All right. Good morning, everyone. My name is John Greenwood. I am going to be your moderator today. And I have the pleasure of introducing two amazing speakers. We started the conference with the non-SADA trial, talking about the value of sedation in the ICU. But today, we're going to be focusing on the other side, the analgesia side. And so it's my pleasure to introduce our first speaker, Professor John Devlin. He is a professor of pharmacy at Northeastern University, as well as a clinician researcher at Brigham and Women's Hospital in Boston, Massachusetts. He's going to be giving two talks, the first one really focusing on analgesia and what goals we should be reaching for, and followed by the second, which are going to be talking about some alternatives in terms of analgesia, including ketamine and a few others. So let's give a warm round of applause for Professor Devlin. Great to see so many of you this morning. So welcome to this session. I think it'll be provocative and raise a lot of questions, because as many of us know, the evidence for some of this isn't as strong. But we obviously have a lot of societal and institutional pressure. And obviously, we have the patient in front of us, but we really want to optimize their comfort when they're critically ill. But the question continuously comes up is, what is the role of opioids? And if we're not using an opioid or trying to reduce or decrease our opioids, what are really the alternatives? So that's really what Megan and I are going to be focusing on. But I think before we kind of jump into the non-opioid analgesics that could be alternatives in our patients, it's good just to kind of go through a little bit, set some basic information about assessing pain and setting goals, and talking maybe a little bit of opioids in terms of their advantages, but also some of their limitations in certain patient populations. These are my disclosures. I'm going to spend just a couple slides. I'm sure many of you have read the PADS guidelines from 2018. This is an area that there was a lot of content regarding pain, opioid use, and non-opioid analgesics. And it's good to have a little bit of an idea of the methodology when we're bringing this up and potentially discussing some of the agents and the recommendations and how we evaluated things. So it was really an update, but also kind of a revision of the PADS 2013 guidelines, two new topics, rehab, mobilization, and sleep disruption. Interestingly enough, we had patients very involved. And in fact, one of our patient ICU survivors is an anesthesiologist, but she'd had a critical illness. So she certainly wasn't talking about what she felt should be done with choices of agents, but she really had some amazing insights that really did inform the guidelines in the panel. And I think by being an anesthesiologist, she was just really able to communicate sort of the pain state. Because it's obviously much more complex when we're at the beds, and then we realized in some patients. We also added experts from Europe and Australia, which is interesting to this whole area of analgesia, because things are done differently in Europe in terms of choice of agents, the care that's provided, and sort of the approach to pain assessment and outcome measurement. Unfortunately, only two of our recommendations are rated strong, so that's a really important thing. And then a number of weak or conditional recommendations, and then a number of ungraded statements. Our data ends in October 2015, although we did look back before publication to see if there's any real seminal publications that were published. We primarily focus on randomized control studies, but especially in the non-opioid analgesia world when we're focused in the ICU, we did look at two areas, the observational studies in the ICU with use of these agents, and also there's a huge body of literature for many of these agents in the surgical world. And this was something that we struggled with as a group, is a patient that transitions really quickly through the ICU, say after cardiac surgery, and is on the floor the night out that night, are they really a critically ill adult, or are they just someone that needed some post-operative support in the ICU? I don't think there's an easy answer for that, but it's an important nuance when we're presenting some of the data. And then this is just how we voted. So just with strong versus conditional, conditional really generally applies to subgroups of patients, so it might just be surgical patients, it might just be cardiac surgery patients. Supporting evidence, generally conflicting, lower quality, if you had one large randomized study that could easily change the whole recommendation. There could be a much closer balance between benefit and risk, so that's where you're getting into, well, it could be good if this patient doesn't have kidney disease or any signs of liver dysfunction, maybe this drug would be okay, but it might not be good in all these patients. So you're really getting into a lot of slicing and dicing subpopulations and risk factors. Influences of future research we talked about. And then these are the hardest questions to even come up and make recommendations for, so it's these types of questions. We had a live meeting three congresses ago, and we spent probably about six, seven hours just talking about the non-opioid analgesic recommendations, whereas other things we flew through because it was such an easy recommendation. These are just all the pain things, this is a little bit more of a focus on, we looked at assessment, protocol-based pain, multimodal analgesia, procedure analgesia, and then non-pharmacologic strategies. Most of today's session is going to be focused on this area, obviously. So just a little bit of stepping back a minute from the guidelines and really looking at how we approach any patient that we know is going to be in pain and require analgesia. And obviously, whether it's an acute illness, and certainly pain is prevalent and discomfort is prevalent in medical ICU patients, even if they have no more of a procedure than intubation and maybe a central line, we're trying to optimize pain control, certainly pain vacillates throughout the ICU stay based on procedures and a ton of other factors. And then what we're really trying to do is make the patient comfortable, mobilize them, start mobility and rehab as early as possible, and then discharge. So this is kind of the ideal pathway for any patient that enters the ICU. Unfortunately, we sort of kind of get in this, what one of my colleagues at the Brigham called the wheel of pain misfortune, where, you know, we have an acute illness, if we have inadequate pain control, this can lead to, you know, lack of rehab. If we really, you know, miss the boat with pain or mismanage it, certainly chronic pain syndromes could ensue post ICU, could increase the like the stay and, you know, lead to other sequelae of prolonged ICU. So you can kind of get the circle and I think we go in and out of these sort of wheels with our patients, but we really need to be thinking about how to stick to this pathway. Certainly pain is complex and, you know, I didn't realize until I was, you know, on some calls with, you know, some really, really talented nurse and anesthesiologists where this is their