Good morning, and thanks to Carlos and the organizers, thrilled to be here today. So pro, pulmonary artery catheters are a must in cardiogenic shock. As Carlos said, I'm an advanced heart failure and transplant cardiologist, and I run our cardiac intensive care unit, so you already know my bias. I have nothing to disclose. My objectives today are to show you that invasive hemodynamics are required to identify the phenotype of cardiogenic shock, and that we use those invasive hemodynamics to triage severity and mortality risk in these patients. And that finally, we need invasive hemodynamics to tailor targeted therapeutics for the right ventricle separately from the left ventricle. So we will have some audience response. So if you'd like to get your QR codes ready. So we begin this by setting the stage. Phenotypes of cardiogenic shock. Which phenotype of cardiogenic shock is most prevalent in the United States in the current era? acute MI shock, post-cardiotomy shock, obstructive shock, heart failure shock. Sorry, it should be coming up for you. All right, we've got about 120 votes in, and let's see what the, okay, heart failure shock. Fantastic. And in fact, that is the most common type of cardiogenic shock we see in the current era. 40 years ago, it was acute MI shock, hence the moniker of coronary care unit. And of course, since then, we've evolved to the cardiac intensive care unit, where we actually manage all four phenotypes, but the predominance is now in patients with acute decompensated heart failure. In this phenotype, but in all phenotypes, invasive hemodynamic profiling is a must. We need to know how the right ventricle functions separately from the left ventricle, because the mechanism of the right ventricle and what it's built for is completely different than that of the left. It also helps guide us on mechanical circulatory support decisions. Importantly, cardiogenic shock in the last five to 10 years has adopted a model of a hub and spoke, similar to STEMIs or strokes, where referring centers call us early to communicate data about the patient so that we can triage them in quickly. And invasive hemodynamics help us make decisions on which patients need to come first and who needs to go to the operating room for mechanical support. But it's not just that. In other forms of cardiogenic shock, invasive hemodynamics can help guide other interventional therapeutics. For instance, in acute MI shock after a STEMI, if a patient's hemodynamics fail to improve, perhaps our interventional team will go after more aggressively a secondary coronary lesion. Or in a pulmonary embolism, an obstructive shock that doesn't seem very central, that perhaps our interventional team doesn't look to go to initially, hemodynamics can perhaps help guide their decisions to move towards catheter-directed therapies, even for a more distal lesion. In 2022, the Society of Coronary Angiography and Interventions updated our cardiogenic shock staging guidelines. And you see that on the right-hand side of the screen. The base of the pyramid are your most stable patients, and stage E at the top are your sickest patients. These are patients requiring ECLS. Even in the updated sky shock guidelines, they note that sky staging alone is not enough to help predict mortality risk for our shock patients. And as you can see circled on the top, they recommend having hemodynamics in order to prognosticate how the patient is going to do, which can be very important when we speak to our referring centers and also in shared decision-making with patients and their families when we're deciding on more advanced mechanical support. We know pulmonary artery catheters accurately phenotype and prognosticate cardiogenic shock. In this paper published in Cirque Heart Failure in 2020, we see in green acute MI shock and in blue heart failure shock. And what you see, interestingly, is more RV shock being identified than even by V shock in patients who have a heart failure phenotype. And using invasive hemodynamics, we can guide therapeutics for the RV more quickly in this phenotype of patients. On the bottom of the screen in the red bar charts, you see in the middle there heart failure shock and in the far right acute MI shock. And what you notice is based on sky stage classifications, so patients are getting sicker as you move towards the right, there's a variation in how much RV failure you see. So even though there's more RV failure and heart failure shock, we can't discount it in acute MI patients who are in stage D or stage E. And invasive hemodynamics can help us identify this early. So given that pulmonary artery catheters guide their targeted therapeutics, let's look at a case study. This is a 41-year-old man. He has a non ischemic cardiomyopathy from a titan gene mutation, and he is in your CICU on an Impelify 5 and cardiogenic shock awaiting heart transplant. Overnight he has worsening hemodynamics. His CVP is now over 20. Normal CVP is about 2. His mean PA pressure is 20. His wedge pressure, though, is 15, and his cardiac index is less than 2 liters per minute despite the fact that he is on a 5-5. You come in in the morning and he's on escalating IV inotropes, and his labs demonstrate worsening renal failure. On exam, he's hypotensive. He is tachycardic, but he's not that hypoxic. So what is the next best step for this patient? Is it A, add a third vasopressor? He is hypotensive after all. B, percutaneous RV support? C, bedside VA ECMO? Or D, do we need our palliative care consultants? Okay, so we see a majority of folks said percutaneous RV support, but at least a third thought about VA ECMO, and that's a really important point, and in fact, makes my point that pulmonary artery catheters are required to help guide this decision. So we use the data from PACs to optimize ventricular arterial coupling. So when you ask a cardiologist which pressor is the right pressor, we're not going to tell you levophed and then vaso, because we don't have the sepsis literature to back us up. We instead want to know what the individual hemodynamic profile of a patient is. Here I have given you a table that I've written about what optimal ventricular arterial coupling looks like for each ventricle, and the goal of management in cardiogenic shock is to get patients back to this level of VA coupling, meaning making sure RV pressures are low, making sure that the pulmonary vascular resistance is in fact about a fourth to a fifth of systemic vascular resistance. In a normal RV, stroke work is a quarter to a fifth of what the LV does. So in a normal patient with a CVP of 2 and a wedge of 12, a ratio of about .2, our patient has a CVP of 20 and a wedge of 15, a ratio of about 1.5. What this tells you is perhaps this RV has done too much work and may be on the other side of being able to be recovered, and we need to think about targeted therapeutics to help the RV get back on the Frank-Starling curve, not just the LV. And sometimes we use pressures as inotropes in cardiology, oftentimes epinephrine, for example, and our threshold for mechanical support is very low. So we use that data from pulmonary artery catheters to guide our MCS decisions and configurations. So in our patient with an Impella 5-5, that is an endovascularly inserted device requiring an axillary artery cut down and then requires fluoroscopy in order to place the device into the LV, it can flow five to six liters a minute. But in a patient who already is on that support and now is demonstrating RV failure, you might choose to go for percutaneous