All right, thank you everybody for coming today and to our session on pain, agitation and delirium in the ICU prevention and management. I would like to, we're going to start, we've changed the order a little bit. Now I'm all confused because, I'm going to say your name, Dr. Torbek is going to actually speak to us about pain today in the ICU and Dr. Gerard is actually going to speak to us about delirium. So it's, it was a little switched from what the initial slide show mentioned and so I'd like to welcome Dr. Torbek to the podium today to start us off with speaking about pain in the ICU. Good morning everyone, thank you for this opportunity to be here today. So my name is Heather Torbek, I am a clinical pharmacist at the Cleveland Clinic in Cleveland, Ohio. I practice in the medical ICU and today I'll be reviewing pain management strategies in the ICU. So in terms of pain management in the ICU, it can be very challenging because pain patterns are highly individual, it can be acute or chronic, it arises from different sources of pain, so somatic or visceral and patients have subjective perceptions of pain and different tolerability. Patient surveys and literature estimate that at rest over 50% of patients experience pain and over 80% of patients experience pain during invasive procedures. According to patient surveys, the procedures associated with the greatest amount of pain include arterial catheter insertion, chest tube removals, wound drain removal, turning and repositioning, and tracheal suctioning. Studies have also identified baseline risk factors for perceived pain. These include younger age, patients who have had prior surgeries, history of anxiety or depression, patients who require support for activities of daily living, patients who have higher number of comorbidities, and female gender. So in this presentation I will be reviewing some of the pain management strategies that we have available for patients in the ICU. Appropriate pain assessment is important for critically ill patients to ensure that they do not experience consequences of undertreated pain, but also to ensure that they don't experience overtreated pain. One thing to remember about pain scores is they are highly variable and patients are going to have different pain thresholds, so it's important that we use validated tools to assess and deliver pain management strategies. If the patient can reliably self-report pain, that should be our gold standard. And the numeric rating scale and the visual analog scale are both validated tools and easy to use. When patients cannot reliably report pain, then we have a few behavioral assessment tools that are available to us. The behavioral pain scale and the critical care pain observation tool are both highly valid and reliable across medical, surgical, and trauma ICU patients. The behavioral pain scale ranges from 3 to 12 and it assesses facial movements, movements, and ventilator synchrony. The critical care pain observation tool ranges from 0 to 8 and also includes muscle tension in addition to those assessments from the behavioral pain score. In terms of proxy reporters like family and friends, it's been reported that family and friends can reliably identify that patients are in pain up to 75% of the time. However, they tend to overestimate pain by 50% when compared to using an actual pain assessment tool. And then finally, vital signs such as heart rate and blood pressure respiratory rate should only be used as proxies to go and then use a validated pain assessment tool to assess pain. The primary guideline for the pain management in critically ill patients are the PADIS guidelines, which were last updated in 2018. They recommend using a non-pharmacologic management approach first and then thinking about pain management strategies that include pharmacologic interventions. So first of all, minimizing painful stimuli. Are there any tubes, lines, and drains that can be removed? Can we coordinate lab draws and procedures so that we're only impacting patient pain once or twice a day versus every two hours? Also things like music therapy, massage therapy, warm and cold compresses, relaxation techniques can also be considered. In terms of physical therapy, it may be acutely painful for patients, but it does help reduce the incidence of chronic pain in patients. So in terms of pain management in the ICU, it should not be a one-size-fits-all approach. Before making recommendations for pharmacologic pain management in the ICU, I think about a patient's past medical history. What comorbidities do they have? Do they have a history of chronic opioid use dependence? Have they already had baseline pain regimens established? What is that pain regimen? So if a patient is taking 100 milligrams of oxycodone a day, it's probably not appropriate for me to only recommend a starting dose of 5 milligrams. Also I like to consider the pain type. Is this neuropathic pain? Is this localized? Is it diffuse? That can help me think about what pain options or pain treatment options would be most appropriate. What routes are available to me? The pain medications come in a variety of routes, but what routes are available to me with this current patient? And then thinking about PK and PD for the patient. Is there any end organ damage, such as liver and kidney? And how will those affect the management approach for these patients? And then constantly re-evaluate using those pain assessment tools that I referenced in a few slides prior. So as a general overview, the PADIS guidelines state that all available IV opioids, when titrated to similar pain intensity endpoints, are equally effective. As a whole, these medications have a class effect in terms of adverse effects, including constipation, hypotension, and delirium. I would recommend selecting an agent based on patient-specific factors and PK and PD targets. Fentanyl tends to be one of the more commonly used opioids in this class of medications. It tends to be more hemodynamically neutral compared to the other opioids. It has less hypotension. It does, however, have a context-sensitive half-life, so it will accumulate with higher doses and longer durations of use. It's highly lipophilic, so it can accumulate in obesity. It also can be associated with chest wall or cause chest wall rigidity, which can be masked as ARDS at higher doses. It has CYP3A4 interactions. And it does have an affinity for serotonin receptors, so it can cause serotonin syndrome. Serotonin may be a better option for patients following fentanyl use, particularly in patients in ECMO, since it is less protein-bound and less lipophilic. Morphine tends to be of limited use in the ICU because it does have more histamine release, which results in hypotension and bradycardia. And it does have metabolite accumulation, particularly in patients with end-stage renal disease, so it leads to CNS toxicity and can cause cholecystitis. Remifentanil does have—it's more potent than some of these other opioids. It's rapid onset and offset, but