Good afternoon. Welcome to the Optimizing Management of Severe ARDS Symposium for two hours this afternoon. Thanks for sticking around late on one of the later days. We have five talks in two hours, so I'm not going to spend a lot of time talking to you. My name is Rich Branson, and I'm a respiratory therapist, and I'm a professor in the Department of Surgery at the University of Cincinnati. Our first talk is going to be called, Are We Utilizing Proven Therapies for ARDS? And our presenter will be Michael Shishadi, who's an assistant professor of medicine, pulmonary allergy, and critical care division, and associate scholar of the Center for Clinical Epidemiology and Biostatistics at the Perelman School of Medicine at the University of Pennsylvania. Dr. Shishadi. Thanks very much, and thank you to SCCM for inviting me. I was hoping, actually, that we would be able to present in a large forum, but we'll just have to make do with this little cozy space for right now. So my disclosure is I'm actually funded to study acute kidney injury and lung transplantation, which begs the question why I'm presenting on ARDS, but I am a clinician who takes care of ARDS patients all the time. However, I have to give credit to Dr. Meetha Prasad Kerlin, who has given me a lot of information for this talk, and is really a visionary in the field of ARDS therapy implementation. So here's the overview of the talk in answer to the title question. The answer is sometimes, but SCCM asked that perhaps I could flesh this out a little bit. So here's the outline of the talk. We're going to talk about which ARDS therapies are proven, or quote-unquote proven. Are we using these therapies? What are the barriers to use of the therapies, or what do we know about that thus far? And then just a little bit at the end about how can we do better, although several of the talks that are going to follow mine I think will focus on several aspects of that. So starting with proven therapies. This is just a summary of the 2017 ATS, ESICM, SCCM guidelines for ARDS in adults. And they identified several interventions that they reviewed and then made recommendations for. At the top of the list was so-called lung protective ventilation with low tidal volumes. They gave a range of four to eight cc's per kilo based on a number of studies with a plateau pressure goal of less than 30, although in some of the studies it was less than or equal to 30. And this was a recommendation that was strong, and it was given for all ARDS groups. Prone positioning for greater than 12 hours a day was a recommendation that they made for severe ARDS patients and gave a strong recommendation. And then there were several other recommendations down the list. The only other one that was strong was actually against the use of oscillatory ventilation. But higher PEEP and moderate to severe ARDS, as well as potentially recruitment maneuvers, received conditional recommendations. And then ECMO received no recommendation because the feeling was there were not enough data at the time. I am going to focus on, for this talk, the lung protective ventilation and the prone positioning pieces. So just as a review, this is actually not a picture of lung protective ventilation. This is a picture of the opposite. I am not a basic scientist, but I can kind of tell that that one set of lungs that's like the big bag and bloody and that sort of thing, that's not a good thing. So we know and have known for a long time that ventilation, and particularly ventilation with high tidal volumes, can result in additional lung injury in patients who already have something wrong with their lungs. So this is review. Many of us, most of us, all of us know this. But this is a summary of some of the ventilator procedures from what was called the ARMA trial, originally CARMA when they had the ketoconazole arm. And the initial tidal volume that they looked at was 6 cc's per kilo versus 12 cc's per kilo, and then a different plateau pressure goal. And I just mentioned that the respiratory rate that they used was between 6 and 35. That becomes slightly relevant with something I'm going to show later. And in terms of the outcomes, there was a 9% decrease in mortality, absolute decrease in mortality, in the group that received lower tidal volumes, as well as some of the other corresponding measures, including, for those of us who are kidney enthusiasts, there was a greater number of days without renal failure, as defined by a creatinine. The failure itself was two or greater. And then prone positioning, some background on that. So what we know is that in the presence of gravity when somebody is in the supine position, the alveoli are not distended in a uniform way. And in fact, down at the bases where gravity is acting the most, you can get a lot of collapse, whereas you can get alveolar over-distension in the non-dependent portions. Whereas when a patient or an animal is prone, there is greater homogeneity of lung inflation. There's greater homogeneity of stress and strain to the alveoli, and there's improved VQ matching. And then you see some CT scans that demonstrate some of that difference. And then these are four randomly selected children who are showing that in addition to the physiologic benefits of prone positioning, prone positioning can really just be very enjoyable. Okay. So, and if that doesn't convince you, all right. So prone positioning, the trial, the PROCEVA trial, was the big one upon which the recommendations from the task force were based. They recruited patients who had ARDS and were intubated for less than 36 hours. And then they had to have a P to F ratio less than 150. And that was confirmed after 12 to 24 hours. They didn't want patients who had transient hypoxemia. They needed to be on PEEP. They needed to be on lung protective ventilation. And they were randomized to supine and prone. And you can see there was a large difference in mortality or survival between the prone group and the supine group. So that's a list of at least we'll say best proven therapies that we have right now. So then we can say, well, are we using these therapies? So I'm going to have some assumptions here up front which form what I'll call the top bun of our collective feedback sandwich, if that makes sense, right? So the buns of the feedback sandwich are supposed to be the part where we're saying, yay, we're doing a great job. And then the stuff in the middle is like, well, maybe we could do a little bit better. So the top bun part is I think we all want to deliver the absolute best care for our patients. We all work very hard to achieve that goal. And we are excited that there are therapies to improve outcomes for our patients with ARDS. But now we need to move on to the meat or beyond ham as the case may be, if that's a thing. So there have been a number of studies looking at low tidal volume use for ARDS patients. So I say this is since that ARMA trial. It's not every study that's been done since then. But in a study that was done by Weiss at Northwestern published in 2016 and CCM, the authors looked back at several other studies that had been published looking at the rates of use of low tidal volume ventilation in ARDS patients. And what you can see on the y-axis there is the percentage of patients who were receiving low tidal volume ventilation. These were ARDS patients. They did have varied definitions of low tidal volume ventilation. Some used strict criteria, 6, 6 1⁄2. Some flexed it up to 7 1⁄2, maybe 8 cc's per kilo. But the range in those groups of utilization of lung protective ventilation was between 15 and 70%. So then in 2016 in JAMA, the Lung Safe Study was published. This was a study that involved 435 ICUs across the globe. It involved a four-week screening of ICU patients. It was done in 2014. They had prospective daily evaluation for acute hypoxemic respiratory failure. And then they collected a bunch of additional data, defined ARDS by the Berlin definition. And that was using site investigator input. So the site investigators input the components, P to F ratio, chest X-ray interpretation. But then an algorithm stated, okay, the patient has ARDS or not. And they ended up with around 2,400 patients who had ARDS and required invasive ventilation. So I'll just orient you for a sec. The one chart shows tidal volume, just the array of tidal volumes that were used. And then the three lines are based on mild, moderate, and severe ARDS. So you can see there's not much of a separation in the tidal volume utilization between those three groups. Whereas the plateau pressure, they found that the plateau pressures were roughly increased in the patients who had severe ARDS, if you look at that curve. Now overlaying a few things here, that dashed line is the median tidal volume that was used in the ARMA trial. So in the ARMA trial, they allowed patients to go to 7 to 8 cc's per kilo if they needed to. But 6.2 cc's per kilo was still the average. And the standard deviation was 1. And then the gray box there just shows the 4 to 6 cc's. So in actual practice, we have a lot of tidal volumes that are certainly well above the median from the ARMA study and clearly above where the goal was. Plateau pressure, and this is just based on the average of all these groups, we were actually closer to about 90% in terms of folks who were able to get a plateau pressure less than or equal to 30. And then this graph, and now they didn't have plateau pressures on every single patient, but on the patients that they did have plateau pressures, this graph shows a relationship between tidal volume and plateau pressure. And I have to get my orientation here so I can see what you're seeing. So in the lower left, in the lower left is the group that they were considering lung protective ventilation. That is less than 8 cc's or less than or equal to 8 cc's per kilo and less than or equal to 30 centimeters of water plateau pressure. Now once again, I've superimposed what the median tidal volume was in ARMA. And you can see that a large, large percentage of those patients, including patients with severe ARDS were fitting into that 6 to 8 category. Many more so than were in ARMA. And among the severe ARDS group, this is part of why I referenced the respiratory rate previously. The average respiratory rate was 23 and the interquartile range was 22 to 24. Whereas the average tidal volume was 7.5 cc's per kilo and a pH that was not great, but that was not terrible. And so I just raised the question, are there opportunities to decrease the tidal volume, keeping the minute ventilation the same and potentially increasing the respiratory rate? I'm not there. None of us is there. Each of these patients is an individual, but it's a reasonable question to ask, I think. And then looking at PEEP and FiO2, this is a chart that shows PEEP on the y-axis by inspired fraction of oxygen on the x-axis. And what you can see is that the average PEEP levels that were used did increase as FiO2 increased. However, how much did they increase and did they actually go according to some sort of protocol? So what I've superimposed here were the goals within the ARDSnet protocol. This was the actual ARMA study, but what we now call ARDSnet low PEEP target. If you superimpose the ARDSnet high PEEP targets, really any one of them at the higher ranges of FiO2 are utilizing a lot less PEEP than at least would be prescribed on the table. Certainly in the high PEEP category, they weren't anywhere close to that in terms of what we are actually doing in this observational study. So I asked the question, do we have opportunities to increase or become more aggressive about our utilization of PEEP? The caveat being we didn't know specifically whether a bunch of these patients had contraindications to increased PEEP. They were trialed on increased PEEP and it didn't work out. And then for prone position utilization, there is the APRONET study, or in French, APRONET, which would be based on I think 10th grade was the last time I had French and now I will never be invited to give a talk in France. But this involved 141 ICUs, 77% of which were European. They screened for prevalent ARDS on four dates during 2016 and 2017. So this is a number of years after the PROSIVA trial came out. They identified 735 patients with ARDS. Eighty-two of the patients met PROSIVA study criteria and 40% of those patients were prone. The severe ARDS patients, that was a broader group probably because they may have met some of the criteria but then improved some or there may have been other exclusion criteria they might have met for PROSIVA, 33% of them were prone. And what you can see down here, I think my pointer is working, this is the odds ratio for receiving prone position intervention based on P to F range. So what you can see is in the very low P to F range, it was much more likely for patients to get prone. I will note that in the LUNGSAFE study, which was done after PROSIVA, about a year after PROSIVA, at that point only 16% of severe ARDS patients were prone. So you compare that to 33% in a pronent, perhaps that's some degree of progress. And then the Canadian Critical Care Trials Group did a study in 24 ICUs, 2014 to 15, again after PROSIVA was done and these were all moderate and mostly severe ARDS patients. And what I've highlighted over here is the use of prone positioning, about 10% of all of these moderate to severe ARDS patients, and still only about 20 to 23% of patients with refractory hypoxemia, which was a PAO2 of 60 or less than 60 on an FIO2 of 100%. So some necessary caveats, like I was mentioning before, it is difficult for these studies to account for all patient level variables affecting therapy decisions. Differences in things like defining ARDS, changes in clinical condition over time, patient goals, they are likely to contribute to and potentially confound the findings, and I think it's probably a non-trivial effect. However, I still think that the signals are pretty substantial. These are not great rates of using lung protective ventilation. These are not great rates of using prone positioning in potential candidates for prone positioning. And so I think they're still informative. So that leads to the question, what are the barriers to use of these therapies? Okay, so time, and I don't mean nobody has time on our hands. I mean, has enough time passed? Well, just to kind of jog your memories here, when ARMA was published on May 4, 2000, hanging chads