Hello, and welcome to today's Journal Club Critical Care Medicine webcast. This webcast, hosted and supported by the Society of Critical Care Medicine, is part of the Journal Club Critical Care Medicine series. This webcast features two articles that appear in the April 2022 issue of Critical Care Medicine. When the webcast is being recorded, the recordings will be available to all registrants on demand within five business days. Just simply log into mysccm.org and navigate to the Mind Learning tab. My name is Tony Gerlach, and I'm a clinical pharmacist at Ohio State University Medical Center here in Columbus, Ohio, and I will be today's moderator. Thank you for joining us. Just a few housekeeping items before we get started. There will be a question and answer or Q&A session at the conclusion of both presentations. To submit a question throughout the presentation, just type into the question box located on your control panel. If you have a comment to share during the presentations, you may use the question box as well. And finally, everyone enjoying today's webcast will receive a follow-up email that will include an evaluation. Please take the three to five minutes to complete the evaluation as your feedback is greatly appreciated. Please note that this presentation is for educational purposes only. The material presented is intended to represent an approach, view, statement, or opinion of the presenter that may be helpful to others. These views and opinions expressed herein are those of the presenters and do not reflect the opinions or views of SCCM. SCCM does not recommend, endorse any specific test, physician, product, procedure, opinion, or other information that may be mentioned. And now I'd like to introduce today's presenters. First is Dr. Seth Bauer, who is a clinical pharmacist in the Department of Pharmacy at the Cleveland Clinic and an assistant professor of medicine in the Cleveland Clinic Lerner College of Medicine. After finishing his PharmD training at Drake University, Dr. Bauer completed his postgraduate training at Mayo Clinic in Rochester, Minnesota. His research centers on vasoactive agents and the use of circulatory shock, and he's authored or co-authored more than 85 peer-reviewed manuscripts and book chapters. Dr. Bauer is a fellow of the American College of Critical Care Medicine, and he's also a member of the Surviving Sepsis Campaign Research Committee. Our second presenter is Dr. Gretchen Sascha, who's a clinical pharmacy specialist practicing in the medical ICU, also at Cleveland Clinic at their main campus. Dr. Sascha received her B.S. in pharmaceutical sciences and her doctorate of pharmacy at the University of Toledo. She completed two years of postgraduate training at the Cleveland Clinic and is board certified in critical care. Her research focuses on hemodynamic, septic shock, and vasoactive agents. She is authored or co-authored over 30 peer-reviewed manuscripts and two book chapters. Thank you both for joining us today, and I'll turn the presentation over to Seth. Thank you, Dr. Gerlach, for the introduction, and I'm delighted to participate in the webcast today to discuss our article entitled, Association Between Vasopressin Rebranding and Utilization in Patients with Septic Shock. I don't have any actual conflicts of interest relative to this presentation, but in the interest of full disclosure, I will mention that I was formerly a consultant for Walters Kluwer, and I also am appointed by SCCM to serve on the Surviving Sepsis Campaign Research Committee as well as an albumin use guideline. Vasopressin is not a novel drug to this audience, so I won't dwell on the background too long. I think vasopressin can be best described as a non-catecholamine vasoconstrictor without inotropic properties. Several studies have shown that vasopressin is a catecholamine sparing agent. However, the effects of the drug as an adjunct in septic shock are unclear, specifically some meta-analyses showing a mortality benefit and some meta-analyses not detecting a mortality benefit. The most recent guidelines from the Surviving Sepsis Campaign suggest vasopressin as a norepinephrine adjunct. Specifically, in patients with inadequate mean arterial pressure, this is an alternative to escalating the dose of norepinephrine. Vasopressin has been used in clinical practice for circulatory shock for over two decades. It was first utilized for research in 1928. This finding is important because drugs marketed prior to 1938 are exempt from the Food, Drug, and Cosmetic Act review by the FDA. The FDA, however, in 2006 launched the Unapproved Drug Initiative. The details surrounding this initiative are nicely outlined in the accompanying editorial with our article by Drs. Peterson and Devlin, and I invite you to read that excellent editorial. Briefly, this process is called rebranding, and it was intended to decrease the number of drugs that are used off-label, specifically looking at medications that are used in practice without a new drug application. If a manufacturer were to submit a new drug application and have it approved, they would be afforded patent protection and market exclusivity for the drug. This is the process that the manufacturer of vasopressin undertook, and vasopressin was rebranded and entered the market in November 2014. Before vasopressin was rebranded, there was a large, well-done study describing the drug's utilization for septic shock. In this study, the authors used the Premier Healthcare Database to evaluate utilization of vasopressin specifically in patients with septic shock. The authors found that at the beginning of their observation period, around 14 or 15 percent of patients were receiving vasopressin in 2008, but at the end of their observation period, about 20 percent of patients were receiving vasopressin in 2013. Notably, there was significant variation in utilization by hospital. The utilization rate ranged between 0 percent of patients in some hospitals all the way up to 70 percent of patients in other hospitals. Patient demographics and clinical