its use is generally limited by adverse effects and cost. And finally, I included oxycodone, which unfortunately is only available as an oral dosage form but may have a greater prolonged effect compared to some of these other short-acting IV opioids. It can accumulate in renal and liver dysfunction, particularly liver dysfunction. Notably, I've left off a number of long-acting opioids, including methadone and buprenorphine. These may be used later in the course of a patient's ICU stay as you're weaning from an acute-to-chronic pain management strategy. In terms of opioid comparisons, there is limited evidence evaluating the comparison of opioids in a critically ill patient population. Since the publication of the PADIS guidelines in 2018, there have been a number of drug shortages that have impacted our ability to get some of these IV opioids. And so a number of small retrospective studies have been published looking at the comparison of what they did prior to the shortage and what they were doing during the shortage. One of the largest studies was a study of 487 patients that looked at fentanyl versus hydromorphone. They found that hydromorphone was associated with decreased mechanical ventilation duration and no difference in adverse effects such as delirium, ileus, and pneumonia. In a phase two cluster crossover study, fentanyl was compared with morphine. In this study, fentanyl was associated with higher ventilator-free days and decreased ICU length of stay with no difference in mortality and adverse effects. These differences were more highlighted in patients with end-stage renal disease, which makes sense with the potential of accumulation of morphine in this patient population. And then finally, in 2022, there was a meta-analysis that compared fentanyl to all other opioids, including IV opioids like hydromorphone, morphine, and oxycodone was included as well. In this study, there was no difference in mortality, ICU length of stay, and mechanical ventilation duration, and adverse effects. So I think ultimately, with no new data and no great data comparing the opioids, ultimately we should be selecting IV opioids based on patient-specific factors, considering PK and PD, and really thinking about the type of pain that we're trying to treat for patients to identify the best agent that would be most appropriate for patients. Although opioids tend to be our workhorse in the ICU, there are a number of non-opioids that I think we should consider. These non-opioids, particularly acetaminophen, have been shown to reduce the amount of opioids that we need to give patients to target pain goals. Acetaminophen can also be used for mild or moderate pain as monotherapy. NSAIDs, although they do come in a number of dosage forms, they should not be routinely used in critically ill patient population due to their increased risk of bleeding and acute kidney injury. The gabapentinoids, like gabapentin, pergabalin, and carbamazepine should be considered if neuropathic pain is available. However, these medications are only available as oral dosage forms, so your ability to administer them to all patients in the ICU may be limited. And finally, I've included lidocaine, but really should not be used for routine use, particularly the IV formulation in the ICU population due to its risk factors for CNS toxicity and arrhythmias. The guidelines also suggest considering other potential adjunctive agents, such as volatile anesthetics like isoflurane, nitrous oxide, tricyclic antidepressants like amitriptyline, SNRIs such as duloxetine, and other topical agents. Again, there is limited evidence for using these adjunct agents and probably have a very limited role in our ICU population, but are available if appropriate for your patient. So at this point, you may be thinking, all of this sounds like a lot of polypharmacy. Is there just one agent that I could prescribe for both pain and analgesia management in our patient, or pain and sedation in our patients? So maybe ketamine and dexmedetomine are the solutions. Ketamine is an NMDA receptor antagonist and has been both analgesic and sedative properties. It has a number of unique adverse effects, including emergence reactions, increased secretions, respiratory depression, and potentially ICP increases. The PADIS guidelines currently recommend using ketamine as an opioid sparing agent in the post-surgical patient population. In terms of dexmedetomidine, it's an alpha-2 agonist, which has, again, both analgesic and sedative effects. It can safely be used in renal impairment, but its use should be used with caution in liver dysfunction, as it can result in severe accumulation. Unlike ketamine, it does not have respiratory depression. It does not cause deep sedation. Its use is generally limited, though, by adverse effects, including hypotension and bradycardia. And with long-term use, you can have issues with withdrawal. So ketamine was actually approved for use by the FDA in 1970, but data evaluating its use in the critically ill patient population is very limited. In a recent survey of ICU practitioners, 55% of respondents reported never or infrequently using ketamine preferentially for pain management, and over 60% said infrequently or never using ketamine for sedation. Many of the most common barriers for ketamine use include its adverse effects, lack of familiarity, and lack of evidence. And to date, there is limited evidence using ketamine, particularly in a non-surgical population and particularly as monotherapy. In 2022, a meta-analysis was performed for continuous infusion ketamine versus non-ketamine regimens for analogous sedation. They included 19 studies for a total of over 2,200 patients. The ketamine was associated with decreased opioid consumption compared to other regimens for analogous sedation. However, it did not reduce sedative consumption, duration of mechanical ventilation, median ICU length of stay, intracranial pressure, or mortality. I think at this time, based on this data, it probably is best to reserve ketamine as an adjunct to our other therapies, particularly in a non-surgical patient population, and thinking about using it when you have gone through other medications that are not meeting your pain and sedation goals. Unfortunately, the data for dexmedetomidine as a dual pain and sedative agent is even more limited, and the benefit to date is most likely related to post-operative ortho and cardiothoracic surgery patient populations. The benefit of using dexmedetomidines for its proposed mechanism for analgesia is related to its inhibition of neurotransmission through the posterior horn of the spinal cord, thus inhibiting pain signals to the brain. So although dexmedetomidine has not been directly studied as an analgesic in a non-surgical patient population, you can see from the data here from the CEDCOM, Midex, Prodex, SPICE 3, and MENS 2 trials that the amount of fentanyl consumption when using dexmedetomidine compared to its comparator was no difference between dexmedetomidine