were not yet a thing. We were seven years away from the first iPhone and 10 from the first iPad. Michael Phelps had zero Olympic medals. The New England Patriots had zero Super Bowl victories. The Deutsche Mark was the official currency of Germany, and Lord Voldemort was still a non-corporeal wizard, which I can explain later to anybody who doesn't understand, just come up and ask, okay? 2013 was also kind of a while ago when you think about the Proceva trial, and at that point, the Eagles had not given the beat down to New England and whatnot. That was just gratuitous. Okay, is cost effectiveness an issue? I hope that we can agree that kind of coming into this, we're going to hypothesize that changing some settings on the ventilator, that's not likely to be super expensive. And even prone positioning, certainly if you watch that New England Journal video, it definitely takes people to do, but there's not a ton of fancy equipment that's needed. There's some. There may be some. There were a couple of studies done on cost effectiveness. Colin Cook looked at lung protective ventilation in 2009, and what I will give you is the take home from this, which was they found with an intervention increasing the use of lung protective ventilation from 50 to 90%, it would be a good value if that intervention cost less than $9,000 per patient. Cameron Bastin did a study published in CCM in 2019. They assumed an intervention that would increase the proning rate from 16 to 65%, and they looked at both societal and hospital perspectives and found that from the societal perspective, there would be a good value to the intervention if the intervention cost less than $51,000 per patient. And from the hospital standpoint, it would be good value, and this is based on quality adjusted life years, if the intervention cost less than $5,000 per patient. So that gives us a little bit of a ballpark here, but I think we probably can come up with some interventions that might meet that price point. There's probably underrecognition of ARDS. In the Lung Safe Study, site investigators who were actually entering the data on P to F and chest x-rays and whatnot, 34% of them recognized ARDS on day one of ARDS. Sixty percent of them recognized, or they recognized 60% of the cases by day 28. And some of the risk factors for not recognizing ARDS were things like higher staffing ratios, older patient age, higher P to F, heart failure, and lack of ARDS risk factors. And then there are other lung protective ventilation barriers as well. This is a fairly old study done at our institution by Mark Michelson, and he looked at, so our center was an ARDSnet center, and sort of just following the publication of the ARMA trial, they looked at a series of patients, ended up looking at 75 patients who met ARDS criteria, and they found that 32% of them were on lung protective ventilation by day two after ARDS diagnosis. They wanted to give a little bit of lag so they could, you know, you get your act together, you get them on lung protective ventilation. And they identified some things. They really looked specifically by chart review to say, well, what were the reasons why this may have happened? Well, in the cases where lung protective ventilation was used transiently, there was the concern expressed about diagnostic uncertainty. This patient may not have ARDS. Then they fell off lung protective ventilation. Diagnostic uncertainty, even if LPV was never used. It was intended to be used, but it was not implemented. And then in some cases, there was no documented reason. And so these represent some potential opportunities for improving things. And then we have a problem of perceptions versus implementation. So this is a study, again, by Weiss at Northwestern that interviewed, did a survey of physicians, nurses, and respiratory therapists. When asked the question, what percentage of your patients with ARDS warrant treatment with low tidal volume ventilation based on the best available evidence, 93% of physicians said yes, 61% of nurses said yes, 68% of RTs said yes. What percentage of your patients with ARDS have contraindications, 8%, 25%, 21%? What was the lung protective ventilation use among these same practitioners? 7.4% among physicians. And, well, let's just call it low for everyone across the board. I think we can just conclude that. And then the prone positioning barriers. I really like the title of table one in the APRONET study. It is literally titled, The 734 Reasons for Not Proning 634 ARDS Patients, which sometimes sounds like we're talking to our kids or something. But we're not kids, and these are actually real reasons. I think they were just being factual. They weren't complaining. Many of these reasons were actually appropriate. The hypoxemia was not severe enough because they included mild ARDS patients in this study. There was an end-of-life decision, hemodynamic instability. But 23% of patients with severe ARDS, the reason listed was hypoxemia not severe enough. And then there are some perceived proning risks. I've included the table here from Proceva, which just showed there was no increase in adverse events in the proning arm. But there are multiple additional potential concerns, which I'm not going to go into right now because that's the subject of a subsequent talk. And then I say this frequently. Well, the trial populations don't match my patients. ARMA excluded patients who had severe respiratory disease, a six-month mortality greater than 50%. Certainly, the sedation practices were different then. In Proceva, they excluded patients who had a life expectancy less than one year. And 85% of them had neuromuscular blockade, not necessarily like my patients. But, and this is not the only study to look at this. Is lung protective ventilation, talk about proning separately, but is lung protective ventilation really risky? And I think the summary of the studies that we have out there is there are some potential concerns, and we should pay attention to them. But they're probably small. And when you looked at signals for harm in this meta-analysis, the pH was 7.37 in the lung protective ventilation group. And the PaCO2 was 41. So how can we do better briefly? And I really do mean briefly. I absolutely had to include this picture because this is the computer order entry system that I used when I was an intern. It's called TDS, and literally, yes, like very well-named, TDS. The point that I have here is this was an order set that was put in when I was an intern low these many years ago. These ideas are not new to say order sets, e-alerts have accountability. But there are probably new ways that we need to implement some of these things. And I'm hearkening to some of the ideas that Meetha Kurlan has. Can we use defaults for lung protective ventilation within the order set? Can we deal with the diagnostic uncertainty gap and the implementation gap? And then can we implement processes by which providers are held accountable? Not you must use this, but just here's a reminder, and is there a reason that you're not using lung protective ventilation? Proning interventions, I'm not going to talk about a proning protocol, but I will simply say that in my experience, and I think in the experience at Penn, we really needed to muster activation energy to get a proning program going. But once we did, it was a lot simpler. And then we felt like we could just apply the evidence. So returning to the feedback sandwich, this is like the bottom bun, we're passionate about delivering outstanding evidence-based care for our patients with ARDS. I think observational studies suggest that there's still substantial room for improvement in utilizing lung protective ventilation and prone positioning, but better understanding and overcoming barriers to implementation may help to bridge the gap between evidence and practice. Thank you. Thank you. So our next presentation is about prone positioning. It's called Implementing a Proning Protocol, and our speaker will be Kathleen Vollman, who's a clinical nurse specialist consultant for advanced nursing in Northville, Michigan. Kathleen. Thank you so much. I'm excited to be here. This is a topic that has been near and dear to my heart and my practice. My first prone positioning was in 1982. Needless to say, as a nurse, mobility was part of our world. And so we had read some research, and there was a group of us on the midnight shift that decided we'd try it. And we did and saw a benefit, and then we started working with our interdisciplinary colleagues to make it happen. So in terms of disclosures, the only one that may be relevant is there are some turning and slide sheets that are available to help make the prone positioning process easier. And in the past, I have consulted for Stryker. So the objective is really to talk to you about the logistics of how to make this happen from a multidisciplinary perspective. This look familiar to you? So I wanted to ask the individuals in the audience, how many are, when they recognize a patient with severe ARDS, are proning within the first 12 to 24 hours? Awesome. How many are still last ditching effort on the prone? Because it sort of becomes, it's self-fulfilling prophecy, is that if you wait that long and you turn them, oftentimes they're more unstable and they don't respond. And then we say, well, that didn't work, and we move on to the next thing. So it's important to pay attention to that. I also wanted to make a note that the first author in this research was a registered nurse. It was the first patient study using the circle electric bed in prone positioning, and it was published in Critical Care Medicine in 1976. So how do you get started with this? Most of us have some form of interdisciplinary quality team, hopefully, within our environment. The usual players are present. The providers, a clinical nurse specialist or a clinical lead, some frontline staff nurses, the pharmacist, your respiratory therapist. And I would make sure to include physical therapy because proning is actually the start of some form of mobility journey that your patient will be on. And because one of the major risk factors is skin to incorporate your skin experts. So overall, looking at the process, where we want to start, you want to make sure the entire team is on the same level with the science. So there's an understanding of the research, and people are committed to working to make this happen. And then development of a guideline, and we're going to talk about each of these steps. Another step in this process is doing a gap analysis in your environment. Where is the education level of the practitioners with regards to the science around prone positioning? What is their comfort level? What are their concerns? So that as you build the guideline and the education, you can start to develop that stakeholder buy-in as a part of it. And I'm not going to go in detail because it was eloquently addressed in the first presentation. But really part of this is about building the why. Why are we doing this? And understanding the impact it has on the lung and the ability to recruit the alveoli from a non-invasive way is one way to look at it, and reducing any type of lung injury. And then oftentimes, I don't know if any of you have experienced it, even patients on high PEEP. When you turn them, sometimes you can get a rushing river coming out of them. And so lung clearance is a part of that journey. The ATS SCCM guidelines, as you saw, made a recommendation for positioning prone patients greater than 12 hours a day in patients with severe ARDS. And one of the challenges we have is actually recognizing the severe ARDS population in the clinical arena in order to apply this in a timely fashion. And again, this is some of the data from the SAFE study that you heard. Overall, the Lung SAFE study covered the world, whereas the most recent study that was published was a European prevalence. But interesting enough, the two major European countries participating in this were France, where the original PROSTIVA trial was, and Spain, which has been doing prone studies for quite a long time, and then some Italy. And those are three of the major regions that have been working on prone positioning research for a long time. Again, because the trial was presented, I wanted to point out, and it was addressed, within, they prone within one hour of randomization, and it was initiated within 12 to 24 hours of recognizing the severity of ARDS. They had criteria for the patient going in, and they had criteria for when to stop. And so, that's exciting for us in the fact that as we build a guideline, we have inclusion criteria, and we have exclusion criteria to help with that process. And again, not only was there an improvement in mortality, but there was a significant reduction in, well, it wasn't significant, but it was clinically significant, in that it was an ICU reduced length of stay by two days, and ventilator-free days of four days. So, when we come to inclusion criteria, basically, you're going to make it up based on the guidelines and the science. Try within less than 48 hours of the onset of severe ARDS. Some, you know, depending on what criteria you use, are moving it to PF ratios of 150. They're on a 60% oxygen and 10 of PPR greater. And that they can be, you can think about hemodynamics, but you can be on a vasoactive drip. You can turn the sickest of the sick. You just need to be mindful. What would I classify as absolute exclusion criteria? If you have unstable cervical spine, thoracic, pelvic, neurologic, if you can't control the ICP because turning is an oxygen demand maneuver, and so, it's going to cause some challenges, huge hemoptysis that you may not be able to control the airway, or if you have a do not resuscitate goal as a part of it. There are relative contraindications that look at a number of things, but I would say relative basically means there needs to be a frank discussion by the team before a decision is made about these patient populations. But interesting enough, we have multiple publications on patients that have been proned with open abdomens, pelvic fractures, external fractures, multiple trauma. So, some of the things that are classified as the relative contraindications, there are case reports and publications regarding those patients and being proned with ECMO and being proned with CRRT. And so, that sort of leads to the process component of this, which looking at