Our next speaker is Dr. Eithan Pelton, who is an associate professor at Intermountain Health and an adjunct professor at the University of Utah. Dr. Pelton is also the medical director of acute care research for the Intermountain Health System. His work focuses on the prevention, early identification, and management of critical illness. And today, he'll share with us some of that work in his talk titled, Sooner or Later, When Should We Start to Add Vasopressors? Thank you very much, Elizabeth. I do have some industry disclosures that are not related to the present talk. So I was asked to sort of address two questions today. The first being, when should we start vasopressors and septic shock? Now, the question of when is really a question of thresholds. It could be a time-related question, or it could be a question of threshold, a physiologic threshold. And really, most of the data that we have is about physiologic thresholds, not about time, not about how long we tolerate hypotension. But at the threshold at which we begin to initiate vasopressors. And the first question is, what is our vasopressor target? We do have some randomized trial-level evidence from this, from two major trials. First is the sepsis PAM trial, which compared blood pressure targets of 65 to 70 versus 80 to 85 in 29 centers in France. And these were patients who were on norepi of at least 0.1 micrograms per kilogram per minute for at least six hours, and they randomized 776 patients and saw no difference between the two treatment groups in terms of the primary outcome of mortality. Importantly, the high-target group and the low-target group both overshot their target range. The 65 to 70 target group was really almost entirely in the mid-70s range throughout the trial. And potentially, one of the consequences of this is, even with that overshoot in the low group, we saw in the higher-target group that those patients did have more arrhythmia, statistically more atrial arrhythmias, and numerically more potentially life-threatening ventricular arrhythmias, which provides to me pretty clear evidence that, in general population, we do want to target that lower target range. Another trial that we have recently is the 65 trial against 65 UK hospitals with targeting patients above age 65 and who are on vasopressors. This isn't restricted to septic shock. And this was a very large trial, over 2,400 patients, and compared a slightly low blood pressure, 60 to 65, sort of mild permissive hypotension versus usual care. In contrast to the sepsis PAM trial, they did achieve a nice separation of arms, but once again did not see a difference in the primary outcome, although there was a numerical trend toward benefit with permissive hypotension. It didn't achieve or approach statistical significance. The reduction in mortality was more pronounced in the permissive hypotension group and who were identified to be chronically hypertensive, so exactly the opposite of what you would have expected. I think, in general, what we know now, let's target 65. Next question is, what do we do in terms of fluid resuscitation against the threshold for other non-vasopressor interventions? The first thing we know, of course, is that the surviving sepsis campaign downgraded its recommendation for 30 milliliters per kilogram of initial fluid resuscitation within the first three hours after sepsis identification from a strong to a weak recommendation. That reflects the extensive evidence, shown by this slide, that underlies the 30 milliliters per kilogram recommendation. Now, I'm being a little facetious there, but what we really have is information about after initial fluid resuscitation, how much more fluid should we give? After at least some initial fluid. We now have the Clover's trial, and Ivor Diaglos is around here somewhere, so some of the folks who led the trial are around, but the Clover's trial asked, after initial fluid resuscitation, how much fluid should be given? These were adult patients, suspected or confirmed infection, and systolic blood pressure less than 100. They'd gotten at least a liter, but less than three liters of fluid. This study compared a liberal fluid resuscitation strategy, beginning with a one to two liter bolus of fluid, followed by up to additional three liters, total of five liters of fluid, triggered by hemodynamic parameters. Fluid was given 500 mils at a time, and you could reach from rescue vasopressors before that five liters if there were essentially extreme hypotension or refractory hypotension was being shown. Conversely, the restrictive strategy said stop IV fluid and start vasopressors for MAP less than 65 or systolic less than 90, and these patients could also receive fluid if there was refractory hypotension, but that was really limited. Both strategies were enacted for 24 hours, and the study did achieve pretty good separation of arms. There was about a two liter fluid