Well, good morning, everyone, and thank you for staying in town, coming to the Congress on this last day on Tuesday. You're not going to be disappointed, I don't think. My name is Jerry Zimmerman, and I have the pleasure of introducing our speaker today. What we're here for is to celebrate the Asmund Laerdal Memorial Lecture and the award that commemorates Asmund Laerdal, who was the creator of the resuscit-anti-mannequins that are utilized in CPR training. And this recipient for this award is chosen annually from the SCCM Critical Care Congress faculty for that person's extensive involvement in critical care research and publishing, particularly as it may relate to resuscitation research. So Robert Sutton, this year's recipient, is the division head of critical care medicine and director of clinical resuscitation science at the Children's Hospital of Philadelphia Research Institute Resuscitation Science Center. Dr. Sutton is the Robert A. Berg Endowed Chair and Clinical Resuscitation Scientist, and he is an associate professor of anesthesiology and critical care medicine at the Perlman School of Medicine at the University of Pennsylvania. His research develops and evaluates novel interventions that are both educational and technological to improve cardiopulmonary resuscitation performance and ultimately improve outcomes following pediatric cardiac arrest. He is an internationally recognized expert in pediatric resuscitation that is ascertained by his more than 100 peer-reviewed publications, multiple NIH funding awards, his appointments to numerous national and international committees and task force concerned with resuscitation and authorship of several American Heart Association CPR guideline documents that are used to train millions of CPR providers around the world. Dr. Sutton's multiple, and I would say very significant, contributions to resuscitation science really epitomize the spirit and the intent of this Laerdal Memorial Lecture Award. So please join me in welcoming Dr. Robert Sutton, who will present to us the evolution of pediatric CPR, more than just push hard, push fast. Thank you for that introduction, Dr. Zimmerman. Thank you to all of you for sticking around here in this last session. And thank you to the SCCM organizers for giving me this opportunity. But I think the biggest thanks, I just want to start saying thank you to the Laerdal Foundation and Laerdal family for giving me this opportunity to give this award. To say that I'm feeling a little bit of an imposter syndrome here, kind of standing up giving this award is like the understatement of the year. I mean, I am truly humbled and honored to be here and hopefully enjoy the talk. So here are my disclosures. Dr. Zimmerman referenced some of these, some grant funding to look at physiologic directed CPR, and I also volunteer for the American Heart. But I want to start with just one more thank you in the words of Dr. Berg, who's there on the fourth row. CPR is a team sport, and so I want to thank many of you who are in the audience right now and those individuals at our resuscitation science center who have really kind of pushed and moved the needle over the last decade for kids in cardiac arrest. And lastly, thank you to the new SCCM president, B. Nanette Kearney, who I just noticed in the fifth row, as well as Bob Berg, who were my mentors, and all of the other individuals who work not only in our clinical resuscitation science program, but also in the lab under the direction of Todd Kilbaugh and Ryan Morgan. All right, so the objectives of this talk is to understand the justification in real world implementation of physiologic directed pediatric cardiopulmonary resuscitation. I'd be remiss if I didn't talk a little bit about some of the controversies related to ventilation during pediatric CPR with the recent guideline change and some of the hate mail coming our way. And then I will describe potential future non-invasive therapeutic approaches to titrate CPR to improve neurologic outcomes. So as a little bit of a lecture, I figured I would start with some incidents. Our best data right now for the incidents of pediatric cardiac arrest comes with the Get With The Guidelines Resuscitation Registry from the American Heart by Matthias Holmberg. This is looking at pulseless and non-pulseless arrests in A and B, respectively. And what we've seen in kind of the most recent data that we're now over 15,000 in-hospital events occur every year in pediatrics. And if you use the data from the ROC Consortium by Erica Fink and Diane Atkins, or the two studies that kind of give you another 7,000 to 8,000 per year, we're now having more than 20,000 events, CPR events, in children each year in the United States. And so for those of you that are adult providers in the audience, probably few of them with pediatric in the title of this talk, but probably thinking that seems rare, but if I run you down this little math of, let's say, 22,000 events per year, roughly a 65% mortality rate, obviously more kids survive in hospital, but out of hospital is still not great, an average age of arrest of four years, a U.S. life expectancy of 78 years. You're really talking that each time a kid dies from cardiac arrest, that we're talking about a loss of 1,000, or sorry, a million quality of life years annually in the United States. And so now that really puts us on the map, similar to the adults. And while they kind of get, you know, the adults, cardiac arrest gets a lot of the press, we are, this is a big public health problem that we deal with as well. And when you look at pediatric cardiac arrest from our standpoint, from an in-hospital standpoint, it's really an ICU problem. So this was a study that Bob did using the American Heart Association Get With The Guideline Registry, where we actually looked at where the in-hospital events were occurring through the decade of 2000 to 2010. And what we actually found is that 95% of the events were actually occurring in the ICUs. And why is that important? Well, when I start talking about physiologic-directed CPR, it means a lot of our kids that are resting in hospital are actually going to have the monitors in place to guide resuscitation. And what we actually found, if you look on the left-hand side there, it's the proportion of events. We saw this shift around 2000 and 2004, which we think was probably due to MET teams kind of the same time that this data was coming out. And it really shifted in pediatrics, these events from the floors to the ICUs. So giving this talk, I have to actually, you know, it'd be a big mistake not to mention the landmark article in 1960 where this all started. The article in JAMA that described closed chest cardiac massage. But I really wanted to show this article because not only is this the first time that we have had this, you know, in the literature in JAMA, but not only were there great survivals to talk about, but there were actually kids in the five cases that they talk about in the back of this. So there were two kids that, you know, had undergone general anesthesia. And so the report of pediatric CPR goes right back to the beginning. And even in the study right after that, 1961, looking at arterial line tracings during arrest. So this is what we've dedicated a lot of the kind of 2010 to 2020 science on. They were already doing this in the 60s. So this has been going on for 50 years. This was an eight-year-old post-op cardiac surgery, getting compressions, getting outstanding blood pressures during CPR. So this stuff works. So I want to go through my journey. So I'm going to cheat and do the evolution of what I know. So I kind of got into pediatric resuscitation science around 2005. I was a resident when Dr. Ned Carney pulled me aside and got me into this. There were two articles that came out in 2005, one by Benabella and one by Lars Wick, using CPR quality monitoring defibrillators