and the comparator. The only difference was in the MENS trial in which dexmedetomidine was compared to lorazepam. Dexmedetomidine patients actually received four times the amount of fentanyl than patients in the lorazepam arm, likely because patients were more lightly sedated and could maybe better report pain. But I think from this data, the use of dexmedetomidine as a pain agent in a non-surgical patient population is probably very limited. The adverse effects that I have listed on this slide are potentially either due to under or over treatment of pain, but also as a direct result of the pain medications that we provide to patients. It's well documented in the literature that undertreated pain can lead to adverse consequences, including PTSD, PICS, impaired sleep, impaired physical function, and chronic pain. These consequences may also have a financial impact for patients in the healthcare system with patients unable to return to work and need for hospital readmissions or new outpatient visits. Although we know that there are consequences for prescribing antipsychotics, sleep aids, and stress ulcer prophylaxis after ICU or hospital discharge, there's less that is known about the continued prescriptions of opioids following an ICU admission. In a 2021 study looking at over 200,000 patients who had an ICU admission, 10.8% of patients went on to develop chronic pain, which was associated with greater mortality six to 18 months following their ICU discharge. Over the 24 months that the patients were followed, the opioid consumption steadily declined but it never returned to baseline. So I think some of the risk factors that they identified for chronic use included older age, female patients, patients with lower education, history of substance abuse, and greater comorbidities. So I think we also need to keep in mind that there are consequences with prescribing opioids in the inpatient setting and not thinking about and reevaluating the pain regimens that we set up for patients to ensure that they're discharged on the right pain regimens. The ABCDF bundle is recommended in addition to a daily ICU checklist to ensure that we do reduce exposure to analgesics and prevent long-term consequences. Studies clearly demonstrate that when an assessment tool is used with a protocolized approach, including education of your healthcare team, that pain scores are improved, duration of mechanical ventilation, ICU and hospital lengths of stay are decreased, and need for sedatives and analgesics are also reduced, and medication adverse effects are also reduced. Some strategies to reduce analgesics in a ICU setting include setting a pain goal and reassessing it frequently, making sure that everyone on your multidisciplinary team is aware of that pain goal, creating patient-specific regimens, thinking about patient-specific factors, PK, PD, and the type of pain that we're trying to treat. Consider a multimodal pain regimen that includes non-opioids as well as non-pharmacologic interventions. Use a symptom-triggered drug administration strategy rather than continuous infusions without reassessing a pain target. Think about sedation and analgesic holidays, medication rotations, and finally, look for opportunities to discontinue or de-escalate pain regimens, especially as patients are transferring out of the ICU. Key takeaways include using the ABCDEF bundle to reduce exposure to analgesia, develop regimens that account for patient-specific factors, including PK, PD targets, and optimizing pain control and minimizing adverse effects, and then finally, monitor for adverse effects associated with agents that we are using for pain management and intervene early. Thank you. Thank you, Dr. Torbek. We're going to actually hold questions because we will come up as a panel at the end. So today I'm tasked with talking to you about anxiety and agitation prevention and treatment. I have nothing to disclose. And so the learning objectives for today are understanding what the risk factors for anxiety in the ICU are, recognizing the causes for agitation, understand how to assess agitation, and learn what we can do to address this agitation and anxiety in the ICU. So I probably don't need to define anxiety because I'm sure we've all felt it, but it's defined as a feeling of worry, nervousness, or unease, intense worry about something with an uncertain outcome. And so you can imagine being in the ICU is something that's going to cause anxiety. That being said, there's not actually a lot of data that looks at anxiety in the ICU. We know that prevalence data is limited, and what we know about it usually comes from intervention studies where they happen to ask patients, are you feeling anxious? Or from qualitative studies, so no studies specifically based on anxiety. But those studies do suggest about 40% to 50% of patients report anxiety, and what we do know specifically is that there is a lot of anxiety post-ICU, so about 50% of patients after an ICU stay will experience clinically significant anxiety. So again, really limited data to look at risk factors for anxiety in the ICU, and some of these are actually retrospective studies that looked at patients who reported anxiety and then looked at variables that were documented in their ICU charts. And not surprisingly to anybody, the things that predict whether you're going to have anxiety in the ICU is if you have a history of anxiety, substance use disorder, current prescriptions for benzos, or current prescriptions for opioids. So again, all of those things kind of tie together for your predisposing risk factor. And then again, when you look at ICU-related variables, you're going to see that this all ties together with everybody's talk today. So agitation, delirium, severity of illness if you're on mechanical ventilation, and again, if you're getting benzos or antipsychotics. So again, it's kind of all confounded together, these patient-related variables prior to the ICU and what happens in the ICU to increase your risk for anxiety. So we don't typically measure anxiety in the ICU. There's no guideline that says to measure anxiety in the ICU. There are some research studies that talk about tools to use to measure anxiety. And so on the bottom, you can see a FACES scale. It's a visual analog scale, similar to what you use to measure pain. And that can be used for anxiety. And then the other thing that's been used in research is something called the State Anxiety Inventory. And there's questions to answer. And so you can imagine that these have to be done on people who aren't delirious, who are awake and can communicate. And so you can see that we don't often measure anxiety in a structured format, but we get it because they tell us they're anxious. The families say they feel that they're anxious. And the nurses hear it a lot from patients about how anxious they're feeling. So there's not much to say about anxiety and anxiety treatment. I actually looked, and there's no study that talks about treating anxiety in the ICU. And