SCCM Pod-441 Connections Between COVID-19, Sepsis, and Fluid Management

---

Learn about the importance of utilizing dynamic assessments of fluid responsiveness to guide treatment in patients with viral sepsis, including COVID-19, and understand how they can be used to help improve patient outcomes in sepsis patients. Host Pamela M. Peeke, MD, MPH, FACP, FACSM, is joined by Patrick Troy, MD, to discuss how only 50% of hemodynamically unstable patients are fluid responsive and that the same treatment paradigm in fluid management applies to both viral sepsis and COVID-19 patients (Douglas IS, et al. Chest. 2020;158:1431-1445; Latham H, et al. J Crit Care. 2017;42:42-46). This podcast is sponsored by Baxter.

Patrick Troy M.D., is the Division Director Pulmonary Critical Care and Sleep Medicine Physician at Hartford Hospital, Assistant Professor of Clinical Medicine at the University of Connecticut School of Medicine.

*If you are unable to play the podcast please click here to download the file.

Transcript:

This podcast is sponsored by Baxter Healthcare. Baxter's Starling Fluid Management Monitoring System provides continuous, accurate, non-invasive, and personalized hemodynamic monitoring to support individualized fluid management. The Starling system empowers you to make patient-specific fluid management decisions wherever your patient is in the hospital, so that you can deliver the right therapy to your patient every time. Visit starling.baxter.com to learn more. 

 

Dr Pam Peeke:  Hello and welcome to the Society of Critical Care Medicine's iCritical Care podcast. I'm your host, Dr. Pam Peeke. Today's topic is connection, the connection between COVID-19 sepsis and fluid management. I'm joined by Dr. Patrick Troy, who is the Division Director of Pulmonary Critical Care and Sleep Medicine at Hartford Hospital. He's also Assistant Professor of Clinical Medicine at the University of Connecticut School of Medicine. Welcome, Dr. Troy. 
 
Dr Patrick Troy: Thank you, Pam. I'm very pleased to be here. 
 
Dr Pam Peeke: Awesome. Before we begin, do you have any disclosures you wish to report? 
 
Dr Patrick Troy: I do. So I would disclose that I am a paid consultant for Baxter. And they have a Starling device which measures stroke volume, so I think that's relevant to today's conversation. 
 
Dr Pam Peeke: Excellent. And thank you so much. Now, regarding our topic, which is the connection between COVID-19 sepsis and fluid management, I want to run through the learning objectives so everyone understands what we're going to learn here. The main learning objective is to understand the importance of utilizing dynamic assessments of fluid responsiveness in patients with viral sepsis, including COVID-19. 
 
Now, why is his podcast needed? It's needed because learners out there need to be aware that the same treatment paradigm in fluid management applies to both viral sepsis and COVID-19 patients. 
 
Now, there are knowledge gaps that we'll also be addressing. The first is that learners may be unaware that only 50% of hemodynamically unstable patients are fluid responsive. Number two, learners may be unaware that utilizing a dynamic assessment of fluid responsiveness to guide treatment can improve patient outcomes in sepsis patients. 
 
All right. Now, as we begin, Dr. Troy, can you please tell us a little bit about your background and your role, your focus and your main specialty? 
 
Dr Patrick Troy: Absolutely. Happy to. So I am a pulmonary critical care physician, also a sleep medicine physician by training. I went to the University of Connecticut for medical school and then down to Hopkins for my internal medicine residency. And then I did the Harvard Combined Pulmonary Critical Care Fellowship, and then a Sleep Medicine Fellowship also at Harvard. 
 
And then, after all that traveling, came right back to my roots at the University of Connecticut, and worked at their main teaching hospital, which is Hartford Hospital, where I've been since 2011. And my main focus is sort of a balance between outpatient sort of bread and butter pulmonary and sleep medicine. And then about half of the time or a little bit less, well, maybe more than half of the time these days, unfortunately, is critical care, working in a large ICU in a 900-bed hospital. We were the, right after New York at least, ground zero for the initial COVID outbreak, which seems like 10 years ago, but it was only about 18 months ago in March of 2020, and then saw all the rise and fall of the various variants and the joys of the vaccine. And now, we find ourself smack dab in the middle of our good friend, the Delta variant. 
 
