So, good morning, everyone. Can you all hear me? Okay. All right. My name, again, is Andre Holder, and I'm an assistant professor at Emory, as was mentioned earlier. So, today, I'd like to talk about just an update on vasopressor and fluid resuscitation in shock states with an emphasis on the fluids, given the context. So, again, this is my bio. I'm sure you can read it online. I don't need to focus on that. However, I do need to talk about my financial conflicts of interests. I'm currently funded by the NIH to look at predictive algorithms for sepsis, but none of that will really come up in this talk. I also do get consulting feeds from Philips, and, of course, also from Baxter. So, the objectives here is to help understand... By the time we're done with this talk, I'd like for all of you to be able to understand that this field is evolving around early fluid resuscitation. I'm specifically going to talk about two areas in this space, so the amount of fluid that we give early on, as well as the fluid type. So, I also would like for all of you to walk away with some understanding of a concept of something... An emerging topic called vasopressor responsiveness, very akin to what you might have heard about fluid responsiveness, and then also to identify the impact of MAP goals, specifically on the outcomes in the elderly, based on a trial that we'll talk about later. So, I'd like to read all of you a bedtime story now. Once upon a time, in a land far, far away, there were doctors who didn't really pay a whole lot of attention to their sepsis patients very early on. And along came a person named Emanuel Rivers, who created a clinical trial that tried to hone in on that very topic. Now, many of you are familiar with this, perhaps. This came from the original Early Goal-Directed Therapy paper. And really, the paradigm shift here, of course, was the fact that we were focusing on these patients early, but also, how can we identify targets for when patients might have had enough volume? First off, knowing that we need to provide organ support for those patients. And volume and achieving some kind of MAP goal to achieve adequate perfusion were hallmarks of that trial. Now, of course, this trial has been poked and prodded since then. We've looked at the marker that was used as that target, CBP, and of course, we know the studies around that. And then, of course, there was actually a question about whether or not the protocolization of care in these patients was even helpful at all. Of course, there's a lot of conversation about that that I'm not going to focus on in this talk, but the bottom line is, our practices have changed because of these trials and the fact that, perhaps, maybe it was just the early recognition. At least, it was what came from these trials. And so, that's where the conversation and the focus of our care tended to pivot. Now, why fluids? So, fluids are obviously great in terms of, particularly in patients with sepsis, where you have issues with insensible and sensible volume losses to increase perfusion to potentially damaged organs. However, there are some consequences to that, and I'm going to focus you here on the microcirculation. So things, first off, around endothelial cell dysfunction, and of course, these come from the inflammatory markers that are released, as well as the damage-associated molecular patterns that we see in sepsis, as well as in the later forms of other types of shock, like ischemia, that we can see in any type of shock. So the inflammatory process tends to be present in all of those types, and so we see that endothelial cell dysfunction as a result of those inflammatory markers, disruption of the glycocalyx, widening those tight junctions that we have between the endothelium and also endothelial cell death, as well, and then, of course, the inflammatory cascade is just a positive feedback, and we start seeing microvascular obstruction, and then we see this capillary leak that's pretty common, again, in patients with sepsis, as well as, again, later other types of shock when they're later in stage. So of course, when we give volume, we know that there's consequences to doing that. As we see capillary leak, we know that that's going to affect the viability of that tissue and the functionality of that tissue, wherever it might be. So of course, those factors contribute to some of the things that we see in clinical practice, right? So we see our patients, some of that may be cosmetic. We may see a lot of peripheral edema, right? However, there are some other consequential things that happen in these states, depending on where we see that edema. Of course, the lung, which I'm sure many of you have seen in your practice, and this increasingly recognized phenomenon of abdominal compartment syndrome that our surgical colleagues have recognized for decades, but we're now starting to understand, especially given the context of large volume resuscitation that has been the mantra for at least two decades. So we don't want to do that, right? Now let's ask some questions. So how often will fluids increase stroke volume when given by providers in the right clinical context? Flip a coin, right? So this is important, because we tend to give fluids indiscriminately if we don't have some way of assessing that. Too much fluids, of course, is bad, and I'm sure you're all familiar with John Boyd's trial from a few years ago, looking at the VAST trial, sorry, his study looking at the VAST trial, secondary analysis, that showed, depending on the quartiles of fluid resuscitation that was received, that was directly related to mortality. So maybe we should revisit this idea of protocolized care. Maybe we shouldn't throw the baby out with the bathwater. What we're doing may not be the right approach. So we're sort of at this deja vu moment where we're thinking again about protocolized care in a slightly different way. So along comes these two trials that came out within the past year in New England Journal. I'm going to focus a bit on these to talk about some of those points. So I'm going to have them side by side for comparison, because many of the things here are common themes. So the classic trial, their specific clinical questions were, can protocolized care decrease fluid volumes over wild-type care, and then secondarily, if we see that decrease in fluid volumes, does that translate to a mortality benefit? So this was a multicenter European trial, RCT, and again, the arms that patients were randomized to were restricted fluid versus control or basically wild-type, so do whatever you normally would do in your clinical practice. Patients and providers were unblinded in that