when you actually look at it, I was surprised, because I'm like, oh, I'm just going to go get data from outpatient studies. And surprisingly, there are no data about treating anxiety in general. If you look, you can find specific small studies about how to treat anxiety in dementia patients or in certain subsets of population. But no studies that have really looked at it in a systemic fashion or looked at medication use or other things to treat anxiety. So turning to agitation, which we all see in the ICU, again, agitation is a nonspecific response to a variety of stimuli. And all of these things that happen in the ICU, physiologic abnormalities, pain, anxiety, delirium. And why we're concerned about it is because it leads to potentially dangerous behaviors, and it impacts our ability to care for the patients. So there is a recommendation to assess for typically sedation, but the sedation scales also measure agitation. So what is used most frequently and most reported in the literature, although there are others, is the Richmond agitation-sedation scale. And you can see that scale is centered around zero, which is alert and calm. And for agitation, it's in the positive zone. So plus 1 is restless, plus 2 is agitated, plus 3 is very agitated, and plus 4 is combative. And so what I typically say is I'm OK with plus 1 if they're restless and moving around in bed. And then we start to talk more seriously about what we're going to do when the patients start to become very agitated. So what I'm going to do is take you through some case presentations to how I think about agitation in the ICU. So I'd like to start with this patient who I took care of. He was a 65-year-old man who had a history of end-stage renal disease, was on hemodialysis, was admitted from the dialysis unit with fever, hypertension, hypotension, and respiratory failure requiring intubation. Actually, I was a brand new attending, and I think this is why this case sticks with me because on morning rounds, the nurse reported that while he was clinically stable overnight, he was really agitated, not following commands, and had received escalating doses of both benzodiazepines and Haldol for this agitation. And so what happened was I went into the room to see the patient, and he would wake and attend to me. And I was watching him interact with the nurse as well, and it seemed like he was attending to her, but he was really agitated, like making eye contact, but agitated, and so you're thinking of all the things. What could this be? So was it his end-stage renal disease? He had metabolic abnormalities, so was it calcium or potassium derangements? Was it is uremia? Was he in pain? Was he having withdrawal, either from alcohol or medication he'd been receiving as an outpatient, which we weren't giving him? Is it something else, another medication that had anticholinergic effects that we had given him that made him so agitated? Was he uncomfortable on the ventilator with the ventilator settings? And so we reviewed the possible etiologies, and then the more I watched the interaction between the patient and the nurse, I was like, can he hear us because he's attending to us, and it seems like we're talking to him, and he's like all over. And she was like, Dr. Pisani, don't be ridiculous. You know, this, you know, he can hear, whatever. She kind of really cut me down, and I was like, okay, I'm a brand-new attending. Maybe not. But I was actually vindicated because the renal fellow came by and said he's profoundly deaf, and he can't hear a thing without his hearing aids, and as soon as we put his hearing aids in and told him what happened, he calmed down. Like he was in dialysis one minute. He wakes up in the ICU with a tube in his throat and is agitated, and then we're, you know, sedating his agitation and not addressing it. So that was a simple thing, and so that lesson, I mean, and it's, if you read the literature about things that cause delirium and agitation, we talk about vision and hearing impairment, but to actually have that experience actually always makes me ask, you know, the family and everybody, do they need a hearing aid? Can they hear? Do they need a hearing aid? They don't need a hearing aid. You know, was their hearing impaired before? So think about causes of agitation, and so here's a schema that talks about this. So we talk about potentially life-threatening, and these are the things you don't want to miss. So gas exchange, metabolic abnormalities, again, thinking about the ventilator on the intubated patients. Is the ET2 malposition? Do they have a tension pneumo? Infection, sepsis, drug and alcohol-related, so intoxication, and we think about intoxication a lot. We also like to think about withdrawal as commonly, and especially for patients who are in the ICU for a while and have been on meds, are they withdrawn? And then ischemia, so myocardial, but also intestinal and cerebral. And then non-life-threatening things, which, of course, the hearing aid was a non-life-threatening reason for his agitation, but if we just kept sedating him for days and not figuring it out, you can imagine the adverse consequences from that, and all of these other things that can potentially make you agitated, pain, inability to communicate, sleep deprivation, urinary retention, nausea, nicotine withdrawal, dry mouth, delirium, and drug reactions. So here's another case. This was a 37-year-old man with prolonged ICU course secondary to COVID. He had been on ECMO. He was decannulated, had a trach place, was in the ICU with ongoing respiratory failure, and had gotten large amounts of IV sedation, and you can see them all listed there. And since he had been there for so long, every attending was trying something different. And every time his sedation was lightened, he became agitated. And so in this case, I came in and I got the story of what was happening, and so I'm like, and I said, well, let's try, and they tried, and he got agitated, and they sedated him again. I'm like, all right, let's try with me in the room. I want to actually see what was happening. And so what was happening was that he became agitated. He desaturated. But when I stood in the room and watched him, when he woke up and started to do some breathing on his own, with the ventilator settings he was on, he was inverting his I to E ratio and becoming really agitated because of the way the ventilator was set. And so once we realized that and adjusted the ventilator, the problem was resolved. And we didn't notice it when he was so sedated. And on the standard vent settings. And so I want to posit for you guys to think about ventilator as a cause for agitation, especially as you lighten the sedation for patients. So is the patient air hungry? So oftentimes, the flow rates are low. We tend to set lower flow rates. Especially in ARDS, it helps improve oxygenation. But if patients are awake, they don't like those low flow rates. They want the breath faster. So think about the way that the ventilator is being