Dr Pam Peeke: Oh, my God. All I can wait for now is Lambda to show up, but, you know, we're just going to hold our breath on that one. All right. So let's hop right to it. What is the relationship between COVID-19 and sepsis? 
 
Dr Patrick Troy: So I think it's all the same at the end of the day, Pam. So whether it's COVID or a bacterial pneumonia, you really want to take a step back and think about with COVID or even a bacterial pneumonia, when it comes to sepsis, fluid management is essential. And to understand that, what I would tell you, and that's not just my opinion, by the way, that is CMS guidelines. They have the sepsis CMS core measurements, these SEP-1 highlights, which now dictate how we manage sepsis. And it's all about understanding the history and the history is really, if we turn back the clock to when I was a resident at Hopkins when the Surviving Sepsis Campaign guidelines first came out sort of circa 2004, there was this huge push, right? Huge push on volume resuscitation. Don't leave a patient in the corner of the emergency room without giving them volume. And then those days I remember furiously targeting the CVP goal of 8 to 12 or slightly higher if the patient was ventilated. 
 
And I always remember thinking why am I just doing this reflexively without any real sort of guidance? And what we were seeing even in those days was we would flood them with fluids and then, not surprisingly, a lot of them would end up in respiratory failure and then they'd end up intubated. Then we'd spend the next seven days trying to diurese them or worse with renal replacement therapy. And so we got them extubated and I always found it like deeply unsatisfying to be so rote in terms of just fluid, fluid, fluid, fluid, without really any kind of physiologic guidance from that. So I think what's changed over the last couple of years, and I think even CMS recognizes this now, is a sort of a migration, if you will, away from this idea of, "Hey, look. instead of following just rotely giving people IV fluids until they're foaming at the mouth, maybe there's a more intelligent way, a more physiologically principled way of doing that." 
 
And so I think that began sort of circa 2015 and beyond, and we can talk about that data, and has historically been exquisitely important for bacterial sepsis all comers. It's just as important in viral sepsis specifically COVID because these patients are already sort of struggling with high-grade hypoxic respiratory failure, and they behave very much like other septic patients. They can be hemodynamically labile. You've got to be really thoughtful about do they need fluids or not and not make any assumptions. And to the point you brought up as one of the teaching objectives, simply looking at these patients who are hemodynamically unstable, about half of them are not fluid responsive. So what are we doing giving a hundred percent of them fluid challenges? 
 
Dr Pam Peeke: So, what you're really talking about is customizing and instead of, as it were, doing everything rote, you're actually saying, well, why don't we have any metrics here? Where are the guidelines? Where are the numbers? Where's a tool that we can actually utilize? So take us to that place now. So you've been talking about this evolution over time of managing these patients, especially as it relates to fluid. Now, what are you doing now to be able to address that? 
 
Dr Patrick Troy: Right. Good question, Pam. So as I said, historically, there is really a catastrophic misunderstanding about that fluids because at the end of the day, what we care about is whether when you give a patient a fluid bolus, at the end of the day, are you going to increase their cardiac output or index, depending on how you want to look at it? And so the question then becomes, well, what's going to tell me that and the answer is historically central venous pressure. And I mean, I can even remember arguments about making measurements of pulmonary capillary wedge pressures with PA catheters. And even if you look at the CMS guidelines now, they still recommend the measurement of central venous pressure. 
 
The problem is that, and this is not my opinion, this is what the data supports, it's essentially flip of the coin if you're looking at the CVP as to whether that accurately reflects in any real sense what you care about, which is if I give a patient a fluid bolus will, not their blood pressure, but will their cardiac output/index increase? And the answer is with the central venous pressure, what we thought chasing this number is completely, essentially unrelated. It's flip of the coin, whether actually numbers mean anything. 
 
