SCCM Resource Library-Controversies in Delirium: This Drug, That Drug, No Drugs at All!

Video Transcription

Thanks, everybody. I'll try to make up for a little time. I guess one of the most important things if we're looking at reducing delirium with sleep improvement is quietness.  So we certainly had a little bit of that. I'm going to be focusing on analgesics and sedatives, sort of in the context of preventing and treating delirium. These are my disclosures.  Obviously we know one of the hallmark things that we can do on a daily basis in the ICU is reduce risk factors, evaluate and reduce risk factors that are modifiable in our patients.  And medications play a big role here. The other thing when we're talking about delirium in terms of preventing it and treating it,  and treating it, obviously one of the most important outcomes is reducing symptoms, but also reducing its duration.  And this can kind of have a profound effect on the post-ICU outcome of our patients, which  can go long beyond the time of hospitalizations as they hopefully, like this lady, has a pretty successful outcome because she's back walking her cat, actually.  I didn't know you could walk cats, but there we go. I just had to find someone with gray hair all the way across, so it's not the same lady.  And of course, most people are familiar with the A to F bundle. And choice of analgesia and sedation is a big part of this. Some of it is choice.  Also a really important part of this is de-escalation of medications on a daily basis. I'm not going to be talking about that.  So certainly everybody knows this famous Jose Maldonado slide in terms of all the potential pharmacologic pathways which we could use in delirium.  I'm going to be talking about gamma inhibition, MDMA antagonists, and alpha-2 agonists. So certainly there's sort of a hit list of medications that has been—it's available  in journals, websites. We have them certainly in our hospital delirium pathways and up at our COE thing.  And certainly analgesics, particularly opioids and then sedatives, benzodiazepines and propofol are sort of the culprits or the potential culprits, which I think is a better term  to use, that I'm going to focus on today to really talk about the potential risk and association with delirium in terms of both preventing delirium and treating delirium.  There's not a lot of great data in the ICU that have carefully evaluated the association between opioid use and delirium.  Margaret Pisano, in a cohort of older adults at Yale, did find a pretty strong association between benzodiazepines or opioid use.  But when she found opioid use alone, there wasn't a significant association. In a large clinical trials group study, we looked at a cohort of patients that were matched  with delirium or without delirium and found as you increased fentanyl dosing in the ICU by 0.1 milligrams, there was virtually no association between delirium and opioid exposure.  When you look at the non-ICU literature where there's a lot more data, it's kind of all over the place, but really not a strong signal in hospitalized non-ICU patients that if you  give patients opioids that they're going to become delirious. I think the other important thing in all of this is we do know that post-operative  pain, this hasn't been well validated in the ICU, but moderate pain post-operatively, this is a VAS of one to four, is two times greater risk of developing delirium, and as you get  to severe pain, a VAS greater than five, nearly a four times greater risk. So pain itself is a potential risk factor for delirium, and it makes it kind of complex  to really figure out the association, particularly in most patients in the ICU if they have pain,  you know, it's fluctuating, delirium fluctuates, and you know, we're treating the pain and often going up and down in our opioid doses, so it's kind of a complex relationship.  Really the only way we can evaluate this, and I don't want to get deep into a pharmacopoeia  epidemiology course here, but is we need to, as we've done with benzodiazepines and anticholinergics,  et cetera, we need to use sort of a daily Markov multivariate time-dependent analysis. And this is simply where we're looking at the daily risk of transitioning from a patient  that doesn't have coma, so they're awake without delirium, the next day moving to a delirium state, and of course there's baseline risk factors that we're all fairly familiar with,  but there's also daily risks. So for example, on day one, this patient's on an opioid.  Day two, they're on an opioid, but day three, they actually stopped the opioid and had delirium. Is that a good association?  The opioid stopped on day four, no opioid here, but now they're awake, but the complex thing here is, you know, what if I tell you a midazolam infusion started here?  Is it the opioid or is it the midazolam that caused the opioid, right? We have daily, we know severity of illness is a big risk factor for transition of delirium  risk each day, so what if the patient had new sepsis and they got a lot more sick? So it could go on and on.  The only point I'm bringing up here is these relationships are really complex to evaluate and they haven't yet been done.  And certainly the things we do are looking at, you know, admission service, age, severity of illness, you know, Charleston comorbidity, hypertension, you know, factors that could  affect the PKPD response of medications. And then the time-varying ones are, again, daily severity of illness, sepsis, as again,  presence of pain is really important to evaluate each daily. And then what if the patient got dexmedetomidine, which I'll talk about in a second, could potentially  have a delirium-reducing effect. And so this is a benzodiazepine association study where we found a 1.04 odds ratio, which  doesn't look particularly high, but it's actually the, it's a 4% daily odds of transitioning  from an awake state, say on ICU day three, to delirium on day four for every five milligrams of midazolam administered. So this is the kind of data we need with opioids.  So we can't rule out that opioids have no association with delirium, but we have to really be careful until these rigorous studies are done to really say that opioid is an important  risk factor for delirium. I wanted to step back, though. It doesn't mean that we should be giving opioids necessarily to every single ICU patient.  So one of the things that we started to evaluate in the PET-IS guidelines that I chaired was the potential use of non-opioid analgesics.  And there's some interesting studies that are starting to come out looking at the delirium risk with these non-opioid medications.  And these are from the PAD-2013 guidelines, and we make pretty strong recommendations from with opioid use. It's really based just on one cohort analysis.  And certainly I think we're learning a lot from our perioperative colleagues, only some of which these patients transition through the ICU, although maybe more of a multimodal  approach where we're reducing opioids could improve outcomes. Delirium potentially could be one of them.  