difference, resuscitation fluid difference, over the first 24 hours, and about 60 versus about 35 percent, more or less, of patients received vasopressors, so good separation. Once again, however, we do not have a difference in mortality, and also no difference in mortality in any of the pre-specified heterogeneity of treatment effect analyses. This study did provide great information on peripheral vasopressors. Dr. Monroe's going to speak about that, but it does under a lot really provide backing for the surviving sepsis guideline that is new in the last iteration, which says don't wait for central venous access before starting vasopressors, so that shouldn't be one of our thresholds. Next question I want to address is when should we add vasopressin? So vasopressin has been, as you know, generated by the posterior pituitary. It has a role in both osmoregulation, sodium level regulation, as well as blood pressure regulation. The fact of a pituitary substance on raising the blood pressure was well recognized since the 1800s, and actually, as a drug agent, the peptide vasopressin was isolated and actually marketed in the U.S. since the 1920s for treatment of vasodilatory shock. There's particular interest in vasopressin in septic shock because of data that suggests that patients with septic shock are actually vasopressin deficient. This is data from a very influential, if small, physiologic study from the 1990s that showed that compared to patients with cardiogenic shock, and again, you'll notice just 12 patients in the cardiogenic shock, 19 in the septic shock group, patients with septic shock had relatively lower vasopressin levels, and this is potentially due to a biphasic release of vasopressin in these patients. So we know that, so using vasopressin to go out and bind our V1 receptors and assist with reversal of vasodilatory shock is, of course, particularly interesting in this population. I think most of the folks in this audience are probably familiar with the VAST trial led by James Russell and published, gosh, 15 years ago now in the New England Journal. This trial enrolled patients who had serious criteria, it's an old concept now, but still relevant, who were on norepi for at least five micrograms per minute for at least six hours and had the failure of at least one other organ system. Patients with MI, mesenteric ischemia, heart failure, severe heart failure were all excluded. The trial enrolled 772 patients comparing addition of vasopressin at a fix, at a moderately low dose to escalation of norepinephrine therapy. Primary outcome was 28-day mortality, and there was no difference in the overall population in that primary outcome or at the 90-day time point. But of course, what's influenced us so much is this heterogeneity of treatment effect analysis. So when the study looked at patients who were randomized and began treatment when they had more severe shock, defined by being on norepinephrine at 15 micrograms per minute or more, versus less severe shock, less than 15 micrograms per minute or more, there did potentially appear to be a mortality benefit in the patients who received, who started on vasopressin when they had less severe shock compared to the norepinephrine treatment group. This has been very influential in continuing the, and strengthening the recommendation that vasopressin is our second-line vasopressor in septic shock. But we do have some additional trials that have, that we need to consider. The first is the VANISH trial. This was a little bit different. This was actually a factorial trial that looked at vaso-versus-norepi and hydrocortisone-versus-placebo. And it didn't require patients to be on a particular dose of norepinephrine, or even to be on norepinephrine to be enrolled in the trial. It really, it technically was simply a first-line vasopressor trial, although 85% of patients were receiving norepinephrine. And they were actually receiving it at a pretty good dose, 0.15 micrograms per kilogram per minute on average. And the trial randomized patients to titrated vasopressin in the 0.01 to 0.06 range versus norepi in the 1 to 12 range. And like I said, 408 patients. This had a slightly complicated primary outcome. There's a lot of interest in whether vasopressin could improve renal outcomes particularly. And that was actually the primary outcome for this trial. Renal failure-free days. But the way it was analyzed, I've always had a hard time interpreting. And so this is actually not exactly the primary analysis data. This is because the analysis is focused on whether patients survived or didn't survive, as opposed to compositing mortality with renal failure. With that caveat, we really didn't see any difference for the renal failure-free days, the AKI, or 28-day mortality. Although there was a statistically significant reduction in the need for renal replacement therapy in the patients who received vasopressin. I really take this data with a large grain of salt. There were a lot of secondary outcomes in this analysis. And there was no adjustment for multiple comparisons, which means that there's a decent possibility that this could be a statistical fluke, as opposed to a true difference. Another trial that's looked at this question, again, looking at first-line vasopressors is a single central randomized trial from Brazil. 