to kind of document that our quality wasn't that great. And these were kind of the first studies that really kind of put CPR quality on the map. And these, if you look kind of where we've come over the last 15 years, there are tons of studies now associating individual metrics, rates with outcomes, depths with outcomes, release velocity. So getting your hands up off the chest to let blood flow come back. Chest compression fraction, keeping your hands on the chest. And then more recently, some higher-level analytics where we're actually looking at the right combinations and trying to use heat maps to find the right combination of rate and depth. And this really spoke to me. This was a quote. I guess as the urban legend goes, Bob told me that Max Harrywell, one of the giants in our field, at the 1996 Wolf Creek Resuscitation Conference said, performing resuscitation without measuring the effect is like flying a plane without an altimeter. So with that quote and Vinay pulling me into this, when he asked me if I wanted to be the fellow to kind of work on this in pediatrics, of course, I jumped at the opportunity. So in 2009, we published kind of our first single-center data where we actually looked at our in-hospital cardiac arrest. There were 20 events, very small study, 18 patients. And what we actually found is just like the adults, we weren't that great either. So when you look at the amount of our CPR epics that were meeting the targets of rate, depth, and leaning force, we still had 40 percent roughly noncompliance across the individual metrics. And then we were able to actually do a small study in comparison to the adults to actually associate depth with outcomes. So when we got up to 85 in-hospital cardiac arrests, we did this where we looked at associating CPR quality with outcomes, controlling for calendar year, and we were able to show that if you performed chest compressions in older children that hit the AHA target, you were roughly more than four times likely to have event survival, ROSC, and 24-hour survival. While discharge in good neuro did not meet statistical significance, you can see that they were more than three times higher, just limited by power. And so this was kind of one of the first times in 2014 that we kind of actually got on the map that we can actually collect large-scale data and actually associate CPR quality metrics with outcomes. And if I fast-forward just six years, now under the direction of Vinay and Dana Niles, we now have a large quality improvement network, the PD Rescue Network, that is actually collecting this in an international fashion across hundreds of children. And similar to that data, we still have, when you look at compliance across age groups, we're still having about 30 to 40 percent noncompliance. But this is now rapidly growing. If your hospital is interested in joining, they are definitely enrolling more hospitals, but growing so that we can actually take this data in a large-scale fashion, in an international fashion, actually associate these metrics with outcomes. And this whole time, this kind of 2005 to 2010, was really where this catchphrase came from of push hard and push fast. This was the idea that we needed to perform high-quality CPR, and it was kind of really the birth of this part of the science in the early 2000s. Well, one of the things that I always felt weird is I spent a lot of time actually trying to associate these metrics with physiologic parameters or outcomes when it was more like, well, why aren't we just focusing on the physiology? And so around 2015, this came out. We had a collaboration with these individuals to actually look at some heterogeneity when you looked at depth of compression and the actual blood pressure that was achieved. So this was a small study. It was one of the first ones. It was in and out of hospital cardiac arrest. It was about 39 patients, 42,000 compressions. And what they found was there's substantial heterogeneity to kind of hitting these metrics. And so what I'm showing you there on the left is depth of compression versus the diastolic blood pressure obtained. And when you look at an overall kind of guidelines fashion, do the model, do the most good for the most people, on average, as you push deeper, your blood pressure does improve. And if you're that person, please keep pushing harder. If you're this person, please stop helping me. And so you really have to pay attention to your patient's response. And the other thing that I really liked about this study, and I'm not up here arguing that everybody should be interrupting CPR to try to put in arterial catheters, but they actually published this in this, showing that they were able to get radial artery recordings from the time of the arrest to the time of the measure in a median of four minutes. It is something possible. It's something that we kind of talk about at CHOP when we're in resuscitations and somebody has their hand on a pulse, we're usually like, well, aren't you using the other hand to actually put a catheter in the artery? Stop taking up real estate if you're not going to do something. So it is something that we think about. And it leads me to this kind of rescuer-centric versus patient-centric CPR. And so the metrics, as we talk about, that's really monitoring how that person is doing. Are they meeting the right rate, depth, release velocity, getting their hands up, and minimizing interruptions? And this is that kind of most model for the most good. It's the push hard, push fast model. But the person that we should really be shifting the focus to is the one actually getting the CPR and start looking at their physiologic response. And this has really been a focus, probably since the 2015 guidelines, to look at coronary perfusion pressure, diastolic blood pressure, end-tidal CO2, or even cerebral oxidation monitoring. So I'm not going to talk about NEARS. And if you've seen this joke, I apologize. It's just too good not to show again. So I'm not going to talk a ton about NEARS, because there was this landmark study that came out in 2018 that showed that NEARS have similar, sorry, vegetables have similar NEARS measurements to healthy humans. One of the things that I was taught by Bob is when you're interpreting studies, you always have to look at the control group. In this case, it was cardiothoracic surgeons and anesthesiologists. So I will let you be the judge of whether or not this is a good study. But sorry, no NEARS data. So the whole idea of monitoring physiology during CPR really came out and was emphasized in the 2013 consensus statement that was authored by Pete Meany, where in the back, you have to look for it. But they talk about monitoring the patient's response to the resuscitation effort using coronary perfusion pressure, diastolic pressure, and end-tidal CO2. And that data, kind of the first data to really push the idea of monitoring coronary perfusion pressure comes from this classic from JAMA in 1990 by Norm Paradis, where they looked at ED and pre-hospital arrests, about 100. And they actually found that if your coronary perfusion pressure never hit 15, there was no ROSC. But as you kind of hit these higher categories, the probability of ROSC or the percentage of patients with ROSC improved. Fast forward, what is it, 30 years, and the 2020 guidelines now have a recommendation for kids. And we have actually focused on diastolic blood pressure rather than coronary perfusion pressure because it's a little more readily available. And this data actually has come out of the Capcorn network. It's an NICHD-funded network. And we looked at 164 children who had an ICU arrest. They had catheters in place at the time of the arrest. And when we set out to do this, we really picked two a-priority cutoff targets. So in infants less than a year, we were looking at 25 millimeters of mercury. And those greater than a year, those older