're getting there. Hi, great talk. My name is Terry, I'm from Children's National, circadian sleep researcher myself, and in looking at a lot of the cellular molecular basis for circadian rhythms, at each cellular level, they're all desynchronized, right? And as you go forward in thinking about ways to try to re-sync them, what is your feeling about which one of the Zeitgebers, if any of them, are the best at doing it? Is light the answer, or is the SCN so messed up that it's not actually gonna respond? Is it gonna be feeding, or is it gonna be everything? So I think the answer is everything. I think the question before your question is, so just to step back, so really a question about how are we going to create synchrony, not only earth, so to say, earth to brain, but also brain to body. And so I think the first thing is that many of us are concerned that the brain's on one time zone, the heart's on another, the gut's on another, the immune system's on another. We not only don't have the diagnostics to tell us what each of those are on, we don't even have the diagnostics to rapidly tell us where someone's brain is. So we can do a 24 hour urine curve and measure it, and by the time we've measured it, they've shifted further. So we can guess how far further they've shifted. So we need some real time diagnostics. But then I think it is gonna be a bundle and we're gonna have to approve the bundle. I know folks in the circadian world, some of them literally when they travel, they're like, I'm a central person, I expose myself to light and I pre-delay myself and blah, blah, blah. And someone else is like, oh no, I'm a peripheral person. I use meals and I only orient to meals. And so I think there's believers in both camps and we just don't have the diagnostic precision yet to even know that. But I think it's probably gonna be everything. I think we're gonna have mobility and feeding and sleep and lights and all that together. Thank you. Go ahead. Hi, my name's Erica. I'm a clinical nurse specialist and former night shift nurse. So this is great work, fascinating. My question is around almost for us as clinicians, what are some recommendations about some places where we might need to shift our workflow and expectations around things like labs are done between two and four because we gotta have them when we round before surgery, some of those things that are put in the middle of our patient's sleep cycle because of centering care around providers versus around patients. And what's the balance? Just any commentary on that. So yeah, so I think we've got a lot of disciplines here and it really is, when I dug into making a sleep protocol, well it turned out that there was tremendous workflow issues. So maybe the lab counts on the certain amount of labs getting done before morning shift comes in or the pharmacy counts on, we do tubing changes for the children's hospital here and we do tubing changes for the adult hospital here. So actually I think it's understanding what's causing the problems at your institution. And so the labs are pretty easy. It's just, you know, if someone said the house was on fire, you'd be like, okay, don't get the labs. And so could we get labs at 9 p.m. and trust that they don't change between 9 p.m. and the 7 a.m. surgery time? Could we? A lot of that stuff, like the radiology in the labs, happens in the middle of the night because of tradition, because it used to take six hours to develop the film, and now it takes, you know, 20 minutes. And so, you know, we have providers come on. I do night shift myself, and they'll just say, oh, well, 2 a.m. is when we do nighttime labs. That's just, that's what we do. And I'm like, no, it's not what we do. And so it, to me, it's this constant debunking. On a patient-by-patient basis, I look, and I look, when are there labs? Did somehow their Colace get ordered for the middle of the night? You know, my mom was in the hospital, and they woke her up every midnight and asked her if she was in pain and needed Tylenol. You know, because someone said, well, we can't have patients in pain. And so you just have to keep sort of, and I often look back at the MAR and say, well, what did they get in the middle of the night? When did they get their labs? And try to troubleshoot why it happens. And so kind of at the individual and institutional level. Yeah. Okay. I think we'll take maybe one last question here. Oh, thank you. Peter Yim, Columbia. I'm also a night shift worker. But I guess my question is, a lot of these receptors for circadian treatment are wavelength specific. Has your thoughts on using wavelength filters or during the daytime versus nighttime? So yeah. Again, so the light circadian experts met this year and decided to come up with a different measurement of light, super. And so they really are pushing for this melanoptic equivalence. So really understanding, using spectra, using the wavelength of light, and understanding how important that light is for the circadian system. And so I think we're close to having a unit of measure. But then we need it to be adopted. Unfortunately a lot of the LED lights are blue spectra, so very circadian. And we do need to start to think about that. It may be as simple as filters, films, and so on. Those tend to be fire issues. And that's why me and the fire marshal are like this. But it may be that if we can get some data, we can even get to the point that this is part of the design. And again, it would be nice to be able to control it. So maybe programs and so on. But it is a big problem. I think what we're going to do is just take some questions at the end as well, just to make sure we stay on track. But we will leave a few minutes at the end for additional questions. Thank you all. So our next speaker can go ahead and come up, Dr. Paul Wischmeier, who will talk about intermittent enteral feeding, the ICU's new fad diet. So a few what I'll call alignments of interest, because I would like to believe I'm not in conflict with these groups, including the NIH and various industries, but that we're in alignment to try to improve patient care together. So I'll call these my alignments of interest. I'll also disclose that I am not a circadian rhythm researcher, but I am highly intrigued by intermittent feeding and its effects that it may have in benefiting our patients and how best to do it. So I think ultimately our goal in the ICU, of course, and particularly as we think about how best to bring therapies to our patients, is to create survivors of our patients, right? We should be most worried, I think, or equally as worried about the outcomes after the ICU as the outcomes in the ICU, so that our patients don't survive but end up as victims, which I fear as we're improving things happens all too often. I think the real dream for our patients is, right, to go back to the life that they came to us to restore, so they can walk down the street with the people they loved again or hold their grandchildren again. And ultimately that's the goal that I hope what we can do with the treatments we develop in the future that will allow them to achieve that. So the good news is, of course, and we should all feel good about this, is ICU mortality is decreasing in the hospital. We are saving more lives, less people die in the ICU over the last 10 years. We should applaud ourselves, everyone should feel good about that. But unfortunately are we winning a lot of battles, as I often say in these lectures, and losing the war, ultimately the war on the actual survivor part. And so is it perhaps, if it will advance, can you help me advance this slide, it's stuck. We're losing the war right now. If you escape the presentation, go to the next one, maybe it'll come out of whatever is sticking in it. Just go to the next slide. There you go. Just click on that one, see if we can advance. Awesome. Maybe go to that one. Oh, yeah, it's not playing. Oh, that's scary. Never had that one before. Okay. Do you have a connection I can hook my computer up? I can put it up my computer. Can you? Oh, I have a remote, can you hook it up there? Yeah, yeah, but if you go from my computer, it probably will present. If you want to go ahead, if the people who had questions for Dr. Knauer, if you want to ask those while we're connecting the computer, somebody else had a question, I think, no? Go ahead. I have two questions. Venkatesh Hayagari from UT Southwestern in Dallas. So I work in a neuro ICU, and the way we assess patients is to shine a light in their eyes every hour and, you know, pinch them at various unmentionable places. So how do we try to promote sleep in that environment? That's the first question. Second question is, is there any evidence to suggest that, for example, given that fragmented sleep is going to happen in the ICU, are four blocks of two hours each any better than eight blocks of one hour each? Okay. So how to do this in neuro ICU when you're shining lights, and what's the best sleep we can achieve? Oh, good. I don't have to answer. So I think the neuro, everything you do is you have to decide how important it is. And so is it critically important that you shine that light? At least that is contained. You could pick your favorite flashlight that maybe was more red-toned and so on and so forth. Maybe say, you know what, clinically, I can live with spacing it out to three or four hours in the middle of the night. Or clinically, I can wait until they turn over and call me for something else, maybe not if they're severely brain injured. So I think as with anything, if you broaden the question, you just have to decide clinically what's the most important. And if someone literally needs that, that you need to know it because you need to know if you need to scan their brain, you need to take them to surgery, I think their sleep is going to get disrupted that day. But if you can clinically space it out, then you're going to benefit them. As far as the, is it eight-hour block or two-hour blocks, I think the easy answer is that we could never, ever achieve the eight-hour block. We did a pretty intensive patient-by-patient. I combed through the record. We had a team go in, go through a five-page checklist, and we got about 46 or 47 minutes at a time. So that 27 minutes I said that patients were left alone, I got that to 46 minutes, taking each patient and basically having their own sleep coach comb through their record. So I think what we have to do is