set. Oftentimes, I'll try to switch them to a different mode of ventilation. I find that patients seem to tolerate pressure control ventilation where they can set the flow rate and it can vary better. And then other things. If they're double triggering, do they want a larger tidal volume? Are they trying to trigger but not getting the breath? Do we need to change the sensitivity on the ventilator? And then if they're having persistent dyssynchrony, are they somebody that could tolerate a CPAP mode or a pressure support mode? And then obviously, getting them extubated as quickly as possible will fix the whole problem. And so this is another case from COVID. He was a 45-year-old man with prolonged ICU course, again, secondary to COVID. He, again, was trach because he failed multiple extubation attempts. And again, like the former patient, was on large amounts of different sedations. We were trying everything. And again, every time sedation was lightened, he became agitated. And so you could see he was agitated. And you'd go in the room. And I realized, you know, and he spoke Spanish. And so, but again, he was purposeful. When you'd go in the room, he was agitated. But it wasn't like he could make eye contact with you. So he was purposeful. And so we finally determined, he kept trying to say something over and over. And I'm like, guys, he's speaking Spanish. This is not English. And we figured out he was trying to say bathroom. And then I went back. And I looked at the sheet. And he hadn't had a bowel movement in like five days. And so once we started an aggressive bowel regimen and worked on providing time to sleep and started PT, his symptoms of agitation resolved. So again, this was another case of somebody who was constipated. When we woke him up, was uncomfortable. And when we addressed that, the agitation resolved. And so this is a figure which kind of ties in all of our sections today that we're talking about of pain, delirium, and agitation. And I've kind of blacked out my little area, which is agitation. And again, just to reiterate these things of what we're talking about, what are the things that cause it? Anxiety, frustration, physical restraint, ability to communicate, ventilator dissynchrony. And you can see these arrows kind of go back and forth between delirium and pain and agitation. And so they all kind of tie together. So it's a sedative, but the side effects are anxiety, fidgeting, excitability, and insomnia. So I'm just going to say a few brief words about sedation. And so why do we sedate patients in the ICU? So right through our mechanical ventilation, we need to oxygenate them. We need to prevent self-harm. We do it for comfort and easier management. Remembering that it evolved from the operating room, right? And so historically, intensivists were anesthesiologists. And so they kind of took that practice of sedating people in the operating room and brought it into the ICU. And so really thinking about, is our practice in the ICU should be this? Well, I say it shouldn't be the same, but we shouldn't be sedating people to these really deep levels that you would in an operating room unless there's an absolutely good reason. And so why don't we just not sedate everyone? Well, it's not always necessary. And there are studies that have shown that you can get by with minimal or low sedation and that those patients have better outcomes. It can obscure life-threatening problems, side effects. And you heard Dr. Torbik talk about the side effects of the pain medications. And often, it can be treated specifically and more effectively without sedation. And as I said, the less sedated patients tend to have better outcomes. And so that brings us to case four. This was a 26-year-old man brought to the ED, clonging, kicking, after what they thought was an apparent cocaine ingestion. He stood up on the stretcher. He defecated, was placed in restraints, sedated, and intubated. And so, OK, young guy, probably drug overdose. Let's just sedate him, keep him safe, and we'll sort it out. Well, what actually happened, and luckily, somebody thought to get an ABG. After he was intubated, his pH was 6.85, PCO2 was 110, and his PO2 was 15. He actually was having a severe asthma attack. His tox screen was actually negative. And so luckily for him, he was treated with steroids, bronchodilators, and recovered. But you can imagine if someone was just like, this is just a drug ingestion. Let's just sedate the patient. This young man would have died. And so this is always a reminder, again, what are the underlying physiologic abnormalities that can be causing agitation? So acidosis, electrolyte abnormalities, hypercarbia. So I guess it's just the theme of what I'm going to keep saying is look for the underlying cause. So sometimes the solution is simple. So people are itchy. They're uncomfortable in bed. Just being able to take the time to communicate with the patient and being able to try to figure out what the etiology is. And so again, as I'm going to keep saying, assess for the etiologies. Remember that sedation is a nonspecific treatment to prevent or alleviate the agitation. It causes depressed consciousness. For analgesia, as Dr. Torbek talked about, you want to treat the pain. So really assessing how much of the agitation is underlying pain, if you can. And then anxiolysis, if you can treat the anxiety. And so again, back to the whole concept of what are we going to use to treat anxiety. And so there's no good medication studies, as I mentioned in the beginning, for either anxiety or agitation in the ICU. And so thinking about small doses of maybe a benzo or something else, but really trying to individualize it to the patient. And then the other thing we often forget is that we stop meds when patients come into the hospital. The risk factors for anxiety are pre-existing anxiety and use of pre-existing meds. So really looking back at their medical history, seeing if they were on stuff, continuing their medications, or restarting them as soon as possible. Dr. Torbek already talked about the ABCDEF bundle. And I'm going to give it another plug here. Really, it reduces delirium. There's reduced mortality and more coma and delirium-free days. And so I would encourage everybody to use this. As far as another thing that will help with anxiety and agitation is everybody knows what it feels like to be sleep-deprived. And if you've come from the East Coast and have been up every day at 3 AM, you know you start to feel, all right, by the end of the day, I'm like, I need time alone. I'm agitated from all this lack of sleep and stimulation. So really thinking about sleep promotion in these patients, and especially in the patients who are chronically critically ill and have been in the ICU for a while, so really trying to get them on some kind of a sleep-wake cycle. And there's literature to support this. So I am going to conclude here with, again, saying assess causes for agitation, control pain and anxiety, minimize the use of sedatives, and use protocols. So