So we made a multimodal analgesic question, critically ill adults in the ICU. And these aren't patients that are just quickly transitioning through the ICU.  These are critically ill adults. We looked at acetaminophen, nephepen, ketamine, lidocaine, and NSAIDs, looking at scores. And we did look at delirium if it was reported.  This is the algorithm we created, and I'm just going to spend a little bit of time talking about acetaminophen and low-dose ketamine.  DEXACET randomized control study came out in JAMA. This is after our guidelines were released and published. This is an interesting study. So it was cardiac surgery patients.  They were randomized to receive either propofol or DEX, and also, so it was a double-arm study, acetaminophen or placebo.  When you look at the acetaminophen versus placebo results, in-hospital delirium was significantly reduced. Days with delirium was reduced by half.  And this is a Kaplan-Meier score of the time to develop delirium. And there was about a 20%, 35% decrease in opioid dosing.  So it's impossible to answer the question, is the reduction in opioids associated with the use of acetaminophen versus placebo? But this is kind of a hypothesis generating.  And then, interesting enough, as I sort of transition into the sedative part of my talk, in this study as well, realizing it's cardiac surgery patients, which have many differences  to our general medical-surgical ICU patients, which could be quite a bit sicker, they actually  found no difference in transition to delirium between the propofol and DEX arms, which is kind of interesting. So what about ketamine?  We were only able to find one single dose, one single randomized controlled study of ketamine use. This is after major abdominal surgery in patients that stayed in the SICU.  Of course, we all know there's many, many studies evaluating ketamine use in non-ICU populations.  The authors did not evaluate delirium, but they did bring up potential for delirium and confusion in patients.  And this is just sort of a summary of these many, many randomized controlled studies in non-ICU studies. And it's interesting.  Of course, I didn't try to summarize all the results of these individual studies, but delirium certainly pops out, even though we're using low doses of ketamine.  So ketamine certainly could be a risk factor for delirium. There's not a strong signal that it reduces it.  Dexmedetomidine versus morphine has been compared in, again, cardiac surgery patients in the DEXCOM study.  And they did find a reduction in delirium, but it wasn't statistically significant between dex and morphine as the primary sedative agent.  So we look at delirium pharmacologic prevention and critically ill. Pam's going to talk about some of the other interventions, but focusing on dexmedetomidine.  So this isn't comparing dex versus another sedative. I'll get to that in just a sec. This is dexmedetomidine versus placebo. This is a large Chinese study.  And these are patients that had non-cardiac surgery, so primarily major abdominal surgery. But their APACHE-2 scores were around 8 to 9.  The average median time in the ICU was less than 24 hours. And they did find an association between administration of dexmedetomidine versus placebo in terms  of reduction of delirium. What we decided, though, based on these studies, particularly given how much you could apply  this study to the general adult critically ill population who's sicker, we suggest not using dexmedetomidine to prevent delirium.  But there could be patient populations where this might be optimal. There's a study that I helped work on.  And this was a nocturnal dexmedetomidine study that was published in the Blue Journal a couple years ago, where we gave low-dose dexmedetomidine at night versus placebo.  These were all patients that were delirium-free at time of randomization. And delirium occurred less frequently in the dex patients than placebo. We also looked at looking at sleep.  We couldn't find a sleep signal at all. But of course, that's very hard in a study like this.  There's a lot of confounders in all of this, though, because these were all mechanically ventilated patients that were fairly sick, already getting a lot of propofol, and in  some cases, benzodiazepines. So it's really hard to sort out what the signal here. But it does appear that adding dexmedetomidine, whether it's the dex itself or reducing the  gabaminergic agent, seems to have a delirium-reducing effect. So we suggest looking at comparing sedative agents in the PAD guidelines.  We suggest using either propofol or dex over benzodiazepines for sedation in critically ill mechanically ventilated adults.  And of course, we had multiple outcomes for this question. But delirium was one of them. And I'd just like to focus on dex versus benzodiazepines and dex versus propofol.  Interesting enough, most of the propofol versus benzodiazepine studies are from the 80s and 90s, and delirium wasn't evaluated in any of those.  So when we looked, this is our pitot diagram for the dex versus benzo studies. Again, non-cardiac surgery, mechanically ventilated adults.  And the only study that really showed a significant difference is the CEDCOM study that compared dex versus midazolam.  A little bit of a trend here with pentahyripondi, but overall, it was almost significantly different, but it wasn't.  So I think it's important that we realize when we summarize these studies that we didn't find a significant association across these four studies. This is the CEDCOM data.  And you can certainly see with the Kaplan-Meier, the prevalence of delirium here reducing each day faster in the patients getting dexmedetomidine versus midazolam.  But again, it's sort of the question is, is it just like this nocturnal dex study, is it reducing the midazolam, it's decreasing the daily prevalence of delirium, or is it  actually a specific dexmedetomidine effect, and we're still not really sure. This is data from the MEN study, again, not statistically significant for delirium.  This is dex versus lorazepam. However, when the investigators looked at the subgroup of patients that had severe sepsis,  there was almost a significant difference in the delirium-free days here. It was a close confidence interval.  Looking at dex versus propofol, the only study that we found, and it actually didn't look at delirium on a daily basis, it looked at delirium 48 hours after ICU discharge,  and found a trend for reduced delirium with dex versus propofol. However, many of you are familiar with Dr. Shahabi's recent New England Journal study  comparing dex versus sort of standard treatment, much of which was either benzodiazepines or propofol, and there was no difference in days free from coma or delirium across the  two groups. So I think we have to, you know, questions arise about what exactly is the treatment effect, specifically looking at delirium when we use dex versus other agents.  