study, and the primary outcome here was 90-day mortality. Compare that to the Clover's trial, and we are fortunate to have one of the primary investigators sitting next to us. If I'm saying anything wrong, you'll let me know. But the main question here was, can a decrease in fluid volume that's given through protocol lead to a mortality benefit? So this was a multicenter US-based randomized control trial, and here, the arms were a bit different. So instead of restricted versus wild-type, here it's a restricted fluid protocol versus a liberal fluid protocol. And both patients and providers, again, were unblinded, and the same outcome was assessed here. So let's talk first about the classic trial. Their intervention arm protocol was as follows. So everyone received the usual 20 to 30 cc's per kilo of fluid at the outset, and the restrictive arm basically replicated what we would normally do in clinical practice. If we see someone that looks like they have severe hypoperfusion or severe circulatory impairment, and this was specifically defined by all the things that you would use in your practice, lactate, blood pressure, et cetera. They were given 250 to 500 cc's of IV fluid and no more. And then this would be reassessed. And the standard care arm, again, it was just based on what we would normally do, and presumably they were abiding by the 2016 Surviving Sepsis Campaign guidelines, since that's what was the contemporary practice at that time. Now contrast that with the Clover trial, and again, it's a bit more complicated because we have two different protocolized arms. So I'm going to try to highlight the things that are common and the things that were different between these. So everyone, again, received a certain amount of fluid prior to randomization, up to about two liters of fluid. However, the differences here is there was less that would trigger the intervention, and the intervention was focused primarily in the restrictive arm, of course, on increasing MAP using vasopressors as opposed to fluids. And you see here that the triggers for that intervention in the control non-restrictive protocol was a little more voluminous. No pun intended. Now there were also some pop-off valves for safety reasons in this protocol. Patients were pulled out of their randomization arm if they had signs of severe injury, either ischemia or what have you. And then there was rescue with the other intervention. So if they were in the restricted arm, they would get fluids if they received an increase in their MAP from vasopressors and they still had signs of hypoperfusion. And the same is true for the control arm in the sense that they would get more vasopressors if the fluids didn't correct that. So long story short here, the patients were generally similar in both arms, and I'm just going to highlight, and this is on the left, by the way, is going to be the classic trial, on the right is going to be Clover's. So I just highlight here that the amount of fluids that patients received in the classic trial 24 hours before randomization was similar. And the same was true of the median fluid that was given in the Clover's trial. And also, less fluids and more negative fluid balance was achieved in the restrictive arm for the classic trial. Again, this is comparing to standard control, so they needed to justify that what they were actually hoping to find was actually what they found. And what we note in the Clover's trial, we see that there is, as we expect, a less fluid given in the restrictive arm. And we see that there were, conversely, more vasopressors and a higher frequency of administration of vasopressors in the liberal fluid arm. Again, this is what we would expect. That said, there was no difference in mortality in these, within these two groups in the classic trial. However, there was also no serious adverse reactions. In other words, at least it was, we find that it's at least safe in the restrictive, in the classic trial, in the restrictive arm. And the same was true of the Clover's trial. So we didn't see an improvement in the restrictive fluid management, but perhaps we're asking the wrong question, right? So we're assuming that anyone who's presenting with hypoperfusion in either of these arms just needs a one-size-fits-all approach. And perhaps what they need is something that's tailored to their specific physiologic need. So what do we really want to know, right? What we care about in fluid, in giving patients fluids, is whether or not they're responsive to it. Whether their stroke volume would increase as we administer the fluids. And that wasn't really assessed in these two trials. So again, how do we really know that the fluids are doing what we expect them to do? That is to say, increasing stroke volume and therefore increasing perfusion at the tissue level. Did the treatment teams actually assess response? So probably Dr. Douglas can speak a bit more to that, but at least based on what was published, it didn't seem as if that information was captured. So this goes back to the point of fluids as drugs, right? It's the most commonly prescribed drug in the ICU, if you think about it in that way. So with any drug, you want to give the right drug, the right dose, the right day and time, and the right duration. So up until now, we've been talking about the right day and time. Should we tailor the fluids that we give depending on the physiologic need? I'm going to switch gears here and now talk about the right drug. That is, which fluids should we be giving patients? So this idea of balanced crystalloids versus normal, or as many of my colleagues would say, abnormal saline to all the nephrologists out there. So many of us have used it in practice, especially those who have been practicing long enough. However, we are understanding the physiologic consequences of using 0.9% saline. Hyperkalemic acidosis, oliguria, which would lead to oliguric AKI, and also potentially death based on some of the studies that you see down below. However, even though this is changing, it's still a fairly common fluid that we are administering in medical resuscitation of patients with shock. So of course, I'm going to focus a bit more on the trials that have come up recently that talk about this topic. So the SMART trial, the question here is, does resuscitation with crystalloids, does the resuscitation of the crystalloid type actually impact outcomes? So this was a U.S. based study. It was a pragmatic cluster randomized multi-crossover trial, a long way to say basically we're trying to make it practical, so it wasn't your standard RCT that individualizes by patient level. This was an unblinded study, and the arms were either