the methodology, the equipment, and the staffing, so in other words, the guts of making this happen in your environment. So, I'm very excited to say that myself and a couple other colleagues have authored, and this is, I think, the third revision of this protocol in the American Association of Critical Care Nurses Procedural Manual. So, it has three types of proning. One with a frame, because the frame is still out there in the world. It may be in the bowels of your hospital, but it may be out there. Manual proning, and then proning with a prone bed, so that no matter how you're proning, there is a guideline. Now, if you're lucky enough, you may have Xena the Warrior on your staff, and you won't need a whole coordinated team to make this happen. And early on in the proning, I ran into some individuals, and there was a group up in Minnesota that there were like two of them, and they were just flipping it and making it happen. But most of us don't have Xena on the staff. So, if you take a look, you're familiar, if you aren't, the rotoproner. This was a positioning frame, and it is no longer on the market, but it is still in various hospitals. The manual proning, using the sheet methodology, which you can see on the New England Journal website. When we look at the sheet, some people affectionately call this the burrito method, as a part of it, because you burrito the patient up, and I'll show you that in a little bit. And then, there are some devices that help facilitate, they don't turn the patient into the prone position, but they make it easier for the clinicians that are doing it. And these are some examples. And then, the one on the bottom is actually wedges, if you want to elevate the patient and support various components of their body, as a part of it. So, when we look at creating these guidelines, I think it's important for each of the disciplines to know what their role is in implementing prone positioning when we're ready with the patient. So, on the provider side, consulting with the team, looking, touching base with the intensivist, explaining the purpose to the patient and family, and then looking at what kind of clinical interventional needs do we need to put on board in order for this practice to be safe. Do we need to change some vet settings during the turn? What's the sedation like? Taking care of the eyes and looking at the skin. From a respiratory perspective, I will tell you, they own the tube, not the ventilator tubing. They own the endotracheal tube at the head of the bed. It's interesting for me in the research literature to see a number of extubations. And it concerns me because, literally, in the work we've done on proning protocols, is that respiratory therapist is holding the endotracheal tube the entire time until the team lead calls a release, that everything is safe and out of the way, so that airway is protected during that process. They may also pre-oxygenate before the turn. You want to ensure the education. Now, I will tell you, and I think I heard this in another presentation, is that if the patient is used to sleeping on their stomach, there's every likelihood they will need less sedation. If they're not used to sleeping on their stomach and it's not a natural position, it may be uncomfortable for them. And so, you need to determine that. And also, the work of breathing and some of the other components. In terms of getting things ready, because pressure injuries are a significant challenge, a couple places that I've been to recently have literally put protective dressings in a baggie that pretty much cover the major areas, so nobody's having to run and get a million different dressings. They've got it and they call it a prone package. And they get it ready so that they can put it on all the areas that are at risk for shear or pressure injury. Another piece here is aligning the tubes. You want to cut the body at the middle. Everything that is inserted here up goes over the top of the bed. Everything inserted here down goes to the base. The only exception are your horse tubes, your chest tubes, your ECMO, and your CVVH. They usually run at the bottom because you're turning the patient in the same gravitational plane. The biggest challenge for us as nurses is all the spaghetti that happens with regards to the tubes and lines. And what I share with nurses is don't uncross that because, by the way, when they go back this way, it'll be in the exact correct spot. So, you'll be okay there. And then making sure we have protective dressings in the various locations. So, I want to give you the manual sort of burrito method. And, again, you can see this on the New England Journal site as well from a video. Taking a flat sheet, putting the patient in, always orient your prone positioning where you're going to be turning the patient prone towards the ventilator because then it's a direct line for your endotracheal tube as you move. As I said, the respiratory therapist is at the top. And then what you do is place a flat sheet around the arm so you're sort of burritoing or tucking the arm so that as you move into that 45-degree position, the arm flows under as a part of that. A second flat sheet is then placed and tucked under the patient. And this is going to be your pull-through sheet as you make the turn. I always recommend that when you're doing, before you do a full prone, you actually do a 45-degree tilt, do a safety check to make sure the tubes and lines and everybody's in the right place before you do the final prone. And then you move the patient over and then you can make the determination whether you want to suspend the chest or pelvic area and how you want to manipulate the head. One of the things that I do know, and these are some of the things to consider on the maintenance component of this, is it doesn't, it's a challenge to manage the head. The key is move the head every hour, just a small amount. But man was not meant to be face-planted this way, so there is zero support. And so being able to do small adjustments, you can significantly reduce the pressure injury risk for the head, whether you use a gel or a regular foam or pillows or whatever you happen to be using. Range of motion, if the patient can move this, swimmers is probably the best on a two-hour basis to get that rotation. Support the feet and correct anatomic alignment. Literally when I finish proning, I actually step back and I take a look at the patient and look at those pressure points and look to see if they're anatomically where they need to be. If you're doing hemodynamic monitoring, do the reference point. That's been evaluated in the literature, as long as you're doing the phlebostatic axis and re-referencing the zero reference, then you're going to have accuracy. If you are using a special ET device, you want to be mindful, because there's a fair amount of plastic here, that when you're facing down this way, you can get into some real pressure injury trouble. Some organizations, when they prone, actually go back to double taping to reduce the risk of that pressure injury. Eye ointment, and if you do decide to tape, don't tape this way, always tape this way. Because if the patient opens their eyes, it will reduce any chance of corneal abrasion. Also, people swell, because they're not used to being like this. So some do reverse Trendelenburg to address the edema component, but you have to be mindful of shear. So how do we return them to the supine using a flat sheet? You pretty much follow the exact same process. Move them over closer to the vents, because you're going to roll them the opposite way. Place the flat sheet around the arm, just like you did, because it's going to be your pull-through sheet. And the sheet then gets pulled through as you turn and tuck them, do the 45 degree, make sure everything's in the right place, and then return them and straighten your lines. Now, you can see why this would be easier with a glide sheet. That would reduce the worker potential for handling injuries in that process. So when to stop proning. In the PERCEIVA trial, the prone positioning was stopped when they maintained a PF ratio greater than 150, or after being in the supine position for longer than four hours, where their PEEP was less than 10 and their FIO2 was less than 60%. So that's sort of what people start with. And then you have to individualize for, you know, you use as much of the science as you can, but the team has to decide, because that's where the comfort level is. So what are some of the challenges and issues you are likely to face? Pressure injuries, as I shared. And that's a really important component, which is why you get the right people on board and you use some of those protective strategies. Major airway problems in this research, seven of it was unplanned extubation. And I truly believe holding onto that endotracheal tube through the entire process. Not the ventilator tubing, but the tube itself will make a difference. And then how you position the head can affect the tube obstruction. And so that is something we have to pay attention to. Additional complications, we talked about the fact that they may be overwhelmed with secretions. Vascular kinking, what I've discovered, if the lines are in the right place anatomically, being able to do the half body that you won't end up with kinking. I also don't recommend disconnecting any lines when you do the turn, because that increases infection risk. You can also have elevated abdominal pressure. You can have increased gastric residuals. And one of the recommendations is to stop the feeding about an hour before you're turning. Because the risk of aspiration is during the actual turn, not while you're in the prone position. And then there are, based on arms, the brachial plexus potential injury. So one of the other challenges, how do we find these humans in a fast time and not wait until they've passed the mark of benefit? And so I would ask, and this would benefit low tidal volume ventilation as well, is who's screening PF ratios? Is it happening? Is it being routinely looked at in your environment? And if the patient's moving in the wrong direction, is it increased from every eight hours to every four hours? So that the team is made aware that the severity of the ARDS from an oxygenation perspective is challenged and may be able to then make a decision for implementation. Another, and this was brought up, is the concept of hemodynamic instability. And I've published on this, and there is work in the aeronautic science looking at stretch receptors. And discovering that the longer a patient remains in the same gravitational plane, when you turn them, the more unstable they are likely to be. And we keep people flat a lot. And so they establish this equilibrium. And so one of the things that they do with astronauts is they train them at a half a G a time so that they get used to the change in G force. And so that is one strategy, is keeping the patient laterally turned so that the prone turn is not as severe from that perspective. Another is we tend to turn people too fast. And so these same researchers that worked on the stretch receptors, they also worked on inner ear equilibrium. And they found that there is a direct connection between that and cardiovascular response. And when we turn people, do we go like this? No, it's like whew. And that is not a smooth change in gravitational equilibrium. So I wanted to share with you, and this is published, is an algorithm to help frontline clinicians evaluate that hemodynamic instability and begin to train the patient to tolerate movement. Another last piece, and you went over the neuromuscular or discussed a little bit about it. The challenge, especially with the ROSA trial results, is where does neuromuscular blockade land now with prone positioning? Because the ROSA trial was a negative trial not having impact. And I think part of a component that we need to pay attention to is the severity of illness of the patients in the ROSA trial. And literally only 16% of those patients were prone in the ROSA trial. So I don't know if I can make that assumption yet of one versus the other. So that's a question that's still out there. And I'm going to end with I think in addressing the process piece of how do we get through this is, and this was recently published and it's an expert opinion, but it's here's the gradated science and here's where I need, at what point I need to implement. And so this was published and I also have one from my colleagues at University of Michigan, which by the way have been proning for 40 years in their surgical intensive care unit. So they've got a structured process down pretty well. So thank you very much for your time and happy turning as a part of that. Thank you. So our next presentation will be on balancing sedation, paralysis and mobility. And our speaker will be Dr. Russ Miller from First Health of the Carolinas. Thank you. And thank you, Kathleen. You know, I remember one of my colleagues at Intermountain several years ago, Polly Bailey standing up and talking about mobility at ATS and the excitement about the how-to of it. It's really refreshing. Thank you. I do not have any disclosures. My overview is to frame the relevant overarching question that is inherent with balancing to provide some hard data about sedation paralysis and mobility. And that is totally with a pun intended. And then unfortunately, although we recognize some of the benefits of many of the things we're going to discuss or have discussed today, how do we put it all together with the harms that are inherent in some of our therapies? That is to say, perhaps when you're stuck between a rock and a hard place. So, do sedation and paralysis, for example, run counter to important patient outcomes? And to the extent that they might, are these acceptable consequences? I suppose the question might best, in fact, be described or stated as, what are we really balancing? Lots of knowledge over the last 20, 30, now 50, 70 years. I'm going to take us back in history a little bit. How do we balance it all? Well, it turns out that after Galen recognized that air goes in, and that must be the way in which circulation is achieved, and Vesalius furthered some of this work, we only recognized oxygen and the relevance of it a couple hundred years ago. Through the ensuing 19th century, there was work that led to the development of negative pressure ventilation that in the ensuing 20 