of critical care medicine that I much appreciate. So we're going to talk about the SkyShock pyramid and some validation. In terms of disclosures, I've worked with a variety of the device companies. Obviously, this is all a CME lecture. This has very little to do with anything like that. And in terms of an outline, we'll talk a little bit about what existed. What was the evidence gap or the hole left that led to the decision to categorize shock? We'll talk about the SkyShock classification just briefly, though, and my colleague and friend, Dr. Jensen, will follow up on some of the newer iterations in his extensive lecture, right to follow mine. And we'll talk about validation. So does this actually work? And some conclusions. So the shock trial is, in fact, since 1999, the only trial in cardiogenic shock that actually has ever shown a benefit. And that's really remarkable. If you consider now 23 years later, we don't have any therapies that work. Maybe I should be invited to an SCCM meeting, because it's like septic shock, lots of treatments and not necessarily anything dramatically better. So in cardiologists, we tend to be an inpatient bunch. So for us, 23 years is several light years. And what was the definition of cardiogenic shock was, to everything, to really understand it, you need a proper definition, and this is not particularly rocket science. This was a trial of AMI shock, and the question was, do you emergently revascularize the coronaries versus initial medical stabilization at an era when the only destabilization besides inotropes were an intratic balloon pump? And the definition of shock was that of end-organ hyperperfusion due to cardiac failure with cool extremities or less urine output and so forth, clinical variables, and a blood pressure less than 90 for 30 minutes, an index of less than 2.2 if they were on support such as a balloon pump and they were adequately hydrated. And the problem is that really describes an entire range of patients between mildly ill and likely to get better no matter what you do, just sort of stay out of the way, versus extraordinarily critically ill. And we felt that actually it was really important to get the language right. And I like this little cartoon. So if you're standing on the left of the bars, the perception oftentimes is incorrect and it leads to misunderstanding and conflict. And you can clearly see from the perspective of the gentleman on the left, there are four bars without a doubt. And if you look on the right, actually there are three. And we were held back in 1918, we're thinking about how is it that we don't have a common language to describe a disease that so many of us were involved in taking care of. And it's important to be able to take care of the patient, it's important to be able to provide the proper level of care, and to avoid futile transfers. And actually the idea of the SkyShot classification came from a meeting. And right before that meeting, I had a patient with a referring physician call, described somebody sounded desperately ill, sounded like they really needed ECMO right away. And so we accepted the transfer, that was back when I was in Virginia. And I actually had my perfusionist, my cardiac surgeon, the intensivist, were all waiting for this person to come and assume that we'd be lucky to get them off the stretcher before crashing on ECMO. And in fact, the patient came in and were like, you must have sent us the wrong patient. The patient looked like he was probably good enough to go to telemetry or step down. And to the sending center, this was the sickest patient they'd ever seen in their lives. And to us, well this was maybe, you know, very, very straightforward overnight night float patient. And we felt that really it's important to develop a language. And let me just explain then in the setting of three scenarios are these shocks. The first one is a 48-year-old man with a chronic dilated cardiomyopathy who tells you that he stopped taking his Lasix because doctor, it was set for 30 days and it's been 35 days. And since that time, he hasn't breathed so well and notices actually that now it's okay as long as he's up and close with his lazy boy. His blood pressure is not so good, 80 over 50, he's heart attack or cardiac. Maybe he's got some edema but he's not particularly cold and he has a bag of Doritos and kindly volunteers you look like you're hungry, would you like some? Is that shock? Yes, blood pressure is low. Or is this shock, the 62-year-old man in the ER with chest pain and obvious anterostomy? Same blood pressure, maybe a little bit more tachycardic, but he has cold extremities. He's profusely diuretic, diaphoretic. He has no murmurs and basal crackles. He's going to the lab for a PCI. Well, you don't have a swan. You don't really have much. Is this patient in shock? Or is this patient, the 78-year-old man brought in by the EMS after collapsing at the Home Depot? He had V-fib, got a quick shock, was intubated on site, intermittent runs of VT. And when the EMS staff come to give you the patient, they say, yeah, okay, he's stable. He's stable and you notice that he's got wide open fluids and dopamine at 30, probably would have had another but that's all they had available. He has a similar blood pressure and even more tachycardic, a bunch of crackles. Everything seems cold. EKG is not very helpful to you. Are they all the same kind of shock? And so because of those discrepancies, I had the privilege of co-chairing the committee that several of you in the room were involved with where we developed an expert consensus on the classification of cardiogenic shock. And we developed this pyramid starting with A and the idea was actually based on the ACCHA classification of heart failure between stage A and stage D. And as we deliberated over the year, we decided to treat shock in five stages, the familiar to many, A, being patients at risk, B, with beginning cardiogenic shock defined by hypotension or tachycardia but adequate perfusion, classic cardiogenic shock like was shown in the previous slides, patients who have hyperperfusion by whatever way you define it, and we felt that the additional two categories were really important, that somebody who was deteriorating on whatever first medical management was employed, whether that was ambulance management, whether it was management ER or so forth, was on a different trajectory and therefore they should be classified differently as stage D for deteriorating. And then a category at the pinnacle of the triangle of extremis of the patient you can all