Hi, and welcome to Conversations in Interventional Cardiology. My name is Andrew Goldsweig. I'm the Director of the Cardiac Catheterization Laboratory and Cardiovascular and Conclinical Research at Baystate Medical Center in Springfield, Massachusetts. I'm fortunate to be the Associate Editor of J-SCI, the Journal of the Society of Cardiovascular Angiography and Interventions, and I'm honored to represent J-SCI and our Editor-in-Chief, Dr. Alexander Lansky, this evening. Today we're here to discuss an important recent publication in J-SCI entitled, Acute Myocardial Infarction in Sky Stage E Shock Patients Treated with Impella and PCI, Insights from the RECOVER3 Study. We're very fortunate to be joined by an esteemed panel of really internationally renowned experts. The first author, Dr. Ivan Hansen, is the Medical Director of the Cardiac Cath Lab and the Program Director of the Interventional Cardiology Fellowship at Corwell Health, William Beaumont University Hospital in Royal Oak, Michigan, a place near and dear to my heart because a couple of years ago I was born there. Also Dr. Sreehari Naidu, who is Director of the Cardiac Cath Lab at Westchester Medical Center, SCI President-Elect and Senior Author of the SCI Shock Classification System. We've got Dr. Bavar Basir, who is the Director of Mechanical Circulatory Support at Henry Ford Hospital in Detroit. He's an author of the present paper and he is a founding member of the National Cardiogenic Shock Initiative, as well as the annual SCI Shock Meeting. And a man who really needs no introduction, Dr. William O'Neill, the Emeritus Medical Director of the Structural Heart Disease Program at Henry Ford, also a founding member of the National Cardiogenic Shock Initiative and a real pioneer in interventional cardiology and management of cardiogenic shock. So gentlemen, welcome and thank you so much for joining. I'm going to start by directing a question to you, Dr. Hansen. In this paper, your work really on patients with acute MI complicated by SCI Stage E shock sheds new light on which patients may benefit from mechanical circulatory support. So we'd be grateful if you could summarize your findings for our listeners and I see you've got some slides to share, which is great. Thank you. Thanks, Andy. Thank you for the introductions and on behalf of the co-authors, I'm delighted to be able to present our study and what we found. RECOVER-3 is an FDA post-approval study. The Impala device was approved by the FDA for treatment of cardiogenic shock complicating acute MI in 2016 and the post-approval study was conducted between 2016 and 2020. It enrolled 418 patients with cardiogenic shock complicating acute MI undergoing revascularization of their coronaries with stenting and Impala device. The Impala could have been placed before, during, or after revasc and looked at patients across 41 sites in the United States. The way we looked at this in RECOVER-3 is with the SCI shock stage system, which is becoming more and more familiar to all of us in the space, but basically a quick review. Stage A patients are patients that are at risk, so that would be somebody that has an acute MI but are hemodynamically stable, haven't displayed any evidence of hypotension or malperfusion. B for beginning shock is a patient that has an acute MI but now they're developing some early signs of shock. C for classic shock would be a patient that now is requiring pharmacologic support with pressors or even mechanical circulatory support, such as Impala. D for deteriorating, those patients are, despite all those measures, getting worse. Then E for extremists, these are the sickest of the sick and the subset that we looked at with this study. These patients are arresting, they have impending circulatory collapse. With the update in 2022, which was chaired by Dr. Naidoo, importantly, we have that there's a modifier here. We used to consider all cardiac arrests as a stage E. Now we've modified that to say that really it has to be a clinically significant arrest. These aren't just defibrillations that are quickly and easily treated with a defibrillation. These are patients that require CPR for some time and there's concern for anoxic brain injury. Granted, it's difficult to tell in real life if someone's going to have anoxic brain injury because a lot of these patients are sedated, they're on mechanical ventilators. As far as categorizing these patients, we were careful to take the modifier into consideration. What we did is we looked at patients at baseline and then within 24 hours, so two separate stagings. At baseline, for this study, they were all stage E. Then we reassessed that within 24 hours. They could have changed stages. That's what we assessed here. We looked at their baseline clinical angiographic and procedural characteristics and their survivals. Then we looked at a multivariate analysis to try to figure out what predicted responsiveness. What made a person respond after their PCI and Impella support to a better sky-shock stage versus staying in stage E? Essentially in Recover 3, this was our original publication last year in Journal of the American Heart. We saw a large proportion of patients in Recover 3 coming in at stage E. Again, sickest of the sick. What happened with these patients over the ensuing 24 hours is that about half of them stayed in stage E. The other half improved their sky-shock stage to either C or D. As would be expected, the patients that did not respond, they had their Impella device, PCI. Despite that, they remained stage E. The survival was very poor, as you can see there. If they responded in their sky-shock stage, then their survival doubled, going from 31% to 62% significant change. The next thing we looked at, of course, is what were the predictors of improvement here? How can you tell who's going to get better and who's not? I think it's