whole world and, you know, pain is so complex. All the domains of pain, whether it's somatic, visceral, neuropathic, you know, baseline pain and analgesia requirements is really important. So often these patients are coming into the ICU with chronic pain syndromes. They're already on opioids or other analgesics, they have tolerance. There's certainly a lot of critical illness factors, ICU related factors that affect pain and how patients report pain. And then certainly many things that we do in the ICU, the way it's structured and how pain analgesia is delivered can affect all of this. So there's lots of painful procedures in the ICU. These are probably some of the larger ones, both medical and surgical. And so the one question we posed is what is the most reliable and valid pain assessment method to use in critically ill adults? And you know, one of the most important things, and I've seen it at our institution where we've put a lot of training into CPOT, but really trying to see if the pain, you cannot, the most important thing is trying to get the patient to self-report pain, even if they're intubated and mechanically ventilated. So that generally is going to be a, you know, NRS scale of 0 to 10. So patients self-report in, you know, all, and really the whole pain literature is much, much better than using a behavioral assessment tool to estimate pain. We looked at proxy reports, if patients are able to self-report. I think, you know, we suggest that families maybe can be involved in pain assessment, but there's not the data to suggest that the family member should truncate the expertise of certainly if the patient can self-report or what the clinician can assess. This is a big area of research. I think we have to really be careful about vital signs. It's easy to look at the monitor, maybe at a more sedated patient, seeing a high, you know, tachycardia, hypertension, other things that, you know, might be changing and suggesting, oh, the patient must be in more pain. It could be a trigger for further assessment, but it shouldn't be the be-all and end-all that this patient needs in analgesia. So this is just obviously, you know, using a numeric rating scale on an intubated patient. Generally four, greater than four is clinically significant, but this is very dependent on the patient. You know, there's a lot of variability here. This is a CPOT, which relies on facial expression, body movements, muscle tension, compliance with the ventilator. It's a generally, if it's three or above, you should be thinking about do I need to deliver or change, you know, analgesic or change what I'm doing. It's important with CPOT, it's not a longitudinal scale, so four is not twice as much pain as two. I'm just going to skip over the BPS. This is more widely used in Europe. We did raise the question about using a protocol-based pain approach, which could include just simply a protocolized pain assessment, using protocolized analgesia with a different agent, or it could be using sort of the classic analgesial sedation where you're giving an opioid or analgesias before you deliver sedation. This could be a whole lecture, obviously, but we know by using a protocolized approach to pain in the ICU, lots of important outcomes, including better pain control, better IC throughput, decreased consumption of sedatives, greater use of non-opioid analgesics, and greater reduced reductions in opioid-related adverse events. So we just made, too, a practice statement that ICU patients should be guided by routine pain assessment, and pain should be treated before a sedative agent is considered. I think this is kind of a no-brainer for any of us in this room, but it's funny when you translate into practice how things will come up at the bedside where we're like, well, maybe not, and we should think of this. So try to think of this principle. Let's just move right to opioid use in the United States. This is obviously not news to anybody in this room. Four-time increase in opioid prescriptions between 1997 and 2002. Obviously a long, torturous history here. Many states liberalized opioid prescribing practices. JACO came in about 15 years ago. Pain was the fifth vital sign. Aggressive opioid manufacturer marketing. U.S. accounts for 5% of the world's opioid use but consumes 80% of the population, but 88% of the world's prescriptions. Obviously the whole opioid use crisis. Opioids have many potential unwanted effects. Some of these opioids certainly have some advantages in some populations. Once you've altered, say if someone's in ventilator dyssynchrony and you've altered the ventilator status and they're still tachypneic and they're not as sick as the ventilator, maybe an opioid would be an option. There's some rules for opioids in some of these things, but generally there's a lot of class effects with many of these agents and then there's specific agents, side effects with the agents that we might potentially be using. Interesting enough, if you go back to 2013, there was a clear message from the 2013 guidelines that if someone's in pain, they should get an opioid. This wasn't really based on a lot of literature. This is a question that our guideline group in 2018 really spent a lot of time thinking about among many people, particularly in I think the perioperative surgical world. I think just some simple things with opioids, tolerance starts generally within 24 hours. When you give an opioid, the pain response is not a nice even thing. It's not like the more opioid you give, the better the pain response. This has really led to a huge amount of interest in multivental analgesia over the last decade. I think a lot of this started in the anesthesia world and it's sort of extended throughout critical care practice. These are the agents obviously we could be using. We've sort of targeted the ones that we think probably have the best evidence and clinicians are most interested. Megan and I will focus on that. Then obviously, whether you have an ERS protocol in your ICU or you have patients post-surgical that are followed through an ERS protocol, some of which move through the ICU. This is obviously a standard part of practice. We're not going to go into this, but I just wanted to sort of make the connection here. That's all I had sort of in the first section just to kind of get us warmed up and thinking about this topic. Next slide. Okay, so we're going to talk a little bit about ketamine, lidocaine, and gabapentin. So ketamine is a really interesting medication, right? The pharmacology is fairly complex. It's structurally related to PCP. It's considered a non-barbiturate anesthetic. It's got a complex sort of molecular structure that leads to multiple different clinical effects when it binds to the MDA receptor and it also has effects on the mu and kappa opioid receptors. So it's really a complex pharmacologic effect. This is very much generally a dose-related effect. So at very low doses, it's more of an antidepressant effect. As you move up in dosing, it's going to be more analgesia. Then you're going to start seeing some of the more psychomimetic effects. And then at higher doses, you're going to be looking at sedative sedation and hypnophysis. And I realize we're not going to be talking about ketamine as a primary sedative here. We're strictly going to be looking