years, we turned into positive pressure ventilation. And part of the impetus for that was very practical. And it, in fact, undergirds the very notion of activity, mobility, proning, ECMO, any of these therapies, how do we get to the patient to do things to them if they sit in a box with negative pressure in a polio ward? So, the revelation of PEEP in 1938 furthered this, and then, perhaps surprisingly, the whole notion that respiratory failure was, in fact, a cause of death and could be treatable is not that old a notion. While first described in the 1770s, it was 200 years before we really began to appreciate potential harm of said ventilation, such as ventilator-induced lung injury. So, in the early, at least, 2000s, we have a position, we had a position that positive pressure ventilation can be harmful, of course, also beneficial. That being placed on a ventilator may induce harm, in fact, leading to concerns about patient recollection of that, of being on the ventilator. And also to the oft-worried issue of patient-ventilator asynchrony that frankly could be traumatic either to the lung or, per my prior comment, psychologically to the patient, him or herself. So, in other words, while sedation and paralysis can mitigate these harms, is putting the brain down really the way to go? And that's where we found ourselves 20 years ago. Now, truth be told, I also, two decades ago, trained with Wes Ely and grew up in an era of sedation and delirium management and concern. Putting the brain down may not be the way to go either. As an homage to the late Dr. Wheeler, I did want to include a quote, which I will not read, but simply to state, perhaps rephrase the second sentence of this, which is that sedation and paralysis are not for the benefit or comfort of those caring for the patient, or at least they shouldn't be. And while I stand here in 2020, hopefully at a different place in time than where I began my career, the reality is it still happens for those reasons. So, what's the probably known? The probably known, and I'm going to go through this quickly, and it's in words rather than pictures, in fact, to emphasize the fact that we should go through this quickly at this point. Light sedation is better. That's all prior to 2012, more than half my career, regrettably, at this point. Persistent coma, it turns out, leads to death. So, let's get away from heavy sedation. And that led to an onslaught of investigation over the ensuing couple of decades now, or decade, where beginning with J.P. Kress and then furthered by one of my colleagues at Vanderbilt, Tim Gerard, looking at sedation awakening trials, and then to light sedation compared to sedation awakening trials, spontaneous awakening trials, and then to how do we think about no sedation? How do we feel about that? But also, what happens to patients? And just, obviously, recently published some of this work building upon the work of Strom from nearly a decade ago now. The probably known also includes that short duration of intubation may lead us to choose one agent over another. And I'm not, that's not the point of this talk and not where I'm going to go with this, but I did want to highlight that we still have relatively few options. We need to be thoughtful about how and where we use those options, because they're not equal in yielding the desired results. We, I suspect most of us have had at least one patient where the drug withdrawal state did not respond as favorably to one agent or another. So, it's not as if one could very easily drop one strategy onto this, and, of course, that is inherent in the sedation guidelines last published a couple years ago now. So, the seemingly known, and then I'll move on, is that opioids probably should be used over sedatives, although there was a nice debate about that a couple of days ago, nice pro-con. Sedatives likely contribute to muscle weakness and loss. This notion of that being due solely to immobility is interesting, but not completely well tested. And then, sedatives and opioids, just as a reminder, impair restorative sleep. There's not a bit of what we do in terms of use of sedation that fosters restorative sleep with the caveat to the dexmedetomidine crowd that there may be some mechanism there, I recognize. What about paralysis? So, suspected, and I have to give a nod here to, again, to a pro-con that I added on to from John Marini and Norbert Schmidt a couple of days ago, also looking at the evidence behind paralysis. Yeah, reducing oxygen demand is a very noble goal. That sounds good, and there are lots of circumstances where we might imagine that that would work and be a good idea. Of course, it also led, prior to my career in the 1990s, to a desperate search for the ideal oxygen delivery number, the oxygen delivery, the 600 target that led us down a rabbit hole, as it was, at least so it looks 25 years later. Paralytics, do they avoid asynchrony? Gosh, we need that, because that darn lung protective ventilation makes people breathe funny. And that was some of the, some, I suspect, that's my bias, of the heritage in looking at paralytics. And, after all, we desperately want to avoid where this all started, and that was the ventilator-induced lung injury or diaphragmatic dysfunction. But the reality is, while, yes, paralytics can be used to reduce oxygen demand, prolonged use have consequences, specifically on diaphragm function. And also on clinical outcomes that matter. Not surprisingly, you immobilize the diaphragm, you will get an endotracheal tube or tracheostomy tube for longer. What we know, then, if I may, is that there is a modest, perhaps at most, contribution, and this was a systematic review from three years, four years ago now, of paralytics to subsequent, at least global, neuromuscular weakness. And, to speak nothing of the clinical results of the ACURASYS trial, the short-term use of paralytics in that trial did not appear to contribute to weakness up to 28 days later, at least by the measure that they were using to assess for weakness. And added, therefore, some doubt into the causal chain between paralysis and long-term weakness. So, to the clinical data, of course, yes, there was initial excitement that appears at least, depending upon your take of the differences between the studies, not to be confirmed. So, whence the third element of this? And that is mobility. So, in many ways, of course, the ICU liberation movement began and set as a primary target to my understanding, length of stay. But, no mistake about it, the target was getting people up and moving them. And the acknowledgment that undergirded this was that weakness correlates both with all of lower survival, lower physical function, and lower reported physical function, or lower quality of life. At the same time, rehabilitation safely improves muscle strength, decreases, you can read the slide. What's interesting about mobility itself is that in the latest guidelines, of course, in the absence of significant or sufficient data, there was not the strongest of recommendation because of the lack of good outcome data around mobility at this point. Now, it turns out you can't mobilize, well, you cannot easily mobilize a comatose or profoundly weak patient. It is hard to drag them through your ICU. So, you have to have a strategy of lightening sedation. And, again, that was one of the original points of the ABCDE bundle. And then, once providers got out of the way, if you will, and nurses got involved, and respiratory therapists got involved, and our colleagues reminded us that it turns out the family's got to know what's happening. They got to tell, they got to be engaged, and that opened, I think, our eyes to the notion of having a culture of wakefulness. What clearly does remain unknown is that it depends on how long-term you want to look at benefits as to whether we have clear data here. We know that certain other sedatives, like ketamine, again, a talk in the last couple of days on this, or other routes of administration, when available, the enteral route, might be equally or more beneficial. And then, of course, what really are long-term consequences of immobility from whatever it may be, whether it's the sedative itself, the paralytic, the lack of moving in general that so many of our patients are left in in the ICU still. The meta-rules for balancing this, if you will, I think, are to prove that sedation is needed at all. Not to assume that it is. To favor spontaneous breathing in almost all cases, and if spontaneously breathing have the plan for mobility in place, there may be a role for a trial dose of paralysis, and then a ready assessment of the effect of that, both clinically, if possible, but certainly physiologically. And if ongoing paralysis is deemed beneficial, then probably a relatively short course with a plan, again, for that next step. And I'm going to give you, to allow Kathleen not to gasp at the 48 hours before moving to proning, I'll offer instead this slide, which is completely hypothetical. I know of no one, including myself, who's abiding by this. But I wanted to offer it as a potential way to think of balancing. You will note that this starts with severe ARDS. Now, we've, I think, highlighted in the prior two talks, more to come, that moderate ARDS perhaps has a different set of, a different approach. Severe ARDS, even that, if it's not severe, by gosh, let's not sedate much. Let's let the patient breathe, and let's be planning for mobility. If severe, are we still in that early time? And if we're in that early time, is the patient uncomfortable? Some might argue, some have argued, for using measures of transplantary pressure here instead, that I think are perhaps just not ready for routine use in my stated, if not so humble, in that case, opinion. And then, consider sedation, if there is discomfort. Sedation, and maybe even a trial paralytic. And if it works, continue it for 6 or 12 hours with the intent of stopping it within that time frame. The justification for this coming back to data I did not describe around the development of diaphragmatic weakness, or diaphragmatic paralysis, if you will, from neuromuscular blocking agents, which is thought to occur as early as the 16, 18 hour time frame. So, before you get to that point, make sure it's really working. It also allows you the opportunity to come back and reassess the sedation of the patient, if you got it wrong, or if your train of four was wrong, or your EEG monitoring was wrong, or whatever you are using to consider paralysis. And then, at that point, you have two options, at least. One would be to reassess the patient's comfort. Because if they are comfortable after the initial 6 to 12 hours of paralysis, one might consider not re-paralyzing, and deciding where to go after a period of intentional observation. Or, progressing, I put it as a dotted line, but progressing toward proning, or ECMO. And for those of us who now, having moved from a big center to a smaller center, appreciate, in some cases, practically speaking, this is perhaps a referral decision. Now, ironically, and I'm not going to show this, I suspect this will come up later. Now, ironically, when we do ECMO, when we do proning, and Kathleen touched on this, we don't ignore the idea of mobility. We don't ignore, or at least physical activity. We don't ignore these things. Great pictures of mobilizing ECMO patients, I believe, originally from Duke. But certainly, existing in many places at this point in time. So that, even once you refer, or start to prone, or start to do ECMO, we have to frequently come back to this notion of, how much sedation do we need? Are we allowing that patient to breathe spontaneously? Of course, that has a separate category within ECMO. And then, are we thinking about mobility? It is, if you will, I think, relevant for us not to get so enamored of our numbers and our data at any point in time, but to understand our history and to appreciate how that has led us to the decisions that we are making today, and whether those, in fact, represent the best decisions for us. We have to change culture. This is happening in most of our ICUs. It has to happen in all of our ICUs, because it turns out, a lot of critical care occurs in non-academic ICUs. And then, finally, when we innovate, because we will, we have to think about this balance. Care of the ventilated patient in 2020, I would submit to you, is much more like this. That we continue to pile on new ideas, and we continue to come up appropriately and innovate and think of new approaches, new applications of old technologies, if you will, but all the while maintaining a balance between the harms that our well-intended therapies may be providing and the actual benefits. Thank you. Thank you. Thank you, Dr. Miller. So, I was asked to talk about driving pressure and PEEP, and trying to stay within our 20 minutes. I'll go through and try to do a little bit of science here at the beginning, and then try to show you some things that are more to my concept of what this looks like and to make it practical for use at the bedside. I do have research support or interest in these companies. None of them have anything to do with what I'm going to talk about today. I do have some objectives. I'm not going to read them for you. So, let's talk about what this driving pressure is, which has become so popular. So, driving pressure is really just a way to normalize the tidal volume to the lung compliance, using it as a ratio, as an index, indicating the functional size of the lung. You're trying to use driving pressure kind of to talk about the baby lung. So, driving pressure is the change in pressure from, and I'll show you a diagram of this here in another slide as well. The change in pressure is the tidal volume divided by the respiratory compliance. So, you've all seen this before. In order to do this, you have to have a constant flow, volume control breath, where flow becomes zero at end inspiration. This is the peak inspiratory pressure. This is the plateau pressure. The difference between these two is due to airway resistance, and the difference between plateau and total peak is the actual elastic or driving pressure, okay? And one of the things that I, I saw John Marini was here two days, and I'm about to show you a half dozen slides from a very recent paper that he wrote. That's intriguing, and one of the things he said is that at a very basic level, it's energy that must be involved in