Hello and welcome to today's Journal Club Critical Care Medicine webcast. The webcast hosted and supported by the Society of Critical Care Medicine is part of the Journal Club Critical Care Medicine series. In today's webcast, we feature two articles from the October issue of Critical Care Medicine. The webcast will be available to registrants on demand within five business days. Please log into mysccm.org and navigate to the My Learning tab. My name is Tony Gerlach and I'm a clinical pharmacist at The Ohio State University Medical Center here in Columbus, Ohio, and I will be the moderator of today's webcast. Thank you for joining us, but just a few housekeeping items before we get started. First, during the presentation, you'll have the opportunity to participate in several interactive polls. When you see a poll, simply click the bubble next to your choice. Second, there'll be a question and answer session at the conclusion of the presentation. Please submit questions throughout the presentation and just type into the question box located on your control panel. Third, if you have a comment to share during the presentation, you may use the question box for that as well. And finally, everyone joining us today for the webcast will receive a follow-up email that will include an evaluation. Please take five minutes or so to complete the evaluation. Your feedback is greatly appreciated. This presentation is for educational purposes only. The material presented is intended to represent an approach, view, statement, or opinion of the presenter, which may be helpful to others. The views and opinions expressed herein are those of the presenters and do not necessarily reflect the opinions or views of SCCM. SCCM does not recommend or endorse any specific test, physician, product, procedure, opinion, or other information that may be mentioned. And now I'd like to introduce today's two presenters. Dr. Tina Chen is an attending physician of critical care medicine at Montfior Medical Center and an assistant professor of clinical medicine at the Albert Einstein College of Medicine. Since coming to Montfior in 2015, her research interest has been primarily on volume resuscitation in patients with shock and acute respiratory failure, specifically acute respiratory distress syndrome. She obtained her master's in clinical research methods at the Albert Einstein College of Medicine to further pursue her career in clinical research. Since working on volume chasers, she has been involved with informatic projects, harnessing the power of electronic medical records and machine learning to identify and manage patients in shock and ARDS. Next is Dr. Tom Miano, who is a critical care pharmacist at the Hospital of the University of Pennsylvania and an instructor of epidemiology at the Perlman School of Medicine. Todd received his doctor of pharmacy degree from Virginia Commonwealth University, completed postgraduate pharmacy residency at Wake Forest, and received a doctor of philosophy degree in epidemiology from the University of Pennsylvania. He is elected a fellow of the American College of Critical Care Medicine and associate editor of BMC Pharmacology and Toxicology. He practices in a high acuity surgical ICU where he optimizes pharmacotherapy as a member of the multi-professional and disciplined critical care team. Dr. Miano's research leverages electronic health record data to study drug safety and comparative effectiveness in the critically ill population. His work seeks to understand the source of drug heterogeneity in acute ill patients and quantify the impacts of variable response on outcomes. Areas of emphasis include drug-associated acute kidney injury and drug-drug interaction. Thank you both for joining us today. Before we begin, could you each tell us if you have any disclosures to know? Tina? I don't have any disclosures. Thank you very much. And Todd? I also have nothing to disclose. And now my time is on the wane, so Tina, it's your turn to be our first presenter. Great. Thank you again to SCCM and Discovery Network for letting me present the study by our volunteer study group. I'll be presenting the variation in fluid and vasopressor use in shock with and without the physiologic assessment. All right. Next slide. Okay. So, as we all know, in the past six years, there are a few studies comparing early goal directed therapy in a protocolized fashion with usual care. And all three studies, including ARISE, PROCESS, and PROMISE, they did not find a survival benefit in this protocolized management of early septic shock comparing to usual care. So, really, the question now is, what is usual care? What are we actually doing in taking care of these patients in shock? So, with the help of SCCM and Discovery Network, a group of us from SCCM, UCEDD, and CCPTM come together to form a volunteer study group. We aim to conduct a multi-center observational cohort study to determine the variability in fluid and vasopressor resuscitation in shock patients in a broad range of treatment area in the hospital, because, as we all know, shock doesn't exclusively happen in the ICU. The primary aim is to describe the variation in fluid and vasopressor use in the first 24 hours after shock onset with and without the use of physiologic assessment. Here, we define physiologic assessment as static measurements, such as central venous pressure, wedge pressure, dynamic measurements, such as stroke volume variation, pulse pressure variation, and critical care ultrasound, passive late-grave tests. And this is compared to empiric management of these patients in shock. And when we said empiric management, we're