SCCM Resource Library-Optimizing Fluid Stewardship in Your ICU

Video Transcription

Good afternoon, everybody. My name's Mike Pisa. Welcome to our session on optimizing fluid stewardship in the ICU. Appreciate you all coming. There'll be three talks this session.  I'm the session moderator, and I'll also be giving the first talk. We have a talk from Ty Kaiser from the University of Colorado  and Timothy Ellender from Indiana University as well. I'm just going to get started, get rolling. So this first talk is designed to just kind of give  an overview of what fluid stewardship is and some sense of how to start thinking about implementing it. I have no disclosures related to this topic.  I'll start by talking a bit about the current state of fluid therapy, how we ended up where we are, then some about how to conceptualize  fluid stewardship, how to start thinking about our fluid resuscitation in a different way, and then some clinical application of fluid stewardship in the ICU.  Raise of hands, who knows Rivers Early Goal-Directed Therapy? We all do, right? So this landmark study demonstrated improved outcomes  when you put patients in a bundled care associated with septic shock. Kind of hallmark to this approach is using pretty aggressive early fluid resuscitation in septic shock.  Folks enrolled in this trial received a lot of fluid and a lot of fluid early, so five liters over the first six hours of resuscitation,  approximately 13 liters over the next 72 hours of resuscitation. And this fluid administration was directed at achieving several specific goals, one of those being CVP and driving  essential venous pressure to a goal of 8 to 12. Currently, Surviving Sepsis Campaign also recommends early aggressive fluid resuscitation  with a prescribed 30 cc's per kilogram of volume resuscitation in the early stages of septic shock. And this is echoed by the Centers for Medicaid and Medicare  Services, who use the 30 cc's per kilogram as a performance benchmark for the treatment of septic shock and getting that fluid in early.  So we're in a spot where our thinking is directed at getting volume into a patient in septic shock, and there's strong physiologic need for that.  Shock states arise from one of two things, either loss of volume or the increase in size in the container holding that volume.  Very simplistic kind of way to represent a shock state. So our therapy, and it stands to reason that aiming our therapy at filling the tank,  whether that tank be bigger or that tank being emptied by loss of volume, should help with treating shock. And indeed, especially in the early stages, it does.  But there becomes a decision point where you need to decide when is enough, when you'd be given enough fluid. And I think our current guidelines guide us  in that direction by saying that once you've given 30 cc's per kilogram, and maybe even before that 30 cc's per kilogram is in, your patient's hypotensive.  You should probably initiate some vasopressors to start squeezing down the tank.  This kind of approach of administering volume to drive CVP kind of makes an assumption that your CVP is a surrogate for preload and the actual stretch on the heart, which should  increase cardiac output. But there's another conceptual way to look at this, and this is looking at CVP as a competitive force to venous return.  So the higher you drive CVP, the harder it is for fluid to get back from the periphery. Into the central circulation, because you've  created this pressure head preventing it from doing that. And in effect, you see a backup of fluid because you have this pressure head that the fluid can't push against.  This leads to all sorts of untoward effects on the body, and it affects multiple systems. And certainly, I'm not going to read through all of these.  This is a great graphic that gives a kind of remarkable look at how many of our systems would be affected with the overdosing of intervascular volume.  Leads to things like peripheral edema, bowel edema, intercapsulary hypertension in vital organs, which just contributes more to the known capillary leak that  occurs during shock states, and contributes to anaerobic metabolism occurring in these kind of almost compartment syndrome type physiologies in these encapsulated organisms.  So the overdose of fluid is bad.  This is a recently published article that demonstrates that well with some data. And it shows that if there's a 10% increase or a 10% volume  overload for patients, there's a dramatically increased chance of having some ongoing kidney injury, some major adverse kidney events, as they say.  And additionally, this study demonstrated an increase in mortality. So it's helpful to realize that shock is a dynamic state that changes over its course.  And our therapy in shock should change over the course.  Useful to look at things like this with some conceptual framework. And one conceptual framework to look at this is encapsulated by the acronym ROSE.  So in this model, there's four distinct phases of shock and four distinct phases of resuscitation. The first is a resuscitative phase where it's necessary to rapidly correct  hemodynamic compromise. So this is a phase that's marked by a first physiologic hit where the organism sees profound cardiovascular and hemodynamic effects.  And it's here that rapid fluid resuscitation probably has the biggest effect. But even in this early stage, you have to start thinking about adding something other  than volume to the system in the form of vasopressor. Next comes an optimization phase. And this is where you really have to start looking at some dynamic features of resuscitation  and dynamic features of your patient to help guide resuscitation. So protocolized care that directs resuscitation at static measures like CVP lead to those things associated  with the overdose of volume. So using metrics like the ongoing assessment of things like pulse pressure variability, responses to small amounts of fluid administration,  or things like straight leg raises give you some sense of where your patient lies on this continuum of needing fluid or not. The next phase is the stabilization phase.  This is where hypoperfusion has resolved. And our goal of fluid management at this stage is focused on maintaining an even to negative fluid balance.  So we try to restrict intravenous fluids because we've already given enough to kind of get the patient over the initial hump. Then follows the evacuation phase  where you start to pull fluid off, either with hemofiltration or diuretic therapy. You want to get the patient negative at this point to correct the volume you've administered,  now that they don't need it.  There are several different options we have for the administration of volume. These are the most commonly prescribed drugs in the intensive care unit are crystalloid fluids,  intravenous fluids. None of these are physiologic. They should all be thought of as drugs. They all have adverse side effects associated with them.  Often, we don't think about it enough when we're thinking about which fluids to choose. We really need to think about physiology and what's going on with our patient at a given time  as we're deciding which fluids to prescribe. The misuse of these fluids can lead to things like iatrogenic acidosis, iatrogenic electrolyte abnormalities.  It's very important to consider.  This leads us to the concept of stewardship, which we're very familiar with in the setting of prescribing antibiotics. It's kind of commonplace to have in place  a stewardship framework of thought surrounding antibiotics where we're thinking critically, very critically, about what drugs we're administering, what drugs we're initiating,  tailoring those drugs to specific microorganisms,  considering at the outstart the duration of those drugs, and then figuring out how to de-escalate our therapy with those drugs.  We should be doing the same thing with fluid therapy. So Milbrain et al recently proposed the four Ds of fluid therapy. It's a very similar model to that  used with antibiotic prescription where you need to critically consider your drug in this context of the physiology of the patient. So think about which fluids you're administering  and what's gonna benefit the patient best. At the onset of fluid administration, start thinking about how long you're gonna continue it  and what your endpoints will be for getting rid of it.  