There's probably a signal here, but it's probably not as strong as many people might assume that it might be.  When we're looking at delirium treatment, we did investigate this question, and so these are people with delirium. Can we use dexmedetomidine to treat it?  There's one Australian study, and these were patients that were mechanically ventilated who became agitated during their SBT, and they were randomized to dexmedetomidine versus  placebo. And they did find a significant increase in ventilator-free hours with the dexmedetomidine-treated arms. However, there's some external validity issues with this study.  Out of 1,000 patients were intubated to screen and enroll 71. Alcohol withdrawal patients were not excluded. The study was terminated early.  Many patients did not receive opioids, so with some of their agitation, and potentially even delirium pain-related, and no effect on ICU or hospital length of stay.  So we do make a pretty specific recommendation in the PAD guidelines suggesting dex for patients  that have agitation accompanied by delirium during SBTs where it's precluding weaning and extubation.  So in summary, the association between opioids and delirium, I think, remains still quite unclear.  IV sedimetidine, based on the one study in cardiac surgery patients, may reduce delirium after cardiac surgery, so I think we need some more investigation on this.  Evolving evidence suggests that dex may help prevent delirium, but there's a lot of factors  of course that could affect this relationship, and they haven't been really well evaluated. Among patients requiring continuous sedation, dex appears to reduce delirium prevalence  compared to a benzodiazepine, but at this point, there's little evidence to suggest that there's a change in delirium when dex is compared to propofol.  And then as I mentioned, dex may help facilitate extubation. Thank you very much. Thank you. Thank you.  Good afternoon, everyone. I have the privilege of following Dr. Devlin to speak about antipsychotics and the treatment and prevention of delirium. I have no conflicts to disclose.  So with this portion of the session, I'm going to really focus on recognizing situations in specific patients where antipsychotics may be appropriate for prevention and treatment  of delirium in the ICU. So first of all, I'd like to just give a very brief background of antipsychotic use for psychotherapy.  In the early 1950s, chlorpromazine, a typical antipsychotic, was originally used as an anti-emetic.  A Paris surgeon saw that his patients that were receiving chlorpromazine had less anxiety going into surgery.  So this individual believed we should maybe try this to augment anesthesia for surgical shock and was very pleased with the results.  He communicated these results to a Canadian psychiatrist who then began using chlorpromazine for his most agitated patients.  And from this bore really the use of antipsychotics for agitated patients. Since that time, we have had more agents enter this class.  They're divided into our typical antipsychotics, just as a background, and atypical. Our typical antipsychotics, chlorpromazine, heloperidol, primarily antagonize dopamine  receptors, whereas your atypical antipsychotics antagonize both but have more affinity to the serotonergic receptors.  A brief table of the different affinities for the typical antipsychotic heloperidol and your atypical antipsychotics I thought were useful just to begin our talk.  As you can see, our heloperidol has antagonistic affinity for our dopamine receptors, and our  atypical antipsychotics, specifically zeprasidone, quetiapine, risperidone, and lalanzapine, are more active with our serotonergic receptors.  I would like to call out, if you look, there are differences within the atypical antipsychotics,  even beyond that of the serotonergic antagonism with the adrenergic, histaminergic, and muscarinic receptors.  And we'll refer back to this slide as we talk more about the studies in this area, but it's a good baseline to look at the differences with affinities.  So as we move to talk about antipsychotic use with delirium, as Dr. Devlin mentioned and showed the very busy slide with different pathways, delirium itself, the pathophysiology  currently is very complex. We don't have a complete understanding of it. And there's multiple potential hypotheses how neurotransmitter abnormalities come into  play with delirium in the ICU, specifically dopamine and serotonin dysregulation and potentially  a decrease in acetylcholine and an increase in dopamine in the brain as two potential mechanisms.  As we know, antipsychotics alter neurotransmitters and alter neurotransmitter concentrations, which might lead to prevention or treatment niches for these drugs.  It's also postulated this class of medications have neuroprotective and potential immunomodulatory effects, which may decrease critical illness CNS derangements.  When we specifically talk about ICU prevention, I'd like to begin with the PAD-IS guidelines  suggestion of not using haloperidol, an atypical antipsychotic, dexmedetomidine, or an HMGA-CoA reductase inhibitor or ketamine to prevent delirium in the critically ill patient.  I'd now like to focus on some studies, specifically I included randomized trials that highlight specifically antipsychotics, as that was the focus of this specific talk.  The first investigation, a very high-level overview of these studies include with Hakeem  and colleagues evaluating cardiac surgery patients greater than 65 years old with subsyndromal  delirium and randomizing them to receive either risperidone or placebo enterally with a primary outcome of occurrence of delirium.  This study found that patients that receive risperidone had a less frequent occurrence with delirium than placebo.  In another study by Wang and colleagues in non-cardiac surgery patients, haloperidol and placebo were the two agents where patients were randomized to with a primary outcome  of incidence of delirium. There was, again, a positive outcome with patients receiving haloperidol versus placebo having less of an incidence of delirium.  In 2018, Kahn and colleagues conducted another randomized trial in thoracic patients where haloperidol or placebo were compared with an incidence of delirium being the primary  outcome, and there was no difference between the two groups. Switching gears slightly to medical and surgical patients in the MIND trial by Dr. Drodd and  colleagues, mechanically ventilated medical and surgical patients were randomized to either  receive enteral haloperidol, enteral zeprazidone, or placebo with a primary outcome of delirium-free coma-free days. This investigation found no difference between the groups.  In the HOPE ICU trial, again, adult mechanically ventilated patients were randomized to haloperidol  or placebo, and patients were able to receive open-label antipsychotics with a primary outcome of delirium-free coma-free days. Again, no difference was found between the groups.  