really seeing management based on clinical signs and symptoms, such as heart rate, blood pressure, physical exam, determined by the treating physician or clinician. Secondarily, we wanted to determine the association between the use of physiologic assessment and clinical outcomes, like in-hospital mortality and change in SOFA score between the first 24 hours and 48 hours. So, we included adult patients with intention to go to the ICU. So, again, part of what is unique here is that we're looking at shock that could occur anywhere else, anywhere, not only in the ICU. And we define shock as someone who needs a vasopressor to maintain normal tension or systolic blood pressure less than 90 or a MAP less than 65. We excluded patients who were previously enrolled, so one patient per enrollment. The patient who had shock in the operating room at the time were patients coming to the ICU directly after cardiac surgery, because these patients' physiology is slightly different from your usual patients. And ultimately, we did exclude patients who were transferred in from outside hospitals or outside emergency. And the reasoning for that is we may not have a complete record of what was done at the other hospital. So, we ended up having 34 sites and 1,639 patients included. Sepsis was the most common primary etiology of shock, including 60%. The following shock type was hypovolemia of 19%. Cardiogenic shock was 12%. And we included 9% of other shock types, such as neurogenic shock, anaphylactoid shock, and other non-distributive shocks. And we're looking at where shock happens. Most of the shock did happen in the ICU, comprising 49% of the location. The second location of shock was emergency department at 33%, and ward at 14%. And PACU and other location in the hospital of 2% each. I'm showing you the baseline characteristic of this cohort of the 1,639 patients. And it's quite a thick population, right? The mean age is about 60, with the cardiogenic shock patient being the oldest, and other shock type being younger. The APACHE score is, the average is 86, suggesting a sick group. Septic patients were the sickest, with the highest APACHE score average in the 90s. And hypovolemic patients had the lowest APACHE score at 78. Predominantly male, every shock type is greater than 50%. And there is a small variation in the distribution of races. In terms of this cohort outcome, a quarter of the patient with shock passed away. And this is consistent of what we found in the APACHE score, right? And the other thing is, septic patient with the highest APACHE score also had the highest mortality of almost 30%, with hypovolemia having the lowest mortality rate. Coming to the next slide, okay. So, my first polling question for this group is, when you're managing a patient in shock, what helps you decide how to resuscitate the patient? Do you, one, use clinical signs and symptoms, such as heart rate, blood pressure, and exams? Do you use a critical care ultrasound? Do you do passively grazed tests? Do you use stroke volume variation and pulse pressure variation in the dynamic measurements? Or do you use a static measurement, such as central venous pressure and wedge pressure? Okay, so the polling result, it seems like 69% of you chose clinical signs and symptoms, 23% says critical ultrasound, critical care ultrasound, and then I'm going to go back to the presentation, 8% of the patients said they will use passive late-grave tests. All right. So, this is what we found in our population. So, similar to what the poll show, 61% of the patient enrolled did not receive any physiologic assessment. I think this poll says about 69% would use empiric management. And similarly here, in our cohort, critical care ultrasound was the most common type of physiologic assessment performed when they were performed. And this is interesting because we are moving away of using these static pressures such as the CVP and wedge pressure. The other thing that was interesting to note in our finding is that the use of physiologic assessment is predicted by having higher APACHE score. Patients with cardiac dysfunction as a secondary cause of contributor of shock. So, sicker the patient is, the more likely they would use physiologic assessment in their shock management. In terms of fluid and vasopressor use in the first 24 hours, we definitely saw a lot of variation. Here, I'm showing you the cumulative fluid and cumulative vasopressor use. On your left-hand panel, this is the fluid. Blue bar represent the percent of patient receiving fluid and deep different shock type. And that orange dot and line is the medium volume of fluid. And as you can see here, both septic shock and hypovolemic patient were the most likely to get fluid and cardiogenic shock and other shock types. Although getting fluid, most of them because medication and whatever, but in their 24-hour, the volume was much less than those in septic and hypovolemic patients. The other thing that's interesting in the fluid data is that there is great variation in how much fluid someone was getting. As you could see here in the septic patient, medium volume in 24 hours is 3 liters. But if you look at the interquartile range, patients were getting about 1 liter up to 4 or 5 liters in that first 24 hours. Similarly, coming to your right-hand panel, showing the cumulative vasopressor use by shock type, the green bar represent the proportion of patient receiving vasopressor and the orange is their medium vasopressor dose and norepinephrine equivalent. Septic patients were most likely to get pressure during their first 24 hours. And that dose also varies with different