That brings into dosing. Critically examine the dose and use dynamic measures to determine your dosing of fluid. And then start thinking of ways to de-escalate your therapy.  So it's a quick talk to the basic conceptual framework.  And the probably primary thing to remember is that IV fluid administration isn't free. There's consequences. Though a bag of LR or a bag of saline can do harm.  And we need to think about that critically as we prescribe these medications.  Thank you. I'll bring up Ty Kiser now. We'll talk some about that de-escalation.  Excellent. Thanks, Mike. You set me up perfectly.  I'm gonna talk a little bit about managing fluids and the use of diuretics. And I have no disclosures related to this talk except for the fact that most of the data  that I'm gonna be using is large cohort observational data. And I'll be making some assumptions which we know our ICU patients are very heterogeneous.  So it's all gonna be very individualized. But ultimately what I wanna convince you guys when we get done is that there are differences in outcomes based upon fluid imbalance.  There is an optimal time. And just like most things in the ICU, the sooner we address this, the better. There are impact of other things besides resuscitative fluids.  I'm gonna show you some data that I think is kind of intriguing and things that we don't think about quite a bit in our patients. And I'm gonna try to give you a practical approach  to managing your fluids and diuretics. And ultimately, I'll talk about a few strategies and maybe some things we could do daily on rounds because how many of you in the room  have an actual protocol for de-resuscitation of your patients in the ICU?  I did not see a single hand. Okay, so I saw one abstract presented at this meeting from the University of Kentucky. So maybe they're at the forefront. So let's talk a little bit  about fluid imbalance and outcomes. So what we're getting at and what Mike did a really nice job of explaining is that there seems to be this happy medium in a U-shaped curve  where hypovolemia is associated with risks of complications and hypervolemia is associated with risks of complications, looking at intravascular volume. And I won't belabor these  because I thought Mike went over several of them, but in either case, you can end up with multi-organ failure. So you're really trying to keep your patients  in the middle part of this or the optimal range of fluid. Now, this is a paper done by my colleague, Paul Reynolds at the University of Colorado. And it's kind of nice  because it takes that same U-curve and puts some volume associated with it. So in general, this is a paper that's just published ahead of print in pharmacotherapy last month.  It looks at albumin, crystalloid, balanced crystalloids or saline. And this is the U-shaped curve looking at sepsis patients. And you can see that somewhere in this U-shaped curve,  positive fluid balance early on is somewhere between two and four liters, which is somewhere around that 30 cc per kilo. And most patients seems to be reasonable.  But as you get to over six liters, you're starting to run into trouble and anything under resuscitation is bad as well.  So what does this mean if you have fluid imbalance? Well, if you have fluid imbalance, and mostly we're talking about a positive fluid imbalance after our resuscitation efforts,  you have risk of acute kidney injury and ultimately mortality. And if you have patients with no acute renal failure are gonna be less likely to have positive fluid imbalance.  But if you have early acute renal failure or late acute renal failure, these patients are gonna be at the highest risk and are gonna be ones that we wanna pay close attention to.  So this is a study of 198 ICUs, of course, 24 European countries. So look at kind of the SOAP data. And the positive fluid balance was independently associated with mortality,  even when you took away other risks.  So positive fluid imbalance might be bad. So what do we do about that? So let's talk about timing. When is it optimal to address this? So looking at timing and balance  of fluid intake in ICU patients, mostly looking at the first seven days of ICU stay. This is 1800 patients with sepsis, a little over half of them had septic shock.  And on the left, you have non-survivors, on the right, you have survivors. The key here that I wanna show you is that whether you're a non-survivor or survivor,  the fluid intake is about the same on hospital day one. But really, we're looking at in this first 72 hours, the differences start to occur around day three,  where survivors are getting a pretty even balance of intake and output by day three. Whereas in the non-survivors, there's a substantial positive fluid balance at day three.  So I'm going to try to show you data that I believe that this is really our window of opportunity for starting fluid stewardship and de-resuscitation in our patients.  And as you can see, as you go later, so by day three, end up with even fluid balance. And then as you go through days three through seven, starting to work your way negative,  negative being here below the line.  And the reason that I think this is kind of important is as you look at more and more data and you look at these large cohort studies,  if you look at 24 hours, the first 24 hours, you can't really see a huge difference in death based upon your early resuscitation efforts and fluid balance early on.  So adequately resuscitating patients is still very important. Other talks here have talked about that, how much to give we won't get into today.  But really, as you get into that 72-hour window, that's when we really start to see major differences in the lines and the risk of death. So after you've adequately resuscitated  and immunized the patient, we need to start looking at correcting the positive fluid balance. And this may have implications or may be associated with a lower mortality.  When we think about fluid management during de-resuscitation, I think it's important to think beyond just resuscitation. So we'll start with resuscitation.  We just talked about the Manny Rivers Study, Early Goal-Directed Therapy, PROCESS, ARISE, PROMISE, all these studies. If you look at the first six hours,  whether you're in the standard of care or Early Goal-Directed Therapy, somewhere around 30, 60 cc's per kilo of resuscitation is what we're seeing in these studies.  But the big difference is, as we've started to improve our practice, is that when you looked at the 72-hour cumulative dose of fluids, you're talking about 13 liters  in the Early Goal-Directed Therapy study. Whereas now we're talking about seven, eight, nine liters in our most recent studies of sepsis.  So this is still a significant amount of fluid intake in these patients. So what do we do with this positive fluid balance? Well, I give you an example of a case I just grabbed  from our unit on the way here. And I'm not saying we were doing well or not. This is a 45-year-old male with gram-negative bacteremia, came in otherwise healthy from the community.  Turns out it's a new diagnosis of leukemia. But this is looking at a patient with sepsis and what happened on the first three days of fluid resuscitation.  And as you can see here, yellow is urine output. Blue is things we're giving intravascularly. And not much is going in orally, PO, IV. We've got things like maintenance fluid  or resuscitation fluid, our continuous infusion drips, dexmedetomidine, fentanyl propofol. This is making up 2.4 liters of intake in this patient. And this is just a 24-hour period.  The next up, the thing that I don't think we think about quite frequently in our patients is what is going into the patient besides those continuous infusions.  And we miss this quite a bit unless someone's looking very closely at it. So look at this. This is a patient with leukemia. We're throwing five kingdom coverage at them  with a variety of different antibiotics, antivirals. And here we have half of the IV fluid intake coming from non-resuscitative fluids or continuous IVs.  So things like acyclovir, 400 cc's. And things we may not think about, giving potassium runs, using an electrolyte protocol. Here we have 600 cc's of volume going in this patient  between potassium and phosphorus in a 24-hour period. That's a significant contribution. So and overall, the patient's only making about a liter of urine output.  