right ventricular support, such as a Protec Duo or an Impella RP-Flex. Both of these are dual-lumen catheters that are inserted via the right internal jugular vein with an inflow cannula pulling blood out of the RA and an outflow depositing blood back into the main PA. This may be more than adequate for this patient, given that he's oxygenating okay. However, depending on the expertise of your interventionalist and your surgeons, you may move towards other forms of ECMO support that you see on the right-hand side of the screen. Whether it be central or peripheral VA ECMO, you may even go for alternative ECMO configurations. If the main issue is just unloading the RV and you don't really need that much flow, you may actually elect to do axillary ECMO, upper extremities. And therefore, you don't need to flow the patient as much, but you're still appropriately unloading the ventricles. And so using pulmonary artery catheter data can really help you refine these decisions instead of simply putting patients on femoral VA ECMO. We know pulmonary artery catheters improve mortality. In 2020, the team at Columbia published on how patients who had pulmonary artery catheters and cardiogenic shock actually had an association with less mortality. The patients in black in the pie chart were managed with a pulmonary artery catheter. On the bottom with the bar graphs, you see that at every sky stage of shock, their mortality was less. Just this year, in 2023, in jack heart failure, similar data were re-demonstrated. Though there is enormous variability in the use of Swann-Gans catheters and cardiogenic shock in the United States, in centers where pulmonary artery catheters were used, mortality was 28% in cardiogenic shock, as you can see on the bottom right of the screen, as compared to 35%. That's an adjusted odds ratio of 0.79. The bottom line is pulmonary artery catheters determine care pathways in cardiogenic shock. We need it up front as a hub hospital to understand from our referring centers how sick these patients are, who needs to come in first, do we need to have the operating room ready for them, and what therapeutics do we need on hand. It may also help us determine which of these sick patients are actually going to move towards the pathway of recovery and change the kinds of oral medications we put them on and how quickly we put them on. On the other hand, it may also tell us which patients will require durable LVAD or transplant and who can be candidates, in addition to what mechanical support they need. So take-home points, pulmonary artery catheters determine the phenotype of cardiogenic shock. They help improve early triage to hub centers. They guide our decisions on vasoactive agents and mechanical support and help us prognosticate long-term outcomes. The bottom line is they improve mortality when you know how to use them. Thank you. Thank you so much, Lakshmi. That was absolutely amazing. Looking forward for the home part by our colleague, Matt. So for those who are standing in the back, I invite you to please come. There's plenty of seats in the front, so we can potentially allow more folks to come in. They're not allowing some of the people outside to come in because y'all are standing. So please do come up, and then more folks will be able to join us. Thank you. Okay. So, Carlos, thanks for inviting me to talk today, and thanks to the organizers. I don't really believe in the pro-con structure. I think it's a false dichotomy. So instead of saying con, I'm going to say this is a vision for non-invasive, POCUS-driven, hemodynamic monitoring, and cardiogenic shock. I'm Matt Shuba. I'm a medical intensivist at the Cleveland Clinic in Ohio. So this is kind of a weird talk for me to give for a couple reasons. I actually love PA catheters. I think they're pretty neat. They can provide some useful information. But then on the other hand, I've got this whole minimally invasive, maximally attentive thing that I favor. So I'm kind of a little bit in a bind. And then probably most of all, I'm the only one up here that's not a cardiologist. So I feel a little bit like I'm a lamb being led to slaughter. So we'll talk about a couple of items today. If you had the perfect tool, what would it look like to hemodynamically monitor this patient group and talk about systems of care, advanced POCUS applications, cardiogenic shock, what the hemodynamic monitoring continuum looks like, and asking the right questions of the tools that we have. So for me, the ideal tool in cardiogenic shock would be able to measure pressure-volume relationships, including the VA coupling relationships that were mentioned. I want to be able to measure filling pressure, cardiac output. It would be great if it's a continuous system. It's great if it's minimally invasive, and it needs to be routinely available. So we'll start with the gold standard technology, which is conductance catheter used to measure pressure-volume relations. It's great for filling pressure assessment, cardiac output assessment, pressure-volume relationships. Otherwise, not a great fit. And then the PA catheter. OK, we can measure filling pressures, cardiac output continuously. It's not minimally invasive for sure. And depending on the unit and the hospital you work in, it may or may not be routinely available to you. And then comes POCUS. There's something that can be said about the way it can help us understand pressure-volume relationships, especially VA coupling. Filling pressures can be estimated, cardiac output can be assessed. It's certainly not continuous. It is minimally invasive, and it's usually available. And then the last thing is minimally invasive cardiac output monitors, which I'm not going to talk about in great detail today due to time limitations. But the main advantages of these things are that they're continuous and they're minimally invasive. They may or may not be available in your unit, depending on what you have in front of you. They certainly don't tell us anything about filling pressures or pressure-volume relationships. In fact, most of them assume a CVP of 7 when they give you all the calculations they give. In cardiac output assessment, most of these things are non-calibrated, so I'd be cautious about using them to make a diagnosis. So I'm tasked to talk about problems with pulmonary artery catheters, so let's start there. We don't even necessarily agree on what we're measuring when we put these things in. If you ask an interventional cardiologist versus heart failure cardiologist, they measure different components of the assessment at different points. So we can't be consistent when we measure a wedge, where we measure the mean PA pressures, things like that. Some people, shockingly, just take the computerized output from the machine, like the MATLAB system. That terrifies me a little bit. And then how we measure things can change the diagnosis. This is a paper of ours about comparing end-of-expiry measurements of wedge versus mean measurements of wedge, and then also whether you use thermo versus direct FIC, and how it changes the diagnosis in patients with pulmonary hypertension. So there's a lot of nuances in the way that you interpret the information that you're being given. Then once we get to that point, can we even agree on what we do with the information once the information is there? As you can see from some of the poll questions, there's not even consensus. And as was mentioned, doesn't recent data suggest maybe there's some benefit to PA catheters in cardiogenic shock? On the left