A third investigation in ICU non-invasively ventilated patients randomized patients to  receive a continuous infusion of dexmedetomidine or a continuous infusion of haloperidol versus  placebo, again, with an incidence of delirium being the primary outcome, and no difference  was found between the haloperidol or placebo group, however, dexmedetomidine did show benefit.  And finally, two more, a study by Al-Khatib and colleagues randomized medical and surgical  ICU patients with subsyndromal delirium to haloperidol or placebo with, again, the incidence  of delirium being the primary outcome, and they found no difference between these groups.  And finally, the REDUCE trial, randomizing patients to haloperidol 1 milligram, 2 milligrams, or placebo with a primary outcome of days survived in 28 days.  The haloperidol 1 milligram group was stopped early by the Data and Safety Monitoring Board due to presumed futility.  And when they evaluated the 2 milligram to placebo group with the primary outcome, there was no difference in days of survival.  And a secondary outcome of incidence of delirium, again, found no difference. So to summarize, I know there were quite a few studies, but really wanted to point  out several things when we look at them, is the diversity between the patients included, the interventions, as well as the delirium assessment tools and the frequency of assessments  that were taking place. So globally, there's really no role for antipsychotics for prevention of delirium in the ICU.  There were some perceived benefit in our cardiothoracic and non-cardiac patients or surgery patients,  but as clinicians, we need to keep in mind these patient populations are very different than most of our ICU patients with potentially a lower severity of illness and potentially  less risk factors. So really taking the results of those studies and extrapolating to a medical ICU mechanically ventilated patient should be done with extreme caution.  Also looking at the protocols and length of prophylaxis. If you recall back to the slide with the different affinities for receptors of a specific antipsychotics  with your atypical and typical, there were a variety of agents used in these studies and a variety of different routes, IV versus enteral.  So more work really needs to be done to consider could there be a difference with the different antipsychotics used because of the receptor affinities.  And then finally, could longer lengths of prophylaxis really impact outcomes? The reduced study, the median length of prophylaxis was two days.  So could longer treatments potentially impact the development of delirium? However, when these investigators conducted a subgroup analysis, this showed not to be the case.  So really I'd like to highlight that the decision to use antipsychotics with prevention is not benign.  We put patients at potential risk for over sedation, potential ADRs, and drug interactions specifically with concomitant medications that could prolong the QTC interval.  While most investigations did show that the ADRs were either mild or not statistically different between groups, I would like to caution, again, using a medication is not  without its potential pitfalls. Changing gears slightly to treatment of delirium, I again would like to highlight the randomized trials in this area.  The PAT-IS guidelines suggest not routinely using heloperidol, an atypical antipsychotic, a HMG-CoA reductase inhibitor for the treatment of delirium.  So beginning with an investigation by Jana Skrobek and colleagues in 2004, patients were randomized either to heloperidol or olanzapine, both given entrally, with a primary outcome  of change in delirium index. This investigation did find that over time the delirium index did decrease within both  groups, but there was no difference between the heloperidol or olanzapine group.  In the study conducted by my colleague, Dr. Devlin, which no pressure here talking about someone's study who's sitting right up in front. Adult ICU patients were randomly assigned to either quetiapine  or placebo with rescue haloperidol being allowed within the groups. The primary outcome was time to resolution of delirium. And these investigators found  that patients randomized to quetiapine had a shorter time to resolution than patients in the placebo group. And a study conducted by Bakri and colleagues  of postoperative trauma patients, these investigators randomized patients to either dexmedetomidine on danzotron or haloperidol with a primary outcome of incidence of delirium.  And no difference was found between the three groups. And then finally, the MIND-USA trial recently published in 2018, this investigation is the largest trial  to date randomizing 566 medical or surgical patients that were receiving invasive or noninvasive ventilation. Vasopressors are an intra-arterial balloon pump  with delirium to either haloperidol IV or ziprasidone IV or placebo with a primary outcome of delirium-free coma-free days. And the results of this investigation show  that there were no difference between the three groups. With this being the largest trial to date, it had many strengths that I'd like to highlight just briefly.  It had a diverse patient population and a broad inclusion criteria as well as a large sample size. To note, most of the patients that developed delirium, however,  did develop hypoactive delirium. And these patients received an antipsychotic, which in many individuals' practice, hypoactive delirium patients many times would not be treated.  So while we can say based on the rigor of this investigation that there is most likely no role for antipsychotics with treatment of hypoactive or mixed delirium,  it's still unclear if solely hyperactive delirium would benefit from potential use of an antipsychotic. So to summarize, globally,  antipsychotics really have not been shown to impact delirium or the course of patient outcomes. However, there are still several questions that remain to be answered.  For example, could antipsychotic choice play a role? For example, Dr. Devlin's investigation used quetiapine, where other investigations used  for their atypical antipsychotic ziprasidone. Could different receptor affinity impact outcome? So that's one question I'd leave you with.  And then secondly, what is the impact specifically in patients with hyperactive delirium? Would antipsychotics in specifically that cohort show a difference?  But for now, we can specifically say, if we're thinking about using antipsychotics in the delirious patient, that to use it for symptom management, not for treatment  of the actual CNS derangement. And we should consider using it when patients are experiencing symptoms that are bothersome or distressing to the patient, such as our delusions,  our hallucinations, or these symptoms are impacting care, maybe preventing extubation, for example. Or patients become a danger to themselves or other caregivers.  