children, we were looking at 30. But we did use a spline analysis, and that's what I'm showing you there on the left, mean diastolic pressure versus probability of survival. And you can see as the diastolic blood pressure improves, the probability of survival improves. Interestingly in this group, and when I talk about the animal data, I'll kind of refer back to this. The optimal cut point that discriminated survivors between non-survivors was actually 34 millimeters of mercury. But when we ran our model with our a-priority targets there of 25 and 30, we found that survival was more than doubled when you hit these DBB targets. This is clinical and multicenter, but observational. And also just kind of the derivation set. And so fortunately, NHLBI has now funded a multicenter validation study with nearly three times as many children, over 400 kids, where we actually looked at those same targets in a different sample. And we were able now to actually show that in a validation cohort, that both survival to discharge and ROSC are significantly improved if providers achieve these diastolic blood pressure targets during CPR. Got very close on survival with favorable neurologic outcome, but did not meet statistical significance. So if blood pressure matters during pediatric CPR, and so we've moved from kind of the depth and rate quality to over to more blood pressure quality, how are you going to get there? And so I think the first thing you can do is actually adjust the mechanics. So this was actually the study in 2013 that got me thinking about maybe I'm kind of got this graph shifted the wrong way and I should be focusing on the blood pressure. So we actually used CPR quality monitoring defibrillators and actually associated individual chest compression depths with arterial blood pressure. We expressed the depth and the percent anterior posterior chest depth as we do in the American heart, one third of the AP chest depth, and what I've kind of highlighted there for you is that target. And you can see is that the percent of AP chest depth, deeper chest compression, you actually get better blood pressures. But what you can also see here is actually as we got in the deeper side of this, you also started to see a little bit of an idea that the blood pressures were starting to dive, that maybe there could be too deep of a compression, and actually looking at the physiology might be a better way to go. But we took this and we actually took the regression curve that came out of that relationship between depth and pressure and actually built it into a mannequin, see if I can get this to work. Oh, there we go. And what the mannequin did is when you pressed on it, it actually gave you an arterial line tracing. And so the idea is we could actually train people to a blood pressure target that was based on kind of the depths that were associated with blood pressures. And we used our rolling refresher training that we were using, and Heather Wolf, one of our faculty, was actually able to show that if you trained people to a clinical outcome, 12-hour retention and three-month retention was significantly better compared to just training people to a certain depth. And it makes a lot of sense, right? We can all remember what our patient's blood pressure is supposed to be, but when the guidelines change every five years or every two years now on a continuous basis and that depth changes, it's hard to remember what the actual target is. So if you train a clinical provider to a blood pressure, they can remember it. So one of the ways you might be able to kind of improve physiology during CPR is actually adjusting drugs, and maybe that's more or less. And this really gets into the idea of physiologic-directed CPR. So we've been looking at this in our lab for a long time now, over a decade. Our primary model is an asphyxia model, where these animals undergo seven minutes of asphyxia. These are anesthetized animals. We clamp them to a tracheal tube, and then at seven minutes, we induce VF so we can study the CPR techniques. And we have two different techniques that we use. One is our standard care, so that's doing kind of the standard AHA care, one-third AP chest depth or greater than five centimeters in the adolescent-adult model. But in the hemodynamic group, we actually adjust our depth of compression to assist on blood pressure, 90 for the smaller pigs, 100 for the adult model. Two minutes in we start giving drug and we give it every four minutes just as Piles would tell us to do in the standard care group. But in this kind of blood pressure directed group we actually can give epinephrine more frequently upwards of every minute and if two doses of epinephrine don't work we actually rescue with vasopressin and then we kind of repeat that cycle two minutes after the vasopressin dose. We'll do this for ten minutes, start shocking the animals and then after another ten minutes we'll end the experiment if we don't get ROSC. Everything else between the animals is similar, the chest compression rate, the ventilation rate and the FiO2. And what we have found is that not surprisingly if you target coronary perfusion pressure during CPR over time you're able to improve coronary perfusion pressure. This is that first ten minutes of resuscitation and while in the guideline care group and kind of the light gray dash line there while they get a couple bumps in their coronary perfusion pressure with epinephrine administration it never quite hits our target of 20. This corresponded to a survival benefit in these first experiments, 70% good neuro and the blood pressure group compared to a complete lethality in our death-guided group. And we've done work, a lot of people in this room have done work in different models. We've done work in pediatric asphyxia models, adult asphyxia models, adult VF models. This was work done by Miriam Name, Stu Freese who's now at Wash U and continues under the direction of Todd in the lab. But we also wanted to look at a shock state as we know kids suffer that and so we used an LPS model and the first three animals that we did with HDCPR died. It was clearly wasn't going to work and so pulmonary hypertension seemed to be a key physiologic roadblock. And so it was that time that Ryan Morgan in his K23 looked at us and the mentors and was like, hey dummies you're focusing on the blood pressure why aren't you focusing on the pathophysiology? And so to focus on this pathophysiology if you're a pediatrician what do you do? You add nitric to combat pulmonary hypertension. And so what I'm showing you here is there's kind of those same first ten minutes of resuscitation and the higher graph they're getting above the the green target of a coronary perfusion pressure of 20 is actually adding nitric on top of our blood pressure directed therapy. And with that not only did it improve the coronary perfusion pressures we actually went over to have a survival benefit in a Blue Journal paper for Dr. Morgan. One of the things we don't use in our lab but we actually do sometimes use at the bedside is actually an epinephrine infusion. So there are times when we're doing a resuscitation if somebody's on an epi infusion we will take it all the way up to one micro kilo per minute so basically a code dose every 10 minutes. And this was work that was presented by Bob Neumar from Johansson. It was published in Resuscitation in 2003 where they looked at laser Doppler determined cerebral blood flow in 24 animals and compared kind of both groups got a bolus but then one just continued in a bolus fashion versus the other one actually started a continuous infusion. And not surprisingly just kind of visually you can see there's a higher error under the curve under that infusion group and it was statistically significant. So this is sometimes we do use this clinically when we're at the bedside. But this isn't I'm not standing up here I know that there