So that idea however has been so entrenched in the critical care literature for the better part of 20 years now, to be perfectly honest, right? That it's been very difficult for people to kind of migrate around that. And so the idea is really going back to physiology, which is really why most of us got into the game to begin with, right? It's to understand the Starling law. And to do that, you've got to basically think about, well, what does it say? And what it says is, well, not to go back to physics here, but Ohm's law, right? That basically V equals IR or your MAP equals your cardiac output times your SVR. And so what you're trying to really figure out is if the patient has stroke volume responsiveness, right? And we can talk about how to measure that in a second. If I can move that number, I'm moving someone up the Starling curve. Now, you might not see changes in the blood pressure because of what Ohm's law looks like if you were to draw it out in front of you, but you're doing the good work of increasing the cardiac index. So you're having a positive benefit. 
 
And so where we've moved to, and there's a lot of literature on this and it sort of began really with some of the seminal stuff back from 2017 and then the University of Kansas Group did a lot of this. And what they're talking about is non-invasive cardiac output monitoring. What used to be called Nicom, now it's called the Starling device. And so what this device really does at the end of the day, at the simplest level, is it measures a change in stroke volume to create an index, a Delta, if you will, right? And there's good literature, which suggests that a change in this of more than 10% is indicative of someone who is stroke volume responsive. 
 
What does that mean? Just think about the steep limit of the Starling curve, they're sitting there, right? Meaning essentially, their cardiac output or index is recruitable. So if they are that, then the goal is to utilize this tool to kind of move them up that curve when you're encountering them in that hypotensive state. And the initial studies were back in 2017. If anyone wants to look them up, they were in the Journal of Critical Care. And I can certainly send you the references for that. But what was impressive about them, again this is retrospective stuff, is that it was pretty clear that this data demonstrated more or less what you would think is that if you used a stroke volume-guided approach, your fluid balance was about 3.5 liters less than sort of standard care, that your ICU length of stay was significantly less, that there was less use of pressors, there was less mechanical ventilation. And then this last one is really interesting, there was less use of renal replacement therapy. So that got people thinking about migrating away from these static measurements that don't really relate to volume responsiveness and migrating into this more dynamic measurement using the Nicom or what we now call the Starling, and it's non-invasive. 
 
Dr Pam Peeke: I love it. This is fantastic. The reference you were talking about with stroke volume-guided resuscitation in severe sepsis and septic shock and how it improves outcomes. And the main author is Steven Simpson as a senior investigator and Heath Latham, L-A-T-H-A-M. And this was in the Journal of critical care as you noted. And their study basically demonstrated that stroke volume-guided fluid resuscitation of patients with severe sepsis and septic shock was associated with reduced fluid balance and improved secondary outcomes. So that's fantastic. And I really think everyone should be very well aware of that. I think that was a 2017 study. And you also cited another one more recently, 2020, from Critical Care, fluid response evaluation in sepsis hypotension and shock. And this was a randomized clinical trial. Tell us a little bit about that. 
 
Dr Patrick Troy: No, that's great, Pam. And thank you. I was going to take a breath and lead into that one, because this is the same essential group, the University of Kansas folks, because, look, everyone's going to knock a retrospective trial. I would too, quite frankly, because you're going to say, "Oh, well, there's inherent bias in there. Let me see what you got with a prospective randomized controlled trial." So here it is, right? So basically this trial and, let's be honest, this one got lost. It was published in CHEST, but it got lost because unfortunately I think it came out in April of 2020, not exactly a time that a lot of us were, you know, looking at the literature for this kind of stuff. We were up to our eyeballs in sepsis with COVID, and just trying to kind of figure out our way, but this study essentially looked at the same things, looking at a stroke volume index guided again with a Starling device, noninvasively, to assess in these septic patients, right? Can we replicate what we saw retrospectively in a prospective fashion? 
 
And again, the advantage here is we're reducing selection bias. We're going to be able to demonstrate causality, it's level one evidence versus retrospective, which is more level two evidence. So it's the highest level, right? And so they're looking in ER patients with refractory septic shock who've gotten less than three liters, and then they are going to use this device to sort of guide fluid resuscitation, and make key decisions on basically fluid versus pressors. 
 