balanced crystalloid, either in the form of plasmolyte or lactated ringers, or abnormal saline, pardon the expression. And the primary outcome here is major adverse kidney events within 30 days. So this is a composite outcome of either death, need for renal replacement therapy, or signs of renal dysfunction as evidenced by creatinine. The PLOS trial asked the same question. However, it was based in Australia and New Zealand. It was a randomized controlled trial. In the classic form, it was double-blinded, unlike the SMART trial. And here, the arms were simply plasmolyte versus normal saline. The outcome here, you'll note, is also different. It's 90-day mortality. The BASICS trial asked the same question as well. However, this one was in Brazil. It was a factorial RCT, so they looked not just at the type of fluid, but also the rate at which it was infused. And I'm not going to focus on that particular study, but just understand that that was part of the study design. This was a double-blinded trial, and they looked at either balanced crystalloid, just like the SMART trial, versus normal saline. And again, the same primary outcome as the PLOS trial. So what are the results? Patients in the two arms were found to be similar, again, at baseline in all trials. So some interesting things. The SMART trial, balanced crystalloids actually did decrease their composite MAKE30 outcome, as is evidenced here. And you can see that on this side of the forest plot, balanced crystalloid is better. Normal saline better. We see that our overall composite outcome is more on this side, as evidenced also by our odds ratio and the statistically significant p-value. Now, one thing to note, though, is that the absolute risk reduction of that composite outcome was only a 1% difference. Now, of course, they had 15,000 patients to choose from, which allowed them to power to this kind of absolute risk reduction and still see a statistically significant benefit. The PLOS trial, conversely, did not find a decrease in their outcome, which was 90-day mortality, as you can see here. Now, why is that? Why the discrepancy? So there's a number of reasons for this, and I obviously can't go over all of them in the context of the 15 minutes that I'm given. Just to highlight some here, the SMART trial, if you recall, looked at this make-30 composite outcome versus the others looked at mortality. So perhaps that might be the reason why we're not seeing a signal with these other trials. But again, keep in mind that that study was huge, the SMART trial. So it was able to see a statistically significant difference in that outcome. And again, it wasn't just looking at mortality. The SMART trial was unblinded, and there was a low baseline mortality. The patients in general were less sick than was expected. And then in the PLOS and BASICS trial, there was significant crossover, and this actually had to do with prior to randomization. The randomization time that was allowed from the time of admission was fairly long. So they could have received either fluid during that period of time, and in fact, that's what happened. So it's possible that that's why we're seeing a washout of the potential benefit we might have seen had we not had that. And then BASICS, one can argue that it was also underpowered. So they expected to have a mortality of 35% in their control group, when they actually had, I think, about 26.2%. So because of that, they may not have been adequately powered based on their power calculations to see the outcome that they wanted to see. Now, might we have had a consistent benefit if we had the right subpopulation? And I'm going to focus a bit on secondary analysis that was done for the SMART trial by Brown and colleagues a couple years ago that showed that if you look specifically at the septic shock patients, you actually do see a signal here, a statistically significant one, and not just statistically significant, potentially clinically significant as well. We see a decrease of almost 5%. And another question, might we have seen consistent benefit if there was minimal crossover, as I mentioned earlier? So I'm going to switch gears here now and talk about the other drug that we like, or drugs that we like to use in resuscitation. So the second heroes, if you will, of early resuscitation in shock states are vasoactive agents. Now, a question could be, are vasopressors all the same or interchangeable? In some respects, maybe, perhaps, based on some of these landmark trials that have come out over the years. However, one exception to that may be septic shock, where in one meta-analysis by Dan DeBacker, we see that there is a higher mortality with dopamine in use of septic shock compared to norepinephrine. And presumably that might be due to higher rate of atrial dysrhythmias. So how do we select the best drug otherwise? Outside of the garden variety shock, and after we use a first-line agent for septic shock like norepinephrine, what might the second-line agent be that we pull from our Pyxis or whatever you use? So we have an assortment of drugs to choose from, but how do we know which one is the appropriate one to choose? So there's this concept of vasopressor responsiveness in the same way that we have about flu responsiveness. That is to say, if we give a vasopressor, is it actually going to improve what we want it to improve? And our surrogate that we use would be MAP. Now, this is data that came from a secondary analysis of the ATHOS-3 trial looking at angiotensin 2, and we see here that time is money. If we have people who are vasopressor non-responders, we see an increase in mortality in that group. So perhaps we should be focusing on that. And by the way, this extrapolates to other drugs like vasopressin. So perhaps we should be focusing on this concept of vasopressor responsiveness. The conclusion of that study was pretty provocative, and they talk about perhaps we should have a paradigm shift of looking at broad-spectrum vasopressor use, considering the fact that if we put patients on a drug that they're not responding to, then we may see untoward consequences later on. Now, once we have a vasopressor that we choose to use, what should be our MAP target? Has anyone ever thought to themselves, why MAP of 65? Can we set lower goals to minimize vasopressors and untoward outcomes? So the 65 trial is one such trial that looked at this more recently. And the question here is, if we look at a lower MAP goal, how could we potentially reduce 90-day mortality because of all the untoward consequences that we have with the use of vasopressors otherwise? So this was a pragmatic, open, multi-center, parallel-group, randomized controlled