Sepsis-induced cardiomyopathy: What is it and what do we do about it? Try out using Rapid Sequence for groups, in conference, or as an asynchronous learning activity: rapidsequence.org/for-educators References: Jentzer et al. Systemic Inflammatory Response Syndrome Is Associated With Increased Mortality Across the Spectrum of Shock Severity in Cardiac Intensive Care Patients. Circ Cardiovasc Qual Outcomes. 2020;13(12):e006956 Chavez et al. Pathophysiologic Vasodilation in Cardiogenic Shock and Its Impact on Mortality. Circ Heart Fail. 2024;17(9):e011827 Shapiro et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. NEJM 2023;388(6):499-510 Sato et al. Effects of Inotropes on the Mortality in Patients With Septic Shock. J Intensive Care Med 202;36(2):211-219 Tarbaghia et al. Dobutamine in septic shock, when and why? Am J Respir Crit Care Med 2025;211:A3748 Sato et al. Sepsis-induced cardiomyopathy: Mechanism, Prevalence, Assessment, Prognosis, and Management. Chest 2025; 3:S0012-3692(25)05143-8 Vallabhajosyula et al. Impact of Right Ventricular Dysfunction on Short-term and Long-term Mortality in Sepsis. Chest 2021;159(6):2254-2263
Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.
Transcript
From the show. Tap a time to play from there.
SPEAKER_00Hey everybody, this is Sarah Gregor. I am an emergency physician and intensivist, and this is the ICU EDU podcast. Today we're talking about sepsis-induced cardiomyopathy. But first, quick update on rapid sequence. So, um rapid sequence is the clinical cases resuscitation game where you get to choose your own adventure and put into practice a lot of the concepts and physiology we talk about on the podcast. I've heard from a number of people that people want to try playing it in groups, in conference, or asynchronously, and then maybe discussing it as a group.
SPEAKER_00So we've created a test with the demo case for people who want to try that. And it's a link that gives you a really easy plug-and-pay option for people to try it out either asynchronously or in person in educational conference. So link will be included in the show notes. And overall, keep the feedback coming because it's been incredibly useful. Now let's talk about sepsis-induced cardiomyopathy. Sepsis-induced cardiomyopathy, what is it? The definition of sepsis-induced cardiomyopathy is septic shock with new onset cardiac dysfunction or worsening of pre-existent cardiac dysfunction.
SPEAKER_00And whenever I see one of these patients, it always makes me want to be like, can you just please pick one? You only get to have one, not both. So just commit to whichever one of these, sepsis or heart failure that you want to run with. Because these patients can be hard to take care of. They're almost always really sick, and they have competing physiologic priorities. And how you approach them requires being very thoughtful about their pathophysiology, but also being very careful to trend that pathophysiology to figure out what is the most important component of their shock right now.
SPEAKER_00But probably the most important thing I'm going to say about sepsis-induced cardiomyopathy is actually before we get into the details of sepsis-induced cardiomyopathy itself. We need to go back to step number zero, which is recognizing that this is a diagnosis of exclusion. So before I head down that path, there are three major things that I want to make sure I've thought about first. One, misdiagnosed cardiogenic shock.
SPEAKER_00Two, sick myocarditis, and three, septic shock with baseline cardiac dysfunction. Okay, so the first one, misdiagnosed cardiogenic shock. I think one of the biggest mistakes we make either in the ED or on the floor or certainly in the ICU, is when we are initially assessing a hypotensive patient, especially a hypotensive patient with a high lactate, we are tempted to go straight to septic shock. Now, part of that is not our fault, part of that is we have all these metrics to meet and so forth and so on.
SPEAKER_00But it's really tempting because it's also often perceived to be easier to manage septic shock rather than cardiogenic shock. Question is, well, did we just get it wrong in the first place? We labeled it as septic shock, but is that actually what it was to begin with? Or did we just miss cardiogenic shock? Do they actually have a source of infection? Let's just start there. But it's also sometimes not totally straightforward to tell the difference. So, SERGE's criteria, for example.
SPEAKER_00I was certainly taught that yes, SURS criteria have something to do with sepsis and infection, and it turns out that is simply not the case. SURS criteria, all they mean is an inflammatory state. Now, that could be from infection, it could be from pancreatitis, or you know what causes a lot of systemic inflammation, shock of any kind. When a patient's in shock, the body doesn't get perfused. And when the body doesn't get perfused, it gets kind of pissed off and it throws a tantrum by being inflamed. And if you look at some of the data on this, SURS criteria, positive SURS criteria, is actually quite common in patients coming in with just straight-up cardiogenic shock.
SPEAKER_00So this was a study from one of the American Heart Association journals that came out in 2020. And as usual, all of the literature I talk about here will be included in the show notes. But this particular study, retrospective, looked at 9,000 some patients admitted for cardiogenic shock to the cardiac ICU. And a third of them, a full third of them, were positive for SERS criteria. And not only that, but patients who were positive for SURS criteria, that was associated with increased mortality.
SPEAKER_00And that's not enormously surprising to me, because if the worse your shock is, the more your body is inflamed, the more SERS criteria I might expect you to have. And that sort of makes sense. Now, what about this idea of quote, warm shock, vasodilatory shock? Because I frequently have people come to me and they say, well, you know, I saw them and they were hypotensive and their lactate was high, but they can't possibly be in cardiogenic shock because their extremities weren't cold.