characteristics were responsible for the majority of the observed variation. However, the hospital of admission also contributed substantially to the observed variation. However, the epidemiology of vasopressin use after rebranding in 2014 was unclear. This use pattern was of particular interest to us because of the rising vasopressin acquisition cost over time associated with rebranding. In late 2014, the acquisition cost of vasopressin was approximately $4.27 per 20-unit vial. However, as soon as rebranded drug entered the market, the price immediately jumped nearly sixfold to about $25 per vial. And as you can see on the slide, the cost of vasopressin sequentially increased almost on a yearly basis, such that in 2020, the cost of vasopressin in the U.S. was about $197. In a 2016 survey of critical care pharmacists, 72 percent of respondents reported implementing some form of vasopressin cost containment initiative as a result of this cost increase. However, studies had not yet evaluated large-scale vasopressin utilization after rebranding. It's this gap in knowledge that our study aimed to address. We partnered with the Colorado Pulmonary Outcomes Research Group to assess the association of vasopressin rebranding and subsequent cost changes over time with vasopressin utilization in patients with septic shock. Our central hypothesis was that across the United States, vasopressin utilization for septic shock decreased after rebranding. Our study was a retrospective multicenter cohort study using the Premier Healthcare Database that evaluated patients admitted between January 2010 and March 2017. We created cohorts labeled as before rebranding and after rebranding, and I will note we used the beginning of 2015 as a surrogate marker for the vasopressin rebranding time point to allow analysis of the data in quarterly intervals. We included patients above age 18 in the ICU for two days or more coded as having severe sepsis or septic shock with a pharmacy claim for at least one vasoactive agent for two or more calendar days. Consistent with previous studies, we excluded patients from a pediatric facility or from the lowest fifth percentile of sepsis admissions. Our primary outcome we assessed was the proportion of patients receiving vasopressin before and after rebranding. Secondary outcomes included vasopressin cost, the proportion of patients receiving other vasoactive agents, the cost of those agents, and clinical outcomes. For the sake of time in today's presentation, I will only discuss the primary outcome and the vasopressin cost results, but I invite you to review the manuscript for results of these other analyses. Our statistical approach used both traditional and modern techniques where we analyze our cohorts using chi-square and Wilcoxon rank sum tests. However, we also use interrupted time series analysis, also called segmented regression. This is not a method that is commonly used in the critical care literature, so I'll just take a moment to explain it in a little bit more detail. This method uses aggregate data that are collected over equally spaced intervals. In the example of our study, we used quarterly intervals, and then the method plots these data points on a time series graph and draws a regression line between that time series. The different regression lines are plotted for the before intervention period and the after intervention period, and then the slope of the regression lines are compared. Additionally, the method can evaluate the origin or level of that second regression segment to see if it changed immediately after implementation of an intervention. Important for our study was that this method accounts for pre-intervention trends. Again, it was important for our study because we knew prior to vasopressin rebrowning, the utilization of vasopressin had an increasing threat, and I will note this method is recommended for evaluation of longitudinal data sets as a pre-post comparison. Of the nearly 5 million patients in the PREMIER Healthcare database during our study period, most were not included in our study due to no sepsis diagnosis or less than two days in the ICU or less than two days receiving vasoactive therapy. We included about 290,000 patients, with about 187,000 in the before vasopressin rebranding cohort and about 100,000 in the after rebranding cohort. Patients and hospital characteristics of these two cohorts were similar. Most patients were above age 60, cared for in a medical ICU, and had multiple comorbidities. Represented hospitals were predominantly large urban hospitals. About 40% of hospitals were teaching hospitals, and the southern U.S. was the most frequent region. In our primary analysis, we found that before rebranding, about 26% of patients with septic shock received vasopressin. This proportion increased to about 31% after rebranding for about a 5% increase in utilization. Clearly, this finding is inconsistent with our operatory hypothesis. Incompetently, the cost of vasopressin per patient increased by about $450. The interrupted time series analyses completed and extended our primary analyses. On the slide here is a figure representing the interrupted time series analysis of quarterly vasopressin utilization. The y-axis represents the proportion of patients receiving vasopressin within the study quarter, and the quarter and year are on the x-axis. The four rebranding quarters are represented in blue and after rebranding in red. The regression lines for these two time periods show increasing utilization over time in both study periods. The slope of the regression lines are quite similar, with a post-rebranding change in slope of 0.06% per quarter. Essentially, vasopressin utilization was increasing before rebranding and continued to increase at a similar rate after rebranding. These data contrast the interrupted time series analysis of quarterly vasopressin cost. This figure is similar to the previous one, but vasopressin cost per patient is on the y-axis. As you can see, we found an increase in the slope of the regression line after rebranding, as