And good afternoon, everyone. I'm Tim Buchman, and it's my pleasure to welcome you to this year's late-breaking Thought Leader Session. Each year, the Society of Critical Care Medicine identifies and shares late-breaking articles at Congress from the top peer-reviewed journals. These are released close to or simultaneously with their presentation at Congress. SCCM recognizes the importance of and the need for the latest in critical care research findings to be shared as quickly and as widely as possible, and we use Congress as the vehicle for sharing this information with you. Starting this year's late-breaking session and to introduce today's first late-breaking article is Dr. Emma Granger. Thank you, Tim. So I'm delighted to introduce the first study today, and that is on conservative versus liberal oxygenation targets in critically ill children, the OxyPico trial. It's a multi-center, open, randomized clinical trial. And that will be presented today by Dr. Mark J. Peters, and the study was recently released in the Journal of the Lancet in December. Thank you. Thank you. Thank you for the invitation to present a pediatric trial in this conference. I'm presenting on behalf of the United Kingdom Pediatric Critical Care Society Study Group and our colleagues in the ACNARC Clinical Trials Unit. I have no relevant conflicts of interest, but we were funded by the UK NIHR, Health Technology Assessment Agency. My interest in this was started more than 10 years ago now when I was part of the Extreme Everest trip to examine the effects of hyperbaric hypoxia. One of the outcomes from that and others was this idea that oxygenation might follow the same U-shaped curve of association with risk that most other physiological parameters do. Perhaps hyperoxemia wasn't sort of uniquely safe when compared to other physiological parameters. So by both ends, there's potentially a risk. Interventional data support this. Interventional data less so, but the question for us, for children, is what is the optimal strategy for systemic oxygenation to target? I can tell you in some preliminary work that the current practice is very much hyperoxic. These are 7 million oxygen saturations from our PICU and a huge proportion of superphysiological as shown here. As I said, this has got quite a long history. We did a pilot trial about 5 years ago of 120 patients and we confirmed that the trial was feasible and there were no obvious safety concerns associated with it. On the basis of that trial, we designed the following research question. Target population is emergency PICU admissions who are intubated, newly intubated, not using LTV patients, and receiving supplemental oxygen for abnormal gas exchange. The intervention and comparative groups then are two different targets of peripheral oxygen saturation, but we don't care how you get there. You can use whatever ventilator strategy or inspired oxygen fraction you want as long as you're trying to hit this target range. And our outcome measure was informed by PPI work with families and former patients. It's a combination of survival and days of organ support. I'll say a bit more about that in a moment. Study has these adjectives. It's pragmatic, it's inclusive, it's multi-center. Our randomization was undertaken one-to-one between the two groups and the following criteria were used for minimization to try and ensure balance between our two groups. Study schema looks like this. We screen for inclusion, then we randomize, and as I've said, one-to-one to the two groups. You may have noticed that randomization has occurred before consent. This is our standard approach for emergency trials in UK research without prior consent or deferred consent, and we again have family support for doing that, which we can evidence. Our inclusion criteria are simple as part of a pragmatic trial. You had to be of an appropriate age and not a premature infant or an adult as we define it in UK for PICU purposes, and you also had to be enrolled within six hours of meeting all three of these following criteria, that you're accepted to a PICU that's participating, you are indeed intubated and on oxygen, and you're physically in the same room as PICU staff. So if you're intubated at a local hospital, that doesn't qualify. It's only when the transport team comes to the door that eligibility is confirmed. Perhaps not surprisingly, we have a collection of exclusion criteria. These reflect the equipoise of risks and benefits of oxygen at the time we designed the trial. We might do these differently now, but the big ones that we were expecting to cause most of the exclusions were acute encephalopathy and known or suspected uncorrected congenital cardiac disease. The primary outcome measure I mentioned is this combination of these two factors. So this is an ordinal endpoint, and the way this works is that we have a rank-based system of days of organ support according to the numbers of days of organ support by these predefined criteria, which we've already been using in our national registry for 20 years. And then on the end of that, we tack death, and that has the advantage, dissimilar