SPEAKER_00Now, in cardiogenic shock, can you absolutely have cold extremities? Yes. Am I worried about that? Yes. But there certainly at the same time is a phenotype of cardiogenic shock that has a vasodilatory phenotype rather than a vasoconstrictive phenotype. So this is from a study that came out in 2024. And this was a study that looked at vasodilation in patients with cardiogenic shock. And what was convenient about this study is that they actually had swan sense.
SPEAKER_00So when they were saying vasodilation, it wasn't like, oh, I don't know, did that patient look vasodilated to you? They could actually just measure an SVR. And these were patients admitted with cardiogenic shock who did not have concurrent sepsis and were not post-recent cardiac surgery. These were just straight up cardiogenic shock. And what they found looking at the SVR is that 38% of them actually had a low SVR, actually had this vasodilatory phenotype.
SPEAKER_00And as the SVR got lower and lower, that was associated with higher and higher mortality. Now, I think part of that probably is, again, a mark of illness severity that patients who are sick or have more inflammation, probably have more vasodilation. But also in cardiogenic shock, it can be much more difficult to manage the cardiogenic shock patient who's vasodilated because it can make it more difficult to volume offload them. It can limit my choice of inotrope.
SPEAKER_00So I can totally see that might be the case. So the conclusion here is that not only do we see this SRSy vasodilatory phenotype in something like a third of cardiogenic shock patients, that phenotypes associated with higher mortality. And so it could be very easy in that first initial evaluation of a patient to just say, oh yeah, clearly this is septic shock. So go back to the drawing board, and before you're saying, oh, sepsis-induced cardiomyopathy, make sure they don't just have straight-up cardiogenic shock in the first place.
SPEAKER_00Next item to think about on your list of things that you need to kind of mentally exclude before running with the diagnosis of sepsis-induced cardiomyopathy. And that is sick myocarditis. Now, these patients are particularly difficult because they actually do have an infection, right? They are definitely sursey, they are febrile, they're virally, they'll give you a history often of infectious symptoms. But that is almost always not the thing that's actually driving their shock physiology.
SPEAKER_00And these sick myocarditis patients, they usually don't come in shocky from their viral infection and sepsis. They come in shocky because they are in cardiogenic shock. And once I've identified that this patient just is really sick from myocarditis and that's why they're in shock, I just treat them like straight-up cardiogenic shock, the way that I would really treat just about any other cardiogenic shock patient in terms of their hemodynamic management. Now, are they more likely potentially to have a vasocilatory phenotype of cardiogenic shock?
SPEAKER_00Sure they are. That being said, as we just talked about, that's something that I'm trying to work through. What is their phenotype in terms of vasoconstriction versus phasodilation in any kind of patient with cardiogenic shock, regardless? So, sick myocarditis, I just treat like cardiogenic shock. Now, what about the third thing to think about when we're saying, okay, what else do we need to consider and exclude? And that is septic shock with baseline cardiac dysfunction.
SPEAKER_00So patients walking around on a normal day and they just live in an EF of 35%. And they come in and they have eurosapsis and they are in septic shock, and their EF is 35% as it usually is. Don't have worsening dysfunction, don't have new dysfunction. Now, it's kind of nice if this is the kind of patient who sees their primary care doctor four times a year and has three recent echoes, and you can be like, ah, this looks like their baseline cardiac dysfunction. One caveat about that is that often these patients will come in and you look at their echo report and you're like, oh yeah, their EF is only 35% at baseline.
SPEAKER_00And then you echo them and you're like, oh, actually, they're healed. It looks a lot better than that. Um, that's probably not what happened, is probably not that the sepsis magically healed their cardiomyopathy. What probably happened is that because of the sepsis, they're so vasodilated, their SVR is low. So now that 35% heart does not have to push against any afterload. So it's not so much that the heart got stronger, it's that what it's pushing against got lighter. Sometimes you can see a worsening of baseline cardiac function, but that then gets into the territory of true sepsis-induced cardiomyopathy.
SPEAKER_00And if I have a patient with septic shock, I believe that their cardiac dysfunction is essentially at their baseline, well, my only acute problem is actually the septic shock. And I treat those patients just like septic shock. Now, what about fluid evidence in these patients? The patient with baseline cardiac dysfunction who comes in in septic shock. We do have some evidence, mainly from the Clovers trial, that in terms of patients with baseline heart failure, baseline cardiac dysfunction, there didn't really seem to be a difference in those patients between a fluid liveral and fluid restrictive strategy in patients with this chronic heart failure.
SPEAKER_00And so sometimes you do actually need to give them just as much fluid as you would give any normal patient with normal cardiac function with septic shock. Now, that being said, in patients with baseline heart dysfunction, you know, I'm not as likely to give it all at the same time. Because those patients often have maybe a little bit more of a smaller therapeutic window in which they are fluid responsive. And so I like to do possibly the same amount of fluid, but rather than doing it all in one go, I will sometimes give aliquats, reassess, aliquats, reassess.