trial. Very a mouthful. It was based in the UK, and it was also unblinded, and their primary outcome was, again, 90-day mortality. And to make a long story short, there was no difference in mortality. However, a key note here is that it was not inferior. So in other words, even though we didn't see a benefit to it, we certainly didn't see a harm. And on top of that, in the patients who were randomized to that arm, we saw that vasopressors were taken off five hours earlier, again, with no untoward consequences. Even though we don't see a mortality benefit, perhaps that might have some impact on ICU workflow. So the take-home points from my talk and conclusion is that individualized, protolyzed care and careful phenotyping may help us optimize food resuscitation timing and type, respectively. Understanding this concept of vasopressor responsiveness may actually help us improve the time it takes to get our patients to their MAP goal, and again, hopefully improve outcomes. At least in the elderly population, a MAP goal of 60 may be a safe goal, even if we do not see an improvement in patient outcomes. And with that, I thank you for your time. Again, good morning to those of you who have joined us during the presentation. Please do feel welcome. Please come in. This is meant to be a little interactive. I'm going to encourage, although the last presentation was a little longer than planned, we actually do pause here, just because the flow may value a couple of direct questions of clarification, and it also gives us an opportunity to recognize and thank our colleague here for an excellent presentation on really some key points. So Dr. Holder's presentation open for a couple of quick questions or requests for clarification. Okay. Dr. Holder, you should take that as a sincere compliment of absolute clarity. Or everyone's just super confused, and they don't know. Well, I doubt that. May I borrow your clicker? It's up here. Oh, it's up here, yeah. Folks, my name is Dr. Ivor Douglas from the University of Colorado in Denver Health. I'd like to just start off, before I talk about the content, just to get a sense of who's with us this morning. For those present, whom of you are bedside or critical care pharmacists? Okay. Whom of you are critical care nurses, APPs? Very good. And are you folks involved in the direct care and resuscitation of critically ill patients on a regular basis? Yeah. Whom of you are clinicians, by which I mean MDs, other clinicians? Good. And amongst you, whom are anesthesia critical care, EM critical care, pulmonary critical care? Okay. Good smatching. Trauma critical care, out of interest? No. And then, obviously, some very important folks that have joined us. So from industry representatives, a few of you. Who did I leave out? Raise your hands. I think we've got everybody. So I really need to reiterate that, with a diverse audience like this, it's imperative that we don't just sit and accept. I really do need us to challenge some of the paradigms. We've heard Dr. Holden give us a really outstanding overview. It'd be nice if my advancer worked, right? It's down. Andre, I'm going to go in advance manually. Thank you. I trust you, but not that much. Okay. Let me see if I can. Yeah, there we go. Okay. We got it. Okay. So important areas of alignments and conflicts, funding to my institution from the NIH, DOD, and Baxter for work ongoing and published. There we go. Dr. Holden has highlighted to us that the paradigm that we exist in is focused on macro-circulatory resuscitation, whether it be with a variety of crystalloid-type fluids, selections of vasopressor agents. The targets of our resuscitation early in shock states, and particularly sepsis, are to address this perturbed vasodilatory response, transvascular leak, relative circulatory defilling, and the almost perturbed counterbalance regulatory effect that both volume resuscitation and vasopressor administration may actually amplify the cyclic microvascular injury that occurs in sepsis. So even though we are working to fix the big numbers, macrocirculation, we have an ongoing challenge in addressing the microcirculation. Dr. Holden alluded to a group of recent studies looking at the fluid-restrictive, fluid-liberal paradigm. My intention here is not to repeat that, but to put in context that the question of which is more beneficial, even before the publication recently of the Clovis trial, really suggested equipoise around whether more restrictive strategies or more what I would term liberal strategies resulted in differences in survival. Dr. Holden highlighted the publication of the Clovis trial that demonstrated no difference in 90-day survival, but it was a highly informative study. Let me reiterate that. A neutral outcome, well-powered study that's highly informative because it tells us a lot about our ability to drive differences in our own protocol. Importantly, despite large differences in fluid administration and vasopressor hours, important patient-sensitive outcomes and organ failure-free days were no different. And this leaves the field with a very challenging question, which is, where next? Is there anything more to be done to optimize the circulation that could potentially improve outcomes? And we are really here at an inflection point. Clearly, I would argue that vasopressor-centric fluid restrictive resuscitation strategies are probably equally effective and safe that fluid bolus administration in both the liberal and restrictive arms of, for example, Clovis, were given agnostic to the fluid responsive state. Now, let me say that again. There are patients in all of these trials that you've heard about that had a low blood pressure who either administered fluids when in more than half the cases it's likely they were unlikely to benefit from fluids or were administered vasopressors when there was no clear evidence that there was residual systematic vascular resistance reserve that could be recruited either on the venous or arterial side. And lastly, we acknowledge that while hemodynamic monitoring was permitted in these studies, they were very infrequently used and they weren't organized. At the end of the day, we may conclude that whether we pursue a fluid more liberal or fluid more restrictive approach, there is a prevailing question about alignment. And that is that administering bolus fluids and maybe presses in a fashion that is agnostic to the circulation may well mean that we are exposing patients to unneeded treatments and unnecessarily unbalanced harms. Let's look at this. In the response to the question, will my patient respond to a fluid bolus? Numerous studies show that in people who are in hypertensive