well as a change in the regression segment level. These results were quite intriguing to us because they do represent price inelastic demand for vasopressin. Our study wasn't designed to evaluate the reasons for these findings. However, from an economic perspective, it appears that clinicians, at least in the study time frame, considered vasopressin a necessity because utilization did not change despite rising costs. Alternatively, ICU practitioners may not have been aware of the increased drug cost. This rationale is supported by previous work showing ICU providers generally have low awareness of drug costs, and they may have continued to use vasopressin as if it were a low-cost drug. The findings from our study contrast other reported data with rebranded drugs, where nitroprusside and isoproteranol use decreased when prices increased. This may be because clinicians felt, at least during the study years, there were no acceptable alternatives to replace vasopressin in practice, whereas alternatives exist to other rebranded drugs such as nitroglycerin to replace nitroprusside, for example. Our study has important limitations that must be considered. First is we relied on the accuracy of a retrospective database that's primarily composed of administrative encoding data. We also did not have access to a lot of discrete data that we were interested in, such as laboratory values or medication dosages. Further, medication utilization was limited to calendar day billing. Lastly, we included mostly large hospitals in urban environments. Whether our results are applicable to smaller hospitals or those in rural environments is unclear. These hospitals, in fact, may be more susceptible to pricing constraints. The findings from this study, though, have important clinical implications. Since it appears that clinicians are using vasopressin despite its rising cost, we need further evidence to elucidate the optimal initiation criteria. This manuscript suggests that there are large evidence gaps, but I think Dr. Sasha will have something to say with the paper that she'll be presenting. Hospitals constrained by the cost of vasopressin may want to focus their efforts on the recovery phase through discussing vasoactive discontinuation order and vasopressin cessation methods, such as abrupt or tapered cessation. The policy implications for our findings were not ones we discussed in our manuscript, but they are nicely outlined in the accompanying editorial. And I'll quote the words directly from Drs. Peterson and Devlin, who state, the story of vasopressin is an important reminder to policy makers and clinicians that the rewards given to manufacturers selling a newly branded drug are commensurate with their research and development investment. They go on to say, a decision on market exclusivity should depend on the generation of new and ideally novel research data and investment, the investment the company is planning to make in the drug product, and the price it plans to charge. Unfortunately, none of these objectives were met with vasopressin rebranding. In light of the unintended consequences with the FDA unapproved drug initiative, it's currently being reconsidered. This concludes my presentation and I'll now turn it over to Dr. Sasha to present. Thank you, Dr. Bauer. I think the discussion, as he mentioned, that discussion segues really nicely into some discussion on when to utilize vasopressin, since we did just learn that despite its cost, its use still increases. So before I begin my portion, I will mention that I have no conflicts of interest to disclose regarding this presentation or the content therein that I'm going to talk about. And I'll start off by revisiting the background slide that Dr. Bauer discussed. So vasopressin is our non-catecholamine agent with unclear overall mortality benefit. And I'm not going to belabor that piece right away, but what I will expand upon are the 2021 surviving sepsis guidelines. The prior 2016 iteration suggested that either vasopressin or epinephrine be added to norepinephrine. However, the 2021 guidelines separate out the two adjunctive agents and suggest the addition of vasopressin as a second line to patients on escalating doses of norepinephrine over epinephrine. So for the first time, giving us some guidance on what they pose as the second line agent. Now I want to discuss the trials that evaluated vasopressin and tease out some of their nuances. I know that everybody on this call is familiar with the landmark VASH trial that was published in 2008, in which over 800 patients with septic shock were randomized to either norepinephrine monotherapy or vasopressin in addition to norepinephrine. Overall, they found no global difference in 90-day mortality between groups. However, I want to discuss a few of their subgroup analyses. So first they took a subset of patients that they classified as having less severe shock, which they defined as those on or requiring less than 15 mics per minute of norepinephrine at the time of randomization. And it was in that subgroup that they found that patients who were randomized to receive vasopressin had an improved mortality rate in comparison to those who did not receive vasopressin. Similarly, they looked at a different subgroup of patients whose lactate concentrations were less than or equal to 1.4 millimoles per liter at the time of randomization. And they found similar outcomes, improved mortality rates with those randomized to receive vasopressin. But with this piece, I will note that there was a significant interaction detected and these results should be interpreted cautiously. And then finally, when they looked at the subgroup of patients based on when the study drug was initiated, they did not find any difference between intervention arm and mortality. But regardless, this is what leaves people wondering if there is something more to vasopressin's use. And in spite of the vas subgroup analyses potentially indicating this benefit from vasopressin's use in patients who are less severely ill, in practice we can see great