SPEAKER_00Okay, so those are the three things that I need to make sure I've mentally checked off before I run with a diagnosis of sepsis-induced cardiomyopathy. One, misdiagnosed cardiogenic shock, two, sick myocarditis, and three, septic shock with baseline cardiac dysfunction. Now, let's get to true sepsis-induced cardiomyopathy. Septic shock with new onset or worsening cardiac dysfunction.
SPEAKER_00Now, why? Why would this be the case? You know, it's not completely obvious why having septic shock of any stripe would cause you to go into heart failure. I mean, myocarditis, that makes sense. You know, the virus is messing with the heart, but why would eurosepsis cause you to go into heart failure? What is going on here? Mechanistically, we used to think that it was maybe just an oxygen demand issue, and these patients were getting ischemia on a macrovascular level, but it turns out that's actually not one of the mechanisms, as far as we can tell.
SPEAKER_00There are three major categories of mechanisms that I like to think about when I'm trying to understand how sepsis is causing acute cardiomyopathy. The first one is direct myocardial depression caused by cytokines, caused by the oxidative stress you get with septic shock and that excess nitric oxide. And so both those things, cytokines directly and oxidative stress, both can act as myocardial depressants.
SPEAKER_00Two, microcirculatory dysfunction. So in your myocardial tissue, just as in all of your other tissues in sepsis, you have leaky vessels. You get an endothelial leak. And what that does is causes myocardial tissue edema. And this edema causes you two problems. The first one is that it limits oxygen diffusion. The second is that it stiffens the ventricle, which, as we'll talk in a minute, is one of the subphenotypes of sepsis-induced cardiomyopathy because it functionally gives you diastolic dysfunction.
SPEAKER_00And then the third, the third mechanism is decreased sensitivity to the things that normally stimulate cardiac contractility. So both calcium and beta adrenergic stimulation are through various mechanisms that I'm not going to get into right now, partially because I only partially understand them. But the bottom line is that things that normally stimulate inotropic responses in the heart, calcium, beta adrenergic stimulation, your heart is less responsible to those things.
SPEAKER_00So, mechanisms that we know of. It's multifactorial, but one, direct myocardial depression, two, microcirculatory dysfunction, and three, decreased sensitivity to endogenous inotropes. Those are the things that appear to give you this pathophysiology of sepsis-induced cardiomyopathy. Now, the next big question here is whether cardiac dysfunction in septic shock is actually just a marker of illness severity.
SPEAKER_00Because the sicker you are in septic shock, the more things start to fail. You know, your kidneys are failing. You start getting ARDS, and now your heart is failing. And so is it purely a marker of illness severity? Um, it's certainly associated with increased mortality, but is there a causal relationship here or just a correlational relationship here? And it's not a hundred percent clear. That being said, for me, this is an example of when your shock precipitant and your shock physiology are not necessarily the same thing, because in different patients, the same inciting event can be associated with different shock phenotypes.
SPEAKER_00The right question I think to ask here are: what are the primary driver or drivers of my patient shock physiology right now? And so even though I don't for sure know the degree to which sepsis-induced cardiomyopathy is just a marker of illness severity or not, I like to hedge my bets and treat it as a specific septic shock phenotype. And there is some data that this is a reasonable thing to do. So, you know, what we know from multiple studies is that if you nonspecifically throw some inotropes at all comers who happen to be septic, it is not helpful.
SPEAKER_00Giving dobutamine or Epi to every single septic patient who walks in the door seems to have no effect on mortality whatsoever. However, if you look at a subset of patients that you are worried have sepsis-induced cardiomyopathy, that you do actually have a problem with LV inotropy, there was a study that did exactly that. Came out in the American Journal of Respiratory Critical Care Medicine in 2025. And what they did is they specifically looked for patients with an LVOT VTI of under 12, which suggests that the LV is having trouble pushing against the afterload that it's seeing in those patients, in those patients using dobutamine, actually did help them.
SPEAKER_00And I don't think this is surprising, but it does argue that we should be viewing and probably treating patients with sepsis-induced cardiomyopathy as a phenotype of septic shock. And so, really, the way that I like to think about it is that these patients are in a complex mixed shock state, especially because sometimes it truly can be hard to tell the difference acutely between a patient who is in cardiogenic shock with superimposed sepsis, septic shock with superimposed cardiomyopathy, or if I don't have prior records on somebody, do they have new onset cardiomyopathy?
SPEAKER_00Does it always look like that? I don't know, which is why I just think about it as complex mixed shock state somewhere on the spectrum of cardiogenic shock complicated by sepsis on one end and septic shock complicated by cardiac dysfunction at the other. Now, when we're talking about those patients on the spectrum of septic shock complicated by cardiac dysfunction, it turns out that we actually have three defined subphenotypes of the kind of cardiac dysfunction we're looking at.
SPEAKER_00And all phenotypes are associated, all of these three are in fact associated with increased mortality. And so I think this is significant. The first phenotype is LV systolic dysfunction. And depending on the study you read, the prevalence is all over the place. Uh, you know, a review article on this that came out in 2025 cites anywhere from 5 to 32%, depending on the study. But that's the first one. LVsystolic dysfunction. And I think that's the one that we all usually think about.