or shock states, fewer than 50% of the time do people have a fluid responsive state, and I'll describe to you what that is in a minute. Measuring fluid responsiveness is different than measuring preload, importantly preload, a valuable marker of venous capacitance and congestion, if you need say, is not the same as asking whether circulation can be responsive to a challenge. Numerous patients with high CVPs continue to have a fluid responsive state, and when you look across large numbers of studies that seek to address whether a static measure like CVP is informative of whether there is recruitable stroke volume, again, no better than a flip of the coin, the area under the receiver operating curve is around 56%. So important that in answering the question, can we make a more informed approach to the way we titrate care in either a background restrictive or fluid liberal approach, we need to understand that it's insufficient to use static measures of hemodynamic circumstance. To appropriately predict fluid responsiveness then, what we really do need is strategies that more rationally align, I'm sorry, align between changes in the circulation and dependent variables like stroke volume. Resuscitation strategies that target only static hemodynamic endpoints such as MAP or CVP generally result in no difference than protocolized care. And so I'm going to argue to you based on some data from a completed clinical trial and ongoing research that we need to layer on top of these domains of fluid restrictive or more liberal approaches a less treatment naive and less agnostic approach as to the responsiveness of the circulation. To do this work, we have leveraged an important piece of reserve physiology in the circulation, which acknowledges that the unstressed volume of the circulation can be primed, preload, either with a bolus, a passive leg raise, or microbolus trending therapy. And if stroke volume augments more than a pre-specified 10%, we term this state a fluid responsive state. This does not mandate fluid administration. It does characterize the state of circulation. And by contrast, a less than 10% response infers that the end diastolic volumes are higher, there is less reserve volume in the circulation, and there's less augmentation of stroke volume in response. Targeting this dependency on this curvilinear relationship might augment hemodynamics if we then administer a bolus. Now, this approach is not well supported by even the current versions of the critical care guidelines. And my contention is it's time to update the guidelines. How you measure stroke volume is really a matter of regional preference. But in the work I'm going to show you, we have very much anchored and focused our attention on a non-invasive approach that allows for very rapid, highly accurate, and reproducible estimates of stroke volume change using the Starling Nikon device that relies on an electro-volted box created by these four double sets of electrodes and converts a flow signal into an electrical signal that can be then integrated in near real time using a version of the Doppler shift phenomenon. Compressive waves, the ambulance coming towards the ER, wee-wah, wee-wah, wee-wah. Your ER is on divert like mine is almost all the time. The ambulance goes to the other hospital, wee-wah. That Doppler shift has some correlates that can be derived mathematically, and you can look at changes in the amplitude or the frequency, and the device that we've used has leveraged the frequency shift to look at the velocity change and integrates heart rate to estimate stroke volume in a very sophisticated algorithm. I'm giving it to you as basically as I can. This has been extensively validated in anesthesia operating rooms, acute resuscitation, pulmonary hypertension, and is very sensitive to changes in response to passive leg raise where blood flow increases or a volume infusion across numerous cardiac and respiratory cycles. Here's Carlos, our volunteer research coordinator who was voluntold that he was going to have a PLR, and so he comes in the morning, we do a PLR. This is Carlos' baseline stroke volume. This is his recruited stroke volume. It's gone up 25%. I told you that's fluid responsive. Are we worried about his prognosis? Not at all. Why we're not worried about his prognosis, he doesn't have tissue hypoxia or tissue hyperperfusion. He doesn't have a likely infection. He just didn't drink his coffee, and so we bought him a triple grande latte with a cream, a frou-frou, and his stroke volume responsiveness went back down to normal. There's a lot of other really interesting data that comes out of it, and our group currently is very focused on measures of thoracic fluid content that correlates with resuscitation. We looked at this approach in a prospective randomized control trial that's been published, and I won't expend too much time talking about it in terms of design, because you can read the details, but this was a two-to-one randomization study that was informed very much by the work of our colleagues at Kansas University. Keith Latham was the PI for that, that in a pre-post model had identified some really compelling and interesting findings about using a protocol that was leveraged off fluid responsiveness to guide resuscitation, and so we set up a two-to-one randomization protocol. We allocated patients through that if they had septic shock, were enrolled early on in the ER or ICU, had received no more than three liters of fluid, had persistent hypertension, and had none of these significant and potentially problematic contraindications. The patients were either allocated to usual care or to a PLR-guided protocol, and in that protocol, if they had a stroke volume change of more than 10% and remained hypotensive, a 500 mil bolus was administered. We strongly recommended LR, and if they remained persistently hypotensive after that cycle, they went back to the top of the algorithm. At any point, if they were no longer fluid responsive but remained hypotensive, vasopressors were prioritized in the other limb of the protocol, and you can see that there were many important escape clauses from the protocol for patient safety. These were your run-of-the-mill, critically ill, septic patients with pneumonia. We had a female-male gender imbalance that was not statistically important in adjusted sensitivity analysis, and the primary findings in this pilot were to really ascertain if we could achieve separation, and there was huge separation in 72-hour fluid balance, not just fluid administered, highly statistically significant in the order of a liter and a half at 72 hours. There was a reduction by two-thirds in the likelihood