at it for analgesia here. So it's generally lower doses. Like any drug that shows these different effects, it's somewhat of a dose-related effect. There's a huge amount of individual variation between patients. And so I didn't even bother trying to put dosing ranges here because we see a large crossover with these effects as we increase doses, depending on what's going on in the patient and the patient themselves. You know, there's some notable adverse effects. Recovery reactions, emergent reactions can happen in up to 12% of patients. Recovery agitation in 0 to 30%. Again, huge variability. MS is 5 to 10%. The one advantage of ketamine is it doesn't have effects on the respiratory drive. But if you give rapid IV pushes, you can see temporary respiratory depression. It certainly can increase the mean arterial pressure, cardiac open, and ICP. So depending on the patient and where you're at, these are effects that could happen. You know, there's many, many. I'm kind of going to start with the much broader perioperative population and then sort of focus it down more into the ICU. So there is consensus guidelines on the use of intravenous ketamine for acute pain management. This is from the American Society of Regional Anesthesia and Pain Medicine, also the American Academy of Pain Medicine, and also ASA. Overall, these guidelines were fairly well done, although they didn't quite focus as much in terms of their PICO process and carefully look at individual outcomes and the effects. They sort of mixed things together a little bit. But you know, probably their most – I'm just going to focus on two other guideline questions is which patients in acute pain conditions should be considered for ketamine treatment. And overall, we conclude that sub-anesthetic ketamine infusion should be considered for patients undergoing painful surgery. So it's a grade B recommendation. While this would be sort of a conditional or weak recommendation, A would be like a strong recommendation, moderate level of certainty. Ketamine may be considered for opioid-dependent or opioid-tolerant patients undergoing surgery. And there's obviously far more in this that I'm not going to go through today. You know, what's amazing is how many randomized control studies have been completed. Obviously many, and there's been many systematic reviews looking at these randomized studies. This review in 2011 by Zaleskowski included 70 randomized control studies. Now these studies have a huge amount of heterogeneity in terms of dosing, patient populations, the outcomes, exactly when ketamine was given. Was it given intraoperative? Was it continued postoperative in the ICU? And then with that, and then also the outcomes that were evaluated, generally the one thing I was kind of interested in looking at too is safety. And it's sort of mixed in terms of delirium and hallucinations with this agent. So in terms of efficacy and safety, overall they've shown effects, but it really depends on the individual studies. And so the guideline group sort of concluded, we conclude moderate evidence supports the use of seven anesthetic IV bolus doses and infusions as an adjunct to opioids for perioperative analgesia. And again, not a lot of these studies looked at its use postoperative in the ICU, which is certainly an area that I think has really increased over the last five to eight years. One group from Brigham looked at lotus IV ketamine infusions for adjunctive analgesia. And they looked at 396 consecutive patients receiving an infusion less than 0.6 for analgesia while receiving opioid therapy. And these were all patients getting it postoperative. So that's why this is kind of an interesting study, even though it's observational in nature. Surgical 70%, ICU 25%. The duration was 35 hours. The starting infusion was 0.2. Maximum infusion was 0.2. So a fairly tight range of very low dose. Visual analog pain scores decreased after initiation from 7.1 to 6.4. However, 21% of the patients developed hypertension. This was defined as the systolic blood pressure increased greater than 25% from before the initiation of the ketamine. Getting more into the guidelines, there's actually only one, at the time of the guidelines, one randomized control study looking at postoperative ketamine use in patients that were critically ill in the ICU. This was a single center perspective randomized trial of 93 patients. And it was major abdominal surgery who were all ventilated in the SICU after surgery. Patients were randomized to receive either morphine or placebo or ketamine. As needed, morphine boluses were allowed. And this is the pain scores. And on any particular day, and then also across the whole state, it was significantly reduced with the exposure to ketamine versus no ketamine. And then they looked at PCA morphine administration. And every day, it was significantly reduced in patients exposed to ketamine. However, there was, interesting enough, when you read the study, they really didn't collect data on potentially over sedation, delirium, or any other related safety concerns, including hypertension in this population. So sort of the considerations that the PET-IS group made when evaluating this is only one randomized control study is available, very, very high risk of bias. Lots of issues with this study. Data limited to abdominal surgery for patients only, safety, particularly delirium not reported, role of sedation on analgesic effect or pain assessment unclear. And probably what drove us to make a conditional recommendation was just the considerable observational data in ICU. And of course, you can't ignore the non-ICU perioperative data in this. So we suggest using low-dose ketamine, 0.5 milligrams per kilogram, 1.2 mics per kilogram per minute. The dosing is funny here, because we had to base it on that one study, realizing most people would not be dosing it or ordering it at this dosing level. As an adjunct to opioid therapy, we would seek to reduce opioid consumption in post-surgical adults. So I'll turn to lidocaine now, again, another agent with kind of interesting fibricology, blocking high-speed voltage-gated sodium channels in neurons and monocytes during the action potential. So certainly, it's widely used as a local anesthetic. But of course, it also, because of these properties, has antiarrhythmic properties. It also has immune-modulating effects. Half-life is about 60 to 120 minutes, so a longer duration of action, which is advantageous. But if you have safety concerns, it might not so be. And it has a little bit of a tricky pharmacokinetic profile, because it has biphasic distribution. It's primarily cleared by the liver. Certainly, we're all familiar with the contradications to administering lidocaine. And of course, many of these I would consider relative contradications. It very much depends on the patient and, obviously, expert assessment by the team. And then I think many of us are aware of the toxicity that can devolve and result from lidocaine infusion. And I think probably the way I always best remember this is mild symptoms are generally more peripheral symptoms, so numbness, lightheadedness, dizziness, ringing in the ears. Moderate is much more central, so nausea and vomiting, severe dizziness, hearing tremors, starting to see hemodynamic changes, confusion. And then severe