SPEAKER_00But there's two more. The second one is LV diastolic dysfunction, and in fact, that appears to be more common. They're citing 20 to 57% of patients with septic shops show up with LV diastolic dysfunction. Again, big range of numbers, but significant on either end of that range. And then lastly, and most interestingly to me at least, is the patients with the third phenotype, which is right ventricular dysfunction. Now, this is a big range of patients. They cite 18 to 79%.
SPEAKER_00You know, uh, the study that I saw that was most convincing was somewhere in the 35% range, regardless, a significant number of patients. And each one of these phenotypes, LV systolic dysfunction, LV diastolic dysfunction, and RV dysfunction, are all associated with increased mortality. So what are you gonna do about this? Um I about to cite another paper that is about to give you the tried and true one-path algorithm that you are gonna use to treat sepsis-induced cardiomyopathy?
SPEAKER_00Sadly, no. Because if I had that paper, I would just save myself and you guys a lot of time and hand it out to you instead of making you all listen to this podcast. Um, what do we have in terms of the literature? Well, probably the best recent review that I've read on sepsis-induced cardiomyopathy is this one that came out in 2025 in Chess. And what they say about management is as follows: Management requires phenotype-guided, individualized approaches rather than standard resuscitation strategies.
SPEAKER_00Now, I fully a hundred percent agree with that statement. I just don't find it all that practically useful. I mean, what exactly do we mean by that? Because at two in the morning, when you have a shocky patient with sepsis-induced cardiomyopathy, and the patient's hypotensive, and the nurse is like, yeah, what would you like to do about this hypotension? And you're like, I would like to take a phenotype-guided, individualized approach. They're like, Cool, but what pressure do you want me to start?
SPEAKER_00And fair enough. So, like with anything else, we need to be specific and systematic and have a mental model of the strategy we're going to use in these patients. When I'm thinking about my mental model, as usual, I think we need to be physiology-based here, especially with something that is a complicated mix shock state. So, my physiology guide, the three categories of things that I think about in terms of both deciphering their physiology and then deciding how to manage that physiology, are not mysterious, they are afterload, preload, and inotrophy in that order.
SPEAKER_00Let's start with afterload. We have to remember we're talking about two different afterloads. We're talking about both the left ventricular afterload and the right ventricular afterload. But let's start with the left ventricular afterload, because that's the one that we're all very comfortable with. And so, left ventricular afterload, that is your systemic vascular resistance, that's your arterial tone. And the important thing here is that the left ventricle is putting out a bunch of cardiac output, in fact, preferably five liters per minute, something along those lines.
SPEAKER_00But the left heart can put out as much volume as it wants to, but that volume is only usable once a week. It's pressurized. And if you don't pressurize that LV output, it is totally useless to your body. I mean, this is why we talk about blood pressure, not blood volume. You've got to pressurize it. Pressurizing the LV cardiac output becomes much more difficult when you lose some of that arterial tone, when you have arterial vasodilation.
SPEAKER_00And so your left heart has to put out more cardiac output to get the same bang for its buck when your arterial system is not doing a great job of pressurizing that volume. And so that's one of the reasons why this is one of the very first things I do when I see one of these patients is I try and pressurize that volume because that is going to help make the best use of the cardiac output that my LV is giving me. Sepsis causes peripheral vasodilation, right?
SPEAKER_00We all know that. But an interesting thing that actually becomes kind of important is that's not what happens in the pulmonary circulation. So the right heart is seeing something very different than the left heart is in septic shock. Here's the deal. In the pulmonary circulation, inflammation does not actually cause vasodilation. In the pulmonary circulation, inflammation actually causes pulmonary vasoconstriction. And so these patients who are vasodiled systemically are vasoconstricted in terms of what the right heart is seeing.
SPEAKER_00And they also have other reasons, often, that they're vasoconstricted in the pulmonary circulation. One is we know that sepsis can cause this widespread microvascular thrombosis, and microthrombi within the pulmonary circulation lead again to increased pulmonary pressures. When the lungs get sick and get inflamed, they soak up fluid like a sponge, and that, that pulmonary edema, can create additional resistance to flow, additional increased pulmonary vascular resistance.
SPEAKER_00And then if you top that off with maybe a patient who has pneumonia as the source of their sepsis, and now maybe they're hypoxemic and you have hypoxemia-induced vasoconstriction, well, now we have an RV who's really, really hurting here because RV is very dependent on afterload in terms of what it can put out, much more so than the left ventricular is. The right ventricle is afterload dependent, and increased afterload can very, very significantly drop the right-sided cardiac output.
SPEAKER_00And don't forget that the right ventricle also has all the same problems the left ventricle does in terms of depressed myocardial contractility. Because it's also seeing the cytokine storm. It is also seeing that myocardial edema, and it is also seeing that it's less responsive to beta adrenergics and calcium. So your right ventricle is really taking a second hit here. That the left ventricle, it's like, well, yeah, I'm taking a hit. I'm not squeezing as well, but I don't have to squeeze against as much afterload.
SPEAKER_00So there's that. The right ventricle's like, what? Wait a minute here. This is so not cool. On the one hand, I'm having cardiomyopathy, I'm having trouble squeezing. On the other hand, at the same time, you're going to increase my afterload. So not cool. And this is probably why we see RV dysfunction in so many septic patients and why it's associated with mortality. Okay, that is thing number one. Afterload, what it does to your systemic versus pulmonary circulation, and what the right versus the left heart has to push against.