are microvascular effects from epinephrine front in the cerebral circulation so there may be what we need to do is actually find a subset of patients that would actually benefit from the epinephrine and and what we have kind of noticed is maybe you know sometimes that be works and sometimes it just doesn't and so trying to find those patients is really I think where the next bit of science is going to go. And this whole idea really came out the first publication that I saw on this was in 2019 from the group at Hopkins where they looked at their animals and found that when you looked at survivors and non-survivors survivors are in the blue triangle non-survivors in the red circle they kind of clustered together from a diastolic blood pressure response after the adrenaline dose. Here are the non-survivors kind of in that lower category and then the survivors all had better DBP responses. And if I just kind of do the math for you if you had a diastolic pressure response less than 10 millimeters of mercury those animals about a 1% rate of survival versus 54 and kind of these higher responses upwards of almost a hundred percent when you had extreme diastolic pressure responses. And so we took this and actually are now looking at this Ryan is doing this looking this in a multi-center fashion as a secondary study to the ICU recess trial which I'll talk about in a bit where we actually look at the first epinephrine dose and our patients these are pediatric patients their clinical response to epi. We deem these responders versus non-responders. The graph that I'm showing you there is systolic on top diastolic on the bottom and our responders are kind of the dashed line or the dotted line on the top there you can see that those ones really get a nice response if I kind of hone you in on where the epinephrine was given versus kind of the non-responders and the dashed line they just kind of stay flat. And so we defined a responder as five millimeters of mercury increase when you kind of give your epinephrine within you know the first minute or so and when you look at this there's a much higher DBP response and responders and it actually ROSC is 1.6 times more likely if they respond to that first epinephrine dose. We took it one step further and we actually looked at different cut points and what we actually found is if you actually have higher diastolic response to your epinephrine dose you're now more likely to have improved survival to discharge and improve survival with favorable neurologic outcome. So Ryan has submitted this adjournal and this is something that we're going to look at in the future of how you may be able to pick out these patients and determine who's going to respond to the drug so that you actually get the benefit and not necessarily some of the side effects. So is it the intensivist mindset right if it works we should be giving more. So this was work I'm saying that as a joke but nobody laughed so apparently it fell on deaf ears. So this was work that was done by Martha Kinsley at the Children's Hospital Philadelphia using our single center database where we looked at our patients who had at least two doses of epinephrine and we looked at frequent epi administration so less than or equal to two minutes versus greater than two minutes. So these are both kind of more than what is currently recommended in the guidelines and we looked at 125 patients in the PICU and CICU. So you're asking like why y'all really give epi that frequently. Yeah we do. There was some bleed over from the way we did it into the lab to the bedside and so we do give epi in certain cases this frequently about 25 percent of the time. But when we looked at this in a model adjusting for the things that you would expect age rhythm time of day time to epi administration patients who actually had frequent epinephrine administration were more likely to have raw survival to discharge and survival with good neuro. This was completely I shouldn't say completely is about 70 percent mediated by the duration of CPR. So the idea was we give them a little bit of drug up front actually got them back faster and cumulatively these patients actually get exposed to less epinephrine overall. So but we try to tie it back to the lab because if we're the DBP people in the lab we should actually look in the patients and so we were actually able to show on the left panel there you have the interval greater than two minutes on the left. No change in diastolic pressure if you wait more than two minutes to give your epi. If you give it less than two minutes we had a significant increase in diastolic pressure and in panel B I'm actually comparing the deltas and they're statistically significant. So this gave us a little bit of physiologic kind of justification of where how we were getting these patients back faster. But again this isn't to say that you can keep doing frequent epi and actually having any benefits. So in the same study we looked at probability of Ross curves over time and if you see where these curves really splay out it's really in kind of the first six minutes so about three doses and at that point they're all kind of similar right. And so it may be that the benefit is kind of up front up to three doses but after that occurs you probably should look at exit strategies if you haven't gotten your patient back maybe ECMO. And so we took this back to the lab and you're kind of getting an idea of what we try to do is we have an idea clinically it goes to the lab or an idea in the lab it comes clinically. We looked at this in piglets this was work done by Konstantin Mevrudis who's a CT surgeon at CHOP. Looked at one month old piglets and an asphyxia model this was the AHA high quality model not the blood pressure directed and what I'm showing you here is the individual epinephrine response pre and post epi for five doses. And you can see it from pre epi to post that actually really drops off again between three and four doses so it does look like the effects or the benefit of epi does kind of diminish over time as well. All right so I've talked a lot about pressure for those of you in this audience you probably know pressure is not flow and so I think this is when we start talking about end tidal CO2 as a CPR quality monitor. So the first stuff and I like to use one of the classics was from Sanders and Jim of 1989 where they looked at 34 patients of adult in and out of hospital cardiac arrest and this is kind of where you found that or we had that idea that low end tidals less than 10 were associated with mortality. None of the survivors in this study actually had end tidals or should say they all had end tidals greater than 10. From a pediatric standpoint since that's what we're talking about we've had a little bit more trouble and again it took another 40 years for us to get this data and I'm off the center fashion but this was data that came from the Capcorn Network where we looked at 43 intubated children was a small study to start and we looked at that AHA target of trying to keep your end tidal above 20 during CPR was not associated with improved ROSC or survival to discharge nor was the mortality kind of target of an end tidal less than 10 associated with bad outcomes. Fast forward to 2022 and we have some work coming out of Hopkins this is single center where they looked at a mixture of in and out of hospital cardiac arrest. They used a ZOL device to look at end tidal CO2 and their primary predictor was the median end tidal CO2 in these events. Here is their patient flowchart they started with 460 events and got down to 143 events and 2,800 minutes of end tidal and actually CPR mechanics data and while they weren't able to kind of associate with that target of 20 they were able to find a statistically significant difference between patients who did and did not have ROSC so the end tidals were higher on the patients with ROSC. This is