shock types, with septic patient receiving the most amount of pressure. So coming to my second polling question, how do you think the use of physiologic exam impacts the fluid and vasopressor administration? Does it give you more fluid, less fluid, more vasopressor and less vasopressor, or ultimately no difference in the fluid and vasopressor use? Okay. Here, I could see the results. So, it seems like people are thinking the use of physiologic assessment is associated with more fluid and more likely for vasopressure. And a good proportion of you also think that it makes no difference on the fluid and vasopressure use. Let's go back to the slide. Great. Okay. So, I will show you what we found. So, surely, we thought that cumulative fluid was slightly higher in the group receiving physiologic assessment. So, on your left-hand panel, I'm showing it at each time period of data collection. The blue box represents empiric management, and physiologic assessment represents the group that has physiologic assessment done, the orange part. So, as you can see, at each time period, the amount of fluid administered was actually no different. But if you add them together, it achieves statistical significance at hour 12 and hour 24. When we look at the actual volume difference between the two groups at hour 12, it was only 157 cc, and at hour 24, it's about 300 cc. So, even though it's statistically significant in this univariate analysis, the difference, I think, clinically speaking, is really not that significant, not that meaningful. And indeed, we did a multivariate adjustment using random effect modeling, adjusting for patient-level severity of illness, hospital-level differences, such as ICUs, different types, and shock location, and actual hospital site differences. We saw no association between the use of physiologic assessment and cumulative fluid in the first 24 hours. On the other hand, we saw that patients who had physiologic assessment were more likely to be on vasopressor during their shock management. The dose was slightly higher as well, but it's striking here in this graph, the green represents people receiving empiric management in that 24-hour period, and orange part representing patients who have physiologic assessment. And during each time period, a significantly higher proportion of patients receiving vasopressor. The secondary aim was to look at association between physiologic assessment with in-hospital mortality, and here in the univariate analysis, we saw that the use of physiologic assessment was associated with maybe higher risk of in-hospital mortality, so with fluid and definitely in vasopressor at 3.68 times higher risk. As we adjusted patient-level differences, such as age, gender, race, Apache, SOFA, and hospital differences in terms of location of shock and what type of ICU patient was admitted to, and at different hospital level, this relationship disappears. So after adjusting for these differences, the use of physiologic assessment was not associated with in-hospital mortality, in fact, it was not associated with cumulative fluid in 24 hours, and it was still associated with the use of vasopressor. This is a graph kind of representing how the differences between center to center. Your x-axis is the different hospital sites. Blue bar represents in-hospital mortality, and this is sorted on in-hospital mortality percentage, and then the orange bar is physiologic assessment. In our random effect modeling, we felt that there was significant variation between hospitals if physiologic assessment was used. In terms of our strength and limitation, this is a study that's documentation-based, so the quality of physiologic assessment and the direct impact on decision-making could be lacking in our method. The overall frequency of physiologic assessment use was way less than expected. In terms of strength, this is a real-world practice in shock resuscitation for this observational study. We included a multitude of shock location and ICU types, and this is a large cohort with a very collaborative effort through Discovery Network and SCCM. The takeaway is that physiologic assessment is not as commonly used as we thought. When used, critical care ultrasound is definitely the most frequent method. We're moving away of using the pressure-based system. Physiologic assessment was not associated with how much fluid patients received, but it did guide the use of vasopressor. We thought that physiologic assessment was more frequently used on sicker patients. I want to say thank you to all my co-authors and all the administrative support I've received from mentors and people we're working with. It's an amazing effort of 34 sites with all the PIs, coordinators, residents, and students. Without Discovery Network and the efforts of the Society of Critical Care Medicine, we really wouldn't have been able to get this study done in such a short time. At this point, I'm going to turn the attention to Todd. Thank you, Tina. That was a really fantastic presentation highlighting key aspects of shock resuscitation monitoring and treatment. I hope to add to that with a deeper dive into the use of vasopressors in patients with a diagnosis of septic shock. I'd like to start by just saying how lucky I feel to be part of this research team. I think the volume chaser cohort is a great example of how the benefits of multidisciplinary collaboration that have long been observed at the bedside can also translate into the research domain. Our research questions focus broadly on septic shock resuscitation. Fluid administration and vasopressor therapy are obvious cornerstones of shock