try to do the best we can to get even an hour or two blocks and maybe put a few of those together during the night. All right. So can I ask a question now? Go for it. Okay. Good for you. I'm pleased that you're giving a topic about sleep. It's very important. I'm a pulmonary sleep medicine intensivist in Miami. Question number one is simple things. I know we'd love to do studies, but simple things. Can you give an advice on, A, should we be offering earplugs and eye blinders to all our patients in the ICU, number one? Number two, should we be prescribing melatonin at night since it seems like our ICU populace may be melatonin depleted? And I think I'd like to know, should we also redesign our vital signs so that if it's a stable patient, a non-STEMI possibly, but stable, instead of doing hourly vital signs, maybe every four and change our vital signs from Q1 hour to maybe Q4 to five hours? And then also, should we be measuring decibels in our ICUs so that when the decibel light goes on, there'll be a red light so everybody, and I totally agree not to have our WOWs, which we used to call cows, no longer in the room at all. I know you have to barcode, but maybe there's another simpler way of dealing with that. So I'd like your recommendations. Yeah, so I'm going to let him get back to his talk. I agree, we do a lot of silly stuff, and we have a whole talk on melatonin. So stay tuned. Okay, let's hope we're winning the battle now. Okay, so the question is, are we winning many battles and perhaps losing the war in our ICU survival battle that we have fight for our patients? Or is it the survival is really a PUREC victory, unfortunately, for so many of our patients, where although sepsis deaths have fallen by half, we've at the same time tripled the number of people not going home but going to rehab, and we don't know if any of these patients survive to actually ever go home again or have any meaningful quality of life. And so clearly, the emphasis now has been shifted from mortality in our endpoints of trials to quality of life and how we can support that. With the question that I want you to ponder at all the interventions we do and all the activities we do around the patient is, are we creating survivors or victims of our patients? And of course, we know that the long-term outcomes have not been great. These are 30 and 40-year-old ARDS survivors, half of whom won't go back to work for a year and a third of whom will never go back to work because of disability they've suffered. And no one, of course, is immune. This affects children. This affects young people. This affects elderly people after ICU. We know that 60 to 80 percent of people on a ventilator for even five days will have significant, often lifelong, debilities. Why are we losing this quality of life war? Well, of course, there's many reasons. But one reason, and this gets back to why maybe we should feed differently, is a critically ill patient can lose up to a kilogram of lean body mass a day. And the catabolism and the hypermetabolism, which the circadian rhythm affects, can persist for months to years after patients are well. The body doesn't know when to turn it off. A burned child will remain catabolic for two years after the burn injuries healed. So although patients gain weight back, much of this is fat mass. It's not muscle mass as the body remains catabolic and hypermetabolic and not back to its normal rhythm yet. And again, Mother Nature never intended any of our ICU patients to survive any of this, which is why the body doesn't know when to correct itself and return to its normal rhythms and normal metabolism. So this is truly an epidemic for us as ICU caregivers and for our patients, and we need to do better. But how can we do that? And how much of this is related to how we feed patients, or don't feed them, unfortunately, in our care? And we believe some of this, or maybe a significant amount, could be due to that. And how do we actually feed now, and could we do better? Well, we know from the International ICU Survey that Darren Hyland and others do, and thousands of ICUs around the world, and hundreds of, thousands of patients, sorry, around the world, hundreds of ICUs, that we underfeed not for a day or two, but we feed about half of what our patients need for two weeks on average. And the U.S. is always the worst, always the worst. And our protein delivery, the part that actually sustains the muscle to allow physical recovery, we give about a third of what our patients need for up to two weeks, not the first three days when perhaps trophic is a reasonable thought, but for up to two weeks in general is that underfeeding period. And so again, we need to be better than this, and how can we do that? Is there a better way to feed in the ICU? Is there a new fad diet? And what about intermittent or bolus feeding? Can we do better? And can we learn something from athletes in the process? Clearly, continuous feeding is the preferred method. How many of you continuously feed in your ICU? Raise your hand. Almost the whole room. How many of you intermittent or bolus feed? About half my patients are intermittently bolus fed. Before I ever got to Duke, actually, our trauma guys and our NT guys have always done it, so a fair number of people in the room, actually, more than I expected. So there may be physiologic benefits, and we'll talk about that, but the studies are limited. And so these are, if you want to take a picture of something to read about this further, these are excellent review articles that drove a lot of the content I'm going to talk to you about. These have summarized all the literature that exists to date on these topics, and so are really good resources if you want to learn more about this. Danny Bear and Zuden are doing a neat study I'm going to tell you about later. We don't have all the results yet, but we have some of them. I'm going to use some nomenclature. I'm going to show it to you. This is what continuous feeding looks like. It's around the clock. There's three kinds of discontinuous feeding. There's cyclic feeding. Usually it's nighttime feeding, for better or for worse. There's intermittent feeding, where you give Ian over 20 to 60 minutes every four to six hours with or without a pump. And then there's bolus feeding, where you actually push it in or drip it in with a syringe over a very short period of time. This is another way to look at it. You can see the bolus feeding, short intermittent boluses, and intermittent feedings over a little bit longer period of time. You're going to see this table again. That's why I'm showing it to you. These are the, according to Danny and the work she did, Danny's an RD PhD in the UK, for those of you who don't know the RD nutrition world, NICU. There's some benefits to continuous feeding, and those include some of the things like blood sugar and GRV management. And then we think there's benefits to intermittent feeding, particularly protein synthesis. And we're going to talk a lot about that. Ability to catch up on misfeeds. Perhaps more freedom to do rehab, which is essential, and I know we're going to talk a little bit more of that in the next talk. This is a paper by Paul Merrick. Those of you who don't know Paul, he did the vitamin C steroid thiamine trial. I'm sure you know about that. He published a few years ago, and again, coming back to this muscle issue as possibly being important to why continuous feeding is less than ideal. And so this is our goal, right? I mean, it's reasonable, we think, and this is my take on how feeding should be done in the ICU. It's reasonable to think we perhaps need a little bit less calories in the first few days of ICU, but we need to get to protein goals more quickly, especially in malnourished patients perhaps, where they won't tolerate trophic feeding as well. But it needs to be the complete package, and you saw the protein went up faster, and that's the essential nutrient we think may be so essential to improving long-term outcomes in the ICU. Can't build a house without bricks, and protein is fundamental to preserving muscle mass, preserving immune function, and a lot of other key things we need to recover. These are new recommendations that have come out in the literature for how we feed people, especially elderly people. They're anabolically resistant, which we'll talk about, but up to 2 grams per kilo probably are needed to have any meaningful muscle recovery in an elderly ICU patient, and we'll talk more about why that is. So again, you can see, even early on, there isn't much need for increased calories until about day 20, and that only goes up about 1.8-fold, the metabolic need for non-protein calories. But the loss of protein and the protein needs go up in 24 hours as much as 4-fold, and so this is really the key time, and protein being delivered is a key goal we should have. And that's because, of course, we all have the same response to stress, whether the car hits us, the surgeon hits us, or the tiger bites us. We break down muscle to generate energy to run away from the tiger or the surgeon, and ultimately—I'm an anesthesiologist, I can say that—ultimately, early protein delivery may be essential to do this, and the body can generate its own non-protein calories if you're reasonably well-nourished, at least for a few days it can. This is where it gets really unfortunate for our elderly folks. So if I was to lay one of you under the age of 40 in the room down in a bed for 28 days, you wouldn't be able to walk afterwards. NASA did those experiments. If I laid an elderly person down in a bed for 10 days, they'll lose three times as much muscle mass as a young person would. If you lay a sick elderly person down, they'll lose that in three days. So the muscle wasting that I described before is catastrophic as we age. And why is that? And this is an outstanding paper. If you want a really great paper by Stuart Phillips, one of the major authors, he's one of the true protein and muscle growth experts in the whole world, and really a great guy. Young people can build more muscle per protein gram delivered than elderly people. It's called anabolic resistance. As we age, we need more protein to gain muscle. So in our everyday lives, this is important too. You don't have to lose muscle as you age, as the data might say. This guy's 70 and he looks pretty good. He's doing a little more than just eating right to look like that, but maybe