is muscle twisting, convulsions, cardiac arrhythmias, and loss of consciousness. So this is always playing in the background, particularly in our guideline group when we're trying to figure out in an ICU population where maybe people weren't used to running IV lidocaine infusions for pain a little different than the perioperative setting, the safety really came to sort of the forefront in terms of our concerns. Looking at IV lidocaine for postoperative analgesia in non-ICU population, there's really good systematic review by Weeble and colleagues in the British Journal of Anesthesia. And basically what they found when they looked at the pain score is early pain was more efficacious. So this is defined as first four hours after surgery. That's where lidocaine seemed to have the biggest effect in those first postoperative hours. As you moved out from 24 hours to 48 hours, lidocaine seemed to have less of an effect on the visual analog scores. But of course I think there's lots of variables that could affect pain as you move out from surgery. But that seems to be the trend across these studies. When they looked at pain score changes here, they were significant in terms of the side effects where postoperative ileus was reduced because of the reductions in opioids. And then also time to first bowel movement was significantly reduced. And this is an important part of the equation that also I should mention has really been poorly evaluated in a lot of these non-opioid analgesic studies is are you really seeing safety benefits by not using opioids when you use a non-opioid analgesic? This is the same meta-analysis. And this is looking at difference in opioid exposure. And as you can see, there's a lot of heterogeneity across studies. And even when you divide it by open abdominal surgery, laparoscopic abdominal surgery and other surgeries, it keeps crossing zero here. Overall, there was a significant effect with a p-value of .06, but certainly this is not an overwhelming reduction in opioid use with lidocaine. And again, this isn't the ICU population. This is just sort of the general perioperative population. Moe and colleagues looked at trying to, in 25 patients, this is critical care now after ICU, they looked at reductions in the pain score. And they did find a reduction after the initiation and also in morphine. But this is just a case series of 25 patients because there's just not a lot of really good ICU data in critically ill adults. There is one single center by Inser and colleagues that was published almost 20 years ago, 100 cardiac surgery patients requiring post-operative ICU stay. They gave lidocaine 1.5 milligrams, IV bolus over 10 minutes during surgery, followed by a 10-mite per kilogram per minute continuous infusion for 48 hours versus placebo. And they did not show any difference in self-reported pain, fentanyl, sedative consumption, time to extubation, or ICU length of stay. And neurologic and cardiac safety concerns, as we already talked about, are potentially of a concern in ICU patients, particularly that are hemodynamically unstable or have other degrees of end-organ dysfunction, particularly evolving litter disease. So we made conditional recommendations against the routine use of IV lidocaine. You just have to be kind of conservative when you're in the guideline world because if we say, you know, use it, but then really be careful and don't do it, that's kind of worse. So we suggest against the routine use. But of course, there's going to be patients where, you know, particularly anesthesiologists, surgical intensivists are going to think this has a role to extend into the ICU stay. And obviously, it's simply, you know, when you read our evidence gaps, part of our manuscript is there's just so little data. The question hasn't been answered. So this whole area is really ripe for further research. I also wanted to throw in, we did look at lidocaine for procedural pain and generally chest tube removal, sort of the gold standard procedure where procedural pain is best evaluated. Three studies have looked at that, all very small, and absolutely no difference when we're looking at VAS pain scores. I mean, you can't get much less different than this. So there's no evidence against lidocaine's beneficial concerns, obviously, about toxicity even with a single dose. So again, we make your conditional recommendations against using for chest tube removal. And there's certainly, as Megan's going to talk about, there's certainly potentially other agents that could be used. I just wanted to close looking at adjunctive neuropathic pain medications. This is another challenging area because many of these patients are on these agents before they come into the ICU. Our question in what I'm going to talk about today is not looking at should we continue these agents and trying to figure out what was the pain before they became critically ill. That's another whole population of patients, right? But we looked at really two questions, two bodies of literature. We looked at three studies that all have been completed by the same research group in evaluating gabapentin in patients with Guillain-Barré syndrome, which is a good model in patient population to truly evaluate neuropathic pain and very strong benefits in terms of opioid use and pain control. There's two other studies, one by Josh and colleague and one by Persona and colleagues, that looked at the use of pregabalin, 150 milligrams daily starting the day before surgery and then 150 milligrams daily in the ICU. And again, these were both patient populations of cardiac surgery patients that were going to stay in the ICU at least for 24 hours. So when we looked at these two studies, there was no difference in time to extubation and I see like the state did not change either. So based on the effects though of the pregabalin on pain control and decreasing opioid use in this population, we do make a conditional recommendation for their use in after cardiovascular surgery and we make a strong recommendation for treating neuropathic pain medications in general critically ill adults that have sort of proven neuropathic pain. I didn't include in my talk in the interest of time, but these agents are all tricky to use and titrate and optimize use in the ICU if we do suspect neuropathic pain and that's another whole area. So I didn't even feel like I could give recommendations or guidelines because there's so many variables that might affect dosing and and then of course as you might dose things more aggressively in terms of getting a therapeutic effect, then important safety concerns come in. So I just wanted to highlight sort of some of the general evidence gaps that we brought about multibunnal analgesia, and I'm sure Megan's gonna, you know, highlight some of these as well in the context of her agents. But you know each adjunct of non-opioid analgesic requires far larger studies really in critically ill surgical adults to evaluate their open sparing effect properties and their ability to reduce opioid associated effects. There's actually been some very small studies in medical ICU patients that actually haven't really shown benefit and raised potentially some safety concerns, particularly with use of ketamine. So we have to read tons more literature in medical ICU patients. As we all