recognize, you're sitting and writing your notes, you see people running in and out of a room, lots of yelling going on, the patient where you're literally actively trying to keep them alive. Clearly all different trajectories, at least we thought, and the question is does that actually work? So if you go back to the cases then, you're a patient who comes in with decompensated heart failure but is walking wounded as a Sky B patient because he's tachycardic, he's on the way to trouble, but he's not hyperperfused, at least not you can tell. The patient in two in the ER is obvious classic cardiogenic shock, and indicating that you don't need a SWAN or even POCUS or a HOCUS or whatever else you want. You can use your eyes and ears to see that that patient is in cardiogenic shock based on the history just on the slide and on your eyes. And number three, of course, this patient is deteriorating based on initial EMS therapies with clearly going in the wrong direction, and he has the A modifier because he's had a cardiac arrest. And reflecting that all these three patients have a different trajectory to go down. Does the Sky Stage predict the outcome? So privileged to present work that I co-authored with the first author, Dr. Jenser, who's going to be speaking next. Dr. Jenser and I met at a meeting right before the SkyShock classification was presented in Houston. He came up to me and said, you know, this is sort of interesting, and I've got a large data set. And so these kudos are all to him for having this beautiful data set we published in JAK. It was the first classification of validation of the SkyShock, and he had a database at Mayo Clinic Rochester of more than 12,000 patients after we excluded things like readmissions of those who didn't have requisite research consent. There was about 10,000 patients. And you can see that, just like a pyramid is shaped, the majority were Sky A, but followed by a very important minority of Sky B, C, D, and E. And we actually set up definitions before analyzing the data that were based on the Sky criteria, and they're here. And this is from the central illustration. Those patients were either not hypotensive or tachycardic, but not hyperperfused were A. B were the tachycardic or hypotensive without hyperperfusion. In the paper, it defines the very specific criteria for hyperperfusion, so C being classic, D being patients who've deteriorated, E being extremists. And on the right, you can see this first illustration in a very large data set that actually this simple even EMR-capable definition of cardiogenic shock really predicted not only ICU, but also hospital mortality for the first time. And then he followed up by an analysis of post-discharge, and it was a privilege to work with Dr. Jensen on this paper, where you could see that this actually didn't just predict what happened in the ICU, in the hospital, but also post-discharge, really reinforcing the idea that if you clinically identify what trajectory the patient's on, you know, if the car is headed off the cliff, it doesn't really matter whether the cliff is a 100-foot drop or a 500-foot drop, there's going to be a longstanding problem, and you're going to even have differences in one-year death and readmission. So very helpful to use this. So as we were defining this classification, many in the group were wondering, maybe this is too simple. Maybe we need an app. Maybe we need a score. Where's the square root sign? This is just so darn simple. How could it work? And it's been gratifying to see that, in fact, it does work. In Europe, my colleague, Dr. Benedict Schrag, looked at his very large cardiogenic shock registry, more than 1,000 patients with a large MI, heart failure, and or ACS, so big, well-defined, single-center group, and he was able to show that the sky stage there and that large cardiogenic shock population also predicted mortality. And so that actually, if you have this sky class at baseline, you can actually show what the survival will be. So particularly helpful, and these are all retrospective validations. At the same time, my group at Sentara in Virginia, we were using the sky shock classification as soon as we had settled on the derivation and so forth. And we had 166 patients in our shock team, and we looked retrospectively. Importantly, we also carried patients who didn't get a device. So if the shock team were called, we'd watch, and of course, nobody calls for a sky A patient, but there were sky B through E, when we borrowed a similar color scheme. And in this paper, we were able to show, interestingly, because it was single-center and prospective day zero shock team evaluation of sky stage, before we knew what the end of the story would be, we were able to make some unique insights. And you can see that when you look in the table, cardiac output and hemodynamics do vary with sky stage, but not as dramatically as you might think. So in fact, it's helpful to have a SWAN, but your initial assessment of sky stage is probably as valuable. Some people love SWANs, some don't like them. I personally fall in the camp that feels they're incredibly useful. Point being, clinical evaluation is also extremely useful, and if you look even at lactate, except for sky D, many patients actually, lactate does not necessarily predict what sky stage you're in. So plug for being at the bedside and talking to patients. And the signature figure, I believe, from this paper is this one. Importantly, we also assess sky stage at the 24-hour mark, the day two when the shock team evaluated, or day one, when the 24 hours after initial shock team assessment, and we're able to show, then over time, actually that was extraordinarily predictive. On this figure, you see that patients who either had no change in sky stage, in green, had very similarly and horrible mortality as compared to those whose sky stage worsened. And while this may be logical, it was interesting to see that if you improve sky stage by even one, so from D to C or C to B, dramatic difference in the mortality, the P value for this graph is highly, highly significant. And if, in fact, you improve two stages or more. And the relevance to this is that if you're in the outside hospital, we tell our partners all the time, if the patient's not improving sky stage at 24 hours, either you understand there's nothing that can be done here or elsewhere, or it's time to consider moving the patient. You know, like Einstein's definition of insanity, doing the same thing over and over and expecting different results is truly insane. So if the patient's