variations in when it's initiated. VAS initiated vasopressin when the patients required an average dose of 20 mics per minute of norepinephrine, and when the patient's lactate concentrations are around 3.5 millimoles per liter, and then within about 12 hours. Now compare that to the VANISH trial, which is our second landmark vasopressin trial, which was set up to look at early vasopressin use. You can still see there was a wide range of norepinephrine requirements at the time of vasopressin initiation, with a median IQR of a norepinephrine dose at randomization between 0.1 and 0.31 mics per kilo per minute. And in a 75 kilo patient, that range is around 7 to 23 mics per minute. So pretty wide, and some might argue not always early initiation. Finally, the last study I want to briefly mention is a study that Dr. Bauer and I conducted in addition to other colleagues of ours that looked at vasopressin use at our institution. And I use this as a representative of when vasopressin is initiated outside of clinical trials. And you can see that on average, in practice, vasopressin is initiated at a dose of around 28 mics per minute of norepinephrine, and when patients' lactate concentrations are around 5 millimoles per liter. Now the combination of all these points that I just mentioned, so the changing recommendations in the Surviving Sepsis Campaign Guidelines, and now I didn't mention this, but they added this dose range in which to consider the initiation of vasopressin, pairing that with the unclear overall mortality benefit, and then the data that seemed to indicate that earlier initiation may be better. All of these factors lead to the remaining question of when vasopressin should be initiated. And as we've discussed, timing, and I say that in quotes, of vasopressin initiation can be thought of in many manners. So is it the severity of illness of septic shock dictated by their norepinephrine equivalent dose? Is it the severity of illness dictated by their lactate concentration, or even the temporal time from shock onset? So we wanted to evaluate each of these three timing variables. And that was what we set out to do with this study. So our study goal was to evaluate the association of norepinephrine dose, lactate, and timing of vasopressin initiation with mortality in patients with septic shock receiving vasopressin. And our hypothesis going into this study was that early initiation of vasopressin would be associated with mortality, and early referring to vasopressin initiation at lower norepinephrine doses, at lower lactate concentrations, and earlier from the time of shock onset. This was a retrospective multi-center observational study that included in patients admitted to our institution, our enterprise institution, between 2011 and 2017. Patients were included if they were over the age of 18 and met the CDC adult sepsis event definition. And some of those components, but not all of the components, include the receipt of blood cultures, antibiotics, and a lactate greater than two. And I refer you to the full text for the entire definition of this component. Patients also had to have received continuous IV norepinephrine, as well as continuous infusion adjunctive vasopressin. Patients were excluded if vasopressin was initiated before or at the same time as norepinephrine, or more than 48 hours after the first catecholamine agent. Our analytic plan included conducting multivariable models to evaluate the association of each of our three timing variables on in-hospital mortality. We also wanted to determine if there was and estimate a change point in which the slope of the odds ratio between each of our variables and its association on in-hospital mortality changed. So this could indicate a potential clinical cutoff that could be utilized in practice. Now a change point was only detected for norepinephrine equivalent dose, and that was at a dose of 60 mics per minute. And if you use weight-based dosing, that can be around 0.8 mics per kilo per minute in the average 75 kilo patient. And I'm going to go into what this change point means as we get to the results. So we ended up including over 1,600 patients, and you can see our baseline characteristics. And I'm not going to focus on the comparisons of survivors to non-survivors, as it's not surprising there are pretty vast differences in these patient groups. But what I want to point out is when vasopressin was initiated. So if you look to the total patient column, you can see that on average, vasopressin was initiated when the norepinephrine equivalent dose was around 25 mics per minute, when the lactate concentration was around 3.9 millimoles per liter, and within about five hours from shock onset. Now I'm going to touch on each of our three timing variables of interest separately. So first, looking at norepinephrine equivalent dose. This figure depicts the predicted probability of in-hospital mortality over increasing norepinephrine equivalent dose at vasopressin initiation. And to note, these probabilities are adjusted for many variables that are detailed for you on the slide, but include several markers of severity of illness, so attempting to adjust for all of our known compounders. Now we found increasing predicted probability of mortality as the norepinephrine dose at vasopressin initiation increased. And to spell this out, what we found, and this is looking at our odds ratio, what we found was that the odds of mortality increased by 20.7 percent for each 10 mic per minute increase in norepinephrine equivalent dose at the time of vasopressin initiation. And this was only up to that change point of 60 mics per minute that we detected. So beyond the initiation dose of 60 mics per minute, there was no association between dose and in-hospital mortality. And I'm going to again emphasize that the 20 percent piece is not an absolute mortality rate, but refers to our odds ratio and the finding that the odds of mortality increased with increasing norepinephrine equivalent dose at the time of vasopressin initiation. The next timing variable is lactate