know, we really can't extrapolate data from a perioperative patient in the surgical ICU to medical ICU populations, even though there I think there could be a huge role for ketamine in many of these patients. Potentially safety concerns that are elevated in critically ill adults, dosing, route of administration, really unclear. And then mostly studies, there's not really good studies looking at combinations of agents, and that's the way most of us are operating, where we're using more of a protocolized approach trying to use lower doses at different times to optimize care. So, and just I just want to refer you, we did publish an algorithm in the PAD-IS guidelines, which really just highlight the you know, that kind of the recommendations I've started to deliver. And also it does highlight that there is non-pharmacologic approaches that can be used in some of these patients, particularly surgical patients, and particularly for procedural pain. So thank you very much. All right, so I think what we'll do is we'll have a little bit of time for questions at the end. So I'd like to introduce our next speaker. She's a friend and colleague of mine, Dr. Megan Lane-Fall. She is an associate professor of anesthesiology and critical care at the University of Pennsylvania. So let's welcome Dr. Lane-Fall, who's going to discuss potions or poisons, acetaminophen, paracetamol, COX-2 inhibitors, and NSAIDs as another choice for analgesia in critically ill patients. Good morning, everybody. My family hails from the South, so I have to wait until you actually say good morning back before I can start. So good morning. There we go. Hi. Welcome to the last day of the conference. It's my great delight to be here to address you in this session. I can sum up my comments by saying acetaminophen probably, NSAIDs maybe, COX-2 probably not. But I will walk you through how we got there. So my objectives for the short time I have with you is to talk about non-opioid analgesic alternatives, aside from those that Dr. Devlin mentioned, and then to explain some indications and contraindications specifically for those three types of drugs that I mentioned. I don't have any conflicts of interest here. I am associated with the groups on the screen, but it has really not much to do with what I'm talking about, other than that I'm an anesthesiologist. So the key assumption here dovetails with what Dr. Devlin talked about, which is that we're assuming that you're looking at multimodal analgesia for your critically ill patients. When we say multimodal analgesia, we're talking about combining different analgesics, that act by different mechanisms, at different sites in the CNS. And the caveat that Dr. Devlin mentioned as well is we're sort of flying blind when we're thinking about medical ICU patients. So we use these drugs a lot in surgery, and that's where a lot of our information comes from. Certainly medical ICU patients have pain, but there's still a lot of opportunity there. So for those junior investigators in the crowd looking for an area of focus, this might be a good one. Why are we having this talk? So I gave sort of a version of this at a previous Congress, and someone came up and they said, well, what's the evidence that we should actually be decreasing opioids in the ICU at all? I said, that's fair. You know, we, in the face of the opioid epidemic, are being forced to re-examine our opioid use, and I think that's reasonable. You know, as critical care clinicians, practitioners, and researchers, we always want to think about why we're doing what we're doing. So I think it's fair to re-evaluate. But we also know that we tend to continue meds that have been started in the ICU once people have been discharged to rehab or to the ward. And so if we are continuing people on opioids that were started in the ICU, there's a risk of continuation after that and potential tolerance and addiction. We also increasingly are seeing that patients who come in with an opioid use disorder, either an active opioid use disorder or a history, are asking us to avoid opioids. And they're saying, please don't give me any opioids. And I've certainly had patients come in who've been hit by a car and they say, can you just not give me any opioids? And I go, so let's have a talk. That may or may not be reasonable. I can certainly try to minimize them, but we're having more of these discussions up front about patients and families wanting to avoid opioids. There are a couple of good resources that I'll steer you to. There's a paper in BJA Education, the British Journal of Anesthesia Education, that talks about analgesia and intensive care that covers a lot of what we've talked about. And then also, of course, the PAD-IS guidelines. So first, Tylenol or acetaminophen. As you know, acetaminophen is essentially acting analgesic and antipyretic, and we don't know a lot about how it works, but we think it decreases prostaglandin synthesis. This is really the cornerstone of multimodal analgesia. And if you look at the PAD-IS algorithm, it's right there at the top, sort of after the branch point, is that most folks should be getting acetaminophen. The evidence here isn't fabulous, but we think that the risk-benefit ratio strongly favors using acetaminophen. This is a table that comes from the Anesthesia Patient Safety Foundation newsletter. It's a nice sort of overview of different analgesic drugs that you can give. This is in the setting of peri-op care, so it has that caveat. But if you look here at acetaminophen, there's recommendations for PO and IV dosing. This is freely available online, by the way. We don't want to exceed 4 grams per 24 hours. And in general, in folks that have liver dysfunction or chronic alcohol use, you want to lower that even further. We do incidentally use acetaminophen in our post-liver transplant patients, as long as their graft is functioning well. So this is a discussion that we have with our transplant surgeons, but they're usually very willing to give our post-liver transplant patients acetaminophen. Does the route of administration matter? So how many of you are using IV acetaminophen in your practice at all? And most of you, right? There is a clear cost difference, which our pharmacists remind us of and our administrators remind us of. So if you look at the pricing in the U.S. here, you'll see a capsule of Tylenol is 18 cents. So you could give maybe three of those to get almost up to a gram, whereas it's 5.70 per 100 milligrams down here for the IV. So it's about $57 for that one bottle of 1,000 milligrams of acetaminophen. So there is clearly a cost difference. There does not appear to be strong evidence supporting that IV actually works better. So there was a systematic review published in 2015 that looked at IV versus PO, and they really didn't find a difference. This is why my pharmacist won't just give me IV acetaminophen, much to my chagrin. And certainly there are anecdotal stories of people saying, oh, it was so much better. It was amazing. And we don't know whether it was the speed of delivery or that there's greater bioavailability and they were getting a higher dose. One of these people is a good colleague and friend. So I don't want to say it was placebo, but maybe there's a little bit of placebo in there, too. But certainly there's no empiric