targeting sedation levels using validated sedation scales, using daily awakening protocols, and using the lowest doses of medications for the shortest amount of time. So I'm going to stop here, and thank you for your attention. And I'm going to invite Dr. Girard to the stage. Thank you. OK. Good morning, everyone. I'm going to speak about strategies to prevent and treat delirium. And I'm going to try to move pretty quickly because, as Dr. Pisani said, we wanted to leave some time for questions at the end. And if we run over the hour-long session, I can't speak for you two, but I'm willing to stick around afterwards. I'm Timothy Girard. I'm an intensivist at University of Pittsburgh, and you see my disclosures there. I have three learning objectives. The first is to recognize the negative effects of delirium on clinical outcomes. The second, to briefly discuss how to diagnose delirium, specifically in the critically ill patient population. And then finally, to discuss some strategies for prevention and treatment of delirium. So what do we know about the outcomes? And even before we look at outcomes, we have to know how often delirium occurs in this patient population. Well, I think most of you realize, because you see it in your own practices, delirium occurs frequently. These are just a few of the many studies that have examined the prevalence, overall prevalence, of delirium in critically ill patient populations. And when you look specifically at those in lighter blue here, which are the studies that limited enrollment to mechanically ventilated patients, you can see that the prevalence in that population ranges anywhere from 50% to 80% of patients will develop delirium at some point during their ICU stay. But what does that mean for these patients? If you look back now 20 or 30 years ago in the literature, it was often just called ICU psychosis. It was thought to be sort of a nuisance, and it would resolve when the patient was no longer critically ill, no big deal. It turns out it is a significant factor in terms of associating patients with adverse outcomes. There are numerous adverse outcomes that are worse for patients who experience delirium. These include the duration of mechanical ventilation is longer for patients who are delirious than those who are not. And every one of the studies I'm about to show you adjusted for multiple potential confounders. So it wasn't just a matter of, well, older patients take longer to get off the ventilator and they're more likely to get delirium. Actually, when you adjust for things like older age, higher severity of illness, and other risk factors for delirium and duration of mechanical ventilation, you still see this association. Patients who are delirious, and especially those who are delirious for multiple days, tend to stay on the ventilator longer than those who are not delirious. The same is true for mortality. These are data that were published by Dr. Bazzani, and we see that those who develop delirium have higher mortality than those who do not after adjusting for confounders. And not only developing delirium at some point, but the more days that these patients were delirious during their critical illness, the worse their overall risk of mortality was. Similarly, the duration of delirium also predicts long-term cognitive impairment in patients who survive critical illness. So if you get off the ventilator after being delirious and you survive, you're still not out of the woods. If you were delirious for numerous days compared with those who were delirious either not at all or for only one or two days, you had a significantly higher risk of persistent cognitive impairment up to one year in this study as well as several others. So this has been repeated in multiple different studies, different patient populations. Patients who are delirious during their critical illness, especially for numerous days, have worse long-term cognitive impairment than those who do not experience delirium. So with all of that in mind, what is it that we can do to prevent and treat delirium? Well, before we answer that question, we have to actually be able to recognize it. And so what do we know about assessing patients for delirium and diagnosing it in the ICU? It's a challenging population. And I should define delirium briefly to say it's a disorder of both consciousness and cognition. So and these symptoms occur acutely and often fluctuate over time. So you need a change in level of consciousness as well as abnormalities in cognition. Just one or the other alone are not delirium. But when you have both, that is what we're describing as delirium today, according to the American Psychiatric Association. In this study done in the Netherlands by Peter Strunk and colleagues, they looked at numerous patients who were in the ICU and they asked the question, how often did physicians recognize delirium? And every one of these figures here represents a patient day in the study. And those with the dark blue were the number of patient days during which a physician recognized or documented that they recognized delirium. I guess it's possible they recognize it and didn't write it down. But in this study, if they wrote it down, they got credit for recognizing it. The nurses were a little bit better. The blue plus the purple were the number of patient days that nurses recognized delirium. But all of these orange days were patient days that delirium was actually present according to a validated assessment conducted by a research staff and yet was overlooked or missed by the clinical staff. So when you put this all together, they found that the sensitivity of physicians and nurses when just using their routine clinical judgment and no sort of structured or validated approach to assessment for delirium in this ICU patient population, the sensitivity is quite low. This actually wasn't very surprising. The same thing has been shown in other hospitalized patient populations. Sensitivity of clinical judgment for delirium is not very good and it's actually even worse in the ICU. I think the main reason for that is because we often think about the patients that Dr. Pisani described, the agitated patients, as being the delirious patients. But we tend to overlook many other patients who have delirium because they have hypoactive delirium where they have a reduced level of consciousness and changes in cognition and those are the manifestations of delirium for those patients. So how do we detect delirium in the ICU? Well, there have been multiple studies that have evaluated the sensitivity and specificity of tools that were designed specifically to detect delirium in ICU patient populations. That means not only patients that were in the ICU but patients who were sedated often in these studies, although not heavily sedated because if a patient's sedated to the point of being unresponsive to verbal stimuli, then they can't be assessed for delirium at that time. But they could be moderately sedated, sedated such that when you