from ventilator-free days, if you like, that it doesn't equate death to still being ventilated. It has the disadvantage or the compromise, if you like, that it means we cannot then handle this as a continuous distribution, we can only handle it as ranks, because being ventilated for 30 days is not one unit worse than death. So we have to analyze it by something called a probabilistic index, which again needs a little explanation. This is the probability, if you sample both groups, a random pair of patients and one from each side of the trial, it's the probability you've got a better outcome in the conservative group, and 0.5 would be no effect. Perhaps more familiar to you, we've also included ordered logistic regression as an analysis, and that gives you an odds ratio estimate that has the advantage we can do the adjustment for the acute physiological severity and primary diagnosis and site and that sort of thing. I'm only going to present a subset of the secondary outcomes in view of time that are listed here. In the UK, we worry about costs, and I'll report those briefly. Our sample size calculation was that we needed 2040 patients. We made an assumption of no impact on mortality from our pilot trial, which was 7.5% in that population. We wanted 90% power to detect half a day's reduction in duration of organ support, and we allowed for a 10% attrition rate in the study. Results look like this. First of all, our recruitment, we did it ahead of time and target. We had been slightly conservative in our estimate of the number of patients available. That dip in recruitment is the second wave of the COVID pandemic. Single interim analysis halfway through. Those of you who know the UK will know that the 15 sites represent a geographical spread across the whole country, which means we believe our results are generalizable to the wider UK population. Consort diagram I'll take you through. I hope you can see that 6,000 admissions at the top who meet the sort of screening criteria, a number of those were then excluded on the basis of age or had already been outside of the time window when they were screened, giving us 5,500, and half of those meet an exclusion criteria. The first, as anticipated, was acute encephalopathy, and the second biggest was uncorrected congenital cardiac disease. There were 777 who were eligible but not randomized, but that gives us 72% of the eligible patients who were included in the study, which again we think supports the validity of our findings. And it's worth noting, we were proud of the fact, this is embedded right into NHS Care, that the median time between meeting inclusion criteria and being randomized in the study was just over two hours, and just under 40% of the patients were randomized by the emergency transport teams at the local hospitals. So 1,000 patients in each group, the majority of those, consent was available for analysis of the primary outcomes, we have 92%. Patients look like this, baseline characteristics, you can just scan through there, I think I'll just highlight the fact that there's significant comorbidity in about half the patients, and anybody who does paediatric intensive care nowadays will recognize that's a common feature. The ethnicity mix and gender and age are absolutely identical to the UK registry data, so this is a representative population. The physiological severity then, just note there that the risk of mortality calculated from the paediatric index of mortality, the second line, put the mortality risk at about 3.5%, so half of what we had anticipated in the study beforehand. You'll see from the physiological severity parameters there that they are not overwhelmingly unwell, it's a population with a lot of RSV bronchiolitis, so there is significantly abnormal gas exchange in a high proportion at the top line. Okay so what about adherence, it's quite difficult to capture adherence in a study like this, we have hourly values of FiO2 and SpO2 for the first week, and we can show here, this is the time-weighted averages against the oxygen saturation, we can see that there is separation between the groups, there's a good proportion of the time spent in the conservative group in the target range that's shown on that graph. That's a little bit more difficult to see then is the separation in the inspired oxygen fraction, this is the average again, over the first seven days, and the blue is the conservative group and the red is the control group. There is a separation, it works out about 7 or 8% inspired oxygen percentage, but that only tells part of the story and perhaps this graph helps us understand it a little more. So this is a plot of the x-axis of the inspired oxygen fraction and you can see that it baselines both groups are just over 0.5 and the baseline oxygen, peripheral oxygen saturation which is 97-98%, but immediately after randomisation one hour later the groups separate quite dramatically, but notice that everybody is weaning, so you're trying to hit a moving target, although you're reducing the inspired oxygen fraction you're not yet down into target range, if we let the hours run forward you