Hi, everyone. I'm going to talk about Introduction to Refractory Shock Assessment of Perfusion. My name is Zainab Gandhi. I'm a third-year internal medicine resident, so just be kind, and I'm too young to have any disclosures. All right, so I'm going to review the pathophysiology of refractory shock. I'm going to review the definition of refractory vasodilatory shock and factors to determine refractory shock and initial steps to pursue. I wanted to start with talking about how do we maintain homeostasis. It's a combination of three systems, that sympathetic system, vasopressin system, and renin angiotensin system. And during systemic inflammatory response, we've seen an insult to the homeostatic mechanism, but it's interesting to note that there's disproportionately low vasopressin levels and downregulation of angiotensin receptor during shock status, which is kind of counterregulatory to the mechanism that maintains homeostasis. So then we ask the question that what is the pathophysiology of refractory shock? And this is by far the best three-step algorithm that I've found in literature, which explains the pathophysiology, with the initial hypoxia and hyperlactatemia, which leads to an overproduction of nitric oxide, reactive oxygen species, activation of potassium channel, and adrenal insufficiency, which leads to a metabolic dysregulation and decreased response to the vascular smooth muscle, which leads to relaxation, endothelial dysfunction, coagulation pathway dysfunction, which we see as impaired responsiveness to catecholamine, and call it refractory vasodilatory shock. Then the next question that we ask is then what is refractory vasodilatory shock, and how do we define it? And as I stated earlier, I'm a resident. I'm only looking for a number so I can look smart doing rounds. And this is a fairly comprehensive table of landmark studies that have been conducted to study severe shock. And it's interesting to note that each of these studies have used a different definition, which first renders them incomparable, but also doesn't give me the answer that I need. But there have been some proposed definitions in literature, such as failure to achieve blood pressure goal despite vasopressor therapy, need for rescue vasopressor therapy, or need for high vasopressor doses, such as norepinephrine equivalents. But it's important to note that we do not have a consensus on the definition of refractory vasodilatory shock. All right. So before we say that somebody is in refractory shock, the first question is how do I know if the treatment for shock is working? So important parameters that we've used over time is lactate, urine output, capillary refill time, or mixed venous oxygen saturation. Early goal-directed therapy also used lactate as a marker to assess and start fluid resuscitation. And meta-analyses have shown that lactate clearance, if used as a guide as a therapy for sepsis, has shown mortality benefit. But then what about the non-hypoxic hyperlactateemia? So this study that was done used all patients who had elevated lactate and studied that who had capillary refill time, which was abnormal, and found only 30% had an abnormal CRT. And those people who had abnormal CRT had worse outcomes. So I guess we have two winners here, which showed up in the surviving sepsis guidelines in 2021, that you can use lactate as a marker for guiding fluid resuscitation and also use capillary refill time as an adjunct to it. Now before we say that somebody is in refractory vasodilatory shock, we must make sure that it's vasodilatory. So if patient is persistently hypotensive, despite being on vasopressors, it's important to assess cardiac output and fluid responsiveness, and make sure that we are not having a picture of undifferentiated or mixed shock with hypovolemic shock, cardiogenic shock, or obstructive shock. So when we talk about fluid responsiveness, how would I know if I've truly fixed the preload? And traditionally we have used static markers such as CVP and wedge pressure, which have become outdated for a while. And more recently, dynamic parameters such as mimicking fluid challenge has been used, amongst which passive leg raise test is the most validated in the literature. And this was a really good pictorial depiction of how one must do a passive leg raise test. The other mechanism would be heart-lung interaction, amongst which pulse pressure variation and stroke volume variation have been the most validated in the literature. But we're going to talk about it a little bit more. And the other one that I want to talk about is end expiratory occlusion test. So here, I wanted to simplify it for myself. So here are the four tests that I think are most validated in the literature. So in terms of mimicking a fluid challenge, passive leg raise test is most validated. It's irreversible. It's easy to do. It does not require external administration of fluids, but does need some provider expertise for direct estimation of cardiac output. Only confounders being compression, stocking use, or elevated intra-abdominal hypertension. It was also interesting to note that you can use change in end tidal CO2 to assess passive leg raise test, which is exciting because, you know, I'm still learning POCUS. The other two tests for heart-lung interaction, which are widely available, are un-widely validated are pulse pressure variation and stroke volume variation. While they're easy to do, it's important to remember they have a lot of confounders where they cannot be done in patients with arrhythmias, spontaneous breathings, if they have low tidal volumes, or if they have intra-abdominal hypertension. Which in my mind, I think, was overcome by the end expiratory occlusion test, which can be done in all those patients. But it does require interruption of breathing for 15 seconds, which if not done, becomes a confounder for this test. This also requires assessment of cardiac output, VTI, or perfusion index. Now, if you are assessing fluid responsiveness, it's important to remember these are continuous variables, and one value cannot be taken at a time as putting a patient in either the group of preload responder or a non-responder, and a continuous reassessment is recommended. All right, so next, can we say it's refractory unless source control is achieved? And