our patients should too. But there's a lot of things we can learn from athletes like him. This is how they eat. 0.4 grams per kilo per day and a bolus of protein every four to six hours. And there's data for this. And they evenly distribute their meals and they take protein in before sleep. We're going to talk about all those things and what effects they may play. And then, of course, exercise is essential too. So of course, no athlete would ever do continuous feeding. There are not bodybuilders running around with tubes in their noses feeding themselves all night long. That doesn't exist. It doesn't work. And I'm going to show you why it doesn't work and why we should think differently about how we feed. And hopefully some of you will go away from continuous feeding. And maybe I'll show you some data to help that too. Bolus protein and feeding is much more likely to optimize muscle gain. And there's a study going on. Zuden Puducherry and Danny Baer have got results. Chest is almost accepted, we hear. I'm going to give you some of the safety data from that study that was presented at ESICM, but I don't have the muscle results. But it's essential in critically ill patients we pay attention to this because they have the most anabolic resistance. So they gain the least amount of muscle per protein delivered. And they probably need help from some other agents and supplements, much like an athlete would. So data for intermittent feeding. What does it do to the muscle and how does circadian rhythm affect that? So muscle protein synthesis peaks 45 minutes after a bolus of protein. This is basic physiology data, not in ICU patients. This is what all of you and I do as humans. You can take in about 50 grams of whey protein and in 90 minutes the muscle is full and it returns to baseline, your muscle protein synthesis does. It doesn't matter if you keep taking it in. It takes time to recover this synthesis ability. And this is where the muscle full effect, which is a catchy term now in ICU nutrition, comes from. This is from Danny's paper. And it's the idea that muscle is protein full after a certain period of time and you can't keep gaining if you keep feeding continuously. If you intermittently deliver protein and nutrition, you can maximize your protein synthesis by allowing this to recover. So it's the baseline to peak repeated concentrations of amino acids that benefit the muscle the most. And so this is one of those studies that supports that. These are elderly adults who are given boluses of protein once a day or more than once a day with their meals. The people that took in multiple high-protein meals gained much more muscle mass and strength. And these are not just young people. These are elderly people. And so 30 to 45 grams, that's a reasonable dose, that's a reasonable-sized protein shake for those of you who think about it, produced the greatest increase and this is likely how we should feed through the day. This is probably what we should do for ICU patients too. This has not been studied yet. Now the flip side of this is continuous feeding may suppress protein breakdown. So there may be a role for bolusing protein in the midst of trophic feeding, perhaps to protect the gut, protect the microbiome, and do other things. So there's a potential for a combination of effects to be had here. But clearly protein synthesis is important with bolus feeding. So could timing of protein or EN boluses around the clock or at certain times in the clock be the best? Should all of our patients perhaps get a bolus of protein, perhaps not the rest of the calories, before they go to bed? And maybe all of you should too. There's data that shows in well people it increases muscle synthesis 20% if you take branch genes in at night. That's in normal people. So there could be. For one is, your growth hormone, your anabolic hormone that's most important to your ability to gain muscle, peaks right at the first half of sleep. And so if you're going to maximally utilize an anabolic substrate like a bodybuilder would, you would take it in then. And all bodybuilders take in protein or amino acids right before they go to bed, usually branch genes. I do too, I'll hint. This is the circadian rhythm chart that many of you have probably seen. And all of metabolism is affected by the time of day that is occurring for our patients or for you as individuals. Anabolic hormones are largely secreted overnight. Growth hormone in the evening right after bed. Testosterone early in the morning between 6 a.m. and 9 a.m. in most people. And clearly there's many effects. And I won't go into them. You can look at this or take a picture of it. But this is a great paper if you want to read about it. It's secreted through the day. And should we target boluses to evening and morning anabolic peaks? Perhaps. The other question is, if we could normalize sleep, we would help our patients as well. Clearly deranged sleep patterns, being jet lagged, working at night, derange all kinds of things. Insulin resistance, muscle gain, and many other things. I worked on this talk the other night on call for my second night in a row on call and realized that being up all night, I'd never read the circadian rhythm, being up all night increases your risk of hypertension, early death, all kinds of diseases, and that we should all stop taking colic. Unfortunately, the patients don't go home, though. But I did learn it was really bad for you. So this is actually real interesting. And actually, there is a best time to work out and a best time to consume protein for muscle synthesis. So although testosterone peaks in the morning, so does cortisol. So you actually should not work out in the morning. You should work out in the evening when your growth hormone is about to peak. And this is when maximal physical ability occurs. Perhaps we should be giving protein boluses to our patients around this timing, too. It's a hypothesis, right? And so because there's a lot of things that clearly are affected by the time of day that we take interventions upon our patients. That's the best time to work out, but getting high testosterone secretions in the morning as well. So these are things we might be able to take into consideration. Why is normal sleep so vital in the ICU? Perhaps it has a muscle effect in addition to all the other things you've heard today. If you sleep-deprive someone or alter their sleep pattern, you ablate their growth hormone spike. So you take away a major metabolic stimulus, a major anabolic stimulus, from your patient. So loss of sleep pattern disrupts growth hormone. It disrupts all the clocks. Every organ I learned has a clock. I didn't know that. And it's massively disrupted by sleep disruptions. So metabolism is also massively disrupted. This is an excellent paper. It's a long paper, but it's a great one. It has a lot of good data in it. So are there trials of this? There's a few. I'll try to cover some of the key ones briefly. This was a comparison of the metabolic effects of continuous feeding versus cyclic feeding. And cyclic, in this case, means at night. So plus-minus on what eating at night does to you and to your sleep. But that's with this study. Now, also remember, this is not a critical care population. It's a head and neck surgery population, but it's an interesting paper. And what they found was that feeding just for 12 hours a day at night reduced hypermetabolism and reduced the negative effects, perhaps, of feeding continuously, which the body is not evolutionarily meant to do. So this is quite interesting. Perhaps you can reduce some of this persistent hypermetabolism. The other reason, it reduces catecholamine release. That's what causes the hypermetabolism. That's why we give propranolol to all the burn patients. Propranolol is one of the only things that will make a burn child anabolic, because it reduces the hypermetabolism of the persistent catechol release. Feeding at night versus continuous feeding does the same. And so there's perhaps benefits to cycling feeding. This is a paper looked at 50 patients who had PEG placements in the ICU. Now, that means they probably weren't acutely ill. They probably were late in their ICU stay, perhaps, when how you feed someone may not matter as much. But maybe it does. And David Evans, a name known to us in the ICU nutrition world, he's a trauma surgeon, did this. No difference in glycemic variability, so at least bolus feeding doesn't disturb glycemic variability. No difference in insulin utilization, so it doesn't lead to needing more insulin, getting bolus feeds versus continuous feeds. No difference in getting to goal 2 feeds, although I would say as late as these patients were studied, probably getting to goal 2 feeds in a PEG patient isn't that hard. And no other real differences in other nutritional or glycemic outcomes. Again, these are later, usually, patients in the ICU, from what I could gather. So maybe not the best acute care patients. This is an early paper in trauma patients, early feeding in trauma critically ill patients. Using bolus feeding or intermittent feeding leads to reaching goal earlier. So there is some data to support the idea in your head that you could do make-up feeds or get to goal sooner and avoid some of the interruptions. So these are different paper, review papers, benefits and risks of continuous feeding. So bolus central feeding seems to have more physiologic effects and may have some advantage for rehab. It may increase your risk of aspiration if you gave too much, perhaps, although I'll show you some data that says it's probably not true, but it also affects which feed you can use. You can't just use any high-fat tube feed as a bolus feed. This is Danny's. You can see some of these things we talked about, protein synthesis being one of them. Guidelines, at least the Canadian ones in 2013, didn't give us much guidance. There wasn't much data then. The European ESPN guidelines from last year do. They studied this. They meta-analyzed the data that exists. No difference in GRVs in continuous versus intermittent feeding. So perhaps the aspiration risk is not as large as we thought. The GRV volumes are not a problem. There was a bit more diarrhea in the continuously fed patients. And I would say the two feeds you choose probably makes a difference here, and we can talk about that more if you'd like. But I think that has something to do with this, but it's important to know