Now, what they saw was essentially what you would expect. It's more or less a near replication of what was seen in the retrospective data, not completely, but fairly close. And so what they saw was as expected, there was a lot less in terms of fluid balance, there was dramatically decreased fluid balance. Then specifically, there was also decreased renal replacement therapy, which was the same kind of signal that pinged as well in the retrospective study. So, interesting stuff that kind of confirms, and I think from my perspective, one of the curious things was the renal replacement therapy because I kind of hinted at that when we first started talking. You know, back in 2004, when we were plugging away with liters and liters of normal saline targeting CVPs, not only did we intubate a lot of patients, but we ended up finding that a lot of them ended up going into renal failure. So it turns out that, everything in moderation, right? So like too little fluids, I get that and not good, right? That has end-organ damage. No one's going to argue that. But I don't think what's fully appreciated until recently over the last five years is too much fluid may be just as bad or even worse as too little fluid, which if I said that out loud five years ago, would have been really contentious point. If I'd gone to critical care, the annual meeting, I probably would have been laughed out of the room. Now, I think that people are recognizing that in fact, that actually is in fact a true statement. 
 
Dr Pam Peeke: Yeah, it's actually a thing. And actually, if you went to the meeting, they wouldn't laugh you out. Today, they just cancel you. You're not there anymore, and there you are. But your point is extremely well-taken. And again, it's so interesting looking at management of COVID-19 in the ICU, I think, this has really opened up the door to discussions about the need for individualizing treatment in general. And instead of using what has been in the past, sort of a, well, you know, it's a recipe. Here, follow this and we do that, without really having one metrics to follow that are unique to that patient, which is so terribly important. And that's why I think this podcast is so relevant because I think that COVID-19 and maybe you could tell me your thoughts has really pushed us to rethink a lot of how we've been managing viral sepsis. 
 
Dr Patrick Troy: There's no question about it. I think that in this population, when you think about COVID in particular, we are fairly limited in the therapeutic modalities that we can direct against COVID, right? So if you have severe COVID, we have a pretty standard playbook in the sense that you're going to get steroids, you're probably going to get remdesivir, you may or may not get an IL-6 inhibitor like tocilizumab, and you may or may not get convalescent plasma depending on kind of which way the wind is blowing that day. That's everybody. Now, and how you respond to that is entirely sort of the idiosyncratic to that particular patient and your comorbids and your age and how long you waited and what stage of the disease you're in. What I have found with COVID is once you check those boxes, you better make sure, you better make sure that you, as the critical care team, do not do anything that iatrogenically makes the patient worse. 
 
What do I mean by that? Well, I mean, you don't want to create a blood clot, so you better prophylax them because you know that this patient's at higher risk. You don't want to create a secondary superinfection. So you better be very careful with your vent management. You better be very thoughtful about whether you want to ventilate them and you really better be careful about not knocking out other organs or causing or worsening respiratory failure with too much fluid, right? Because if you do, this patient, which is already kind of borderline for their ability to get through this, you've kind of knocked something else out, because you know, with this Pam one of the beauties of COVID, if there is one, is they tend to be one organ, one organ only, right? They tend to be respiratory and that's it. Where we run into trouble is when something else goes. And then, you know, that's not too different than in some ways how I think about bacterial sepsis. We're okay when you're one organ down with bacterial pneumonia, but it's when the kidneys go or when a blood clot shows up or there is a GI bleed, that's when trouble comes. 
 
So the question is, what can we do to keep ourselves out of our own way at some level? And the answer is instead of just rotely following something or a guideline, which is quite frankly at this point is dated, let's try to, as you say Pamela, let's individualize the treatment to the patient. Okay, so they're hypotensive. All right. Before we say let's give them fluid, let's make an assessment, which would suggest they may or may not respond to fluids that are next step is the correct one, because to be honest with you, you could have a CVP of 2 and need pressors and a CVP of 20 and need three liters. That's the whole problem with the CVP. It's the flip of the coin, it's not accurate. It doesn't reflect what you care about. 
 