some unpublished data that I'm sharing that hopefully I'll have some time to get to journal at some point but this is a cap again a Capcorn study NHLBI funded where we looked at 234 children. We primarily evaluated that AHA target of an end tidal greater than 20 and encouragingly we were able to find pretty strong benefit for both ROSC and survival to discharge if you were able to keep your end tidal above 20 during pediatric CPR including kind of looking at an area under the curve so the percentage of epics that were actually at the target both of these were significantly associated with improved ROSC and survival to discharge but in this same group we saw no association with low end tidals and outcome so in pediatrics we think about this a little differently probably because we think about exit strategies like getting on to ECMO and so the group at Hopkins similar to how we have been working on kind of blood pressure directed CPR the group at Hopkins has really been looking at an animal model to kind of titrate the technique of CPR to improve end tidal and improve outcomes so their group has a feedback optimized group is their control group they use depth markers they use coaches they really try to make sure that they're hitting the targets from kind of the kind of standard CPR and then the other group they adjust their technique primarily chest compression rate to optimize end tidal and this their model has a little bit longer asphyxia this is a neonatal model it's 20 minutes about six minutes prior to starting BLS they induce fibrillation similar to our group so they can study the CPR and they kind of have a BLS portion up front and then they start their ALS portion but the chest compression aspect is being altered to maximize end tidal in that end tidal guided group and what they've been able to show the end tidal group there in the circles compared to the squares in the standard group that they're able to improve physiology during CPR mean arterial pressure and cerebral perfusion pressure is what I'm showing here and then they can actually that has corresponded to an improvement in survival more than 30% improvement in ROSC they've now taken this and they've actually moved it on to a pediatric model this is also an asphyxia model they all stay kind of adjusted the target that they were going for so they looked at their old or kind of their earlier pigs and found that a target of 25 didn't discriminate survivors from non-survivors enough so they up the target and actually what I liked about the newer work that just came out in 2021 is we actually have their algorithm and kind of how they're going through to adjust their end tidal so in their BLS portion I told you they started with BLS they actually started a chest compression rate of 100 and then every minute they up their chest compression rate by 10 up to a max of 150 they're not giving drug yet but then when they move into the ALS portion if their end tidal is low they will actually move towards a more frequent epinephrine approach to try to get the end tidal up and with this they've been able to show that obviously if you're trying to maximize end tidal and their algorithm directed CPR and the triangles there they have better end tidal co2 during these 20 minutes they have higher chest compression rates because that's primarily what they're adjusting they have higher cerebral perfusion pressures and that ultimately when you looked across all the animals the 20 animals they've done was about more than a 40% improvement in ROSC when they titrated to end tidal during CPR so this kind of set up a little bit of a showdown which was great because when we did the P's critical care colloquium in Hopkins we kind of talked about end tidal versus blood pressure directed CPR and so dr. Morgan actually looked at our animals to see what better discriminated survivors from non-survivors was an end tidal or diastolic pressure we looked at 60 animals and what we found was kind of not surprisingly in our model since we were targeting blood pressure that's actually what we targeted it definitely discriminated survivors from non-survivors better that was true across all models both asphyxia and VF when I'm showing you there on the left are actually ROC curves with diastolic blood pressure having an error into the curve 0.82 versus 0.6 for end tidal but what was interesting what I liked in a lot of this data is the optimal target to discriminate survivors from non-survivors and our pediatric animals was 34 millimeters of mercury which was exactly similar to what we saw when you used the pediatric data and took all patients so one of the other things that I thought I would talk about was leveraging technology and this is not something we're doing yet at least not to the extent that I'm going to talk about but there is some work that was started by Felipe Turan when he was at University of Pennsylvania he's not Cornell where he actually these are adult arrests where they actually drop a te probe during cardiac arrest to try to look at hand positioning which are actually pushing on and so in his first study what he it was a small study of 17 cases what they actually found using this long access gastric view was that we're pushing on the LVO LV outflow track frequently is more than 50% of cases and what they would do is they would actually position use the te to kind of move their hands around so that they would get better flow so this is actually an echo that Felipe shared with me from some of his initial work and if I kind of zoom you in here on the LV outflow track you can see that those chest compressions are occurring right on the LV outflow track and just smashing the valve during CPR and so what they would do they would adjust their hand position under te guidance and now you can see all the valves are open and things to be moving a little bit better during CPR and recently just came out in 2019 there's actually some hint that we may be able to use this to improve outcomes so this was a study of refractory out-of-hospital cardiac arrest so these were patients that were coming in these are adults that were coming in for eCPR small numbers 19 patients but they use te to kind of adjust and then look at the whether or not the LVT was open during CPR and its association with the outcome kind of an indicator and so on panel a there you can kind of see more black the LV outflow track is open and panel B you can see it's closed and what they found in this small number some of these patients had Ross some different but when you looked at survivors versus not survivors the LVT was open at a hundred percent of the survivors and only one patient or eight percent of the non-survivors so it's something that we're not looking at in kids I think there's some some concern about putting a te probe and rupturing the back of the heart or the esophagus but it's definitely something that may come down the pike in the future so we tried to put this whole concept all together with a large trial recently called the IC recess trial so I'll talk about it briefly the IC recess trial was a randomized trial of physiologic directed point-of-care training and debriefing for cardiac arrest and so what was the actual intervention well I haven't talked a lot about this but one of the brainchilds of V and I was actually this idea that you could do point-of-care training for CPR to improve retention talked a little bit about it with our blood pressure directed mannequin and if you have an RQI program at your hospital some of the work that actually supported that was done at CHOP why why did we put this in part of the bundle well there was pretty good work that this improved skill acquisition and retention and then I showed you the study that Heather Wolf did were actually training to blood pressure improved retention but what we did at these individual trainings in the trial was actually review the