evidence to support that IV is better than PO. If you have a patient who has a contraindication to PO meds, if their mouth is not accessible or their rectum is not accessible or their bowel is in discontinuity, which I see a lot, unfortunately, or if there's an ileus, you may want to use IV. And it's certainly part of our ERS pathway, our enhanced recovery after surgery for people that have had abdominal surgery, especially on the bowel. We'll put IV Tylenol in there. But usually it's just for the first day or so. And then we transition over to enteral formulations. In terms of contraindications, it's really just hepatic toxicity. So that's that's the limitation there. Limit the dose in people that have either actual or concern for hepatic dysfunction. Adverse reactions to Tylenol are super uncommon. There are some case reports of hypotension with the IV formulation, and it's not entirely clear why that might be. Certainly it could be the med. It could be how it's formulated. It could be that if someone is normotensive because they're in pain, so relative hypertension for them and you relieve their pain, you may then unmask hypotension. So we'll move on to NSAIDs. Here, I think of catoralac and ibuprofen, but clearly there are many more NSAIDs. These inhibit COX-1 and COX-2, and they act as analgesics, anti-inflammatories and antipyretics. There's very limited data here to support their use, but there's a clear demonstrated benefit in the one case of rib fractures. So this is a study that was published in 2014 of about 600 patients where 400 of them did not have catoralac and 200 of them did. And clearly there were improvements in outcomes in pneumonia and all of these different ways of looking at pneumonia that favored catoralac. There was also decreased vent days and decreased ICU length of stay. So we tend to use catoralac quite a bit in rib fractures, sometimes with erector spinae blocks, and we haven't talked about blocks as another adjunct. We could certainly talk about that in the Q&A, but there, at least in this population, is very good evidence for use of catoralac. In terms of downsides, GI upset for sure is a concern for these nonspecific COX inhibitors. Nephrotoxicity is a concern. And so if someone has acute kidney injury, which many of our patients do, or if you're giving nephrotoxic drugs along with them, you may not want to use these NSAIDs. Our surgeons worry a lot about bleeding and oozing. So if you have a post-surgical population, you ideally would talk with the surgeon and have that conversation about whether the NSAIDs are acceptable. There's some theoretical concern for delayed wound and delayed bone healing. This is pretty surgeon specific. So sometimes our spine surgeons really worry about this. The evidence isn't great, but certainly if someone has a bone and cervical fracture, for instance, and you're worried about their bone healing, otherwise their spine will be unstable, then maybe you don't give catoralac. And then if you have a patient who has aspirin sensitive asthma, you can certainly exacerbate that with NSAIDs. When you're looking at NSAIDs versus acetaminophen, there was a 2015 Cochrane review that showed low quality evidence supporting a lower risk of GI side effects with acetaminophen, which goes with what we know about these drugs, right? Acetaminophen has a very favorable risk profile, side effect profile, and NSAIDs are associated with GI upset. So this isn't entirely surprising, but this group did not find any difference in pain, swelling, or functioning that would support using NSAIDs over acetaminophen. So from Pat IS, they said, we recommend not routinely using a COX-1 selective NSAID as an adjunct to opioid therapy for pain management in critically ill patients. So if you have selected populations, you may want to use NSAIDs, but it's not routinely recommended. Finally, COX-2 inhibitors. So in the U.S., we only have one, it's celicoxib. We had a bunch of them before, but they were taken off the market because of cardiovascular problems, cardiovascular morbidity. So it's really just this one. The benefits of COX-2 selective inhibitors is that you avoid those GI side effects, but you still have analgesia. And we're not really sure about if it provides the same relief as opioids. There are some studies that were based on valdicoxib, which is one of those that was removed from the market. So whether celicoxib is equivalent to opioids is really unknown, but at least a drug in the same class had similar pain relief as opioids. So there are some limitations here with the use of the COX-2 inhibitors. There's a PO dosing, so they're only available PO. So if you're not giving your patients PO, that's off the table. And you will still have some platelet dysfunction and bleeding concerns related to the COX-2 selective inhibitors. So you have those sort of combined reasons for not necessarily giving them. That probably has a limited role in the ICU. So to pull it all together with these two drugs, I feel like Professor Devlin got the more fun drugs, but that's OK. We can talk more in the Q&A. So considerations by ICU type. So if you're in a medical ICU or you're taking care of patients with medical critical illness, you want to think about the possibility of inducing hepatic toxicity with acetaminophen. Certainly, we worry about nephrotoxicity with NSAIDs. And you want to think about covering episodic pain. So ideally, these patients aren't having pain at rest, although if they come in with pain states, if they come in with neuropathic pain, if they come in with fibromyalgia, certainly you worry about that. But just sort of sitting in the bed, you're thinking about, well, what am I going to give them when I turn them or when I take out their chest tube or when I when I do something to them? Whereas surgical ICU patients, not only do they have that episodic pain, but they also have the pain associated with whatever brought them to the ICU. So you may need to cover them sort of continuously for that. We worry about POMEDs in surgical ICU patients and bowel continuity or an ileus. And then we worry a lot about bleeding and oozing. Fever is one thing to think about. So the drugs that I have talked about here have the potential to mask fever. And so you want to think about if you need to know whether your patient has a fever. Our practice at Penn has evolved to not worry about this as much because we have other ways of picking up whether people are manifesting signs of what we would call systemic inflammatory response syndrome. But if fever is your diagnostic trigger, then you need to think about a different trigger. Right. So if fever is the thing that makes you check for pneumonia, then you might miss it if you mask a fever. Or if fever is the thing that makes you think about a DVT, then you might miss it. So you'll want to think about whether your diagnostic algorithms rely on fever before you think about giving an antipyretic routinely. In terms of personal reflections, really there aren't a lot of reasons to not use Tylenol, to not use acetaminophen. The side effect profile is very favorable. There are many different ways to give it. It seems to be a useful adjunct. So for most of our patients that have pain, that's our starting point. I will tell you that surgeons are very persnickety about NSAIDs. So