can see that process continues, this is now 24 hours, you're down in 30% oxygen, but the majority have not yet hit the target range. This is quite a complex slide, but again trying to capture adherence, the two groups here shown over the first 30 days post-randomisation, the first thing to say is that the darkest colour at the bottom is the time below 88% saturated, and that is infrequent in both groups, so we can put aside that worry that you're sort of next to a physiological limit. Then there's the grey part when you're undoubtedly adherent in the conservative group, there's also the top part, the light pink, where although you're above 92 you're already breathing air, so it's more difficult to drop you into the target range, we would consider that adherent. There's another population here where you're still in some oxygen and you're above range, but the majority of those times you're weaning towards the target range, there's only just under 4% of the time were we able to identify that patients were neither in the target range nor weaning towards it. Okay, so that's all background, here's the primary outcome, I showed you as a spectrum from one day of organ support up to death on the right hand side, that's the control group or liberal oxygenation. Here's the intervention group, there is a small shift in favour of conservative oxygenation, which does hit the pre-specified threshold for statistical significance. You are 6% more likely, 47-53%, to have a good outcome with conservative oxygenation than with usual care. If you prefer to speak in adjusted odds ratios, the same effect is visible as a 16% reduction in the adjusted odds of a worse outcome. It's always reassuring to see that the components of that outcome go in the same direction. Here's 30-day mortality, and although the numbers are small, and half what we anticipated in the trial at baseline, they also favour conservative oxygenation. Duration of ventilation is the principal component of organ support, and again there is a small, I mean really small, 3 and a bit hours reduction in duration of ventilation. We attempted to measure functional status at PICU discharge, this is difficult, partly because of prior morbidity, but also the granularity of the tools is limited, there's no signal in either direction. Healthcare costs are shown here, they're in sterling, I'm sorry it's just too unstable to do the calculation, depending on which Prime Minister we got this week, it varies too much, but it's about 7% reduction in healthcare associated costs at 30 days. We had some subgroup analyses, they show no heterogeneity, I'm just displaying the heart rate centile adjustment here, and you can see there's no visible signal. Perhaps a hint that if you're in the highest centile of heart rates, you may be close to the limit of physiology, it's possible the effect may disappear at that point. Safety reporting, there's nothing significant here, but isn't it interesting, this may be human behaviour in a trial, that all the things you'd think would happen with low oxygenation were reported more frequently in the liberal oxygenation group. So in an attempt to summarise that, we did a graphic, I'm not sure this displays the differences very much, but perhaps you can see there are slightly fewer deaths at the bottom, or slightly fewer 30 days of organ support, the large black oblongs, and there's slightly more of the very short stay in the conservative oxygenation group. Difficult to say. If we could summarise it like this though, if the point estimates are accurate, and are reproduced across 200 patients, we would anticipate seeing one fewer death, 123 fewer days of organ support, at around half a million dollars, well it might be more than that, and your bed day costs are suspected more than ours, but in the UK, 400,000 saving. So my conclusion is there is significant, but acknowledging it is a small benefit of conservative oxygenation on the duration of organ failure and death in 15 UK NHS PICUs. The secondary outcomes are also supportive of this, so I'll make the clear recommendation to you that you should be targeting 88% to 92% in the population representative of this trial. The last thing to say is I acknowledge the huge number of people who made this contribution, those on the trial committees, the parents and families who allowed us to study them. And I'll just leave you with the link to the paper if you want to hear more details. Thank you. Thank you, Dr. Peters, for a tremendous presentation. Now, the next presentation, I will tell you the paper was published yesterday along with an accompanying editorial in Critical Care Medicine. It is online for you to read after this talk in this session, and you will want to read the paper and the editorial when you hear these top-line information presented by Dr. Paul Pepe. The study he's going to be presenting this afternoon is titled Survival for Non-Shockable Cardiac Arrests Treated with Non-Invasive Circulatory Adjuncts and, Importantly, with Head-Thorax Elevation. As again, it has simultaneously been published in the journal. Paul, welcome. Tim, thank you so much for