So really thinking of IV haloperidol for the acutely agitated patient, where there may not be enteral access or the patient might not be able or willing to take anything enterally.  Or using atypical antipsychotics if the risk versus benefit of IV haloperidol is the case. And again, keeping in mind, use of antipsychotics are not benign.  We do have the risk of ADRs, of drug interactions, of over sedation that we must keep in mind. And as my next colleague will be discussing,  the use of nonpharmacologic should always be on the forefront of our treatment regimens, even though sometimes giving a drug is certainly easier than mobilizing.  And if we choose to use an antipsychotic, I really encourage everyone to remember we need to deescalate therapy, as Dr. Devlin mentioned earlier.  It's not uncommon for an antipsychotic that's being started in the ICU, specifically for ICU delirium, to be continued upon patient transfer from the ICU.  One investigation, 80% of patients that had a new start of an antipsychotic continued to receive this medication while they were on the floor.  And about 30% of these patients were discharged home with an antipsychotic. So really, we need to remember to discontinue these agents upon symptom resolution.  Even if the medication is as needed medication, many times these medications are forgotten about. Oh, it's as needed. I mean, they won't get it unless they need it.  Well, maybe the patient's discharged home with olanzapine as needed. So really, discontinuing it when patients leave the ICU  or documenting in the chart when to discontinue this medication if there are ongoing symptoms upon transfer. This will help our patients really be less exposed  to potential long-term side effects of these agents, the potential for sporadic QTC monitoring throughout the rest of their hospitalization, drug interactions with other agents  that may prolong the QTC, and unnecessary costs, not only to the healthcare system but to the patient. So really, in summary, for delirium prevention,  there's currently globally no role for antipsychotics to be used for the majority of our ICU patient population. And then secondly, for treatment,  we should really consider antipsychotics only for treatment management in that acutely agitated patient where they're distressing to themselves or impacting care.  And then finally, if we decide to use these agents, it's extremely important that we remember to discontinue and de-escalate. Otherwise, our patients are going home on agents  that we started in the ICU for a specific ICU reason. And five years down the line, they just keep reordering this medication because no one really remembers where it came from.  So thank you very much for your time. And it was a pleasure speaking. Thank you.  Good afternoon, everyone. Can you hear me? Okay. All right, let me just acclimate myself here to slide. Okay. So, my name is Julianna Jablonski, and I'm thrilled to be here today.  I'm going to try to be really tight with time knowing that we started late and allow time for questions. I have no actual or potential conflicts of interest  in relation to the presentation. And what I'd like to do in the next few minutes is to really talk about are non-pharmacological  interventions effective in the prevention or treatment of delirium in critically ill adults. So, I want to start by looking at this multifactorial model  of delirium, which was first published in 1996 by Inouye and Carpenter. And this is a really important model that is still very relevant today, although it's been over 25 years.  And it speaks of predisposing and precipitating risk factors. And that a person's vulnerability towards delirium is based on the predisposing risk factors.  Someone with high vulnerability may require a less noxious insult or exposure to, say, one risk factor versus someone with low vulnerability may need repeated or noxious insults  to start developing delirium. The key implication here is that targeting one single risk factor is unlikely to resolve delirium. And when you're developing a delirium prevention  and mitigation program, it's the multi-component interventions that are most likely to be most effective. So, on your right-hand side, you can see the predisposing risk factors,  which we're all very well aware of, that have the highest rate of exposure to someone developing delirium. You can see this is for both ICU and some of the non-ICU.  And then it's the modifiable risk factors, which are the precipitating, sometimes we call them triggers. These are the things that we feel are modifiable.  And when you're creating a program of prevention and mitigation, these are the trigger points that you can start to focus your interventions.  So, I really want to talk about Dr. Sharon Inouye, who is really internationally known as a geriatric researcher as well as delirium researcher.  And she was really the first person to develop and validate the confusion assessment method, which Dr. Wes Ely then, in 2001, adapted this model  to the ICU and for the mechanically ventilated patient. Dr. Inouye developed this way of identifying delirium, and it was targeted towards the geriatric,  non-critically ill patient population. And she then converted this into what's now known as the Hospital Elder Life Program, which she established in 1993.  And the essence of this program is identifying delirium using the CAM, assessing for risk factors, and then applying multi-component interventions based on those risks.  And you can see the interventions listed there. And I also just want to highlight Dr. Inouye also since 2013 has what's called Cedartree, which is a Center  of Excellence for Delirium and Aging Research. So, when you look at all of the literature in the non-critically ill geriatric patient,  using this similar model of assessing for risk factors and applying key interventions, as you can see many of those listed there, to get into some of the detail.  And I like to do that throughout the talk because I think sometimes when people say non-pharm interventions, it feels very large and not to the granular level.  But I'm hoping to show you a little bit at the granular level of what we can do. So, this meta-analysis of 14 studies looking at incidents of delirium and falls showed  that there was a 53% odds reduction of developing delirium and a 62% odds reduction of falling when you were following this type of delirium prevention and mitigation program.  And then in 2018, another meta-analysis showing all of the studies were around implementation of the HELP program, 12 studies included for delirium outcomes.  