if you're dealing with a surgical patient, it's useful to have that discussion with your surgeon. And opioids still are a mainstay. So it's sort of all of these things we think about along with opioids. Unless you have a patient who's older, I'll sometimes use just acetaminophen with an older patient because of my concerns about delirium and all of the other side effects that come along with opioids. Respiratory depression, urinary retention, which can lead to delirium, nausea and vomiting. Sometimes patients and families request otherwise, as I mentioned. And then if you have a patient who has a tenuous respiratory status who doesn't have a secured airway and you're really worried about respiratory depression, you might go to one of these other drugs first because you are sort of teetering on the edge of, well, if I depress the respiration even just a little bit, I'm probably going to have to tube them. So maybe I'll not use opioids. So these adjuncts are useful. They tend to decrease opioid use, but they don't actually eliminate it for most of our patients. So by way of conclusion, multimodal analgesia is reasonable in caring for our ICU patients. Acetaminophen is super well tolerated, so it tends to be one of our next go-to drugs after opioids. And then NSAIDs can be effective, but you really do want to think about the side effect profile. So thank you for your time, and I think we'll move on to questions. So, yeah, we have time for at least two questions. Anyone in particular, please feel free to step up to the microphone. And if not, I have one to start. Okay, yes, sir, in the front. Hey, I'm Dr. Lindberg. I'm a critical care fellow at Roanoke, Virginia. I noticed none of the clinical series are trauma patients. They're going to have, I'm sorry, I noticed none of the patients were like trauma patients where they're not going to recover from their, quote, major surgery in 48 hours. They're going to be in pain for weeks, some of them. And if you found any data on that, I also wondered about the gabapentin data because 150 milligrams is pretty low dose. I mean, I usually start patients on 300 a day and go from there. So I'll let Dr. Devlin talk about the gabapentin. I agree, it's a small dose, but for trauma patients, I think, you know, I think we're flying a little bit blind in terms of the evidence, especially looking at patients longitudinally. For our trauma patients, I mentioned the cotorilac for rib fracture, so that's one representation of the trauma population. But oftentimes, we're trying to get their acute pain under control, and we're using a lot of opioids, and we try to escalate pretty quickly if we can. But it's really just an individualized approach to treating their pain and not necessarily resigning yourself to saying, oh, well, they're going to go home on 20q4 of oxycodone. We really try to push all the other ways of treating that sort of acute phase of pain, and then we're going to expect it to decrease in the acute phase. And then oftentimes, we're getting them on to standing acetaminophen, and then hopefully lowering doses of oxycodone or something like that going out over time. But it's unlikely that you're going to discharge them from the hospital without any pain medicines, without any opioids, with a sort of polytrauma, lots of injuries. So I think your point is well taken, that we aren't necessarily going to be able to get everyone off these drugs by the time they leave the ICU. But certainly, we're thinking about de-escalating, and we're having those conversations with the teams on the ward to say, hey, by the way, this is not the home dose, right? Like keep de-escalating as you can. I think the other thing, too, is there's many other pain strategies in terms of regional anesthesia and blocks and all kinds of things that we we've just focused on the drugs. And I think putting that in context, and I would suggest with trauma patients, probably, I mean, the injuries are so heterogeneous, so it's kind of hard to say, well, what's the average trauma patient in terms of injuries and what to do? But I think it's it's a huge area. And obviously, it all gets back to in common traumatic injuries, you know, more research. With gabapentin, I agree. Most people, I think, in the ICU would start it if they suspect neuropathic pain, as you suggested, at a much higher dose. Certainly, in sort of chronic neuropathic pain studies, you know, most of the evidence suggests that its dose is far greater, you know, than 300 milligrams a day that might optimize benefit. You know, with gabapentin, there is sort of a tolerance effect, though, too. We are worried about sedation in our patients, depending on what their ventilator status is. So I think 150 milligrams is pretty low. I think anybody can make the argument of 300. But interestingly enough, giving 150 milligrams at that low dose really had a pretty profound effect on pain and treatment. So I would suggest we probably need more dose, you know, response type studies to really see what's the optimal starting dose and how high should we titrate up in the context of safety. One more thing about the trauma patients. I wouldn't underestimate the importance of fixing fractures. So if they're hip fractures, if they're femur fractures, you want to fix them. There is potentially a role for rib plating, depending on the number of rib fractures that you have and where they're located. And if there's a flail segment, there are other non-pharmacologic approaches to treating pain and trauma, but they tend to be adjunct. So use of abdominal binders if someone's had an open surgery. But I'm glad you brought the population up. Thank you. It looks like maybe there was another question. Any comments on the cardiovascular events associated with the NSAID use? Cardiovascular events associated with NSAID use? I don't see them very much. The COX-2 inhibitors, Valdecoxib, and I forget the other one, were taken off the market because of cardiovascular events, but I don't see them very much anymore now. But Professor Devlin, I don't know if you have any comments about that. I mean, I think there is a preponderance of using just often, you know, for very short periods, even with selects. So like, you know, at the brink of a lot of the protocols are just one single dose preoperatively. And so I think people are concerned, but we're not really sure. You know, is it one day? Is it two day or what exactly the risk is? It just hasn't been well-evaluated, especially when you look at risk of strokes and other cardiovascular outcomes, because these are fairly rare things and you need a large database to look at this. Especially in patients post-cardiovascular surgery, graft patency, any comments on those things? I am not aware of any evidence that speaks to graft patency or any issues after cardiovascular surgery and the use of NSAIDs. I would say in my own personal experience, the cardiac surgeons I work with are not super jazzed about using NSAIDs because they're already concerned about bleeding and they're probably already on an antiplatelet agent of some kind, like a low dose baby aspirin. And so adding additional NSAIDs on top of that is not something I've seen in my practice, but I'm just not aware of any evidence in that in that setting. Thank you. And I think I think with NSAIDs, too, I just