Well, good morning, I'm honored to be with you guys. My name is Pete Johnson. So a little bit about me, I'm a professor with the OU College of Pharmacy, have an adjunct appointment in the College of Medicine. I'm a pediatric clinical pharmacist, have practiced in the PICU and the cardiac ICU, but more recently in the cardiac ICU and have a definite research interest, practice interests in delirium, sedation, analgesia, etc. So my objective, really straightforward, is talk about prevention and pharmacologic strategies. In terms of disclosures, no financial disclosure, but I guess I have a bias, and that is as a pharmacist, I think that drugs with delirium are often the problem, and as we'll talk about, not the most effective or safest solution. So in terms of pharmacologic prevention strategy, so I did have an honor of working with Christina and Stacey on the PANDEM guidelines. So just to make sure we're all on the same page, so one of the first ones, the strongest recommendation is recommending against routine use of benzodiazepines as they've been independently associated with delirium. The next one, a conditional recommendation is to minimize strategies whenever feasible for sedation. And then also do not recommend the routine use of haloperidol or atypical antipsychotics for the prevention of delirium. So I also would relate back to brain maps, and the way that I think about this is this is a very helpful tool in thinking about prevention strategies from a pharmacologic realm. So one of the first one, pretty simple, the R, remove or reduce medications associated with delirium. So in terms of medications associated with delirium, there's definitely been a number of epidemiologic studies that have looked at this. Really, as we talked about, benzodiazepines have been independently associated with delirium, as well as anticholinergics. And what I will say about this is many medications actually have anticholinergic properties. Promethazine, diphenhydramine that we might not often think about, and also the thing that I think of is the anticholinergic burden that many of our patients have if we start drugs like oxybutynin or diphenhydramine that they may receive. So it's not just one drug, but maybe multi-factor when it comes to anticholinergics. As well as opioids, that one has been associated in some studies, although I would argue that many of our patients do have pain requirements, though I think it's prudent to use the lowest dose when possible. Antiepileptics have been associated, but I think that that's probably a marker of, again, one of the risk factors, cognitive impairment that some patients may have that have seizures. And then vasopressors, that could just be a surrogate marker for their critical illness. Looking back at the brain map strategy, or the framework, rather, again, certainly other metabolic disturbances, if they have electrolytes, impairments, or metabolic acidosis, alkalosis, as well as pain and sedation. So in thinking about sedation, it's definitely recommended in the PANDEM guidelines that dexmedetomidine is the most recommended sedative for most of our critically ill children. But what about its role as a prevention strategy for delirium? So compared to benzodiazepines, which do negatively impair sleep, dexmedetomidine has not been associated with altered sleep patterns like benzos. And there is some data in adults that you might be familiar with. These are older studies that have looked at the role of comparing dexmedetomidine versus benzodiazepine-based regimens, and have found, in these two studies, that there was a decrease duration of days without delirium when using dexmedetomidine as opposed to midazolam. However, that being said, and noted in the PANDEM guidelines, there really are only two studies in the operating room that have shown, at least in children, a decreased emergence delirium. So there really, at least to date, is not a randomized trial, like in our adult colleagues that have shown the role that dexmedetomidine may have, I'm delirious, in preventing delirium. So what about the role of melatonin? So many people think, oh, this is great, it's really safe. As we'll talk about, maybe not. So this possesses anti-inflammatory, antiepileptic, and antioxidant effects. It's definitely involved with synchronization and regulation of the circadian rhythm to help with sleep management. So one study in the operating room setting actually noted a dose-dependent effect of melatonin in the decrease in emergence phenomenon in children in the operating room. But in terms of its role in the ICU setting, there's probably a little bit more published data, not a little bit, there's a lot more published data in the adult critical care world. So Ng and colleagues conducted a meta-analysis of 16 studies in 1,600 adults. They found that when using melatonin, there was no significant decrease, rather, in the development of delirium. And they also did note a decrease in mean length of stay in melatonin versus non-melatonin patients. So what about its role in children? Really, there's limited evidence in hospitalized