So this is giving us a 7,500 positive fluid balance in this patient over a 24-hour period. And as you can see, this compounds on itself over the first three days of hospital stay.  And this is corroborated in the literature. So I know this is hard to see and it's really not important. You can read all the data.  The key here is that when you look at resuscitation fluids, that's the orange part of this pie. Blood products are the red. And then nutrition, other types of maintenance  or replacement fluids. And then there's this yellow section which is called fluid creep or extra fluid that we're not really thinking about. So this is the medication diluents,  other types of electrolyte runs, et cetera. This is a huge chunk making up liters of fluid in these patients causing a positive fluid balance.  It's something we don't think about quite a bit. So what does this translate to? It translate that 25% of your IV volume is coming from your maintenance and carrier fluids.  63% of your IV volume is coming from medication diluents on average in most of your ICU patients. So what are some things that we should do?  We should think about restricting the diluents. We should minimize our maintenance fluids and focused on the stewardship that Mike was telling us about.  And this is kind of compounded on a daily basis. So this is looking at that same pie. So day one, yes, resuscitation fluids are a big part.  But as you go to day two, day three through seven, these other fluids, this fluid creep is really where the bulk of your fluids are coming from in your patients.  And it's something that we don't pay a lot of attention to. And as you can see, your urine output and your output generally is not keeping up with this.  So you end up with a heavily positive fluid balance in these patients. And we spend a lot of time talking about resuscitative fluids. Do we use a balanced crystalloid?  Do we use normal saline, et cetera? And what this study, I think elegantly did is it looked at a bunch of different ICUs and it looked at medication diluents.  And if you look at our drugs, which are mostly diluted in saline, you actually see a lot of chloride and sodium imbalances based upon the medication diluents.  So we talk all the time about giving a liter of lactated ringers versus a liter of saline, but we don't think about the liters of fluid we're giving with the medication diluents.  So D5 versus saline in this case, you have higher sodium values with saline, higher chloride values with saline as your diluents. So medication diluents are making a difference  in your anion difference. So let's talk a little bit about deresuscitation. So now that we figured out that positive fluid balance may be something we want to avoid and we  need to do something about, how do we go about deresuscitating a patient? And there's really not a lot of great randomized studies on how to do this.  So I'm going to give you a couple approaches. I think first thing is we need to think about passive deresuscitation. So if the patient has adequately functioning kidneys,  then they may be able to keep up with diuresis themselves. So what we need to do is we need to take a critical appraisal of the fluid intake and the volume status of the patient.  And Tim's going to go over some of our endpoints in monitoring, so I'm not going to talk a lot about looking at intravascular volume. But if you look at the patient I showed you,  as well as the patients in your unit, they're going to be getting well over a cc per kilo per hour of fluid intake on an hourly basis. So anything we can do to minimize  those intravenous fluids, medications, electrolyte replacements, and other diluents or things to keep veins open, that will help. And then we've got to support the kidneys.  So our kidneys are going to be our main mechanism for diuresis. So we either have to support the kidneys in a way that they will function themselves,  or we may need to give medications to do that. And of course, if you have an acute renal failure, then we may be talking about dialysis or other mechanisms.  So let's just look about, if you take an observational study and look at thousands of ICU patients, what does this look like?  Well, this is the data that I kind of showed you already. So this is the intake of fluid, which can be blood products, medications, nutrition, et cetera.  And you can see that these maintenance fluids and medications are making up the biggest part of the fluid intake. And then here's urine output.  And in almost all these cases, the input is greater than the output. So what ends up happening as you move through days 1 through 3 is about 40% to 50% of your patients  are going to need some diuretic assistance. In addition to stopping all the intravenous intake you can take, the urine output will just not quite keep up.  So we're looking at about 40% to 50% of patients by day 3. If you really want to hit that day 3 target of matching your input and output,  you may need to give some diuretics to these patients. And if you're not getting there, you can see as you get into the later phases,  more and more patients are getting renal replacement therapy in order to meet these or to address the fluid imbalance. And again, just to show you kind of the data of what  happens to the association here, as you get patients with a less positive fluid balance, you get improvements in survival or associations with improvements in survival.  And again, the key here is the major significant difference is somewhere around that day 3 mark. So we really need to be, as soon as the patient stabilizes  after on day 1 through 3, this is where we really need to start making our interventions. Because the numbers are not as significant as you move through later in the ICU stay.  So let's talk a little bit about diuretics. So diuretics, after you've done the passive deresuscitation, you've minimized your fluid intake,  now we've got to do something to help boost the output. So this is the active part. And I show this slide because we are one of the sites for the FACT trial.  And this is what protocolized diuresis might look like if you're trying to do a clinical trial. So pick one of 18 boxes that your patient might fit in.  And you're either going to be giving fluid, pressure support, or different doses of diuretics. And that can be quite difficult, right?  So let's talk about it just from a more practical approach. So if you think about patients and their kidney function, the people that have normal functioning kidneys,  they're going to be the ones that are going to be most capable of the passive deresuscitation. And their kidneys are going to have the best chance of keeping up.  So about 40% or so of these patients are going to need diuretics. And you can see blue here means a loop diuretic, so Lasix being the most common one.  And then the other lighter colors are additional types of diuretics. And as your serum creatinine goes up, you can see that not only do you use more of a diuretic,  but you also may be combining your diuretics together to get a synergistic approach. And or you may be using continuous infusions. What I would say here is that, yes,  passive deresuscitation is a good way to start. But expect that, you know, a flip of a coin's chance, most of your patients are going to need  a little bit of diuretic to help them out. And in this case, most of the starting doses of Lasix were around 20 milligrams, regardless of serum creatinine.  But what I would say, and that most of you know as clinicians, is that as your serum creatinine goes up, the amount of Lasix you're going to need  is going to be substantially higher. And also as your serum creatinine goes up, you're going to be likely adding on orange here, which is the thiazide diuretics to it  to get a kind of synergistic approach. And as you get to serum creatinine of five, you're probably looking at more of dialysis and other mechanisms depending on how sustained  their kidney injury is. So think about serum creatinines of one, 20 milligrams is probably a reasonable starting dose of Lasix.  As you get to two, you're probably needing to double that. Three, you may end up having to start with something like 60, 80 milligrams of Lasix to get what you want.  