important to look at why our study is important and how it fits nicely into the existing literature. I borrowed this slide from Hari Naidoo's same paper from 2022 with the update on the sky classification. If you look across the board there, at least the time this was published, all of the existing studies at that time had a relatively low proportion of stage E patients that were represented. Then you look at our study, and now we've got at least some literature now on a very sick stage E population. I think it's important to point out that stage E patients just aren't very well represented in other studies across the board, whether you look at acute MI patients or sick ICU patients with advanced cardiomyopathies or cardiac arrest patients. Here's what we found. For baseline laboratory values, really the only thing that stood out was the white blood cell count was lower in responders versus non-responders. The pH was higher, and heart rate for hemodynamics, the only thing that was different was heart rate barely reaching statistical significance, but it was lower in the responders. As far as admission characteristics, the things that were different here in the responders versus non-responders is that the duration of shock pre-impello is actually longer with non-responders, which was a little bit counterintuitive to us, but that was what we found. When we looked at procedural characteristics, the main things here to note is that inotropes and pressors were fewer in the responders than the non-responders. Whether that was patients that were treated before impello support or on support, those patients that improved had fewer pressors on board. Another thing that we found is that implanting the impella prior to the PCI actually portended a more favorable response, which is also seen in prior studies. And then impella P-level, so the higher the P-level, in other words, the faster the impeller is moving, the more support patients being given, that correlated with responders. Not exactly sure why the non-responders had a lower P-level. Maybe it was because they had things like hemorrhage or their ventricles were smaller. Maybe they had malpositioning of the device requiring operators to turn down the flow rate and maybe that's what translated into poor responsiveness. We didn't have that granular of data to be able to answer that question. And then as far as angiographic characteristics, we found a higher proportion of patients who responded had multivessel disease treated in the lab. Also the LAD was more often treated in responders and the baseline TIMI flow grade was better in responders. So we took all covariates that had a P-level of less than 0.15 into the multivariate analysis and this is what we found. So the things that corresponded to responsiveness were impeller pre-PCI versus after the PCI. Putting the impeller in first resulted in improved shock stage. Also more lesions treated resulted in improved shock stage. And then patients that had more pressors did worse. So that's essentially the results of the multivariate analysis. In conclusion, what we found is half of patients in this real-world registry that presented with stage E actually improved to stage D or E within 24 hours after the impeller and PCI. So that's what we call responders. They had actually much better survival. The survival doubled if they did improve their sky shock stage. The predictors of responsiveness, as we showed, were more likely related to the treatment strategy for their shock, not the baseline clinical characteristics. So it didn't matter that the blood pressure, the cardiac output, things like that really didn't predict who would respond. It was more related to pressor use, the number of vessels treated, and putting the impeller in prior to the PCI. Those were the things that really mattered. We think shock stage really should be used as a clinical tool. If you think about it, these really, really sick patients, you have a lot of practical considerations such as, do you even do anything? Or do you just approach them with a palliative mode of care? Probably just one assessment of sky shock stage at baseline isn't enough because half these patients actually improve to a shock stage of C or D, which correlates to a much higher survival, which is what we showed in the prior study with Recover3. And then future trial design, because as we know, shock is a spectrum. It's not a black and white disorder that you either have or you don't. So the sky shock stage showed that there's basically a direct correlation between stages and outcomes. So we can compare apples to apples if we look at patients within an individual shock stage. Thank you. Wow. Fantastic. Thank you. Thank you very much for sharing that, for sharing the slides and for sharing your findings. Clearly the study provides a great deal of information to allow more careful prognostication as well as some important insights, I think, in terms of clinical practice. Let me direct a question to Dr. Basir here. So, Babar, you've been designing shock algorithms for years now. So the question for you is, how should this knowledge change our practices in terms of how we manage cardiogenic shock? How should we incorporate this information into our practices in terms of patient selection, in terms of the care we provide, in terms of the monitoring that we perform? Yeah, no, thanks, Andy. Firstly, let me just thank Ivan. He did a tremendous amount of work to be able to lead this study, and I think the results are really meaningful. I think the characteristics that we're seeing out of the ReCOVER-3 study are what's happening in real life, which is that we're using Impella support devices in the later stages of shock, when it's very apparent that the patient's in shock, when they're on multiple vasopressors and inotropes, and when they're very sick. And that becomes a really