resuscitation. However, there remains substantial uncertainty in terms of what is the right balance between these two therapies. Since the publication of the first early goal-directed therapy nearly two decades ago, aggressive fluid resuscitation has been the standard of care. We give fluids to increase preload, cardiac output, and hopefully tissue perfusion. And we often give quite a lot, as Tina described. However, a growing body of evidence has described the potential adverse events and complications associated with overly aggressive fluid resuscitation. Multiple studies have linked fluid overload with higher morbidity and mortality through a variety of potential mechanisms. In light of this data, there's been growing interest in approaches to minimizing the amount of fluid that we give to these patients, which often implies earlier and more aggressive use of vasopressor therapy. We give vasopressors to augment afterload, cardiac output, MAP, and hopefully tissue perfusion. But excessive vasopressor exposure can also mediate important adverse effects, myocardial injury, ischemia of important tissue beds, and arrhythmias. So we are kind of confronted with the common rock-and-the-hard-place dilemma that we face at the bedside. In a patient who has refractory hypertension, do we give more fluids, or do we give more oppressors? What is perhaps less understood are the potential interdependent effects of these two treatments, meaning, for example, that the effect of vasopressors on outcome may depend on the volume status of the patient at the time of administration. So that the balance between beneficial hemodynamic effects and adverse effects depends on an interaction between these two therapies. With these ideas in mind, our aims were to first determine the association between vasopressor dose intensity in the first six hours and first 24 hours after septic shock and 30-day in-hospital mortality. We also determined whether the effect of vasopressor dose intensity varies depending on volume status, and whether this association varies depending on the pattern of dosing titration. This was a prospectively defined sub-study of the volume chaser cohort that Tina just described. Of the 34 sites included in the parent study, we included 33 sites, with one site being excluded due to the criteria for entry into our cohort. We restricted the parent cohort to patients who had septic shock and received vasopressors within 24 hours of shock onset. Sepsis diagnosis was made in accordance with the surviving sepsis campaign guidelines surviving sepsis campaign guideline definitions, and shock was defined as Tina previously described. We excluded patients who received inappropriate initial antibiotics, those who received mechanical support of any kind, and patients who had missing data. This is the timeline for data collection, which is important in terms of understanding our methods. The X signifies the onset of septic shock. We defined a 24-hour period around this time beginning 12 hours before shock onset for the collection of data required to estimate APACHE-3 scores. And then the 24-hour period after shock onset was divided up into four time periods. 0 to 3 hours, 3 to 6 hours, 6 to 12, and 12 to 24. Within each one of these time periods, we collected the total amount of fluids given, the minimum vasopressor dose, and maximum dose, and whether patients were mechanically ventilated during each time period. Using this data, we estimated vasopressor dose intensity, which was the total vasopressor dose in norepinephrine equivalents expressed in micrograms per minute. We calculated this by first averaging the lowest and highest dose of each vasopressor for each time period, converting to norepinephrine equivalents, and then estimating the cumulative average over 0 to 6 hours and 0 to 24 hours by taking a time-weighted average over the corresponding time periods. We also collected the cumulative amount of fluids, again, for the time periods of 0 to 6 hours and 0 to 24 hours. Lastly, we created titration categories as follows. We first defined high vasopressor exposure, or dose, as at least 15 micrograms per minute based on the threshold used in the VAST study, and then we created the following categories. Those who were always low, meaning that during the 0 to 6 time period and the full 24-hour time period, their dose was always below this threshold. Those who were early high, late low, meaning during the first six hours, they were titrated to above 15 micrograms per minute but were subsequently titrated down by the end of the 24-hour period. Those who had low dose exposure during the first six hours with subsequent titration to above 15 by the end of the 24-hour period, and then those who were on high dose early and had that dose sustained over the entire 24-hour period. We associated our exposures with outcome using multivariable logistic regression models. We constructed separate models for the 0 to 6 and 0 to 24-hour time period, and in each model, we tested for interaction between vasopressor dose intensity and fluid administration. By including these interaction terms, this allows the association between vasopressor dose intensity and outcome to vary depending on volume status. We also conducted a logistic regression model associating our titration categories within hospital mortality at 30 days. Our models were adjusted for a number of potential confounding variables, including key demographics, measures of severity of illness, and treatment characteristics, including APACHE-3, mechanical ventilation, use of corticosteroids, and key baseline comorbidities. These variables were selected a