generating ventilator-induced lung injury. And that input of energy is needed to both inflate or overstretch, deform, and potentially damage tissue structures, and I'm going to show you some pictures of that. So, again, if you haven't had a chance to see this paper, this is really probably one of the best papers there's written about this idea, and John says, well, you know, you know, things we've talked about so far, everybody started after Argent Armor with low tidal volumes, and then there was the emphasis on, well, even at low tidal volumes, there's no safe plateau pressure. So, let's keep the plateau pressures low, and then there were some argument for high PEEP and an open lung approach. And then Amato's paper about low driving pressure, and then Gattononi introduced the idea of power, which is basically driving pressure times respiratory rate. It's more complicated than that mathematically, but at the bedside, that's what it is. So, should we target low driving power? And then there's this idea of John's, that is reducing cumulative injurious strain. So, I'm not going to make you study this. There's no test afterwards, but these are, you know, basically parts of the total pressure is related to the pressure to overcome resistance and elastance, and that can be described by the equation of motion, and then, of course, then you have flow and resistance, tidal volume and compliance, and then the presence of PEEP. And then what I think John has added to his discussion is this integral of flow over time as a potential injurious factor. I don't know if any, many of you know it or not, and it's, the papers are not all that well described. There are at least two papers in the literature of animal studies that show that pressure control is more injurious to the lung than volume control, and it has to do with how rapidly the flow is delivered. So, here's a newer diagram of the same picture I showed you before. So, this is the work being done either by the, in this case, by the ventilator, okay? So, this is the distending pressure or the peak inspiratory pressure. Here's the plateau or the static pressure, and again, these have to be done under passive conditions and constant flow. I can tell you I review a lot of papers for critical care medicine and other journals where somebody has used big data to look at driving pressure, and the problem is if you don't know if the patient triggered the breath, then you can't really know the true driving pressure, and if you don't know that there was really a plateau as opposed to just perhaps even in pressure control, the set pressure, then you still can't know the total driving pressure, and again, without the knowledge of just the set peep here in gray, but also auto peep, you can be fooled about what driving pressure really is, but these are all the components that balance the work of inflation. Again, just some definitions, and they're a little small. The ones that are, I think, important are this idea of power. Power is the energy expended per unit time, and in the ventilator world, it's product of inflation energy times the ventilating frequency, and we're going to get into why that is probably more important, okay? And then think about this idea of when we inflate the lung, some of the energy is absorbed into the lung, and what happens with that energy? So one of the things that gets talked about is the use of peep, or why is pressure or peep sufficient in some cases and injurious in others, and it does have to do with the lung compliance. So here we have a compliant lung with a pressure of 5, and then we increase the pressure to 25, and because it's very compliant, there's a high energy input, and it's overstretched. So this is what we see in the baby lung of ARDS, the nondependent regions of the lung. In a noncompliant lung, so with the stiff parts of the lung, you have a pressure of 5, and you just examine, figure this as either the chest wall or just the stiff lung itself, there's a small volume change for the exact same pressure. So there's a low energy input, and that strain is tolerated. So that's why, you know, we always look at that. Pressure and volume are inexorably linked, right? You can't give a bigger volume without using a higher pressure, but this is why sometimes the pressure is safe in one patient, is not safe in the other. Sometimes the pressure is safe in one part of the lung and not part of the other. So I like this picture. It's a little bit more complicated, but here they're making a point of this is that area, perhaps, of threshold for lung, for damage because of airway pressure, and at the bottom we're using low levels of peep, and the width of each of these boxes, okay, kind of describes the junctional regions that are capable or predisposed to having reduced or increased lung injury, and this is from paperback by Jerry Mead back in the 50s looking at how alveoli are interdependent. So as these get smaller, though, as you increase peep, the intensity increases, okay? So peep reduces the number of junctional interfaces. The boxes aren't as wide, but it increases the strain on ones that are unrecruited. So alveoli that remain collapsed, you're right in this increasing driving pressure, and in this case, high peep can be associated with high driving pressure and lung injury, and it just depends on where you are at in the alveolar units. So the start of this was basically this paper by Marcel Amato that was in the New England Journal, and they basically took, again, this is not a research original trial or a prospective trial. They took existing data from some of the ARDSNet trials and other trials and looked at 3,500 patients and found that driving pressure was, in fact, an independent variable associated with survival, and here's probably the slide that makes the most sense from this group, and on the bottom here, you see the x-axis. You see the driving pressure in centimeters of water pressure, and on the y-axis is the multivariate relative risk of death in the hospital. So they kind of came up with the idea that somewhere around 13 or 14 centimeters of water pressure, a driving pressure above that was associated with poor outcome, and a driving pressure below it was associated with good outcome. Now, this is where it gets a little bit tricky, the difference or the relationship between driving pressure and peep, and again, this is not original data. These patients weren't randomized to these kind of relationships, but here you have a constant peep with an increasing driving pressure, okay? So here you're increasing the plateau pressures. Here you maintain the driving pressure constant with increasing peep, and here you have increasing peep with falling driving pressure, and if you look at the bottom, and these are a little bit complicated, but what they basically demonstrate is that just because you have high peep here, it isn't always lung protective, and just because you have a high plateau pressure doesn't necessarily mean that you always have a poor outcome. We've seen the Lung Safe Study already once, so this is just, again, the patients within that lung protective group below 8 mLs per kilo and less than a plateau of 30, and if you look at that data related to driving pressure, so driving pressure quintiles in the risk of hospital death from 10 to 22, there tends to be an increase in the probability of death associated with driving pressure, and if we look over here real quickly, again, mortality rates in a Kaplan-Meier driving pressure greater than 14 on the top, driving, or on the bottom, driving pressure less than 14 on the top. So is it true in everybody? Is this relationship the same? And this is a paper from Gordon Rubenfeld and the group in Toronto, and they're looking at driving pressure in patients with ARDS, patients who don't have ARDS, patients or may have developed ARDS, non-ARDS, and then patients who are hypoxemic, patients who are non-hypoxemic, patients triggering the ventilator, and patients not triggering the ventilator, and what they found was they confirmed this relationship of driving pressure to outcomes in patients with ARDS, but they didn't see that in patients who were non-hypoxemic, and if you look over here to the right on the top here, this is non-ARDS, and this is ARDS, and this is the probability of mortality, and with changes in, increases in driving pressure, you end up with increases in mortality only in the patients with ARDS. So this, their conclusions from that study, that it was not independently associated with hospital mortality in patients without ARDS, and it did confirm the relationship without, of patients with ARDS, and Eddie Phan, and again, the group from Toronto looking at the papers that were available by 2018 looking at driving pressure, Marcello's original study, the Lung Safe Trial, David Chmielo, Claude Guerin's paper, and Jesus Villar's paper, and if you look, I think this is more, the most important thing I want to tell you is in these study designs, save the Lung Safe Trial, which again, was not randomized, was just a snapshot of what was happening, all these are secondary analyses of other randomized controlled trials, but when you look at these things in a meta-analysis, what you find is that you still have a higher driving pressure being associated with a significantly higher mortality, so here's all the studies together, and this would result in higher driving pressure being related to worse outcome, and here, for ones that use a low cutoff for driving pressure, still demonstrating that same effect. Now, there have been some contrary opinions, and this is the paper by Claude Guerin's group that looked at two different trials from the French studies of ARDS, and they actually found that there was no real advantage of driving pressure over plateau pressure, they both were related to outcome in ARDS, and you kind of can see how that might happen, because plateau pressure is one of the components used to determine driving pressure. Then this paper that was in intensive care medicine looked at driving pressure and mortality and obesity, and again, I don't want to go through all these studies and all the complicated issues, but here you see with obese patients really no difference with driving pressure, so we start to wonder is there something happening if the increase in, or decrease in compliance is caused by chest wall compliance and obesity. So, Eddie and Gordon wrote this editorial in critical care medicine a few years ago now, and I think anybody who's done research in critical care can appreciate these things, and that's that just because you observe a physiologic profile in survivals doesn't mean if you target that intervention that somehow that will result in better outcomes, and we have lots of examples of that, giving people blood earlier, I was telling somebody earlier, when I first started working in the surgical ICU, which was in 1980, if you got a hemoglobin less than 12, you got a blood transfusion, you know, and all, because that couldn't possibly be a good idea, right, to let somebody be anemic, and all we ended up finding out was that patients who got blood transfusions had more complications. Clearly, blood is important when it needs to be given, but when we start shooting for trying to correct the albumin, trying to correct the anemia, as somebody else mentioned, trying to get super normal oxygen delivery, all those things have resulted in worse outcomes, better physiologic variables, worse outcomes in critical cares. So, it's possible that the target's not in the causal pathway. It's also possible that the intervention could cause harm, and again, with driving pressure in particular, there's both mathematical coupling between plateau pressure and driving pressure, because one variable's derived from the other, and there's physiologic coupling, because one variable that you change can change the other. If I change the PEEP, I might change the driving pressure. So, kind of a first conclusion, then, before I go into the simpler, I think, looking at what this means. Driving pressure allows a compliance-adjusted tidal volume to allow you to individualize settings. It's associated with mortality, but we've yet to see a study targeting it, and we don't know that it's going to translate into better outcomes, and you can use driving pressure to help you set the PEEP while targeting the tidal volume. So, why does that happen? So, here's the part that I think, explanation-wise. So, there are some patients who have ARDS and don't have a lot of lung to be recruited. You might think of patients with chronic lung disease and pneumonia, and severe hypoxemia, or patients with pre-existing lung disease. When you go from this PEEP to this PEEP, you increase the driving pressure, because you haven't recruited any lung. You're just over-sending lung that's already open, and then that results in this higher driving pressure, and in this case, the higher PEEP, the injury it causes in the non-dependent lung is greater than the benefit it produces in the dependent lung. But there are patients who are highly recruitable, and in those patients, when you increase the PEEP from blue to green, you don't get this increase in driving pressure. You get a decrease in driving pressure, because compliance improves. So, this is my simplest version, probably too simple, maybe, simplified of why this driving pressure phenomena seems different in different patients. So, let me show you something that John's been doing demonstrations for me in conferences over the last couple years, and this is where he is really taking the issue of how we deliver the breath might be very important, and he makes the point that if the coffee cup is here, and I move it down a step, and I move it down a step, and I move it down a step, I can move it down a step, and then the coffee cup never breaks, but if I just take it from here, and I drop it, then, obviously, there's catastrophe, and he believes that there's a similar principle here about the transition from PEEP to plateau, and for those of you who are still high-frequency fans, especially high-frequency percussive ventilation, this might explain how giving the tidal volume with a small increase at a time slower flows might somehow impact ventilator-induced lung injury and power. And then, again, another example that's more, I think, real world is he says the lung is like this stress-amplifying principle, and again, this is from