it almost sounds intuitive, right, that if there is source, we must control it. We've been talking about it for decades. Even surviving sepsis campaign says if septic shock is present, you should administer antibiotics within one hour, and even if sepsis is possible and shock is absent, you must still administer antibiotics within three hours. But then why is source control important? And I kind of chose these two studies because they give two different parameters to it. The first one on the left is the study that compares two groups, patients who received source control and patients who did not, and they found that patients who received source control within 12 hours, they had a mortality benefit with a statistically significant hazard ratio, but it was interesting to note that the bundle compliance had decreased in the source control group. So the study on the lower right is implementation of sepsis protocol and outcomes associated with that. And they found implementation of that sepsis protocol led to decrease in vasopressor duration, ventilator duration, and ICU duration. So what I learned is that not only source control is important, timing is important, and also along with that, implementation of sepsis protocol can impact outcomes for our patients. But then what about those patients where you cannot find source? So this study that was published in 2016 took patients who did not have a source after first 24 hours in their ICU stay, and almost 50% of them had a sepsis mimicker, either such as adrenaline insufficiency, HLH, acute mesenteric ischemia, abdominal compartment syndrome, or toxic shock syndrome. And it's just a reminder that if you cannot find a source, make sure we are not missing a mimicker. All right. So I'm going to cite Dr. Sato here, just because, you know, I don't know why I'm here. These two people have done really good work in this area. They took patients with maximum norepinephrine equivalent dose, and they followed them with increasing vasopressor requirement and found increasing in-hospital mortality rate with increasing time with vasopressor doses. And I think going back to the initial part of the presentation where we stated that homeostasis multisystem, there is metabolic dysregulation with refractory shock. How can we truly just treat our patients with one vasopressor? So then the question that arises is, which adjunct agent should we use? And again, I'm citing my co-speakers here, who are much more accomplished, where Gretchen Sato and Seth Barr, they initiated early vasopressin and found mortality benefit to it last year. And then ATOS3 trial, where they also showed some positive signal with initiation of angiotensin early on. And this most recent study that came out from Brigham's used adjunctive vasopressin and corticosteroids early on and found decreased vasopressor duration overall. So I'm ending my presentation with some questions, which I hope will be answered in the next 45 minutes, is at what point do you consider catecholamine treatment failure? Do we need a consensus on a number? And when do you apply a secondary vasopressor? Which secondary vasopressor do you apply? And if homeostasis is truly multisystem, is there a role for multimodal vasopressor therapy? And can we call it refractory unless we've tried it all? And thank you. And I have to thank Dr. Dugarry, who is here, who has helped me a lot with this presentation. Thank you. Thank you so much for starting off our session really well. I'm going to be that moderator that says, for those in the back, there are seats up front. I don't want them to have to close this session. We want a lot of people to be able to attend. If you have a seat next to you too, maybe move in, make some room so that people can come and sit down before they have to spill us over into the overfill. So that actually leads us perfectly into the next session, which I'm going to lead, which is going to be talking about vasopressin as the first-line adjunctive agent. Or I'll either refer to it as the first-line adjunctive agent or your second-line vasopressor overall in norepinephrine refractory shock. I do have some audience response questions. Just one polling question. So the QR code is here. But it will pop up too once we get to it. What we're going to talk about — sorry, do you want me to go back? Okay. I talk really fast. I move fast. So I will remember to slow down. All right, perfect. What we're going to talk to you about in this part of the session is we're going to understand the role of vasopressin, talk about some patient populations that may benefit from its use, and then focus on the timing of its initiation in the course of patients with septic shock. Now to get right into the meat of the talk, we know that there are two large randomized controlled trials that have evaluated vasopressin, both VAST and VANISH. We're going to touch really briefly on VAST today, which randomized about 800 patients with septic shock to either norepinephrine alone or norepinephrine in combination with vasopressin. What this study found was overall no difference in 28-day or 90-day mortality between the treatment arms. When they broke down patients based on what they classified as having more or less severe shock, in the population of patients that had more shock, they also had similarly no differences in 28-day or 90-day mortality. However, when they looked at patients that were classified as having less severe shock, which they defined as having less than 15 mics per minute of norepinephrine requirements at the time of randomization, they found that patients who were randomized to receive vasopressin had lower 28-day and lower 90-day mortality. Now in light of these results, let's quickly refresh on the guideline recommendations regarding vasopressin. So although it's been talked about in each iteration of the guidelines, it wasn't until the 2021 iteration where vasopressin was actually identified as the first-line adjunctive agent or your second-line overall vasopressor in patients with septic shock. And historically, it had been recommended as either vasopressin or epinephrine as your second-line agent or your first-line adjunctive agent. The 2021 iteration also states that initiating vasopressin when the norepinephrine dose is between 0.25 and 0.5 mics per