The key here, and I think something we don't do very well, is when you're using your diuretics, we need to be giving a dose, setting goals for our nurses  who are excellent at achieving whatever input-to-output ratio we want. And they can come back to us on an every four- to six-hour basis and tell us whether we're  hitting that balance of input and output, and we can adjust our therapy accordingly. And some different mechanisms, like continuous infusions,  may allow you to be more stable with your fluid removal so that you can allow redistribution into the intravascular space of fluid and excretion.  So I won't belabor you with every positive study I found. I wanted to show you one that also shows that maybe there isn't a difference with early diuretics and outcomes.  This is 13,000 patients. The good news is that diuretics work. You get a better fluid balance. There may or may not be an associated with improved mortality or shortened  mechanical ventilation. The downsides are something that I'd be remiss if I didn't talk about a little bit. So if you over-diuretics the intravascular space,  the diuretics are not directly nephrotoxic, but you can see a little bit of acute kidney injury. The other thing you're going to want to monitor for  in this deresuscitation phase is electrolyte abnormalities. Almost all the deresuscitative, the diuretic studies show that you're going to see alterations in sodium  and potassium content, as well as have increased risk for contraction alkalosis in these patients. So definitely something to look out for and help  you decide how much of your diuresis you're going to do. So how do we practically do this combination of active and passive deresuscitation?  Here's what I think is a good approach in your ICU. First thing is we need to be starting this early. So as soon as they hit that stabilization phase  that Mike was talking about, this is where our interventions need to start. And stabilization does not always mean that you're completely off of vasopressors.  The patient has been adequately fluid resuscitated. They're stable. Their pressors are going down. There's actually a couple of prospective studies  going on right now looking at starting deresuscitation when your norepinephrine dose is less than 0.3 mics per kilo per minute. So that is reasonable. So stabilization.  Then we can start thinking about what IV fluids can we minimize? What medication diluents can we minimize? At our institution, we do not have the microinfusion pumps.  Some people do. That's one way you could do it. But think about how much IV fluids you're giving patients. Diuretics. I think furosemide is probably going to be,  or whatever, your loop diuretic is going to be your workhorse for your institution. But let's be kind of systematic about our dosing and approach.  And let's frequently monitor our patients and adjust. This could be like an every six hour basis as far as do we need to give more or less Lasix to this patient?  And what sort of electrolytes do we need to keep up with? And then I think the goal is to achieve a balance of input and output by day three of the ICU stay.  And we want to work our way negative as tolerated between days three and seven. Now, as much as we can, I think those are kind of our targets. And things to look out for are going  to be hypotension, the electrolyte abnormalities, worsening kidney function, and then the metabolic alkalosis we talked about. So I'll give this to you as a beginning approach  as you do your rounds in your ICU. So most people use Fast Hugs or Fast Hugs BID in their unit as kind of a checklist to keep up with the things we need to keep track of.  I think that the F needs to be two Fs now. I think fluid stewardship becomes part of our F in our Fast Hugs. And I think, as you're talking about,  as Mike said, de-escalating antibiotics, I think we need a second D, which is de-resuscitation of fluids in our patients. So as you're going through your checklist on a daily basis,  I would suggest adding fluid and deresuscitation to your checklist in your ICU. At least put it on your radar to keep track of, so you don't end up with that patient  that I just showed you in our unit currently while we're here at SCCM, who's 8 liters positive, and not on vasoactive agents. So let's talk about some conclusions.  I think deresuscitation, it's going to be an individualized approach. But we need to make it practical and easy for our nursing staff and others to manage. Think optimally.  Let's get it done in the first three days and start our approach then. And then let's keep a close eye on all the other types of fluids that we're giving patients  from an input standpoint. And let's try to be as minimal as we can in that approach. And then diuresis is going to be needed.  But we're going to have to be careful with our diuretic approach in these patients to avoid adverse events.  And with that, I'll pass it over to Tim, and we'll finish up and take questions at the end. Thank you. All right, excellent. So as mentioned, my name is Tim Ellender.  I'm an associate professor at Indiana University, and I practice both emergency medicine and critical care. And I'm going to talk to you today a little bit about the endpoints.  So how do we make that transition? We talked eloquently about the need to identify resuscitative phase versus deresuscitative phase.  But just where does my patient fall, and how do I measure sort of basic endpoints to decide when to say when?  So again, I have no true disclosures to discuss, but I am a member of the International Fluid  Academy, and Manu Malbrain and I have been in contact with regards to some of the content in these slides.  So I think what we're trying to say to you is that early adequate management is still a priority. We're not asking you to under-resuscitate your patient, right?  We still have a role for fluids in the hemodynamically unstable patient, especially as they augment cardiac output and improve peripheral perfusion deficits, correct?  So fluid outputs are important. Fluid intake and boluses are important, but only in the context that they truly augment cardiac output.  And I think many have argued that point now for well over five or more years. And of course, there's an uprising in Europe about how fluids can be dangerous and how  fluids can be bad. We're not saying that all fluids are bad, but fluids are a drug. They have a dose, a duration, and a delivery.  And those fluids, with regards to bolusing for resuscitation, have meaning when they augment cardiac output.  When they no longer augment cardiac output, there probably is no longer a role for fluids as a bolus agent.  So the big quagmire, or the thing that we have to answer is, when do we say when, right? When is my patient resuscitated?  When do they make that transition to the stabilization phase? And how do we use these endpoints to measure when do we start the de-escalation and de-resuscitation phase?  And that's sort of the context of the next 10 minutes of our conversation. So fluids are a complex item, right? We know that there are certain pragmatic reasons to use them.  However, I think increasingly we know that fluids can be evil, and they can cause harm to our patient. They are, in fact, an iatrogenic harm, right?  They do induce some level of histologic inflammation, especially in overload, in high balance fluid overload.  And so how we couch fluid resuscitation in the individual phases with which we resuscitate and de-resuscitate become a very important item.  So positive fluid balance, we will kind of quickly review this. I think my colleagues before me have shown that positive fluid balance have impact, right?  I think the FACT trial probably did the most in the last 15 or more years to really highlight  that fact and show that, in fact, a conservative management strategy in actually taking fluids away from patients had a major impact.  And in this scenario, it related to ARDS with prolonged recovery. Shortly thereafter, the SOAP study showed a very like signal in patients who had acute kidney injury.  And followed that in the very same year, the VAST trial showed a very similar model in patients with septic shock, in fact, improving mortality.  Rosenberg and colleagues showed not only did quantitative balance, not only for seven days or 14 days, but in fact, fluid balance over the first three to four to five days had an  impact. And in fact, when we began to impact that balance and drive patients toward a negative  fluid balance, we were able to show mortality early or mortality and benefit, and again, an ARDS population.  