of progressing to renal replacement for renal failure, a halving in the probability of the need for mechanical ventilation, and those data very much echoed the findings from the pre-post study of our colleague, Dr. Latham. There was numeric differences in the right direction for length of stay and discharge home, but this study was underpowered for that observation. Indeed, as the data from these approaches has been accumulating, there's an interesting accumulation or aggregation of mortality benefit from several trials, now in about 460 patients. Point estimate is to the favor of passive leg raise guided dynamic resuscitation, but we still lack the power from these studies to be definitive about a patient-sensitive outcome. We've learned several important additional things from these studies. I'm gonna touch on just two of them in the interest of time. The first being that patients do not remain in a static state of fluid responsiveness. So while initially more than half of patients have a fluid responsiveness state, that decreases quite quickly in the first six to 12 hours, but by the second 12 hours of care, you see that many patients return to being fluid responsive and understanding the dynamic evolution of the state of the circulation is rather crucial in these studies. I'm gonna skip through that just in the interest of parsimony and time, share with you the idea that our approach currently then would be to really acknowledge the dynamic assessment which can result in low-net fluid balance and reduce organ failures, when coupled with a rational approach for fluid bolus administration in a fluid restrictive regime, which is what I believe we can now safely and comfortably infer as a new standard for primary care or primary resuscitation. Really now we could argue should be augmented with a protocol that leverages this approach for a more thoughtful, physiologically guided and personalized approach for administration of fluid. This is an algorithm derived from what I've shown you. We're in the process of trying to set up a big randomized control trial with mortality as the endpoint. You'll recognize the same kinds of triage and endpoints around patient decisions for hyperperfusion. And this protocol is available through clinicaltrials.glob. Finally, I will suggest to you that our work is far from done. We've spent a lot of time understanding how to resuscitate in a average treatment response, restrictive, more liberal. We've understood that agnostic approaches to how you give fluid can result in misaligned bolus treatment. We really should be understanding whether the circulation remains fluid responsive. And then finally, I'll leave with you this notion that we need to really get our arms around the venous capacitance side in a more formal fashion. So our group is adding on and has current studies funded, in fact, by Baxter, looking at some discovery science approaches to evaluating venous capacitance and overfilling using bedside point-of-care ultrasound. We're gonna be looking forward to presenting that work. In addition to our research group at Denver Health and the University of Colorado, my collaborators in the PETL network, the fresh investigators, Dr. Holder, thank you for your collaboration, Dr. Martin here. I really wanna spend a second to acknowledge my collaborators and supporters at Baxter who have been phenomenal scholars through the discovery science and continue to do that. Now you know why they call it the ICU. I've long believed that you've got to believe in what you do, and I thank you for your time. With that in mind, I am going to skip taking questions in the interest of parsimony and time and pass the mic to my colleague. So Dr. Latham is going to review with us the last part of this presentation using some clinical examples to really dig into some of this data. Dr. Latham. Thank you. Yeah, so when we were discussing how to go about doing this, our presentation certainly had great work being presented by my colleagues Dr. Douglas, Dr. Holder, and Lee. You got it? Is that better? There we go. Thank you. That, how about we discuss some cases that can represent kind of basically what we're talking about here. So I put together four cases that represent resuscitation or shock patients in which we were utilizing stroke volume to really assess and guide our therapy for the patients. I do have a disclosure for getting honorary from Baxter for talks or consulting. These are the four kind of patients and cases that we will discuss. All these patients I have taken care of in the last couple of years, some more recent than others. And when I say fluid bolus, I will just say in our institution, that's almost exclusively LR in our ED and our ICU, medical ICU. And when we use our stroke volume assessment, as Dr. Douglas alluded, our institution and our medical ICU, we are using the non-invasive Starling device. So AR, 72 year old, 80 kilo lady, basically comes to us from her skilled nursing facility. One day history of altered mental status and fevers. She's hypotensive on presentation with evidence of a UTI and indwelling Foley. History of known ischemic cardiomyopathy, EF down in the 20% range. Also chronic renal failure that complicates her course. Baseline creatinine's low twos, but at her presentation, she's three and a half. She's given a 500 mil bolus in the ED, given appropriate antibiotics based on her prior culture data, and they call us for admission. After that fluid, she remains febrile, hypotensive, tachycardic, tachypneic. She requires six liters of oxygen, that's above her baseline of two. They've done a POCUS that confirms she still has this reduced LV function. No pericardial pleural effusions, no B-lines, no pneumothorax as contributing factors to her shock. And basically they also say, oh, there's a dilated IBC. We're not gonna give her any more fluids. We don't have to do so anymore because of the new guidelines. We can just document that she doesn't need the 30 mils per kg, and they'd like to send her up to us. We accept her, she comes up to us, and we wanna make sure that she doesn't actually need more fluid. So we do our stroke volume assessment. Because she is a large lady, she also has already an sarca, we go ahead and start with just a passive leg raise as opposed to giving her more fluid by chance if she actually doesn't need it because of her known cardiomyopathy. That passive leg raise reveals a change of improvement in her stroke volume index by 15%. So then we do give her a 500 mil bolus. Also reveals that she improves her stroke volume by 16%. We give her another 500, and she improves that by 7%. So she is now, in our term, fluid refractory, and