kilo per minute is, quote, sensible. However, in spite of all of this, there's still a lot of remaining questions regarding vasopressin's utilization. So I'd like to take a poll of the room and get a lay of the land, if you will, on how everybody in here utilizes vasopressin in your practice. So do you utilize it as your second-line vasoactive agent of choice at either higher norepinephrine requirements to trigger your initiation or lower norepinephrine requirements to trigger your initiation? Do you initiate it at the same time as norepinephrine is initiated? Do you have it as your third-line agent or do you not have a place in your practice for this agent? I'm going to give it a couple more seconds because we're getting a lot of votes trailing in. All right. Three, two, one. Okay. All right. Wow. So actually, I was going to think there was going to be more variation. You guys all started early, so I think we're done here. So I think I've convinced you all to start early. So I think I'm going to go home and enjoy the beach. Okay. No. So let's talk about why we do this. So I think when we talk about vasopressin response, it really starts to solidify why we continue to utilize this agent in light of those unclear mortality signals with its use. So the data regarding vasopressin response come from one retrospective analysis of 938 patients that attempted to look at how many patients who receive fixed-dose vasopressin respond to its initiation, and response being defined as achieving a goal map or mean arterial pressure of 65 or greater, and having a reduction in their background catecholamine dose six hours after the time of its initiation. Now what this study found was that 45% of patients who receive vasopressin respond to its initiation. And this response is independently associated with improved ICU mortality and 28-day mortality. And this is after adjustment for several severity of illness confounders. Now when looking at other measures of clinical outcomes, vasopressin is also associated with improved clinical trajectory. So you can see vasopressin responders less often experienced early death and more often experienced rapid recovery. Now breaking down these data further, it seems that these clinical outcomes differences can be mediated by differences in severity of illness. So essentially looking at less organ dysfunction and more vasoactive free days in vasopressin responders. So I take two points home from these data. One, this response definition of vasopressin can be used as an easy bedside indicator of patient's prognosis. And second, we should probably be looking at this outcome in our clinical trials as a potential outcome measure associated with the benefit to vasopressin's utilization. And we can use this definition potentially to pull out what patients, what phenotypes if you will, of patients are going to respond to its initiation and we can target our use in. So there have been several populations or subgroups that have been evaluated to determine their association with vasopressin response. We'll talk about a couple of these. So to first address arterial pH, vasopressin is often thought to retain its vasoconstrictive effects in the setting of acidosis. However, what you can actually see is lower rates of response, of vasopressin response, at lower pH levels. To break this down even further, for each .1 unit, the arterial pH was below 7.4, the odds of vasopressin response reduced by 21%. What these data indicate is that vasopressin's vasoactive effects are likely impaired in the setting of acidosis, similar to our other catecholamine agents. Now shifting gears into the genetic realm, there's one study that I actually like to talk about that looked at a single nucleotide polymorphism for the primary protein that metabolizes vasopressin, the leucineal cysteinyl aminopeptidase, and specifically found that patients with the TT genotype of this protein had higher vasopressin clearance and then lower survival. Now although this study did not look specifically at vasopressin response, to me it's telling us that there's a lot of genetic factors that may be associated with how vasopressin works in our patients, and we haven't even studied them yet. Now when we talk about vasopressin's utilization as an endocrine replacement, it makes sense to talk about vasopressin plasma concentrations. Now unfortunately the data at this point in time that we have available does not indicate an association between plasma vasopressin concentrations and vasopressin response, but we do still need more data in this realm. Similarly when we look at body weight of our patients, there has been no detected association with BMI or body weight and vasopressin response in patients with septic shock. Now with all of that, that leads to this question of okay, vasopressin response is great, maybe we can start to tease out some populations of patients that benefit, we need more data, but when should we start this drug, when we choose a patient that we are going to initiate it in? Now when we talk about timing of vasopressin initiation, you can think of it in many different ways at the bedside. So you can think of it as a specific norepinephrine or norepinephrine equivalent dose threshold that will trigger its initiation. You can think of it by how severely ill the patient, if you use lactate as a marker of how hypoperfused they are. You can even think of it temporally, as the time from shock onset or the time from vasoactive agent initiation. You can see here when VAS and VANISH initiated vasopressin in their trials, which I would argue is relatively early overall, maybe some room for improvement if you ask me. But when we look to clinical practice and how it's initiated and when it's initiated in clinical practice, so these data come from a retrospective evaluation of over 1,600 patients, you can see that it's pretty similar to overall to VAS and VANISH on our clinical trials, but I would argue later when we look at the norepinephrine equivalent dose. Now these data also went a step further and looked and tried to determine the association between these three timing variables and its association in hospital mortality. This was conducted at my institution, multiple hospitals across one hospital center with myself and Dr. Bauer as my colleague. What I