this is the data that exists, although limited. This is the trial I told you about. So this is a trial that was meant to study this in a meaningful ICU population in the UK with muscle data. I don't have the muscle data yet. It's going to be published soon. It was a critically ill population of patients on the ventilator and acutely ill early in their stay with four-hourly intermittent feeds for 10 days versus controlled continuous feeding. And they looked at outcomes. They also looked at muscle, which these slides are from Zuden himself. He sent them to me a few days ago. This was presented at ESICM. So they have the outcomes data, not the muscle data. And there was no difference in pulmonary aspiration. There was no difference in serious adverse events. There was a little bit of glycemic variability changes in a few patients, which is reflected in the next data I'll show you, but that was about it. The only thing was there was some glucose variability changes in the bolus-feeding patients. And so that would be something you'd have to watch for if you were to do this. What he did show, or what Zuden and Danny did show, the two of them, was that this is safe. Gastric intolerance is the same. Same amount of insulin use. And this may be a tool for preventing muscle loss, and the muscle data will follow the papers and final review at CHEST. So hopefully we'll have it soon. So this is one strategist from the Merrick paper. This is how he does his bolus feeding in his institution. So this is one strategy you could employ. I'm going to propose another to you that would go based on the physiology data we know about muscle. And that would be to bolus proteins early in the morning to take advantage of the testosterone peak. After physical therapy, every athlete knows the best time to take in protein and the muscles most likely to use it is after exercise. We do this in all of our surgical prehab studies. We give them protein and HMB after that. And then before they go to bed, perhaps not with any other calories, just the protein itself. There's some pretty good data for that in healthy elderly adults. I would add these things as well. I think all of these things are vital. I'm not going to go into great detail. Copper and carnitine are deficient in 90-ish percent of your CVVH patients after seven days if you are not checking copper especially, but carnitine too in your ICUs. After five to seven days in CVVH, you're having deficient patients. Long-term copper deficiency leads to permanent neuromuscular weakness and leads to pancytopenia and immune dysfunction. But the muscular weakness is irreversible if it's let go too long. You should be checking for that. Vitamin D, muscle can't form normally without vitamin D being sufficient. So every bodybuilder knows that, but we're just learning that in the ICU. So why am I confident we can do better? And why do I hope you are confident too? And why do I care so much about nutrition as an anesthesiologist intensivist? And that's what I'm going to finish with and tell you a little bit more about what I do with nutrition. Well, Derek Angus has shown us that all of us will average 1.7 ICU admissions over our lifetime, so don't use them up too fast. So someday this will be you or someone you love. And for me, I've used up a few of your lifetimes of ICU admissions because this was me at 15 on TPN. So nutrition and dietitians saved my life. And I've been in the ICU many times. I've had 22 surgeries and live with 160 centimeters of small bowel. So nutrition's a big part of my life. And in fact, I was in the best shape of my life a few years ago. This is me with my son at my grandfather's house until I ate some quinoa. Don't ever eat quinoa. It's the best advice you're going to get because it causes bowel obstructions. At least if you've had surgery, it does. And that did. And my wife is sitting in the front row. We flew all the way back from Baltimore to Denver through Houston because we want to control our care as physicians. We're terrible patients. And I showed up in my own emergency room, seeing my own residents looking like this with a lactate of 10 and bowel edema with a severe bowel obstruction. Went to my own OR with a surgeon who, first the surgeon that came was an intern I had trained, and she was a new attending. And I said, no, not you. And then bring someone older. And she did. But they spent eight hours cleaning quinoa out of my gut. I ended up in my own SICU with an open abdomen myself. And was hospitalized 23 days on parenteral nutrition, continuous feeding, right? Went from looking like this to looking like this in 17 days. Lost 20 kilos of lean body mass in 17 days. No one's immune. This happens to everyone. Couldn't walk down the block without being short of breath. Couldn't pick up my own child five months later at Christmas. He wasn't that big. This is the weakness that happens. So I see recovery and prehabilitation and nutrition delivery as an everyday part of my life. I like to show this picture because a lot of my patients who get ostomies feel like their life's over when they have an ostomy. I was surfing a few minutes before this picture was taken on the North Shore of Hawaii. And nothing you can't do with an ostomy that you could do before. But I take in exactly what the bolus protein requirements I'm suggesting, I take that in every day. And it's interesting because Tim Buckman came to me a few years ago, and he said, I have a friend having surgery, Tim Buckman, the editor-in-chief of our journal. He said, how do you get better so quickly? I see you at SCCM, and I know you've just had surgery. I know you've just been in the ICU. But you gain it all back so quickly, and you seem like nothing happened. And I said, so, Tim, this is what I do. I put a list together. I said, you have to bring a separate bag to the meetings to carry all this, which I do. But these are all the things I do. I've been using intermittent bolus protein for years. I learned it from the people who do it best, the bodybuilders and muscle physiology experts. This is what I would advocate our patients need. I know this gets people better. I know you can recover from the ICU with this. So we have opportunities to do better for our patients, and I think there's great promise for the physiology of what this can do for our patients on many fronts. But we need more research. We have very limited data. What's fascinating is the things that an elite athlete would do to recover and to function at a high level, we don't do almost any of for our patients who invest millions of dollars and many hours of our lives in. And there's some basic things we could all do for them. This is a personal passion for me, because as I tell my residents, I'll be in the ICU again. And that 2 a.m. phone call they get with the old guy with the bad abdomen, that might be me. But someday, that might be you or someone you love. And so I think it behooves us all for our patients and for ourselves to be better at this. And so in the end, I hope we can learn from athletes to help us eat better, eat right, and do it right, and take patients who look like this, have them walking like this within a day or two, have them using machines and beds that do this. Anyone have this bed? I don't have it either. But they do sell them, and I can't wait to get one, the beds that turn into leg presses. And then ultimately like this, or like this. Or at least hopefully like this. And with that, I look forward to your questions. Thank you. I think to keep things moving along, what we might do is just reserve questions for all the rest of our speakers at the end. So please save your questions, since we will have some time for that. Oh, just kidding. We will take a few questions while we fix the technology. Thank you for the audience's flexibility. Can you comment on how to connect some of the previous lectures with the nighttime wake cycles, and how you could change feeding? Because if we're trying to do the bolus feeds, what do you think, based on what you saw earlier, if you could hypothesize about when would be the appropriate time to put someone on intermittent feeding? I think we've always wondered if we started feeds intermittently, and the last protein dose perhaps was given before bedtime, but not too late, like 8 or 9 o'clock. And then we didn't feed overnight. I think we can all learn from some of the slides we saw today that perhaps there's adverse things about feeding at night when we're not supposed to be fed. Perhaps we could restore sleep-wake cycle, and the next boluses could all occur during the day, like we've evolved as humans to eat during the day, where we'd restart the boluses in the morning. And that there could be a period overnight where the bolus feeds weren't given, and all the caloric delivery could occur between wakefulness and bedtime, and we wouldn't need the continuous feeding overnight, which probably isn't ideal for restoring. And you guys down here who do this for a living would know better than I, but I think it would make a lot more sense. Unfortunately, there's a lot of favoritism towards feeding at night continuously, right, or intermittent feeding at night, because people say, oh, then they'll eat more during the day. That data's pretty soft. People have looked at that in small studies. The data's not really that strong for that, so I don't know how much advantage we're gaining. Again, cycling overnight, there isn't a lot of data for that, improving appetite during the day either. But who knows? I think that study needs to be done. Thank you. Thank you. I think we'll keep moving with our theme. So next we have ICU mobility, workout today, sleep better tonight, Dr. Lauren Front. All right. Thanks, everyone. I will try to keep my voice up so you can hear me over the next room. So I'm Lauren Fronte. I'm an assistant professor of pulmonary and critical care medicine at Yale. Despite the name of our section, I would like to disclose that I am not a sleep medicine physician. My expertise is in function, long-term outcomes, and mobility. But I'm really excited to talk about how we can unify these two themes. I have no conflicts of interest to disclose, and I'm currently funded by a Beeson Award from the NIH. So my objectives for today are first to briefly understand the relationship between exercise and circadian clocks, and then to talk about a couple of questions that I think are applicable when we're thinking