Dr Pam Peeke: Absolutely. No question about it. Well, yeah, I'm looking through, good grief, the knowledge gaps and the objective. And I really think you've nailed it in a big way. As we bring this to a close, what would you like to tell your peers out there and people who are learning about better fluid management as a closer? 
 
Dr Patrick Troy: So, what I would say is, first, look at the CMS core measurements, right? And take a hard look at those and think about, "Hmm. do I agree with this?" Because those measurements and a part of the problem with those measurements is it's in some senses how healthcare systems are reimbursed and/or penalized, right? I get that they're there, but even there, there's an emerging sense of we need to migrate over way from static measurements, right? And I want people to individualize and tailor their approach to the individual patient, which means make no assumptions, make no assumptions whatsoever about the patient and realize that you have tools now, which are available to help guide you in your initial assessment, but not just there, right? 
 
I think one of the big mistakes that I was guilty of making early on was "Okay, I'm going to measure this sort of stroke volume index with the Starling device and I'm going to make a measurement and then that's it." No. No, no, no, that's not it. That's just the beginning actually, because what you're going to do is sequentially and dynamically use this tool to help guide you throughout the patient's critical care course. And by doing that, I think what we end up with is a much more tailored approach, a much more physiologically sound approach, right? I mean, it's Ohm's law that we're utilizing here. And I think that the outcomes in terms of retrospective and prospective data speak for themselves, less renal replacement therapy, lower fluid balance, less respiratory failure, retrospectively less ICU length of stay. And I mean, it doesn't touch mortality, but I think that's one of the mistakes we make, Pam. We say, "Oh, you know, it's got to be mortality. It's mortality or bust." Really? You know, ICU length of stay doesn't matter? Vent days doesn't matter? Pressor day doesn't matter? Those things matter too, just as much maybe, perhaps as mortality does. So I think there's a better way things are emerging. And I would challenge the audience to sort of think about, is their practice more consistent with kind of the former, kind of traditional management or would they be willing to consider a more dynamic approach to things? 
 
Dr Pam Peeke: Absolutely spot on. I just love it. So everyone out there we've been talking to Dr. Patrick Troy. And specifically the focus was the connection between COVID-19, sepsis and fluid management. Dr. Troy, thank you so much for being on this podcast. 
 
Dr Patrick Troy: My absolute pleasure. Thanks for having me. 
 
Dr Pam Peeke: And this concludes another edition of the iCritical Care podcast. For the iCritical Care podcast, I'm Dr. Pam Peeke. 
 
This podcast is sponsored by Baxter Healthcare. Baxter's Starling Fluid Management Monitoring System provides continuous, accurate, non-invasive, and personalized hemodynamic monitoring to support individualized fluid management. The Starling system empowers you to make patient-specific fluid management decisions wherever your patient is in the hospital, so that you can deliver the right therapy to your patient every time. Visit starling.baxter.com to learn more.  
 
Pamela M. Peeke, MD, MPH, FACP, FACSM is a nationally-renowned physician, scientist, expert and thought leader in the field of medicine. Dr. Peeke is a Pew Foundation scholar in Nutrition and Metabolism, Assistant Professor of Medicine at the University of Maryland, holds dual master's degrees in Public Health and Policy, and is a fellow of both the American College of Physicians and the American College of Sports Medicine. 
 
Dr. Peeke has been named one of America's top physicians by the Consumer's Research Council of America. She is a regular in-studio medical commentator for the National Networks and an acclaimed TEDx presenter and national keynote speaker. Dr. Peake is a three-time New York Times bestselling author and is a science and health advisor for Apple. 
 
The iCritical Care Podcast is the copyrighted material of the Society of Critical Care Medicine and all rights are reserved. Statements of fact and opinion expressed in this podcast are those of authors and participants, and do not imply an opinion or endorsement on the part of the Society of Critical Care Medicine, its officers, volunteers, or members, or that of the podcast commercial supporter.

Disclaimer