targets that I talked about earlier in the talk so the diastolic greater than 20 greater than 30 and older children and kind of talked a little bit about how you can use your art line during CPR as a physiologic quality monitor and in the individual sites and I'll talk about a little bit more of the design but they had to do 60 of these every month to be compliant with the intervention the other part of the intervention was actually a physio physiologic focused multidisciplinary debriefing everybody you know there's been a lot of talk at this conference and others that debriefing helps with cardiac arrest and improves quality and outcomes we did this to really focus on the physiology and we did it across all parts of the cardiac arrest did you need help at the individual site from early recognition getting on the chest faster did you need help maximizing the physiology during CPR or did you need help with post arrest management so arterial blood pressure or temperature control and it were part of the bundle because we did a single center study that actually supported this multi-center trial where we put this this bundle in place at chop and we saw near doubling of good neurologic survival and this was published in critical care medicine by Heather but the the real big thing that we did here and I hope is going to lead to some some newer work and hopefully try to answer some questions as we build a physiologic waveform data collection platform so all of the sites there were 10 distinct clinical sites 18 ICUs they actually submitted research quality waveforms to us through their either their central monitors or some platform that they had homegrown or sick bay or something like that and we were able to get these at chop where we actually had a team who would manually review each one of the waveforms and so what I'm showing you here is kind of a cartoon of what we would do ventricular fibrillation it starts at point A we would annotate that as a rest start at 228 seconds we would then you know put when we started CPR there point B at 247 seconds the start stop of CPR but then through the entire event upwards of 20 minutes our team would go through and hand annotate was it a real pause in CPR ie was the art line flat was there some underlying ROSC that we should annotate so some bumps on the arterial line, or was it just bad data that we didn't think should be included? And so we did this, and we now have data on 400 children with invasive arterial lines and all the other waveforms that go along with that, end-tidal CO2, ECGs, and pulse ox. But what we did for the trial was actually make report cards. And so we would take this data, we would come to 30-second report cards, and we would actually feed these out weekly to all of the sites so that they could see not only how they were doing from a depth, rate, and fraction standpoint, but also how were you doing from a physiologic standpoint at the unit level. And that's why they're bigger at the bottom of the screen there. So what was the percentage of events or epics that were actually hitting these diastolic targets that we previously talked about? And as I said, we did this across 18 distinct ICUs in the United States. It was 10 sites. It was a mixture of general medical and cardiac ICUs. The core was the eight sites of the NICHD-funded CapCorn network, and then we added a couple sites on to achieve our enrollment goals. We used a hybrid stepped wedge, clusterized, randomized trial. What do I mean by hybrid? Well, in this situation, you actually have two sites that actually never went into the intervention. They remained controlled the entire time. And you had two sites who were always in the intervention, so you actually stepped them over to the intervention before you start collecting patient outcomes. And then the other sites actually stepped into the intervention in kind of a classic step wedge trial, which we actually deemed them to be transferred after they had actually done their fourth debriefing. So this is a compliance chart to give you an idea. This was a four-and-a-half-year trial. Darker here is better. You want this to be black. From a debriefing standpoint on the left side, so through this four-and-a-half-year study, we did 440 of these post-cardiac arrest debriefings at an interval in these sites of about 30 days, which is what we were hoping. We did over 21,000 of these bedside physiologic trainings, about 50 per month across these ICUs. And I think everyone can probably know what I'm denoting there around February of 2020, what happened right in the middle of this trial. And so we definitely had a pause. So from when COVID hit, we actually had to move our debriefings to a virtual platform. I will say that actually was relatively seamless. We kind of all did that pretty easily, although you'll see that our compliance was never as dark after COVID hit. The point-of-care trainings, not surprisingly, everyone was busy and nor did they want to be right on top of each other trying to do these bedside trainings. So we took a three-month pause, kind of met with the sites, what we thought was safe, and we kind of got the programs back up and running. But you can see it looks more speckled there after COVID hit, as opposed to being completely black for the first three years of the study. So what did we find? Well, it turns out that this was a negative trial. When we looked at our primary outcome of survival to hospital discharge with favorable neurologic outcome, there was no difference between the groups. But when you look at these numbers, for those of us that do in-hospital pediatric cardiac arrests and ICU patients with arterial lines, survival rates approaching over 50% in the control group was just phenomenal. What we were able to do is show that this intervention, this bundle, was able to get people to our process of care variable. So there was a significant improvement in the amount of CPR that was achieving the diastolic blood pressure targets, 10% improvement. Interestingly, although it wasn't focused on, there was actually, people were not ventilating as quickly in the ICU recess group. Still though, if you look at AHA compliance, if there's any NIH reviewers in here and think that we actually targeted compliance, please see that this was 3% and 6% compliance. Pediatric still has a lot of trouble getting to the compliant ventilation rates. But what we were also able to do is show that we were able to improve post-arrest care. So post-arrest systolic hypotension, which is associated with outcomes in kids, was actually less common once we kind of did the ICU recess intervention. So why weren't we significant here? And so what we kind of interpreted this as, unfortunately, when you think about this, all of that work, you probably heard me say Capcorn Network about five times, all of the preliminary work, all of the observational work that supported this trial was done in a group. It clearly seemed to have drank on the Kool-Aid that physiologic-directed CPR or paying attention to blood pressure during CPR was the way to go. And then that was further amplified by the fact that the PEDS guidelines took the data from the lab, took these observational studies, and actually started talking about diastolic pressure starting all the way back in 2015, kind of in the middle of this trial. And so when you look at this from a numbers standpoint, if you look at our control group, which happened after these previous Capcorn publications, when we designed the trial, did the power calculations, this group was actually hitting these DBP targets about 62% of the time. And our control group during the trial was over 80%. And when you look at the survival rates, which we had great data that was published, it was a 40% survival rate with good neuro. And by the time we started the trial, it seems the control group was up to over 50%. So it seems like we may have