not getting better, and you think there's a possibility, that's the time or sooner with reevaluation to send the patient, because the outcome is very, very clear. And there are actually numerous other validation studies. And the group from Sky reconvened, and I was privileged to author this along with Dr. Gensler and Dr. Nadeau, first author, and Jack. It was a reevaluation after many, many different consensus studies had been, and many different validation studies had been done. And I'm going to not go into great detail, because Dr. Gensler has much more detail about this, and a lot of the work is actually his, so I won't repeat that. But importantly, the figure was revised to show the sky pyramid that you may be familiar with, gradations of color. And I think this is probably the most important first aspect, the idea being that even among the stages, obviously, they're not all the same, and whether you call them early Cs or late Cs or early D or late D, the point being that shock is dynamic. And for those in a critical care audience, that part is obvious. But in cardiology, oftentimes, we're very focused on doing something. OK, I stented the left anterior descending. My work is done. No, it's not done. Yes, the LAD is beautiful, but your patient is now teetering. And reevaluation and the continuous reassessment is really important, and more to come with Dr. Gensler's lecture. There are other ways to assign sky stage as well. So this is by colleagues of mine in the Cardiogenic Shock Working Group, where they felt that it was important to give, at least for retrospective data sets, a way of assigning stage. And so this little figure shows if you have no drugs and no devices in their studies, they call those patients Sky B, because they did not prospectively assess sky stage. And if you have either one device or one drug, you're a C. But if you have both drugs and devices, D and so forth. And this is a very sort of helpful way to look at data sets. There's also a very recent paper a couple of months ago by their group, where they said, all right, that first definition was fairly simple. Then they went ahead and added this one. And this has made a fair amount of controversy, actually, because they have patients with a very complex figure, where you define it based on blood pressure, perfusion status, amount of treatments. And so I think this is interesting. And they have a large data set that they look at this, and it definitely would refer you to the full paper. I still go back to my nature of being a simple-minded person. Simpler is better. This is beautiful. But I'll ask you if any of you is going to take it and put it on a pocket card tonight in the ICU or next week when you're back on service. And then a very recent paper that came out from Ricci et al., looking at sky stage reclassification predicting outcomes. And this is actually just very nice, showing that patients progress. Even the patients that come in relatively well in Sky B, you could see that 41% of those patients actually deteriorated over time. Deterioration a little bit less common in Sky C, more common in D. And the point being that reassessment and by stratifying by a common language is very important, because you can recognize things, and machine learning all is nice, but ultimately, we're all the clinicians at the bedside, and we have ourselves first. What are future directions? So I think one of the key future directions is going to be robust registries, and this is a registry that I'm privileged to lead along with my colleagues. It's called Vanquish, so it's a multicenter collaborative to enhance biologic understanding quality and outcomes in cardiogenic shock. And this is a four-center registry led by our group in Cleveland Clinic, Western Florida, along with Inova Fairfax, with Dr. Charani, and Dr. Batchelor, and University of Utah, and University of Toronto, with Dr. Billia. And the idea here is that a small group, where they leverage the power of our shock teams, and actually follow longitudinally with prospective assessment of sky stage, where one doesn't need a figure. We just have the team consensus, what is sky stage? And then, in addition, layer in biomarkers, prospectively calculated, because it's really important to be able to follow serially. As well, we're going to follow patients for a year, including discharge assessments at 30 days, six months, and one year. And the team in Toronto's been the first, actually, to get this up and running, has enrolled, as of last week, 100 patients. We anticipate, between the four centers, 400 to 800 or 900 patients among the four centers in a year. And we think that this registry will allow us to really answer questions that have really plagued cardiogenic shock research, because the inclusion and exclusion criteria are challenging. What about other efforts? The American Heart Association has a cardiogenic shock registry, which is kicking off, which will be nationwide. However, we're involved in that, as well, but the case report form will be relatively not deep, but very, very broad, over many, many centers. So I think these are all complementary. And actually, my partners and I are in the midst of investigation of the Society for Thoracic Surgery Registry, more to come soon, about post-cardiotomy shock, which looks like it may be amenable to similar classifications. And so, in general, I'll leave you with this figure, that cardiogenic shock sky classification is this simple. Patient at risk is like your matchbook. Nothing may happen, but it has potential to cause big problems. Once you have a lit match, you're in B. If you have a wastebasket fire in your office or in your house, that's classic cardiogenic shock, still quite manageable. You could still get away. But I would posit to you that once the curtains have caught on fire, you're on a different trajectory in a worse circumstance. And by the time you have E, you have lots of those patients where we struggle and spend much effort trying to rebuild the house that's already burned down. And in conclusion, the cardiogenic shock is like a fire. And for those of you in the ICU, when you see it, it's really, really important. Dr. Jensen and others will describe how, in fact, sky classification may have value even in non-pure cardiac failure. And using the sky shock simple classification to communicate, prognosticate, and motivate. Please don't wait until your patient's house is burned down. I thank you.

SkyShock classification system

critical care medicine

cardiogenic shock

common language

mortality rates