priori based on clinical judgment and in consideration of the number of outcomes that were observed in our study to avoid overfitting our models. And for our results, so we, of the 1,639 patients in the parent cohort, we included 666 patients who had septic shock and received vasopressors at any time point during the first 24 hours. After exclusion criteria, we were left with 616 patients in our analysis population. Key characteristics include an average age of about 64 years, a high severity of illness as represented by APACHE-3 and SOFA scores. Most patients were in a medical or mixed med-surg ICU, and about 45% of patients were, had shock onset in the emergency department. About 62% of patients received mechanical ventilation at some point within the first 24 hours. A quarter of patients received renal replacement therapy at any time during hospitalization, and 191 patients, or 31% of our cohort, died during the first 30 days of admission. These figures describe, or some descriptive data on vasopressor use. On the left is the percentage of patients who received various vasopressors. By far and away, norepinephrine was the most commonly used vasopressor. Over 90% of patients received this at some point within the first 24 hours, but with a meaningful number of patients who received vasopressors at any time during hospitalization, but with a meaningful number of patients also receiving a vasopressor at some time point. On the right is the number of vasopressors at each time point over the first day. Most patients were on one vasopressor at any given time. You know, during the, at the six-hour time point, there was a, you know, an increased or meaningful amount of patients who were on two or more. The average dose, in terms of norepinephrine equivalents, was about 10 micrograms per minute over their, the full 24-hour time period. This is fluid resuscitation over the first 24 hours in milliliters and milliliters per kilogram of ideal body weight. At six hours, patients received about a liter and a half, or 22 milliliters per kilogram per kilogram during that first six-hour period on average, with the number increasing to about three and a half liters by the end of the 24-hour period. In terms of our primary analysis at six hours, we observed a significant interaction between vasopressors and fluid and the association of mortality, meaning that the association between vasopressor dose and mortality varied depending on the amount of volume given to the patient over that six-hour period. In the setting of no fluid resuscitation, mortality increased sharply as vasopressor dose increases. However, with increasing volume, that association weakens, and it's no longer significant at about two liters and really negligible, you know, above two liters. So, consistent with our initial hypothesis. We did not observe interaction between vasopressors and fluid when considering the entire 24-hour period. Here, we just observed a consistent increase in the odds of mortality with increasing vasopressor dose. In our analysis of titration categories, compared to those who never received high dose, those who received brief early high dose with subsequent down titration had similar mortality, which was lower than patients who received early low dose with late titration to high doses and those who received sustained high doses. To determine whether the outcome patterns across dosing categories is explained by differences in achieved map, we explored the trends in average map over time, which I show here, stratified across the different groups, showing that all patients on average achieved a map of at least 65 during each of the time periods with no significant differences between the four categories. And so, in summary, our key findings are first that the association between early vasopressor dose during the first six hours and mortality depended on fluid resuscitation volume, and secondly, that brief early high vasopressor exposure was associated with lower mortality compared to later and sustained high doses of vasopressors. Based on this, our key hypotheses coming out of this work is that early aggressive vasopressor titration may be preferred to slower titration and that there is some minimum threshold of fluids that is required in patients treated with vasopressors after the onset of septic shock to avoid adverse effects of those pressors. I would also like to thank our team, you know, the lead investigators, all the site investigators, the Society of Critical Care Medicine, and obviously the Discovery Network, without which the study certainly would not have been able to be completed. And with that, I will turn it over to Tony, who will moderate our question and answer session. Thank you very much, Todd. And thank you, Tina, both for a presentation. And for everyone out there, please feel free to answer or ask any question by using your question tab on your control panel. We would greatly appreciate it. For both of you, since I've seen at least in Tina's study and then obviously in the sub-study that Todd presented, there were a lot of septic patients here. And I just had a question about how did you guys assess source control and really time to appropriate antibiotics in your studies? So, Tina here, I'll take this question. So, for our data collection, we conducted multiple training sessions to make sure all of the study coordinator and site PI, the data collection methods were standardized. So, one of the questions in our survey was appropriate antibiotics dose. And this answer will be based on whatever the physician documented in their note. And that's how we collected that