speak to them, they have some sort of response. Those patients could be evaluated with these tools and many of these patients were mechanically ventilated and therefore nonverbal. They could nod their head, they could shake their head, they may even be able to write but they can't speak to you in response to your questions. These tools have been assessed in those patient populations and found to be valid. There are two that are recommended by the Society of Critical Care Medicine, clinical practice guidelines that we've heard about today. One is the confusion assessment method for the ICU. I'm not gonna go through these two tools in detail because we simply don't have enough time this morning but you can look this up. It's pretty easy to find online. The Society of Critical Care Medicine guidelines include the tool and ICUdelirium.org includes the tool. This is the confusion assessment method for the ICU and the other one that's been well validated in this patient population is the intensive care delirium screening checklist. Now let me go back just briefly. You can see on the right hand side here that in order to be considered positive for delirium on the confusion assessment method for the ICU or the CAM-ICU, you have to have the first two features, that being a change or fluctuation in mental status and inattention and then either feature three or four, altered level of consciousness and disorganized thinking. Now it turns out that most patients who are delirious in the ICU have all four but you don't have to have all four features and then on the intensive care delirium screening checklist which looks for eight different items in a checklist fashion over the course of a nursing shift, you have to have at least four or more of these items to be considered quote positive for delirium. Now that you know how to diagnose delirium, what can we do about preventing and treating it? Well, you've heard a lot about this already this morning. The ABCDF bundle, often referred to as the ADEF bundle, is probably the most evidence-based approach to treating delirium or rather preventing delirium and once it's occurred, trying to reduce the occurrence of delirium in ICU patient populations. It certainly is not the entire answer for every patient. There are many factors that are specifically important for certain patients that you need to identify. You heard some case examples of patients that were agitated and may have also been delirious that had specific reasons for being delirious so anytime a patient is delirious in the ICU, you wanna ask what potentially modifiable risk factors can I identify? In addition to that, you wanna think about these major areas of risk that are included in the ADEF bundle. Pain is a risk factor for delirium. Oversedation is a risk factor for delirium. The specific medications that you use when you sedate patients can increase risk for certain patients of delirium and then just evaluating those patients so that you know when delirium occurs and when you can look for something that may have triggered it is important. Mobility and family presence. Now I'm not gonna in the time I have go through the evidence for each of these six items as it relates to delirium but I wanna highlight a few components. We've known for a long time now that benzodiazepines in a dose response fashion increase risk of delirium and this is after adjusting for numerous potential confounders. It's not just that well, patients who are agitated and therefore maybe were delirious in the first place get higher doses of benzodiazepines. It's also that if you receive a higher dose of midazolam or lorazepam in the ICU and you're not delirious today, you have a substantially higher risk tomorrow than if you had not received a benzodiazepine. And in randomized controlled trials, when you prioritize or you choose a non-benzodiazepine sedative compared with a benzodiazepine sedative and this is just one example that compared dexmedetomidine with midazolam, you see a substantially lower risk going forward after that first day of randomization for patients who are sedated with the non-benzodiazepine than for those who are sedated with the benzodiazepine. This has been repeatedly found in numerous studies. Just another example here, the CEDCOM trial compared dexmedetomidine and midazolam. Actually, I think, yeah, the first one I said midazolam, this is lorazepam. If you sedate patients with a non-benzodiazepine or over a benzodiazepine, you see lower risk of delirium. Turns out you also see a lot of other outcomes are improved such as reduced risk of over sedation, reduced risk of prolonged mechanical ventilation. So except for specific instances where it's indicated, for example, benzodiazepine withdrawal treatment, in most cases, you shouldn't be using benzodiazepines as a general sedative in the ICU. What about other aspects of the ADEF bundle? Well, we know from multiple studies, I'm not gonna get into the effects on other outcomes here, but mobility in studies that examined it and then assessed delirium in those randomized trials, it results in lower risk of delirium overall and a reduced duration of delirium. This is one of the first trials done by Bill Schweikert and colleagues of early mobility. They had a early physical and occupational therapy strategy in this trial of mechanically ventilated patients and found that patients who were treated with early mobility had duration of delirium reduced by 50% compared with those who were not. If you put all of this together in the ADEF bundle, then the probability of delirium goes down substantially as you do more and more components of the bundle. So better compliance with the ADEF bundle is significantly associated with reduced risk of delirium in ICU patients. So that's focused on treatment, although also reduction of delirium once it occurs. What about, I'm sorry, prevention. What about treatment? Well, one of the most common medications used to treat delirium in ICU patients is antipsychotics. These have been studied now in multiple placebo-controlled randomized trials, none of which have found that delirium was reduced with the treatment of haloperidol. This is one study that we conducted early on. It was the first placebo-controlled randomized trial comparing haloperidol with placebo. We also looked at zeprazidone, an atypical antipsychotic, and we found that overall the percentage of patients without delirium or coma improved with time in all three groups and there was no significant difference between the three groups. So treatment with haloperidol or an atypical antipsychotic did not speed recovery from delirium in this patient population. Similarly, in the HOPE ICU trial, which was larger than the previous trial, that compared haloperidol with placebo, again, in mechanically ventilated ICU patients, there was no significant improvement of those who received haloperidol relative to placebo in the speed of recovery from delirium or coma. One reason that all of these trials also include coma is