want to discuss is each of these three timing variables and their results individually. First when looking at the norepinephrine dose or norepinephrine equivalent dose at vasopressin initiation, what we found was a linear association between norepinephrine dose at vasopressin initiation and in-hospital mortality. This figure shows you the predicted probability of in-hospital mortality at the various norepinephrine equivalent doses at vasopressin initiation. These were all accounted for different confounders including severity of illness that are listed for you on the slide. So to spell this out for you, the odds of mortality increased by 21% for every 10 mic per minute increase in the norepinephrine equivalent dose at the time of vasopressin initiation up to a norepinephrine equivalent dose of 60 mics per minute. Once that dose was exceeded, there was no association between the dose and in-hospital mortality. It was likely that it was futile to initiate that late. You can also see the results when broken down by clinically relevant breakpoints, what we deem to be clinically relevant breakpoints. So when initiating vasopressin at a norepinephrine equivalent dose of 25 mics per minute was associated with a 33% increase in the odds of mortality in comparison to just initiating at 10 mics per minute. Initiating at a dose of 60 mics per minute was associated with a two and a half fold increase in the odds of mortality in comparison to initiating at a dose of 10 mics per minute. Now when looking at lactate concentration at the time of vasopressin initiation, there was also a linear association between the lactate concentration and the in-hospital mortality. Now this got a little bit more complex because we detected an interaction between the lactate concentration and the timing of vasopressin initiation. So to evaluate their association, they had to be evaluated together. So you can see the various associations at different time from shock onset. So when vasopressin was initiated two hours from shock onset, the odds of mortality increased by 12% for each millimole per liter increase in the lactate concentration at vasopressin initiation. Similarly, when vasopressin was initiated 12 hours from shock onset, the odds of in-hospital mortality increased by 18% for each millimole per liter increase in lactate concentration at vasopressin initiation. And then finally, the timing, that temporal relationship, there was no association detected between the increasing time from shock onset and in-hospital mortality. And again, because of that interaction, this was evaluated at different lactate cutoffs. Overall, what I take home from these data though is that if you're going to initiate vasopressin in a patient, you should be initiating it earlier. Of course, we don't have a crystal ball to determine who I'm going to end up giving vasopressin to. But if you're on that fence, if you're thinking they're escalating their doses of norepinephrine, start that trigger, start it early in our patients. A couple of reasons why this may make sense. One of my hypotheses is that it could make sense when you think of vasopressin as an endocrine replacement. So in patients, when we're talking about having this depletion of vasopressin in this vasopressin depleted state, it could make sense that restoring your vasopressin concentrations early when they are reduced helps restore tissue perfusion more appropriately and quicker. Secondly, it could be that initiation of vasopressin early limits your overall norepinephrine and catecholamine exposure, thus preventing those untoward effects that we've seen and that we've talked about with our catecholamine agents like the immunomodulatory effects that we've been seeing more recently. So I'd like to revisit our poll. I know you guys all were so great in initiating vasopressin earlier, but I'd love to see if I've convinced anybody else in this group. So this is the same question, just trying to revisit to see now how you would utilize vasopressin going forward vasopressin in your practice. I'm going to go till 100. And we're there. OK. All right. So it's pretty similar, but I've definitely convinced more of you. I feel pretty proud. I'm accomplished today. I can go home really happy, go to the beach really happy. So thank you guys so much. To conclude, I know this was a lot of content for this particular session. We still have a lot more to go. There are still many remaining questions with vasopressin. Who responds? How do we determine? Why do they respond? Most importantly, what do we do when they don't respond? So with the data that we have at hand, though, right now, it's telling us, it's indicating to me we should start this drug earlier. And we should use this definition of response as a potential trigger point to think this patient is not responding. Maybe I need to pivot and look at alternative therapies in this patient population moving forward. Hello, everyone. Today I'm going to talk about lesser-used agents for reflactory shock. So what I mean by lesser-used agents is basically third-base pressors. And I'm mainly talking about angiotensin II and epinephrine. So my name is Ryota Sato. I'm working as an intensivist at the Queens Medical Center in Hawaii. And I work in the medical ICU as well as cardiac ICU. So today's objectives is, before getting into the basal pressors, I'm going to go into the running angiotensin ultrasound system. And also I'm going to talk about evidence of angiotensin II and also the potential of personalized basal pressor selection. And then lastly, I'm going to talk about epinephrine. So when you think about angiotensin II, you have to think about whole picture. So when your blood pressure decreases, your kidneys, actually it's a glomerular cell, release pro-raining. And then pro-raining is going to be converted to raining. So the raining is going to increase. And then that will increase angiotensinogen. Still, this doesn't have any hemodynamic effect, but this is going to induce angiotensin I. Mostly angiotensin-converting enzyme is going to convert angiotensin I to angiotensin II. And there are four known angiotensin receptor, but major hemodynamic effect is driven by AT1 receptor. So when angiotensin II is increased, most likely angiotensin I receptor is mostly activated. And that