panel, I'll show the study that was already shown, which basically showed if you measured more things, then the patients did better, and I'll ask you to consider whether that strains credibility or not. And then the other part of this is, in the panel on the left-hand side of the screen, you can see that pulmonary artery catheterization was associated with a decreased risk of death, but also an increased risk of advanced therapies in the hospital. And this is, I think, kind of the authors telling on themselves, saying that maybe there's other reasons that people would be more likely to receive those things. So what does that actually mean? So if we try to escape the ESCAPE trial, then we move into a mortal time bias. So in these studies, we can't necessarily quantify which of these patients were, they don't have the same time zero. So if you died before you could get a pulmonary artery catheter, you died because you didn't have a pulmonary artery catheter, those things are viewed as being the same. We all know that's not exactly the truth. And then with the other study, the one that was referenced in the last talk, we have to talk about confounding. And by that, we have to use something called a direct acyclic graph. So if you're going to say that PA catheter placement is associated with survival, then you also have to consider the confounders in that equation. Things like, what are the system of care that this PA catheter is being placed in? Is this patient a candidate for advanced therapies? Because if they're not, they may not be likely to get one of these catheters. And then a whole host of unmeasured confounders, which we can't account for. But I think the point was well made, that we need to get objective data in order to be able to properly characterize patients with cardiogenic shock. So if we look at the sky scoring, which was mentioned, there's a number of things that we need to look at, filling pressures, cardiac output, and then some equations that are based on those values. So can we do that with POCUS? And the answer is yes. So to start with, we can assess cardiac output using LVOT-VTI, which can give us a stroke volume and then lead us to cardiac output. But how useful is that? Or how does that compare to gold standard? Well, it correlates pretty darn well with thermodilution cardiac output. And the agreement is decent. And probably more importantly, we can argue whether actually the precise number of cardiac output is important, rather than the relative value. And then also, how does it change? And this four quadrant plot shows that changes in cardiac output by pulmonary artery catheterization are reasonably assessed by trans-thoracic echo. If you want to learn more about that, there's a whole systematic review about it. So what does this mean? POCUS estimates of cardiac output can be used for diagnosis and therapeutic monitoring in some cases of cardiogenic shock, some. So now we'll talk about filling pressures. So this is a giant table. Don't read it. But if you want to take a picture of it, you can read it. You can read it later. So there's a number of ways you can assess right atrial pressure in patients with ultrasound. We all know about IVC dimensions, but there's a number of other things that we can use to assess that. I will just remind you that you need to be cautious in using these things on patients on mechanical ventilation, because it's not validated in that setting. And then probably more importantly, once again, the exact number of right atrial pressure is probably not super important, but the relative value. And then also, what is the organ-specific effects of this? So we have this beautiful scoring system called venous excess ultrasound, where we can actually see what is the effect on the portal vein, the intrarenal vein, and things that actually correlate pretty well with outcomes like acute kidney injury. There is also the opportunity to use POCUS to assess the left atrial filling pressure. This paper is so good. I'm going to show you the reference again in the next slide, but all my references are QR codes. But this is a really nice paper about assessing left atrial pressure in critical care. And there's a number of ways you can do this. And once again, you don't want to just assess a raw number. You want to assess what are the end-organ effects of this. So if you have somebody that has estimated high left atrial pressure on the basis of their echo, and they have B lines, then you say, okay, this is a consequential elevation in left atrial pressure. So this is the paper that I'm referencing, and most importantly in this paper is a whole table that talks about the limitations of both pulmonary artery wedge pressure and echo estimates compared to the gold standard of LVADP measurements. So what does this all mean? POCUS can help us estimate filling pressures and can also tell us about end-organ effects via vexus and lung ultrasound. And you can even take this a step further, and you can actually measure VA coupling. And this is definitely advanced, advanced skills. There's probably not many of us that do this routinely. But these are things that you can do. You can assess LV arterial coupling with echo-based measurements, and you can also assess RVPA coupling measurements. And there's two papers cited there, but if you want to learn about this more, on my Twitter page, the pinned tweet is a 20-minute lecture on non-invasive estimation of RVPA coupling. What are the caveats here? If you don't have good windows or your technique is bad, you're going to have no data or you're going to have bad data. Filling pressure estimation is not validated truly in ventilated patients, so I think you need to regard this carefully. And this is, by definition, non-continuous monitoring. Every time you want to get these estimates again, you've got to reel the machine back in the room and do it all over again. And really, probably importantly, if we're talking about putting somebody on mechanical heart, you really can't echo through that. If you try to do an LVOT-VTI through this, people are going to laugh at you. So what are the key points of my presentation? The right tool really depends on the team, the patient, and the situation we're in. There's no such thing as all patients need this or no patients need this. Dynamic assessments always trump static assessments. So it's great to have the information from a pulmonary artery catheter, but you really have to be really careful of how you use it. It really hurts me when I hear something like, well, the wedge is low, so we should get fluids. No. That is not how this works. Pulmonary artery catheter, make a Starling curve. It's sitting there right in front of you. Having the right information can't make good decisions for you. That's super important. This is my mental model of hemodynamic monitoring. As the severity of illness rises, especially if there's reversibility, you may tend to be more invasive, and that might be appropriate to the patient. Or if you have diagnostic uncertainty, you might need to be more invasive. You have to have the right information to take care of the patient. If you don't know why the patient's in shock, you don't get to guess. You have to figure it out. Most importantly, no monitoring tool will ever save lives unless it's tied to an intervention which does. POCUS, PA catheters, regardless. It does not matter. But you have to take the information and use it in a way that actually the patient can benefit from it. And what this all means is context is everything. Thank