SPEAKER_00Now let's talk about preload. Preload is one of those concepts in physiology that I think people see as a very easy thing, but it's actually one of the most complicated things to think about and to assess. When I'm thinking about preload, there's two different components of the preload. The first one is you gotta get the preload to the left ventricle. The second component is what is the left ventricle then going to do with that preload?
SPEAKER_00So let's start with we gotta get that preload, we gotta get that volume to the left ventricle in the first place. How are we gonna do that? Well, the first step is venous return. You gotta dump some volume into that right ventricle. Now, again, remember, volume in the circulatory system doesn't count. It doesn't do anything for you unless that volume is pressurized. And that is just as true on the venous side of the circulation as the arterial side. And so what happens when you're vasodilated?
SPEAKER_00And you're vasodilating your veins as well as your arteries, and what we are now creating is this unstressed volume sink, where we haven't added or taken any volume away from the system. But what we have done by veno dilation is made a lot of that volume unpressurized and therefore unusable. So when we think about venous return to the heart, we don't talk about central venous volume, we talk about central venous pressure, because it's not just the amount of volume, it's the percentage of that volume that is pressurized, which we call the stressed volume.
SPEAKER_00And so this is another reason that one of the very first things I do will be to give a vasoconstrictor because I want to use, I want to optimize and recruit as much of the volume I already have. I want to get more bang for my buck, particularly because in patients with sepsis-induced cardiomyopathy, this is a state that I often need to be very, very thoughtful about exactly how much volume I'm giving, especially if they start having renal dysfunction. And so I first want to make sure I'm making the best use of the volume I already have.
SPEAKER_00And it's hard to do that if I'm just pouring volume into this big sink. That is thing number one. We gotta get that preload to the right side of the heart. Great. Okay, cool. We've gotten the preload to the right side of the heart. But do not forget, the right side of the heart has to now hand that volume to the left side of the heart. And as we just talked about, if your right ventricle is struggling, even if your left ventricle is doing fine, but maybe the patient has pneumonia and ARDS and is very inflamed, and their RV pressures, well, they're actually really quite high, and the RV is suffering.
SPEAKER_00The problem may be: do we need to give that patient a little right ventricular support so we can actually get that stressed volume all the way over to the left ventricle? So, step one, getting that preload to the left ventricle. Okay, we've gotten the preload to the left ventricle. Now, what is the left ventricle going to actually do with that preload? There are two physiologic principles here that we need to think about in this context. And the first one is the starling curve. We are all familiar with it.
SPEAKER_00The idea that when a patient is on the steep part of the starling curve, they will increase their stroke volume if you add more preload. The second part is, of course, the flat part, where once the patient has arrived at the flat part of the Starling curve, they will no longer increase their stroke volume if you give them more preload. Now, one of the most important things for our purposes here is that you cannot conflate the slope of the Starling curve with the patient's position on the Starling curve.
SPEAKER_00Because patients who have baseline, LV, systolic dysfunction, the slope of the Starling curve is more flat. But they still have a steep part and a flat part of their Starling curve. And so I think I often hear this thing where somebody will echo a patient and be like, oh, no, no, no, I can't give them fluids because their EF looks poor. Well, their EF looking poor may tell you something about the slope of their Starling curve, but it doesn't necessarily tell you anything about their position on the Starling curve.
SPEAKER_00And to figure out that kind of thing, what you actually need to start doing to figure out where they are, are they on the steep part, are they on the flat part, just one view of the LV doesn't do it. This is when you need to start doing things like stroke volume variation, passive leg raise. Those kind of things are even a fluid challenge. So just keep in mind, you can't conflate the slope of the starling curve with the position of the starling curve. And your patient coming in with baseline heart failure, who also is profoundly septic and maybe has been vomiting for three days and is profoundly hypovolemic, they certainly could be on the steep part of their starling curve, even though their LVEF doesn't look great.
SPEAKER_00Conversely, a patient with a great LVEF with a very steep starling curve could in fact be on the flat portion of their starling curve, so you can't conflate those two things. You cannot extrapolate how much fluid a patient needs just from the appearance of their LV on your initial echo, particularly because if their afterload is low, if they're vasodilated, then your LV often looks artificially good, not because it's strong, but because it's not having to push against anything.
SPEAKER_00Now, the second piece of physiology, in terms of figuring out what is the LV going to do with the preload it's being handed, how is it going to make use of it, has to do with LV diastolic dysfunction. Because patients with LV diastolic dysfunction, if you look at the pressure volume curve of how much volume creates, what pressure in the heart of a patient with LV diastolic dysfunction, their pressure volume curve is very, very steep.
SPEAKER_00What does that mean? What that means is a little bit goes a long way in both directions. So if you give them just a little bit of volume, that small amount of volume results in a disproportionately large increase in pressure. On the other hand, if you're just a little bit volume down, you take away just a little bit of volume, now that also has a disproportionate effect, where now the pressure goes disproportionately down. These patients can be so difficult to manage in terms of fluid.