about how early mobilization programs in the ICU can tie in with sleep, including whether exercise has been shown to improve insomnia and whether the timing of exercise matters. And then finally, to talk about how this knowledge is applicable to mobilization and future directions for research. So first, I just want to take a step back and just briefly touch on what is exercise. So it is defined as activity requiring physical effort carried out to sustain or improve health and fitness. And generally, we have four categories of exercise. We have cardiovascular or aerobic exercise, which I think is what most of us think of, strengthening exercise, flexibility, and balance and agility. And most of our early mobilization programs in the ICU, as I will show you, are programs of progressive active mobility that, when our patients are executing them, tend to fit into the more moderate aerobic category, although we are also working on balance and agility. And so as you heard earlier from Dr. Knauert's talks, basically our circadian clock or the mammalian circadian clock includes photic entrainment and non-photic entrainment signals. And skeletal muscle has an extensive network of clock-controlled genes. And exercise is a robust zeitgeber of skeletal muscle clocks. And so while light directly entrains the SCN, feeding, exercise, and stress can reset peripheral clocks. And so this nice review showed the possible effects of aerobic physical activity on sleep. And I included this because, again, generally for early mobilization, we're in the moderate intensive aerobic physical activity category. And this review nicely outlines pathways that have been shown or are hypothesized to affect sleep time and slow wave sleep. And so I want to just revisit this circadian clock because I wanted to take a step back and ask. I run our early mobilization program at Yale. And I have been asked before, is the timing of exercise important? And I think if we're building and maintaining a program, this is something we have to think about. And so when we look at this circadian clock, as we've heard a couple of times, early in the morning we have elevated testosterone and a peak in alertness. And it's around 5 o'clock that we have our greatest increases in cardiovascular and skeletal muscle strength. But also I'm sure for those of you in the room who exercise, you've probably heard that maybe we shouldn't be exercising before bed. And so this was something I wanted to look into for this talk. And so when you look at the data studying the timing of exercise and sleep, there were two meta-analyses from the late 90s that evaluated sleep onset latency and decreasing wake after sleep onset. So there was one that showed that there was a positive influence of exercise when you, and that's supposed to say sleep, with sleep four to eight hours before bedtime. But a negative influence when exercise occurred greater than eight hours or less than four hours before bed. However, the authors point out that the evening exercise findings may have been biased because they were performed in sedentary women or after an extreme race, which doesn't really apply to most of us or our patients. And when you look at more recent data, such as this 2014 Sleep Medicine article, more recent studies suggest that exercise before bed may not disturb sleep. So I found this reassuring. In our ICU, actually, our physical therapists and occupational therapists generally work with patients between 7 a.m. and 5 p.m. So they aren't mobilizing patients right before bed anyway. But if we have patients who want to get up and move around before bed, we certainly don't need to discourage them from doing so. So the next question I wanted to answer that ties mobility to sleep is, does exercise improve insomnia? So in this 2010 article, the investigators studied 48 chronic insomnia patients assigned to moderate-intensity aerobic exercise, high-intensity aerobic exercise, moderate-intensity resistance exercise, or a control group. And the only significant findings was actually in the moderate-intensity aerobic exercise group, and that group had a 55% reduction in sleep onset latency, an 18% increase in total sleep time, and 13% increased sleep efficiency via polysomnography. So again, I had mentioned that our early mobilization programs probably fit best into this category among the four that were studied here. So I find this reassuring that maybe our early mobilization programs might actually help patients with insomnia in our ICU. But again, this has not been studied in the CRITIC-LEO. There was a follow-up study of a six-month moderate-intensity program on insomnia that showed the same improved outcomes and also showed that morning or evening activity did not matter in terms of its effect on sleep. And I really wanted to find one larger study, but this was the best one I could find. There were about 2,600 patients. However, the intervention was a lifestyle modification program that included both exercise and stress reduction. This did show improvements in insomnia through a questionnaire. It's just a little bit hard to disentangle the exercise piece from the stress reduction program, I think. Okay, so now coming back to the ICU. So this is really the other extreme. This isn't the outpatient setting where we're looking for the best timing of activity. And as Melissa showed earlier, we have patients who are connected to lines, tubes, vents. And I showed this the other day in my mobility talk because I just found it striking that in this editorial from 1998, the author describes how when critical care first started in 1964, patients were awake, alert, interacting with families, mobilizing around the intensive care unit. And then the pendulum swung completely in the other direction where we had patients sedated, intubated, and just snowed for the whole day until they're better. And now we're all trying to shift the pendulum back the other way. So what is mobilization in the ICU? So I just wanted to bring up two more definitions so we can think about how they tie to the exercise definition from earlier. So rehabilitation is a set of interventions designed to optimize functioning and reduce disability. And mobilization is a type of intervention within rehabilitation that facilitates the movement of patients and expends energy with a goal of improving patient outcomes. And we know that early mobilization programs in the ICU are mostly programs of progressive active mobility. We generally don't include passive range of motion in these programs, although passive range of motion is, of course, encouraged. Patients' families can help with this or nursing. But nursing, physical, and occupational therapy are generally participants and helping to execute programs of progressive active mobility. In our ICU, I've included the steps that we have here in our program. So we start with therapeutic in-bed exercise and progress the patient to supine to sit and dangle. Then we progress to sit to stand. Transfer training, which is bed to chair or about five feet of ambulation. And then gait training with increasing levels of patient independence. You might ask, what if the patient isn't awake and alert enough to participate in therapeutic in-bed exercises? So this is an area that has been studied as well. There are several studies, including Michelle Koh, who has her cycle RCT, where they're studying in-bed cycling. This can be passive or active for patients who are not yet ready to participate in this first step of progressive active mobility. And we know that these programs have a lot of data supporting their efficacy and improving patient outcomes. While some outcomes have been mixed, such as hospital length of stay, other outcomes have been pretty consistent, such as improved muscle strength at discharge, increased participation in activities over the six months after the ICU. And also, so there have actually been improvements in six-month outcomes as well as in-hospital outcomes. And so do we have data showing the association of early mobilization with improved sleep outcomes? So we actually don't yet, although Biran Kamdar and Dale Needham have studied the opposite question, which is an evaluation of the association of sleep quality, delirium, and sedation with participation in physical therapy. So in this study, perceived sleep quality was not associated with participation in PT the next day, but delirium and sedatives were. And when we think about this question, flip-flopping it to the other way, delirium is an important thing to think about because we actually know that early mobilization decreases delirium incidence in the ICU. So since there wasn't actually a study evaluating the association of mobility with sleep outcomes and circadian disruption, I thought that I would propose one as part of this talk. So what I would love to see is a study where we have a progressive active mobility program and then we study delirium outcomes and circadian disruption, and we want to know whether progressive mobilization is associated with circadian or I guess less circadian disruption and improved sleep outcomes and whether it's being mediated by delirium. And of course we wouldn't want to just know about PSG, polysomnographic outcomes and circadian disruption. We would also want to know whether the patients have improved sleep quality in a qualitative way, functional outcomes, cognitive outcomes, and mental health outcomes because I think it's pretty easy to see how all of this can tie in together. We just need the studies evaluating these outcomes now. And then I think once these first sets of studies are done, what I'd like to see is some follow-up studies. When passive cycling is the exposure, do we still see the same improvements? And also does the time of day of mobilization improve these outcomes among patients who are already delirious? So anyway, Melissa, if there's an X like 10 years, do you want it? Okay, great. So for now, based on what we know, I would suggest to ensure that your ICU has a program of progressive mobility because we know that this does improve outcomes for patients. We just don't know yet about sleep outcomes, but we can hypothesize that it probably does help. And second, to plan on progressive active mobility for patients during the day, but they could certainly be out of bed in the chair until ready for bed. And if they want to mobilize before bed, there are no data suggesting