you. Thank you, Maddox, fantastic. We're probably gonna take a couple of minutes for Q and A's and I probably wanna start with kind of contextualizing the two talks. You talk about the when to use a PA catheter or a non-invasive tool. And if we use a PA catheter, how to use it because this paper from the Journal of Cardiac Failure was very interesting, how interventional leads in heart failure approach it. And then once you have that information, if you did it in the right way, what to do with that? So perhaps for, maybe for Lakshmi initially, what do you think will be in the how? We have to educate our colleagues, we have to educate our trainees, we have to educate ourselves on how to deal with this. And I think there's a lot of literature in pulmonary hypertension, when to measure these patients, when to measure the patient in the cath lab, we have in very nice settings, maybe a little bit of sedation, as opposed to the crash and burning cardiogenic shock in the unit who is the synchronous with the vent. How do we improve the education? I don't know, what we think would be the best way to address that. So Carlos, so let me understand, you're saying how do we improve that education? In terms of PA catheter. Yeah, and I think that's the point exactly. The fact that we don't use them enough means that folks aren't going to get comfortable with them. But the truth is, that's true of any tool, advanced focus imaging or PA catheter utilization. I think for us, in our unit, we put swans, our APPs are trained to put swans in by themselves in the middle of the night. And that requires a lot of attending involvement, that requires buy-in from our whole cardiogenic shock team. And that also, I think, requires that you have a systems-based support where you're making a multidisciplinary decision on how to manage shock patients. And for that, you may say in some patients, hey, this is what the interventionalist said when the patient was sedated, but when we were actually managing the patient in the unit, these are the numbers we got. And the surgeons may then say, well, if the pulmonary pressures are that variable, this patient is too high-risk for transplant. And I think that's the point, to be able to integrate all that data. So when the question is, how do you educate and get better at using them, I think you just have to start putting them in more in order to try, fail, and learn again, because that's how we get better in our unit. Perfect. Yeah, I agree with all that. I think in general, the problem with any of the technologies is if you're hinging anything on what you're doing on one data point or on one set of measurements, you're going to be led astray. So similarly, if you apply an intervention to a patient and then you don't reassess the patient, there's no device in the world that's gonna help you. So the problem is, we'll see, okay, the index is this, we put them on dobutamine, we did our job, let's go home. No, you actually have to reassess all the parameters that you assessed the first time. And then also recognize, as Lakshmi said, that there are times where the patient changes and you need to be able to reassess that because none of these things are static. That's a great point, and it actually transitions well into what I wanted to ask. You talk about the dynamic nature of going back and reassess. And I think when we have invasive hemodynamics, we can look at the continuous linear change. Is the PA pressure going up or down? Is it responding to the strategies that we're doing, let's say diuretics, pulmonary vasodilators, MCS? What do we do with the data of non-invasive assessment, particularly echo? You know, the LA pressure is more semi-quantitative. Vexus itself is a score, but it's also semi-quantitative. And there's some folks that really criticize Vexus because of the issues with the IBC. But for example, how do you integrate portal vein pulsatility or those more not linear parameters in reassessing? You do it in the next six hours. What is your specific practice? So for me, it depends on what interventions that you've made and how quickly you expect to see an effect from them. So for something like Vexus, depending on what, if you treated, by whatever means you want, RV dysfunction, and by the way, come see us talk about RV dysfunction tomorrow. If you treated RV dysfunction and you think whatever treatment you gave, some sort of reduction of afterload or some sort of inotropy or something, that should be a quick effect. So maybe you should reassess that in the next couple hours. If it's something that you expect slower, I'm decongesting the patient, they're on CRT or I'm diuresing, maybe then you're checking six to 12 hours. I think it's all about on what time horizon you think your intervention should have had an effect. Because realistically, things like portal vein pulsatility could be, it is quantitative and it is dynamic. So it's something that could change quickly. That's a great point. I think it calls again for the integration of the data. You have given the patient diuretics and they didn't put out any urinary output. You don't expect to see a significant change unless something else happens. So that's great. So now we're gonna move into our next topic which is going to be mechanical circulatory support. It's been a matter of controversy in the last few years, in particular with more recent data that is finally growing in the areas of MCS and randomized clinical trials. It is my pleasure to introduce Anne Gage, friend and colleague who is gonna be talking about the pros of MCS. Hi everyone. Good morning and thank you to CHEST for hosting this session. I'm going to spend the next 15 minutes debating the illustrious Dr. Jentzer on the use of mechanical circulatory support in cardiogenic shock. So my name is Anne Gage. I'm an interventional cardiologist and also a critical care physician in Nashville, Tennessee. These are my disclosures but probably the most pertinent disclosure that I have is actually that Dr. Jentzer and I are friends so any justs or jabs are all for the sake of science, nothing personal. So my objective is to convince the audience there exists a role for mechanical circulatory support in cardiogenic shock. When the audience leaves this session, you will more astutely understand the limitations of the data cited by those not in favor of mechanical circulatory support and I believe that you will understand how mechanical circulatory support devices likely play a role in protocolized, goal-directed care for cardiogenic shock. So the question central to our debate today, is there a role for mechanical circulatory support in the management of cardiogenic shock? Well, the challenge for me is that the data is not on my side. I have very little data on which to argue this but I'm up for the challenge. So here are some of the many trials of mechanical circulatory support in cardiogenic shock, randomized trials that are all have been proven to show no mortality benefit. IABP shock, IABP shock two, tandem heart versus IABP, Impress, Impella stick, ECMO-CS, ECMO shock. Taken together, it is fair to say that randomized trial data for the utilization of mechanical circulatory shock or support has failed to show a mortality benefit. Furthermore, observational studies have not shown a benefit or have been inconclusive. But my objective is to convince this audience that there exists a role for mechanical circulatory support in cardiogenic shock. How can I do that? Well, the answer is that we have to move past our reliance