challenging place for us to be able to have optimal and good outcomes. One of the things that was mentioned was, is it futile to actually use this support device in this case? And I would just say that it's probably not the case, because here we have a 30% survival, even in patients who had stage E shock in 24 hours. So I think it's really useful information in terms of building algorithms, but I would make us all cautious about not treating stage E patients. Fair enough. So I guess those are a couple of important insights then. Let me drill down a little bit on some specifics. Three things that came to my mind, and hopefully the whole panel can weigh in on this. The use of pressors, is it a marker that we're too late to put the Impella in? Impella pre-PCI being a winning strategy, should that be our standard? And what about indications for non-responders, indications for escalation of mechanical circulatory support? So pressors. Can I comment on a little bit? Because we've looked at multiple databases, the NCSI and R3 both showed that increasing doses of vasopressors increase mortality. And there's one thing that's not very well, two things that are not well appreciated about alpha agonists. First is that they cause profound pulmonary vasoconstriction, so the TBR goes up. And if the patients have any evidence of RV dysfunction, or the RV collapses, and you can't get blood flow from the right to the left, and therefore the left becomes oligemic. And so the left support device doesn't work. And secondly, they cause very profound splenic vasoconstriction. And so that's gonna markedly increase lactic acidosis, and liver dysfunction, intestinal dysfunction, which if the patient survives initial insult, is gonna kill them down the road with SIRS. So just one simple thing, there's increasing evidence suggesting that the more you use, the worse the patients become, and therefore mechanical support would be a better strategy than pharmacologic support to get the blood pressure elevated. Yeah, I was gonna weigh in that I think, everybody's known this for some time, we've been teaching it for a while. I think the challenge and what we have to get out there is that we're never gonna know if people are on multiple pressors, because they just let them stay at that level and did not titrate them down. The deescalation protocols need to be as aggressive as the escalation protocols. We're all very good at escalating. We're not very good at deescalating. So I think one of the good things about the SCAI definition is that it creates a very fluid and dynamic process because we outlined very clearly in the last document that you can go up and down the stages. And that was something that we all recognized, but did not look at actively in terms of, how can we drill these down? So to drill these down, meaning pushing them to a lower shock stage, one of the reasons would be to try to withdraw some of these pressors as much as possible. Bobber, you've done that in your algorithm. So I think as word gets out more, we'd like to see if people can withdraw the support, maintain the lactate below two, how do we find that balance and really have that be a goal of therapy? So this study is great, but we have to know whether they were just staying on the pressors because of kind of their protocols or the status quo, or were the patients who were coming off actually better or the patients who were on multiple actually worse? That's a challenge. What about Impella pre-PCI? Should that be the paradigm? Yeah, I think what Impella pre-PCI really represents, Andy, is using it as the treatment strategy in cardiogenic shock. And luckily for us, even though these patients weren't treated in the post-danger era, these were patients mostly treated before danger shock, we live in the post-danger era where we know that Impella is associated with an improvement in survival in patients with AMI cardiogenic shock. And so it is really a tool that we need to use to be able to start to reverse the cardiogenic shock, which along with revascularization is gonna improve outcomes. And so there's no reason to wait because it is really part of the treatment strategy. And I think the data is always gonna be biased, right? The later cases are always those cases who you thought you would try to get away without using them and then you ended up using a support device. And so the data is always biased and you're not gonna be able to get a good answer that way. But there's so many observational studies that have shown the same thing. And physiologically, mechanistically, it makes a lot of sense. I would agree. And Dr. O'Neill and I have been around long enough to know that certainly as I trained, you have to make it a clinical assessment of the patient needs support or not. And back in the day, we would put balloon pumps in, stabilize the patient at least, and then move on to the PCI where you can get a more durable result. Now we have better tools and we have tools that also maybe unload the heart as well as support the heart, support the end organs and allow you to begin the process of healing while you do PCI, which might also improve microvascular dysfunction and whatnot. So I do think that clinically people, this makes sense that you try to support the patient and then do the PCI for shock. Yeah, we've been collecting data on shock since 2007 for use of Impella. So it's been kind of almost a 20 year journey. And when people first started using the device, large, clunky, difficult to use, big access. And so they hesitated to use it. So the patients would come in, they would get a PCI, maybe a balloon pump, pressers. And then if they failed all of that, so at the last resort, you would put in an Impella. And what we found relatively early is if you actually move that strategy where you move the Impella much further upfront, that you would get a better outcome. And