that they shouldn't. So for take-home points, first, we've learned that skeletal muscle has an extensive network of clock-controlled genes and exercises a robust zeitgeber of skeletal muscle clocks. Early mobilization programs have been shown to improve outcomes, so again we have more work to do in this area in terms of sleep outcomes. And based on these existing data, we recommend implementing a program of progressive daytime mobility for our ICU patients. So with that, I'll say thank you, and I'll take any questions. Thank you. Thank you very much. Great talk. Joe Tona from the University of Utah. So I have a question about your study design because this is something that I think about a lot. I've been fortunate enough to work with a very aggressive physical therapy group of colleagues in the ICU, and so we've done a number of interventions to get these patients up and moving. And one of the issues that we came up with was that we did a study design that was a stepwise intervention implementation because I didn't feel that we could ethically randomize to yes versus no mobility. And the reason why I bring that up is that with the study design that I'm interested to know your thoughts on is there seems to be a time-dependent exposure risk for the development of delirium within critically ill patients within an ICU and also probably for circadian disruption. And so I wonder how do we set that study up because I love the idea. I'd love to do this, but how do we set that up because you're going to see this evolution within patients over time in the ICU. So, I mean, is it a one patient versus another patient randomization or is it something that we control for within patients? I don't know. I just know. What's your outcome? What you're saying if the outcome is a sleep outcome? Well. What's your primary outcome? Sorry. If you want to look at the relationship, can you flip back to the slide? Oh, sure. So you're talking about my proposed study? Yeah, exactly. Okay. You're not talking about mobility and outcomes overall? No. No, no. Your proposed study. So you're talking about progressive mobilization, which is going to progress within patients within a period of time. And there's a time dependent risk of delirium and circadian disruption. How you're going to statistically tease out that relationship? Yeah. You're going to have to capture. So the first thing I'll say is I would not necessarily. I think you're going to have to start with an association study, right? So you're going to, because as you said, ethically, you can't really mobilize patients to mobility or not because there isn't really clinical equipoise there. We know that it improves outcomes for patients. So one thing that we've done in our mobility research, not the QI program that I mentioned, is we've captured mobility as a time varying covariate where we basically have the length of stay as the denominator. And then we have the increased levels over the denominator of the length of stay. And then we have that for every day. And so you're going to have to do something complex in terms of operationalizing your exposure. And then also capture delirium for each day. So you're going to want days of delirium. And then probably looking at the associations of the time varying mobility variable with the sleep outcomes. And then doing a mediation analysis to see if it's being mediated through delirium, I think is probably your best bet. Thank you, Dr. Ferrante. Thank you. Okay. And last, but certainly not least, Dr. Mojde Hebner will be talking about melatonin use in critically ill patients, dabbling with the hormone of darkness. All right. Who wants to talk about medications? Yeah. I'm really excited about this. I hope you are too. And I'm glad we got through our technical difficulties. So as the pharmacist in the group, I'm going to be specifically talking about melatonin and the melatonin agonist, remeltion. And I know there was a question about that earlier. So this is perfect. And I have nothing to disclose. So here we are just miles away from Disney World. So who better to help us understand the neurotransmitters involved in sleep than Snow White and the Seven Dwarves? So here we have Dopey. He's the youngest of the Seven Dwarves. And he's kind of just silly. He just wants to stay up all night. He's kind of like my seven-year-old, and just play and not go to sleep. So the neurotransmitters that we would associate with him would be those that are really promoting wakefulness and not necessarily promoting sleep. However, of course, if we were to antagonize these pathways, that's where we would really promote sleepiness. And that, of course, brings us to Sleepy. He is the most mellow of the Seven Dwarves. And he's often seen kind of falling asleep at the most inopportune times, sometimes even sleepwalking in the midst of all the chaos around him. And so, of course, I'm going to associate those neurotransmitters that are going to promote sleepiness with Sleepy. And of course, GABA has been a primary pharmacologic target for a lot of the medications that we use for management of insomnia. And that brings us to Melatonin, the focus of this. And I think melatonin is a rational target when we're talking about, of course, circadian rhythms in our ICU patients and sleep in those patients. Before I get into the meat of the talk, I really want to kind of review some of the basic concepts that have already been touched on, but with regards more so to the melatonin aspect. So of course, during the day, the sun activates melanopsin in the retina. And that causes pupillary constriction and promotes wakefulness and alertness. And that also suppresses release of melatonin from the pineal gland. And then as nighttime is upon us here in Orlando, that change is also sensed by the retina. And that causes, actually, release of melatonin from the pineal gland through activation of postganglionic serotonin. So that's release of melatonin. And then synthesis of melatonin is also very highly related to the circadian rhythm and, of course, very important when we're talking about sleep in our ICU patients. And it all starts with tryptophan. I always think about that turkey on Thanksgiving. But actually, the most important step is the conversion of serotonin to N-acetyl serotonin via serotonin N-acetyltransferase. And it is notable that that activity is increased 100-fold at night. So we talked about release of melatonin. We talked about synthesis of melatonin under normal circumstances. So what about the very abnormal circumstances that we have in our patients? And to orient you to this figure, we have time on the x-axis. On the y-axis, we have the concentration of the urinary melatonin metabolite as expressed by nanograms over four hours. And you can see by the healthy patients, as represented by the green line with the triangles, that their excretion starts off around 3,000 nanograms in the morning, drops off by about 10 a.m., and then stays low until about 10 p.m., where you see that rapid rise in the excretion of melatonin up to a peak of somewhere around 5,000 nanograms. And then compare that to our non-septic ICU patients, who are represented by the red line with the squares, and you can see that their excretion starts off a little bit higher than our healthy patients, or our healthy volunteers. Drops off by around the same time, stays low appropriately, and goes up around the same time as well. But you can see that that peak that's obtained in those non-septic ICU patients is pretty significantly lower than our healthy patients or healthy volunteers. However, the most profound abnormality that we see is in our septic ICU patients, and they're represented by the blue line with the diamonds. And you can see very clearly that their levels start off, you know, around where our non-septic ICU patients are, but they stay persistently high. So essentially they have no variation in their excretion of melatonin over the 24-hour period, and presumably, similarly, their melatonin serum levels as well. So what happens after recovery from sepsis? Well, again, you can see our blue line, which is our patients who are undergoing sepsis, kind of that persistently high melatonin level or excretion over the 24-hour cycle. And then after recovery, a subset of those patients who actually go on to survive, they're represented by that red line with the squares. And you can see, if you remember that green line from the previous slide, this actually looks very similar. So essentially what this is telling us is that after recovery from sepsis, our patients do recover their normal excretion of melatonin and presumably their cycles of melatonin in the serum as well. So I know that was a lot. So to kind of summarize the types of abnormalities that we're seeing in our patients with regards to melatonin, and I know that was touched on in the question earlier as well, some of these patients are just missing that peak melatonin that we would see at nighttime. Some of them are having that abnormally high daytime level, essentially that high level throughout the entire course of the 24 hours. Or they can have insufficient nighttime levels. And so all of these things are going to blur the distinction between day and night and therefore cause sleep disruption. So what causes this? Well, the simple answer is we don't really know. But we think it's likely multifactorial and, of course, heterogeneous. And at least part of this is related to the underlying disease states in our patients. And so some of these patients may have pineal gland dysfunction, just like we see organ dysfunction in our patients who have poor perfusion of their organs. Conversely, they can have compensation. So as we saw with our septic patients, they can have abnormally high melatonin for some unknown compensatory reason. And then, of course, just like with anything we do in ICU patients, a lot of this can be caused by us. So either through abnormally high overnight light levels, through the cows or the wows or whatever, or the medications that we administer to these patients that can also have an effect. And I know that was another question as well. So of the medications that would increase melatonin levels, opioids are notorious for actually increasing melatonin. However, this becomes problematic because, of course, we're often administering opioids throughout all times of the day, at night, so really throwing off the normal variation of