concentration at the time of vasopressin initiation. Now when looking at both lactate and timing, this got a little bit trickier because what we found was that there was an interaction between timing of vasopressin initiation and the lactate concentration at vasopressin initiation. So to evaluate their association, we had to evaluate them together. What we found was that there was a linear association with lactate and increasing – lactate and in-hospital mortality, meaning that in-hospital mortality linearly increased as lactate concentration at vasopressin initiation increased. So when vasopressin was initiated two hours from shock onset, the odds of mortality increased by 12% for each millimole per liter increase in lactate concentration at the time of vasopressin initiation. So that's that top line in the table on the slide for you. And then when vasopressin was initiated 12 hours from shock onset, the odds of mortality increased by 18% with each increasing lactate concentration at the time of vasopressin initiation. Now lastly, timing. We actually found no association between increasing time from shock onset to vasopressin initiation and in-hospital mortality. And again, because of that interaction between timing and lactate, this was evaluated at various lactate concentrations, which is why you'll see three curves on the figure. Overall, what I take home from these data is that if you're going to initiate vasopressin in a patient with septic shock, it should be initiated at lower norepinephrine equivalent doses and lower lactate concentrations. And these findings could be rationalized by the utilization of vasopressin as an endocrine replacement, and it may be that restoration of vasopressin concentrations should be targeted early when vasopressin concentrations drop to more quickly restore tissue perfusion. And it could also be that the early initiation of vasopressin limits norepinephrine and catecholamine exposure, preventing untoward effects such as the immunomodulatory effects recently discovered with norepinephrine. But regardless, I think these data are important in the canon of literature regarding vasopressin and septic shock. Now that's not to say that these data are not without limitations. First and foremost, I want to stress that these data can only be applied to those patients who receive vasopressin and make no conclusions to vasopressin's use compared to no vasopressin use, like with VAS and VANISH, as we have no comparator or control group in this study. And because of its observational nature, we relied on medical record charting for our data extraction, and we were unable to account for cardiac function in this population. But what I want to leave this talk with is that I believe these findings corroborate the subgroup analyses from VAS that I described at the top of my portion, that early vasopressin may be integral to its benefit. And my take-home point is that I believe these data support the initiation of vasopressin when the norepinephrine dose is around 10 mics per minute, or when the lactate concentration is 2.3 millimoles per liter or below. And if you use weight-based dosing, this would be 10 mics per minute would be equate to about 0.13 mics per kilo per minute in a 75-kilo patient. And this is now when I start the discussions of initiating vasopressin in my practice. And I leave with this table presented in our supplemental appendix that reports out the predicted probability of in-hospital mortality at these and other time points which support the recommendations and the conversation that we had today. And now we invite any questions that you may have, and I'm going to turn it back over to Dr. Gerlach for the moderation of the Q&A portion. Thank you very much, Seth and Gretchen. I personally learned a lot about this. And I do very much appreciate, as we were talking before, since I am an institution that uses weight-based dosing, that you talked a little bit this on weight-based as well. So thank you very much. And I think our first question that we have today, and feel free for anyone in the audience to type a question at any time, is based on this, have you guys developed a protocol or algorithm of really when to start vasopressin in these patients? And if so, what exactly is in your protocol, and was that really hard to get buy-in for? Sure. That's a really great question. I will try to keep my answer brief because I could probably have an entire discussion about this. But I will say that actually, historically, I'll go back a little bit, we used to recommend the initiation or reserving vasopressin until the patient's norepinephrine dose exceeded 50 mics per minute in our practice for septic shock patients. Again, this is specific to septic shock. And this was based on a trial that we saw that was published out of Vanderbilt that did a similar practice. And it actually wasn't until we did this study that we started to change and look into that practice too. And I will say, historically, we did compare and look at that protocol that we initiated of reserving vasopressin until a dose of 50 mics per minute of norepinephrine and did find no differences globally in our patients. So we felt comfortable at the time with the practice that we were utilizing. Now when we published and looked at our evaluation here, what we wondered is actually whether or not we were even initiating vasopressin early enough before that protocol was initiated, which is likely why we didn't see any difference when we initiated it. So because of these data, as well as these new data that I just presented, we actually have regrouped in our medical ICU and are now going to start initiating vasopressin earlier in the patient shock course. So our new guidelines are going to have that conversation and recommend its initiation once the dose is around 15 mics per minute. It actually was pretty easy to get buy-in. We discussed our results with our physician colleagues, many of whom were actually involved in this study that I presented today. And based on this literature and knowing that it was