will induce basal constriction, other sternal secretions, stuff like that. There are some other pathways. When angiotensin-converting enzyme II works on angiotensin I, instead of angiotensin II, angiotensin I-9, angiotensin I-7 will be created. And these are going to be attaching to mass receptor, and this has opposite effect from AT1 receptor. So knowing that, Dr. Kanner and his colleagues published angiotensin II randomized control trial, which is called ATOS-III trial. So what they did was basically they compared angiotensin II and placebo in the setting of basal dilatory shock. So they included a patient with shock who is on above 0.2 microgram per keg per minute of norepinephrine for minimum six hours. They also mandated the patient to have high-output shock, meaning SCBO to about 70%, and also indexed above 2.3. Then they compared blood pressure. So the primary outcome here is a blood pressure increase. So this is what they got. So 70% of angiotensin II group actually had a blood pressure increase, and this is statistically significant. Something to note is in the secondary outcomes, they also compared all-cause mortality, even though that's not statistically significant. If you look at the numbers, it's actually lower in angiotensin 2 group. And then again, if you look at the 95 percent CI, it's close. It's not there yet, but it's close. And then, obviously, this study is not powered to make a difference for the mortality. So it's underpowered, likely to be underpowered for the mortality. So also, the conclusion for this study is angiotensin 2 effectively increased blood pressure in those who was high-output-reflecting basal dielectric shock. There are a couple of studies actually generated from ATOS 3 trials. So this is actually the study done by Dr. Belomo, who is also a part of the ATOS 3. So they actually checked renin, angiotensin 1, and angiotensin 2 before and three hours after angiotensin 2 administration. Then they noticed that the renin was actually way higher, you know, than normal limits. So anybody in shock who is receiving 0.2 microgram per keg per minutes of norepinephrine, they have high renin. And also, angiotensin 2 group actually had a significant renin reduction by administering angiotensin 2. So when you compare the ratio of reduction of the renin after starting angiotensin 2 versus placebo, angiotensin 2 group had almost 50% of reduction of renin level compared to the placebo. And actually, if you look at the median of renin level, anybody above that range, actually the angiotensin 2 was associated with reduced 28-day mortality. So this implies that if you have higher renin level, you might respond more to the angiotensin 2. Here's another study from Dr. Belomo. They actually included patients from ATOS 3, and also they included healthy volunteers. And they compared shock population and control population. When they checked angiotensin 1 level, they noted that angiotensin 1 in shock population is way higher than normal control population, whereas angiotensin 2 were similar. As a result, angiotensin 1 and angiotensin 2 ratio was so much higher than control population. And this was actually associated with higher vasopressor dose, and this is actually survival, lower survival. So angiotensin 1 and angiotensin 2 ratio, if it's higher than median number of this population, compared to the lower than the median number, the mortality was actually significantly lower. And then another study from also ATOS 3, basically they compared the population based on when angiotensin 2 was started. So because everybody was on angiotensin 2 after they received norepinephrine of 0.2 microgram per keg per minute, they divided the population into two groups, which is low group and basically patients on norepinephrine between 0.2 and 0.25, and high group, which is basically about 0.25 of norepinephrine. What they found was when they started angiotensin 2 earlier, which is basically 0.2 to 0.25 of norepinephrine, that was associated with better outcome compared to the high group, really not showing any significant difference. So this implies that if you start angiotensin 2 earlier than later, that might be better. And again, going back to the RAS system, so when you are in shock, your blood pressure is decreased, your renin's going up. This is shown by study. As a result, angiotensin 1 is also going up. And then actually angiotensin 2 is similar, but relatively, compared to the renin and angiotensin 1, it's relatively low. So that actually gives the feedback, and then try to create more renin. And then because angiotensin 2 is not created as requested, relatively, AT1 receptor stimulation is decreased. Instead, this is going to this pathway, so angiotensin 1-9, angiotensin 1-7 is more created. As a result, mass receptor is stimulated, which cause basal dilation and antithrombotic effect. So this is what's going on in the refractory shock. So what happens when you are administering angiotensin 2 in this setting? So what happens is, you cancel this out. So by giving angiotensin 2, this is a sort of hormonal replacement, you're basically suppressing this feedback. And then again, you are restoring this pathway, stimulating AT1, and then you're making basal constriction. So here, what we know about angiotensin 2. So angiotensin 2 effectively reduced other vasopressor dosage and increased blood pressure. And per protocol, or per the current evidence, angiotensin 2 may be indicated when norepinephrine is about 0.2 microgram per keg per minute, with high output status. This is a key. When patient's having refractory shock with a cardiac dysfunction, it's most likely not going to be indicated. This is more to come, but when renin is high, and also angiotensin 1 and angiotensin 2 ratio is high, they're most likely benefiting from angiotensin 2. And this may help us to personalize the basal pressure selection. And finally, when you start angiotensin 2 earlier, ideally between norepinephrine, between 0.2 and 0.25, that may be better than starting angiotensin 2 at a later stage of this shock. So this is what we know about angiotensin 2. So now we're moving on to epinephrine. We use this epinephrine a lot. And then oftentimes, it's used as inotropes or basal pressor. And then I will tell you, most of people here is going to use epinephrine