children, much less critically ill children. In terms of dosing, I think this is one area of controversy. So the range is really broad. So that's been reported, 0.3 to 10 milligrams a day. Many of you may take melatonin. So in terms of pediatric patients, some have suggested going off the premise of using the lowest dose possible of a weight-based dosing strategy of 0.05 to 0.15 milligrams per kilo. Our colleagues in Europe, the European Pediatric Neurology Society, recommend that they're less than 40 kilos, a maximum of three milligrams a day, and then greater than 40 kilos, five milligrams per day. So in terms of the ICU setting, there are two very, very small retrospective, so full of flaws, studies that looked at the role of melatonin. So the first one from UNC, and you can see here the median age was around nine. They utilize this to prevent and possibly treat delirium, but you can see the majority of these patients still required antipsychotics. They did note a decrease in RAS scores before and after the use of melatonin. One from our own institution, this was a study specifically looking at infants, so included 5.5 months of age. The median dose, you can see, is higher than the dosing reported below, but logistically, which we're going to talk about in the next slide, that had to do with the dosage formulation that we had on formulary, we did find that this did lower the cumulative opening dosing before and after the use of melatonin, though. In terms of considerations with this, again, I think some of these things are really problematic with melatonin. It's available in all sorts of routes, but it is a supplement. So one thing that's very important to note is it's obviously not regulated in terms of what is actually in it. So there was one study that looked at 31 different melatonin supplements, and as you can see here, this is not a typo. There was a 465% variability between the different dosing formulations that are out there. I think one thing that could be utilized to kind of standardize this is what's called the United States Pharmacopeia. They do look at, not the efficacy and safety, but the reliability of what's in the product. And I think I would argue or recommend that you work with your pharmacist to try to get a formulation in your hospital that might have or be regulated by that. But I will say that most of the time, the oral solutions that we might use for our young infants, there really is no USP brand. So that is unfortunate. In terms of adverse events, this is where I also push back on, it's really safe, we should just give more. Drowsiness maybe has been reported, but doses greater than three milligrams in adults, and who knows what that might be in a neonate or a small infant, that's actually been associated with hypothermia and increased melatonin concentration. So again, extrapolating that to our infants, it's really hard to say. In terms of treatment recommendations, one conditional recommendation is suggest that in refractory delirium, so bolded there, that haloperidol or atypical antipsychotics could be used for severe delirium. And really our only strong recommendation is that because of how toxic these drugs are, is recommending a baseline EKG followed by routine electrolyte monitoring for these agents. So when thinking about first versus second-gen antipsychotics, so really our first-gen agents, just kind of a reminder from a pharmacology standpoint, includes haloperidol, chlorpromazine. Our second-gen agents, quetiapine, olanzapine, risperidone, aripiprazole. And really in terms of the comparison of these agents, there certainly is less extra pyramidal side effects, sort of dyskinesia, neurologic effects, and also decreased effects with prolactin with the atypical versus the first-gen agents. However, there's an increased risk of cardiac toxicity rather with our first-gen agents compared to the atypicals. But on the downside, in terms of side effects with the atypicals, there is an increased risk of metabolic side effects, which we'll talk about in a moment. So my colleagues and I did a systematic review a few years ago on single center, retrospective case reports, case series, et cetera, on agents that have been published in delirium and have added to it on this slide based on more recently published studies. So you can see haloperidol had the most number of patients. In terms of adverse events, about 14% had dystonia. And just some pearls is there is an oral liquid formulation available. And as we noted, that this agent should have the highest risk of QTC prolongation, but I think it's probably a publication bias in that nobody looked for it or they didn't report it. So I would argue that that risk is probably much higher than our other agents on the slide. With quetiapine, been some of the more recent reports with this. There have been six patients with QTC prolongation. And then one patient, I will definitely note, going back to the premise that these can cause metabolic side effects. So a two-month-old infant that was on quetiapine for less than seven days actually had a triglyceride of 573. So