SPEAKER_00And when I hear somebody say, oh yeah, yeah, this patient's blood pressure, it's just super labile, I start thinking, could this patient have LV diastolic dysfunction? Because they can be very frustrating. Because you're like, oh, I gave you 250cc, your blood pressure was amazing. Or when you're diuresing them, you're like, I just gave you 20 of herosapide and now your blood pressure's tanked. Now, what this means in the context of a patient who's septic, who's vasodilated, is that these patients often need more fluid because they just need higher filling pressures.
SPEAKER_00They need more volume into that LV to get the same amount of pressure as you would with a normal floppy compliant ventricle. And don't forget that, you know, there are a lot of patients who have baseline diastolic dysfunctions. It's common in elderly patients. Maybe they have some underlying thing like aortic stenosis or just baseline diastolic heart failure, but you can also have diastolic dysfunction as a phenotype of sepsis-induced cardiomyopathy, with all that edematous stiff myocardium.
SPEAKER_00So those patients may need a little bit of extra volume to give them those higher filling pressures that they want, but don't forget it cuts both ways, because also they have a very narrow therapeutic window. And if you give them just a little too much, the next thing you know, off we go into pulmonary edema world. So that's the second thing. The physiology of preload is both getting the preload to the LV, thinking about our stressed volume, and the fact that the RV has to do something with that volume, hand it to the LV.
SPEAKER_00And then in terms of what the LV is going to do with the preload, we can't conflate the slope of the Starling curve with the position of the Starling curve. And we keep in mind that patients with LV diazolic function have that very steep pressure volume curve, where a little bit of volume gives you a big change of pressure in both directions. Now let's talk about our third big physiology principle, and that is ionotropy. Now, there's a reason that I talk about this last.
SPEAKER_00Because whether and how much ionotropic support you need to give your patient can be very difficult to figure out until you've dealt with both the afterload and preload issues. Because remember, your left heart might look really artificially good until you've fixed the afterload. And so this happens all the time where we start out, we do an initial echo, we're like, yep, left heart banging away. We start a little nor epi, we give a little volume, we come back and we're like, eh, we broke it. Um you didn't, but all you did by doing that was just reveal either baseline or new cardiac LV dysfunction.
SPEAKER_00And so this is why I just I want to do those other two things first. Once I have fixed my LV afterload and my preload, then now I'm like, okay, now I can hurt assessing my heart. We'll get to what we're gonna do about the RV afterload in just a minute. Okay, so I've started some Norepi, I've given some volume, and now I'm ultrasounding and I'm like, okay, we're not looking all that great here. Um, what am I going to do with this?
SPEAKER_00Well, this one is actually the simplest one. Because if I've corrected my afterload, my preload, and my left ventricle doesn't look so good, the squeeze is getting worse, or the squeeze is still bad, now I'm like, if my patient still looks shocky. And that doesn't just mean if they're hypotensive. Yes, if they're still hypotensive, but also how's their cap refill? Is their lactate starting to look better? Is their base excess starting to look better? Are they perking up?
SPEAKER_00Are they looking less kind of pale? Are they still diaphoretic? How is the patient looking? And so if after correcting, after load and preload, I'm like, okay, I feel pretty good about this. My patient's looking less shocky, they're looking pretty good, I'm like, great. Um, if not, that's when I start some inotropy. And one of the things you'll often see as you get to this transition point that you need to start thinking about inotropy is initially in these septic patients, they frequently have a very wide pulse pressure. Their diastolic pressure will be very low.
SPEAKER_00And then as you fix afterload and preload and reveal this underlying cardiomyopathy, its underlying problem with contractility, you'll see that pulse pressure start to narrow. And so your systolic pressure won't necessarily go up, but your diastolic will. And so maybe you started out with a diastolic of 40 and now your diastolic is 70, but your systolic is still maybe 90. And now that is also suggesting to me that okay, I fix these things, but I need to give my patient some contractility support, some inotropy.
SPEAKER_00How am I gonna do that? You know, a lot of studies of this, um, like dobutamine, I don't totally see why. That is just not my first inotrope of choice here. Because, as I just said, I just finished fixing their SVR. And so what I don't want to do is now go back after all that nice fixing I did and break it again. And I don't really want an inodilator here. You know, if I'm going to an inodilator, it either means I don't think that vasodilation and septic shock is really the primary thing driving their shock physiology anymore, or it means that I just want to aggressively balance it out with a vasoconstrictor, and I think I've bought myself room, I guess.
SPEAKER_00But what I'll do instead that makes a lot more sense to me is just start an inoconstrictor. And to me, that means epinephrine. Now, here's the deal with this, and here is how to use and to not use epinephrine in this situation, because a lot of people are like, ah, but epi will increase your lactate, and whatever shall we do? Well, the first thing is that we should not be using lactate by itself to decide whether we're making the patient better in the first place. Um, the second thing is that it's not clear that lactate elevation intrinsically through a type B lactic acidosis is actually bad or doing you any harm whatsoever.
SPEAKER_00And three, you know, if I'm titrating my epi drip and I'm at one of epi or 0.5 of epi, yeah, okay, maybe we maybe now have something significant. It's maybe muddling my picture a little bit, but that's not what I'm talking about. Because here, I'm not proposing that we'd use epi as a vasopressor. I want to use it as an inotrope. So when I write this order, I write it as a fixed dose. I'll start 0.04, 0.05, 0.06 of epi. I throw it on there and I leave it because I'm not trying to use it as a titratable vasopressor.