as a third basal pressor, especially in the setting of septic shock. And first choice is norepinephrine, second choice is basal pressor, basal pressing, sorry. And then it's not really working, then you would probably add epinephrine as a third pressor. But there is no consensus about when to use or how to use. And surviving sepsis campaign guidelines suggest epinephrine as a monotherapy as an alternative to norepinephrine and dabutamine. In the setting of myocardial dysfunction was decreased perfusion despite adequate resuscitation. What that means. So in other words, in my mind, this is telling us that, okay, basically they are saying you could use dabutamine on top of norepinephrine or epinephrine monotherapy in the setting of mixed septic and cardiogenic shock. So here's a study from Critical Care Medicine. Basically they compared epinephrine and other basal pressors and inotropes in the setting of critically ill patients. And then they found there's no mortality difference. But they compared epinephrine and norepinephrine dabutamine or norepinephrine plus other inotropes. And there is no difference. But this is what we knew. But my question is, what happens when epinephrine is just added on the regular medications or used as a third pressor? Or maybe you might expect some inotropic effect. But that's what we want to know. So this study was done a couple years ago at the Queens Medical Center in Hawaii. And we included 400 patients with septic shock. And even though this is a retrospective observational study, to minimize the bias, we adjusted for LVEF, E over E-prime, CBRK scores, norepinephrine dosage. So we adjusted variables as much as we can. And what we found was epinephrine and dabutamine were associated with high mortality. And I know some people may say, hey, this is still a retrospective study. We cannot conclude yes. We cannot. However, what we know is epinephrine and dabutamine are associated with high mortality in this population, despite all the variables that I just said before. And also, this was time-dependent and dose-dependent. What that means is, if you're on epinephrine longer, or if you're on epinephrine higher dose, then that's even more associated with worse outcome. On top of it, epinephrine, dabutamine, merinone, all these inotropes were associated with AFib as RPR, which is known to be associated with worse outcome. So this is something to note, especially when you are just adding epinephrine just in case, or because other medications are not working. We just don't know if it's really doing a good thing. And then, because I mentioned the cardiogenic shock in the surviving sepsis campaign guidelines, this is also from cardiogenic shock population, from acute myocardial infarction. And this is, as far as I know, only decent size, well, actually, so small, but RCT, comparing norepinephrine, epinephrine head-to-head. The primary outcome was change in the cardiac index, and they had some other endpoints, but they basically included a cardiogenic shock from AMI. What they found was, yes, index increased in the first several hours, but then after that, index eventually became same with the norepinephrine and epinephrine. And then, obviously, as expected, heart rate, lactic acid, cardiac double product, epinephrine was way higher than norepinephrine, but these parameters, MAP, stroke volume, SbO2, were eventually same. And then, most importantly, epinephrine group had more reflective shock, worsening of organ failure, and, you know, lactic, well, organ failures. And then, eventually, this study was terminated earlier than expected because of this. So they couldn't include, you know, planned population because of this. So early termination because of this. So now, what we know about epinephrine is, existing evidence is mostly regarding epinephrine versus norepinephrine, plastobutamine, or other inotropes. And this is the reason why Surviving Sepsis Campaign guidelines recommending to use norepinephrine and dobutamine. As alternatives, they suggest using epinephrine as monotherapy. But this doesn't tell us if epinephrine is helpful as a third precursor. And what we know is, even in the setting of cardiogenic shock, pure cardiogenic shock from AMI, inotropes, especially epinephrine, we don't know if it's going to be helpful. And then, the current evidence that we have is actually all suggesting some harmful effects from epinephrine. And then, I would say, though, we have some population that responds to the epinephrine for sure, but we just don't know who. So if you see obvious, clear, clearly reduced ejection fraction or decreased cardiac function, and you suspect cardiogenic shock in the context of septic shock, then epinephrine could be tried. But you still have to be very cautious, because you might be contributing to worsening of outcomes by doing so. And then, if LVEF is maintained, let's say EF is 50%, 60%, or other parameters are OK, you're not specifically suspecting cardiogenic shock, epinephrine may not be helpful or even harmful, based on the quality of evidence. If you have to choose third base oppressors, angiotensin II, in that setting, may be preferred over epinephrine. I'll try to put all of this together and follow these three previous outstanding speakers to try to leave you with something to take home to your practice. My name is Seth Bauer. I'm from Cleveland Clinic. I will disclose that my institution has received grant funding from NIGMS, but it doesn't affect the content of this talk. And here, we'll try to develop an evidence-based regimen for patients with refractory shock. As Dr. Gandhi mentioned, we really don't know how to define refractory shock. There was a recent systematic review that tried to pick up how do we do this in septic shock. And they found that the definitions that were identified were inconsistent and highly variable. But they did have some common elements. The first is, we want to establish a hyperdynamic state, meaning this patient actually has vasodilation driving their shock syndrome. And also, some degree of hypotension despite some norepinephrine high dose. What that high dose is, we don't exactly know. They said the most common here was above 1 microkilo per minute. But I would submit that refractory shock may actually be in the eye of the beholder. What