specifically in our patient population, we all felt very comfortable and supported our findings of initiating it earlier. So it's been going along really well. I think the one thing that I'll add to that as well is some of the logistical challenges with vasopressin are real, meaning when many people think about it's time to start vasopressin, it's not right there in their ICUs available for administration. So now that ready-to-use vasopressin is available, it may be easier to adapt our plan of starting vasopressin earlier. Now, I think you do talk about that. In fact, I was talking with a nurse today who is traveling, who is at a small hospital here in Ohio, and she is making all her drips on nights. And I still didn't think that that happened here. So it is something I think we have to think about is how do you get the drug from the pharmacy to the patient in a timely manner? So I think that's a good point. And Seth, this follow-up question is really for you to ask her first. But, you know, I think you guys have made a good point, and we're getting better literature of where we define when to start vasopressin and septic shock, as it seems that the use of early vasopressin really might confer a benefit, but it's at an increased cost. For clinicians that might have to justify this increase in cost, what should they do? Yeah, I think this is an important question. And part of the rationale for why Gretchen and I started down this path of researching, and that is, can we justify the increased cost? We're grateful that we're at an institution where patient outcomes are prioritized and cost becomes a discussion item, not a driving factor. But I'm completely cognizant that other hospitals, cost becomes much more of a relevant piece of the conversation. For that reason, our group actually did a formal cost-effectiveness analysis looking at second-line vasoactive agents in patients with septic shock. And that study found that vasopressin is a cost-effective therapy compared to simply increasing norepinephrine alone, and is more cost-effective than alternative agents such as angiotensin II. That study actually found that the incremental cost-effectiveness ratio of vasopressin was about $18,000 to have one improved quality-adjusted life year. So I think from a hospital's perspective, it makes sense to use more vasopressin in septic shock. However, the reality is that hospital budgets are often siloed, and what I mean by that is the pharmacy budget is different than the critical care department budget. The critical care department may see better patient outcomes and lower costs in that silo as opposed to the pharmacy budget exploding and not seeing realization of any benefits. I think if you're at an institution where that is the case, then I think that the discussion of vasopressin is a great item to bring to administrators to have discussions about overall patient benefit and cost-effective care from an institutional perspective and not relying on these individual silos to drive decision-making. Well, thank you very much. Here's a question from the audience, and I think it's a very good question. Would cessation of vasopressin be at the same dose of norepinephrine that's equivalent to the initiation, or is there a time point on vasopressin after which cessation does not impact mortality? I think that's a really great question, and I am going to preface my thoughts on we don't really know the answer, but what I can say is at least from a cessation standpoint in general, we have evaluated this internally at our institution as well, at least looking at the discontinuation order. What we found and published those results in a meta-analysis, because there are several observational trials as well as one randomized controlled trial that looked into this, and what we found was that discontinuing vasopressin first was associated with some increased hypotension, which was driven by just needing to increase your norepinephrine dose, which was not associated with clinical outcomes. In our practice, we do recommend discontinuing vasopressin first before norepinephrine to also add to the cost-saving discussion that Seth just posed, but I think the question of then when does it get discontinued is still something that we don't fully know the answer to, but what I feel that I can hypothesize from the research and the data that is out there is that the vast majority of the studies discontinued vasopressin when the patient was fully in the recovery phase of septic shock, and they were no longer on escalating doses. They had been in the co-populace. They were improving clinically, and there was the one randomized controlled trial that did find differing results actually discontinued vasopressin maybe a little bit earlier when they weren't actually in that recovery phase, so because of that, I think that we might be seeing that the benefit, or I should say when it is safe to discontinue vasopressin is once you feel that patient is in the recovery phase and is clinically improving, so in practice, I think if you're going to implement our recommendations of initiating vasopressin around a dose of 15 mics per minute, I think that that is a very sensible and reasonable point in which to discontinue it once that patient is on the path to clinical recovery. I don't know if Seth, if you have any thoughts as well. I would just echo what Gretchen is saying that we don't know the answer to this, and this is a great research question for somebody that wants to design a trial to answer this question. I think local practice dictates a lot of how these drugs come on and off, and I will in full transparency mention that we did a study, an observational study looking at this, and a group from Mayo Clinic did a similar study that was even larger that did find a benefit of leaving vasopressin for last drug, and I think those competing observational studies just go to show that we really don't know the answer to this, and we need well-designed studies to help answer the question. Absolutely. Well, thank you very much. I have two questions