SPEAKER_00I'm trying to use it as a stable inotrope. So that's my inotrope of choice. Now, the other reason I like Epi is because the right ventricle likes Epi as well. And the right ventricle, remember, is suffering more than the left ventricle because it has all the myodepressant things that you have going on with the left ventricle, but you're also pumping against increased afterload. And so you often do need to give that poor right ventricle a little support, and more so than the left ventricle, because you do have that other problem of increased afterload.
SPEAKER_00Now, if I'm really, really in trouble with the right ventricle, and I have started some Epi, and I feel pretty good about that, and my LV inotropy, I think that that's looking better. My RV is still struggling, it's still limping along. Well, maybe I need to do something about that pulmonary vasoconstriction. And I particularly consider doing this in a patient who has ARDS, who has pneumonia. And this is when I think about an inhaled pulmonary vasodilator. Because the nice thing about those guys is they specifically vasodilate the pulmonary circulation, not the systemic circulation.
SPEAKER_00Because for the same reason, I wouldn't want to give dobutamine or milrenone. I mean, the pulmonary circulation would vasodilate and like those things, but dobutamine or iV milrenone will also cause my systemic vascular resistance to go down, which I don't want because remember I just fixed it. So what I can do is have this setup where I give IV vasoconstrictors and inhaled pulmonary dilators so I can do what I want. Vasoconstrict my systemic circulation, vasodilate my pulmonary circulation if my RV is hurting in a patient who has other things like hypoxemia pneumonia, and it's especially looking like it's hurting after I've already started an inotrope.
SPEAKER_00So those are three physiologic principles. Afterload, preload, and inotropy. Now, how do I go through the process of sifting through this? This is not gonna be an extensive podcast on use of ultrasound to assess these patients, because one, there are many more better qualified people than me to talk about that, and two, that would add another hour to the podcast, and you guys are probably already getting bored and want it to be over. So, real fast.
SPEAKER_00The key things that I do on ultrasound, I mean, if you're not using ultrasound, you're kind of flying blind. If you happen to have a swan, fantastic, but I'm assuming most of you don't. So, what are the key ultrasound studies I'm doing? Well, I like everybody else almost certainly start with a visually estimated LVEF. Um, because that's great, but you gotta keep in mind that as we said before, you gotta reassess it after vasopressor initiation. And if it looks really bad prior to vasopressor initiation, that's one of the things that makes me start questioning: is this really septic shock to begin with?
SPEAKER_00Especially if I see a bad-looking LV and a dilated IVC on first pass, I'm like, wait, let's just re-examine whether this is even septic shock in the first place. But if I'm seeing like a, okay, either baseline bad or contractile LV, then I start vasopressors, I gotta reassess the LV once I've normalized the afterload. Now, the thing that is really, I think, often very helpful in these complex mixed shock patients, and especially in sepsis-induced cardiomyopathy, to help you figure out what the primary thing driving the physiology is, is an LVOT VTI.
SPEAKER_00Because this is telling you how is the LV contractility coupled to the peripheral vascular resistance? Are these things well matched? Are they not well matched? Now, a normal VTI is 18 to 22. And if I have a normal VTI, that suggests to me that, okay, my contractility of the LV is handling an appropriate SVR. What if I have a high VTI?
SPEAKER_00Well, if I have a high LVOT VTI greater than 22-ish, what that suggests to me is I'm still having a lot of peripheral vasodilation. I'm still having a lot of vasoplegia, and maybe my main problem remains that my peripheral vascular system cannot pressurize the cardiac output that my LV is putting out. So if I have a high LVOT VTI, I'm thinking, do I need more vasoconstrictor? Or let's say I had started dobutamine and I'm like, oh, maybe I should take that off.
SPEAKER_00That's with a high LVOT VTI. What if I have a low LVOT VTI, less than 18 ish? That now suggests to me. That I maybe have enough vasoconstriction, or maybe I don't have enough, but the primary thing driving my shock right now is my low cardiac output. That whatever kind of SVR I have, whatever kind of afterload is present at the moment, my LV is not actually able to effectively push against it.
SPEAKER_00And that is especially when I want to start an inotrope. And the study I referred to earlier on starting inotropes in particular phenotypes, you'll remember, started dobutamine selectively in patients with a low LVOT VTI because that is very suggestive of me that they may in fact need some inotropic support. So those are the ways that I look at the LV, visually estimated LVF, take it with a grain of salt, pre-imposed norepi or whatever presser you want to use, and then an LVOVTTI. Right heart.
SPEAKER_00I am looking for RV dysfunction. But as we often talk about, you can't really eyeball RV systolic function in the same way you can eyeball LV systolic function. Now, if it obviously looks bad, if it like the RV is blown up and clearly not squeezing, one, you should probably re-examine whether this is septic shock in the first place. But, you know, if you're like, oh, I don't know, I can't quite tell about systolic function, but ooh, am I seeing septal flattening? And especially if I'm seeing a dilated IVC, I'm like, eh, my RV might not be loving this.