do I mean by that is, we have all heard these terms at our institution of the norepinephrine dose is maxed out. But maxed out is often an arbitrary dose limit that has been put in place at institutions. And I will state definitively here, there is no maximum dose of norepinephrine, no matter what anybody says. The maximum dose of norepinephrine is when your predicted adverse effects exceed the predicted efficacy. And we do not know where that is for any patient. It has to be individualized. It also could be in the eye of local culture. How do you use vasopressin? Are you using epinephrine as a vasopressor? Or are you only using one dose of vasopressin? Do you titrate the vasopressin? Do you use higher doses? Do you have angiotensin 2 on your formulary? All of these pieces come together to answer this question of, what is refractory shock? And I would submit, we just don't know. However, we may want to try to put all of this together in some sort of treatment algorithm. And this is just one example that you may be thinking about for your institution. Where if you have a patient that has septic shock, you're going to give them antibiotics and you're going to assess fluid responsiveness and use that to guide your fluid resuscitation and ensure adequate oxygen delivery. And if their MAP is low, then, sorry, if their MAP is adequate, great. Don't do anything else. Keep doing what you're doing. But if the MAP is low, start norepinephrine, titrate to your goal MAP. And if you achieve your goals there with relatively low doses of norepinephrine, you've achieved the goal. However, if the norepinephrine dose is a little bit higher, then you may be following Dr. Sasha's recommendations. And most of you do this anyway, where you start vasopressin early. And continuing on this algorithm, we think about I started vasopressin, if my MAP is still low, then I'm going to increase the norepinephrine dosage for these patients. And if it maintains relatively low, just keep on with norepinephrine. Or maybe you are aware of the data that Dr. Sato mentioned. You might want to think about adding angiotensin II here at relatively low norepinephrine doses. On the other side, you may think, oh, okay, if the norepinephrine dose is even higher, then I'm going to be thinking about adding corticosteroids. And maybe this is where you use epinephrine or angiotensin II in your practice. And algorithms like this are used all the time at the bedside. We use norepinephrine doses to guide our thinking about vasoactives and shock. But I would submit that this is only a starting point. This is where we might want to use this construct to approach the average patient. But we know most of the patients are not average. What I mean by that is we have to individualize this algorithm to our patients. And I would submit that the way to individualize this algorithm is detailing the shock profile. We know from several lectures earlier today that echo is the preferred approach to detailing the shock profile. And here we want to think about hemodynamic components. We like to use the simplification at our institution of filling, flow, and tone. So here we're thinking about assessing RAP or CVP to estimate cardiac preload. The LV function is an indicator of flow, and maybe you use surrogates of SCV02 and others if they're applicable. And tone with surrogates such as diastolic blood pressure. The key here though is these parameters are used to detail your shock profile. They are not indicators for treatment. Hopefully we have learned that lesson from CVP and giving fluid to these patients. And in fact, we shouldn't use numeric thresholds to make decisions for treatment approaches. What we should do is use adequacy assessments of is my LV function adequate, not just is my cardiac index 2.2 and I expect it to be higher. And we should use tissue perfusion markers such as capillary refill time to help guide this thinking of adequacy. This determination of the shock profile was highlighted in a recent study that was published related to initiating vasopressin. And I'm grateful that Dr. Sasha left it out of her presentation so I could talk about it. This was a study that assessed cardiac function in patients who ultimately received vasopressin. But the key here is all these patients had septic shock, receiving catecholamines, and they had an echo while they were receiving catecholamines but before the start of vasopressin. So this is a study trying to evaluate if echo parameters can have some indication of responsiveness to vasopressin. Here the exposure of interest was pre-vasopressin ventricular dysfunction and compared to non-existent ventricular dysfunction with the outcome of hemodynamic response which is the same vasopressin response Dr. Sasha already highlighted. Here in the primary analysis patients were grouped by responders versus non-responders and the only ventricular dysfunction that was different between these two groups were patients who were hemodynamic responders had a lower frequency of LV systolic dysfunction. Now at the bedside you don't have a crystal ball as Dr. Sasha said. You don't know who's going to respond. So this is a little bit hard to apply at the bedside. So the authors reoriented the data to answer the question that you may be asking is if I see LV systolic dysfunction what is the likelihood of a patient actually having hemodynamic response and in those that had LV systolic dysfunction they had a lower frequency of vasopressin response. In fact in patients who had LV systolic dysfunction again these are all vasopressin recipients mortality was higher in those who had LV systolic dysfunction and I'll call your attention to the first five days of therapy because this is when 90% of the patients received vasopressin and here you can see clear separation of these two curves suggesting that if we give vasopressin to patients with LV systolic dysfunction we may actually be causing harm. So I would submit that we should go back to our algorithm and maybe rethink this step of if the norepinephrine dose is 10 to 15 mics per minute maybe not just start vasopressin but this is the time when we should be assessing our LV function and this is the time for an echocardiogram and we ask the question is your LV function adequate? If you are thinking about adding