on a randomized controlled device trials. There is no mechanical circulatory support device that is the silver bullet. Put another way, we cannot continue to design trials that fail to recognize both the troubled engine and the flat tire. While a new engine is necessary in this car, this car isn't going to go anywhere until we treat the flat tire. Dr. Jimser is going to show you some very well done randomized controlled data done by many of our very intelligent colleagues. But the underlying issue is that all of this data was designed to test if putting a new engine, aka mechanical circulatory support, in this car would change outcome. And the answer is obviously going to be no. So here is where we should look to one of the rare positive trials in critical care. I think this is a paper that this audience is very familiar with. Published in the New England Journal in 2001, Dr. Rivers and his colleagues developed a protocolized bundle of care for the treatment of sepsis. Prior to this trial, sepsis trials frequently focused on singular interventions such as hemodynamic optimization, immunotherapy, trials for pulmonary artery catheters. And they often enrolled patients that had been hospitalized in ICU for up to 72 hours. This study protocolized bundled care and targeted multiple endpoints. And while this data has evolved over time and certainly isn't perfect, this is the type of intervention that we should be looking to for our care of cardiogenic shock patients. And so herein lies my argument. Do not vilify the device, vilify the data. There is data that we can alter cardiogenic shock outcomes via a protocolized bundle of care targeting multiple endpoints, I'll walk through that shortly. And inappropriately selected patients, mechanical circulatory support will be a part of this bundle of care. So what is the current state of the data for mechanical circulatory support? I adapted one of Jake's own figures for this point. So as Dr. Jimser is going to tell you over and over, they're all randomized control trials have been negative. But again, rather than vilify the device, I'd suggest these trials have been negative due to lack of timely recognition of cardiogenic shock, lack of timeliness to treatment initiation, failure to account for the heterogeneity of etiology and phenotype of cardiogenic shock, lack of support for concomitant pathologies that may increase mortality, inclusion of patients unlikely to benefit, and finally, failure to protocolize additional components of the patient's care. So for the sake of time, I'm gonna highlight only a couple of examples of how these issues plague the data that Dr. Jimser is going to use to make his argument. But please know that these issues plague the research in this space, everything that he shows. So the lack of timely recognition of cardiogenic shock means that patients in clinical trials, the ones that are about to be cited by my colleague, these have enrolled patients that have progressed from hemodynamic failure to hemometabolic failure due to lack of timely recognition of their shock. We know that the hemodynamic failure associated with cardiogenic shock is associated with decreased cardiac output, increased SVR, decreased MAP, and systemic hypoperfusion. But with time, hemometabolic failure ensues. And this is now characterized by patients who have more hypoxemia, higher levels of lactatemia, increased metabolic acidosis, onset of multisystem organ failure, and systemic inflammatory response syndromes. Temporary mechanical support is best matched to the physiology of hemodynamic failure. This physiology of hemometabolic failure is not expected to respond to hemodynamic support alone. This lack of timely diagnosis then lends itself to the timeliness of treatment initiation. So this is from a study of Dr. Tarani and his colleagues in 2019. This is a group out of Inova. They published their experience prior to and after the development of a shock team. And they showed that in patients requiring mechanical circulatory support that every one hour delay in escalation of therapy was associated with a 9.9% increased risk of death. I think that these findings suggest that in appropriately selected patients, mechanical circulatory support has a benefit. But the randomized controlled trial data that you're going to see has failed to really get to this point because we failed to recognize the shock in a timely fashion initially and then we certainly haven't initiated treatment fast enough. The current trial data we have also fails to account for the heterogeneity of phenotype of cardiogenic shock. So you'll see this, you've seen this slide before, but ultimately this is greater than 1,400 patients from the cardiogenic shock working group. They adjudicated the hemometabolic profile and congestion profile using invasive hemodynamics of over 1,400 patients. And what you can see from their shock population was that greater than 58% of patients actually had biventricular failure or right ventricular failure. When you look deeper at this data, so these are the tall gray bars, that the mortality associated with RV and biv failure is much higher than LV predominant failure. And that the figure on the bottom left, that if you look at Sky Stage C, D, and E patients, those patients that we would be talking about putting on mechanical circulatory support, that over 50% of these patients had biv failure or RV failure. Well, why does that matter? In these patients, which is greater than 50% of patients, any trial of a percutaneous LVAD or a balloon pump is destined to fail. These devices treat LV dysfunction, not biventricular or RV dysfunction. Our trials fail to always phenotype these patients appropriately. So current randomized control data also fails to account for the heterogeneity of etiology for cardiogenic shock. So Dr. Jenser will inevitably cite this paper, the IABP Shock 2 trial, which was a randomized control trial to compel you not to use a balloon pump in the treatment of cardiogenic shock. This trial used balloon pumps and acute myocardial infarction shock for patients undergoing early revascularization and then randomized them to insertion of a balloon pump or no balloon pump. And as you can see from the central figure of the paper on the right, the conclusion was that the use of an IABP did not significantly reduce 30-day mortality in patients with cardiogenic shock complicating acute myocardial infarction. The problem is that what the IABP Shock 2 trial failed to recognize is that not all shock is acute myocardial infarction shock. As you learned in our first lecture, this is actually becoming the minority of shock in this country. So we have to account for what the etiology is and that the etiology may respond to mechanical circulatory support in different fashions. So what you see here, this is retrospective cohort data from Columbia, and they looked at their patient population that had cardiogenic shock who were treated with IABPs. And what you can see from a cursory glance here is that there is significant variability in the hemodynamic response to the IABP based on the underlying etiology of the patient's cardiogenic shock. So as our trials have historically done, the problem is that we ignore heterogeneity of disease state with cardiogenic shock and that given data demonstrating a variable response to mechanical circulatory support by etiology, we cannot simply generalize acute myocardial infarction trials. So one of the last points I want to highlight about the flaws in our currently available