from the audience that are very similar. So, first of all, they wanted to state, thanks for the excellent presentation, so thank you very much, and basically the question is, is there a difference or do you think we should go beyond 0.3 to 0.4 units per minute with vasopressin and either septic shock or vasoplegic shock? Is there any benefit for going higher than 0.04 units per minute? So, I think that's another really great question, and again, I hate to say it, but one, I will be transparent about that we also, again, don't really know the right answer. At least if I am going to relate it and extrapolate it from the data that we presented here today, that we definitely can't answer that question with the study that I just presented. So, the vast majority of patients received an initial dose of 0.03 or 0.04. I think it was only 30 patients or so that received a dose higher than that, so we can't utilize that to help us in this sense, but what I will say is I'm sure we're all familiar with the Torgerson trial that was published in 2010, I'm pretty sure it was, that did compare doses of 0.033 units per minute of vasopressin to 0.067 units per minute in patients with vasodilatory shock, and they did find significantly lower norepinephrine dose requirements in those that were randomized to receive 0.067 units per minute of vasopressin that maybe this indicates there's some hemodynamic benefit to the use of higher doses, but to date we just we really don't have outcomes data to support this practice. So if I ever do it, I really do reserve it for pretty much very salvage therapy until we get more clinical trial data associated with not only the benefit, but the potential harms and consequences of using doses that high. So I can't support it as general practice at this time, but I again, I think it's another really great clinical question. I think the only thing I would add is that there are clinical trial data with vasoplegic shock after cardiac surgery, which in my mind, I separate that scenario from septic shock, even though they behave similarly, because there can be safety concerns with high dose vasopressin. So in vasoplegic shock, I think it's reasonable to use doses above 0.04 units per minute, because we have trial data showing that it's safe. I guess that kind of begs the next question is, with this increase in vasopressin use and using it potentially earlier than what we've been doing, that shows there's at least it seems like there's some clinical benefit. Have you seen any adverse effects with either starting it early or using some of these higher dosages? I will say, and I hate to say anecdotally, I will say anecdotally, no, we have not seen that. And I, but I will say what our study showed. So unfortunately, our studies weren't able to evaluate adverse consequences associated with vasopressin use. So first we couldn't detect if there was a change because we didn't have a comparator group at least. But I think that, I think the adverse consequences is something that should be noted and should be evaluated. But I believe it's very difficult to evaluate in an observational retrospective manner, because looking at ischemic type outcomes is not a easy variable to capture in looking back again, retrospectively. So it wasn't something that we found, but not something that we were able to look at fully. But I think it's something that should be kept on our radar as we continue to utilize these. But remembering that we're still just using one vasoactive agent instead of another, and that hopefully we might be limiting other effects by limiting our catecholamine dosages. So I'm not sure where that will ultimately end up, but it's something that we should keep at the forefront of our mind when we increase its use. Meta-analysis suggests that the incidence of digital ischemia is probably higher with vasopressin. As the audience probably sees, we use a lot of vasopressin here, and I can't specifically point to more digital ischemia with vasopressin, but of course, we'll defer to the randomized trial data. But the incidence is quite low. So we just may not be detecting the digital ischemia. And I guess that kind of begs also the next question is, are there any patient populations that you think vasopressin might be even first line before norepi? And I'm not necessarily talking about septic shock, but different types of shock. I think this is really challenging. I think that the data from the Banks trial with vasoplegia after cardiac surgery did use vasopressin first line, comparing it to norepinephrine first line and found benefit. I think that we could have a conversation about what that outcome was, but that may be a scenario where vasopressin makes sense. There also could be scenarios where a clinician wants to pick a drug that is a vasoconstrictor without inotropic properties. Often we think about phenylephrine in those cases, but I'm sure a lot of folks are hesitant to use phenylephrine, particularly in septic shock based on data available. So if that situation were to arise, such as a patient with left ventricular outflow tract obstruction, either from SAM or hypertrophic cardiomyopathy, it might make sense to use first line vasopressin or at minimum, very early vasopressin in higher doses. But I think that that needs to be an individualized decision where patient specific data come into the decision-making process for why a clinician would want to start vasopressin first line. Of the available data, the data do not suggest that as a whole, vasopressin first line is beneficial. Well, thank you very much, and this one is a little bit probably more for Gretchen, but I think it's quite interesting with vasopressin being costly when we compared where we start vasopressin in the medical and surgical ICU, there's completely a different effect where we tend to use it in the surgical ICU at a lower dose, but your data kind of shows that that might be right, or there might be a benefit, and then it's really hard to argue. But more importantly, with your algorithms, where do you have hydrocortisone? Did