we did that with NCSI and got people treated much quicker. And I think actually had a fewer proportion of patients that ended up in class E. The R3 registry actually were more traditional referral places where the patients were sick elsewhere and then were transported. So they got in later in the natural history. So I think the quicker we get the Impella in, the better, the higher likelihood you're gonna have an improvement in survival. What about the number of lesions? The question about the number of lesions doing more lesions than shock, which is kind of against the grain here in terms of what we advise. But is this a marker of a more stable patient or is this a marker that actually more lesions improve the shock state? What do you guys think? Well, I was gonna bring this up, Ari. I think it's really important because it's counterintuitive after culprit shock. And we really have to differentiate the two studies that have been done. I mean, culprit shock looked at patients where revascularization was done all in one procedure. And one of the things Alejandro Lemoore showed in the Cardiogenic Shock Initiative was that patients did better when they were revascularized more, particularly stage E patients. Now, Ivan has shown in this study that similarly, the more lesions that are treated in stage E patients, the better these patients do. Now, we can only assume that these are proximal mid-blood vessels, relatively large blood vessels that are being intervened upon, but it seems to at least generate a hypothesis that we need to relook at how much revascularization is appropriate revascularization, particularly with more supportive devices like Impella. So let me ask then, let me direct this one to you, Dr. Naidoo. The Sky Shock Classification System was really revolutionary when you and colleagues published the original paper in 2019. It's at this point become very much the standard for research and is becoming more important clinically. In 2022, you published an updated document. Ivan showed some figures from that that incorporated multiple validation studies. Now, in light of the present findings where we're looking at people who are all class E, but some are responders, some are non-responders, how should this change the way we think about and classify shock? How could these findings potentially influence future versions of the Sky Shock Classification System? Well, one thing I'll say that I think you guys made the point already, which is that in that second document, we made the implication that stage E, especially in the setting of cardiac arrest, may be futile. And I think that point that Barbara made, which is that even in the non-responders, you have one in three patients surviving, that's actually quite good when historically the survival rates are 40 or 50%, that's a 30%, and this is quite good for stage E shock, even as a non-responder. And the responders are 60%, which is quite good. So I think this is exactly what the Sky Classification was intended for, not just to be a static document or a static grade, but to see how these patients do and to be used proactively to push down the grade during the hospitalization and commensurate with that to have a better prognosis. And so this data is actually showing that in the real world that if you can get them to a lower stage, especially in the first 24 hours, which I think many of us are calling sort of that golden day of shock, that is the time to do it because it really has a prognosis effect on survival. In terms of future documents, when there is one that we are, the third iteration is just launching, we're in the process of putting together that writing group, and there'll be several things we're looking for. And one of the things is to look at how do we make the classification easier in terms of the types of input that you need to get to that definition? Because what we'd like to do is make sure that all of these registries have that data going forward. So that means that as new registries come along, they should look to see what do they need to more accurately represent the sky shock stages so we can be much more granular and also not just treat the patient in front of us with the data that we have, but make sure that we have the data that we need to get the actual definition that predicts the prognosis. Yeah, Andy, if I could just jump in for a second. The other really important finding, not in this paper, but in things that we're looking at down the road is that the patients that are in class E, even if they survive 30 days, that's great, but then there's a huge drop off in survival at a year. And so I think it's really gonna be a call to action that if a patient is getting ready to go home that came in in class E, they have to have a good handoff to the heart failure folks and possibly electrophysiology because looking at arrhythmias, early implant of an AICD and kind of a heart failure focus to improve survival in those patients, but as a real marker, not only of early mortality, but if they survive, then a one year mortality. That's a great point. Well, there's a lots of very useful clinical information that comes out of this. And obviously there are a lot of very important questions that are raised. We're running low on time. So I wanna thank everyone for really an awesome conversation. Thank you, Dr. Hansen for your presentation and Drs. Naidoo Basir and O'Neill. We are thrilled to publish this important paper in J-SKY and to share this discussion with this incredible panel. Please follow J-SKY, submit your own work to J-SKY. J-SKY is the official journal of the Society for Cardiovascular Angiography and Interventions. You can find us online at jscai.org, J-S-C-A-I.org and on exit at my J-SKY. Thank you very much.

Interventional Cardiology

RECOVER3 study

Impella

cardiogenic shock

revascularization

Sky Shock Classification