having me out here. Okay. So, thank you, Dr. Pepe. Okay. We have a slide show up here. Good. All right. So, thank you very much, everybody, for being here and just wanted to say that it is really a privilege to be up here once again with all of you. And thank you very much. I've been doing this for over half a century on the streets or in the ICUs and promoting and researching CPR. I really feel I can say this has been the biggest advance that we've had in almost six decades now. And you'll see why. I hope you agree with me when we're finished here. And thanks again for the society for having me up here. All right. So, as we heard the title of this thing, and we don't have any titles there. I used to be a distinguished professor. I'm more of an extinguished professor now. So, we'll see as we get along. Before I get started, you've got to do the usual disclosures, clinical trials, IRB, you know. These devices we're going to talk about today are all FDA cleared. And none of my speakers, I'm sorry, co-authors here, including Dr. Kerry Bacista, which is the one right next to me there, the first author. Nothing, no disclosures whatsoever. All right. Let's move on. So, this is kind of like a fun, provocative question of the day, right? Why are you upright? Why are you listening to this thing sitting or standing, you know? Why aren't you, wouldn't get better blood flow to your head if you were in a supine condition? So, I'll leave that with you as the provocative statement. And you come to your own decision after we go through this today, okay? So, let's begin. In the U.S. alone, 1,000 persons will die today from sudden unexpected out-of-hospital cardiac arrest. And CPR, as we've known it for the last, is a clear miracle of modern medicine. I've witnessed the cases and myself have brought back people from the dead, so to speak, clinically dead, with such a, you know, this action. It's been fantastic. The problem is that most cardiac arrests do not receive CPR and also the majority are not shockable cases, which are often the ones that we can get back. In fact, most, or 80% at least, are non-shockable cases, both in the hospital as well as out of hospital. And so, people will present with asystole, like a brady-asystolic arrest, or get near there with pulseless electrical activity, PEA as we call it. And generally, that's associated with longer periods of rest and that's why people have always felt that they've carried very poor prognosis. In fact, the survival odds with good recovery nationwide remain about 1.5% across the U.S. systems and those are the progressive ones that are actually monitoring this. And keep that number in mind, that 1.5% and you'll see why that's going to be kind of a fun number to remember. But here's the but part. It's not just long response times. Yes, they've been unwitnessed arrests. You'll see in almost half the cases it's asystole. But there's physiological limitations of conventional supine CPR we began to appreciate about two or three decades ago. And even if it's performed optimally and early supine chest compressions themselves, I mean they do create an arterial pressure wave that gets stuff up to the brain, right? But on the other hand, there is also a problem on the other end because we're not only, when we're compressing the chest, we're also get relatively equivalent retrograde venous pressure waves going up various venous channels up there and they collide in the brain. And when they do that, you get spikes in ICP at that time and of course that would limit blood flow not only into but across the brain as well. So, outward blood flow has been measured over and over again in various studies to be only 15 to 20% of normal and maybe even less in the brain given the situation. But what's a potential resolution for this limited blood flow into and across the brain? One thing is maybe to pull some of that blood out of the brain and back into the chest. If you think about something as simple as like a toilet plunger-like device that can not only push down but then pull up, create a vacuum, negative enthoracic pressure that helps drain blood out of the brain and back into the chest. It literally helps to pull it that way, okay? Now, another thing you just keep in mind, by the way, some of you, there are studies that show even by itself this thing works on its own. But there are mechanical versions of it. I wanted to make sure you see that because in various slides you'll see this equipment and a lot of you are already using it. But keep in mind what those have on are those plunger devices that are doing this. They may not pull up as far, and hopefully they will in the future, as the pump itself. But we'll see. It's been very useful in some of the studies we're about to show you. Now, there's another device called the impeded threshold device. And that thing is by itself, again, it's applied to the airway and it's a valve that prevents air during the recoil phase of CPR. When you're recoiling, you can suck things in. You're sucking air in. But by inhibiting the airflow for just a brief period of time, you can tug in a little bit more blood. And