data. Todd, do you have anything to add? No. I mean, I think that's, yeah, that's how it was implemented. And I think, you know, the training sessions were key in trying to ensure that that was a standardized process. Perfect. Thank you very much. Now my next question kind of is a stem from that as well, is really especially because there were some people, obviously, that looked like they had steroids in here. Did you guys really look at the addition of flugelcortisone by itself, or did you just really look at the use of hydrocortisone in these patients? So as a pragmatic study, we wanted to maximize the state-up points that we could receive and how many patients that were enrolled. So one of the questions in our survey was if steroids was used during the time period. I think there is a group who is planning on doing an ancillary study looking at the use of steroids in shock. We did not collect specifically what type of steroids was used. Well, thank you very much. I think it's a fairly safe assumption that stress dose steroids, by far and away, would at least certainly include hydrocortisone. I think it would be rare that flugelcortisone by itself would be considered stress dose steroids. Very good. Also, another question out there is, looking at the morbidly obese patients, did you guys look at those people, or was that something that is potentially going to be another sub-study of the volume two search? That's exactly right. So one of the fellow working with us is going to look at the weight-based fluoresuscitation based on people's BMI. As you can see, the current guideline is suggesting 30 cc per kilo, but what is that kilo, right? Is it actual weight, adjusted weight, or ideal body weight? So there is a study using this data to assess what is the best weight to use in fluid resuscitation. We didn't account for necessarily definitions of obesity, but all of our models included as covariates, both height and weight. So in that way, an analysis of micrograms per minute of vasopressors is kind of mathematically equivalent to an analysis of a weight-based. We did some sensitivity analyses included in our appendix where we repeated analysis based on both weight-based fluids and vasopressor definitions, finding consistent results. I think your question is kind of getting at more of an effect modification, meaning that potentially does the interaction between vasopressors and fluid potentially also vary by body habitus, which I think is a really fascinating question, but was outside of the scope of our initial study. But I probably have some more blurry coming from this data. That's very exciting. Perfect. No, I think that's very exciting that you're getting multiple studies out of here. And I do have a question, too, is just to clarify, and this one is more for Todd, about clarifying high-dose vasopressors and volume resuscitation early on, and specifically, how high is high-dose vasopressors? And it might be something, too, I'm not saying I'm not, or if I understand, and my institutionally does norepi per kilo, so sometimes going 10 mics per minute, I have to do the math in my head. So I might put you on the spot, Todd. Yeah, no, that's a great question. And it's funny because we dose by micrograms per minute, so I'm always doing the same math in my head going back the other way. So we chose as high-dose, I mean, that's kind of inherently an arbitrary threshold. And so we chose 15 just because, you know, there at least is some precedent for that number. And we also, in our weight-based analysis, did a similar kind of high-dose analysis where we define that as 0.2 micrograms per kilogram per minute, and that's consistent with the ATHOS-3 study and the threshold that they used. But it turns out, actually, if you consider the average weight in our population, that 15 mics per minute and 0.2 per kilo on average coincide precisely. But those thresholds are obviously arbitrary. But in our primary analysis, we analyzed vasopressor dose and fluid administration as continuous variables. So we didn't use any arbitrarily defined thresholds in our primary analysis. We just let the data kind of speak for themselves. Because of the uncertainty in trying to set thresholds. And so, yeah, so a great question, and hopefully that kind of addresses your concerns. Yes, hopefully. And please feel free, audience, if you have any follow-up questions, feel free to type them in the question tab of your control panel. But I think that brings up a good point. And two, since you work in a surgical ICU, and so do I, I think, comparing to our medical colleagues, maybe even Tina, that, you know, even for 0.2 or 15 mics per minute of norepi, that might be considered high to us. That might be considered a little bit lower to someone else. Now, with that, and you did kind of talk about the ATHOS-3 trial, did you guys collect or were any patients in here on angiotensin-2? Or was that just too small of a sample size? I think that angiotensin-2 was approved in late 2017. Our study was conducted, started in September 2017. So it wasn't an option in terms of the type of basal pressure that were included. Okay. So the answer is probably maybe a little bit, but probably not. So perfect. And then as a follow-up question, did you by any chance stratify the amount of fluid resuscitation in the patients that had AKI or might have had already end-stage renal disease? Or was that just due to the chromatic design, something you didn't look at? And could there be a potential follow-up study? Yeah. So I think in terms of the septic shock population, I think it's a very relevant question. Because of the pragmatic nature of