Again, incidents of delirium and incidents and falls still consistent with that initial trial. Showing a really what I think profound reduction  in incidents of delirium and fall rate. So, what we know is that primary prevention with non-pharmacologic interventions is widely accepted  for delirium in older non-critically ill patients. As I've been working in ICU delirium for, I mean, 14 years now, I'm really looking at the work of Dr. Inouye  as extremely innovative in laying the foundation. And I really think that there's a lot to learn from our non-critical care colleagues. So, what I really want to talk  about are non-pharmacologic interventions effective in critically ill adults. And when you look at this idea of a multi-component intervention,  really you could say the ABCDEF is a multi-component protocol. And you can see Ballas that published in 2014 which was really the first study showing the delirium outcomes  around this bundle implementation. And there was reduction in prevalence, duration, and your odds reduction of turning delirious were all very  positive and statistically significant. And then in 2017, Barnes-Daly looked again at the implementation of the A through F bundle across a seven hospital in California  and over 6,000 patients. And looked at it in the concept of partial and complete total bundle compliance. And again, you're seeing a reduction in delirium-free coma-free days.  Just to point out, many people will say, why is it that partial bundle compliance is leading to a larger reduction in delirium or increase in delirium-free coma-free days?  The reason being when you look at total bundle compliance, it's two variables, either yes completely or even if one element is missing, it's a no.  So, the idea is that partial bundle compliance is probably a better show of the benefit of that protocol. So, and lastly, you're aware of the ICU liberation study  that was just published in 2019. A large QI collaborative of over 15,000 adults. Again, the interventions of the A through F all showing a dose relationship,  including delirium benefits. So, I start with that because that's what we, many people may not think of that as a multi-component delirium prevention protocol. But it is.  And when you look at other studies, which aren't really around ABCDEF bundle implementation, but here's work that was published in 2012 by Colombo, which did a pre-post study.  Phase one you can see where they started to do the delirium assessment, the targeted lidar sedation, the SBDSAT, which is really the ABCDEF bundle to some degree.  And then in phase two is when they started to put in these non-pharm interventions for delirium treatment and prevention. And they used things such as reorientation, wall clocks,  newspapers, books, you can see. And it showed a statistical significantly reduction in delirium occurrence. But what I want to point out here, when through the statistical analysis,  the applied reorientation strategy here in this table is what they found was the strongest protective predictor of delirium prevention.  And what is really interesting, so it are these six questions of the who, what, when, where, why, how that one time per shift someone, mostly the nurse, but it didn't have  to be, had to go through a script of all of these questions with the patient, which I think is just a really interesting way.  We all say that we might do that, but this was done in a way to know that it was consistent and reliably done on every shift. So this is another study from 2015,  a randomized controlled trial. And this is a group that took the help interventions very similar to looking at delirium risk and screening and then putting in interventions based  on your targeted risk factors. The difference here is a research team also intervened daily on all of the patients for about 10 to 20 minutes  to ensure that all of the interventions were followed through and implemented. This study did not show an improvement in incidence of delirium.  But I'm highlighting it here because of the limitation is that delirium before inclusion was not an exclusion. So I think it's hard to make the final conclusion  that this was not effective. So another multi-component protocol that, again, I'm highlighting that it's the essence of the A through F bundle.  They did every four-hour delirium screening, sedation, minimization, early mobilization. And they highlighted this, the non-pharm piece  as being give your patient more with music, opening the blinds, reorientation, and cognitive stim, and an ear and eye protocol. And, again, you can see all of the reduction  in median time spent. Delirious converted into percent of ICU length of stay, which you would say is about a 50% reduction, as well as incidence of delirium was also reduced.  So, and then this work was just published in 2019 by Bannon, which is a pretty large systematic review and meta-analysis of 15 randomized controlled trials  around non-pharm interventions. I have all of them listed here. Four of them highlighting more about bright light therapy, and two specific around occupational therapy.  So as you can imagine, it's the clinical heterogeneity that left for little statistical power for each individual component.  And I'll speak to this a little bit later, how when you try to evaluate an individual component, the studies are lower quality, small sample sizes, and it's really hard  to evaluate what a single intervention is doing for delirium. I also want to highlight when you look at large studies or large numbers of studies,  researchers measure delirium many different ways. Sometimes it's delirium duration. Sometimes it's time spent delirium. Sometimes it's incidence. Sometimes it's delirium free days.  And it's really hard to pull these results into the meta-analysis. When you look at the studies that I've highlighted here, what came out as the most beneficial is this concept  of family voice, which I'll talk to in the next slide, which I find really intriguing. And then of course what we've known, physical  and occupational therapy continues to be highlighted as a benefit for reduction in delirium.  So this is a study that it was a pilot randomized controlled trial that was published in 2017. And I really, for some reason, I just connected with this  and thought that this has to be done on a larger level. So it was just an N of 30. But what this group did is they, on admission to the hospital,  a family member would record a message, and it was four minutes or less, using this script and answering all these questions, speaking to the family member by following all  of these questions. Then they would have a non-family member do the script, and then they had a group that didn't receive any recorded messages.  And they delivered the recorded message over three days every hour during all waking hours. And I think it's pretty interesting, and you can see  that even with that small sample size we were showing reduction. But I thought, hmm, that's something  that is really interesting and maybe grab on that as a larger study. So, and this was also a study that I want to highlight that I also thought was really intriguing.  This is in the elective cardiac surgery population where it was called a psychological educational intervention. This was done by a single surgical resident who was  trained by a psychiatry consultant. And it was pre-op procedure that he would spend the surgical resident spent an hour with the patient really.  