had some control group contaminations that kind of led to some of these negative results. But either way, for those of us that were involved in this, we're quite happy that more than 50% of these kids were surviving during the trial. So sometimes it is good to just be part of a trial. All right. So I'm going to move on and talk a little bit about ventilation. In 2017, when we were working on the guidelines, there was an investigation of chest compression only versus chest compressions with rescue breasts for kids. At that time, there were four large database studies, two after the 2015 release, which is why we re-looked at it. And they really reaffirmed that chest compressions and ventilation should be provided for infants and children. But if you're unwilling as a bystander, some CPR is better than nothing. Some of the work has been done by Miriam Naim in this. This was in JAMA-PEDS. This was looking at bystander CPR from the CARES Registry. And really here, I'll just highlight it. It was conventional CPR with rescue breasts that was actually statistically significantly different from no CPR. And so this really kind of emphasized that we still need to breathe for kids. All right. Well, how much do we need to breathe for kids? So everyone talks about kind of a slow and controlled ventilation approach. And this really comes from some of the landmark work that was done by Dr. Tom Ofterheide in Circulation in 2004. This was a piglet study where they actually looked at three different ventilation rates during CPR, 12, 20, and 30. And not surprisingly, as you kind of squeeze the bag more, blow up the chest, the intrathoracic pressure goes up, the right atrial pressure goes up because it's sitting in the chest. And then because that's a downstream pressure of coronary perfusion pressure, your coronary perfusion pressure drops. They had observed professional rescuers kind of at the same time to frequently hyperventilate. And since that kind of corresponded to decreased survival, prior to 2020, we really talked about 10 breaths per minute for all ages. But for any time we've looked at this in pediatric large studies, it just seems that pediatric providers are just not willing to do that. There's something about the fact that you're bagging a child at 20 or 30 with high pressures and their heart stops, and now you've got to drop down to 10. We physically don't have it in our body to be able to do it. And I think this is some of the reasons why, right? So from a pediatric standpoint, we have more respiratory arrest. We don't have that nice oxygen reservoir from a sudden arrhythmia. Our chests of our kids are more compliant. So maybe they're not actually going to have that detrimental effect of kind of blowing up the chest when the adults are a little more stiff. And so to kind of know that there are differences in the physiology of how they're built and why they're arresting, I think it's always been hard for us to actually do this. And so it didn't make sense for that last line that it was the same for both. So we took a look at this in a recent Capcorn study. Again, this was a small study, this was about 50 kids, and these are spline curves for age less than one year versus age greater than equal to a year. And I was excused from all of these votes. Actually wasn't even in the room. I walked away. Where they kind of looked at the data, and what the guideline group actually decided to do was to change the recommendation looking at this data to 20 to 30 breaths per minute. So why did they do that? So when you look at the spline curves, I'm showing you ventilation rate versus probability of survival. I'm going to highlight the steep slopes of the curves. Nobody in the room, as I was told, was a little uncomfortable recommending 30 or 40 breaths per minute. How are you going to teach it? Is that one like every one and a half seconds? So when you kind of looked at the steep increase in probability of survival was in that 20 to 30, it's teachable one breath every two to three seconds. This is kind of where that recommendation came from for intubated children. Clearly there is a ton more work that needs to be done on the right way to ventilate kids doing CPR. But this was kind of the study that kind of drove that change of the guidelines in 2020. All right. So the last thing I want to talk about is not everyone has invasive monitoring. So a few potential future non-invasive approaches to titrate CPR. One of the things that I think is going to be useful is pulse ox. I think for those of us when we're doing CPR, particularly in hospital, and you see a pulse ox tracing, everybody feels good about that. Nobody knows why they feel good about it, but it's nice to have that pulse ox tracing. And so this was an animal study where they actually looked at arterial line waveforms and compared it to pulse ox tracing characteristics. So what I'm showing you there, arterial line on the top, pulse ox tracings on the bottom. This is from their paper, sinus rhythm on the far left, going to VF, you don't have a pulse, you don't have a pulse ox, nor do you have an art line. And then they looked at low quality versus high quality CPR. And you can see in the low quality CPR, it's poor BP, it's low amplitude. And then when you get a better blood pressure, you get better amplitude on your pulse ox. And so they looked across two different groups of CPR, low quality and high quality. Not surprisingly, end tidal, coronary perfusion pressure, a lot of the stuff that we talked about, these physiologic markers were different between the groups, better in the high quality CPR. But then they looked at simple things like pulse ox amplitude or error into the curve of the pulse ox. And these were just as different, in fact, more different and more statistically significantly different when you looked at low and high quality CPR. So we've been looking at this, this is actually an R01 we have with a collaboration through the Villanova Center for Analytics of Dynamic Systems. These are machine learning engineers. Dieter Bender there on the left is the one that's been driving a lot of this. But really looking at pulse oximeter as a marker of CPR quality. So of that physiologic waveform platform that I told you about, these are these waveforms. So we have simultaneously recorded blood pressure and simultaneously recorded raw pulse ox. And if I, after talking about this, if I ask the audience, which one of these do you think has, is better quality CPR, I think it's pretty obvious that the one with the nice bumps on the left is the high quality CPR. And so they've used machine learning to actually kind of extract out things in the pulse ox tracing to be able to infer whether or not that simultaneously recorded blood pressure would meet the targets of a diastolic of 25 in the infants or 30 in the older children. And it's a little hard to see, but right now the algorithm that they currently have with only a five select, five second delay of getting the data and giving it back to you with a light, they are 90% accurate in telling you whether or not your CPR quality was good. All right. So now for the disclaimer part of this talk. So what I'm about to do, I'm totally stealing AHA PALS guidelines colors, but this does not reflect their policy, SCCM, American Heart, CHOP, PEN, SCCM, VNI, whoever. It's not their opinions. These are my opinions. If you have any complaints, please direct them to my department chair and his email's right there. All right. So when I was preparing this talk, it was kind of cool to think about, you know, the idea of pediatric CPR quality was something that we were thinking about between 2005 and 2010. And now in 10 years later, we have people thinking about blood pressure during CPR and title guided CPR. We have observational trials that have shown what targets are associated with outcomes in CPR. And it makes you kind of think like where