the data collection, we didn't have detailed information on creatinine that would allow kind of a definition of an AKI phenotype. Although we did document use of renal replacement therapy. In terms of potential subgroup effects in those patients, again, really interesting and important. We didn't have enough patients to kind of explore that in a meaningful way. Particularly if the interest is in an interaction between vasopressors and fluids, looking at that in a subgroup is kind of a three-way interaction, which requires kind of substantially more larger sample sizes. Did you have anything to add, Tina? Yeah, that's right. I mean, in terms of comorbidity, we did collect if someone has a baseline dialysis need or at least a chronic renal insufficiency. And then in the overall cohort of the 1600-some patients, looking at the data, about 11%, 12% of them ended up being on renal replacement therapy at the end of the study period. And I think both Todd and my model, when we were doing patient level and severity of illness adjustment, we did include their at least baseline renal function based on data collection. Yeah, that's right. Like ESRD or CKD, yeah. Todd, did you have something else to add? No, yeah. I was just going to say, Tina is correct. Our models also adjusted for baseline chronic renal insufficiency and ESRD variables. And did you do any specific sub-analysis? That was a follow-up question I just received. On those, especially on end-stage renal disease specifically, did they receive less fluid and or higher vasopressor doses? Or is that something you really have not teased out yet? Well, go ahead, Tina. Go ahead. I think in our study, looking at all sort of shock types, end-stage renal disease was associated with less fluid and higher chance of having vasopressor use. But ultimately, we didn't look at it specifically. And actually, I think Dr. Anita Qadir and the UCLA group is looking at a subgroup analysis of these ESRD patients. Yeah, we didn't. In terms of outcome analyses, we did not do any subgroup analyses in those patients. And we did include descriptive data in the paper. But I can say that if you look at the first six hours, the volume of fluid given, it was significantly less in patients who had baseline renal dysfunction. Also less in patients who had congested heart failure, which I think we would expect that pattern with both of those diagnoses. We also saw differences in the amount of fluid given in terms of where the onset of shock was. So about a liter more fluid was given on average when the onset of septic shock was in the emergency department compared to in the ICU or in the ward. We saw that the differences in vasopressor use were less pronounced than the differences in fluid. That is kind of interesting. But when I kind of think about it, it somewhat makes sense, at least in my institution, with the drug delivery models going out there. It's much easier to get things to the ward, I think, or the ICU versus the emergency department. So not super surprising there. I kind of have a next question, and this one's a little bit more for Tina. Specifically with using assessments, it seemed like the most common ones, at least here, using ultrasound. What do you think these results might help with some lower middle resource countries, especially with passive leg raises? Yeah. So the one thing I was reflecting from this presentation is that the poll question I asked actually really reflected what we saw in the study, that most of the patients were resuscitated by empiric judgment of the treating clinician, basing on their blood pressure and clinical exam. And similarly, about 23% of this group said they would use critical care ultrasound, which is close to what we found as well. I think the few takeaway points for me is that we are definitely moving away from these more invasive pressure measurements, like CVP and SWAN and even the stroke volume variation and such. All those things require invasive measurements. But critical care ultrasound is fast, it's easy, and you could have one probe and just clean it well and use it on multiple patients. Like I said earlier, with the limitation of the quality of the study, right? So I think in a resource-limited setting, having critical care ultrasound and good training and good interpretation of the data would really help the decision-making and resuscitation. But again, it's operator-dependent, and I think if we can implement and be able to use this well, it really would help in decision of shock management. Well, thank you very much. Very good answer to that. And then there was a follow-up question for I think both of you. Specifically, like the type of fluid, did it make any difference in there? And I'm assuming they mean type either crystalloid versus colloids or difference between saline versus a balanced salt solution. So in our analyses, so we lumped all of our fluids together. And really, in terms of sample size, we were underpowered to try to look at subgroups of different types of fluid. So I can't really speak to whether or not there would be differences in the interaction between pressors and fluids based on fluid type. But I think that's an important question and certainly something that could be good grist for future work. I think our intention is to document and assess what usual care is. And we saw that most of the patients, the most common fluid type obviously is the crystalloid that people receive. Small proportion of patients receive colloids. Smaller proportion receive any sort of blood type. And we saw this, it's not in the actual supplement paper. But we did see, you know, the septic