what we've shown actually in human studies here where they put art lines out in the field and after 14 minutes of CPR alone, they're getting systolic blood pressures of 85, which is more than double what you usually see out there in most of the circumstances. But tell you what, when you put the two of these together, it's impressive. In an NIH-funded supervised clinical trial, we got 50% improvement in neuro-intact survival in that situation. And for me, this was a true proof of concept that you can pull blood out of the brain and into the chest and get much better outcomes and better flows and so on, even up to one year survival of 49% improvement in neuro-intact survival. All right, so if lowering ICP and enhancing blood flow back to the heart is so important, right, can gravity help? Well, the answer, I think, is yes, and you'll see why. But yes has to be done correctly, and that's really critical to anything you take home today. You can't just lift the head up. In fact, lifting the head up is detrimental because you can't get blood uphill when you're doing standard CPR. You've gotta first do the priming with that ACD and ITD. You've gotta basically get active compression, decompression, ITD going, then you get enough flow, now you can get that blood uphill. And if you gradually elevate the head over a period of time, which we'll tell you about in a second, then you really can get some great effects here. And specifically, over the last 10 years, we've looked at should you just have reverse Trent Dullenberg, can you just put the head up, should you put the chest up? We found having both head and chest up was important. You actually lower comorbid vascular resistance and various other things we found in the lab. But most importantly, as you see here, you got pre-arrest cerebral perfusion pressure in the pigs, and it goes down to whatever, and there we are, 15 to 20% of normal as you would get with standard CPR. And we know that this combination of the ACD, active compression, decompression, and ITD will give you much better flows and better outcomes as we know in humans. But how long should you prime it for? Should it be for one minute, five minutes? And how fast should you elevate the head? Should that be done in two minutes, or 10 minutes, whatever it is? So we carefully worked it out, and it turns out you do about two minutes of ACD, ITD priming, and then you have a two-minute gradual elevation of the head that gets you, basically, it's phenomenal, it's synergistic, you get near-normal blood flows established there. So a device is now available that provides the correct automated gradual elevation of the head, and it gets you sort of up to, I think, around 10 inches. For those here from the U.S. crowd, just wanna make sure you guys know what centimeters are, so that's cool. But it gets you about 10 at the occiput, and the heart's about six inches up as you do it. These, again, are all FDA-cleared devices, and so people have begun to use these, all right? So, let's see now. Oh, let me just make sure you understand the point. Again, we wanna enforce that before you do the elevation, you gotta do those other things first. So what's the study purpose today? Primarily to see if such robust outcomes are also applicable for patients with non-shockable presentations. And I've done studies where we were able to get almost everybody back if they had early CPR in the first couple of minutes and shocking right away, but that's a very small fraction of people. And even those who we think are candidates for ECMO are not in the non-shockable group. So look at this. How are we doing along this line? And more importantly for me, I wanted to find out how fast this should be on. Is there an association with time to application? And I think you'll find out the answer in just a second. So, the methods where we were just comparing the neuro-intact survival for non-shockables, and those of you who wanna know what neuro-intact means, CPC1 and CPC2, and it's in the paper if you don't follow that and you'll see. And we looked at patient care data from our Head Up Registry withdrawal prospectively collected data, very comprehensive data set, and we compared it with corresponding data for conventional CPR. Now where that was taken from and purposely taken from were these NIH trials that were done and we used the controls there because they required electronic recordings, documentation that you're doing quality CPR before they even let you enroll into those studies, et cetera. So we thought that would be the most, I guess, strictest kind of comparison we could make. And just for those of you who are interested, tomorrow I'll be doing a presentation looking at the current data. They haven't changed at all. These are gonna be the same numbers, okay? So, results. Well, first of all, they're good. The survival rates with good neurological outcome for non-shockable patients. Remember, some of these have very lengthy response times already and large percentage, like they're, I think, almost 40% on unwitnessed asystole. And if you've taken all these, even a 20-minute response time or a 