so we're not gonna give her any more. However, she is still hypotensive, so now we initiate norepi, and we increase her stroke volume, or, excuse me, titrate her norepi to a MAP of greater than 65, and that took about 0.08 mics per kilo per minute of norepi. We're very happy with that. Her urine output has picked up at this point. Generally, she's mentating a little bit better, and so we're like, we're on the right track. However, about three hours later, our nursing colleagues contact our team, saying we've had to titrate up her norepi to about 0.15. This is a significant increase, so our question is, well, is she becoming a little more bacteremic, and maybe a little more vasodilatory at this time? Does she have an appropriate stroke volume? So again, we're going to reassess that. So anytime we have a change in perfusion, whether that's increased need for our pressors, maybe it's a rising lactate, change in mental status, maybe a drop-off in urinary output, which she did actually have a drop-off in her urinary output, and her lactate was not clearing as it had been. So we reassessed her stroke volume for responsiveness. We did a passive leg raise again, and now she's 20% responsive to a passive leg raise. So we gave her a 500-mil bolus. She was 22% responsive, and we continued to do that, as you see there, for a total of two liters until she wasn't fluid responsive again. So this is a lady with severe cardiomyopathy at baseline. Nobody in their right mind would have given her three and a half liters of fluid over the course of a six-hour timeframe, myself included, if you're just doing the thumb check. And I think this is reflective of what we've been talking here about these kind of Clovers and classic trial. These patients get randomized to both of these. Some of them probably needed a little more fluid. Some of them probably didn't need any fluid. But which group did they get randomized to? We don't know, because there was no guided resuscitation protocols in either of these trials. So then we titrated her norepi. It went back down to 0.06. Over the course of this 24-hour period of her hospital stay, she was off norepi. Her creatinine was trending back towards her baseline. She was back down to her normal two liters. In the course of about a week, week and a half, she was discharged back home without any worsening renal failure, et cetera, et cetera. This next patient, a common patient that we see in our ICU. We have a relatively large liver transplant center, so we see a lot of cirrhotics. This is 53-year-old woman, end-stage alcoholic cirrhosis, complicated by tensocytes, requiring routine paracentesis, because she's not a TIPS candidate. She presented a two-day history of progressive encephalopathy and weakness. Commonly in this patient population, this represents either infection or profound maybe volume depletion from poor oral intake. In the ED, she's hypothermic, only Ornian did to herself, and just some generalized abdominal discomfort. She's tachy and hypotensive, as you see there. She has some significant metabolic disturbances, as you see by her labs, including a lactic acidosis. She actually was given 30 mils per kg of LR in our ED by our colleagues and appropriate antibiotics. Post-fluid, she's still relatively tachycardic and hypotensive, but a little better than her presentation. And a paracentesis did reveal that she has a stalled blood pressure, excuse me, SBP. In our institution, we do follow, given her the 1.5 grams per kilo of albumin to help improve renal perfusion in hopes of reducing renal failure and adverse outcomes in this patient population with SBP. After that, she came up to us with still a borderline blood pressure, maps of 57. Again, in this population with perhaps renal failure, I will say that creatinine of one and a half was greater than two times her baseline. Her baseline creatinine's 0.05 to 0.06. And so that, in essence, is renal failure. And so we initiated her, the decision was, well, does she need more fluids or does she need pressors to try and get that map closer to 70 in this patient population with cirrhosis and renal dysfunction? Her urine lights also said that she was either volume deplete or in hepatorenal with a low urine sodium. And so we did a stroke volume assessment, but we couldn't really do a passive leg raise. So in this patient population, very tense ascites, that's not the person that you wanna do a passive leg raise for. Very similar to our pregnant patients, or if somebody has an abdominal compartment syndrome or concern for that, you don't wanna do PLRs. There are other contraindications, just profound hemodynamic compromise or respiratory compromise in the uninhibited. So in this particular patient, we didn't do a PLR, but when we hooked her up to the device, her SVI was actually pretty high, it was 58. So generally speaking, if somebody's fluid down, their SVIs tend to be on the low side, low 30s, high 20s, et cetera. And so we looked at that and we're like, she's probably not fluid down with an SVI that high. So we just initiated her on norepi alone. And with doing so, we got her maps right up to around 70. That was our target. And actually, her SVI stayed right there around 60. So it really didn't change all that much. Again, I think her hypotension was more in the distributive, our serotics. They just have a vasodilatory chronic state. And she just needed a little vasoconstriction for her. She didn't need more fluids. And so we avoided further fluid in a lady who was already just chronic hypervolemic, as evidenced by hyponatremia, et cetera, et cetera. Again, over the course of the next 24 hours, renal function improved, hemodynamics improved. She was off norepi over the course of a 24-hour period and overall returned to her previous baseline. In this patient, next one is kind of this undifferentiated shock that we saw here just before Christmas a month ago. And so CA, he's a 32-year-old man, really no significant past medical history. He'd been in an outside hospital for about three days where he was just progressively declining in terms of renal function, was getting worse, more hypoxemic and more shocky. He arrived to us and basically required immediate intubation due to altered mental status and hypoxemia and was in profound two-presser shock and in a very severe metabolic acidosis that required immediate placement of an HD catheter, an emergent CRT start. He had profound lactic acidosis. You can see his vitals there. He had generalized anusarca. It wasn't profound, but it was present and just generalized warm extremities. So part of you is saying, well, he seems distributive, yet nobody had found a real cause for infection