repeated. We're going to get to where it may be that respiratory rate is also a very important component of ventilator-induced lung injury, which has kind of been ignored, but it has to be correct, because if the pressure is injurious, then the more times we deliver the pressure, then the more injury it has to have. And so, here he shows the rope that we have this intolerable strain that concentrates on the fibers that are left, and then he shows a picture of this extracellular stress and microfibers in a rabbit lung, that same kind of sequence, and they have a paper that's just published in Anesthesiology, I think that's still in press, that looks at a group of animals who only get their driving pressure increased by increasing PEEP, a group that only gets their driving pressure increased by increasing tidal volume, and a group that only gets their driving pressure increased by increasing respiratory rate. And the, how quickly the ventilator-induced lung injury occurs is different between the groups, but eventually how it occurs ends up in a very similar outcome. So, when you think about this idea of power and driving pressure and PEEP and how they're all related, is there's a lot of things that are going on. So, there's, you know, things that amplify stress, the heterogeneous nature of ARDS, the viscoelastins of the lung, these stress-bearing elements that we talk about. And then, the energy per cycle, so again, it's per breath, not just the, but the flow amplitude in the profile, the plateau pressure, the PEEP, and potentially even the peak expiratory flow, perhaps modifying the peak expiratory flow might do something to prevent ventilator-induced lung injury. And I think John's point is well taken, that we've kind of ignored the impact of the respiratory rate on power and, or on ventilator-induced lung injury, but it's a key component of power, and whether or not we get ventilator-induced lung injury. So, I'm going to end with about five or six things here that I'm going to say we still just don't know. So, can we assign a direct causal role of ventilator-induced lung injury regarding mortality, and if so, what's the attributable risk? Because you have to think about it, and the next speaker's going to talk about ECMO, but to this day, very few patients die of lung failure. They still die of multisystem organ failure, and we've still yet to identify something that happens in the lung that results in that multisystem organ failure. Can we separate ventilator-induced lung injury from what's actually happening with ARDS? And how do we, at the bedside, assess and personalize the thresholds for strain, stress, and cumulative risk? I think that's really going to be important. Are we going to, which, are we going to measure pressure, flow, power, and exposure, the duration, and the number of breaths? What's the key damaging feature of excessive per-cycle inflation energy? It could be total energy, which would include peak, driving energy, energy above the threshold, and repeated respiratory rate excessive strain. And then, a number of other ones here at the end. I think we're moving in a direction of better understanding lung injury. I don't know that it will end up resulting in how we change the ventilator much. Again, I have some thoughts about that that are all unproven at the moment. But the simple fact of the matter is, is the ventilator guy, the ventilator controls volume, flow, pressure, and time. So, it has four things you can adjust on it in reality. And how those each are important in power, and how they create ventilator-induced lung injury, I think is going to be different. And I'm going to end, I think everybody showed a similar version of this slide. So, this is Eddie Fan, wrote an editorial for Restore Care, and I'm the editor, so I have access to it before it gets published here in the next two months. And he shows mild ARDS, moderate ARDS, and severe ARDS, and the PF ratio that goes along with them. And things that are in blue, light blue, are things that are proven. So, we should do low tidal volume ventilation in everybody. We should use low to moderate PEEP in the mild to moderate ARDS. Non-invasive ventilation probably doesn't work in ARDS. If it's in red, it doesn't work, or isn't proven by evidence. Higher PEEP is good for moderate and severe ARDS. Recruitment maneuvers and high frequency oscillation have gone by the wayside. Again, right now, neuromuscular blockade, inhaled nitric oxide, or prostacycline, no evidence that they can impact mortality. But ECMO clearly has some advantages, and prone positioning, as we've seen twice today, clearly has some advantages. Places that we're still looking at are extracorporeal CO2 removal, and this idea of customized mechanical ventilation, which would include things like using EIT, now that it will soon be available in the U.S., esophageal manometry, to set pressures based on transpulmonary pressure. So, I'm going to end there and appreciate your time. And we're going to move into our last presentation. When we're done, we're going to have everybody come back up here and take questions. So, if you want to ask a question, please come to one of the microphones in the aisle and tell us your name and where you're from. So, our next talk is, where does extracorporeal membrane oxygenation fit in? And our speaker will be Jacob Gucci, who's associate professor at the Perelman School of Medicine at University of Pennsylvania. Thanks. So, I'm going to jump right into it here, where does ECMO fit? Some of this is going to be science, some of this is going to be pragmatic kind of application of ECMO within your hospital. So, three disclosures, really. I perform ECMO on a regular basis. I co-founded a mobile ECMO program, which I'm still very involved with. And third, ECMO is off-label use of this equipment. So, you know, just know that going forward. So, the best trial run right so far for ECMO and ARDS is the OLIO trial. So, they randomized patients with basically super severe ARDS, you know, PAO2 to FiO2 ratios and with a less than 50 for three hours, a little bit less severe for six hours, or severe hypercarbia. These patients were basically randomized to ECMO versus standard of care. Almost everybody in the control group got prone, steroids, neuromuscular blocker, et cetera. Relatively high compliance with some of the standards of care that we all just heard about. And when you look at this trial and the abstract and the P number, it's essentially a negative trial. No benefit for ARDS. And as a scientist, that makes sense to me. You know, I can see the P value. I can see, you know, the problems with the trial were numerous, though. And it's interesting, when I go out in the community and talk about ECMO and ARDS, many of the practitioners, particularly the pulmonologists who send me their patients, consider this a positive trial. And we'll go into a little bit of that. First, 35 patients crossed over into the ECMO group. It didn't seem ethical when they designed this trial, and it doesn't seem ethical to me to withhold ECMO therapy to the control arm. And you can see this is a pretty sick group of people looking at this group. Nine of them had had sustained a cardiac arrest, seven of them had right heart failure, they were on high-dose pressors, and to even be considered for randomization to crossover, you had to have a SAT of less than 80 for at least six hours. So this is a sick group. Now this crossover group had a high mortality, 57%. Now what if they hadn't crossed over? What would their mortality have been? Some have postulated it might have been as high as 100. Certainly a mortality of 67% would probably not be unthinkable. And if that was the case, then that would have been a positive trial supporting ECMO. So that is one of the weaknesses of this trial. The other weakness is the fact that when they powered this trial, which was a very expensive trial, a very long trial, they predicted 40% mortality in the ECMO group, 60 in the control group. They were wrong on both sides. And it turns out they would have needed a far larger number of patients to power this adequately for the reduction in mortality that they were predicting. It would have been virtually impossible to run and would have taken forever. And as we've already seen from the previous talks, the standard of care keeps changing. One year neuromuscular blockers are the standard of care, the next year they're not. So this is why this type of a trial is difficult to run and even more difficult to get meaningful results out of. So I'm just going to, for the purposes of this talk, continue as if ECMO still should be part of the armamentarium. The question is, who should it be provided in, how should it be provided, or what should be provided, when, where, and why? So I'll give you an example case. So a 21-year-old male, influenza, has a moderately bad ARDS, but certainly not maxed out on therapy. So consider this patient has been paralyzed, prone, but somebody wants to be aggressive. And we've all heard of these types of cases. Some people are such believers in ECMO, they think that anyone close to severe ARDS should be considered. Place this patient on ECMO, someplace where it's not done a lot, and there are people out there who literally go to a simulator session and then start putting people on ECMO. Put the patient on ECMO, the patient dies. This is a known complication with this device, and certainly you cannot, it's hard to say that this was appropriate use of this, you know, what many consider a salvage therapy. Now consider the alternative. The same patient, but much sicker. Super high-dose pressors, being prone, paralyzed. On day eight, arrests during the paralysis, during the proning. Patient dies, being placed emergently on ECMO after they flip the patient back over, and dies the next day of cerebral hemorrhage. Now certainly this seems preventable as well. So where should our practice be? Who should be getting this? Nobody really knows. You know, what comorbidities should make you get ECMO versus not let you get ECMO? And does your criteria change by season? I get called once or twice a day during the flu season for ECMO patients. My unit's filled with them. Should I be putting the higher-risk patients on during the ECMO season when I am called almost every day about a 20-year-old who has a long life ahead of them? And this is where it gets hard. How do you learn from your one-offs? There's rare problems that you will get confronted with, or more rare. Should a patient with acute chest syndrome get placed on ECMO? I'll tell you, our experience with acute chest syndrome has been awful. But you can't just say, we'll figure it out with each call because I know what I'm doing. The problem is, there's not great data out there. The trials have very specific indications, but they don't go into every single diagnosis. Kind of these broad diagnoses of ARDS, viral pneumonia versus bacterial pneumonia. Well, what about Wagoner's? What about severe pancreatitis? What about the sickle cell acute chest syndrome we just mentioned? Nobody knows. These are all one-offs. So, a lot of us learn by experience. But the problem is also, the people that are calling me are very well-intentioned. They want to save the patients. They are very persuasive, even in the face of impending doom. I just got called recently about a patient who had a debridement of their pancreas for necrotizing pancreatitis. And if you ever take care of these types of patients, mortality is virtually 100%. Should we be putting those people on? This is why it's hard. This is why you need a guideline. It helps if some of the same people who will be calling you, the emergency room physicians, the pulmonologists, et cetera, help you design that guideline based on your own institutional data and the data that's out there and recommended by some groups like ELSO. So I'll tell you, we have designed our own criteria. It looks a little bit different than what EOLIA used. So for instance, at Penn, our relatively hard limit is 70 for age. If you're over 70, you better have been running marathons before that. Most studies have a limit of seven days. We'll push it to 10 days. If I get called about a person who's 19 who is super healthy and developed flu pneumonia and has been on the ventilator for 11 days, I'll put them on. But I probably won't put the 72-year-old on with chronic renal failure. Size may matter, probably doesn't. Basically, it's just a matter of whether I can actually get the canine in or not. Almost all of our patients who are obese survive. Cancer is another story, though. Severely immunosuppressed patients at baseline do very poorly. Cardiac arrest, they excluded those patients in EOLIA if they hadn't woken up. Most of our patients don't even have a chance to wake up if they have severe RDS. We can't wake them up because they're so unstable. So we have actually had relatively good survival in those patients with witnessed cardiac arrest. The REST score is always touted as a thing you can use. It's a score that's been validated to kind of give you an idea of what the survival will be with your patient. The problem is, if you look at this, this is a patient who has relatively poor survival. And this could be, once again, an 18-year-old who they're saying has a 30% chance of survival. Am I not going to put that patient on? I'm probably going to. We haven't found this very helpful. So we've gone through the who. And like I said, institutions should develop their criteria. And you should stick to it. And you should monitor your outcomes with that. Now the what is also just as important. And this is where the evaluation strategy may matter. The most common modality to put patients on is veno-veno ECMO by cable cannulation. And if you want to sound smart and know what you're talking about, you name the cannula for the machine. Inflow goes to the machine. Outflow cannula goes away from the machine. In this instance, you have blood drained from the inferior vena cava, goes to the machine, oxygen put in, carbon dioxide out, back into the IJ, hopefully to the right atrium in the heart circulates the blood. Very simple, easy way to place this, highly reliable. In the EOLIA trial, almost every single patient had this cannulation strategy. Now almost as commonly used at some centers, particularly in patients who were bridging to say lung transplant or walking ECMO, are these dual lumen cannulas. There's a