patients were getting more crystalloid. The hypovolemic patients who may have trauma or GIP, those patients were the ones who were getting more blood. But overall, the use of blood or colloid in resuscitation is less than 10% in the overall use of fluid. So I think that suggests that we are doing the right thing for guideline and what's published that first line of therapy is using crystalloid and shock resuscitation. Barring trauma and bleeding, of course. No, definitely it's good. I have a follow-up question more in the about the ultrasound monitoring. And that was specifically since this is relatively new. How often do you repeat those measurements, especially going in and out of patient's room? And, you know, I think too, you know, right now it's still 2020. So we always have to think about the pandemic in the back of our minds. Of course. So critical care ultrasound is, it shouldn't be a static measurement, meaning you do the exam once. And I think in practice, in clinical practice, if you are assessing a patient after you provide an intervention, it's always good to see if that intervention made a difference on the patient, right? So if you go in and scan the patient looking at lung, there's A-line, you see the IVC is varying with respiratory on a vented patient. You give a fluid bolus or do a passive leg raise. I think it's prudent to go back and see if that patient responded to the fluid you gave. Did that change their physiologic numbers? Not only just the blood pressure and all that. So I would encourage multiple scan. It's an easy enough thing to do. Some of the sites I know would use, in terms of multiple scan on multiple patients, because this is a machine that you definitely need to clean. I know there are people using ultrasound probe cover between patients to minimize contamination. We also use our purple wipe to really sanitize between patients. I don't think we should limit ultrasound one per patient per day. I think it's a dynamic instrument that should be used when it's clinically indicated, and we should follow up on the intervention that we provided. I completely agree with you. I think a lot of this is, you know, we take these numbers specifically as a word, and it's written in stone, but they're very dynamic. And it's just one piece of the puzzle. I completely agree. And I think we have time for one last question. And I do agree with all those who are writing questions. It's so great to have a conversation that's not COVID-related. For both of you, were there any interventions that were noted that kind of looked, or did you collect, reducing any of the ischemia secondary to the vasopressor use? And I'm assuming they mean things like maybe nitroglycerin or something like that, is what I'm assuming. Yeah, I don't think we, to my knowledge, at least in ours, we didn't have data on potential vasodilator therapies or things of that nature. We did, there was data collected on kind of home antihypertensive medications. So potentially there would be data describing like, you know, beta blockers, which might be relevant from myocardial demand or mitigating some of the arrhythmogenic effects. But we didn't look at that specifically in our analysis. You know, I really, every time when I talk to Todd and Ross about their study results, the fact that within the first six hours, even if you start someone on vasopressor, the use of fluid is so important to mitigate risk of mortality, that interaction. To me, that's one of the most striking things, that you do need something in the pipe in order for the pump to work well. Yeah, I think for me, that was one of the more interesting findings. And I think, yeah, so, and I guess one takeaway from our results is, you know, there's definitely increasing interest in minimizing fluid in perhaps using earlier aggressive vasopressors. And so our data suggests that unless we have a really good understanding of what the balance is, that maybe we should be hesitant in terms of changing our practice. Fortunately, there are at least two large randomized controlled trials that are kind of more directly testing that question. The Clover's trial, which is kind of liberal versus restrictive fluid resuscitation, which is kind of really, you know, during the first 24 hours comparing a early vasopressor strategy versus early fluid strategy. And so I think that will, you know, that's kind of close, that closely mimics the question that we addressed here in our observational study. The classic study is another large randomized controlled trial that is, again, kind of comparing different strategies of fluid management. So, yeah, so I think it's fascinating and excited that we will have some good randomized RCT data and hopefully in the next couple of years. Well, thank you both. And I completely agree with the audience. Those were excellent presentations. And even more importantly, I think an excellent discussion at the end. So I'd like to thank our presenters, Todd and Tina, again, in our audience for attending. Again, anyone who joined us today for the webcast will receive a follow-up email that will include an evaluation. Please take five minutes to complete the evaluation. Your feedback is greatly appreciated. And on a final note, please join us for our next Journal Club on November 19th. And this concludes today's presentation. Thank you. Thanks, everyone.

shock resuscitation

physiologic assessment

critical care ultrasound

fluid and vasopressor use

variation in shock types

vasopressor dose

mortality in septic shock

interaction between vasopressor dose and fluid resuscitation volume

early aggressive vasopressor titration