melatonin. And then, of course, it goes without saying that administration of exogenous melatonin is going to increase patients' melatonin levels. And typically, that's not going to be too much of an issue because we're not administering those medications during the day necessarily. However, this is where it gets really interesting because the list of medications and the classes of medications that have been shown to decrease melatonin levels is quite extensive. And you can see, taking a look at that, that a lot of these medications are things that we give quite frequently to ICU patients and a lot of the time don't necessarily have an alternative agent. So I know you're thinking, well, what's a recommendation? I'm not sure. But at least we are aware that that interaction is there and that that could potentially have an impact on our patients' melatonin levels and potentially, therefore, an impact on sleep as well. So I'm going to spend the rest of the time really focusing on melatonin and remeltyon. But before I get into some of the data behind these drugs, I want to talk about some of the more practical considerations and kind of things to keep in mind with these drugs. So as you all know, melatonin is considered a dietary supplement. And so it doesn't carry the FDA-approved indication for management of chronic insomnia that remeltyon has. Melatonin targets MT1, 2, and 3 receptors, whereas remeltyon is specific for MT1 and then secondarily for MT2. We are not entirely sure what the clinical implication of this difference would be, however. And then, of course, as a pharmacist, I'm thinking, you know, financial impact. And there is a relative cost difference. Melatonin costs pennies, whereas remeltyon is a little bit more expensive. And of course, that dollar amount doesn't look quite as exorbitant as a lot of the medications that we're giving in ICU patients. But when you think about if a large percentage of hospitalized patients are prescribed the drug, that that can have pretty significant financial implications on health systems. So of course, you know, in an ideal situation, we would remove all the iatrogenic influences with noise and light and really improve that. But as Melissa talked about earlier, sometimes it's kind of impractical. And sometimes you're working with the fire department to improve some of these things and have to kind of fight a pretty big battle. So the question that comes up is, can we potentially medicate through this, these iatrogenic influences? And Huang and colleagues did an interesting study where they adapted 40 healthy subjects to the ICU environment. On night two, they randomized these subjects to either quiet and lights off or the ICU environment with the noise and light. On night three, they actually flipped the arms and exposed them to the opposite environment. And then on night four, they exposed all subjects to the ICU environment and continued to expose all those subjects to the ICU environment on nights five through eight. However, they randomized these subjects to one of four arms. So they either received no intervention, placebo, melatonin, or earplugs and eye masks, which I think was another question that came up earlier. So this is the serum plot of melatonin levels in these subjects. And this is kind of the common sense part of the study, where if you expose these healthy subjects to the ICU environment, their melatonin levels are significantly lower. And you can definitely see that by looking at these two graphs here. However, this is where it gets a little bit more interesting. So in these healthy subjects, you can see with the three lines that are kind of clustered there at the bottom, that their melatonin levels were really pretty low and pretty similar between the groups that received no intervention, placebo, and eye masks and earplugs, interestingly. However, in the patients or in the subjects who received melatonin, you can see that after melatonin administration, which was given around 9 PM, then around one hour after administration, they achieved a nice peak of about 1,000 pg per ml, which then, of course, steadily declined over time after administration, but remained significantly higher than the other groups from the periods of about 10 PM through 3 AM. And then getting even more interesting in this healthy subject population is they did do polysomnography, and they did look at subjects' report of sleep. And they actually saw that in the groups receiving melatonin as well as the earplugs and eye masks, they actually saw a significant reduction in sleep on or, yeah, reduction in sleep onset latency as well as a decrease in arousals and awakenings. And then comparatively between those two groups, they saw an additional significant reduction in awakenings and arousals with melatonin administration. In addition, melatonin increased total run time as well as total sleep time. So I know what everyone's thinking, but that was healthy subjects. So what about my really, really complicated ICU patient? Well, there was a recent Cochrane review done looking at melatonin specifically for the promotion of sleep in the ICU, and they ended up looking at four studies that were all single-center randomized controlled trials. Generally, a general adult population or mixed med surge with the exception of one that looked at neurosurgical patients. Two of the studies were administering 3 milligrams of melatonin. One was administering 10 milligrams. And then one had a unique regimen of, which I've actually never thought of doing, 3 milligrams at 8 p.m. and then followed by 3 milligrams at midnight. And as expected, unfortunately there was a lot of heterogeneity between the studies and the methods as well as the data that they reported. And so they weren't able to combine these data into a meta-analysis. However, looking at the overall outcomes related to sleep, it was determined that administration of melatonin either as compared with placebo or usual care was not shown to be beneficial. So that's the overall outcome. However, I think given that the Mistraletti study represented about half of the patients in the Cochrane Review, it's worth digging a little bit deeper to see if there's additional signals there. So again, this was a study that did 3 milligrams at 8 p.m. followed by 3 milligrams at midnight. The primary endpoint that they actually looked at, and I'm sure there are a lot of people in here cringing, is the use, the cumulative use of sedative as defined by hydroxyzine. And so they did see a significant reduction in use of hydroxyzine with administration of melatonin. They tried to look at sleep quantity as per actigraphy. But unfortunately there were some data reliability issues and they weren't able to report that information. However, they actually, when they adjusted for some accumulative melatonin exposure, they actually saw that nurse-reported sleep was significantly improved with administration of melatonin. And Melissa's rolling her eyes. And in addition, also with adjustment for melatonin exposure, they also saw a number of other secondary endpoints that were improved, including pain, anxiety, agitation duration, and need for physical or chemical restraints. So I think, you know, that's certainly some interesting signals there, although some of those signals caused eye rolling in our sleep experts. So I'm not sure how strong they are. And then I think overall the studies with melatonin have been largely underwhelming, probably because of the methodological flaws and the severely underpowered nature of those studies. There have been two studies that have been kind of interesting looking at remeltion, however. So I want to dig into that. And the first was done by Hata and colleagues in Japan. And they actually, so they randomized 67 older patients after hospitalization to receive either remeltion or placebo for seven days after hospitalization. And note that I said hospitalized patients. So only 35% of those patients were actually critically ill. So this does limit somewhat the applicability to our patient population. However, it's interesting that the primary endpoint of delirium was significantly reduced in this patient population. They did try to look at a number of secondary endpoints related to sleep, and unfortunately saw no difference there. So I think what was needed after the Hata study was really a study designed similarly but looking at ICU patients to see if perhaps the signals that we saw with improved rates of delirium would pan out as well as perhaps some of those secondary endpoints related to sleep. And so Nishikimi and colleagues, another study done in Japan, did exactly that. And they randomized ICU patients within the first 48 hours of ICU admission to receive either remeltion or placebo. And they continued that until either ICU discharge, death, or needing to change because of an adverse event. The primary endpoint that they looked at, and I think this is really relevant, is ICU length of stay. My husband always says, well, who cares? He's a physician, so he's like, what's the endpoint? And I think ICU length of stay is a pretty relevant endpoint to look at. And so initially didn't see a difference, but after adjustment for baseline characteristics, they did show that administration of remeltion led to a significant reduction in length of stay. And they also did look at a number of sleep-related secondary endpoints and saw a reduction in the number of awakenings as well as a greater proportion of nights without awakenings. They did not see a change in the mean hours of sleep, however. And consistent with the HADA study, they also did look at delirium, although as a secondary endpoint, and saw a reduced occurrence of delirium as well as a reduced duration of delirium. So I'm pretty excited about the results of this study. So where does this leave us? I think we're in an exciting time because there are some fairly strong signals that there may be a benefit by targeting this melatonin pathway, but we're not quite there yet. And I think there's still some room for more evidence with larger studies. So the PRO-MEDIC study is actually a study being done in Australia, and their plan is to enroll about 850 patients with an expected longer length of stay and randomize them to receive either melatonin or a placebo. And they do plan on looking at a subset of 100 of these patients and specifically provide data on polysomnography to evaluate sleep