randomized control data for the treatment of cardiogenic shock is that we frequently include patients that are unlikely to benefit. I actually kind of find this trial amazing. So this is exemplified by the ECLS shock trial. This was published just about six weeks ago in the New England Journal of Medicine. This trial randomized AMI shock patients for whom early revascularization was planned to either ECMO or no ECMO. And as shown in the central figure on the right of your screen, you can see that this led to the conclusion that in patients with acute myocardial infarction complicated by cardiogenic shock, the risk of death from any cause at 30 days was not lower in patients who received ECMO versus those who did not receive ECMO. So this trial will be weaponized to cite ECMO as an ineffective means of treating cardiogenic shock, but there really are some amazing points about this trial. Namely, 50% of the population was Sky Stage C heart failure. If you look at the bottom left figure, you can see that Stage C heart failure is hypoperfusion without deterioration, and that if you look in the figure on the right, that if you looked at observed mortality of the acute coronary syndrome population with Stage C shock, these people only have a mortality rate of less than 10%. So in many hospital systems, Stage C patients are often being treated on step-down units, not even in ICs, and in most places, I would argue, they should not be offering, be offered ECMO at all, but now we have a New England Journal paper that will be used to say that New England, that these patients should not be treated with ECMO. This was a problem of including patients that were unlikely to benefit. So how can I argue that there's a role for mechanical circulatory support in the management of cardiogenic shock? Well, because we're finally starting to see data emerge that a protocolized approach to the care of cardiogenic shock patients can improve outcomes or survival, and these protocolized approaches confront the limitations of trial design that I just mentioned. So put simply, this data is reminiscent of the protocolized approach to the care for sepsis that we saw Dr. Rivers have greater than two decades ago, and this moves us away from randomized device trials where we simply place an LVAD or a percutaneous LVAD in patients regardless of the underlying etiology, phenotype, or hemodynamic profile of a patient with cardiogenic shock. So in the next few slides, I'm briefly going to show you the data suggesting that protocolized care using mechanical circulatory support can improve survival. So this is data from the National Cardiogenic Shock Initiative. This was published in 2019 by Dr. Basir. This was a trial of 35 different sites across the United States, all of whom agreed to treat acute myocardial infarction shock according to the protocol on the left side of your screen. You don't have to read this, but the pertinent takeaways here are that everyone needed invasive hemodynamics, and there was heavy reliance on the use of mechanical circulatory support. And the National Cardiogenic Shock Initiative showed a 72% survival to discharge in patients who had AMI cardiogenic shock. You can see that compared to historical comparators, this was a significant improvement in mortality. I think that these data suggest that mortality may be modified by mechanical circulatory support when it's used in a timely fashion, when it's matched to underlying hemodynamic profile, and when pre- and post-device care is dictated by best practices to minimize complications. We also have observational data from four single centers, suggesting that the implementation of cardiogenic shock teams is associated with a mortality benefit. Why do I care about this? Well, because these shock teams all use statistically higher rates of mechanical circulatory support in their post-implementation of a shock team. I think that this suggests, again, that mortality may be impacted when MCS is used in a timely fashion, when the right device is matched to the right patient, and when multidisciplinary care is protocolized, or used in a protocolized fashion. And the last small amount of data that I have to stand on up here is that more recently, there is observational data from the multi-center cardiac critical care trials network that again shows that shock teams improve mortality. You can see that on the far right side of the screen here, where the dark bar shows that in institutions with a shock team, their lower mortality is lower. Well, how does this relate to mechanical circulatory support? Well, these teams use statistically significant higher numbers of mechanical circulatory support. So in just a moment, Dr. Gensler will attempt to convince you that the current data does not show, or does not support the use of temporary mechanical circulatory support devices. But I know this audience won't be fooled. The data on which he makes his argument is flawed in all of these many ways. And this data, or in this audience, is too smart to believe that just because a new engine fails to make this car go means that the new engine is the problem. So the answer is that current data suggests MCS may improve outcomes when you utilize in a protocolized bundle care targeting multiple endpoints. And I look forward to the day when we can apply goal-directed therapy to our cardiogenic shock patient population, targeting timely recognition of cardiogenic shock, early initiation of therapy, matching of device to hemodynamics and etiology, and application of best practices pre- and post-device care to minimize complications. And this can be the resulting New England Journal article, which we can use to improve the care of our patients with cardiogenic shock and tell Dr. Gensler for once and for all how wrong he and his data are. Thank you, and that was great. So now we invite Dr. Jake Gensler for the response. Looking forward to it. We'll see if these slides are going to work, although I could probably just use Dr. Gage's slides, which are undoubtedly better than mine. So I'm Jake Gensler from Mayo Clinic, and I'm here to convince you the contrary, that mechanical circulatory support is overused in cardiogenic shock. So as you see, Dr. Gage is a very hard act to follow. You're probably at this point wondering how I could possibly have any hope of holding my own in this debate. Fortunately, I do have a secret weapon. That secret weapon is cold, hard, fast. So there are a lot of reasons to think that we should be using temporary MCS in cardiogenic shock, and this is the reason that thousands and thousands of these devices are used worldwide, not to mention in the United States. So we know that mechanical circulatory support can increase cardiac output, can increase mean arterial pressure. This can restore systemic perfusion. Some devices can unload the ventricle, which is beneficial, and hopefully this will all allow us to wean from vasoactive drugs, which we all know are toxic. Unfortunately, the current generation of devices can do this, but at risk of very substantial complications in many cases. There's no free lunch here. The more support that an MCS device provides, the larger a catheter it requires, and the greater the risk of complications it will cause. And as I don't need to convince you, because Dr. Gage already did, these devices do not improve survival, and they cannot salvage futile cases. It is undoubtedly true that these devices can improve hemodynamics. In many cases, the improvements are less than you might expect. This particularly