SPEAKER_00I might need to do something about this. And conversely, you know, if I have a totally okay IVC, my IVC is small, my IVC is collapsible, I feel better that at least the RV is able to handle the pressure that it's given. It's able to pass it on. It's not congested, it's not lacking up. Now, of course, there's many caveats with IVC, and I would say the major one in patients with septic shock, especially if they have a pulmonary source, is they're breathing really, really hard. And it is very difficult, both technically and in terms of false positives, to assess an IVC in a patient who's breathing in these deep breaths and they're cusmalling at 35 a minute, that is very difficult to do.
SPEAKER_00If I see, though, a dilated IVC in the presence of septal flattening, particularly, I'm like, okay, my RV is really not loving this. And if I'm at the point that I've already started an anotrobe, I'm like, ooh, now is it the time to start an inhaled pulmonary vasodilator? And especially in the situation where I see a dilated IVC, but my LV is really hyperdynamic. That's another situation that I start to wonder: is my RV hurting? Either because I got the cause of shock wrong in the first place, or because my RV afterload in septic shock is becoming a problem.
SPEAKER_00So those are the basic things I do on my echo to try and parse this out. Putting it all together in terms of my management strategy in a patient that I think has sepsis-induced cardiomyopathy. This is how I approach it. One, I start early, early norepi, both because it saves me from giving them too much volume they don't need by using the volume optimally that they already have. And two, in order to truly assess what's going on with their systolic function, I need to first fix my systemic vasodilation.
SPEAKER_00So I will start very early norepi, certainly before I've given them 30 cc's per kilo of fluid. I'm not going to get into how that's not a magic number and you should be ignoring it regardless, but nevertheless, early nore epi. Now, on the one hand, don't be afraid of fluids in these patients. You know, their physiology first started as septic shock. And especially in patients with LV diastolic dysfunction, they actually may need more fluid. They may need higher filling pressures than your average patient, but this is not the time in any kind of sepsis-induced cardiomyopathy that I would want to give big aggressive fluid boluses all at once.
SPEAKER_00I want to give a small bolus, reassess. And that reassessment may be clinically. It may be echo. Usually I do a clinical reassessment, and then also less frequently, but still serially echo reassessment. So don't be afraid of fluids in patients who don't necessarily have the greatest looking function, but be thoughtful about it, know why or why not you're doing it, and give it in small aliquots rather than large boluses. If the LV and or RV systolic function is still looking bad, it's still looking depressed, after you've corrected your afterload and your preload, now it's time to add some inotropic support.
SPEAKER_00And personally, I like fixed dose, low dose epi for that. You can try something else if you want to. I would avoid Milrenone just because it's long acting, you'll be stuck with it. Um, you can try dobutamine if you really like dobutamine, but I personally don't feel like I need to give them any vasodilation if I truly believe this is sepsis-induced cardiomyopathy. And then last, if you believe you've corrected afterload, preload, and you've started some ionotropic support, and your RV still looks pretty unhappy, well, I do a couple things.
SPEAKER_00One, I now think about should I transition to vasopressin? Are these escalating nore epidoses further causing increases in my pulmonary vascular resistance? Are they causing increases in afterload? And can I use vasopressin? Because the lovely thing about vasopressin, it gives you systemic vasoconstriction without pulmonary vasoconstriction. And that's the first thing I'll do. And then the second thing I'll do is, yeah, my RV looks like it's hurting either. The contractility is bad, the IVC is dilated, maybe my septum's not looking so hot, the RV's big.
SPEAKER_00I'm like, okay, do I need to do an inhaled pulmonary vasodilator? And I have a especially low threshold to do that if I have pneumonia, if I have ARDS to help my RV. Alright, so that's sepsis-induced cardiomyopathy. You're gonna start by saying this is a diagnosis of exclusion. I need to make sure I've thought about misdiagnosed cardiogenic shock, sick myocarditis, and septic shock in a patient whose cardiac dysfunction is at their baseline.
SPEAKER_00I then am I gonna think about my physiology principles and manage these patients in terms of trying to parse art their physiology with preload, afterload, and inotropy. And then finally, I'm gonna translate that into my management principles of starting early norepi. Don't be afraid of fluids, but be careful with fluids in small aliquots, making sure that once you fix preload and afterload, you help them out with some inotropy. And then, if your RV really needs additional support, think about switching to vaso plus minus adding an inhaled pulmonary vasodilator.
What are the effect (good and bad) of positive pressure ventilation on hemodynamics in critically ill patients? Link to the visuals for this episode: https://bit.ly/venthemodynamics. Also, checkout Rapid Sequence, our...
How do you manage the vent in a critical polytrauma patient who is hypoxemic? Link to the visuals for this episode: https://bit.ly/traumavents. Also, checkout Rapid Sequence, our new choose-your-own-adventure...
When, how, and most importantly WHY we should be using diuretics. Also, checkout Rapid Sequence, our new choose-your-own-adventure resuscitation game at rapidsequence.org (coupon code for ICUedu listeners: ICUEDU2026)
Acute medical management of valvulopathy. Part II talks about the pathophysiology and management of stenotic valves: AS and MS. Check out Rapid Sequence, ICUedu's new education project at rapidsequence.org Great Core...
Acute medical management of valvulopathy. Part I talks about the pathophysiology and management of regurgitant valves: AI, MR, and TR. Great Core Ultrasound post on POCUS assessment of valvulopathy:...