30-minute response time, what we found was that we were getting threefold odds ratio of getting them back under these circumstances. And remember, we told you that 1.5% number. That's what you're seeing in the gray there at the bottom. And now we're way up in the 4.5% survival rates here. And then we did, by the way, lessons learned from critical care, many of them here, have been propensity score matching for these circumstances. And what we did was, clinical trials have been a challenge. There have been so many confounding variables in the out-of-hospital setting. But what we do know are those factors that are associated with good outcomes. So this is a really appropriate way to do this, and you'll see. And it turns out, that gets you up to maybe fourfold, but you're still way up there in terms of your survival chances, okay? All right, so now, one thing I wanted to know was the time factor. Looking at the application of automatic positioning devices, as we call it, the head-up thing. And what we did is, if you could get it on within 11 minutes, by the way, which is the median time, and there's another little digression, eight minutes, 9-1-1 call, triple nine call, wherever you're from, 1-1-2-2, the start of CPR was eight minutes in every group we looked at, every study, whatever, and even today. So, but the time of application and activation of this thing was, median time was 11, so half the time, you can get there, and look what you're getting there. Odds ratios are off the charts in that situation. And in fact, since there's no more survivors after eight or nine minutes, what we were finding that, in the standard group, that if you look at, I thought quarter hour is a pretty good thing to look at, turns out 80% of the patients can get there in that period of time to them. Well, look at this, the odds ratios still climb even further under those circumstances. Okay, so one of the things I have to note here is that this is facilitated by easy carrying and a pretty correct approach, because when these guys, only two or three people are needed to get this thing there, they open it up and everything's all set and ready to go. So they've really taken it to the next level, because the faster you do this, I was here on stage in 1982 saying, the earlier their intervention, the better results. That's been my mantra, and that's what we're finding here, too. One last thing I said in terms of data, I wanna show you that, did the presenting ECG affect the scale of the differences? You know, because we talked about asystole, just, you know, no chances, and is this all, you know, these PEA cases? Well, interestingly enough, when you do these comparisons, yes, the PEA cases are 10% now versus the, you know, 3% we have been traditionally seeing nationwide and in the past, et cetera, and that odds ratio is good, but if you look at the worst case scenario, asystole, unwitnessed case, you arrive, they're flatlined, most people would say they're gone, you know, that's it. But what we were finding is that we're actually getting neuro-intact survival in these cases in almost 2% of the cases, and what's really striking about that is you do break it down, and you'll see a presentation I'm doing on Tuesday down to, like, you know, less than half the time you get there, 11 minutes, we're approaching 3%. Now, 3% sounds low, but compared to, like, no percent, it's high, and also, you're talking about 400 people a day in the United States alone, you know, this is pretty, it's a great, another proof of concept from my point of view, okay? Just those numbers alone, regardless of your methodology, is important, and by the way, we're seeing a kind of an important new signal. The guys that are doing this out of Edmond, Oklahoma, are reporting that in the past, they would get a mixture of CPC1 and CPC2, now they're seeing nothing but CPC1s. We're looking across the other systems right now, and they're seeing the same thing, but I just wanted to at least let you know that. Before closing, I do have something important I wanna relate to you, is that I'm so grateful to this organization and the people that are in it. Critical care practitioners lead EMS, often in other places around the world, and in your own ICUs, I wanna challenge you to say, hey, take this to the next level, if we can get there so fast, well, you could have it right there, put this on, and you have people instrumented, so you can tell me whether or not, like what I did in a study in the animal lab, is that we found out that after resuscitatum, you get better pulse, cerebral oximetry, you get better cerebral perfusion pressures, but I can't measure that on the streets. I would love to have you, this is just the beginning, a starter kit where you can start doing a lot of great work here. So in conclusion, I wanna say that rapid application of this triad of non-invasive CPR adjuncts is associated with markedly improved out-of-hospital cardiac arrest outcomes, and I bet you it will be in hospital as well. And it is, in a sense, it's an AED equivalent for non-shockable, automated defibrillator equivalent for non-shockable out-of-hospital