is true with the introverted balloon pump, which can have some fairly dramatic variation, and in some cases can actually worsen output. But even some of the more robust devices, including the smaller Impella, they don't increase the cardiac output quite as much as you think. For example, the Impella screen will tell you, gee, it's giving you more than three liters of flow, but because it reduces preload and reduces native cardiac output, the actual net increase in cardiac output may be as little as a liter. But more importantly, we know that many patients with cardiogenic shock die despite the fact that we have successfully normalized their hemodynamics. And they often die from non-cardiovascular causes, such as anoxic brain injury after cardiac arrest, or multi-organ failure, which is all too common. And so I would propose to you that we should not routinely be using devices that can improve hemodynamics unless they can also improve patient outcomes. You've already seen this data. This is the largest randomized trial of mechanical circulatory support devices. This is introverted balloon pump shock two, which again was balloon pumps in AMI cardiogenic shock. It clearly demonstrated no improvement in survival, no improvement in other patient-centered outcomes, and no improvement in hemodynamics. Fortunately, this device was actually relatively safe and didn't harm the patients. Nonetheless, current guidelines from the European Society of Cardiology give this a class three recommendation, which means it should not be used routinely. What about other more expensive devices, percutaneous ventricular assist devices? Well, unfortunately, the data for these is based on some small cohorts, total of less than 150 patients have been randomized, and again, they compared to balloon pump. There were better hemodynamics with these devices that we can all agree on, but there was no improvement in survival, not even the suggestion of improved survival, and there were definitely more complications. So again, no evidence that these devices are better than a balloon pump, which itself is not better than not a balloon pump. What about ECMO? ECMO is, of course, the future. It can do all the things. It can support both of your ventricles. It can replace the entire native cardiac output. You don't even need lungs. It can replace those too. And so this is hot off the presses ECLS shock, and these were patients, again, with acute MI cardiogenic shock. This is one of the largest randomized trials in cardiogenic shock as well, and this was upfront ECMO versus rescue PVADs, which were used only in a very minority, small number of patients, and again, no difference in survival to 30 days, but clearly more peripheral vascular complications and bleeding with ECMO. So that means that ECMO caused no benefit and actually caused harm. So is the next thing we're going to hear about a class three recommendation for routine use of ECMO in cardiogenic shock, and I think that's an important thing to see if that happens. And again, this is a meta-analysis that I did with several folks, including our esteemed Dr. Alviar, and so again, when you put all of the trials together, knowing that yes, many of them are small, yes, many of them are low quality, they still fail to demonstrate any benefit of any of these devices versus any other of the devices or medical therapy. So this is a forest plot showing the comparison of each of these individual devices to medical therapy alone. It would be very hard for me to convince you that there's any actual difference in outcomes. You've seen a version of this slide before, and again, why are these studies negative? Well, if we assume that these devices actually work, which is, I'm not willing to concede, but let's just take that for the sake of argument, there are lots of reasons why trials might still be negative despite devices that can do good things. The key issue is, first of all, many of these trials may in fact have been underpowered, and it's very hard to interpret studies that are underpowered and fail to show a difference. Is that because there is no difference or just because you didn't have enough patients to show a difference? But more importantly, it is definitely true that the enrolled patients in these trials may not have been the ones most likely to benefit for a variety of reasons. Many of them had anoxic brain injury after cardiac arrest. Many of them probably had this hemometabolic shock with established multi-organ failure, which was perhaps too late for the devices to work. Again, these were pretty much all acute MI cardiogenic shock patients, and maybe they're different than these more numerous heart failure shock patients. There are probably a lot of patients in these trials who just weren't gonna benefit. If you look at the average age, the average age is on the order of 70, and we know that many of those patients are not going to be candidates for advanced therapies should they fail to recover. And finally, there was poor matching to shock severity. A lot of the patients in these trials had severe enough shock that a balloon pump probably wasn't the answer, but maybe not severe enough that ECMO was needed. And what I really, I wanna simplify this. There are really three relevant groups in any clinical trial population, particularly cardiogenic shock. There are the patients that are not sick enough to need MCS who are gonna do just fine without it. They're not gonna benefit. They're just gonna be exposed to complications. The other end of the spectrum are the patients that are practically futile. They're so sick that even if you do MCS, they still won't survive. And again, they're not gonna benefit. But the ideal candidates are these ones in the middle. They're sick enough that MCS will be important for them to stabilize their hemodynamics when medical therapy is failing, but they're not so sick that they're in a nearly futile state. And these patients indeed might benefit from MCS. The problem is that this group may be quite small, and we really can't determine who these patients are a priori. We can predict who with cardiogenic shock is likely to die or less likely to die, but that is not at all the same as who is or is not likely to benefit from an intervention. That's a completely different question. And the key thing is that percutaneous or temporary MCS, it's a bridge, it's not a solution. It doesn't really fix anything. It keeps the patient going until you can do something better. Either the patient is kept alive until they can recover or kept alive until they can get an advanced heart failure therapy such as durable mechanical circulatory support or a transplant. So when do I actually use these devices? Well, I sometimes put the cart before the horse and I ask myself, is this a patient who's going to be a candidate for advanced therapies because I only use these devices as a bridge either to recovery, decision, which essentially means they're so sick, I don't know what to do, let's keep them alive and figure out what's next, or someone who's actually going to get an advanced heart failure therapy. I specifically use these devices for patients when vasopressors, inotropes, medical therapy is inadequate to achieve my hemodynamic goals or is causing toxicity that I think is potentially dangerous. I specifically focus on mechanical circulatory support for treatment of low output. We know that patients with cardiogenic shock can get vasoplegia and I typically don't think that that's an effective strategy using MCS for