few different types of these. But basically the patient's blood drains into one lumen of this dual lumen cannula and then the blood comes back out, hopefully towards the tricuspid valve. They can be positional. They can also have problems with suck down frequently. I don't like these devices until my patient has a little bit less severe ARDS than when I usually put them on. So these are the types of things I do when I transition a patient to walking with their ARDS or at least getting out of bed. Now there's a ton of other different ways to use ECMO, including the patient with the failing RV. You can go to a strategy where you have some blood from the outflow going to the right IJ and then to the femoral artery. This can help unload the sick right heart. And I'll be speaking more about this tomorrow. But we have had plenty of patients who have severe ARDS, acute RV failure. We bridge them with this strategy and then transition back to VV ECMO. And finally, there's VA ECMO. Some places will use this very liberally, even in patients with acute ARDS who are very unstable. This is where the outflow is going to the femoral artery instead of the right IJ. Major problem with this is if your lungs aren't working, you're not draining all the blood, patient's heart will still eject this hypoxic blood to their coronaries and their arch. This can cause major problems. This is called differential hypoxia. Some people call it North-South syndrome. And this is why veno-arterial ECMO is, at least from the peripheral standpoint, is a bad strategy for ARDS unless you absolutely have no other choice. So this is what I sometimes feel like when I'm taking calls from ECMO. If you don't know this, this is a scene from Monty Python's Bring Out the Dead. And this is something you want to avoid. But the question is, when do I call for ECMO? And the point is not to wait, like in that second case, when the patient's coding. But it's too early a problem as well. My strategy has always been to push people to do a little bit earlier. And there's a few reasons for that. One is resuscitation. Often when we are managing these patients, we're running them dry, very dry. We're using pressors to maintain the blood pressure, which can have harmful effects on the patient's end organs. And what we have found is, as soon as we're able to implement VVECMO, we can dramatically reduce those pressor loads, which we've published. Now the question is, why does this pressor load go down so dramatically? Part of this appears to be, based on our data analysis, is the ability to change the pH in these patients. You can use the sweep to lower the CO2 dramatically, correct the pH, and the pressors start working. And if you listen to people manage ARDS talk, they say, patients can very well tolerate severe acidosis. It may not be as clear as that, because often when I pick up these patients, their RVs are struggling. And this is a sample of healthy patients, a study from 2003, where they made them hypercapnic for a few hours. And if you look, their pulmonary pressures go up dramatically. And we see this in our RVs. This is a small sample of patients who underwent CABG, who they made hypercarbic for 10 minutes. And you can see these reductions in RV function. And this is what we're seeing in these patients when we pick them up. Their RVs are straining, they're under-resuscitated. And going on ECMO early will allow you to resuscitate them. Now the other benefit is ventilation. And we've heard in the last few lectures that maybe lower driving pressures and lower plateaus are helpful. And when you look at how major centers perform ECMO for ARDS are performing this, this is the Lifeguards study, where they looked at 350 patients at a large number of centers that perform high volumes of ECMO. And what they found was after going on ECMO, the tidal volumes dropped dramatically. And this was persistent throughout the patient's stay at all their centers. And it's not only tidal volume, it was driving pressures and plateau pressures. Many of these centers are looking for plateau pressures of 24. So if you want an answer of how to achieve whether this is achievable, it is with ECMO. And it may have a benefit in these patients' lung recovery. So one of the other things they found, incidentally, when they were looking at survival and factors that impacted it, delay of intubation and each day delay from intubation to ECMO significantly lowered the chance of survival. So this is another thing to consider when you're thinking, should we call earlier versus later? Now proning is possible in ECMO. We haven't published our work. This is an example of a study where they published their work. Increases in saturation, higher PaO2 to FaO2 ratios, not only while prone, but these persisted after unproning these patients. I'll tell you, our experience has been good. We proned almost all of our flu patients last year who were on ECMO in our unit. It says here no flow has changed. We occasionally did. Sometimes you'll have to change the position of these patients to maintain their ECMO flows. So to kind of summarize when I insist on ECMO, because I occasionally get the FYI call. This patient looks bad. We want you to keep them on your radar. But the question is, when do I push back and say, you know what? We should do this earlier. If I get a call and their plateau pressure is in the mid-30s or they're greater than 6 mLs per kilo, I push them to do it early. If they're showing evidence of organ dysfunction, their urine output's dropping off to zero, their liver enzymes are bumping, their lactate is climbing, I ask them to implement earlier because I think that implementation of this ECMO will allow me to resuscitate them adequately and or increase their pH so that we can allow the pressors to wean off. So where should ECMO be performed? There's a lot of debate about this. Just like anything else, if you do it a lot, you get good at it. There's an ECMO network, which is a position of white paper, saying you should do at least 20 cases a year of ECMO if you're going to do it at all, if you're going to run a center. And I know there's plenty of places that have ECMO machines sitting in their cath lab just waiting for a case, and they do one or two cases a year. It's not really clear how well those centers do with their outcomes, but I would guess it's not great. And if you're going to do it, do at least 12 of them with cases for respiratory failure. We have set up a mobile network where we fly out or drive out, put patients on all over the Delaware Valley. And it's actually been a great success for our program in maintaining experience and salvaging patients. And we have 24-7 availability of this program, which is something that's been recommended because not everybody should be doing ECMO, but they should have access to a place that can do it. And when we looked at our outcomes before ECMO and after ECMO, they were dramatically different. So the patient groups were relatively the same, same sickness score, same age, relatively same age. But what we found was a dramatic reduction in mortality, about 50% drop in mortality once we implemented this program. And part of this was because not only were we doing this a lot and we were getting good at it, but patients weren't having a cardiac arrest on transport to get to our program because they were so unstable. And this has gotten down to about zero because now we do all of these at the bedside when we go to the outside hospital. We used to go to the OR, and a couple of those patients had problems. So now we do all of our mobile ECMO at the bedside. So I'm going to end this talk with kind of summarizing why this is a debate still. You will probably not see another large randomized control trial for ECMO. Heolia is probably the last one you'll see for another 20 years. It's too expensive, and it takes too long to run the study. The other problem is ECMO is viewed as failure by a lot of people when it's actually just another weapon in your armamentarium, just like proning, just like neuromuscular blockers. And patients should not suffer injurious ventilation or organ failure because we're afraid to try it. The other problem is, just like a lot of us, we suffer the consequences of our last case. So if somebody went on ECMO and had a horrible outcome, that sticks with us. And you have to look at the data. You have to look at if you're having problems with your ECMO outcomes. Go back to some of the other things I mentioned. Who are you putting on? When are you putting them on? Do you have consistent protocols to manage them? So in summary, I think ECMO can facilitate low-stretch ventilation, which may improve your outcomes. And early initiation of ECMO may be better. Thank you. You guys want to? So we're going to take questions from the audience. I'm going to ask everybody to come up here. We've got 15 minutes left, so. Yes, sir, over here. So I had a question regarding ECMO for a patient who is in severe ARDS but is also in concurrent shock. Now, thought process, some of the shock might be because of sepsis. Whatever else is going on, some of it may be due to acidosis. In terms of rescue strategy, VA versus VV, I've heard differing opinions both ways. Some people say if hypoxia, acidosis is driving, just do VV. Some people say start with VA, then you could switch over to VAV or VV. What are your thoughts about that? My opinion is you, as often as possible, try to start with VV. It has a lower complication profile. We always do a transesophageal echo on every single patient we put on. So if they have severe RV dysfunction or severe LV dysfunction, you're kind of pushed. You have to do VA. But that should also inform your decision. If they have that kind of problem, you have to figure out if it's acute or it's chronic. Because if they have a chronic EF of 5, are they really a good ECMO candidate? But we always push for VV. Not only do you see those pressure requirements go down dramatically, you'll see their RV wake right up within two to four hours of going on, sometimes even sooner. Because that drop in the PA pressures is dramatic. You can immediately change your ventilator settings so that you are not over-distending the lung, increasing the afterload on that right ventricle. Yes, ma'am? Based on your comments on flows being injurious and being responsible for VILI and even pressure control being damaging, I'm guessing you would be against APRV. Any comments? Any work being done? So my concern with APRV and now the new version of APRV, which they refer to as TCAV, which results in a very high prolonged pressure, it's more like I was telling somebody yesterday when John Downs first came up with the idea for APRV, he called it CPAP with release, which is actually a much better term of what it actually is. But my concerns with APRV are mostly not knowing what we're seeing. And you hear the people who are proponents of APRV say, well, that's a release volume, it's not a tidal volume. And that can't possibly be true. My biggest concern about APRV, it just follows the super peep Kirby, Downs, and group at one point, the high frequency oscillation people, that when you use these really high mean airway pressures, you end up having to support the patient, just like in the oscillator trial or the oscillate trial that Neil Ferguson did. The patients who were on the oscillator had more sedation. Because they got more sedation, they got more fluid, and they got more vasopressors, and they did worse. I think if they were an APRV trial, a multicenter trial, I think we'd probably see similar results. And I appreciate some, I hope some of you know what I'm talking about. There have been a number of patient papers from Gary Niemann's group in Syracuse looking at what the lung looks like after being ventilated with this APRV or TCAV technique versus conventional ventilation. And it does look very different. But one of the, you know, the thing I can tell you for sure about those, about that technique is it results in death in every single experiment, right, because the animals get sacrificed after three days. The important thing is sometimes those animals will get 60 liters of fluid over three days to maintain their blood pressure and probably creates a change in chest wall compliance. I just think until we do some more patient trials that are better controlled, I mean, I think I've used APRV, and I've seen it work to help recruit lungs, but I think you have to be very careful that you don't become outside of the lung protective region and don't hurt the hemodynamics. I don't know if anybody else wants to weigh in. Any APRV users on the group? I used to use it before ECMO. Now I never use it. And I would say we used it more before we were proning patients. Hi. Gilberto Vasquez from Mexico. Well, I will try to be clear myself with this, all this excellent session. And many things are right now in my head. For example, you touch the very clear or clearly the continuum of treatment for ARDS patient at the ICU since the arrival to the ICU till the end, either is survived or died. From our point of view in my ICU, we try to fix the earliest treatment as soon as the patient arrives. And during this time, if the patient starts, if we cannot reach the lung protection ventilation strategy, then we turn the patient. If the patient is possible to turn in a prone position, and if it's not enough to do it, well, the next step is to ECMO. We don't have as easy ECMO at this moment, but it seems to be very clear that is the next treatment available. We cannot spend too much time treating the patient, trying to see if for the next hours the patient is going to be better. Even worse, if the team is not prepared to treat the patient and they start doing experimental or new modes of ventilation that are not familiarized, then the patient has a very high risk to be mistreated and the risk of dying. We have to remember that the patient with ARDS, the probability for dying is 30% in the best conditions, and it's very high rate of chances to die. We have just a small time or a few time to treat correctly and to continue to take the right decisions for the best for the patient. Finally, I guess ECMO for ARDS, well, it's a personal experience in the H1N1 pandemic that started in, I don't know, in Mexico, with our H1N1 pneumonia for this virus. I had two of the chances to do ECMO in two patients in our ICU. Very sick patients, yeah, as a rescue maneuver, and