you put it in before you started vasopressin, at the same time, or after? So for what we, if you're referring to our new protocol where we initiate vasopressin or recommend it initiated earlier in the medical ICU for patients with septic shock, within that, we have actually recommended that they get started, they're on the same line. So if you start vasopressin, you start hydrocortisone, now clearly with logistics, which Seth also alluded to as well, availability and things like that, one might get started before the other or at the same time, or there is not a specified order, I will tell you, that we have indicated, but once that norepinephrine dose is around 15 mics per minute, that both should probably be discussed to be initiated. And I will say, too, that I think, I mean, we keep saying it over and over again, there's so clearly, there's so many gaps in the literature when it comes to these agents and patients with septic shock. But I think this is another really great question. We don't really fully know the interaction between these two drugs, as well as if there is an appropriate order in which you should start hydrocortisone in comparison to vasopressin. But my practice now just leans more towards just starting them at the same time is, I'm sure you're probably aware by the study itself, I'm very passionate about the timing of these interventions in patients with septic shock. So my practice is to recommend both get initiated as early as possible. Now, I actually think that that makes sense. And the way I somewhat think about this is no one's in the ICU because they have a norepinephrine or vasopressin deficiency, it's a sign of something else. And what we have to do is make sure are we treating, and in this case, the infection appropriately. And then sometimes you're going to have to let your patient kind of equilibrate and see what they're doing before you move on. I mean, and that's the hard part of what we do, correct? I mean, it's waiting and seeing. But no, I completely agree. Now, I do have another question from online. Can you share your thoughts on the biological explanation of why the mortality benefit described has lost beyond the norepinephrine equivalent dose of 60 mics per minute? Yeah, I think, I think for me, from a, as I kind of mentioned with the biologic plausibility of why early vasopressin is a benefit is that it seems maybe, again, rationalizing that we want to target vasopressin, restoring those vaso, deficient vasopressin stores early in the patient shock state when those vasopressin levels are low. And what I am thinking and what I rationalize is that once you get to that level of severity of illness, the initiation of this adjunctive agent is likely futile. And it's just not harming but not benefiting the patient in any capacity. So I think maybe one of those rationales all has to do with when the patient drops or depletes their vasopressin stores, and then restoring it early as possible. So delaying that restoration maybe is just, you know, causing the benefit of the drug to be futile at that point. Perfect. Do you have anything to add on that, Seth? No, I agree with everything Gretchen says, as always. Well, that's good. Today, I think that's a good choice. She has some sound advice for all of us. And I guess I'm going to also now ask about the, and no pun intended, the $64 million elephant in the room. Where do you really think the role of angiotensin II is in septic shock? I think that's a really hard question to answer right now based on the available data. I think there are excellent, thought-provoking articles recently published about using earlier angiotensin II and thinking about early combination therapy with norepinephrine, vasopressin, and angiotensin II. I think those can make some physiologic sense. But unfortunately, my interpretation of the data are that patient outcomes are unclear when angiotensin II is added. I think angiotensin II has established itself as an excellent vasopressor, meaning we know blood pressure goes up. Whether that translates to improved outcomes, I think, remains to be seen. But it is certainly an enticing opportunity and one that I am very much looking forward to additional research. I think at this point, the thing I'll conclude from my interpretation of the literature is that at minimum, I would not recommend angiotensin II prior to vasopressin based on outcome study and cost-effectiveness based on the currently available data. But I am hopeful that the near future will provide more information to help inform this discussion. I agree completely. I really don't actually have much significance to add on top of that. I look forward to learning more about this drug and how we can utilize it in our practice with future trials. Well, I mean, I think you guys really highlight the points. Obviously, we're talking about vasopressin and norepinephrine today. And these drugs have been out for a long, long time, let alone a drug like angiotensin II that's only been out for a short period of time. And I think we'll probably learn a lot. We do need to figure out what its role is and really when do you add it and what's too early and what's too long. So I think there's a lot more to come as well. So thank you very much. And this concludes our question and answer session. And I very much want to thank not only the presenters, Seth and Gretchen, but the audience. You guys asked some really great questions today. And thank you for attending. And again, for anyone who's joined us today, you will receive a follow-up email that will include an evaluation. So please take the three to five minutes to complete the evaluation. As always, your feedback is greatly appreciated. And then finally, on our final note, please join us for our next Journal Club in the Critical Care Medicine series. It's on Thursday, May 26th. And this concludes today's presentation. Thanks and have a great day.

webcast

Journal Club Critical Care Medicine

vasopressin

septic shock

rebranding

utilization

timing of initiation

mortality rates

cost-effective care