250 patients, specifically patients with cancer who had suspected infection and required vasopressors, were enrolled in this trial. And importantly, at least according to the published data, they were comparing vasopressin 0.01 to 0.06, same as the VANISH trial. But their norepinephrine group was receiving 10 to 60 micrograms per minute, which is not dose equivalent. 0.03 units per minute of vasopressin, or 1.8 units per hour in my system, is about equivalent to 0.1 of norepinephrine, or call it 10 to 15 micrograms per minute of norepinephrine. The 0.06 dose of vasopressin is about four-fold lower than the 60 microgram dose of norepinephrine. So there's a little bit of an issue there. But with that caveat, once again, we don't see a mortality difference, or in fact, in this study, any difference in renal outcomes. So that leaves us with some data suggesting that vasopressin is potentially beneficial in a patient-centered outcome. Certainly it works as a vasopressor, just like angiotensin 2 does. We do care, obviously, about safety. We worry a lot about vasopressin. We remember the studies that reported on mesenteric and soft tissue ischemia in studies where much, much, much higher doses of vasopressin were being used. That suggested that that was a problem. In these trials using lower doses of vasopressin in septic shock, there has never been a statistically significant difference in safety outcomes, including soft tissue ischemia. Although I do recognize that there was numerically more digital ischemia in both the VAST and VANISH trials, but not in the BANKS trial. And that leaves us with our surviving sepsis guideline, which, as the preceding speaker mentioned, is maybe a little vague. For adults with septic shock on norepinephrine with inadequate mean arterial pressures, we suggest adding vasopressin. What exactly do we mean by that? And that's further complicated by the fact that vasopressin, as I mentioned, is a drug that was grandfathered in before FDA authorizing legislation meant that all drugs were approved for safety after review of safety and efficacy. That did, in the Unapproved Drugs Initiative, get actually approval in 2014, which it was based solely on a literature review and led to market exclusivity from 2015 to 2022, and a resulting cost increase ranging from 50 to 100-fold, meaning that what used to be a drug that cost dollars per day for patients could cost hundreds or often thousands of dollars per day for a patient who was needing ongoing vasopressin therapy. A number of systems, as a result, introduced management practices to moderate the dose of vasopressin or restrict vasopressin use because of the only moderate efficacy data. But interestingly, it's not clear that that actually systematically changed behaviors. This is data from over 700 hospitals and nearly 300,000 patients who were on vasopressors for at least two days, and really not showing that the vertical line represents the rebranding point and the cost increase, really not showing a difference in the uptrend in vasopressin use. With that said, we also do see really large-scale practice variation for sepsis. This is what's called a caterpillar plot. On the X axis, we have hospitals. Each dot represents a hospital. The Y axis represents hospitals' case-mix-adjusted norepinephrine dose at which they start vasopressin. And what this kind of data shows, where we see hospitals' average initiation threshold for vasopressin ranging from 10 micrograms per minute all the way up to nearly 90 micrograms per minute, is it means that we have large-scale variation unrelated to patient needs, patient characteristics, or patient preferences. And a lot of equipoise about when and exactly what threshold of norepinephrine we should use for initiation of vasopressin. And we need more data in this space, pretty clearly. And we're excited to be pursuing this as we speak, actually. We will be launching on February 1st what we're calling the VASPR trial, which is a pragmatic, embedded, cluster-crossover, cluster-randomized trial that will examine two different treatment strategies for septic shock, a lower initiation threshold for vasopressin or a higher initiation threshold within the range of current practice. This clinicaltrials.gov number should be available publicly within the next 24 to 72 hours. It may not be publicly available quite yet. And with that, I want to thank you very much for your attention and appreciate your responses.

septic shock

vasopressors

vasopressin

fluid resuscitation

clinical trials