Lastly, but not least, Malbrain and colleagues showed a very similar signal, again, in patients with intra-abdominal hypertension.  And though I don't show and highlight additional studies, this has been shown time and time again in bowel ischemic populations, cardiothoracic surgical populations.  So again, fluid balance has meaning. So positive fluid balance, we talked a lot about it. But really, how do you define that, right?  Well, the duh moment is, well, it's clearly fluid overload resulting from administration of fluids.  And a lot of us think about resuscitative fluids, which in fact, the bulk of that fluid for those early phase patients comes from our resuscitation. But I think Ty mentioned it.  A lot of fluids in what we describe as fluid creep come from other sort of modalities, our therapies that kind of side with our fluid resuscitation.  So the antibiotics, the fluid delivery with electrolytes and whatnot. For research purposes, fluid balance and fluid overload has been defined as a 10% fluid accumulation,  so a 10% cut point above and beyond what is normal balance. And that has been associated with abnormal outcomes. Again, when you look at it, how do we define this 10%?  Well, it's simply you take the total cumulative fluid balance, divide that by the patient's gross weight, and multiply that by 100%.  That gets you the percentile score for fluid balance. Again, that 10% and above tends to portend worse outcomes in one way or another.  So again, not to reiterate and rehash this, but this has been shown time and time again. This is one, again, meta-analysis published in 2014.  There have been three additional meta-analyses published since then that have shown, again, positive balance portends poor outcome, or a signal at least, for negative fluid balance  to portend improved outcomes. That makes sense. So why? Well, I think, again, Michael showed this. This impacts a lot of things, right?  It impacts predominantly, probably, again, a cytokine initiation on a microvascular level, especially three to four days out where that fluid balance is grossly positive.  But it also impacts organ tissue function, such that you have hepatic congestion, cerebral congestion, renal congestion.  So that cosmetic surface edema that we saw probably also relates to the microvascular level and the organ-based level as well.  So again, if we know that fluids are bad, and we know, well, could potentially be bad, the question, the big question for all of us, should be when to say when.  Well, a number of theoretic concepts have been proposed. I think Michael reviewed this ROSE concept, which looks at several phases of resuscitative  strategies where early goal-directed therapy might be important in the early phase, things like late conservative fluid management, and then last but not least, late aggressive  fluid recovery may play a role in the de-escalation phase or the late phase. Another model that has been proposed is sort of this ebb and flow phase of sepsis in sort  of shock states alone. The ebb phase is classically that volume-depleted Grand Canyon-esque effect where we apply fluids to refill the tank, so to speak.  And then as those persons stabilize, that fluid, that cumulative balance of fluid, begins to leak and flow into other interstitial spaces and then begins the flow phase with which  we have the potential to then drive fluid recovery. And that's going to be the focus of the next few slides.  So again, if you overlap the models, the ROSE and this ebb and flow phase do sort of coordinate nicely.  And we're going to sort of review the individual details of this and how our endpoints interdigitate within those phases.  So there are a number of clinical endpoints that we can utilize, things like pedigedemia, fluid balance, cumulative body weight, things like jugular venous pressure.  Those things give us a superficial gestalt for what the patient looks like and perhaps where our fluid balance may lie.  Other biochemical parameters, which are probably less helpful, especially downstream, they're probably more helpful in the resuscitative phase.  Things like albumin ratios, lactate changes, things like osmolar units can be used, but are very hard to use at the clinical bedside in any meaningful way.  Probably more importantly, and things that we've used more successfully are hemodynamic endpoints, things like mean arterial pressure.  Classically, things like CVP, which are large in a way, have been overtaken by things like volumetric measures, passive leg raise, things like stroke volume variation are probably  more telling for how responsive our patient is to fluid volumes. And last but not least, things like imaging, especially with regards to ultrasound, help  us tell more about lung volume, water, perhaps intra-abdominal fluid, hepatic and renal congestion.  So we can use these as adjuncts with our bedside examination to sort of tell us more data. Downstream, I think we have the potential for more organ functioning monitoring.  Again, these may be biomarkers, large scale. Sometimes that can be invasive monitoring as well.  Things like intra-abdominal pressure, ICP monitoring, cerebrovascular sort of monitoring  may have more telling data for us, but that yet has been uncovered with regards to resuscitation strategies.  And last but not least, there are other measures, things like albumin ratios, total body water, which can be used as well.  So again, we're going to take the ROSE analogy and sort of break it down through these phases. So resuscitation, I think alluded to. This is salvage, right?  These are patients who are hemodynamically challenged. They are at risk for hypoperfusion and ischemia.  And at this phase, we essentially turn on the spigot, let the water flow in with hopes to recover cardiac function and improve perfusion deficits.  Classic caveats or endpoints there would be things like mean arterial pressure, again, with suggestion to keep mean arterial pressure greater than 65.  Cardiac indices, cardiac indice measures, where we target a measure of 2.5 or greater. Pulse pressure variation, again, using pulse pressure variation, hopefully less than 12%.  Again, depending on modalities you use at bedside. And if you are an ultrasound modeled sort of unit, then perhaps biometric and end diastolic  areas may be helpful to sort of drive home where you are in the resuscitation phase. From the optimization phase, again, alluded to by Ty, this is where your patient begins  that stage of recovery. This is where fluid bolusing no longer seems to apply, where the cardiac recovery no longer seems to be registering with fluid administration.  These people often are starting to trend downward on their vasopressor needs. They're still sick.  You can recognize them as sick, but they're not quite that de-resuscitated, dire-looking patient that you saw at hour one through six, right?  In this phase, we start to begin, again, to think about positive fluid balance. Where is my daily weight? What is my urine output?  Where are my other fluid creeps starting to play a role? And the question here is this fluid balance that we've alluded to, is this really just  a marker of illness, perhaps, in those that are severely ill, or is it a contributory illness? We don't yet know that, but ongoing studies hope to tell us more.  Again, these people are unstable, but they still need our care. But this is an opportunity where we can start to back away the fluid flow, all right?  Again, mean arterial pressures, cardiac indices, similar to our resuscitation phase, volumetric measurements may be helpful.  In addition, this is where we start to monitor potential downstream organ detriments. So this is an opportunity to measure things like intra-abdominal pressures, perhaps, preload  optimization with, again, end-diastolic volume indices. And we are able to start to say, this is the point where I think we can start to ask, when is enough enough?  For the stabilization phase, again, this is the sort of key role where we need to back away. And I think Ty mentioned that.  That's where we can truly start to pull fluid therapy back whole scale. This is a patient who is in absence of shock.  This is the patient whose vasopressor requirements are little to none. And fluid therapy shows no role in improving cardiac output.  Again, important in this role that we start to really pay keen attention to monitoring of daily weight and fluid balance.  