outcomes as well. So the hope is that the PRO-MEDIC study will better inform us about the role of melatonin specifically and perhaps better elucidate the role of this pathway in sleep in our patients. So in summary, I think it's very clear that melatonin levels in ICU patients are abnormal, but I think the abnormalities are very heterogeneous and the causes are likely multifactorial. Assuming that improving or normalizing melatonin levels and cycles in our patients will lead to better circadian rhythms, better sleep, then I think melatonin and rimothion are rational treatment options to get us there. Unfortunately, I think that there have been some serious methodological flaws and underpowering with the studies looking at melatonin, and there have been some more promise with studies looking at rimothion, but I think we still need the ongoing study to be completed and future larger generalizable studies to really be able to answer the question about the role of this pathway in promoting sleep in our patients. With that, I'd be happy to take questions. Thank you. So I have a question, and then while I'm asking if the speakers want to come back up and give everybody an opportunity to ask questions, but most of the latest iteration of the PADIS guidelines are sort of silent on melatonin and rimothion and sort of say that they can't make a recommendation based on the data. So what are your thoughts about the recommendation not being there and where would you sort of recommend this in your practice? My personal opinion is there is a financial implication of putting every patient on a certain agent. And so before we have really clear data to say that any of these agents are really our standard of care, it's very hard for, I think, guidelines to be able to make a definitive recommendation that either melatonin or rimothion should be administered to ICU patients either for promotion of sleep or delirium or reduction in delirium. But I think it's a rational target in a patient that we think is high risk. I think even though the guidelines don't necessarily speak to it directly, I think that the fact that they're not recommending against it is something that we can consider and definitely think about our patients on a case-by-case basis and whether we want to start those drugs. Do you know, is there data to suggest that melatonin actually improves sleep in anybody? Like in hospitalized patients or outpatients, is there any large stats? We actually, so one of the first things I did when I started at University of Maryland was to request for melatonin to be added to the formulary. So we reviewed the data kind of more broadly and it's certainly not strong. It's actually, it's not necessarily much better than what I presented today. But I think given the perception that the drug is fairly benign and not overly, certainly not overly expensive, I think that was kind of why we had decided to go ahead and add it to our formulary. But not necessarily based on great data. Do you think, is there a reason you think that the studies that looked at remeltion actually show more of a signal than the ones that looked at melatonin? That's a good question. My sneaking suspicion is that they were better designed. And so that's why I haven't given up hope on melatonin just yet. And I think, you know, I don't know that the difference that we saw is necessarily related to like receptor selectivity or anything else specific to the drugs being slightly different than each other. But I think we just need a better powered and better designed study with melatonin to really be able to answer that question. So I'm really looking forward to the results of the ProMedx study actually. Just to piggyback off that, do you have concerns or thoughts on, you know, the thought that melatonin is like an herbal supplement, so not really regulated, so maybe more variations versus remelting on being kind of FDA approved? Yeah. We consider that as well from like a formulary decision. And there are products that are, you know, FDA monitored. So I think that that becomes less of a consideration as long as you're careful about which products you're bringing into the hospital, then it's fine. So I just had one comment. Is this on? My one comment is that I do think it's very exciting, and there's some signal there that we're seeing in these studies. The thing that no one talks about, and it's what I say to my residents when they say, why are you rolling your eyes? I like to roll my eyes. Why are you rolling my eyes about the melatonin? Like we thought you liked that, Dr. Kinnard, is I always say, as soon as you tell me the person's habitual sleep time, and when their melatonin normally onsets, then you can go ahead and write that prescription. So I think one of the problems in the literature is that it's very different what time, you mentioned the doses, but also what time, and also that very unusual dosing of three and three. And so I think one of the things we need to do as a field is maybe standardize that, and we might see some nicer signal, and that might be the remelting. Absolutely. And I just want to add, because I think this is important, and it's simple, but we don't think about it very often, is administering melatonin to a patient, but then waking them up every two hours to do some kind of assessment, or waking them up at 3 AM to give them a bath. Those are definitely the things that we have to think about and just practically take a step away. Why is the patient not sleeping? Is it because they're, potentially the iatrogenic things that we're doing to them? Neil Reynolds, University of Maryland. I have three questions. It's premised on the basis that we've already done all the sleep promotion things we can, so I just want to talk pharmacology. First question, do you have a scheme for escalating the dose from day to day, and what's your max dose of melatonin? Second question, what do you do when you're at max dose and you want something else? Do you have an additive drug? The third question is sort of rumor. There's some scuttlebutt on the wards about midnight or middle of the night awakening when you use melatonin. Have you seen that or heard that reported? So your first question was about the dosing, and do I have like a dose escalation kind of strategy? Not necessarily. I kind of go big, and with like the BORN study that was mentioned in the Cochran review, so they did the 10 milligrams, not because that study showed any benefit, but I just, I tend to go big in general in the ICU with patients, or go home. But I have heard, you know, I would say just anecdotally, I've seen a lot of people start at like 3 milligrams and then go to 6 and 9, and, but that's not based on any kind of evidence whatsoever. Your question about the middle of the night awakening, so that, I haven't seen a whole lot of that. However, it is interesting that given, in the Mistraleti study, that they did the kind of split dose with like the 3 milligrams initially, followed by another 3 milligrams. So it almost suggests that perhaps that middle of the night kind of dosing would be potentially appropriate with melatonin. But again, no data on that either that I'm aware of. And I think you had a second question, and I forgot what it was. What do you do when melatonin, your max dose, what do you add? Oh, yes. Yeah. So when we've gotten to the max dose of melatonin, like if I've done the 10 and it's not helping, I really dig in, obviously, to see if there's anything else that we've absolutely missed that, you know, we're waking up the patient unnecessarily. And then, I don't know, from there, I mean, sometimes, we've occasionally done dexametatomidine if they're delirious and need a little, and they're, you know, anxious and we need a little bit of anxiolysis as well and maybe a little bit of agitation. But not, I don't think there's like a go-to like next step that I've necessarily gone to. Can I make you answer a question then? Do you ever find addition of trazodone useful? Yeah. I think we definitely got into the habit of using a lot of trazodone. I don't, you know, I don't have the data to say what percentage of patients have been prescribed trazodone in recent years. But I can anecdotally feel like that's, it's been fairly high. But not based on a lot of great data supporting it. And I think given there are some side effects related to the drug, we've typically steered away from it. Hi. I'm Kanina Kavarma. I'm in critical care and sleep at OSF Healthcare in Peoria, Illinois. So I'm also a sleep physician, so I just kind of wanted to get an idea from everybody what their thoughts of, because we use, you know, melatonin as a chronotrope, meaning pushing and pulling people's sleep, natural sleep time. So without knowing somebody's natural sleep time and giving melatonin at a specific time, I mean, we could possibly be pushing their own melatonin release and adding to that delayed sleep phase problem that we already have, along with all of the other things that we do in the ICU. So that's something to think about, because we think of sleep time, oh, it's 8 o'clock or it's 9 o'clock or, you know, we need to get them to sleep, and that may not be their sleep time. So by giving melatonin, we could be potentially making this worse for the patient. The other question I have, and this is for everybody up here, is what are you guys' thoughts about the new orexin inhibitors that are out in the market in the inpatient world to use, so such as Suvorexan and the other one that was just approved, which inhibits orexin 1 and 2? Any thoughts about using those in the inpatient world to promote sleep? From my perspective, I'm curious to hear what you guys think, but if I'm worried about the cost difference between melatonin, which costs like $0.26, and remeltion, which is like $14, I have a feeling those agents are going to be quite costly. So they probably wouldn't necessarily fall, you know, we wouldn't necessarily add those to our formulary and start using them more widely given that consideration, especially without data supporting them, especially in ICU patients. Well, thank you, everybody, for coming, and I'd like to thank our speakers for wonderful talks.

intermittent enteral feeding

ICU

underfeeding patients

continuous feeding

protein synthesis

rehabilitation activities

evidence

studies

melatonin

sleep

critically ill patients

abnormal levels

remeltion

delirium

sleep disruptions