want to mention to build on Megan's thing, you know, our group did talk a lot and it all came down when we made our conditional recommendation against. We we actually were thinking about making a recommendation that a single dose was OK. But we just did we went in circles that we just didn't feel because then if we make a recommendation for a single dose, that everybody should get a single dose. And then it all depends on subgroups and risks. We definitely knew most patients should not receive multiple doses. And that's why we thought there's just not the data here to even make any of this. So that's why we actually made a conditional recommendation against. But just because we make a conditional recommendation against it doesn't mean in your individual patient that it might be the most optimal thing. Where guidelines come in is protocolization across common patient populations and procedures. And that's where it's harder to to make those recommendations. I mean, in our institution, we use Tordol a lot. You use Tordol a lot in cardiovascular surgery? Cardiovascular surgery patients use Tordol. Fair enough. Your surgeons are? No problem. Stronger than mine. No problem. I would say in general, we we limit Catorlac use to 72 hours because it can be a potent nephrotoxic agent. So, you know, limit at some point. But usually for us, it's it's about 72 hours of Q6 dosing. Can you say your name and where you're from? Hi, Brittany Orlando, VA health care system. I have a question about your opinion about whether this shifts the paradigm of all this new data coming out and how we really should be pushing for Tylenol, you know, scheduled around the clock in the combination medications that are out there with Tylenol and opioid. A lot of what we see in our post-op patient population anecdotally that I see is that we're not really optimizing the Tylenol component. And then when we try to break that apart, it's, you know, more pills and the nurses sometimes get confusion on which one they should give first. But we really, in my opinion, should be doing that. We should be using Tylenol schedule when patients come out of the OR and then maybe using Oxy for breakthrough pain. I just was wondering about your experiences and how maybe we should be shifting away in the hospital from these combination medications. I'm certainly a fan of Tylenol, of scheduled Tylenol. I think we should. I think we don't often go to it. We don't often start with it when I think we should, which is one place where protocols and pathways are useful. They're not meant to circumvent the clinician's judgment, but they are meant to trigger us to think about what we should be doing. And my sense is that Tylenol, because it's an over-the-counter med, it's something that we give to our kids. It's not necessarily the first thing we go to when we're in the ICU or, you know, I've got my brain and my all my degrees and all my things. So we don't necessarily think about Tylenol. But I agree, I think we should be giving it on a standing basis unless there's a contraindication. I worry about giving things in combination in the ICU because it's sometimes it's useful to be able to sort of change the doses independently. But when someone is moving on to the ward, maybe it's better to use a combination drug to simplify that pill regimen. And especially when someone goes home and you see the pill boxes that people bring in sometimes. So if you can simplify that in any way, sometimes that's useful. And it may actually promote patients taking it, because sometimes they'll look at, oh, I've got an Oxy and a Tylenol. Of course, I'm going to take the Oxy, that's stronger, right? So there might be a role for combination in helping improve the acceptability to patients. But in the ICU, I like to split them. And certainly, you know, in a protocol, post-surgical or ICU, you can always have an opt out if you feel something strongly that the majority of patients, you could just have those as routine order sets and have the surgeon and anesthesiologist opt out. We've done that, too. Yeah. Yeah, altering order sets and creating defaults is a nice way to do this, because, again, you're not circumventing the clinician's authority, but you're just sort of triggering them a little bit. One one final question, sir. Hi, I do emergency medicine, but we're using, you know, the ketamine and the lidocaine. And there's actually pressure on us to avoid as much opioid use in the emergency department. Specific question is, if you give ketamine and you get pretty severe hypertension, when do you treat and what drug do you recommend? Well, I guess, I mean, the first thing is stopping the ketamine. Sure. Which could do it. I think I'll make as an anesthesiologist, but I would just say it kind of depends on the patient and what the value is and what's the risk of having some transient hypertension. You know, most patients, it's probably not a big deal. There could be subgroups where it is. But it sort of depends on it depends on the patient. It depends on their pathophysiology increases in blood pressure of 20 to 25 percent. I usually wouldn't worry about very much. And if you're going much beyond that, then I would worry about a risk of stroke. Certainly in a post-surgical patient, someone who had an aortic repair, I'd worry about keeping their blood pressure with an even tighter blockers. Or would you use something else? So it depends if the patient is tachycardic, then I'd probably use something like labetalol because I'd get the alpha blockade and I'd also get the beta sort of coming down on the heart rate. If they're bradycardic or normal cardic, I may use something like hydralazine or nicardipine to be a vasodilator. So I'd sort of think about it in that way. Whatever you would normally use. Yeah. In general. Yeah. Sort of think about how ketamine causes hypertension and give drugs that would counteract that. But there's no particular thing that I would give. I agree with stopping the ketamine. I would say just for your institutional approach to treating pain, to Dr. Devlin's point, I think there's pretty good evidence for ketamine. Lidocaine in an emergency population, the evidence is really not there. Certainly in a post-surgical population, abdominal surgery, all of that. But in an ED population, sort of all comers. I've used it for renal colic and it's pretty effective. It can work, but there are significant toxicity concerns. And one of the things we didn't mention is if you do get local anesthetic, systemic toxicity, you want to stop the local anesthetic. But you also need to have intralipid around. You need a specific drug to counteract the effect of the lidocaine. And so if I would make sure that all the people that are administering the lidocaine are comfortable with the use of intralipid or at least know where to get it and all of that. A lot of our nurses have never given IV lidocaine. Yeah. And a lot of ours haven't either. We've had people come over from our medical ICU to our surgical ICU just to get a lidocaine infusion because the nurses weren't comfortable. So. All right. Well, with that, I'd like to give another round of applause for Drs. Lane, Fall and Dr. Devlin.

non-opioid analgesics

ICU

analgesia

ketamine

lidocaine

gabapentin

pain management

acetaminophen

NSAIDs

COX-2 inhibitors