At the end of the day you can see what they had done, but developed therapeutic relationship and reviewed delirium as a potential complication.  And then post-op spent another hour sitting with the patient connecting with empathy and providing emotional support. And you can see the significant reduction  in post-operative delirium from 35 to 12.2%. So, whenever you talk about delirium and non-pharm interventions, you have to mention sleep. And I'll do this briefly.  And just in the way that sleep and delirium, as you know, is very complex, could be individual talks on their own. But just in brief, why I want to bring this up, because I want  to talk about the bright light therapy from the Bannon metabolic analysis review that showed that there was no benefit to that.  But what do we know about sleep in critically ill patients? It's fragmented, abnormal circadian rhythm, decreased time in REM and increased time in late sleep.  But does ICU delirium affect sleep? And yes, we believe that there's an association of less REM and increased daytime sleep. But there's also an association believing that patients  who are delirious have a greater disruption of the circadian sleep cycle. Or, yeah, the circadian sleep cycle. And that quality of delirium, I'm sorry, quality of sleep  and is reduced with ICU delirium. And that to date, we know that there's no causal effect. But we believe the hypothesis is that sleep deprivation will contribute to ICU delirium.  So if you're treating sleep deprivation, it is really your non-pharm intervention to treating delirium. And this is done through bright light therapy  as a single individual component intervention. The idea is to get to the disrupted circadian rhythm. And just in short, it's the suprachiasmatic nucleus  which controls our circadian clock. And the way to stimulate this nucleus is through light. But it has to be through perceived light  through open eyes that then will stimulate the nucleus which is then believed to help reset or improve a disrupted circadian rhythm.  So when you look at this study that was published in 2016, it was this concept of using dynamic light application. And this was a randomized controlled trial with over 300  in each of the two groups. And what they did is they used normal hospital lighting. And honestly, I probably couldn't explain it to great detail, but what I can tell you is  that in normal lighting, max lux is 300 and 3,000K. But in this bright light fill-up system, you reach levels of 17,000 lux and 4,300K.  So but what the group did is from 7 a.m. to 10.30 p.m., these were automatic lighting system with gradual increases through wakeful periods to resemble the daylight,  the shifting into night, and to give the patient exposure to this light system. And what it showed is that there was no difference. So and it's a little disappointing  because you think, you know, bright light therapy. And when you look at, you know, why did it not show a difference? Possibly because of the patients included.  If it's patients that are sedated, mechanical ventilation, their eyes aren't open, they may not be reaping the benefits of this light therapy.  But then in another study here that was published in 2013, a before and after. This was a group that you can see in picture A shows the environment  of their ICU before it was going through mass refurbishment. So they did a before and after with over 500 observation days and looked at the incidence of delirium of patients  in this 32-bed mixed ICU before renovation and then in this beautiful room after renovation. And simply by just changing the environment showed  that there was a decrease of .4 days in the duration of delirium for patients now in the new environment. So when you look at innovations in the environment,  and this I think is a really great piece of literature for you to look at, Lutz 2019 published this. But starting with patient reported  through quality work done, published in the Dimensions of Critical Care, showed that top reported patient stressors in the ICU is sleeplessness  and constant exposure to noise and light. So Lutz published, how do you view your environmental space as more of a treatment tool? And these are some of the more innovative things  that I thought were suggested that when you're developing a new ICU, of course, use sound absorbing materials. But I thought what was really interesting is there's,  our ICU studying now, if you have a microphone that is set at the patient's perspective, and it is showing what the patient is hearing in decibels  at any time of the day, and then this is displayed to all of the providers to see exactly what the noise level is of from the patient perspective, which is really interesting.  The idea of having noise canceling headphones right there and ready, having directional speakers for music so you're not disrupting others, as well as creative space  for mobility, there's many of us, I know that through our ICU design that really struggle with the ability to have a space for mobility because everything is extremely tight.  So, what can ICU teams do today? So, I think to create an ICU that makes it easy to provide non-pharm interventions. I think non-pharm interventions is a large term  that we say it's good, let's do it, but if there's not a structure and process in place, it might not reliably and consistently be done on every patient.  And I also say no to drugs, unless absolutely necessary. And I'll just show you a finish-up here in my own Penn Health System where I've implemented a nurse-driven delirium program.  And I think when you have nurses assessing for delirium, there has to be what do I do with that, what's our process? And I do believe in the Sharon Inouye model of looking  at risk stratification and then taking your risks and building in non-pharmacologic interventions and using a multi-component, multi-disciplinary team discussion.  So, in conclusion, as you can see, there's inconsistency in the description and dosing of non-pharm interventions. And honestly, what I can tell you is  when you do the multi-component bundles, none of us can tell you directly which one of those things are helping with delirium minimization.  But when you see, when you do the single intervention studies,  they're not necessarily one of the stronger tools of showing reduction. So, research trials studying multi-component interventions leave the uncertainty then  to which interventions result in the effect. But I think we can conclude that most non-pharm interventions are part of just giving good care. And we shouldn't stop doing that.  And the research needs to develop more in this area. So, thank you. Thank you.  Thank you so much. I apologize. Because we're out of time, we can't have a question and answer at the podium. But our speakers are here.  And they are willing to answer your questions if you are able to come up and speak to them. Thank you so much.