because no one wants to make the mistake of thinking that you've done a great job of treating delirium because you simply made the patient comatose. You can make delirium go away if you make a patient comatose. You can also make the delirium go away if you make them paralyzed or kill them, but that's not a good thing. So you have to consider the holistic goal here. The goal is not just to eliminate delirium, but it's to return that patient to normal mental status. And that's why many of these trials use some sort of composite outcome that reflects alive with a normal mental state. Finally, the MINDUSA study that we conducted compared zaprazedone again with haloperidol and placebo. This trial was multicenter, much larger than the previous two studies, and found, again, no significant difference in the number of days that patients spent with a normal mental state that is alive without delirium or coma. Now, I do wanna highlight the most recent placebo-controlled trial of haloperidol called the AID-ICU trial. I'm not showing you here the effect on delirium, which was that there was no significant difference, but I'm showing you that they did find that the secondary endpoint of mortality was significantly better. Now, I don't know how to explain this, and neither do the authors of the trial. You can see that the authors of this trial, along with I and others, recently wrote an opinion piece in intensive care medicine about haloperidol. We don't have an explanation for this, and we don't actually even know if it's real, and what I mean by that is that all of the previous trials that I showed you comparing haloperidol with placebo in ICU patients who were delirious did not find a difference in mortality. So I think there are some possibilities here, and I'm gonna end on this note before showing you one trial about an alternative antipsychotic agent. One possibility is this is a false positive. It's just not a real effect. We know that that happens in clinical trials. It may be the case. That said, it's a large clinical trial, and we wanna ask whether or not there are other possible explanations. The second possibility is that heterogeneity in delirium in an ICU patient in general affects how different trials, the findings in different trials, and maybe the patient population enrolled in AID-ICU is different than the patient population enrolled in the three previous trials of antipsychotics versus placebo. So I think we need more work in this area. Does that mean that we need another huge clinical trial of haloperidol versus placebo? I'm not sure that that's the answer, but I think we need more work looking specifically at this question of are there subgroups of patients with delirium who respond differently to this treatment and other treatments than alternative subgroups? We need to stop looking at delirium as sort of a monolithic entity that's the same in every patient and start to recognize that delirium in one patient, and you even saw in the examples that Dr. Pisani gave us of agitated patients, maybe some were delirious, some were not, but they all had different reasons for being agitated, and the same is often true for delirious patients. So we can't just treat them all the same. We need to consider what is different about those patients. I think that's a really important area of future work. The last two slides I have here are one on quetiapine and second on what happens after patients have been treated with antipsychotics in the ICU. So this is a really small trial by John Devlin and colleagues, only 18 patients per group. I wish that we had a larger trial to tell us a little bit more about how quetiapine may be treating patients with delirium in the ICU, but I show you this because it's the only positive placebo-controlled trial that I know of, and when I say positive, what I mean is the primary endpoint of delirium was significantly improved for patients who received the antipsychotic relative to those who received the placebo, and that's provocative. That may be important. Maybe quetiapine is an answer. Then again, it's 18 patients, and maybe those 18 patients were highly selected. Maybe they're different than the average patient that ends up in a large randomized controlled trial. There are a lot of unanswered questions here. So do I give quetiapine to every patient? No, I don't. Do I consider it in certain patients? Yes, I do. Can I give you a good evidence-based answer to the question, well, when should I give quetiapine to ICU patients? No, I cannot. We need more answers in this area, and we need more data. What I can tell you is that you need to be very, very careful when you give antipsychotics to patients in the ICU. Number one, remember that on average, you're not giving a medication that when compared with placebo, reduces delirium, on average. It may work for some patients, but on average, the trials say that it doesn't work for all patients. Number two, these medications have side effects. In fact, they have black box warnings. Now, not for use in the ICU, which is an off-label use, but they have black box warnings for being used in older patients in the outpatient setting. And you might think, well, what does that have to do with what I'm doing in the ICU? This is just one study. Many studies have looked at this question. What happens when you give antipsychotics for the first time to an antipsychotic naive patient in the ICU, and it turns out that a lot of them end up going home on the medications. And I think this is potentially really dangerous. So when people ask me, well, why are you so reluctant? I mean, the trials didn't, maybe didn't show that on average it helps patients, but it also didn't show that it hurt them. There are no significantly increased adverse effects according to these placebo-controlled trials. So what's the harm? I think the harm is this. None of the trials looked at what happened to those patients after they left the ICU. But we know from multiple observational studies that around a fourth of patients who were started on an antipsychotic for the first time in the ICU end up going home on it. So maybe the answer is just better med reconciliation before our patients leave the hospital. But I can guarantee you one way to prevent an ICU patient from going home on an antipsychotic that's not indicated is to not start that antipsychotic in the first place. All right, so I did not do a very good job, but maybe none of us did a good job of leaving time at the end of this hour-long session. But I'm happy to stick around and answer questions for those of you that wanna ask, and maybe that's true for the two of you. All right, so please give us a little bit of extra attention here for these questions. We'd be happy to hear them. Thank you very much. All right. Thank you.

pain management

agitation management

delirium prevention

ICU

validated pain assessment tools

individual patient factors

delirium assessment tools

ABCDEF bundle

non-pharmacological interventions

multidisciplinary approach