And hopefully, if you have that opportunity, things like organ function, whether it be cerebral performance, intra-abdominal pressure measurement.  And our targets are truly net neutral balance at this phase. We're targeting negative fluid intake and positive fluid egress.  In this phase, there are no great mediators for endpoints. But there are perhaps indices for things like extravascular lung water, pulmonary vascular  pressure indices, intra-abdominal pressure, again, colloidal oncotic pressure may have roles in telling us how our patients are doing and where we can be more aggressive in this  phase, if that makes sense. So here is what Malbrain et al sort of describe as the sort of late conservative fluid management stage.  And it occurs roughly at this point, where fluids should be utilized, but only for ongoing maintenance of loss.  So if you have a patient with a puritanial process with large scale puritanial fluid leaks, then it's OK to address those leaks.  But if the patient doesn't need aggressive volume, then it's time to turn that volume down, right?  So again, this is the phase where we have to be really careful with fluid creep, as mentioned by Ty.  This is where our hidden obligatory fluids come in, and our overload really starts to be driven home by our co-therapies, if that makes sense.  Again, this is a phase where maintenance and totally fluid should only be used to cover daily losses and daily needs.  And we should be attentive to the role of fluids used for nutritional support, and utilize them probably preferentially over intravenous fluids, if at all possible.  Again, the net goal here is to achieve fluid balance. Again, the hope is that over two consecutive days, roughly within that two to three to  four to five day period, where we would have two net negative fluid days within that period, hopefully within the first week of ICU stay. The evacuation phase is the final.  And I will tell you that not all patients go to this evacuation stage, right? So those of us that have done this long enough, you know that there are a caveat of patients  who come in looking really sick, right? They get aggressive fluid. They get hydration. I think the sign model is sepsis in my mind, but this goes for a lot of different states.  But they get fluids. They get their antibiotics. They get aggressive hydration. They get aggressive ICU care. And within two to three days, they look much better, right?  Sometimes even within six to 12 hours, they look markedly different from that patient that entered the door at hour one. These individuals often will skip this stage.  But there are a high degree, maybe 60 plus percent, that really jump this phase. There is a caveat of patient or a grouping of patients who swell up like Michelin men, right?  They seem to dwindle for a while, right? It takes three, four, five, six days for them to truly make that jump where their stability is easily defined by everyone at the bedside.  And these people seem to fall into that next category that hits this microvascular sort of inflammatory cascade that probably is not only potentially driven by the fluids  that we've used, but perhaps is a marker of this fluid balance issue. And these are the individuals that have more global permeability problems, right?  These are the people that see downstream organ problems that persist sometimes for weeks, right?  These are the people where fluid overload itself may contribute to true end organ injury.  Here we need to avoid enthusiastic fluid removal, potentially, and I think Ty mentioned that, because we can also drive additional hypovolemia, which begins the cycle again, right?  And in this phase, we have to consider perhaps late goal-directed fluid evacuation or removal de-resuscitation, which I think Ty alluded to. So there is some data there.  A lot of it has been largely retrospective, but there are some small prospective studies that have shown us some ability to guide de-resuscitation strategy.  After initial resuscitation and stabilization, again, we would try to encourage discontinuation of invasive therapies and transition to net fluid balance.  This is in the E phase or the egress phase. Late conservative fluid management is tantamount. So again, balance where you can, turn off the spigot.  And then perhaps there is a role for aggressive fluid removal, as Ty has already mentioned. De-resuscitation, again, zero to negative balance by day three would be our goal.  Keep the fluid balance as low as possible throughout that last seven days. And in fact, diuretics and renal replacement therapy can be utilized here in combination  with or without albumin, perhaps to improve diuresis. And we do have, again, some study to show this.  Again, Korman et al. in 2012 published what was at that time the first true prospective look at de-resuscitation strategy.  And they used what they called a Powell approach, which stood for HiPEEP, albumin administration, and LASIK administration, furosemide.  Again, this is a pilot study looking at acute lung injury with the goal of achieving negative fluid balance within a period of three to five days.  And what they found was through this methodology, they were not only able to measure a reduction  in extravascular lung water and intra-abdominal pressure, but they were able to reduce the duration of mechanical ventilation without, again, compromising organ function.  So no AKI and downstream effects with this protocol. Again, what they applied was a HiPEEP for at least 30 minutes, equal to at least intra-abdominal pressure.  And this was in hopes to sort of balance the pulmonary system. And then albumin was administered.  That was 20% albumin administered as a double bolus, 100 mils over a 60-minute period. That was on day one.  That albumin was then followed roughly 60 minutes after by 60 milligrams for four hours per hour. So this was a drip form.  And then titrated between 5 to 20 milligrams to maintain fluid output roughly at 100 mils per hour over that time frame.  Again, albumin was duplicated on day two and three to continue fluid pull. And what they were able to show is not only were they able to show a difference between  control and PAL groups with hospital length of stay, but they were also able to impact not only that, but mortality. And so this has largely taken hold, especially in Europe.  And others have sort of proclaimed this as a methodology. And I think Ty mentioned this in his transition. So again, this was a small pilot study.  So it's hard to come forward today and say that this is the way to do it, right? There is signal that this has some efficacy at bedside.  And ongoing as we speak, enrolling as of 2018, is a large interventional pilot study randomized  amongst multiple centers, trial call radar, which is role of active de-resuscitation after re-resuscitation. So again, it's open label allocation concealment.  The role is active de-resuscitation. So these patients are all being resuscitated, according to appropriate strategies, with  two groups, conservative management strategy, which then undergoes active de-resuscitation  versus a group that goes under standard of care with observation and measurement following. Again, their primary outcome is hopefully fluid balance achieved within day three, change  in fluid balance, and then a host of different mortality, morbidity, and clinical outcome endpoints. So I will stop here and ask my colleagues to rejoin me on stage.  And then we will take collective questions from the audience. I do invite you to come to the microphone. So if you ask the question, please ask in front of the microphone.  We'll do our best to repeat. Caveats of those questions and answer the questions that you provide.  Thank you.  I have a question. Do you think we're headed towards protocolized de-resuscitation? Do you think we should?  So do I think we're headed that way? Yes. I think the signal is that at some point, de-resuscitation will become part of the vernacular.  We will have resuscitation, de-resuscitation. My trauma surgeons and colleagues in other disciplines respect this. They realize this was a need many, many years ago.  I think we're now just starting to gain ground on the evidence that supports this de-resuscitation sort of theorem.  