Video Summary

The speakers discussed the use of non-pharmacological interventions for the prevention and treatment of delirium in critically ill adults. They emphasized the importance of multi-component interventions that target both predisposing and precipitating risk factors. The ABCDEF bundle, which includes many non-pharmacological interventions, was highlighted as an effective protocol for delirium prevention. Other interventions discussed included reorientation strategies, family voice recordings, psychological educational interventions, and sleep hygiene measures. The speakers also mentioned the importance of creating an ICU environment that supports non-pharmacological interventions, such as reducing noise and providing space for mobility. They acknowledged the challenge of evaluating individual components of non-pharmacological interventions and highlighted the need for more research in this area. Overall, they concluded that non-pharmacological interventions play a crucial role in the prevention and treatment of delirium in the ICU setting.

Asset Subtitle

Neuroscience, Pharmacology, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Identify sedatives and analgesics appropriate for prevention and treatment of ICU delirium
Recognize situations and specific patients for whom antipsychotics are appropriate treatment for ICU delirium
Design environmental and behavioral approaches to decrease delirium incidence in the ICU

Ideal Sedatives and Analgesics to Prevent and Treat ICU Delirium

John W. Devlin

Antipsychotics in the Treatment and Prevention of ICU Delirium

Pamela L. Smithburger

Nonpharmacologic Therapies to Prevent and Treat ICU Delirium

Juliane Jablonski

Meta Tag

Content Type Presentation

Knowledge Area Neuroscience

Knowledge Area Pharmacology

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Delirium

Tag Psychopharmacology

Year 2020

Keywords

non-pharmacological interventions

delirium prevention

multi-component interventions

ABCDEF bundle

reorientation strategies

family voice recordings

psychological educational interventions

sleep hygiene measures

ICU environment

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.