cardiac arrest, but with much wider window for life-saving, as we've shown here, and it can be implemented correctly. And again, if it's implemented correctly, I strongly recommend it for all first-in responders, which could be not only firefighters and lifeguards, but ICU, emergency department, cath lab people. And on the road to the 2030s, I hope that we'll get back people we never got back before, right? And make life-saving more routine for future generations. Had to put the perfunctory slide in, the family, right? I'm Paul Pepe, and I approve this message, okay? Thank you very much, everyone. Okay, good, thank you so much. Thank you so much, Dr. Pepe. The scope of this session is for practice-changing findings, and I really think your presentation illustrated that point quite well. So, the final article that's being presented today is entitled, Ceftrioxone to Prevent Early Ventilator-Associated Pneumonia in Brain-Injured Patients. And it's a multi-center, randomized, double-blind, placebo-controlled SSMR superiority trial. So, please welcome to the stage, Dr. Claire Dio-Fazilier to present this article. And this is being simultaneously released in the Lutzer Respiratory Medicine. It was published yesterday alongside Ellen Comet. So, please welcome Dr. Dio to the stage. Thank you. Thank you for the introduction. Thank you to the organizer for the invitation, and thank you to the Lancet to select my study to present today. It's a real honor to be here today to present the results of the PROFIRAP trial, because it's a long history, 10 years of work for our team. The first grant was applied in 2013. The first patient included in the 2015, the last one in 2020 during COVID, and the publication arrived two days ago in the Lancet Respiratory Medicine. So, it's a long story, and it was a great experience, and it finished with this honorific presentation. So, here are my disclosures. As all of you know, ventilator-associated pneumonia are the first cause of infection in ICU patients, and comatose brain-injured patients are particularly re-exposed to the early-onset VAPs. VAPs increase morbid mortality, increase secondary insults in brain-injured patients. It increases exposure to antibiotics, to mechanical ventilation, ICU, and hospital stay and costs. That's why strategies to prevent VAP is very important, and that's why international guidelines proposed several bundles to prevent VAPs. One of these prevention strategies associates STD to SOD, selective oropharyngeal decontamination and digestive tract decontamination during all the ICU stay of patient, associated with short course of parenteral prophylactic antibiotics. Short course in the French guidelines means a maximum of five days of antibiotic, which represents a long exposition to antibiotic finally. That's why in ICU teams in France and Europe, we do not apply widely this recommendation because of the risk of emergence of antibiotic-resistant pathogens. Everybody is quite fear about this recommendation. Another solution has been proposed and studied since two decades. It's short prophylactic antibiotic without STD and prophylactic antibodies in innate antibiotic or IV antibiotics. If we turn to, in this metamorizes results, to the IV antibiotics prophylactic strategies, the auto-selected five STDs and four of the five STDs were performed in brain-injured patient. And if we turn to the results, IV antibiotics seems to protect patient from VAPs with a relative risk of 0.46. Here are the four STDs studied in this metamorizes. Three of them were randomized trials. Two of them with very little number of patient with no blind committed to review VAPs. And the last one on the bottom was on cardiac arrest patient, not on all type of brain-injured patient. The only one with observational cohort was compared to an historical cohort and showed a decreased incidence of VAP after one dose of ceftriaxone. Recently, another positive issue, thanks to this center TBI cohort, here again, antibiotics prophylaxis was reported to be an independent protective factor against VAP. To go further and to try to show other benefits of these strategies, we perform a multi-center randomized double-blind placebo-controlled assessor mask trial in committed brain-injured patient within the Atlan area research network. Eight centers in this research network included patients. And the main goal of the study was to assess efficacy of a single dose of 2 gram of ceftriaxone on prevention of early VAP. Secondary goals on day 28 was incidence of all type of VNP, bacteria-induced VAP, antibiotic, mechanical ventilation exposure, ICU and hospital stay, neurological prognosis, and mortality and safety. On day 60, you had ICU, hospital stay, neurological prognosis, and mortality also. Patients who were eligible were adult, comatose, brain-injured patient who needed more than 48 hours of mechanical ventilation and who were intubated for less than 12 hours. Exclusion criteria are listed here. Patients with tumor, CNS infection, cardiac arrest who were intubated for more than 48 hours after hospitalization and hospitalization less than one month. And ongoing