And we're starting to gain ground on measures that allow us to do it safely. In other words, we have the data that support LASIKs plus albumin, various measures of endpoints  probably can help us do this safely at bedside. I would concur. I mean, it's going to be a multidisciplinary approach. So the nurse at the bedside is going to be the key component.  The pharmacist is going to know the physician. So a protocol allows people to set goals, and we set goals. Things get done timely in the ICU, so I think.  And it keeps us from over or under doing it. So in that sense, I think a protocol makes a lot of sense. Sir? Hi.  We're all familiar with the parameters that equate to volume responsiveness in our patients. Is there any objective parameter to show when our patients may benefit from fluids?  Most of what you've said are clinical, may benefit from a net negative fluid balance. Most of the parameters you spoke about today are still clinical.  Say, for example, would a stroke volume variation or pulse pressure variation below a certain percentage indicate that our patient may benefit from a net negative fluid balance?  I'll take that. So I actually did research that sort of topic. There is a policy of data in regards to stroke volume variation, right?  So a large majority of it guides resuscitative strategy, right, such that a volume variation under 12% is positively linked to fluid response.  There is a very, almost no data supporting changing that variable in a different direction to support de-resuscitative strategy.  Where there probably is a bulk of data is with regards to using Lydco and Pico in extravascular lung volume water.  So measuring volume of lung water and the reduction of that is the best strategy available  if you're looking for true objective measures. So if you're a numbers guy, the best thing to guide your de-resuscitation strategy at  this point is looking at things like intra-abdominal pressure and extravascular lung water. Thank you. Yeah. Hi.  So in 2017, it was an industry-funded study, but it was a prospective look at basically the cheetah system or an electronic bioimpedance system.  It seemed to suggest a signal toward less fluid creep and, in general, some improved patient outcomes.  I was wondering if anybody on the panel had any opinions or personal experience with these kinds of non-invasive newer bioimpedance devices.  My shop's exploring them, and I was just wondering what your thoughts were on it. We don't currently use them. We have used them in our clinical setting.  One thing to remember is they're dependent on a regular heart rate. So if you have irregularity in your cardiac rhythm, the data gets skewed.  So that's one thing that kind of throws their utility off. But yeah, they can be a useful tool. Like any tool, those metrics need to be taken in context of the patient.  So our shop does use cheetah, and as a caveat, I did not speak to that. I was one of the co-investigators at our shop for the cheetah study. We did have experience with it.  It did allow us, in my opinion, to better curtail fluid volumes non-invasively, but there are multiple ways to skin that cat, if that makes sense.  So if your shop can't invest in invasive methodologies, I think that is an alternative strategy. As mentioned by Michael, there are negatives to it.  To plug that as well, bioimpedance, not by cheetah itself, but other bioimpedance measures do allow you to extrapolate lung vascular water and things like that.  So I think in the future, there are going to be other modalities that we can utilize, both invasive and non-invasive, to help guide this de-resuscitative mindset.  But I think we're just at the forefront of that. We have yet to gain ground on high-quality level data to drive that sort of strategy. Thanks.  I think we have time for one more question. I'm not sure who was... Maybe two. Maybe two. We have two more. My name's Lars. I'm over at Cooper in Camden, New Jersey.  One of the things that I've been struck by is how often my intensivist opinion of the patient's volume status is diametrically opposed to the nephrologist's opinion of their volume  status. And I guess my question, particularly as it relates to, as we're de-resuscitating, we're coming down off pressers.  I'm really stoked about where the patient's going, but I can't... My nephrologist, because they were consulted because of hyperkalemia or some other reason  that they were invited to the party, but when I'm like, I want a diuresis too, and they're like, well, he's on norepi.  I was like, yeah, but he's on only 10% of the norepi he was on. Why don't we start pressing on the gas a little bit?  I was just wondering some communication strategies you guys might have used.  Yeah, I think it's a great point. And I think there's going to be some data coming out in the short term that's going to give us an idea of what we should do in those situations.  But just because they're on a low dose of norepinephrine, that may be what's needed to perfuse the kidney and provide support and flow to end up helping you with the diuresis.  So I wouldn't necessarily say that being on norepinephrine means that the patient is not adequately resuscitated.  They may actually be volume overloaded, and maybe the norepinephrine is just helping support your kidneys while you diurese. So I think it's a balance.  It makes you uncomfortable taking volume off sometimes when someone's on pressure support, but I don't know how you navigate that conversation with a nephrologist.  That's not my usual place, but I think if you're committed to diuresing the patient, then it makes sense to move forward with it.  I think in future we're going to have to engage our nephrologist colleagues in this data. I don't think this is a data set that they go to for evidence for what they do on a daily  basis. I think like a lot of things, we have a bevy of information that is common to us, and we have to share it with the outside world and bring everybody up to speed.  And I think those are ongoing conversations that are going to be had in the coming five years. Hi. Thank you again for just a great talk.  So just real briefly, I'm a neurointensivist, so the brain's always foremost on my mind.  In the ARDS cognitive outcomes trial a couple years ago, there was a pretty potent statistical association between a restrictive fluid strategy and worse cognitive outcomes.  I was hoping that you guys can kind of comment on that, and is this something that's kind of context sensitive and only for septic patients, or is this something that's more  generalizable? I'm not familiar with that, Tyrone. Yeah. So in context to your question, there is a bevy of different endpoints if you look at strategy.  There's obviously some signal to where certain things, like Ty mentioned, where perhaps with de-resuscitation you could invoke kidney injury. I think there's a lot of gray there.  And that's why those of us that are intensivists have this job. I think you have to sort of take the clinical picture that you see at the bedside in context  with the objective data, the sort of subjective data as well, and put that together. It's somewhat hard, I think, to cluster everyone into a one-size-fits-all sort of model.  And I'm not sure how those neuroscience patients fit directly into the sort of protocols that he alluded to, especially with regards to FACT and others.  So it's really hard to extrapolate that. I think we have to start to uncover it. And I think there's going to be a role for studying it in a TBI model or in a stroke model.  I think we just have to come to terms with that sort of level of study. I don't yet have enough data to really guide that question, is what I'm saying. Anybody else?  Is there one more? Yeah. Hey, Ty. I think you were alluding to this a moment ago, but my question was, are you aware of  any studies currently looking at the diuresis of patients on moderate to even high doses of vasopressors who have echo evidence of RV dysfunction or venous congestion in the  hepatic vein or the portal vein? I don't know about ongoing. Obviously, the big study currently is CLOVER, right?  So CLOVER is the new NIH-based study looking at sepsis resuscitation and early vasopressor use.  I don't yet know if there's a large-scale study looking at specific ultrasound-derived  parameters like hepatic congestion, vena cava size just yet to guide de-resuscitation strategy. I apologize. Are you familiar with anything specifically? I am not. All right.  Thank you. Yeah. Okay. With that, we'll conclude. Thank you very much.