another vasopressor I would argue you need to detail the shock profile to help guide your therapy further. And as you ask that question of if the LV function is adequate, if it is adequate then by all means start your vasopressin and you may continue that algorithm the exact same way as you were before. However if the patient doesn't respond to vasopressin and norepinephrine with higher doses of norepinephrine I would argue this is a warning signal to you and this warning signal should stick in your brain and will come back to this signal. However if the LV function is inadequate this is the time when we start asking the question of adding epinephrine or dobutamine. I would not add epi or dobutamine purely based on norepinephrine dose. As Dr. Sato mentioned the reason we're picking these drugs is for their inotropic properties not because we think that we need more vasoconstriction from adding these adjuncts. The question will be which one do I pick epinephrine or dobutamine in this scenario and I'm only aware of one trial that has tried to address this question and they kind of lucked in to the target population. This was a study that enrolled patients with septic shock receiving just 0.1 mpq per minute of norepinephrine and they had inadequate MAP. Just so happens the sample had a cardiac index below 2.2 and so it does fit this scenario that we're talking about here. They randomized patients to receive either dobutamine or epinephrine titrated to the same MAP or cardiac index goals and the primary outcome here was change in SOFA over four days and they did not detect a difference between these two therapies. However what they did find was that hemodynamic parameters differed between the two therapies such that index MAP and heart rate were higher with epinephrine. So if I had to pick between these two in this scenario I would probably pick epinephrine and Dr. Sato mentioned though the study that he was first author on. The question will be should you be doing this at all because observational data suggests that maybe we're doing more harm than good with adding these inotropes in patients with refractory septic shock. So if we come back to our algorithm and we ask these questions of well should I pick epi or dobutamine if the index is low. Maybe we should just keep titrating norepinephrine in this scenario because that scenario in that scenario norepinephrine may be an effective vasopressor and inotrope. We know that norepinephrine increases cardiac output by increasing mean systemic pressure and this is particularly effective in patients that have ventricular arterial decoupling. So in these patients norepinephrine alone may actually be the best therapy and has physiologic rationale. Now let's come back to that warning signal. That warning signal there of my norepinephrine dose is still escalating. I would submit this is the time when you need to redetail your shock profile. Yes this means another echo in the same patient. Yes this means maybe the same day or within a few hours another echo. Oh my goodness the cardiologists look yeah I know I see. But this is the time when we need to better understand that shock profile. If you have assessed the LB function and it's inadequate then this is useful information because you are back to that discussion of well should I add an inotrope or not. But I would also say it's also useful because you might think maybe I should be stopping vasopressin. Is vasopressin truly helpful here or is it harmful if I have induced a state where my LB function is now inadequate after adding vasopressin. But what should we do if we assess that the LB function is adequate in these patients. Some here would say I just keep titrating the norepinephrine. I'm on vaso and I have refractory vasodilatory shock. This is what I would do. However maybe you should increase the vasopressin dose. And for some of you this may be scary territory there's vasopressin is you know that that old infomercial set it forget it walk away. Maybe we should be thinking about vasopressin as a vasopressor and not just endocrine replacement therapy. And this was demonstrated in a relatively small study where they took patients who had vasodilatory shock on relatively high doses of norepinephrine and randomized them to vasopressin 0.03 units per minute or 0.06 units per minute. I see more decimals there than we're used to. This is a European study of two units per hour versus four units per hour. So that's why things look a little bit different. Small study but their primary goal here was to look at hemodynamic profile over time and they found lower norepinephrine requirements with the higher dosage without any differences detected in a signal for harm. So here if you're dealing with that refractory shock and you have assessed that the patient's LV function is adequate maybe we should be increasing the vasopressin dose to a higher dosage again using vasopressin as a vasopressor and maybe this is the time when we're also thinking about angiotensin 2. I'm not quite sure how to apply the literature for angiotensin 2. We know it's an effective vasopressor but at minimum I would say if you have it available to you in practice it makes the most sense to add angiotensin 2 to the combination of norepinephrine and vasopressin based on the data Dr. Sato mentioned and other data from this trial. In the end though what makes these vasoactives very hard to evaluate is that they all mimic the endogenous response and the endogenous response has this dynamic interplay where each of these agents actually affects the effect of all of the other agents. What does this mean? This means it's hard to do research in this area and hard to predict for an individual patient who you should use what drug in and that's why individualization is so key for refractory septic shock. In conclusion I'll leave you with a few takeaways. That is assessment is key for creating a systemic approach. By all means use algorithms but don't think of them as the end. Think of them as the start and individualize your approach and use the literature to guide agent selection and don't just rely on pharmacology. Thank you.

refractory shock

angiotensin II

epinephrine

renin-angiotensin system

ATAC-3 trial

vasodilatory shock

mortality

cardiogenic shock

echocardiographic assessment