we're going to be in 10 years from now. So I created this slide a couple years ago, and our whole goal of our center is to start kind of clicking off the boxes. So many people in this room have been helping with this. And so I would hope that in, you know, in 2030, maybe this is what the PALS card looks like. So in the first minute, you're always going to be doing high quality CPR. That's not going to change. You're going to be shocking lethal rhythms. But if you have PEA, maybe we're going to be putting an echo probe on the chest to kind of see if there's anything that we could reverse. And maybe you're going to think about physiologic monitor placement instead of standing there with your hand on the pulse in an ICU. Maybe you're going to put a needle in the other hand. Prepare for vasopressors, fluid, nitric oxide. All of these things may be coming down the pike. We're going to be assessing hemodynamic goals in that first five minutes, and we're going to reassess every minute if you have the data available to you. If your CVP is low, maybe you're going to give a fluid bolus. If you have that kid in profound shock from, you know, volume loss, maybe they need a little bit of fluid to get them back from cardiac arrest. If you have a really high, maybe you need to do a POCUS. Maybe you have a pulmonary embolus. Maybe you have something that's increasing the pressure in the chest. From a diastolic pressure standpoint, I kind of talked you through this. I think a lot of places are already doing this. Alter your mechanics, your depth, your rate. Maybe you give drug a little more frequently if it's working. Maybe you give vasopressin rescue, or maybe you start an infusion. Like I said, we are actually doing stuff like this at CHOP already when kids are already on an infusion. And from a systolic standpoint, it's probably going to be similar where you could alter your mechanics, give drug, or maybe you're going to move your hands into the right spot. If the adults are in the wrong spot with hearts that are this big, what do we think we're pressing on when we're pressing on acorns? But I think the point of all of this is if this stuff isn't working, call for help. And so if you're in an institution that has the ability to actually rescue these kids with ECMO, by minute six we should be making these decisions. And at that point, you know, I get asked a lot of questions, should I continue to give epinephrine in that kind of situation? And the answer is probably not. If you have good blood pressures, why continue to add the afterload? But really get the people there that can rescue this kid with ECPR protocols. And so in conclusion, and I said I would talk for about 40 minutes, personalized physiologic directed CPR is really centuries old. This is stuff that's been going on for a long time. From a pediatric standpoint, I think the best evidence supports diastolic blood pressure. We have lots of options to get there. You could alter your mechanics. You could alter vasopressors, whether that's more, less, or alternative. Target underlying pathophysiology. One of these that we've used in our lab is actually nitric to target pulmonary hypertension. But really, and I use this slide a lot, is just to say have the courage to avoid insanity. If you're doing kind of the same thing over and over and you're not getting the response, think about switching things up. And because changes are, but it's why we do what we do. And so I just want to say thank you for listening to me. This is the obligatory picture of my three kids at the beach. But I really want to end saying especially thanks to the Lairdall family for their undying commitment to saving lives and advancing care for critically ill children around the world. Thank you. Are there any questions from the audience? Yes, please come to the microphone if you have one. Hi, wonderful talk. I'm from Children's Oakland. A quick question and then a question. In the study, which I think you said was adults where they did TEE during chest compressions and they were blocking the left ventricular alveolar tract, which way did they move to fix the problem most of the time? So when I talk with Felipe about this, what he actually said is it depends. And they really, sometimes it's down, sometimes it's over. But he said a lot of it depends on it's not the same in every case because the way the heart sits in all of us is a little different. Okay. I couldn't help when you showed that data thinking about, of course, you know, the conventions we have for neonatal and pediatric, you know, circumferential compression. And I also couldn't stop thinking about my good friend Dan Stromberg's work looking at interposed abdominal compressions in shunted patients. And so it made me wonder, is there some kind of generalized approach or model that we should be using in kids where we're deviating from the kind of the traditional position or should we just be focusing on deeper? Yeah. No, it's a great point. And one of the things, you know, I didn't show the data, you know, kind of building on his work. Actually, one of the things we just submitted to journal was actually abdominal only chest compressions. So when the surgeons don't want to actually like have you pressing on the chest of their newly repaired aortic arch, there's actually, there's a CT surgeon out there that actually he just tells them to just squeeze on the chest, stay off the liver, and they generate a pressure. Is it generating flow? I don't know, but we're actually submitting that right now. They have been able to actually generate pressures with abdominal only CPR that is actually the same as actually pushing on the chest. Don't know which way the blood's going, but when you look at his outcomes and, you know, what we've put in this paper, they're absolutely similar. And it's actually something you can do during cannulation. So our surgeons are like, we've tried it a couple of times to chop now, so we're actually doing this and it's like the surgeons are still putting the cannula in and you never actually stop doing it. Hi. I'm an institution where we mainly use the Entitle and the Nears. So I wanted to ask about the placing an arterial line in a non-ICU settings. Maybe it's not as pertinent for pediatric, but for adult, it becomes the question, you go to the floor, what do you get a supply? Can the nurses prepare the line and level it? And so I just want to give you can comment on your experience doing placing a line and using this information outside ICU. Yeah, I have to be honest with you. I don't think in 2023 we're there yet. Even at CHOP when we've, I mean, this is no offense to our emergency room providers, but even in the emergency room arrests, when we're like, we'd like to set up an arterial line, if the providers aren't used to setting it up, you're just going to distract somebody and then they're not doing the actual important thing that you need them to do. So I don't think we're there yet from a non-ICU setting, particular in pediatrics. Even in the ICU, it's not easy with kind of the chaos that's all around. But my point and my joke about that is it always seems like in pediatrics, we have someone like on the pulse and I'm not really sure what they're doing there on the pulse. And so they could, you know, just have something in their hand. So it's, I don't think it's ready for prime time outside the ICU because it's even just hard when you're in kind of the quaternary centers in the ICU. And Bob's laughing because he's the one that I always get into an argument about this. Thanks again for just an outstanding lecture. All of us appreciate it and congratulations on winning the Lerdahl Award. You really deserve it.

personalized physiologic-directed CPR

diastolic blood pressure

pediatric CPR

chest compression mechanics

end-tidal CO2 monitoring

technology

IC-Cres trial

CPR quality improvement

resuscitation efforts