was very, well, at that time, ECMO was not so popular like this time, but we are talking 10 years ago. And then we did the venous-venous cannulation. In fact, I did it myself because we have no experience at all in Mexico at that time. So how to fix, why not to fix a standard procedure for initiate ECMO earlier at the ICU with venous-venous cannulation that is less complicated than the other ones? As a standard procedures, I think that mechanical ventilation is much more complex than ECMO. ECMO is just three or four buttons to move, and the difficult one is the cannulation, of course. But a very good skilled angiologist or thoracic surgeon can help us with that, and not to lose time to start a very simple or more simple procedure than highly advanced mechanical ventilation procedures. That's my point of view. I guess we have to promote ECMO much more often and in a very short time than other strategies that we are not secure. Thank you. You make a good point in terms of the algorithms that you saw in various presentations are really the stepwise, if this is not working, move to this. If this is not working, move to this. And so that we escalate that, and we don't wait too long to where we lose the efficacy of the therapy. One thing I'll just add to that is I always think of the very early phases as a great opportunity to rescue patients because after they get intubated, when they're developing hypoxemic respiratory failure, the reason that Proceva was designed how it was and the way they did the enrollment, the way they did it, was because they knew there was going to be a very real percentage of patients who would have severe hypoxemia right after they got intubated, but with appropriate ventilator management, actually would no longer even have a P to F ratio that was less than 150. So I think in the early stages, that's a great opportunity to focus on the vent instead of what is, I guess, Jack's simple machine. I guess that's what we're going to call it now. Hi, quick question. I wanted to follow up on the comment about the return of RV function and the setting of a PaCO2 level or a certain pH that you made. Were you implying that you had, with a more normal PaCO2 or pH, that you had a return of myocardial function or just an increase in your pressure and inotrope function? And what were those levels in PaCO2 or pH that you noticed those changes? Yeah, so this was not like one of those stepwise function, kind of like, let's get the pH up to 7.25, see what happens, then 7.3. We basically just, when I titrate the sweep, I ask them to titrate for, not for a CO2, but for a pH. And I ask them to titrate for 7.3. I know everyone knows that 7.25 is acceptable, but if I have a high pressure requirement and an RV that's struggling, I'd like it to be a little bit higher. I've seen over the past five or six years, just too many patients lose fingers, toes, get renal failure from kind of tolerating high pressure requirements. So I feel like getting those pressers off is better, but I don't know if it's because the pressers are working better because of the pH or if their PaO2, their pulmonary pressure are going down. It could be both. Thank you. Over here. Excellent talks there. Kathleen, I have a question for you. So I'm an intensivist in Indianapolis, and we, at my hospital, we are kind of struggling to manually prone people, and there's a lot of tendency to use rotoprone. So from your experience, do you think that, what's the cost effective analysis? And two, do you see a lot more pressure injuries from your experience with a manual versus rotoprone? Actually, both of those are great questions. The first one is there's a market challenge in the fact that when the company that originated the rotoproner sold it, the company that took it over really shrunk the market and really regionalized it. So a lot of places that only sometimes use the rotoproner literally got boxed out of the ability to use it, and it moved over to manual. The cost is there. It's, on average, from what I hear around the country, it's about $1,400 a day. And you look at manual proning, even if you had wedges, and even if you calculated in staff time, it would not reach that. So from a cost-benefit analysis, and I would also say some of the individual studies that were done in Europe, single center or a couple centers were done with the rotoproner. So we have some pressure injury data on the rotoproner. We also have it on manual. And I would say both of them are challenging. One of the things from talking with clinicians in the field, if the patient is not in the rotoproner tightly, you're going to get significant sheer skin injury. If they are in there correctly, and so, again, there's a technological component to utilization, that you'll get the same types of pressure injury that you're seeing with manual. In terms of the second question, just delving into the pressure injuries, both prevention strategies are very similar. Both of them are using protective dressings in the high-risk areas, the forehead, the chin. This area across the chest, the pelvic, elbows, and knees are usually where you're seeing those dressings placed. Did that get your? Yes. Okay. Great. Thank you so much for your fine presentation. My name is Konstantin. I am from Russia. I would like to ask two questions, two short questions. First is, do you routinely use neuromonitoring for ECMO patients, neuromonitoring? And if yes, what kind of neuromonitoring? And the second question is, do you use ketamine for sedation of patients with the ARDS? Thank you. As far as neuromonitoring, no. We try to keep these patients kind of as lightly sedated as possible, so they're not any more deeply sedated than standard ARDS patients. We do monitor limbs on VA ECMO because they can get compromised flow with cerebral oximeters, you know, reflectance oximeters. I guess the only thing I would add to the ketamine component of that is I think many of us in the room have tried it, have used it. I don't know exactly where it fits in the armamentarium just yet, but I look forward to learning more. So I'm David Kramer from Milwaukee at Aurora Health. My question is about the role for extracorporeal CO2 removal. Because your comments lead me to think that adjusting the PCO2 might be very important, ECOR is something that could be done in less sophisticated settings. Where do you see that in the picture? Do you guys mind if I answer that? Go for it. I would like to use it. The trial is in COPD patients, though, so I'm looking forward to getting FDA approved and using it. We often, in patients who are recovering from severe ARDS, have a pure CO2 retention issue, which I would love to transition those patients to that device. But I can't even trial it right now on these patients yet. But yeah, I agree with you. It would be helpful. Yes, sir. Hey, my name is David Silverstein from Los Angeles. Two questions. One related to a lot of the controversies surrounding APRV and driving pressures. Are we actually describing a failure of the Berlin criteria? And the possibility that the simplicity of the Berlin criteria fails to pick up a heterogeneous disease population. And that some patients are just going to get better quickly. And really, what we're trying to look for are the true diffuse alveolar damage patients that escalate rather quickly and need to be put on ECMO. That's number one. And number two, specifically regarding ECMO, what's your experience with bifemoral cannulations? We have done it, but only in patients who don't have a neck. I get concerned about taking up that much real estate in the inferior vena cava and the potential for venous congestion in the liver and the kidneys and the gut. Duke used to do that on a high percentage of their patients, and I think has gotten away from it. Right. It's just from a recirculation standpoint, it has an advantage. But it does. We do it sometimes. And I don't know if it's right or wrong, but I worry about it. I think, just to address the Berlin question, I would not sit here and call it a failure, but I do think it is a, your question is very apt and is appropriate in examining any therapy, any recommendation we make. I think the heterogeneity represented by different positions on this panel reflect the fact that a single PF ratio will not possibly answer therapeutics in the next 10, 15, 30 years. Thank you. Over here. I'm Karthik Jyothi, I'm an intensivist in Nashville, Tennessee. So question about spontaneous breathing, I know we had a pro-con session on that a few days ago. Do you think it's acceptable to allow for spontaneous breathing in non-severe ARDS, but for allowing higher tidal volume, so above 8 mls per kg? Like if you put a patient on pressure support, which is spontaneous breathing, but they're naturally wanting to take bigger volumes, but they're not in distress, do you think that is acceptable in non-severe ARDS? So I mean, I'll make a comment on that. I think the first question, and I'm sure you do ask yourself this, is why do you want to do that? So to me, when I'm managing patients, I will say, what is my default going to be? My default is going to be, well, there was good evidence that was generated for a group of patients, and that whole group of patients had a P to F ratio that was less than or equal to 300, and then we saw a mortality benefit in that group. So that's how I'd like to apply that evidence. If I had an individual patient who I thought required tons and tons of sedation and all sorts of other things that I could tell was really doing something bad to that patient, and their P to F ratio was 295 or something, I would likely liberalize or sort of move towards a weaning strategy earlier. Obviously, there was a weaning strategy in the ARMA trial anyway. So once you get good enough, then you can liberalize, and patients were weaned by pressure support. But I think that my personal feeling is that we kind of need to prove that the patient is not doing just as well on the therapy that we've actually seen to work before moving towards liberalizing them to a spontaneous breathing mode. Thank you. Cesar Rojas from Mexico. What are your thoughts about the randomized control trial published a couple of weeks ago in The Lancet about the use of dexamethasone in ARDS? I suppose my only thought personally about that is that in reviewing the basic science, which I am not the guy to do, about ARDS, it would appear that earlier use of steroids would likely have the greatest benefit relative to later use. And that is all I have to say on that topic. People are, this is the Villar study that was published recently that showed there was a benefit in ARDS by giving dexamethasone early. So yeah, it's going to be interesting to see. I mean, I've seen things on the critical care medicine website and blogs where people say, this is going to change my practice. I don't know if anybody feels that way just yet. So let's take one last question and then we'll be done. Sorry, I forgot to introduce myself the first time. My name's Amar. I'm an intensivist from Georgia. So my question is about sedation, proning, and paralysis. We, I mean, right now with paralysis, the data being questionable, evidence going back and forth, but we had previously come up with an institutional protocol preference, whatever you want to call it, was deep sedation, then paralysis, then proning. And some of it came with, some of the nursing staff thought it was easier to prone with paralytics on board. And I just wanted to know if you guys had any thoughts about that, where should they come in the algorithm? Should it be first, second, or, you know, in that order or not? Go ahead. You start. I guess, I don't know the answer to your question. I don't know that any of us knows the answer to your question. I would submit that the clinical utility of paralysis, while certainly in doubt in 2020, nonetheless, is part of that armamentarium. And I would further submit that, again, that outside of certain or otherwise academic medical centers, we have to entertain the use of alternate therapies early, even perhaps while awaiting arrangement of transport or referral to another place. So, I think it may be relevant to continue to use paralysis in that context. Mike, one caution would be, I don't think deep sedation clearly achieves anything that I'm personally aware of. I think once you paralyze, you are obligated to sedate, of course. But I'm not sure that deep sedation, for the sake of it, has a very clear niche, although unstudied. And I would say it's literally based on the patient, because there are patients where the respiratory ventilator synchrony and how you're managing the ventilator is working fine. And the patient will tolerate the position. I mean, I've proned patients that are awake. And I've instructed and told them. And I usually, if they're not awake, I actually ask the family, are they a belly sleeper or a back sleeper? Because I can sort of, in my original research, I was able to determine that those that were belly sleepers actually did better and required less sedation, and those that weren't didn't. And actually, I feel that's when they lie supine, too. If that is not a usual position of comfort for that patient, it is more uncomfortable. And they usually get more fidgety as a result of it. But I think the other piece is, and this is changing the culture that you mentioned with the A through F bundle in the concept of lighter sedation as our goal, and only moving in that direction when the patient condition warrants it to achieve the therapy. I would just comment, I mean, I think everyone here probably knows that not 100% of the patients who were in the PROSIVA trial were on neuromuscular blockade when they were enrolled into the trial and proned. It was about 85%. And so while it is typical, we certainly don't treat it as a mandatory thing. I would totally agree. It's going to be based on the patient. I would also just hearken back to the quote from Art Wheeler in regard to who we're treating the patient or ourselves. Thank you. Okay. Thanks to the panel. Great job. Appreciate you guys' attention. And thanks for staying.

acute respiratory distress syndrome

ARDS management

mechanical ventilation

extracorporeal membrane oxygenation

ECMO

proning

ventilator-induced lung injury

VILI

optimizing ventilator settings

driving pressure

positive end-expiratory pressure

PEEP levels

APRV

spontaneous breathing

neuromonitoring during ECMO