again, going off the premise that not the safest thing. Olanzapine, risperidone, the things that I'll note with this is that really there's more limited dosing. These are more fixed dosing approaches and just really more limited evidence in that sense. Really, there is only one study that I have found that kind of looked at a comparison of outcomes. So this title slide's a little misleading. They actually didn't directly compare haloperidol versus quetiapine. So this study provided a comparison of outcomes. So it was actually a retrospective study, a really small sample size, in which they matched patients two to one. And they compared the outcomes with the match, the haloperidol versus unmatched patients, et cetera. And their outcomes, they looked at improvement of delirium with CAPD scores, time to resolution. But what's actually frightening with this is that there was really no difference in delirium improvement between the haloperidol untreated versus, and that says risperidone, but quetiapine untreated patients. The haloperidol patients had more days with CAPD scores, higher CAPD scores than the untreated patients. And then quetiapine, a longer duration of stay versus the untreated patients, and more of a functional decline at discharge than the untreated patients. So really, this speaks to the fact that we don't really have good evidence to say that they actually work. In terms of QTC prolongations, again, there's higher risk with haloperidol. There's a variety of different definitions. And as I mentioned on that one slide, many of these reports that have been published may really have underreported how often this may have occurred. The frequency of monitoring, I think, depends upon the number of, what their baseline EKG could be, their number of QTC prolonging medications, electrolyte disturbances, and of course, their past medical history. In terms of the metabolic considerations, so there are several different groups, the American Psychiatry Association, that recommend that patients that are on these for long-term, again, I hope they're not there on long-term, but as I'll show you in a second from some of the studies, it's less than be desired, that they recommend screening for total cholesterol triglycerides and HDL at baseline and at several-week increments. In terms of the duration, so you can see here that in those reports in ICU delirium, some patients were on these from one to anywhere from 22 to 100 days. So definitely not recommending that, but that's what was reported. Again, we noted that one infant with quetiapine that had a triglyceride of 570. Some colleagues and I actually did a survey of clinical pharmacists at a number of different institutions, and we asked a number of different questions, but one was the frequency of metabolic-type monitoring for these agents. So only 30% recommend, or actually utilize routine monitoring for triglycerides on these, and there was a variability of monitoring practices, and many of them did not have an established threshold for when they might stop some of these agents. In terms of tapering considerations, I do think that pretty much any drug that affects your CNS, probably if you're on it for a number of days, does need, may need to be tapered. Again, I'm not recommending the duration that was reported in some of these studies, but one thing that I think this should speak to is trying to get rid of it if it started. However, in the adult world, there was one study that looked at antipsychotic initiation in the ICU, and they actually noted anywhere from a quarter to 85% that were continued on these when patients actually went to the floor, and the majority of them, the use was inappropriate. They no longer had delirium, as shown by delirium screening scores, and then unfortunately, almost half, or up to half, were continued on these at hospital discharge. So one thing that I'll say with tapering considerations is that in adults, akathisia, dyskinesias have actually been reported with abrupt discontinuation, and so I, this is just my opinion, I think if they've been on it for like 10 days, which hopefully not, but if they were, they probably should be tapered just to minimize causing another problem. In terms of additional agents for delirium, there's a little bit more studies in adults, but the reality is that of these agents listed on the slide, none of them are recommended by the PANDEM guidelines, just to the lack of quality of evidence. So in conclusion, going back to my bias, I really think drugs are the problem, and not really the most effective or safest solution, so definitely using non-pharmacologic therapy, and then if they are used, really the choice, because we don't know how effective they are, it's probably based on toxicity, probably based on available dosage forms, clinician preference, and that if they are used, baseline EKG, but another thing would be considerations for dyslipidemia if patients are on these for several days. So thank you for your attention.

delirium

pharmacologic strategies

benzodiazepines

sedation strategies

non-pharmacologic therapy