treatment, antibiotic treatment, a predictable antibiotic prophylaxis, a predictable death within the two-thirds day of admission after admission, and the beta-lactam zoology. After inclusion and consent, patients were randomized to receive 2 gram of ceftriaxone or placebo within the 12 hours after tracheal intubation plus a bundle of VAP for VAP prevention except SDD. The primary outcome was a proportion of patients developing an early VAP. The diagnostic of VAP was defined according to the ATS definition except the threshold of the VAP with seven days and not five days. And all VAPs were confirmed by a centralized blind committee adjudication committee. The secondary outcomes were a proportion of all type latent set VAP, the global VAP, number of ventilator-free days, antibiotic-free days, of ICU and hospital-free days, type of microorganism-induced VAP, neurodegenerative prognosis with a modified ranking score, mortality and safety. 2,230 patients were eligible. We included, we randomized, sorry, to 345 patients. 165 patients in each group received allocated treatment. Three with no consent in the ceftriaxone group and eight in the placebo group. And finally, 162 patients in the ceftriaxone group were analyzed. And 157 in the placebo group. Here are the population characteristics of the population. So, 30% of patients were brain, trauma brain patient. And the other was stroke or SAH. The majority of patients, so nearly 80% was severe comatose patient with a GCS less than nine. And the time, the median time to, from intubation to treatment was seven hours. 160 VAPs were declared by investigator. And after adjudication committee, only 93 VAPs were confirmed by the blind adjudication committee. At least 28% of patients developed at least one VAP during their ICU, the 28th first day. Early VAPs were the majority of these VAP. If we turn to the primary outcome, patient receiving ceftriaxone were at lower risk of developing early VAP compared with those receiving placebo. With a decrease of incidence of the early VAP from 32% in placebo and to 14% in the ceftriaxone group. The same result for all VAP at day 28, with a decrease of incidence from 36% to 20% for all type of VAP. For the ventilator-free days and antibiotic-free days, they were significantly increased in the group, ceftriaxone group, with an increase of four days for mechanical ventilation and six days for antibiotics. For the prognosis, ceftriaxone improved the prognosis of patient, especially thanks to mortality. And with a decrease of mortality from 25% in the placebo group to 15%, with a hazard ratio of 0.6 to 62. If we turn to the microbiological findings, the four more frequent bacteria isolated in our patient were, like in the literature previously published, methicillin-sensitive staphylococcus aureus, streptococcus species, haemophilus influenzae, and Escherichia coli. As you can see, the haemophilus influenzae was 33% in the placebo and 0% in the ceftriaxone group because of the spectrum of the ceftriaxone, probably. And for the safety, three patients developed a clostridium difficile infection, one in ceftriaxone group and two in placebo group. And among the patients, 115 patients with rectal swabs, ESBL products in anterobacteria acquisition was found in two patients with the, with receiving ceftriaxone. For the outcome on day 60, the ceftriaxone decreased, increased the median ICU free days, the median hospital free days. But unfortunately, the mortality didn't reach the significance with a trend of the decrease of mortality from 30% to 20%, but it was not significantly different. So to conclude, PROFIVAB trials confirmed the protective effect of a single dose of two grams of ceftriaxone without application of the SDD in brain-injured patient. It confirms the protection of patients receiving ceftriaxone against early VAP, all types of VAP at day 28, antibiotic and mechanical ventilation exposition at day 28, mortality at day 28, and ICU and hospital exposure at day 60. I would like to thank all the investigators of the PROFIVAB trials, PROFIVAB 2D, to have included during these five years patients. I would like to thank family of the patients who accepted to be involved, to involve their next of kin in this 2D. I would like to thank also the research team and the NCCU team where I work, and the research team, PHAR2, which are involved in research of pharmacology of antimicrobial agents and antibioresistance. Thank you for your attention. Thank you very much, Dr. Deo Fusilier. I hope you have learned about some practice-changing innovations, the administration of oxygen, the use of prophylactic antibiotics in brain-injured patients, and new strategies on resuscitating patients with non-shockable rhythms. Congratulations to all the presenters. Thank you, and thank you, everyone. Thank you to my colleague, Emma Granger. Thank you for attending this late-breaking thought leader session. Have a great afternoon.

SCCM Congress

OxyPico trial

oxygenation targets

non-shockable cardiac arrest

PROFIVAB trial

ventilator-associated pneumonia

critical care advancements

prophylactic antibiotic use