or other cause of a distributive shock like overdoses or drugs, et cetera. Given the fact, you know, we didn't know if he was fluid responsive. We did do an assessment. His SVI pre-PLR was pretty low at 25 and actually post-PLR, it declined. It actually had a negative deflection and declined by 12%. There can be a correlation there between a negative deflection after a passive leg raise or fluid challenge that that may correlate with a cardiac dysfunction. Now, as you can imagine, multiple things were all going along at the same time of doing this resuscitation. This patient got POCUS. We could see that he had LV and RV dysfunction. He appeared fluid up as well, and so he did definitely not need any more fluid. A formal echo demonstrated dilated cardiomyopathy, EF less than 20%. Within five hours, he's in this five-presser shock, and in all honesty, we didn't know if he was gonna make it through the night. We still did not know what was going on and tried to think through what was going on. His family did say, oh, he's just kind of, you know, he maybe drinks a little bit more than he should, and actually, we got a call by his roommate who actually said he actually drinks probably a liter to a liter and a half of liquor a day, so we gave him a high dose of thiamine, and we believe he had a severe kind of cardiac beriberi, and within 12 hours, he was down to two pressers. This gentleman left the hospital to rehab within a week. I will say, one of the most challenging cases I've seen in my going on more years than you might think, and one of the best saves I could say as well. Thiamine. Might as well give it to about anybody if they're in multi-presser shock because it might literally save their lives. And the last patient here I'll say is, basically, it's a guy that presents with hemorrhagic shock. He had known history of gastric malignancy, came in with a big GI bleed, ended up needing massive transfusion protocol and pressers, ventilated, et cetera, to get him through. Our GI colleagues did find the source and got that fixed, but as oftentimes we run into in the kind of next 24-hour period, these patients are kind of in this persistent kind of shock state. I think it's more of just a cytokine problem, maybe a little stunned myocardium, even though if you do echoes, it looks like they function fine. And so the next morning, his AM hemoglobin was about 9.2, maybe a little down from his post-resuscitation from massive transfusion. His vitals didn't look too bad. He was still on a little bit of norepi, but less than the day prior. By the noon check, his hemoglobin had dropped by a whole gram. He still was on the 0.15. Question is, well, is he re-bleeding? He's certainly high risk for a re-bleed. So we did assess for stroke volume, and we did that. He really didn't change his stroke volume index, so he wasn't really bleeding. It's just a little cytokine, and we just need to let that run its course. And over the course of the next 24 hours, his presser came off, he was extubated, and he had complete coverage. So basically, just to sum up what we've kind of said, stroke volume guided resuscitation, I do believe it has clear benefits in the initial resuscitation of shock, but it also has benefits in the reassessment of patients during the course of their septic shock, or other shocks. And it can have significant guidance in multiple complex shock states, as we kind of displayed here. And ultimately, I think a guided approach will allow us to avoid fluid overload. And thank you very much. Thank you. Thank you. Please join me in thanking both Drs. Latham and Dr. Holder for both provocative and informative discussions. The strength of the cases you've presented really is that it gives you a diversity of clinical scenarios for application. And I think the challenge is that folks often don't think of monitoring guided resuscitation in a bunch of these conditions. If I could just briefly ask, what has been the cultural evolution in your institution to think about early non-invasive hemodynamic monitoring, please? Yeah, so we kind of started this back in 20, I think 14, 2016. And the evolution really came in seeing that we were using less pressors, less intubation, less development of renal failure, when we were doing these guided resuscitative efforts for our shock patients. And then when we did our retrospective analysis, and it actually panned out to what people thought was going on, it was very easy to convince everybody involved that this was probably the right thing to do. And so it took a few years, but now it's a very common phenomenon. Thank you very much. Question? Just going back to the first case there. In the Seattle, sorry. Just going back to the first case, I saw that you needed to do multiple assessments of the patient in terms of fluid responsiveness. It was not the case, but do you use the device continuously as well to titrate, for example, an inotrope and to monitor the cardiac index of the patients in this kind of patient, for example? So generally speaking, we don't use it on a continuous basis for the most part. We only have one device per unit or a couple of units. We always, once used, we always leave them on for a few days. But in this particular case, it was kind of on, we did them, patient wasn't fluid responsive anymore, pressors restarted, it was disconnected, it went on to be used somewhere else, cleaned, yada, yada. But in some cases, more on the cardiogenic shock side of things, we have placed the device in place and then used it to see what the changes in our SVI, cardiac index, et cetera, were with titration of pressors. I do want to point out to the audience and certainly happy to discuss it, there is a large registry study going on right now to really capture the spectrum of usual care deployment of non-invasive hemodynamic monitoring. And we'd be happy to share, we actually have some upcoming presentations from that analysis. And I do think it's going to get to the point about you asked, which is, is it used in a continuous fashion? Is it used in a intermittent, periodic fashion? And I think that is going to be super informative. Let's take another question before we wrap up. Folks, I'm very, very delighted that you all came and contributed your ideas. I know that my colleagues will join me in hanging around for a little while if you would like to discuss any other aspects. I do wish you a most valuable and enjoyable remainder of your time in San Francisco. Thank you.

non-invasive hemodynamic monitoring

stroke volume assessment

resuscitation

shock states

fluid responsiveness

fluid administration

case studies

guided resuscitation

patient outcomes

management