Video Summary

This video transcript discusses the concept of de-resuscitation and fluid stewardship in the ICU. The speaker explains that while fluid resuscitation is important in the early stages of shock, continued aggressive fluid administration can lead to positive fluid balance and potential harm to the patient. The speaker proposes different phases of fluid management, including resuscitation, optimization, stabilization, and evacuation, each with different goals and approaches to fluid administration. The speaker emphasizes the need for careful monitoring of fluid balance and the use of dynamic measures to guide fluid therapy, such as mean arterial pressure, cardiac indices, and volumetric measures. They also discuss the importance of de-resuscitation, in which fluid intake is reduced and diuretics may be used to achieve a negative fluid balance. The speaker suggests that implementing protocols for de-resuscitation may be helpful in guiding clinical practice and ensuring optimal fluid management.

Asset Subtitle

Pharmacology, Quality and Patient Safety, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Examine the role of fluid stewardship in critically ill patients
Assess end points of de-resuscitation in the ICU
Examine the role of central venous pressure in starting, monitoring, and ending de-resuscitation as part of fluid stewardship in the ICU
Apply fluid stewardship in special populations, especially obstetrics, neonates, and comorbidities such as diabetic ketoacidosis, burns, and dengue hemorrhagic fever

What Is Fluid Stewardship and How Should It Be Applied in the ICU?

Michael A. Pisa

Diuretics and Managing Fluids

Tyree H. Kiser

Assessing End Points of De-Resuscitation in the ICU

Timothy Ellender

Meta Tag

Content Type Presentation

Knowledge Area Pharmacology

Knowledge Area Quality and Patient Safety

Knowledge Level Foundational

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Mortality

Tag Fluids Resuscitation Management

Year 2020

Keywords

de-resuscitation

fluid stewardship

ICU

fluid resuscitation

shock

positive fluid balance

fluid management

mean arterial pressure

cardiac indices

de-resuscitation protocols

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.