vasoplegia. And finally, I try to focus on patients who don't have irreversible end-organ dysfunction. If the patient has non-survivable anoxic brain injury, there's no reason to think that MCS is going to help them. So this is a different version of the same three-axis model you were shown earlier. So this was proposed by Sky as a way to conceptualize patients with cardiogenic shock for the purposes of risk stratification and decision-making. The first axis is shock severity. How bad is the sum total of the patient's shock? This will help us decide, is their shock bad enough to need MCS? And if so, how much flow or how much support will the patient need? The next is the phenotype and etiology which helps us figure out, is MCS actually the right thing for this patient? What type of device is appropriate? Is this univentricular, LV, RV, or is this biventricular? And what is your target of bridging? Is this someone who's likely to recover or not? And finally, this concept of risk modifiers. These are potentially non-modifiable risk factors for death that might tell you, gosh, this person is not going to be a candidate for advanced therapies or actually maybe the opposite might be true and they do in fact have a high chance of recovery. So the shock severity is, do they need it? The phenotype is, what type? And the risk modifiers are, should I do it? And what this tells us is that individualized use of MCS is the only way to do this. Routine use in unselected patients has failed unequivocally and the best way to do individualized decision-making is using a multidisciplinary team. You've already seen this data on shock teams and I'll sort of go through this a little bit more because I think this proves my point. So this is a multicenter study of high-level cardiac ICUs at academic medical centers, really the apex of cardiac ICU care. And what they found is that the patients with cardiogenic shock who were taken care of at hospitals that had a shock team actually received less mechanical circulatory support, although they did more often receive the more advanced devices proportionally. But despite lower overall use of MCS, patients cared for in these shock team hospitals had better survival. So this tells us that it's not about using MCS more, it's about using MCS better. So I'm going to summarize by talking about facts again. What facts can be shown based on published literature, particularly randomized trials? We can show you that most patients with cardiogenic shock will survive without MCS. We can show you that patients with refractory shock who receive MCS do quite poorly. We can show you that patients who get MCS have more complications and we can show you that temporary MCS is expensive, not just the device costs, but the overall cost of care and overall healthcare costs. What can we not prove? We cannot prove that routine use of MCS improves patient survival. We cannot identify consistently patients with shock who are going to benefit from MCS and we cannot demonstrate that MCS is either cost effective or even safe for patients with cardiogenic shock. So by definition, if we are using temporary MCS devices frequently for patients with cardiogenic shock outside of the context of clinical trials and these facts are true, MCS must clearly be overused. So I thank you very much for your attention and I hope to answer your questions after this. Thank you, Jake, that was truly outstanding and really entertaining from both parties. So let's get into a little bit of the controversy. We only have a couple of minutes, but I wonder if we are perhaps barking at the wrong tree, meaning we just had a paper published from C3TN demonstrated that cardiac output doesn't necessarily predict outcomes when we look at specific hemodynamics. Maybe short answers, we can move to one of the audience response, but what do you guys think? Is it just the cardiac output or how we deal with that? So I think that the problem with cardiogenic shock is we don't view it as a multisystemic disease often enough. We think of it all low cardiac output, fixed low cardiac output, but as I think all of us have shown you, it is a multisystemic disease that affects all the organ systems through a downward spiral of poor perfusion causing secondary complications that compromise cardiovascular function. And so I think that the observation that cardiac output is not the be all and end all is absolutely true and has been actually known for many years and I think the more complex question is among the therapies that we can use to improve organ perfusion, which ones do it in the most safe and effective manner because that's how we're going to hope to improve outcomes. Yeah, I absolutely agree with that. I think that cardiac output becomes less important as you progress down the spiral of hemodynamic failure. And so unless we're catching patients before the onset of their multisystem organ failure, before we are unable to support the liver, the kidneys, these are, cardiac output at that point really becomes fairly irrelevant. And so if you catch them soon enough, modifying cardiac output is going to, I think, be impactful. Great point. So Dr. Barnett, can you share your question? Yeah, Chris Barnett from San Francisco. That was great. I mean, so much of this revolves around shock teams and I was wondering if you guys could talk a little bit more about what you do if you're in a hospital without a shock team. So if you think you have somebody who might need a PA catheter, might need MCS, and you don't have a shock team or you don't have the kind of shock team that you do, like what does one do? I actually think that if you tie some of these things together, what we talked about today, that the initial assessment of hemodynamics, this really relies on astute physicians and providers that are meeting these patients on the front line. So whether that's you're the physician on the floor or you're the physician in the ICU, using hemodynamic parameters, blood pressure, heart rate, that's the first line of defense. Then what does your physical exam tell you? What's their mental status? Are they cold on exam? This is where we should be, I think, focusing kind of a continuum of PA catheter use or invasive hemodynamics, is that most people aren't going to throw in a swan. And so we really need to be teaching our providers to be thoughtful about whether or not this is cardiogenic shock or not. And then you can move on to things like echo, where what is the EF? What is the RV function? What's the LVOT VTI? What's the RVOT VTI? And when you synthesize these things together, you've initiated the shock team process, right? You've said, hey, this looks like we have, this patient possibly is in cardiogenic shock. And then this moves into the realm of really getting your advanced critical care colleagues or your cardiology colleagues involved. And most of the time that's going to be that we really do have to phenotype shock completely to understand how we're going to treat it. And I think that's where you get into invasive hemodynamics. And so you might not have a cardiogenic shock team available, but if you have frontline physicians who are adept and thoughtful enough about making this diagnosis, I think that that's put you well ahead of the majority of institutions in this country.

hospitals without dedicated shock team

frontline physicians

cardiogenic shock

hemodynamic parameters

physical exam

echocardiography

structural abnormalities

invasive hemodynamic monitoring

treatment strategies

multidisciplinary approach